QubesOS
/
qubes-linux-kernel
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity
You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
Branches Tags
${ item.name }
Create tag ${ searchTerm }
Create branch ${ searchTerm }
from '785cbcb166'
${ noResults }
qubes-linux-kernel/patches.xen/xen3-auto-xen-arch.diff

44653 lines
1.2 MiB
Raw Blame History

Subject: xen3 xen-arch
From: http://xenbits.xensource.com/linux-2.6.18-xen.hg (tip 1073:8fe973d8fb98)
Patch-mainline: n/a
Acked-by: jbeulich@novell.com
List of files having Xen derivates (perhaps created during the merging
of newer kernel versions), for xen-port-patches.py to pick up (i.e. this
must be retained here until the XenSource tree has these in the right
places):
+++ linux/arch/x86/kernel/acpi/sleep-xen.c
+++ linux/arch/x86/kernel/apic/io_apic-xen.c
+++ linux/arch/x86/kernel/apic/ipi-xen.c
+++ linux/arch/x86/kernel/apic/probe_32-xen.c
+++ linux/arch/x86/kernel/apic-xen.c
+++ linux/arch/x86/kernel/cpu/common_64-xen.c
+++ linux/arch/x86/kernel/e820-xen.c
+++ linux/arch/x86/kernel/head-xen.c
+++ linux/arch/x86/kernel/head32-xen.c
+++ linux/arch/x86/kernel/ioport-xen.c
+++ linux/arch/x86/kernel/io_apic-xen.c
+++ linux/arch/x86/kernel/ipi-xen.c
+++ linux/arch/x86/kernel/irq-xen.c
+++ linux/arch/x86/kernel/irq_work-xen.c
+++ linux/arch/x86/kernel/ldt-xen.c
+++ linux/arch/x86/kernel/microcode_core-xen.c
+++ linux/arch/x86/kernel/mpparse-xen.c
+++ linux/arch/x86/kernel/msr-xen.c
+++ linux/arch/x86/kernel/pci-nommu-xen.c
+++ linux/arch/x86/kernel/process-xen.c
+++ linux/arch/x86/kernel/setup-xen.c
+++ linux/arch/x86/kernel/smp-xen.c
+++ linux/arch/x86/kernel/traps-xen.c
+++ linux/arch/x86/kernel/x86_init-xen.c
+++ linux/arch/x86/lib/cache-smp-xen.c
+++ linux/arch/x86/mm/dump_pagetables-xen.c
+++ linux/arch/x86/mm/fault-xen.c
+++ linux/arch/x86/mm/init-xen.c
+++ linux/arch/x86/mm/iomap_32-xen.c
+++ linux/arch/x86/mm/pageattr-xen.c
+++ linux/arch/x86/mm/pat-xen.c
+++ linux/arch/x86/mm/pgtable-xen.c
+++ linux/arch/x86/vdso/vdso32-setup-xen.c
+++ linux/drivers/char/mem-xen.c
+++ linux/drivers/hwmon/coretemp-xen.c
+++ linux/drivers/hwmon/pkgtemp-xen.c
+++ linux/drivers/hwmon/via-cputemp-xen.c
+++ linux/arch/x86/include/mach-xen/asm/desc.h
+++ linux/arch/x86/include/mach-xen/asm/dma-mapping.h
+++ linux/arch/x86/include/mach-xen/asm/fixmap.h
+++ linux/arch/x86/include/mach-xen/asm/io.h
+++ linux/arch/x86/include/mach-xen/asm/ipi.h
+++ linux/arch/x86/include/mach-xen/asm/irq_vectors.h
+++ linux/arch/x86/include/mach-xen/asm/irqflags.h
+++ linux/arch/x86/include/mach-xen/asm/mmu_context.h
+++ linux/arch/x86/include/mach-xen/asm/pci.h
+++ linux/arch/x86/include/mach-xen/asm/perf_event.h
+++ linux/arch/x86/include/mach-xen/asm/pgalloc.h
+++ linux/arch/x86/include/mach-xen/asm/pgtable.h
+++ linux/arch/x86/include/mach-xen/asm/pgtable-3level_types.h
+++ linux/arch/x86/include/mach-xen/asm/pgtable_64_types.h
+++ linux/arch/x86/include/mach-xen/asm/pgtable_types.h
+++ linux/arch/x86/include/mach-xen/asm/processor.h
+++ linux/arch/x86/include/mach-xen/asm/smp.h
+++ linux/arch/x86/include/mach-xen/asm/spinlock.h
+++ linux/arch/x86/include/mach-xen/asm/spinlock_types.h
+++ linux/arch/x86/include/mach-xen/asm/swiotlb.h
+++ linux/arch/x86/include/mach-xen/asm/system.h
+++ linux/arch/x86/include/mach-xen/asm/tlbflush.h
+++ linux/arch/x86/include/mach-xen/asm/xor.h
List of files folded into their native counterparts (and hence removed
from this patch for xen-port-patches.py to not needlessly pick them up;
for reference, prefixed with the version the removal occured):
2.6.18/arch/x86/kernel/quirks-xen.c
2.6.18/arch/x86/include/mach-xen/asm/pgtable-2level.h
2.6.18/arch/x86/include/mach-xen/asm/pgtable-2level-defs.h
2.6.19/arch/x86/include/mach-xen/asm/ptrace.h
2.6.23/arch/x86/include/mach-xen/asm/ptrace_64.h
2.6.23/arch/x86/kernel/vsyscall-note_32-xen.S
2.6.24/arch/x86/include/mach-xen/asm/arch_hooks_64.h
2.6.24/arch/x86/include/mach-xen/asm/bootsetup_64.h
2.6.24/arch/x86/include/mach-xen/asm/mmu_32.h
2.6.24/arch/x86/include/mach-xen/asm/mmu_64.h
2.6.24/arch/x86/include/mach-xen/asm/nmi_64.h
2.6.24/arch/x86/include/mach-xen/asm/setup.h
2.6.24/arch/x86/include/mach-xen/asm/time_64.h (added in 2.6.20)
2.6.24/arch/x86/include/mach-xen/mach_timer.h
2.6.24/arch/x86/kernel/early_printk_32-xen.c
2.6.25/arch/x86/ia32/syscall32-xen.c
2.6.25/arch/x86/ia32/syscall32_syscall-xen.S
2.6.25/arch/x86/ia32/vsyscall-int80.S
2.6.25/arch/x86/include/mach-xen/asm/msr.h
2.6.25/arch/x86/include/mach-xen/asm/page_32.h
2.6.25/arch/x86/include/mach-xen/asm/spinlock_32.h
2.6.25/arch/x86/include/mach-xen/asm/timer.h (added in 2.6.24)
2.6.25/arch/x86/include/mach-xen/asm/timer_64.h
2.6.25/arch/x86/include/mach-xen/mach_time.h
2.6.25/arch/x86/kernel/acpi/boot-xen.c
2.6.26/arch/x86/include/mach-xen/asm/dma-mapping_32.h
2.6.26/arch/x86/include/mach-xen/asm/dma-mapping_64.h
2.6.26/arch/x86/include/mach-xen/asm/nmi.h (added in 2.6.24)
2.6.26/arch/x86/include/mach-xen/asm/scatterlist.h (added in 2.6.24)
2.6.26/arch/x86/include/mach-xen/asm/scatterlist_32.h
2.6.26/arch/x86/include/mach-xen/asm/swiotlb_32.h
2.6.26/arch/x86/kernel/pci-dma_32-xen.c
2.6.26/arch/x86/kernel/pci-swiotlb_64-xen.c
2.6.26/include/xen/xencomm.h
2.6.27/arch/x86/include/mach-xen/asm/e820.h (added in 2.6.24)
2.6.27/arch/x86/include/mach-xen/asm/e820_64.h
2.6.27/arch/x86/include/mach-xen/asm/hw_irq.h (added in 2.6.24)
2.6.27/arch/x86/include/mach-xen/asm/hw_irq_32.h
2.6.27/arch/x86/include/mach-xen/asm/hw_irq_64.h
2.6.27/arch/x86/include/mach-xen/asm/io_32.h
2.6.27/arch/x86/include/mach-xen/asm/io_64.h
2.6.27/arch/x86/include/mach-xen/asm/irq.h (added in 2.6.24)
2.6.27/arch/x86/include/mach-xen/asm/irq_64.h
2.6.27/arch/x86/kernel/e820_32-xen.c
2.6.28/arch/x86/include/mach-xen/asm/pci_64.h
2.6.28/arch/x86/include/mach-xen/asm/segment.h (added in 2.6.24)
2.6.28/arch/x86/include/mach-xen/asm/segment_32.h
2.6.30/arch/x86/include/mach-xen/asm/page.h (added in 2.6.24)
2.6.30/arch/x86/include/mach-xen/asm/page_64.h
2.6.30/arch/x86/include/mach-xen/asm/pci_32.h
2.6.30/arch/x86/kernel/apic/apic_xen_64.c
2.6.30/arch/x86/kernel/apic/probe_64-xen.c
2.6.30/arch/x86/kernel/setup_percpu-xen.c (added in 2.6.27)
2.6.31/arch/x86/kernel/init_task-xen.c
2.6.32/arch/x86/include/mach-xen/asm/setup_arch.h
2.6.33/arch/x86/kernel/irq_32-xen.c
2.6.33/arch/x86/kernel/irq_64-xen.c
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/kernel/acpi/processor_extcntl_xen.c 2010-03-22 12:00:53.000000000 +0100
@@ -0,0 +1,208 @@
+/*
+ * processor_extcntl_xen.c - interface to notify Xen
+ *
+ * Copyright (C) 2008, Intel corporation
+ *
+ * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or (at
+ * your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License along
+ * with this program; if not, write to the Free Software Foundation, Inc.,
+ * 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA.
+ *
+ */
+
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/types.h>
+#include <linux/acpi.h>
+#include <linux/pm.h>
+#include <linux/cpu.h>
+
+#include <linux/cpufreq.h>
+#include <acpi/processor.h>
+#include <asm/hypercall.h>
+
+static int xen_cx_notifier(struct acpi_processor *pr, int action)
+{
+ int ret, count = 0, i;
+ xen_platform_op_t op = {
+ .cmd = XENPF_set_processor_pminfo,
+ .interface_version = XENPF_INTERFACE_VERSION,
+ .u.set_pminfo.id = pr->acpi_id,
+ .u.set_pminfo.type = XEN_PM_CX,
+ };
+ struct xen_processor_cx *data, *buf;
+ struct acpi_processor_cx *cx;
+
+ /* Convert to Xen defined structure and hypercall */
+ buf = kzalloc(pr->power.count * sizeof(struct xen_processor_cx),
+ GFP_KERNEL);
+ if (!buf)
+ return -ENOMEM;
+
+ data = buf;
+ for (i = 1; i <= pr->power.count; i++) {
+ cx = &pr->power.states[i];
+ /* Skip invalid cstate entry */
+ if (!cx->valid)
+ continue;
+
+ data->type = cx->type;
+ data->latency = cx->latency;
+ data->power = cx->power;
+ data->reg.space_id = cx->reg.space_id;
+ data->reg.bit_width = cx->reg.bit_width;
+ data->reg.bit_offset = cx->reg.bit_offset;
+ data->reg.access_size = cx->reg.reserved;
+ data->reg.address = cx->reg.address;
+
+ /* Get dependency relationships */
+ if (cx->csd_count) {
+ printk("Wow! _CSD is found. Not support for now!\n");
+ kfree(buf);
+ return -EINVAL;
+ } else {
+ data->dpcnt = 0;
+ set_xen_guest_handle(data->dp, NULL);
+ }
+
+ data++;
+ count++;
+ }
+
+ if (!count) {
+ printk("No available Cx info for cpu %d\n", pr->acpi_id);
+ kfree(buf);
+ return -EINVAL;
+ }
+
+ op.u.set_pminfo.u.power.count = count;
+ op.u.set_pminfo.u.power.flags.bm_control = pr->flags.bm_control;
+ op.u.set_pminfo.u.power.flags.bm_check = pr->flags.bm_check;
+ op.u.set_pminfo.u.power.flags.has_cst = pr->flags.has_cst;
+ op.u.set_pminfo.u.power.flags.power_setup_done = pr->flags.power_setup_done;
+
+ set_xen_guest_handle(op.u.set_pminfo.u.power.states, buf);
+ ret = HYPERVISOR_platform_op(&op);
+ kfree(buf);
+ return ret;
+}
+
+static int xen_px_notifier(struct acpi_processor *pr, int action)
+{
+ int ret = -EINVAL;
+ xen_platform_op_t op = {
+ .cmd = XENPF_set_processor_pminfo,
+ .interface_version = XENPF_INTERFACE_VERSION,
+ .u.set_pminfo.id = pr->acpi_id,
+ .u.set_pminfo.type = XEN_PM_PX,
+ };
+ struct xen_processor_performance *perf;
+ struct xen_processor_px *states = NULL;
+ struct acpi_processor_performance *px;
+ struct acpi_psd_package *pdomain;
+
+ if (!pr)
+ return -EINVAL;
+
+ perf = &op.u.set_pminfo.u.perf;
+ px = pr->performance;
+ if (!px)
+ return -EINVAL;
+
+ switch(action) {
+ case PROCESSOR_PM_CHANGE:
+ /* ppc dynamic handle */
+ perf->flags = XEN_PX_PPC;
+ perf->platform_limit = pr->performance_platform_limit;
+
+ ret = HYPERVISOR_platform_op(&op);
+ break;
+
+ case PROCESSOR_PM_INIT:
+ /* px normal init */
+ perf->flags = XEN_PX_PPC |
+ XEN_PX_PCT |
+ XEN_PX_PSS |
+ XEN_PX_PSD;
+
+ /* ppc */
+ perf->platform_limit = pr->performance_platform_limit;
+
+ /* pct */
+ xen_convert_pct_reg(&perf->control_register, &px->control_register);
+ xen_convert_pct_reg(&perf->status_register, &px->status_register);
+
+ /* pss */
+ perf->state_count = px->state_count;
+ states = kzalloc(px->state_count*sizeof(xen_processor_px_t),GFP_KERNEL);
+ if (!states)
+ return -ENOMEM;
+ xen_convert_pss_states(states, px->states, px->state_count);
+ set_xen_guest_handle(perf->states, states);
+
+ /* psd */
+ pdomain = &px->domain_info;
+ xen_convert_psd_pack(&perf->domain_info, pdomain);
+ if (pdomain->coord_type == DOMAIN_COORD_TYPE_SW_ALL)
+ perf->shared_type = CPUFREQ_SHARED_TYPE_ALL;
+ else if (pdomain->coord_type == DOMAIN_COORD_TYPE_SW_ANY)
+ perf->shared_type = CPUFREQ_SHARED_TYPE_ANY;
+ else if (pdomain->coord_type == DOMAIN_COORD_TYPE_HW_ALL)
+ perf->shared_type = CPUFREQ_SHARED_TYPE_HW;
+ else {
+ ret = -ENODEV;
+ kfree(states);
+ break;
+ }
+
+ ret = HYPERVISOR_platform_op(&op);
+ kfree(states);
+ break;
+
+ default:
+ break;
+ }
+
+ return ret;
+}
+
+static int xen_tx_notifier(struct acpi_processor *pr, int action)
+{
+ return -EINVAL;
+}
+static int xen_hotplug_notifier(struct acpi_processor *pr, int event)
+{
+ return -EINVAL;
+}
+
+static struct processor_extcntl_ops xen_extcntl_ops = {
+ .hotplug = xen_hotplug_notifier,
+};
+
+void arch_acpi_processor_init_extcntl(const struct processor_extcntl_ops **ops)
+{
+ unsigned int pmbits = (xen_start_info->flags & SIF_PM_MASK) >> 8;
+
+ if (!pmbits)
+ return;
+ if (pmbits & XEN_PROCESSOR_PM_CX)
+ xen_extcntl_ops.pm_ops[PM_TYPE_IDLE] = xen_cx_notifier;
+ if (pmbits & XEN_PROCESSOR_PM_PX)
+ xen_extcntl_ops.pm_ops[PM_TYPE_PERF] = xen_px_notifier;
+ if (pmbits & XEN_PROCESSOR_PM_TX)
+ xen_extcntl_ops.pm_ops[PM_TYPE_THR] = xen_tx_notifier;
+
+ *ops = &xen_extcntl_ops;
+}
+EXPORT_SYMBOL(arch_acpi_processor_init_extcntl);
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/kernel/acpi/sleep_32-xen.c 2008-04-15 09:29:41.000000000 +0200
@@ -0,0 +1,113 @@
+/*
+ * sleep.c - x86-specific ACPI sleep support.
+ *
+ * Copyright (C) 2001-2003 Patrick Mochel
+ * Copyright (C) 2001-2003 Pavel Machek <pavel@suse.cz>
+ */
+
+#include <linux/acpi.h>
+#include <linux/bootmem.h>
+#include <linux/dmi.h>
+#include <linux/cpumask.h>
+
+#include <asm/smp.h>
+
+#ifndef CONFIG_ACPI_PV_SLEEP
+/* address in low memory of the wakeup routine. */
+unsigned long acpi_wakeup_address = 0;
+unsigned long acpi_video_flags;
+extern char wakeup_start, wakeup_end;
+
+extern unsigned long FASTCALL(acpi_copy_wakeup_routine(unsigned long));
+#endif
+
+/**
+ * acpi_save_state_mem - save kernel state
+ *
+ * Create an identity mapped page table and copy the wakeup routine to
+ * low memory.
+ */
+int acpi_save_state_mem(void)
+{
+#ifndef CONFIG_ACPI_PV_SLEEP
+ if (!acpi_wakeup_address)
+ return 1;
+ memcpy((void *)acpi_wakeup_address, &wakeup_start,
+ &wakeup_end - &wakeup_start);
+ acpi_copy_wakeup_routine(acpi_wakeup_address);
+#endif
+ return 0;
+}
+
+/*
+ * acpi_restore_state - undo effects of acpi_save_state_mem
+ */
+void acpi_restore_state_mem(void)
+{
+}
+
+/**
+ * acpi_reserve_bootmem - do _very_ early ACPI initialisation
+ *
+ * We allocate a page from the first 1MB of memory for the wakeup
+ * routine for when we come back from a sleep state. The
+ * runtime allocator allows specification of <16MB pages, but not
+ * <1MB pages.
+ */
+void __init acpi_reserve_bootmem(void)
+{
+#ifndef CONFIG_ACPI_PV_SLEEP
+ if ((&wakeup_end - &wakeup_start) > PAGE_SIZE) {
+ printk(KERN_ERR
+ "ACPI: Wakeup code way too big, S3 disabled.\n");
+ return;
+ }
+
+ acpi_wakeup_address = (unsigned long)alloc_bootmem_low(PAGE_SIZE);
+ if (!acpi_wakeup_address)
+ printk(KERN_ERR "ACPI: Cannot allocate lowmem, S3 disabled.\n");
+#endif
+}
+
+#ifndef CONFIG_ACPI_PV_SLEEP
+static int __init acpi_sleep_setup(char *str)
+{
+ while ((str != NULL) && (*str != '\0')) {
+ if (strncmp(str, "s3_bios", 7) == 0)
+ acpi_video_flags = 1;
+ if (strncmp(str, "s3_mode", 7) == 0)
+ acpi_video_flags |= 2;
+ str = strchr(str, ',');
+ if (str != NULL)
+ str += strspn(str, ", \t");
+ }
+ return 1;
+}
+
+__setup("acpi_sleep=", acpi_sleep_setup);
+
+static __init int reset_videomode_after_s3(struct dmi_system_id *d)
+{
+ acpi_video_flags |= 2;
+ return 0;
+}
+
+static __initdata struct dmi_system_id acpisleep_dmi_table[] = {
+ { /* Reset video mode after returning from ACPI S3 sleep */
+ .callback = reset_videomode_after_s3,
+ .ident = "Toshiba Satellite 4030cdt",
+ .matches = {
+ DMI_MATCH(DMI_PRODUCT_NAME, "S4030CDT/4.3"),
+ },
+ },
+ {}
+};
+
+static int __init acpisleep_dmi_init(void)
+{
+ dmi_check_system(acpisleep_dmi_table);
+ return 0;
+}
+
+core_initcall(acpisleep_dmi_init);
+#endif /* CONFIG_ACPI_PV_SLEEP */
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/kernel/apic/apic-xen.c 2007-06-12 13:12:48.000000000 +0200
@@ -0,0 +1,155 @@
+/*
+ * Local APIC handling, local APIC timers
+ *
+ * (c) 1999, 2000 Ingo Molnar <mingo@redhat.com>
+ *
+ * Fixes
+ * Maciej W. Rozycki : Bits for genuine 82489DX APICs;
+ * thanks to Eric Gilmore
+ * and Rolf G. Tews
+ * for testing these extensively.
+ * Maciej W. Rozycki : Various updates and fixes.
+ * Mikael Pettersson : Power Management for UP-APIC.
+ * Pavel Machek and
+ * Mikael Pettersson : PM converted to driver model.
+ */
+
+#include <linux/init.h>
+
+#include <linux/mm.h>
+#include <linux/delay.h>
+#include <linux/bootmem.h>
+#include <linux/smp_lock.h>
+#include <linux/interrupt.h>
+#include <linux/mc146818rtc.h>
+#include <linux/kernel_stat.h>
+#include <linux/sysdev.h>
+#include <linux/cpu.h>
+#include <linux/module.h>
+
+#include <asm/atomic.h>
+#include <asm/smp.h>
+#include <asm/mtrr.h>
+#include <asm/mpspec.h>
+#include <asm/desc.h>
+#include <asm/arch_hooks.h>
+#include <asm/hpet.h>
+#include <asm/i8253.h>
+#include <asm/nmi.h>
+
+#include <mach_apic.h>
+#include <mach_apicdef.h>
+#include <mach_ipi.h>
+
+#include "io_ports.h"
+
+#ifndef CONFIG_XEN
+/*
+ * cpu_mask that denotes the CPUs that needs timer interrupt coming in as
+ * IPIs in place of local APIC timers
+ */
+static cpumask_t timer_bcast_ipi;
+#endif
+
+/*
+ * Knob to control our willingness to enable the local APIC.
+ */
+int enable_local_apic __initdata = 0; /* -1=force-disable, +1=force-enable */
+
+/*
+ * Debug level
+ */
+int apic_verbosity;
+
+#ifndef CONFIG_XEN
+static int modern_apic(void)
+{
+ unsigned int lvr, version;
+ /* AMD systems use old APIC versions, so check the CPU */
+ if (boot_cpu_data.x86_vendor == X86_VENDOR_AMD &&
+ boot_cpu_data.x86 >= 0xf)
+ return 1;
+ lvr = apic_read(APIC_LVR);
+ version = GET_APIC_VERSION(lvr);
+ return version >= 0x14;
+}
+#endif /* !CONFIG_XEN */
+
+/*
+ * 'what should we do if we get a hw irq event on an illegal vector'.
+ * each architecture has to answer this themselves.
+ */
+void ack_bad_irq(unsigned int irq)
+{
+ printk("unexpected IRQ trap at vector %02x\n", irq);
+ /*
+ * Currently unexpected vectors happen only on SMP and APIC.
+ * We _must_ ack these because every local APIC has only N
+ * irq slots per priority level, and a 'hanging, unacked' IRQ
+ * holds up an irq slot - in excessive cases (when multiple
+ * unexpected vectors occur) that might lock up the APIC
+ * completely.
+ * But only ack when the APIC is enabled -AK
+ */
+ if (cpu_has_apic)
+ ack_APIC_irq();
+}
+
+int get_physical_broadcast(void)
+{
+ return 0xff;
+}
+
+#ifndef CONFIG_XEN
+#ifndef CONFIG_SMP
+static void up_apic_timer_interrupt_call(struct pt_regs *regs)
+{
+ int cpu = smp_processor_id();
+
+ /*
+ * the NMI deadlock-detector uses this.
+ */
+ per_cpu(irq_stat, cpu).apic_timer_irqs++;
+
+ smp_local_timer_interrupt(regs);
+}
+#endif
+
+void smp_send_timer_broadcast_ipi(struct pt_regs *regs)
+{
+ cpumask_t mask;
+
+ cpus_and(mask, cpu_online_map, timer_bcast_ipi);
+ if (!cpus_empty(mask)) {
+#ifdef CONFIG_SMP
+ send_IPI_mask(mask, LOCAL_TIMER_VECTOR);
+#else
+ /*
+ * We can directly call the apic timer interrupt handler
+ * in UP case. Minus all irq related functions
+ */
+ up_apic_timer_interrupt_call(regs);
+#endif
+ }
+}
+#endif
+
+int setup_profiling_timer(unsigned int multiplier)
+{
+ return -EINVAL;
+}
+
+/*
+ * This initializes the IO-APIC and APIC hardware if this is
+ * a UP kernel.
+ */
+int __init APIC_init_uniprocessor (void)
+{
+#ifdef CONFIG_X86_IO_APIC
+ if (smp_found_config)
+ if (!skip_ioapic_setup && nr_ioapics)
+ setup_IO_APIC();
+#endif
+
+ return 0;
+}
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/kernel/cpu/common-xen.c 2009-05-19 09:16:41.000000000 +0200
@@ -0,0 +1,745 @@
+#include <linux/init.h>
+#include <linux/string.h>
+#include <linux/delay.h>
+#include <linux/smp.h>
+#include <linux/module.h>
+#include <linux/percpu.h>
+#include <linux/bootmem.h>
+#include <asm/semaphore.h>
+#include <asm/processor.h>
+#include <asm/i387.h>
+#include <asm/msr.h>
+#include <asm/io.h>
+#include <asm/mmu_context.h>
+#include <asm/mtrr.h>
+#include <asm/mce.h>
+#ifdef CONFIG_X86_LOCAL_APIC
+#include <asm/mpspec.h>
+#include <asm/apic.h>
+#include <mach_apic.h>
+#else
+#ifdef CONFIG_XEN
+#define phys_pkg_id(a,b) a
+#endif
+#endif
+#include <asm/hypervisor.h>
+
+#include "cpu.h"
+
+DEFINE_PER_CPU(struct Xgt_desc_struct, cpu_gdt_descr);
+EXPORT_PER_CPU_SYMBOL(cpu_gdt_descr);
+
+#ifndef CONFIG_XEN
+DEFINE_PER_CPU(unsigned char, cpu_16bit_stack[CPU_16BIT_STACK_SIZE]);
+EXPORT_PER_CPU_SYMBOL(cpu_16bit_stack);
+#endif
+
+static int cachesize_override __cpuinitdata = -1;
+static int disable_x86_fxsr __cpuinitdata;
+static int disable_x86_serial_nr __cpuinitdata = 1;
+static int disable_x86_sep __cpuinitdata;
+
+struct cpu_dev * cpu_devs[X86_VENDOR_NUM] = {};
+
+extern int disable_pse;
+
+static void default_init(struct cpuinfo_x86 * c)
+{
+ /* Not much we can do here... */
+ /* Check if at least it has cpuid */
+ if (c->cpuid_level == -1) {
+ /* No cpuid. It must be an ancient CPU */
+ if (c->x86 == 4)
+ strcpy(c->x86_model_id, "486");
+ else if (c->x86 == 3)
+ strcpy(c->x86_model_id, "386");
+ }
+}
+
+static struct cpu_dev default_cpu = {
+ .c_init = default_init,
+ .c_vendor = "Unknown",
+};
+static struct cpu_dev * this_cpu = &default_cpu;
+
+static int __init cachesize_setup(char *str)
+{
+ get_option (&str, &cachesize_override);
+ return 1;
+}
+__setup("cachesize=", cachesize_setup);
+
+int __cpuinit get_model_name(struct cpuinfo_x86 *c)
+{
+ unsigned int *v;
+ char *p, *q;
+
+ if (cpuid_eax(0x80000000) < 0x80000004)
+ return 0;
+
+ v = (unsigned int *) c->x86_model_id;
+ cpuid(0x80000002, &v[0], &v[1], &v[2], &v[3]);
+ cpuid(0x80000003, &v[4], &v[5], &v[6], &v[7]);
+ cpuid(0x80000004, &v[8], &v[9], &v[10], &v[11]);
+ c->x86_model_id[48] = 0;
+
+ /* Intel chips right-justify this string for some dumb reason;
+ undo that brain damage */
+ p = q = &c->x86_model_id[0];
+ while ( *p == ' ' )
+ p++;
+ if ( p != q ) {
+ while ( *p )
+ *q++ = *p++;
+ while ( q <= &c->x86_model_id[48] )
+ *q++ = '\0'; /* Zero-pad the rest */
+ }
+
+ return 1;
+}
+
+
+void __cpuinit display_cacheinfo(struct cpuinfo_x86 *c)
+{
+ unsigned int n, dummy, ecx, edx, l2size;
+
+ n = cpuid_eax(0x80000000);
+
+ if (n >= 0x80000005) {
+ cpuid(0x80000005, &dummy, &dummy, &ecx, &edx);
+ printk(KERN_INFO "CPU: L1 I Cache: %dK (%d bytes/line), D cache %dK (%d bytes/line)\n",
+ edx>>24, edx&0xFF, ecx>>24, ecx&0xFF);
+ c->x86_cache_size=(ecx>>24)+(edx>>24);
+ }
+
+ if (n < 0x80000006) /* Some chips just has a large L1. */
+ return;
+
+ ecx = cpuid_ecx(0x80000006);
+ l2size = ecx >> 16;
+
+ /* do processor-specific cache resizing */
+ if (this_cpu->c_size_cache)
+ l2size = this_cpu->c_size_cache(c,l2size);
+
+ /* Allow user to override all this if necessary. */
+ if (cachesize_override != -1)
+ l2size = cachesize_override;
+
+ if ( l2size == 0 )
+ return; /* Again, no L2 cache is possible */
+
+ c->x86_cache_size = l2size;
+
+ printk(KERN_INFO "CPU: L2 Cache: %dK (%d bytes/line)\n",
+ l2size, ecx & 0xFF);
+}
+
+/* Naming convention should be: <Name> [(<Codename>)] */
+/* This table only is used unless init_<vendor>() below doesn't set it; */
+/* in particular, if CPUID levels 0x80000002..4 are supported, this isn't used */
+
+/* Look up CPU names by table lookup. */
+static char __cpuinit *table_lookup_model(struct cpuinfo_x86 *c)
+{
+ struct cpu_model_info *info;
+
+ if ( c->x86_model >= 16 )
+ return NULL; /* Range check */
+
+ if (!this_cpu)
+ return NULL;
+
+ info = this_cpu->c_models;
+
+ while (info && info->family) {
+ if (info->family == c->x86)
+ return info->model_names[c->x86_model];
+ info++;
+ }
+ return NULL; /* Not found */
+}
+
+
+static void __cpuinit get_cpu_vendor(struct cpuinfo_x86 *c, int early)
+{
+ char *v = c->x86_vendor_id;
+ int i;
+ static int printed;
+
+ for (i = 0; i < X86_VENDOR_NUM; i++) {
+ if (cpu_devs[i]) {
+ if (!strcmp(v,cpu_devs[i]->c_ident[0]) ||
+ (cpu_devs[i]->c_ident[1] &&
+ !strcmp(v,cpu_devs[i]->c_ident[1]))) {
+ c->x86_vendor = i;
+ if (!early)
+ this_cpu = cpu_devs[i];
+ return;
+ }
+ }
+ }
+ if (!printed) {
+ printed++;
+ printk(KERN_ERR "CPU: Vendor unknown, using generic init.\n");
+ printk(KERN_ERR "CPU: Your system may be unstable.\n");
+ }
+ c->x86_vendor = X86_VENDOR_UNKNOWN;
+ this_cpu = &default_cpu;
+}
+
+
+static int __init x86_fxsr_setup(char * s)
+{
+ disable_x86_fxsr = 1;
+ return 1;
+}
+__setup("nofxsr", x86_fxsr_setup);
+
+
+static int __init x86_sep_setup(char * s)
+{
+ disable_x86_sep = 1;
+ return 1;
+}
+__setup("nosep", x86_sep_setup);
+
+
+/* Standard macro to see if a specific flag is changeable */
+static inline int flag_is_changeable_p(u32 flag)
+{
+ u32 f1, f2;
+
+ asm("pushfl\n\t"
+ "pushfl\n\t"
+ "popl %0\n\t"
+ "movl %0,%1\n\t"
+ "xorl %2,%0\n\t"
+ "pushl %0\n\t"
+ "popfl\n\t"
+ "pushfl\n\t"
+ "popl %0\n\t"
+ "popfl\n\t"
+ : "=&r" (f1), "=&r" (f2)
+ : "ir" (flag));
+
+ return ((f1^f2) & flag) != 0;
+}
+
+
+/* Probe for the CPUID instruction */
+static int __cpuinit have_cpuid_p(void)
+{
+ return flag_is_changeable_p(X86_EFLAGS_ID);
+}
+
+/* Do minimum CPU detection early.
+ Fields really needed: vendor, cpuid_level, family, model, mask, cache alignment.
+ The others are not touched to avoid unwanted side effects.
+
+ WARNING: this function is only called on the BP. Don't add code here
+ that is supposed to run on all CPUs. */
+static void __init early_cpu_detect(void)
+{
+ struct cpuinfo_x86 *c = &boot_cpu_data;
+
+ c->x86_cache_alignment = 32;
+
+ if (!have_cpuid_p())
+ return;
+
+ /* Get vendor name */
+ cpuid(0x00000000, &c->cpuid_level,
+ (int *)&c->x86_vendor_id[0],
+ (int *)&c->x86_vendor_id[8],
+ (int *)&c->x86_vendor_id[4]);
+
+ get_cpu_vendor(c, 1);
+
+ c->x86 = 4;
+ if (c->cpuid_level >= 0x00000001) {
+ u32 junk, tfms, cap0, misc;
+ cpuid(0x00000001, &tfms, &misc, &junk, &cap0);
+ c->x86 = (tfms >> 8) & 15;
+ c->x86_model = (tfms >> 4) & 15;
+ if (c->x86 == 0xf)
+ c->x86 += (tfms >> 20) & 0xff;
+ if (c->x86 >= 0x6)
+ c->x86_model += ((tfms >> 16) & 0xF) << 4;
+ c->x86_mask = tfms & 15;
+ if (cap0 & (1<<19))
+ c->x86_cache_alignment = ((misc >> 8) & 0xff) * 8;
+ }
+}
+
+void __cpuinit generic_identify(struct cpuinfo_x86 * c)
+{
+ u32 tfms, xlvl;
+ int ebx;
+
+ if (have_cpuid_p()) {
+ /* Get vendor name */
+ cpuid(0x00000000, &c->cpuid_level,
+ (int *)&c->x86_vendor_id[0],
+ (int *)&c->x86_vendor_id[8],
+ (int *)&c->x86_vendor_id[4]);
+
+ get_cpu_vendor(c, 0);
+ /* Initialize the standard set of capabilities */
+ /* Note that the vendor-specific code below might override */
+
+ /* Intel-defined flags: level 0x00000001 */
+ if ( c->cpuid_level >= 0x00000001 ) {
+ u32 capability, excap;
+ cpuid(0x00000001, &tfms, &ebx, &excap, &capability);
+ c->x86_capability[0] = capability;
+ c->x86_capability[4] = excap;
+ c->x86 = (tfms >> 8) & 15;
+ c->x86_model = (tfms >> 4) & 15;
+ if (c->x86 == 0xf)
+ c->x86 += (tfms >> 20) & 0xff;
+ if (c->x86 >= 0x6)
+ c->x86_model += ((tfms >> 16) & 0xF) << 4;
+ c->x86_mask = tfms & 15;
+#ifndef CONFIG_XEN
+#ifdef CONFIG_X86_HT
+ c->apicid = phys_pkg_id((ebx >> 24) & 0xFF, 0);
+#else
+ c->apicid = (ebx >> 24) & 0xFF;
+#endif
+#endif
+ } else {
+ /* Have CPUID level 0 only - unheard of */
+ c->x86 = 4;
+ }
+
+ /* AMD-defined flags: level 0x80000001 */
+ xlvl = cpuid_eax(0x80000000);
+ if ( (xlvl & 0xffff0000) == 0x80000000 ) {
+ if ( xlvl >= 0x80000001 ) {
+ c->x86_capability[1] = cpuid_edx(0x80000001);
+ c->x86_capability[6] = cpuid_ecx(0x80000001);
+ }
+ if ( xlvl >= 0x80000004 )
+ get_model_name(c); /* Default name */
+ }
+ }
+
+ early_intel_workaround(c);
+
+#ifdef CONFIG_X86_HT
+ c->phys_proc_id = (cpuid_ebx(1) >> 24) & 0xff;
+#endif
+}
+
+static void __cpuinit squash_the_stupid_serial_number(struct cpuinfo_x86 *c)
+{
+ if (cpu_has(c, X86_FEATURE_PN) && disable_x86_serial_nr ) {
+ /* Disable processor serial number */
+ unsigned long lo,hi;
+ rdmsr(MSR_IA32_BBL_CR_CTL,lo,hi);
+ lo |= 0x200000;
+ wrmsr(MSR_IA32_BBL_CR_CTL,lo,hi);
+ printk(KERN_NOTICE "CPU serial number disabled.\n");
+ clear_bit(X86_FEATURE_PN, c->x86_capability);
+
+ /* Disabling the serial number may affect the cpuid level */
+ c->cpuid_level = cpuid_eax(0);
+ }
+}
+
+static int __init x86_serial_nr_setup(char *s)
+{
+ disable_x86_serial_nr = 0;
+ return 1;
+}
+__setup("serialnumber", x86_serial_nr_setup);
+
+
+
+/*
+ * This does the hard work of actually picking apart the CPU stuff...
+ */
+void __cpuinit identify_cpu(struct cpuinfo_x86 *c)
+{
+ int i;
+
+ c->loops_per_jiffy = loops_per_jiffy;
+ c->x86_cache_size = -1;
+ c->x86_vendor = X86_VENDOR_UNKNOWN;
+ c->cpuid_level = -1; /* CPUID not detected */
+ c->x86_model = c->x86_mask = 0; /* So far unknown... */
+ c->x86_vendor_id[0] = '\0'; /* Unset */
+ c->x86_model_id[0] = '\0'; /* Unset */
+ c->x86_max_cores = 1;
+ memset(&c->x86_capability, 0, sizeof c->x86_capability);
+
+ if (!have_cpuid_p()) {
+ /* First of all, decide if this is a 486 or higher */
+ /* It's a 486 if we can modify the AC flag */
+ if ( flag_is_changeable_p(X86_EFLAGS_AC) )
+ c->x86 = 4;
+ else
+ c->x86 = 3;
+ }
+
+ generic_identify(c);
+
+ printk(KERN_DEBUG "CPU: After generic identify, caps:");
+ for (i = 0; i < NCAPINTS; i++)
+ printk(" %08lx", c->x86_capability[i]);
+ printk("\n");
+
+ if (this_cpu->c_identify) {
+ this_cpu->c_identify(c);
+
+ printk(KERN_DEBUG "CPU: After vendor identify, caps:");
+ for (i = 0; i < NCAPINTS; i++)
+ printk(" %08lx", c->x86_capability[i]);
+ printk("\n");
+ }
+
+ /*
+ * Vendor-specific initialization. In this section we
+ * canonicalize the feature flags, meaning if there are
+ * features a certain CPU supports which CPUID doesn't
+ * tell us, CPUID claiming incorrect flags, or other bugs,
+ * we handle them here.
+ *
+ * At the end of this section, c->x86_capability better
+ * indicate the features this CPU genuinely supports!
+ */
+ if (this_cpu->c_init)
+ this_cpu->c_init(c);
+
+ /* Disable the PN if appropriate */
+ squash_the_stupid_serial_number(c);
+
+ /*
+ * The vendor-specific functions might have changed features. Now
+ * we do "generic changes."
+ */
+
+ /* TSC disabled? */
+ if ( tsc_disable )
+ clear_bit(X86_FEATURE_TSC, c->x86_capability);
+
+ /* FXSR disabled? */
+ if (disable_x86_fxsr) {
+ clear_bit(X86_FEATURE_FXSR, c->x86_capability);
+ clear_bit(X86_FEATURE_XMM, c->x86_capability);
+ }
+
+ /* SEP disabled? */
+ if (disable_x86_sep)
+ clear_bit(X86_FEATURE_SEP, c->x86_capability);
+
+ if (disable_pse)
+ clear_bit(X86_FEATURE_PSE, c->x86_capability);
+
+ /* If the model name is still unset, do table lookup. */
+ if ( !c->x86_model_id[0] ) {
+ char *p;
+ p = table_lookup_model(c);
+ if ( p )
+ strcpy(c->x86_model_id, p);
+ else
+ /* Last resort... */
+ sprintf(c->x86_model_id, "%02x/%02x",
+ c->x86, c->x86_model);
+ }
+
+ /* Now the feature flags better reflect actual CPU features! */
+
+ printk(KERN_DEBUG "CPU: After all inits, caps:");
+ for (i = 0; i < NCAPINTS; i++)
+ printk(" %08lx", c->x86_capability[i]);
+ printk("\n");
+
+ /*
+ * On SMP, boot_cpu_data holds the common feature set between
+ * all CPUs; so make sure that we indicate which features are
+ * common between the CPUs. The first time this routine gets
+ * executed, c == &boot_cpu_data.
+ */
+ if ( c != &boot_cpu_data ) {
+ /* AND the already accumulated flags with these */
+ for ( i = 0 ; i < NCAPINTS ; i++ )
+ boot_cpu_data.x86_capability[i] &= c->x86_capability[i];
+ }
+
+ /* Init Machine Check Exception if available. */
+ mcheck_init(c);
+
+ if (c == &boot_cpu_data)
+ sysenter_setup();
+ enable_sep_cpu();
+
+ if (c == &boot_cpu_data)
+ mtrr_bp_init();
+ else
+ mtrr_ap_init();
+}
+
+#ifdef CONFIG_X86_HT
+void __cpuinit detect_ht(struct cpuinfo_x86 *c)
+{
+ u32 eax, ebx, ecx, edx;
+ int index_msb, core_bits;
+
+ cpuid(1, &eax, &ebx, &ecx, &edx);
+
+ if (!cpu_has(c, X86_FEATURE_HT) || cpu_has(c, X86_FEATURE_CMP_LEGACY))
+ return;
+
+ smp_num_siblings = (ebx & 0xff0000) >> 16;
+
+ if (smp_num_siblings == 1) {
+ printk(KERN_INFO "CPU: Hyper-Threading is disabled\n");
+ } else if (smp_num_siblings > 1 ) {
+
+ if (smp_num_siblings > NR_CPUS) {
+ printk(KERN_WARNING "CPU: Unsupported number of the "
+ "siblings %d", smp_num_siblings);
+ smp_num_siblings = 1;
+ return;
+ }
+
+ index_msb = get_count_order(smp_num_siblings);
+ c->phys_proc_id = phys_pkg_id((ebx >> 24) & 0xFF, index_msb);
+
+ printk(KERN_INFO "CPU: Physical Processor ID: %d\n",
+ c->phys_proc_id);
+
+ smp_num_siblings = smp_num_siblings / c->x86_max_cores;
+
+ index_msb = get_count_order(smp_num_siblings) ;
+
+ core_bits = get_count_order(c->x86_max_cores);
+
+ c->cpu_core_id = phys_pkg_id((ebx >> 24) & 0xFF, index_msb) &
+ ((1 << core_bits) - 1);
+
+ if (c->x86_max_cores > 1)
+ printk(KERN_INFO "CPU: Processor Core ID: %d\n",
+ c->cpu_core_id);
+ }
+}
+#endif
+
+void __cpuinit print_cpu_info(struct cpuinfo_x86 *c)
+{
+ char *vendor = NULL;
+
+ if (c->x86_vendor < X86_VENDOR_NUM)
+ vendor = this_cpu->c_vendor;
+ else if (c->cpuid_level >= 0)
+ vendor = c->x86_vendor_id;
+
+ if (vendor && strncmp(c->x86_model_id, vendor, strlen(vendor)))
+ printk("%s ", vendor);
+
+ if (!c->x86_model_id[0])
+ printk("%d86", c->x86);
+ else
+ printk("%s", c->x86_model_id);
+
+ if (c->x86_mask || c->cpuid_level >= 0)
+ printk(" stepping %02x\n", c->x86_mask);
+ else
+ printk("\n");
+}
+
+cpumask_t cpu_initialized __cpuinitdata = CPU_MASK_NONE;
+
+/* This is hacky. :)
+ * We're emulating future behavior.
+ * In the future, the cpu-specific init functions will be called implicitly
+ * via the magic of initcalls.
+ * They will insert themselves into the cpu_devs structure.
+ * Then, when cpu_init() is called, we can just iterate over that array.
+ */
+
+extern int intel_cpu_init(void);
+extern int cyrix_init_cpu(void);
+extern int nsc_init_cpu(void);
+extern int amd_init_cpu(void);
+extern int centaur_init_cpu(void);
+extern int transmeta_init_cpu(void);
+extern int rise_init_cpu(void);
+extern int nexgen_init_cpu(void);
+extern int umc_init_cpu(void);
+
+void __init early_cpu_init(void)
+{
+ intel_cpu_init();
+ cyrix_init_cpu();
+ nsc_init_cpu();
+ amd_init_cpu();
+ centaur_init_cpu();
+ transmeta_init_cpu();
+ rise_init_cpu();
+ nexgen_init_cpu();
+ umc_init_cpu();
+ early_cpu_detect();
+
+#ifdef CONFIG_DEBUG_PAGEALLOC
+ /* pse is not compatible with on-the-fly unmapping,
+ * disable it even if the cpus claim to support it.
+ */
+ clear_bit(X86_FEATURE_PSE, boot_cpu_data.x86_capability);
+ disable_pse = 1;
+#endif
+}
+
+static void __cpuinit cpu_gdt_init(const struct Xgt_desc_struct *gdt_descr)
+{
+ unsigned long frames[16];
+ unsigned long va;
+ int f;
+
+ for (va = gdt_descr->address, f = 0;
+ va < gdt_descr->address + gdt_descr->size;
+ va += PAGE_SIZE, f++) {
+ frames[f] = virt_to_mfn(va);
+ make_lowmem_page_readonly(
+ (void *)va, XENFEAT_writable_descriptor_tables);
+ }
+ if (HYPERVISOR_set_gdt(frames, (gdt_descr->size + 1) / 8))
+ BUG();
+}
+
+/*
+ * cpu_init() initializes state that is per-CPU. Some data is already
+ * initialized (naturally) in the bootstrap process, such as the GDT
+ * and IDT. We reload them nevertheless, this function acts as a
+ * 'CPU state barrier', nothing should get across.
+ */
+void __cpuinit cpu_init(void)
+{
+ int cpu = smp_processor_id();
+#ifndef CONFIG_X86_NO_TSS
+ struct tss_struct * t = &per_cpu(init_tss, cpu);
+#endif
+ struct thread_struct *thread = &current->thread;
+ struct desc_struct *gdt;
+ struct Xgt_desc_struct *cpu_gdt_descr = &per_cpu(cpu_gdt_descr, cpu);
+
+ if (cpu_test_and_set(cpu, cpu_initialized)) {
+ printk(KERN_WARNING "CPU#%d already initialized!\n", cpu);
+ for (;;) local_irq_enable();
+ }
+ printk(KERN_INFO "Initializing CPU#%d\n", cpu);
+
+ if (cpu_has_vme || cpu_has_de)
+ clear_in_cr4(X86_CR4_VME|X86_CR4_PVI|X86_CR4_TSD|X86_CR4_DE);
+ if (tsc_disable && cpu_has_tsc) {
+ printk(KERN_NOTICE "Disabling TSC...\n");
+ /**** FIX-HPA: DOES THIS REALLY BELONG HERE? ****/
+ clear_bit(X86_FEATURE_TSC, boot_cpu_data.x86_capability);
+ set_in_cr4(X86_CR4_TSD);
+ }
+
+#ifndef CONFIG_XEN
+ /* The CPU hotplug case */
+ if (cpu_gdt_descr->address) {
+ gdt = (struct desc_struct *)cpu_gdt_descr->address;
+ memset(gdt, 0, PAGE_SIZE);
+ goto old_gdt;
+ }
+ /*
+ * This is a horrible hack to allocate the GDT. The problem
+ * is that cpu_init() is called really early for the boot CPU
+ * (and hence needs bootmem) but much later for the secondary
+ * CPUs, when bootmem will have gone away
+ */
+ if (NODE_DATA(0)->bdata->node_bootmem_map) {
+ gdt = (struct desc_struct *)alloc_bootmem_pages(PAGE_SIZE);
+ /* alloc_bootmem_pages panics on failure, so no check */
+ memset(gdt, 0, PAGE_SIZE);
+ } else {
+ gdt = (struct desc_struct *)get_zeroed_page(GFP_KERNEL);
+ if (unlikely(!gdt)) {
+ printk(KERN_CRIT "CPU%d failed to allocate GDT\n", cpu);
+ for (;;)
+ local_irq_enable();
+ }
+ }
+old_gdt:
+ /*
+ * Initialize the per-CPU GDT with the boot GDT,
+ * and set up the GDT descriptor:
+ */
+ memcpy(gdt, cpu_gdt_table, GDT_SIZE);
+
+ /* Set up GDT entry for 16bit stack */
+ *(__u64 *)(&gdt[GDT_ENTRY_ESPFIX_SS]) |=
+ ((((__u64)stk16_off) << 16) & 0x000000ffffff0000ULL) |
+ ((((__u64)stk16_off) << 32) & 0xff00000000000000ULL) |
+ (CPU_16BIT_STACK_SIZE - 1);
+
+ cpu_gdt_descr->size = GDT_SIZE - 1;
+ cpu_gdt_descr->address = (unsigned long)gdt;
+#else
+ if (cpu == 0 && cpu_gdt_descr->address == 0) {
+ gdt = (struct desc_struct *)alloc_bootmem_pages(PAGE_SIZE);
+ /* alloc_bootmem_pages panics on failure, so no check */
+ memset(gdt, 0, PAGE_SIZE);
+
+ memcpy(gdt, cpu_gdt_table, GDT_SIZE);
+
+ cpu_gdt_descr->size = GDT_SIZE;
+ cpu_gdt_descr->address = (unsigned long)gdt;
+ }
+#endif
+
+ cpu_gdt_init(cpu_gdt_descr);
+
+ /*
+ * Set up and load the per-CPU TSS and LDT
+ */
+ atomic_inc(&init_mm.mm_count);
+ current->active_mm = &init_mm;
+ if (current->mm)
+ BUG();
+ enter_lazy_tlb(&init_mm, current);
+
+ load_esp0(t, thread);
+
+ load_LDT(&init_mm.context);
+
+#ifdef CONFIG_DOUBLEFAULT
+ /* Set up doublefault TSS pointer in the GDT */
+ __set_tss_desc(cpu, GDT_ENTRY_DOUBLEFAULT_TSS, &doublefault_tss);
+#endif
+
+ /* Clear %fs and %gs. */
+ asm volatile ("xorl %eax, %eax; movl %eax, %fs; movl %eax, %gs");
+
+ /* Clear all 6 debug registers: */
+ set_debugreg(0, 0);
+ set_debugreg(0, 1);
+ set_debugreg(0, 2);
+ set_debugreg(0, 3);
+ set_debugreg(0, 6);
+ set_debugreg(0, 7);
+
+ /*
+ * Force FPU initialization:
+ */
+ current_thread_info()->status = 0;
+ clear_used_math();
+ mxcsr_feature_mask_init();
+}
+
+#ifdef CONFIG_HOTPLUG_CPU
+void __cpuinit cpu_uninit(void)
+{
+ int cpu = raw_smp_processor_id();
+ cpu_clear(cpu, cpu_initialized);
+
+ /* lazy TLB state */
+ per_cpu(cpu_tlbstate, cpu).state = 0;
+ per_cpu(cpu_tlbstate, cpu).active_mm = &init_mm;
+}
+#endif
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/kernel/cpu/mcheck/mce_dom0.c 2009-10-01 11:00:47.000000000 +0200
@@ -0,0 +1,134 @@
+#include <linux/init.h>
+#include <linux/types.h>
+#include <linux/kernel.h>
+#include <xen/interface/xen.h>
+#include <xen/evtchn.h>
+#include <xen/interface/vcpu.h>
+#include <asm/hypercall.h>
+#include <asm/mce.h>
+
+static int convert_log(struct mc_info *mi)
+{
+ struct mcinfo_common *mic = NULL;
+ struct mcinfo_global *mc_global;
+ struct mcinfo_bank *mc_bank;
+ struct mce m;
+
+ x86_mcinfo_lookup(mic, mi, MC_TYPE_GLOBAL);
+ if (mic == NULL)
+ {
+ printk(KERN_ERR "DOM0_MCE_LOG: global data is NULL\n");
+ return -1;
+ }
+
+ mc_global = (struct mcinfo_global*)mic;
+ m.mcgstatus = mc_global->mc_gstatus;
+ m.cpu = mc_global->mc_coreid;/*for test*/
+ x86_mcinfo_lookup(mic, mi, MC_TYPE_BANK);
+ do
+ {
+ if (mic == NULL || mic->size == 0)
+ break;
+ if (mic->type == MC_TYPE_BANK)
+ {
+ mc_bank = (struct mcinfo_bank*)mic;
+ m.misc = mc_bank->mc_misc;
+ m.status = mc_bank->mc_status;
+ m.addr = mc_bank->mc_addr;
+ m.tsc = mc_bank->mc_tsc;
+ m.res1 = mc_bank->mc_ctrl2;
+ m.bank = mc_bank->mc_bank;
+ printk(KERN_DEBUG "[CPU%d, BANK%d, addr %llx, state %llx]\n",
+ m.bank, m.cpu, m.addr, m.status);
+ /*log this record*/
+ mce_log(&m);
+ }
+ mic = x86_mcinfo_next(mic);
+ }while (1);
+
+ return 0;
+}
+
+static struct mc_info *g_mi;
+
+/*dom0 mce virq handler, logging physical mce error info*/
+
+static irqreturn_t mce_dom0_interrupt(int irq, void *dev_id,
+ struct pt_regs *regs)
+{
+ xen_mc_t mc_op;
+ int result = 0;
+
+ printk(KERN_DEBUG "MCE_DOM0_LOG: enter dom0 mce vIRQ handler\n");
+ mc_op.cmd = XEN_MC_fetch;
+ mc_op.interface_version = XEN_MCA_INTERFACE_VERSION;
+ set_xen_guest_handle(mc_op.u.mc_fetch.data, g_mi);
+urgent:
+ mc_op.u.mc_fetch.flags = XEN_MC_URGENT;
+ result = HYPERVISOR_mca(&mc_op);
+ if (result || mc_op.u.mc_fetch.flags & XEN_MC_NODATA ||
+ mc_op.u.mc_fetch.flags & XEN_MC_FETCHFAILED)
+ {
+ printk(KERN_DEBUG "MCE_DOM0_LOG: No more urgent data\n");
+ goto nonurgent;
+ }
+ else
+ {
+ result = convert_log(g_mi);
+ if (result) {
+ printk(KERN_ERR "MCE_DOM0_LOG: Log conversion failed\n");
+ goto end;
+ }
+ /* After fetching the telem from DOM0, we need to dec the telem's
+ * refcnt and release the entry. The telem is reserved and inc
+ * refcnt when filling the telem.
+ */
+ mc_op.u.mc_fetch.flags = XEN_MC_URGENT | XEN_MC_ACK;
+ result = HYPERVISOR_mca(&mc_op);
+
+ goto urgent;
+ }
+nonurgent:
+ mc_op.u.mc_fetch.flags = XEN_MC_NONURGENT;
+ result = HYPERVISOR_mca(&mc_op);
+ if (result || mc_op.u.mc_fetch.flags & XEN_MC_NODATA ||
+ mc_op.u.mc_fetch.flags & XEN_MC_FETCHFAILED)
+ {
+ printk(KERN_DEBUG "MCE_DOM0_LOG: No more nonurgent data\n");
+ goto end;
+ }
+ else
+ {
+ result = convert_log(g_mi);
+ if (result) {
+ printk(KERN_ERR "MCE_DOM0_LOG: Log conversion failed\n");
+ goto end;
+ }
+ /* After fetching the telem from DOM0, we need to dec the telem's
+ * refcnt and release the entry. The telem is reserved and inc
+ * refcnt when filling the telem.
+ */
+ mc_op.u.mc_fetch.flags = XEN_MC_NONURGENT | XEN_MC_ACK;
+ result = HYPERVISOR_mca(&mc_op);
+
+ goto nonurgent;
+ }
+end:
+ return IRQ_HANDLED;
+}
+
+void bind_virq_for_mce(void)
+{
+ int ret;
+
+ ret = bind_virq_to_irqhandler(VIRQ_MCA, 0,
+ mce_dom0_interrupt, 0, "mce", NULL);
+
+ g_mi = kmalloc(sizeof(struct mc_info), GFP_KERNEL);
+ if (ret < 0)
+ printk(KERN_ERR "MCE_DOM0_LOG: bind_virq for DOM0 failed\n");
+
+ /* Log the machine checks left over from the previous reset. */
+ mce_dom0_interrupt(VIRQ_MCA, NULL, NULL);
+}
+
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/kernel/cpu/mtrr/main-xen.c 2008-01-28 12:24:18.000000000 +0100
@@ -0,0 +1,198 @@
+#include <linux/init.h>
+#include <linux/proc_fs.h>
+#include <linux/ctype.h>
+#include <linux/module.h>
+#include <linux/seq_file.h>
+#include <asm/uaccess.h>
+#include <linux/mutex.h>
+
+#include <asm/mtrr.h>
+#include "mtrr.h"
+
+static DEFINE_MUTEX(mtrr_mutex);
+
+void generic_get_mtrr(unsigned int reg, unsigned long *base,
+ unsigned int *size, mtrr_type * type)
+{
+ struct xen_platform_op op;
+
+ op.cmd = XENPF_read_memtype;
+ op.u.read_memtype.reg = reg;
+ if (unlikely(HYPERVISOR_platform_op(&op)))
+ memset(&op.u.read_memtype, 0, sizeof(op.u.read_memtype));
+
+ *size = op.u.read_memtype.nr_mfns;
+ *base = op.u.read_memtype.mfn;
+ *type = op.u.read_memtype.type;
+}
+
+struct mtrr_ops generic_mtrr_ops = {
+ .use_intel_if = 1,
+ .get = generic_get_mtrr,
+};
+
+struct mtrr_ops *mtrr_if = &generic_mtrr_ops;
+unsigned int num_var_ranges;
+unsigned int *usage_table;
+
+static void __init set_num_var_ranges(void)
+{
+ struct xen_platform_op op;
+
+ for (num_var_ranges = 0; ; num_var_ranges++) {
+ op.cmd = XENPF_read_memtype;
+ op.u.read_memtype.reg = num_var_ranges;
+ if (HYPERVISOR_platform_op(&op) != 0)
+ break;
+ }
+}
+
+static void __init init_table(void)
+{
+ int i, max;
+
+ max = num_var_ranges;
+ if ((usage_table = kmalloc(max * sizeof *usage_table, GFP_KERNEL))
+ == NULL) {
+ printk(KERN_ERR "mtrr: could not allocate\n");
+ return;
+ }
+ for (i = 0; i < max; i++)
+ usage_table[i] = 0;
+}
+
+int mtrr_add_page(unsigned long base, unsigned long size,
+ unsigned int type, char increment)
+{
+ int error;
+ struct xen_platform_op op;
+
+ mutex_lock(&mtrr_mutex);
+
+ op.cmd = XENPF_add_memtype;
+ op.u.add_memtype.mfn = base;
+ op.u.add_memtype.nr_mfns = size;
+ op.u.add_memtype.type = type;
+ error = HYPERVISOR_platform_op(&op);
+ if (error) {
+ mutex_unlock(&mtrr_mutex);
+ BUG_ON(error > 0);
+ return error;
+ }
+
+ if (increment)
+ ++usage_table[op.u.add_memtype.reg];
+
+ mutex_unlock(&mtrr_mutex);
+
+ return op.u.add_memtype.reg;
+}
+
+static int mtrr_check(unsigned long base, unsigned long size)
+{
+ if ((base & (PAGE_SIZE - 1)) || (size & (PAGE_SIZE - 1))) {
+ printk(KERN_WARNING
+ "mtrr: size and base must be multiples of 4 kiB\n");
+ printk(KERN_DEBUG
+ "mtrr: size: 0x%lx base: 0x%lx\n", size, base);
+ dump_stack();
+ return -1;
+ }
+ return 0;
+}
+
+int
+mtrr_add(unsigned long base, unsigned long size, unsigned int type,
+ char increment)
+{
+ if (mtrr_check(base, size))
+ return -EINVAL;
+ return mtrr_add_page(base >> PAGE_SHIFT, size >> PAGE_SHIFT, type,
+ increment);
+}
+
+int mtrr_del_page(int reg, unsigned long base, unsigned long size)
+{
+ unsigned i;
+ mtrr_type ltype;
+ unsigned long lbase;
+ unsigned int lsize;
+ int error = -EINVAL;
+ struct xen_platform_op op;
+
+ mutex_lock(&mtrr_mutex);
+
+ if (reg < 0) {
+ /* Search for existing MTRR */
+ for (i = 0; i < num_var_ranges; ++i) {
+ mtrr_if->get(i, &lbase, &lsize, &ltype);
+ if (lbase == base && lsize == size) {
+ reg = i;
+ break;
+ }
+ }
+ if (reg < 0) {
+ printk(KERN_DEBUG "mtrr: no MTRR for %lx000,%lx000 found\n", base,
+ size);
+ goto out;
+ }
+ }
+ if (usage_table[reg] < 1) {
+ printk(KERN_WARNING "mtrr: reg: %d has count=0\n", reg);
+ goto out;
+ }
+ if (--usage_table[reg] < 1) {
+ op.cmd = XENPF_del_memtype;
+ op.u.del_memtype.handle = 0;
+ op.u.del_memtype.reg = reg;
+ error = HYPERVISOR_platform_op(&op);
+ if (error) {
+ BUG_ON(error > 0);
+ goto out;
+ }
+ }
+ error = reg;
+ out:
+ mutex_unlock(&mtrr_mutex);
+ return error;
+}
+
+int
+mtrr_del(int reg, unsigned long base, unsigned long size)
+{
+ if (mtrr_check(base, size))
+ return -EINVAL;
+ return mtrr_del_page(reg, base >> PAGE_SHIFT, size >> PAGE_SHIFT);
+}
+
+EXPORT_SYMBOL(mtrr_add);
+EXPORT_SYMBOL(mtrr_del);
+
+void __init mtrr_bp_init(void)
+{
+}
+
+void mtrr_ap_init(void)
+{
+}
+
+static int __init mtrr_init(void)
+{
+ struct cpuinfo_x86 *c = &boot_cpu_data;
+
+ if (!is_initial_xendomain())
+ return -ENODEV;
+
+ if ((!cpu_has(c, X86_FEATURE_MTRR)) &&
+ (!cpu_has(c, X86_FEATURE_K6_MTRR)) &&
+ (!cpu_has(c, X86_FEATURE_CYRIX_ARR)) &&
+ (!cpu_has(c, X86_FEATURE_CENTAUR_MCR)))
+ return -ENODEV;
+
+ set_num_var_ranges();
+ init_table();
+
+ return 0;
+}
+
+subsys_initcall(mtrr_init);
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/kernel/entry_32-xen.S 2009-05-19 09:16:41.000000000 +0200
@@ -0,0 +1,1242 @@
+/*
+ * linux/arch/i386/entry.S
+ *
+ * Copyright (C) 1991, 1992 Linus Torvalds
+ */
+
+/*
+ * entry.S contains the system-call and fault low-level handling routines.
+ * This also contains the timer-interrupt handler, as well as all interrupts
+ * and faults that can result in a task-switch.
+ *
+ * NOTE: This code handles signal-recognition, which happens every time
+ * after a timer-interrupt and after each system call.
+ *
+ * I changed all the .align's to 4 (16 byte alignment), as that's faster
+ * on a 486.
+ *
+ * Stack layout in 'ret_from_system_call':
+ * ptrace needs to have all regs on the stack.
+ * if the order here is changed, it needs to be
+ * updated in fork.c:copy_process, signal.c:do_signal,
+ * ptrace.c and ptrace.h
+ *
+ * 0(%esp) - %ebx
+ * 4(%esp) - %ecx
+ * 8(%esp) - %edx
+ * C(%esp) - %esi
+ * 10(%esp) - %edi
+ * 14(%esp) - %ebp
+ * 18(%esp) - %eax
+ * 1C(%esp) - %ds
+ * 20(%esp) - %es
+ * 24(%esp) - orig_eax
+ * 28(%esp) - %eip
+ * 2C(%esp) - %cs
+ * 30(%esp) - %eflags
+ * 34(%esp) - %oldesp
+ * 38(%esp) - %oldss
+ *
+ * "current" is in register %ebx during any slow entries.
+ */
+
+#include <linux/linkage.h>
+#include <asm/thread_info.h>
+#include <asm/irqflags.h>
+#include <asm/errno.h>
+#include <asm/segment.h>
+#include <asm/smp.h>
+#include <asm/page.h>
+#include <asm/desc.h>
+#include <asm/dwarf2.h>
+#include "irq_vectors.h"
+#include <xen/interface/xen.h>
+
+#define nr_syscalls ((syscall_table_size)/4)
+
+EBX = 0x00
+ECX = 0x04
+EDX = 0x08
+ESI = 0x0C
+EDI = 0x10
+EBP = 0x14
+EAX = 0x18
+DS = 0x1C
+ES = 0x20
+ORIG_EAX = 0x24
+EIP = 0x28
+CS = 0x2C
+EFLAGS = 0x30
+OLDESP = 0x34
+OLDSS = 0x38
+
+CF_MASK = 0x00000001
+TF_MASK = 0x00000100
+IF_MASK = 0x00000200
+DF_MASK = 0x00000400
+NT_MASK = 0x00004000
+VM_MASK = 0x00020000
+/* Pseudo-eflags. */
+NMI_MASK = 0x80000000
+
+#ifndef CONFIG_XEN
+#define DISABLE_INTERRUPTS cli
+#define ENABLE_INTERRUPTS sti
+#else
+/* Offsets into shared_info_t. */
+#define evtchn_upcall_pending /* 0 */
+#define evtchn_upcall_mask 1
+
+#define sizeof_vcpu_shift 6
+
+#ifdef CONFIG_SMP
+#define GET_VCPU_INFO movl TI_cpu(%ebp),%esi ; \
+ shl $sizeof_vcpu_shift,%esi ; \
+ addl HYPERVISOR_shared_info,%esi
+#else
+#define GET_VCPU_INFO movl HYPERVISOR_shared_info,%esi
+#endif
+
+#define __DISABLE_INTERRUPTS movb $1,evtchn_upcall_mask(%esi)
+#define __ENABLE_INTERRUPTS movb $0,evtchn_upcall_mask(%esi)
+#define DISABLE_INTERRUPTS GET_VCPU_INFO ; \
+ __DISABLE_INTERRUPTS
+#define ENABLE_INTERRUPTS GET_VCPU_INFO ; \
+ __ENABLE_INTERRUPTS
+#define __TEST_PENDING testb $0xFF,evtchn_upcall_pending(%esi)
+#endif
+
+#ifdef CONFIG_PREEMPT
+#define preempt_stop cli; TRACE_IRQS_OFF
+#else
+#define preempt_stop
+#define resume_kernel restore_nocheck
+#endif
+
+.macro TRACE_IRQS_IRET
+#ifdef CONFIG_TRACE_IRQFLAGS
+ testl $IF_MASK,EFLAGS(%esp) # interrupts off?
+ jz 1f
+ TRACE_IRQS_ON
+1:
+#endif
+.endm
+
+#ifdef CONFIG_VM86
+#define resume_userspace_sig check_userspace
+#else
+#define resume_userspace_sig resume_userspace
+#endif
+
+#define SAVE_ALL \
+ cld; \
+ pushl %es; \
+ CFI_ADJUST_CFA_OFFSET 4;\
+ /*CFI_REL_OFFSET es, 0;*/\
+ pushl %ds; \
+ CFI_ADJUST_CFA_OFFSET 4;\
+ /*CFI_REL_OFFSET ds, 0;*/\
+ pushl %eax; \
+ CFI_ADJUST_CFA_OFFSET 4;\
+ CFI_REL_OFFSET eax, 0;\
+ pushl %ebp; \
+ CFI_ADJUST_CFA_OFFSET 4;\
+ CFI_REL_OFFSET ebp, 0;\
+ pushl %edi; \
+ CFI_ADJUST_CFA_OFFSET 4;\
+ CFI_REL_OFFSET edi, 0;\
+ pushl %esi; \
+ CFI_ADJUST_CFA_OFFSET 4;\
+ CFI_REL_OFFSET esi, 0;\
+ pushl %edx; \
+ CFI_ADJUST_CFA_OFFSET 4;\
+ CFI_REL_OFFSET edx, 0;\
+ pushl %ecx; \
+ CFI_ADJUST_CFA_OFFSET 4;\
+ CFI_REL_OFFSET ecx, 0;\
+ pushl %ebx; \
+ CFI_ADJUST_CFA_OFFSET 4;\
+ CFI_REL_OFFSET ebx, 0;\
+ movl $(__USER_DS), %edx; \
+ movl %edx, %ds; \
+ movl %edx, %es;
+
+#define RESTORE_INT_REGS \
+ popl %ebx; \
+ CFI_ADJUST_CFA_OFFSET -4;\
+ CFI_RESTORE ebx;\
+ popl %ecx; \
+ CFI_ADJUST_CFA_OFFSET -4;\
+ CFI_RESTORE ecx;\
+ popl %edx; \
+ CFI_ADJUST_CFA_OFFSET -4;\
+ CFI_RESTORE edx;\
+ popl %esi; \
+ CFI_ADJUST_CFA_OFFSET -4;\
+ CFI_RESTORE esi;\
+ popl %edi; \
+ CFI_ADJUST_CFA_OFFSET -4;\
+ CFI_RESTORE edi;\
+ popl %ebp; \
+ CFI_ADJUST_CFA_OFFSET -4;\
+ CFI_RESTORE ebp;\
+ popl %eax; \
+ CFI_ADJUST_CFA_OFFSET -4;\
+ CFI_RESTORE eax
+
+#define RESTORE_REGS \
+ RESTORE_INT_REGS; \
+1: popl %ds; \
+ CFI_ADJUST_CFA_OFFSET -4;\
+ /*CFI_RESTORE ds;*/\
+2: popl %es; \
+ CFI_ADJUST_CFA_OFFSET -4;\
+ /*CFI_RESTORE es;*/\
+.section .fixup,"ax"; \
+3: movl $0,(%esp); \
+ jmp 1b; \
+4: movl $0,(%esp); \
+ jmp 2b; \
+.previous; \
+.section __ex_table,"a";\
+ .align 4; \
+ .long 1b,3b; \
+ .long 2b,4b; \
+.previous
+
+#define RING0_INT_FRAME \
+ CFI_STARTPROC simple;\
+ CFI_DEF_CFA esp, 3*4;\
+ /*CFI_OFFSET cs, -2*4;*/\
+ CFI_OFFSET eip, -3*4
+
+#define RING0_EC_FRAME \
+ CFI_STARTPROC simple;\
+ CFI_DEF_CFA esp, 4*4;\
+ /*CFI_OFFSET cs, -2*4;*/\
+ CFI_OFFSET eip, -3*4
+
+#define RING0_PTREGS_FRAME \
+ CFI_STARTPROC simple;\
+ CFI_DEF_CFA esp, OLDESP-EBX;\
+ /*CFI_OFFSET cs, CS-OLDESP;*/\
+ CFI_OFFSET eip, EIP-OLDESP;\
+ /*CFI_OFFSET es, ES-OLDESP;*/\
+ /*CFI_OFFSET ds, DS-OLDESP;*/\
+ CFI_OFFSET eax, EAX-OLDESP;\
+ CFI_OFFSET ebp, EBP-OLDESP;\
+ CFI_OFFSET edi, EDI-OLDESP;\
+ CFI_OFFSET esi, ESI-OLDESP;\
+ CFI_OFFSET edx, EDX-OLDESP;\
+ CFI_OFFSET ecx, ECX-OLDESP;\
+ CFI_OFFSET ebx, EBX-OLDESP
+
+ENTRY(ret_from_fork)
+ CFI_STARTPROC
+ pushl %eax
+ CFI_ADJUST_CFA_OFFSET 4
+ call schedule_tail
+ GET_THREAD_INFO(%ebp)
+ popl %eax
+ CFI_ADJUST_CFA_OFFSET -4
+ pushl $0x0202 # Reset kernel eflags
+ CFI_ADJUST_CFA_OFFSET 4
+ popfl
+ CFI_ADJUST_CFA_OFFSET -4
+ jmp syscall_exit
+ CFI_ENDPROC
+
+/*
+ * Return to user mode is not as complex as all this looks,
+ * but we want the default path for a system call return to
+ * go as quickly as possible which is why some of this is
+ * less clear than it otherwise should be.
+ */
+
+ # userspace resumption stub bypassing syscall exit tracing
+ ALIGN
+ RING0_PTREGS_FRAME
+ret_from_exception:
+ preempt_stop
+ret_from_intr:
+ GET_THREAD_INFO(%ebp)
+check_userspace:
+ movl EFLAGS(%esp), %eax # mix EFLAGS and CS
+ movb CS(%esp), %al
+ testl $(VM_MASK | 2), %eax
+ jz resume_kernel
+ENTRY(resume_userspace)
+ DISABLE_INTERRUPTS # make sure we don't miss an interrupt
+ # setting need_resched or sigpending
+ # between sampling and the iret
+ movl TI_flags(%ebp), %ecx
+ andl $_TIF_WORK_MASK, %ecx # is there any work to be done on
+ # int/exception return?
+ jne work_pending
+ jmp restore_all
+
+#ifdef CONFIG_PREEMPT
+ENTRY(resume_kernel)
+ cli
+ cmpl $0,TI_preempt_count(%ebp) # non-zero preempt_count ?
+ jnz restore_nocheck
+need_resched:
+ movl TI_flags(%ebp), %ecx # need_resched set ?
+ testb $_TIF_NEED_RESCHED, %cl
+ jz restore_all
+ testl $IF_MASK,EFLAGS(%esp) # interrupts off (exception path) ?
+ jz restore_all
+ call preempt_schedule_irq
+ jmp need_resched
+#endif
+ CFI_ENDPROC
+
+/* SYSENTER_RETURN points to after the "sysenter" instruction in
+ the vsyscall page. See vsyscall-sysentry.S, which defines the symbol. */
+
+ # sysenter call handler stub
+ENTRY(sysenter_entry)
+ CFI_STARTPROC simple
+ CFI_DEF_CFA esp, 0
+ CFI_REGISTER esp, ebp
+ movl SYSENTER_stack_esp0(%esp),%esp
+sysenter_past_esp:
+ /*
+ * No need to follow this irqs on/off section: the syscall
+ * disabled irqs and here we enable it straight after entry:
+ */
+ sti
+ pushl $(__USER_DS)
+ CFI_ADJUST_CFA_OFFSET 4
+ /*CFI_REL_OFFSET ss, 0*/
+ pushl %ebp
+ CFI_ADJUST_CFA_OFFSET 4
+ CFI_REL_OFFSET esp, 0
+ pushfl
+ CFI_ADJUST_CFA_OFFSET 4
+ pushl $(__USER_CS)
+ CFI_ADJUST_CFA_OFFSET 4
+ /*CFI_REL_OFFSET cs, 0*/
+ /*
+ * Push current_thread_info()->sysenter_return to the stack.
+ * A tiny bit of offset fixup is necessary - 4*4 means the 4 words
+ * pushed above; +8 corresponds to copy_thread's esp0 setting.
+ */
+ pushl (TI_sysenter_return-THREAD_SIZE+8+4*4)(%esp)
+ CFI_ADJUST_CFA_OFFSET 4
+ CFI_REL_OFFSET eip, 0
+
+/*
+ * Load the potential sixth argument from user stack.
+ * Careful about security.
+ */
+ cmpl $__PAGE_OFFSET-3,%ebp
+ jae syscall_fault
+1: movl (%ebp),%ebp
+.section __ex_table,"a"
+ .align 4
+ .long 1b,syscall_fault
+.previous
+
+ pushl %eax
+ CFI_ADJUST_CFA_OFFSET 4
+ SAVE_ALL
+ GET_THREAD_INFO(%ebp)
+
+ /* Note, _TIF_SECCOMP is bit number 8, and so it needs testw and not testb */
+ testw $(_TIF_SYSCALL_EMU|_TIF_SYSCALL_TRACE|_TIF_SECCOMP|_TIF_SYSCALL_AUDIT),TI_flags(%ebp)
+ jnz syscall_trace_entry
+ cmpl $(nr_syscalls), %eax
+ jae syscall_badsys
+ call *sys_call_table(,%eax,4)
+ movl %eax,EAX(%esp)
+ DISABLE_INTERRUPTS
+ TRACE_IRQS_OFF
+ movl TI_flags(%ebp), %ecx
+ testw $_TIF_ALLWORK_MASK, %cx
+ jne syscall_exit_work
+/* if something modifies registers it must also disable sysexit */
+ movl EIP(%esp), %edx
+ movl OLDESP(%esp), %ecx
+ xorl %ebp,%ebp
+#ifdef CONFIG_XEN
+ TRACE_IRQS_ON
+ __ENABLE_INTERRUPTS
+sysexit_scrit: /**** START OF SYSEXIT CRITICAL REGION ****/
+ __TEST_PENDING
+ jnz 14f # process more events if necessary...
+ movl ESI(%esp), %esi
+ sysexit
+14: __DISABLE_INTERRUPTS
+ TRACE_IRQS_OFF
+sysexit_ecrit: /**** END OF SYSEXIT CRITICAL REGION ****/
+ push %esp
+ call evtchn_do_upcall
+ add $4,%esp
+ jmp ret_from_intr
+#else
+ TRACE_IRQS_ON
+ sti
+ sysexit
+#endif /* !CONFIG_XEN */
+ CFI_ENDPROC
+
+ # pv sysenter call handler stub
+ENTRY(sysenter_entry_pv)
+ RING0_INT_FRAME
+ movl $__USER_DS,16(%esp)
+ movl %ebp,12(%esp)
+ movl $__USER_CS,4(%esp)
+ addl $4,%esp
+ CFI_ADJUST_CFA_OFFSET -4
+ /* +5*4 is SS:ESP,EFLAGS,CS:EIP. +8 is esp0 setting. */
+ pushl (TI_sysenter_return-THREAD_SIZE+8+4*4)(%esp)
+ CFI_ADJUST_CFA_OFFSET 4
+/*
+ * Load the potential sixth argument from user stack.
+ * Careful about security.
+ */
+ cmpl $__PAGE_OFFSET-3,%ebp
+ jae syscall_fault
+1: movl (%ebp),%ebp
+.section __ex_table,"a"
+ .align 4
+ .long 1b,syscall_fault
+.previous
+ /* fall through */
+ CFI_ENDPROC
+ENDPROC(sysenter_entry_pv)
+
+ # system call handler stub
+ENTRY(system_call)
+ RING0_INT_FRAME # can't unwind into user space anyway
+ pushl %eax # save orig_eax
+ CFI_ADJUST_CFA_OFFSET 4
+ SAVE_ALL
+ GET_THREAD_INFO(%ebp)
+ testl $TF_MASK,EFLAGS(%esp)
+ jz no_singlestep
+ orl $_TIF_SINGLESTEP,TI_flags(%ebp)
+no_singlestep:
+ # system call tracing in operation / emulation
+ /* Note, _TIF_SECCOMP is bit number 8, and so it needs testw and not testb */
+ testw $(_TIF_SYSCALL_EMU|_TIF_SYSCALL_TRACE|_TIF_SECCOMP|_TIF_SYSCALL_AUDIT),TI_flags(%ebp)
+ jnz syscall_trace_entry
+ cmpl $(nr_syscalls), %eax
+ jae syscall_badsys
+syscall_call:
+ call *sys_call_table(,%eax,4)
+ movl %eax,EAX(%esp) # store the return value
+syscall_exit:
+ DISABLE_INTERRUPTS # make sure we don't miss an interrupt
+ # setting need_resched or sigpending
+ # between sampling and the iret
+ TRACE_IRQS_OFF
+ movl TI_flags(%ebp), %ecx
+ testw $_TIF_ALLWORK_MASK, %cx # current->work
+ jne syscall_exit_work
+
+restore_all:
+#ifndef CONFIG_XEN
+ movl EFLAGS(%esp), %eax # mix EFLAGS, SS and CS
+ # Warning: OLDSS(%esp) contains the wrong/random values if we
+ # are returning to the kernel.
+ # See comments in process.c:copy_thread() for details.
+ movb OLDSS(%esp), %ah
+ movb CS(%esp), %al
+ andl $(VM_MASK | (4 << 8) | 3), %eax
+ cmpl $((4 << 8) | 3), %eax
+ CFI_REMEMBER_STATE
+ je ldt_ss # returning to user-space with LDT SS
+restore_nocheck:
+#else
+restore_nocheck:
+ movl EFLAGS(%esp), %eax
+ testl $(VM_MASK|NMI_MASK), %eax
+ CFI_REMEMBER_STATE
+ jnz hypervisor_iret
+ shr $9, %eax # EAX[0] == IRET_EFLAGS.IF
+ GET_VCPU_INFO
+ andb evtchn_upcall_mask(%esi),%al
+ andb $1,%al # EAX[0] == IRET_EFLAGS.IF & event_mask
+ CFI_REMEMBER_STATE
+ jnz restore_all_enable_events # != 0 => enable event delivery
+#endif
+ TRACE_IRQS_IRET
+restore_nocheck_notrace:
+ RESTORE_REGS
+ addl $4, %esp
+ CFI_ADJUST_CFA_OFFSET -4
+1: iret
+.section .fixup,"ax"
+iret_exc:
+#ifndef CONFIG_XEN
+ TRACE_IRQS_ON
+ sti
+#endif
+ pushl $0 # no error code
+ pushl $do_iret_error
+ jmp error_code
+.previous
+.section __ex_table,"a"
+ .align 4
+ .long 1b,iret_exc
+.previous
+
+ CFI_RESTORE_STATE
+#ifndef CONFIG_XEN
+ldt_ss:
+ larl OLDSS(%esp), %eax
+ jnz restore_nocheck
+ testl $0x00400000, %eax # returning to 32bit stack?
+ jnz restore_nocheck # allright, normal return
+ /* If returning to userspace with 16bit stack,
+ * try to fix the higher word of ESP, as the CPU
+ * won't restore it.
+ * This is an "official" bug of all the x86-compatible
+ * CPUs, which we can try to work around to make
+ * dosemu and wine happy. */
+ subl $8, %esp # reserve space for switch16 pointer
+ CFI_ADJUST_CFA_OFFSET 8
+ cli
+ TRACE_IRQS_OFF
+ movl %esp, %eax
+ /* Set up the 16bit stack frame with switch32 pointer on top,
+ * and a switch16 pointer on top of the current frame. */
+ call setup_x86_bogus_stack
+ CFI_ADJUST_CFA_OFFSET -8 # frame has moved
+ TRACE_IRQS_IRET
+ RESTORE_REGS
+ lss 20+4(%esp), %esp # switch to 16bit stack
+1: iret
+.section __ex_table,"a"
+ .align 4
+ .long 1b,iret_exc
+.previous
+#else
+ ALIGN
+restore_all_enable_events:
+ TRACE_IRQS_ON
+ __ENABLE_INTERRUPTS
+scrit: /**** START OF CRITICAL REGION ****/
+ __TEST_PENDING
+ jnz 14f # process more events if necessary...
+ RESTORE_REGS
+ addl $4, %esp
+ CFI_ADJUST_CFA_OFFSET -4
+1: iret
+.section __ex_table,"a"
+ .align 4
+ .long 1b,iret_exc
+.previous
+14: __DISABLE_INTERRUPTS
+ TRACE_IRQS_OFF
+ecrit: /**** END OF CRITICAL REGION ****/
+ jmp .Ldo_upcall
+
+ CFI_RESTORE_STATE
+hypervisor_iret:
+ andl $~NMI_MASK, EFLAGS(%esp)
+ RESTORE_REGS
+ addl $4, %esp
+ CFI_ADJUST_CFA_OFFSET -4
+ jmp hypercall_page + (__HYPERVISOR_iret * 32)
+#endif
+ CFI_ENDPROC
+
+ # perform work that needs to be done immediately before resumption
+ ALIGN
+ RING0_PTREGS_FRAME # can't unwind into user space anyway
+work_pending:
+ testb $_TIF_NEED_RESCHED, %cl
+ jz work_notifysig
+work_resched:
+ call schedule
+ DISABLE_INTERRUPTS # make sure we don't miss an interrupt
+ # setting need_resched or sigpending
+ # between sampling and the iret
+ TRACE_IRQS_OFF
+ movl TI_flags(%ebp), %ecx
+ andl $_TIF_WORK_MASK, %ecx # is there any work to be done other
+ # than syscall tracing?
+ jz restore_all
+ testb $_TIF_NEED_RESCHED, %cl
+ jnz work_resched
+
+work_notifysig: # deal with pending signals and
+ # notify-resume requests
+ testl $VM_MASK, EFLAGS(%esp)
+ movl %esp, %eax
+ jne work_notifysig_v86 # returning to kernel-space or
+ # vm86-space
+ xorl %edx, %edx
+ call do_notify_resume
+ jmp resume_userspace_sig
+
+ ALIGN
+work_notifysig_v86:
+#ifdef CONFIG_VM86
+ pushl %ecx # save ti_flags for do_notify_resume
+ CFI_ADJUST_CFA_OFFSET 4
+ call save_v86_state # %eax contains pt_regs pointer
+ popl %ecx
+ CFI_ADJUST_CFA_OFFSET -4
+ movl %eax, %esp
+ xorl %edx, %edx
+ call do_notify_resume
+ jmp resume_userspace_sig
+#endif
+
+ # perform syscall exit tracing
+ ALIGN
+syscall_trace_entry:
+ movl $-ENOSYS,EAX(%esp)
+ movl %esp, %eax
+ xorl %edx,%edx
+ call do_syscall_trace
+ cmpl $0, %eax
+ jne resume_userspace # ret != 0 -> running under PTRACE_SYSEMU,
+ # so must skip actual syscall
+ movl ORIG_EAX(%esp), %eax
+ cmpl $(nr_syscalls), %eax
+ jnae syscall_call
+ jmp syscall_exit
+
+ # perform syscall exit tracing
+ ALIGN
+syscall_exit_work:
+ testb $(_TIF_SYSCALL_TRACE|_TIF_SYSCALL_AUDIT|_TIF_SINGLESTEP), %cl
+ jz work_pending
+ TRACE_IRQS_ON
+ ENABLE_INTERRUPTS # could let do_syscall_trace() call
+ # schedule() instead
+ movl %esp, %eax
+ movl $1, %edx
+ call do_syscall_trace
+ jmp resume_userspace
+ CFI_ENDPROC
+
+ RING0_INT_FRAME # can't unwind into user space anyway
+syscall_fault:
+ pushl %eax # save orig_eax
+ CFI_ADJUST_CFA_OFFSET 4
+ SAVE_ALL
+ GET_THREAD_INFO(%ebp)
+ movl $-EFAULT,EAX(%esp)
+ jmp resume_userspace
+
+syscall_badsys:
+ movl $-ENOSYS,EAX(%esp)
+ jmp resume_userspace
+ CFI_ENDPROC
+
+#ifndef CONFIG_XEN
+#define FIXUP_ESPFIX_STACK \
+ movl %esp, %eax; \
+ /* switch to 32bit stack using the pointer on top of 16bit stack */ \
+ lss %ss:CPU_16BIT_STACK_SIZE-8, %esp; \
+ /* copy data from 16bit stack to 32bit stack */ \
+ call fixup_x86_bogus_stack; \
+ /* put ESP to the proper location */ \
+ movl %eax, %esp;
+#define UNWIND_ESPFIX_STACK \
+ pushl %eax; \
+ CFI_ADJUST_CFA_OFFSET 4; \
+ movl %ss, %eax; \
+ /* see if on 16bit stack */ \
+ cmpw $__ESPFIX_SS, %ax; \
+ je 28f; \
+27: popl %eax; \
+ CFI_ADJUST_CFA_OFFSET -4; \
+.section .fixup,"ax"; \
+28: movl $__KERNEL_DS, %eax; \
+ movl %eax, %ds; \
+ movl %eax, %es; \
+ /* switch to 32bit stack */ \
+ FIXUP_ESPFIX_STACK; \
+ jmp 27b; \
+.previous
+
+/*
+ * Build the entry stubs and pointer table with
+ * some assembler magic.
+ */
+.data
+ENTRY(interrupt)
+.text
+
+vector=0
+ENTRY(irq_entries_start)
+ RING0_INT_FRAME
+.rept NR_IRQS
+ ALIGN
+ .if vector
+ CFI_ADJUST_CFA_OFFSET -4
+ .endif
+1: pushl $~(vector)
+ CFI_ADJUST_CFA_OFFSET 4
+ jmp common_interrupt
+.data
+ .long 1b
+.text
+vector=vector+1
+.endr
+
+/*
+ * the CPU automatically disables interrupts when executing an IRQ vector,
+ * so IRQ-flags tracing has to follow that:
+ */
+ ALIGN
+common_interrupt:
+ SAVE_ALL
+ TRACE_IRQS_OFF
+ movl %esp,%eax
+ call do_IRQ
+ jmp ret_from_intr
+ CFI_ENDPROC
+
+#define BUILD_INTERRUPT(name, nr) \
+ENTRY(name) \
+ RING0_INT_FRAME; \
+ pushl $~(nr); \
+ CFI_ADJUST_CFA_OFFSET 4; \
+ SAVE_ALL; \
+ TRACE_IRQS_OFF \
+ movl %esp,%eax; \
+ call smp_/**/name; \
+ jmp ret_from_intr; \
+ CFI_ENDPROC
+
+/* The include is where all of the SMP etc. interrupts come from */
+#include "entry_arch.h"
+#else
+#define UNWIND_ESPFIX_STACK
+#endif
+
+ENTRY(divide_error)
+ RING0_INT_FRAME
+ pushl $0 # no error code
+ CFI_ADJUST_CFA_OFFSET 4
+ pushl $do_divide_error
+ CFI_ADJUST_CFA_OFFSET 4
+ ALIGN
+error_code:
+ pushl %ds
+ CFI_ADJUST_CFA_OFFSET 4
+ /*CFI_REL_OFFSET ds, 0*/
+ pushl %eax
+ CFI_ADJUST_CFA_OFFSET 4
+ CFI_REL_OFFSET eax, 0
+ xorl %eax, %eax
+ pushl %ebp
+ CFI_ADJUST_CFA_OFFSET 4
+ CFI_REL_OFFSET ebp, 0
+ pushl %edi
+ CFI_ADJUST_CFA_OFFSET 4
+ CFI_REL_OFFSET edi, 0
+ pushl %esi
+ CFI_ADJUST_CFA_OFFSET 4
+ CFI_REL_OFFSET esi, 0
+ pushl %edx
+ CFI_ADJUST_CFA_OFFSET 4
+ CFI_REL_OFFSET edx, 0
+ decl %eax # eax = -1
+ pushl %ecx
+ CFI_ADJUST_CFA_OFFSET 4
+ CFI_REL_OFFSET ecx, 0
+ pushl %ebx
+ CFI_ADJUST_CFA_OFFSET 4
+ CFI_REL_OFFSET ebx, 0
+ cld
+ pushl %es
+ CFI_ADJUST_CFA_OFFSET 4
+ /*CFI_REL_OFFSET es, 0*/
+ UNWIND_ESPFIX_STACK
+ popl %ecx
+ CFI_ADJUST_CFA_OFFSET -4
+ /*CFI_REGISTER es, ecx*/
+ movl ES(%esp), %edi # get the function address
+ movl ORIG_EAX(%esp), %edx # get the error code
+ movl %eax, ORIG_EAX(%esp)
+ movl %ecx, ES(%esp)
+ /*CFI_REL_OFFSET es, ES*/
+ movl $(__USER_DS), %ecx
+ movl %ecx, %ds
+ movl %ecx, %es
+ movl %esp,%eax # pt_regs pointer
+ call *%edi
+ jmp ret_from_exception
+ CFI_ENDPROC
+
+#ifdef CONFIG_XEN
+# A note on the "critical region" in our callback handler.
+# We want to avoid stacking callback handlers due to events occurring
+# during handling of the last event. To do this, we keep events disabled
+# until we've done all processing. HOWEVER, we must enable events before
+# popping the stack frame (can't be done atomically) and so it would still
+# be possible to get enough handler activations to overflow the stack.
+# Although unlikely, bugs of that kind are hard to track down, so we'd
+# like to avoid the possibility.
+# So, on entry to the handler we detect whether we interrupted an
+# existing activation in its critical region -- if so, we pop the current
+# activation and restart the handler using the previous one.
+#
+# The sysexit critical region is slightly different. sysexit
+# atomically removes the entire stack frame. If we interrupt in the
+# critical region we know that the entire frame is present and correct
+# so we can simply throw away the new one.
+ENTRY(hypervisor_callback)
+ RING0_INT_FRAME
+ pushl %eax
+ CFI_ADJUST_CFA_OFFSET 4
+ SAVE_ALL
+ testb $2,CS(%esp)
+ movl EIP(%esp),%eax
+ jnz .Ldo_upcall
+ cmpl $scrit,%eax
+ jb 0f
+ cmpl $ecrit,%eax
+ jb critical_region_fixup
+0:
+#ifdef CONFIG_XEN_SUPERVISOR_MODE_KERNEL
+ cmpl $sysexit_scrit,%eax
+ jb .Ldo_upcall
+ cmpl $sysexit_ecrit,%eax
+ ja .Ldo_upcall
+ addl $OLDESP,%esp # Remove eflags...ebx from stack frame.
+#endif
+.Ldo_upcall:
+ push %esp
+ CFI_ADJUST_CFA_OFFSET 4
+ call evtchn_do_upcall
+ add $4,%esp
+ CFI_ADJUST_CFA_OFFSET -4
+ jmp ret_from_intr
+ CFI_ENDPROC
+
+# [How we do the fixup]. We want to merge the current stack frame with the
+# just-interrupted frame. How we do this depends on where in the critical
+# region the interrupted handler was executing, and so how many saved
+# registers are in each frame. We do this quickly using the lookup table
+# 'critical_fixup_table'. For each byte offset in the critical region, it
+# provides the number of bytes which have already been popped from the
+# interrupted stack frame.
+critical_region_fixup:
+ movsbl critical_fixup_table-scrit(%eax),%ecx # %ecx contains num slots popped
+ testl %ecx,%ecx
+ leal (%esp,%ecx,4),%esi # %esi points at end of src region
+ leal OLDESP(%esp),%edi # %edi points at end of dst region
+ jle 17f # skip loop if nothing to copy
+16: subl $4,%esi # pre-decrementing copy loop
+ subl $4,%edi
+ movl (%esi),%eax
+ movl %eax,(%edi)
+ loop 16b
+17: movl %edi,%esp # final %edi is top of merged stack
+ jmp .Ldo_upcall
+
+.section .rodata,"a"
+critical_fixup_table:
+ .byte -1,-1,-1 # testb $0xff,(%esi) = __TEST_PENDING
+ .byte -1,-1 # jnz 14f
+ .byte 0 # pop %ebx
+ .byte 1 # pop %ecx
+ .byte 2 # pop %edx
+ .byte 3 # pop %esi
+ .byte 4 # pop %edi
+ .byte 5 # pop %ebp
+ .byte 6 # pop %eax
+ .byte 7 # pop %ds
+ .byte 8 # pop %es
+ .byte 9,9,9 # add $4,%esp
+ .byte 10 # iret
+ .byte -1,-1,-1,-1 # movb $1,1(%esi) = __DISABLE_INTERRUPTS
+.previous
+
+# Hypervisor uses this for application faults while it executes.
+# We get here for two reasons:
+# 1. Fault while reloading DS, ES, FS or GS
+# 2. Fault while executing IRET
+# Category 1 we fix up by reattempting the load, and zeroing the segment
+# register if the load fails.
+# Category 2 we fix up by jumping to do_iret_error. We cannot use the
+# normal Linux return path in this case because if we use the IRET hypercall
+# to pop the stack frame we end up in an infinite loop of failsafe callbacks.
+# We distinguish between categories by maintaining a status value in EAX.
+ENTRY(failsafe_callback)
+ pushl %eax
+ movl $1,%eax
+1: mov 4(%esp),%ds
+2: mov 8(%esp),%es
+3: mov 12(%esp),%fs
+4: mov 16(%esp),%gs
+ testl %eax,%eax
+ popl %eax
+ jz 5f
+ addl $16,%esp # EAX != 0 => Category 2 (Bad IRET)
+ jmp iret_exc
+5: addl $16,%esp # EAX == 0 => Category 1 (Bad segment)
+ RING0_INT_FRAME
+ pushl $0
+ SAVE_ALL
+ jmp ret_from_exception
+.section .fixup,"ax"; \
+6: xorl %eax,%eax; \
+ movl %eax,4(%esp); \
+ jmp 1b; \
+7: xorl %eax,%eax; \
+ movl %eax,8(%esp); \
+ jmp 2b; \
+8: xorl %eax,%eax; \
+ movl %eax,12(%esp); \
+ jmp 3b; \
+9: xorl %eax,%eax; \
+ movl %eax,16(%esp); \
+ jmp 4b; \
+.previous; \
+.section __ex_table,"a"; \
+ .align 4; \
+ .long 1b,6b; \
+ .long 2b,7b; \
+ .long 3b,8b; \
+ .long 4b,9b; \
+.previous
+#endif
+ CFI_ENDPROC
+
+ENTRY(coprocessor_error)
+ RING0_INT_FRAME
+ pushl $0
+ CFI_ADJUST_CFA_OFFSET 4
+ pushl $do_coprocessor_error
+ CFI_ADJUST_CFA_OFFSET 4
+ jmp error_code
+ CFI_ENDPROC
+
+ENTRY(simd_coprocessor_error)
+ RING0_INT_FRAME
+ pushl $0
+ CFI_ADJUST_CFA_OFFSET 4
+ pushl $do_simd_coprocessor_error
+ CFI_ADJUST_CFA_OFFSET 4
+ jmp error_code
+ CFI_ENDPROC
+
+ENTRY(device_not_available)
+ RING0_INT_FRAME
+ pushl $-1 # mark this as an int
+ CFI_ADJUST_CFA_OFFSET 4
+ SAVE_ALL
+#ifndef CONFIG_XEN
+ movl %cr0, %eax
+ testl $0x4, %eax # EM (math emulation bit)
+ je device_available_emulate
+ pushl $0 # temporary storage for ORIG_EIP
+ CFI_ADJUST_CFA_OFFSET 4
+ call math_emulate
+ addl $4, %esp
+ CFI_ADJUST_CFA_OFFSET -4
+ jmp ret_from_exception
+device_available_emulate:
+#endif
+ preempt_stop
+ call math_state_restore
+ jmp ret_from_exception
+ CFI_ENDPROC
+
+#ifndef CONFIG_XEN
+/*
+ * Debug traps and NMI can happen at the one SYSENTER instruction
+ * that sets up the real kernel stack. Check here, since we can't
+ * allow the wrong stack to be used.
+ *
+ * "SYSENTER_stack_esp0+12" is because the NMI/debug handler will have
+ * already pushed 3 words if it hits on the sysenter instruction:
+ * eflags, cs and eip.
+ *
+ * We just load the right stack, and push the three (known) values
+ * by hand onto the new stack - while updating the return eip past
+ * the instruction that would have done it for sysenter.
+ */
+#define FIX_STACK(offset, ok, label) \
+ cmpw $__KERNEL_CS,4(%esp); \
+ jne ok; \
+label: \
+ movl SYSENTER_stack_esp0+offset(%esp),%esp; \
+ pushfl; \
+ pushl $__KERNEL_CS; \
+ pushl $sysenter_past_esp
+#endif /* CONFIG_XEN */
+
+KPROBE_ENTRY(debug)
+ RING0_INT_FRAME
+#ifndef CONFIG_XEN
+ cmpl $sysenter_entry,(%esp)
+ jne debug_stack_correct
+ FIX_STACK(12, debug_stack_correct, debug_esp_fix_insn)
+debug_stack_correct:
+#endif /* !CONFIG_XEN */
+ pushl $-1 # mark this as an int
+ CFI_ADJUST_CFA_OFFSET 4
+ SAVE_ALL
+ xorl %edx,%edx # error code 0
+ movl %esp,%eax # pt_regs pointer
+ call do_debug
+ jmp ret_from_exception
+ CFI_ENDPROC
+ .previous .text
+#ifndef CONFIG_XEN
+/*
+ * NMI is doubly nasty. It can happen _while_ we're handling
+ * a debug fault, and the debug fault hasn't yet been able to
+ * clear up the stack. So we first check whether we got an
+ * NMI on the sysenter entry path, but after that we need to
+ * check whether we got an NMI on the debug path where the debug
+ * fault happened on the sysenter path.
+ */
+ENTRY(nmi)
+ RING0_INT_FRAME
+ pushl %eax
+ CFI_ADJUST_CFA_OFFSET 4
+ movl %ss, %eax
+ cmpw $__ESPFIX_SS, %ax
+ popl %eax
+ CFI_ADJUST_CFA_OFFSET -4
+ je nmi_16bit_stack
+ cmpl $sysenter_entry,(%esp)
+ je nmi_stack_fixup
+ pushl %eax
+ CFI_ADJUST_CFA_OFFSET 4
+ movl %esp,%eax
+ /* Do not access memory above the end of our stack page,
+ * it might not exist.
+ */
+ andl $(THREAD_SIZE-1),%eax
+ cmpl $(THREAD_SIZE-20),%eax
+ popl %eax
+ CFI_ADJUST_CFA_OFFSET -4
+ jae nmi_stack_correct
+ cmpl $sysenter_entry,12(%esp)
+ je nmi_debug_stack_check
+nmi_stack_correct:
+ pushl %eax
+ CFI_ADJUST_CFA_OFFSET 4
+ SAVE_ALL
+ xorl %edx,%edx # zero error code
+ movl %esp,%eax # pt_regs pointer
+ call do_nmi
+ jmp restore_nocheck_notrace
+ CFI_ENDPROC
+
+nmi_stack_fixup:
+ FIX_STACK(12,nmi_stack_correct, 1)
+ jmp nmi_stack_correct
+nmi_debug_stack_check:
+ cmpw $__KERNEL_CS,16(%esp)
+ jne nmi_stack_correct
+ cmpl $debug,(%esp)
+ jb nmi_stack_correct
+ cmpl $debug_esp_fix_insn,(%esp)
+ ja nmi_stack_correct
+ FIX_STACK(24,nmi_stack_correct, 1)
+ jmp nmi_stack_correct
+
+nmi_16bit_stack:
+ RING0_INT_FRAME
+ /* create the pointer to lss back */
+ pushl %ss
+ CFI_ADJUST_CFA_OFFSET 4
+ pushl %esp
+ CFI_ADJUST_CFA_OFFSET 4
+ movzwl %sp, %esp
+ addw $4, (%esp)
+ /* copy the iret frame of 12 bytes */
+ .rept 3
+ pushl 16(%esp)
+ CFI_ADJUST_CFA_OFFSET 4
+ .endr
+ pushl %eax
+ CFI_ADJUST_CFA_OFFSET 4
+ SAVE_ALL
+ FIXUP_ESPFIX_STACK # %eax == %esp
+ CFI_ADJUST_CFA_OFFSET -20 # the frame has now moved
+ xorl %edx,%edx # zero error code
+ call do_nmi
+ RESTORE_REGS
+ lss 12+4(%esp), %esp # back to 16bit stack
+1: iret
+ CFI_ENDPROC
+.section __ex_table,"a"
+ .align 4
+ .long 1b,iret_exc
+.previous
+#else
+ENTRY(nmi)
+ RING0_INT_FRAME
+ pushl %eax
+ CFI_ADJUST_CFA_OFFSET 4
+ SAVE_ALL
+ xorl %edx,%edx # zero error code
+ movl %esp,%eax # pt_regs pointer
+ call do_nmi
+ orl $NMI_MASK, EFLAGS(%esp)
+ jmp restore_all
+ CFI_ENDPROC
+#endif
+
+KPROBE_ENTRY(int3)
+ RING0_INT_FRAME
+ pushl $-1 # mark this as an int
+ CFI_ADJUST_CFA_OFFSET 4
+ SAVE_ALL
+ xorl %edx,%edx # zero error code
+ movl %esp,%eax # pt_regs pointer
+ call do_int3
+ jmp ret_from_exception
+ CFI_ENDPROC
+ .previous .text
+
+ENTRY(overflow)
+ RING0_INT_FRAME
+ pushl $0
+ CFI_ADJUST_CFA_OFFSET 4
+ pushl $do_overflow
+ CFI_ADJUST_CFA_OFFSET 4
+ jmp error_code
+ CFI_ENDPROC
+
+ENTRY(bounds)
+ RING0_INT_FRAME
+ pushl $0
+ CFI_ADJUST_CFA_OFFSET 4
+ pushl $do_bounds
+ CFI_ADJUST_CFA_OFFSET 4
+ jmp error_code
+ CFI_ENDPROC
+
+ENTRY(invalid_op)
+ RING0_INT_FRAME
+ pushl $0
+ CFI_ADJUST_CFA_OFFSET 4
+ pushl $do_invalid_op
+ CFI_ADJUST_CFA_OFFSET 4
+ jmp error_code
+ CFI_ENDPROC
+
+ENTRY(coprocessor_segment_overrun)
+ RING0_INT_FRAME
+ pushl $0
+ CFI_ADJUST_CFA_OFFSET 4
+ pushl $do_coprocessor_segment_overrun
+ CFI_ADJUST_CFA_OFFSET 4
+ jmp error_code
+ CFI_ENDPROC
+
+ENTRY(invalid_TSS)
+ RING0_EC_FRAME
+ pushl $do_invalid_TSS
+ CFI_ADJUST_CFA_OFFSET 4
+ jmp error_code
+ CFI_ENDPROC
+
+ENTRY(segment_not_present)
+ RING0_EC_FRAME
+ pushl $do_segment_not_present
+ CFI_ADJUST_CFA_OFFSET 4
+ jmp error_code
+ CFI_ENDPROC
+
+ENTRY(stack_segment)
+ RING0_EC_FRAME
+ pushl $do_stack_segment
+ CFI_ADJUST_CFA_OFFSET 4
+ jmp error_code
+ CFI_ENDPROC
+
+KPROBE_ENTRY(general_protection)
+ RING0_EC_FRAME
+ pushl $do_general_protection
+ CFI_ADJUST_CFA_OFFSET 4
+ jmp error_code
+ CFI_ENDPROC
+ .previous .text
+
+ENTRY(alignment_check)
+ RING0_EC_FRAME
+ pushl $do_alignment_check
+ CFI_ADJUST_CFA_OFFSET 4
+ jmp error_code
+ CFI_ENDPROC
+
+KPROBE_ENTRY(page_fault)
+ RING0_EC_FRAME
+ pushl $do_page_fault
+ CFI_ADJUST_CFA_OFFSET 4
+ jmp error_code
+ CFI_ENDPROC
+ .previous .text
+
+#ifdef CONFIG_X86_MCE
+ENTRY(machine_check)
+ RING0_INT_FRAME
+ pushl $0
+ CFI_ADJUST_CFA_OFFSET 4
+ pushl machine_check_vector
+ CFI_ADJUST_CFA_OFFSET 4
+ jmp error_code
+ CFI_ENDPROC
+#endif
+
+#ifndef CONFIG_XEN
+ENTRY(spurious_interrupt_bug)
+ RING0_INT_FRAME
+ pushl $0
+ CFI_ADJUST_CFA_OFFSET 4
+ pushl $do_spurious_interrupt_bug
+ CFI_ADJUST_CFA_OFFSET 4
+ jmp error_code
+ CFI_ENDPROC
+#endif /* !CONFIG_XEN */
+
+#ifdef CONFIG_STACK_UNWIND
+ENTRY(arch_unwind_init_running)
+ CFI_STARTPROC
+ movl 4(%esp), %edx
+ movl (%esp), %ecx
+ leal 4(%esp), %eax
+ movl %ebx, EBX(%edx)
+ xorl %ebx, %ebx
+ movl %ebx, ECX(%edx)
+ movl %ebx, EDX(%edx)
+ movl %esi, ESI(%edx)
+ movl %edi, EDI(%edx)
+ movl %ebp, EBP(%edx)
+ movl %ebx, EAX(%edx)
+ movl $__USER_DS, DS(%edx)
+ movl $__USER_DS, ES(%edx)
+ movl %ebx, ORIG_EAX(%edx)
+ movl %ecx, EIP(%edx)
+ movl 12(%esp), %ecx
+ movl $__KERNEL_CS, CS(%edx)
+ movl %ebx, EFLAGS(%edx)
+ movl %eax, OLDESP(%edx)
+ movl 8(%esp), %eax
+ movl %ecx, 8(%esp)
+ movl EBX(%edx), %ebx
+ movl $__KERNEL_DS, OLDSS(%edx)
+ jmpl *%eax
+ CFI_ENDPROC
+ENDPROC(arch_unwind_init_running)
+#endif
+
+ENTRY(fixup_4gb_segment)
+ RING0_EC_FRAME
+ pushl $do_fixup_4gb_segment
+ CFI_ADJUST_CFA_OFFSET 4
+ jmp error_code
+ CFI_ENDPROC
+
+.section .rodata,"a"
+#include "syscall_table.S"
+
+syscall_table_size=(.-sys_call_table)
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/kernel/fixup.c 2008-01-28 12:24:18.000000000 +0100
@@ -0,0 +1,88 @@
+/******************************************************************************
+ * fixup.c
+ *
+ * Binary-rewriting of certain IA32 instructions, on notification by Xen.
+ * Used to avoid repeated slow emulation of common instructions used by the
+ * user-space TLS (Thread-Local Storage) libraries.
+ *
+ * **** NOTE ****
+ * Issues with the binary rewriting have caused it to be removed. Instead
+ * we rely on Xen's emulator to boot the kernel, and then print a banner
+ * message recommending that the user disables /lib/tls.
+ *
+ * Copyright (c) 2004, K A Fraser
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ */
+
+#include <linux/init.h>
+#include <linux/sched.h>
+#include <linux/slab.h>
+#include <linux/kernel.h>
+#include <linux/delay.h>
+#include <linux/version.h>
+
+#define DP(_f, _args...) printk(KERN_ALERT " " _f "\n" , ## _args )
+
+fastcall void do_fixup_4gb_segment(struct pt_regs *regs, long error_code)
+{
+ static unsigned long printed = 0;
+ char info[100];
+ int i;
+
+ /* Ignore statically-linked init. */
+ if (current->tgid == 1)
+ return;
+
+ VOID(HYPERVISOR_vm_assist(VMASST_CMD_disable,
+ VMASST_TYPE_4gb_segments_notify));
+
+ if (test_and_set_bit(0, &printed))
+ return;
+
+ sprintf(info, "%s (pid=%d)", current->comm, current->tgid);
+
+ DP("");
+ DP("***************************************************************");
+ DP("***************************************************************");
+ DP("** WARNING: Currently emulating unsupported memory accesses **");
+ DP("** in /lib/tls glibc libraries. The emulation is **");
+ DP("** slow. To ensure full performance you should **");
+ DP("** install a 'xen-friendly' (nosegneg) version of **");
+ DP("** the library, or disable tls support by executing **");
+ DP("** the following as root: **");
+ DP("** mv /lib/tls /lib/tls.disabled **");
+ DP("** Offending process: %-38.38s **", info);
+ DP("***************************************************************");
+ DP("***************************************************************");
+ DP("");
+
+ for (i = 5; i > 0; i--) {
+ touch_softlockup_watchdog();
+ printk("Pausing... %d", i);
+ mdelay(1000);
+ printk("\b\b\b\b\b\b\b\b\b\b\b\b");
+ }
+
+ printk("Continuing...\n\n");
+}
+
+static int __init fixup_init(void)
+{
+ WARN_ON(HYPERVISOR_vm_assist(VMASST_CMD_enable,
+ VMASST_TYPE_4gb_segments_notify));
+ return 0;
+}
+__initcall(fixup_init);
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/kernel/head_32-xen.S 2007-06-12 13:12:48.000000000 +0200
@@ -0,0 +1,207 @@
+
+
+.text
+#include <linux/elfnote.h>
+#include <linux/threads.h>
+#include <linux/linkage.h>
+#include <asm/segment.h>
+#include <asm/page.h>
+#include <asm/cache.h>
+#include <asm/thread_info.h>
+#include <asm/asm-offsets.h>
+#include <asm/dwarf2.h>
+#include <xen/interface/xen.h>
+#include <xen/interface/elfnote.h>
+
+/*
+ * References to members of the new_cpu_data structure.
+ */
+
+#define X86 new_cpu_data+CPUINFO_x86
+#define X86_VENDOR new_cpu_data+CPUINFO_x86_vendor
+#define X86_MODEL new_cpu_data+CPUINFO_x86_model
+#define X86_MASK new_cpu_data+CPUINFO_x86_mask
+#define X86_HARD_MATH new_cpu_data+CPUINFO_hard_math
+#define X86_CPUID new_cpu_data+CPUINFO_cpuid_level
+#define X86_CAPABILITY new_cpu_data+CPUINFO_x86_capability
+#define X86_VENDOR_ID new_cpu_data+CPUINFO_x86_vendor_id
+
+#define VIRT_ENTRY_OFFSET 0x0
+.org VIRT_ENTRY_OFFSET
+ENTRY(startup_32)
+ movl %esi,xen_start_info
+ cld
+
+ /* Set up the stack pointer */
+ movl $(init_thread_union+THREAD_SIZE),%esp
+
+ /* get vendor info */
+ xorl %eax,%eax # call CPUID with 0 -> return vendor ID
+ XEN_CPUID
+ movl %eax,X86_CPUID # save CPUID level
+ movl %ebx,X86_VENDOR_ID # lo 4 chars
+ movl %edx,X86_VENDOR_ID+4 # next 4 chars
+ movl %ecx,X86_VENDOR_ID+8 # last 4 chars
+
+ movl $1,%eax # Use the CPUID instruction to get CPU type
+ XEN_CPUID
+ movb %al,%cl # save reg for future use
+ andb $0x0f,%ah # mask processor family
+ movb %ah,X86
+ andb $0xf0,%al # mask model
+ shrb $4,%al
+ movb %al,X86_MODEL
+ andb $0x0f,%cl # mask mask revision
+ movb %cl,X86_MASK
+ movl %edx,X86_CAPABILITY
+
+ movb $1,X86_HARD_MATH
+
+ xorl %eax,%eax # Clear FS/GS and LDT
+ movl %eax,%fs
+ movl %eax,%gs
+ cld # gcc2 wants the direction flag cleared at all times
+
+ pushl %eax # fake return address
+ jmp start_kernel
+
+#define HYPERCALL_PAGE_OFFSET 0x1000
+.org HYPERCALL_PAGE_OFFSET
+ENTRY(hypercall_page)
+ CFI_STARTPROC
+.skip 0x1000
+ CFI_ENDPROC
+
+/*
+ * Real beginning of normal "text" segment
+ */
+ENTRY(stext)
+ENTRY(_stext)
+
+/*
+ * BSS section
+ */
+.section ".bss.page_aligned","w"
+ENTRY(empty_zero_page)
+ .fill 4096,1,0
+
+/*
+ * This starts the data section.
+ */
+.data
+
+/*
+ * The Global Descriptor Table contains 28 quadwords, per-CPU.
+ */
+ .align L1_CACHE_BYTES
+ENTRY(cpu_gdt_table)
+ .quad 0x0000000000000000 /* NULL descriptor */
+ .quad 0x0000000000000000 /* 0x0b reserved */
+ .quad 0x0000000000000000 /* 0x13 reserved */
+ .quad 0x0000000000000000 /* 0x1b reserved */
+ .quad 0x0000000000000000 /* 0x20 unused */
+ .quad 0x0000000000000000 /* 0x28 unused */
+ .quad 0x0000000000000000 /* 0x33 TLS entry 1 */
+ .quad 0x0000000000000000 /* 0x3b TLS entry 2 */
+ .quad 0x0000000000000000 /* 0x43 TLS entry 3 */
+ .quad 0x0000000000000000 /* 0x4b reserved */
+ .quad 0x0000000000000000 /* 0x53 reserved */
+ .quad 0x0000000000000000 /* 0x5b reserved */
+
+ .quad 0x00cf9a000000ffff /* 0x60 kernel 4GB code at 0x00000000 */
+ .quad 0x00cf92000000ffff /* 0x68 kernel 4GB data at 0x00000000 */
+ .quad 0x00cffa000000ffff /* 0x73 user 4GB code at 0x00000000 */
+ .quad 0x00cff2000000ffff /* 0x7b user 4GB data at 0x00000000 */
+
+ .quad 0x0000000000000000 /* 0x80 TSS descriptor */
+ .quad 0x0000000000000000 /* 0x88 LDT descriptor */
+
+ /*
+ * Segments used for calling PnP BIOS have byte granularity.
+ * They code segments and data segments have fixed 64k limits,
+ * the transfer segment sizes are set at run time.
+ */
+ .quad 0x0000000000000000 /* 0x90 32-bit code */
+ .quad 0x0000000000000000 /* 0x98 16-bit code */
+ .quad 0x0000000000000000 /* 0xa0 16-bit data */
+ .quad 0x0000000000000000 /* 0xa8 16-bit data */
+ .quad 0x0000000000000000 /* 0xb0 16-bit data */
+
+ /*
+ * The APM segments have byte granularity and their bases
+ * are set at run time. All have 64k limits.
+ */
+ .quad 0x0000000000000000 /* 0xb8 APM CS code */
+ .quad 0x0000000000000000 /* 0xc0 APM CS 16 code (16 bit) */
+ .quad 0x0000000000000000 /* 0xc8 APM DS data */
+
+ .quad 0x0000000000000000 /* 0xd0 - ESPFIX 16-bit SS */
+ .quad 0x0000000000000000 /* 0xd8 - unused */
+ .quad 0x0000000000000000 /* 0xe0 - unused */
+ .quad 0x0000000000000000 /* 0xe8 - unused */
+ .quad 0x0000000000000000 /* 0xf0 - unused */
+ .quad 0x0000000000000000 /* 0xf8 - GDT entry 31: double-fault TSS */
+
+#if CONFIG_XEN_COMPAT <= 0x030002
+/*
+ * __xen_guest information
+ */
+.macro utoa value
+ .if (\value) < 0 || (\value) >= 0x10
+ utoa (((\value)>>4)&0x0fffffff)
+ .endif
+ .if ((\value) & 0xf) < 10
+ .byte '0' + ((\value) & 0xf)
+ .else
+ .byte 'A' + ((\value) & 0xf) - 10
+ .endif
+.endm
+
+.section __xen_guest
+ .ascii "GUEST_OS=linux,GUEST_VER=2.6"
+ .ascii ",XEN_VER=xen-3.0"
+ .ascii ",VIRT_BASE=0x"
+ utoa __PAGE_OFFSET
+ .ascii ",ELF_PADDR_OFFSET=0x"
+ utoa __PAGE_OFFSET
+ .ascii ",VIRT_ENTRY=0x"
+ utoa (__PAGE_OFFSET + __PHYSICAL_START + VIRT_ENTRY_OFFSET)
+ .ascii ",HYPERCALL_PAGE=0x"
+ utoa ((__PHYSICAL_START+HYPERCALL_PAGE_OFFSET)>>PAGE_SHIFT)
+ .ascii ",FEATURES=writable_page_tables"
+ .ascii "|writable_descriptor_tables"
+ .ascii "|auto_translated_physmap"
+ .ascii "|pae_pgdir_above_4gb"
+ .ascii "|supervisor_mode_kernel"
+#ifdef CONFIG_X86_PAE
+ .ascii ",PAE=yes[extended-cr3]"
+#else
+ .ascii ",PAE=no"
+#endif
+ .ascii ",LOADER=generic"
+ .byte 0
+#endif /* CONFIG_XEN_COMPAT <= 0x030002 */
+
+
+ ELFNOTE(Xen, XEN_ELFNOTE_GUEST_OS, .asciz, "linux")
+ ELFNOTE(Xen, XEN_ELFNOTE_GUEST_VERSION, .asciz, "2.6")
+ ELFNOTE(Xen, XEN_ELFNOTE_XEN_VERSION, .asciz, "xen-3.0")
+ ELFNOTE(Xen, XEN_ELFNOTE_VIRT_BASE, .long, __PAGE_OFFSET)
+#if CONFIG_XEN_COMPAT <= 0x030002
+ ELFNOTE(Xen, XEN_ELFNOTE_PADDR_OFFSET, .long, __PAGE_OFFSET)
+#else
+ ELFNOTE(Xen, XEN_ELFNOTE_PADDR_OFFSET, .long, 0)
+#endif
+ ELFNOTE(Xen, XEN_ELFNOTE_ENTRY, .long, startup_32)
+ ELFNOTE(Xen, XEN_ELFNOTE_HYPERCALL_PAGE, .long, hypercall_page)
+ ELFNOTE(Xen, XEN_ELFNOTE_HV_START_LOW, .long, HYPERVISOR_VIRT_START)
+ ELFNOTE(Xen, XEN_ELFNOTE_FEATURES, .asciz, "writable_page_tables|writable_descriptor_tables|auto_translated_physmap|pae_pgdir_above_4gb|supervisor_mode_kernel")
+#ifdef CONFIG_X86_PAE
+ ELFNOTE(Xen, XEN_ELFNOTE_PAE_MODE, .asciz, "yes")
+ ELFNOTE(Xen, XEN_ELFNOTE_L1_MFN_VALID, .quad, _PAGE_PRESENT,_PAGE_PRESENT)
+#else
+ ELFNOTE(Xen, XEN_ELFNOTE_PAE_MODE, .asciz, "no")
+ ELFNOTE(Xen, XEN_ELFNOTE_L1_MFN_VALID, .long, _PAGE_PRESENT,_PAGE_PRESENT)
+#endif
+ ELFNOTE(Xen, XEN_ELFNOTE_LOADER, .asciz, "generic")
+ ELFNOTE(Xen, XEN_ELFNOTE_SUSPEND_CANCEL, .long, 1)
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/kernel/io_apic_32-xen.c 2009-03-18 10:39:31.000000000 +0100
@@ -0,0 +1,2786 @@
+/*
+ * Intel IO-APIC support for multi-Pentium hosts.
+ *
+ * Copyright (C) 1997, 1998, 1999, 2000 Ingo Molnar, Hajnalka Szabo
+ *
+ * Many thanks to Stig Venaas for trying out countless experimental
+ * patches and reporting/debugging problems patiently!
+ *
+ * (c) 1999, Multiple IO-APIC support, developed by
+ * Ken-ichi Yaku <yaku@css1.kbnes.nec.co.jp> and
+ * Hidemi Kishimoto <kisimoto@css1.kbnes.nec.co.jp>,
+ * further tested and cleaned up by Zach Brown <zab@redhat.com>
+ * and Ingo Molnar <mingo@redhat.com>
+ *
+ * Fixes
+ * Maciej W. Rozycki : Bits for genuine 82489DX APICs;
+ * thanks to Eric Gilmore
+ * and Rolf G. Tews
+ * for testing these extensively
+ * Paul Diefenbaugh : Added full ACPI support
+ */
+
+#include <linux/mm.h>
+#include <linux/interrupt.h>
+#include <linux/init.h>
+#include <linux/delay.h>
+#include <linux/sched.h>
+#include <linux/smp_lock.h>
+#include <linux/mc146818rtc.h>
+#include <linux/compiler.h>
+#include <linux/acpi.h>
+#include <linux/module.h>
+#include <linux/sysdev.h>
+
+#include <asm/io.h>
+#include <asm/smp.h>
+#include <asm/desc.h>
+#include <asm/timer.h>
+#include <asm/i8259.h>
+#include <asm/nmi.h>
+
+#include <mach_apic.h>
+
+#include "io_ports.h"
+
+#ifdef CONFIG_XEN
+
+#include <xen/interface/xen.h>
+#include <xen/interface/physdev.h>
+#include <xen/evtchn.h>
+
+/* Fake i8259 */
+#define make_8259A_irq(_irq) (io_apic_irqs &= ~(1UL<<(_irq)))
+#define disable_8259A_irq(_irq) ((void)0)
+#define i8259A_irq_pending(_irq) (0)
+
+unsigned long io_apic_irqs;
+
+static inline unsigned int xen_io_apic_read(unsigned int apic, unsigned int reg)
+{
+ struct physdev_apic apic_op;
+ int ret;
+
+ apic_op.apic_physbase = mp_ioapics[apic].mpc_apicaddr;
+ apic_op.reg = reg;
+ ret = HYPERVISOR_physdev_op(PHYSDEVOP_apic_read, &apic_op);
+ if (ret)
+ return ret;
+ return apic_op.value;
+}
+
+static inline void xen_io_apic_write(unsigned int apic, unsigned int reg, unsigned int value)
+{
+ struct physdev_apic apic_op;
+
+ apic_op.apic_physbase = mp_ioapics[apic].mpc_apicaddr;
+ apic_op.reg = reg;
+ apic_op.value = value;
+ WARN_ON(HYPERVISOR_physdev_op(PHYSDEVOP_apic_write, &apic_op));
+}
+
+#define io_apic_read(a,r) xen_io_apic_read(a,r)
+#define io_apic_write(a,r,v) xen_io_apic_write(a,r,v)
+
+#endif /* CONFIG_XEN */
+
+int (*ioapic_renumber_irq)(int ioapic, int irq);
+atomic_t irq_mis_count;
+
+#ifndef CONFIG_XEN
+/* Where if anywhere is the i8259 connect in external int mode */
+static struct { int pin, apic; } ioapic_i8259 = { -1, -1 };
+#endif
+
+static DEFINE_SPINLOCK(ioapic_lock);
+static DEFINE_SPINLOCK(vector_lock);
+
+int timer_over_8254 __initdata = 1;
+
+/*
+ * Is the SiS APIC rmw bug present ?
+ * -1 = don't know, 0 = no, 1 = yes
+ */
+int sis_apic_bug = -1;
+
+/*
+ * # of IRQ routing registers
+ */
+int nr_ioapic_registers[MAX_IO_APICS];
+
+int disable_timer_pin_1 __initdata;
+
+/*
+ * Rough estimation of how many shared IRQs there are, can
+ * be changed anytime.
+ */
+#define MAX_PLUS_SHARED_IRQS NR_IRQS
+#define PIN_MAP_SIZE (MAX_PLUS_SHARED_IRQS + NR_IRQS)
+
+/*
+ * This is performance-critical, we want to do it O(1)
+ *
+ * the indexing order of this array favors 1:1 mappings
+ * between pins and IRQs.
+ */
+
+static struct irq_pin_list {
+ int apic, pin, next;
+} irq_2_pin[PIN_MAP_SIZE];
+
+int vector_irq[NR_VECTORS] __read_mostly = { [0 ... NR_VECTORS - 1] = -1};
+#ifdef CONFIG_PCI_MSI
+#define vector_to_irq(vector) \
+ (platform_legacy_irq(vector) ? vector : vector_irq[vector])
+#else
+#define vector_to_irq(vector) (vector)
+#endif
+
+/*
+ * The common case is 1:1 IRQ<->pin mappings. Sometimes there are
+ * shared ISA-space IRQs, so we have to support them. We are super
+ * fast in the common case, and fast for shared ISA-space IRQs.
+ */
+static void add_pin_to_irq(unsigned int irq, int apic, int pin)
+{
+ static int first_free_entry = NR_IRQS;
+ struct irq_pin_list *entry = irq_2_pin + irq;
+
+ while (entry->next)
+ entry = irq_2_pin + entry->next;
+
+ if (entry->pin != -1) {
+ entry->next = first_free_entry;
+ entry = irq_2_pin + entry->next;
+ if (++first_free_entry >= PIN_MAP_SIZE)
+ panic("io_apic.c: whoops");
+ }
+ entry->apic = apic;
+ entry->pin = pin;
+}
+
+#ifdef CONFIG_XEN
+#define clear_IO_APIC() ((void)0)
+#else
+/*
+ * Reroute an IRQ to a different pin.
+ */
+static void __init replace_pin_at_irq(unsigned int irq,
+ int oldapic, int oldpin,
+ int newapic, int newpin)
+{
+ struct irq_pin_list *entry = irq_2_pin + irq;
+
+ while (1) {
+ if (entry->apic == oldapic && entry->pin == oldpin) {
+ entry->apic = newapic;
+ entry->pin = newpin;
+ }
+ if (!entry->next)
+ break;
+ entry = irq_2_pin + entry->next;
+ }
+}
+
+static void __modify_IO_APIC_irq (unsigned int irq, unsigned long enable, unsigned long disable)
+{
+ struct irq_pin_list *entry = irq_2_pin + irq;
+ unsigned int pin, reg;
+
+ for (;;) {
+ pin = entry->pin;
+ if (pin == -1)
+ break;
+ reg = io_apic_read(entry->apic, 0x10 + pin*2);
+ reg &= ~disable;
+ reg |= enable;
+ io_apic_modify(entry->apic, 0x10 + pin*2, reg);
+ if (!entry->next)
+ break;
+ entry = irq_2_pin + entry->next;
+ }
+}
+
+/* mask = 1 */
+static void __mask_IO_APIC_irq (unsigned int irq)
+{
+ __modify_IO_APIC_irq(irq, 0x00010000, 0);
+}
+
+/* mask = 0 */
+static void __unmask_IO_APIC_irq (unsigned int irq)
+{
+ __modify_IO_APIC_irq(irq, 0, 0x00010000);
+}
+
+/* mask = 1, trigger = 0 */
+static void __mask_and_edge_IO_APIC_irq (unsigned int irq)
+{
+ __modify_IO_APIC_irq(irq, 0x00010000, 0x00008000);
+}
+
+/* mask = 0, trigger = 1 */
+static void __unmask_and_level_IO_APIC_irq (unsigned int irq)
+{
+ __modify_IO_APIC_irq(irq, 0x00008000, 0x00010000);
+}
+
+static void mask_IO_APIC_irq (unsigned int irq)
+{
+ unsigned long flags;
+
+ spin_lock_irqsave(&ioapic_lock, flags);
+ __mask_IO_APIC_irq(irq);
+ spin_unlock_irqrestore(&ioapic_lock, flags);
+}
+
+static void unmask_IO_APIC_irq (unsigned int irq)
+{
+ unsigned long flags;
+
+ spin_lock_irqsave(&ioapic_lock, flags);
+ __unmask_IO_APIC_irq(irq);
+ spin_unlock_irqrestore(&ioapic_lock, flags);
+}
+
+static void clear_IO_APIC_pin(unsigned int apic, unsigned int pin)
+{
+ struct IO_APIC_route_entry entry;
+ unsigned long flags;
+
+ /* Check delivery_mode to be sure we're not clearing an SMI pin */
+ spin_lock_irqsave(&ioapic_lock, flags);
+ *(((int*)&entry) + 0) = io_apic_read(apic, 0x10 + 2 * pin);
+ *(((int*)&entry) + 1) = io_apic_read(apic, 0x11 + 2 * pin);
+ spin_unlock_irqrestore(&ioapic_lock, flags);
+ if (entry.delivery_mode == dest_SMI)
+ return;
+
+ /*
+ * Disable it in the IO-APIC irq-routing table:
+ */
+ memset(&entry, 0, sizeof(entry));
+ entry.mask = 1;
+ spin_lock_irqsave(&ioapic_lock, flags);
+ io_apic_write(apic, 0x10 + 2 * pin, *(((int *)&entry) + 0));
+ io_apic_write(apic, 0x11 + 2 * pin, *(((int *)&entry) + 1));
+ spin_unlock_irqrestore(&ioapic_lock, flags);
+}
+
+static void clear_IO_APIC (void)
+{
+ int apic, pin;
+
+ for (apic = 0; apic < nr_ioapics; apic++)
+ for (pin = 0; pin < nr_ioapic_registers[apic]; pin++)
+ clear_IO_APIC_pin(apic, pin);
+}
+
+#ifdef CONFIG_SMP
+static void set_ioapic_affinity_irq(unsigned int irq, cpumask_t cpumask)
+{
+ unsigned long flags;
+ int pin;
+ struct irq_pin_list *entry = irq_2_pin + irq;
+ unsigned int apicid_value;
+ cpumask_t tmp;
+
+ cpus_and(tmp, cpumask, cpu_online_map);
+ if (cpus_empty(tmp))
+ tmp = TARGET_CPUS;
+
+ cpus_and(cpumask, tmp, CPU_MASK_ALL);
+
+ apicid_value = cpu_mask_to_apicid(cpumask);
+ /* Prepare to do the io_apic_write */
+ apicid_value = apicid_value << 24;
+ spin_lock_irqsave(&ioapic_lock, flags);
+ for (;;) {
+ pin = entry->pin;
+ if (pin == -1)
+ break;
+ io_apic_write(entry->apic, 0x10 + 1 + pin*2, apicid_value);
+ if (!entry->next)
+ break;
+ entry = irq_2_pin + entry->next;
+ }
+ set_irq_info(irq, cpumask);
+ spin_unlock_irqrestore(&ioapic_lock, flags);
+}
+
+#if defined(CONFIG_IRQBALANCE)
+# include <asm/processor.h> /* kernel_thread() */
+# include <linux/kernel_stat.h> /* kstat */
+# include <linux/slab.h> /* kmalloc() */
+# include <linux/timer.h> /* time_after() */
+
+#ifdef CONFIG_BALANCED_IRQ_DEBUG
+# define TDprintk(x...) do { printk("<%ld:%s:%d>: ", jiffies, __FILE__, __LINE__); printk(x); } while (0)
+# define Dprintk(x...) do { TDprintk(x); } while (0)
+# else
+# define TDprintk(x...)
+# define Dprintk(x...)
+# endif
+
+#define IRQBALANCE_CHECK_ARCH -999
+#define MAX_BALANCED_IRQ_INTERVAL (5*HZ)
+#define MIN_BALANCED_IRQ_INTERVAL (HZ/2)
+#define BALANCED_IRQ_MORE_DELTA (HZ/10)
+#define BALANCED_IRQ_LESS_DELTA (HZ)
+
+static int irqbalance_disabled __read_mostly = IRQBALANCE_CHECK_ARCH;
+static int physical_balance __read_mostly;
+static long balanced_irq_interval __read_mostly = MAX_BALANCED_IRQ_INTERVAL;
+
+static struct irq_cpu_info {
+ unsigned long * last_irq;
+ unsigned long * irq_delta;
+ unsigned long irq;
+} irq_cpu_data[NR_CPUS];
+
+#define CPU_IRQ(cpu) (irq_cpu_data[cpu].irq)
+#define LAST_CPU_IRQ(cpu,irq) (irq_cpu_data[cpu].last_irq[irq])
+#define IRQ_DELTA(cpu,irq) (irq_cpu_data[cpu].irq_delta[irq])
+
+#define IDLE_ENOUGH(cpu,now) \
+ (idle_cpu(cpu) && ((now) - per_cpu(irq_stat, (cpu)).idle_timestamp > 1))
+
+#define IRQ_ALLOWED(cpu, allowed_mask) cpu_isset(cpu, allowed_mask)
+
+#define CPU_TO_PACKAGEINDEX(i) (first_cpu(cpu_sibling_map[i]))
+
+static cpumask_t balance_irq_affinity[NR_IRQS] = {
+ [0 ... NR_IRQS-1] = CPU_MASK_ALL
+};
+
+void set_balance_irq_affinity(unsigned int irq, cpumask_t mask)
+{
+ balance_irq_affinity[irq] = mask;
+}
+
+static unsigned long move(int curr_cpu, cpumask_t allowed_mask,
+ unsigned long now, int direction)
+{
+ int search_idle = 1;
+ int cpu = curr_cpu;
+
+ goto inside;
+
+ do {
+ if (unlikely(cpu == curr_cpu))
+ search_idle = 0;
+inside:
+ if (direction == 1) {
+ cpu++;
+ if (cpu >= NR_CPUS)
+ cpu = 0;
+ } else {
+ cpu--;
+ if (cpu == -1)
+ cpu = NR_CPUS-1;
+ }
+ } while (!cpu_online(cpu) || !IRQ_ALLOWED(cpu,allowed_mask) ||
+ (search_idle && !IDLE_ENOUGH(cpu,now)));
+
+ return cpu;
+}
+
+static inline void balance_irq(int cpu, int irq)
+{
+ unsigned long now = jiffies;
+ cpumask_t allowed_mask;
+ unsigned int new_cpu;
+
+ if (irqbalance_disabled)
+ return;
+
+ cpus_and(allowed_mask, cpu_online_map, balance_irq_affinity[irq]);
+ new_cpu = move(cpu, allowed_mask, now, 1);
+ if (cpu != new_cpu) {
+ set_pending_irq(irq, cpumask_of_cpu(new_cpu));
+ }
+}
+
+static inline void rotate_irqs_among_cpus(unsigned long useful_load_threshold)
+{
+ int i, j;
+ Dprintk("Rotating IRQs among CPUs.\n");
+ for_each_online_cpu(i) {
+ for (j = 0; j < NR_IRQS; j++) {
+ if (!irq_desc[j].action)
+ continue;
+ /* Is it a significant load ? */
+ if (IRQ_DELTA(CPU_TO_PACKAGEINDEX(i),j) <
+ useful_load_threshold)
+ continue;
+ balance_irq(i, j);
+ }
+ }
+ balanced_irq_interval = max((long)MIN_BALANCED_IRQ_INTERVAL,
+ balanced_irq_interval - BALANCED_IRQ_LESS_DELTA);
+ return;
+}
+
+static void do_irq_balance(void)
+{
+ int i, j;
+ unsigned long max_cpu_irq = 0, min_cpu_irq = (~0);
+ unsigned long move_this_load = 0;
+ int max_loaded = 0, min_loaded = 0;
+ int load;
+ unsigned long useful_load_threshold = balanced_irq_interval + 10;
+ int selected_irq;
+ int tmp_loaded, first_attempt = 1;
+ unsigned long tmp_cpu_irq;
+ unsigned long imbalance = 0;
+ cpumask_t allowed_mask, target_cpu_mask, tmp;
+
+ for_each_possible_cpu(i) {
+ int package_index;
+ CPU_IRQ(i) = 0;
+ if (!cpu_online(i))
+ continue;
+ package_index = CPU_TO_PACKAGEINDEX(i);
+ for (j = 0; j < NR_IRQS; j++) {
+ unsigned long value_now, delta;
+ /* Is this an active IRQ? */
+ if (!irq_desc[j].action)
+ continue;
+ if ( package_index == i )
+ IRQ_DELTA(package_index,j) = 0;
+ /* Determine the total count per processor per IRQ */
+ value_now = (unsigned long) kstat_cpu(i).irqs[j];
+
+ /* Determine the activity per processor per IRQ */
+ delta = value_now - LAST_CPU_IRQ(i,j);
+
+ /* Update last_cpu_irq[][] for the next time */
+ LAST_CPU_IRQ(i,j) = value_now;
+
+ /* Ignore IRQs whose rate is less than the clock */
+ if (delta < useful_load_threshold)
+ continue;
+ /* update the load for the processor or package total */
+ IRQ_DELTA(package_index,j) += delta;
+
+ /* Keep track of the higher numbered sibling as well */
+ if (i != package_index)
+ CPU_IRQ(i) += delta;
+ /*
+ * We have sibling A and sibling B in the package
+ *
+ * cpu_irq[A] = load for cpu A + load for cpu B
+ * cpu_irq[B] = load for cpu B
+ */
+ CPU_IRQ(package_index) += delta;
+ }
+ }
+ /* Find the least loaded processor package */
+ for_each_online_cpu(i) {
+ if (i != CPU_TO_PACKAGEINDEX(i))
+ continue;
+ if (min_cpu_irq > CPU_IRQ(i)) {
+ min_cpu_irq = CPU_IRQ(i);
+ min_loaded = i;
+ }
+ }
+ max_cpu_irq = ULONG_MAX;
+
+tryanothercpu:
+ /* Look for heaviest loaded processor.
+ * We may come back to get the next heaviest loaded processor.
+ * Skip processors with trivial loads.
+ */
+ tmp_cpu_irq = 0;
+ tmp_loaded = -1;
+ for_each_online_cpu(i) {
+ if (i != CPU_TO_PACKAGEINDEX(i))
+ continue;
+ if (max_cpu_irq <= CPU_IRQ(i))
+ continue;
+ if (tmp_cpu_irq < CPU_IRQ(i)) {
+ tmp_cpu_irq = CPU_IRQ(i);
+ tmp_loaded = i;
+ }
+ }
+
+ if (tmp_loaded == -1) {
+ /* In the case of small number of heavy interrupt sources,
+ * loading some of the cpus too much. We use Ingo's original
+ * approach to rotate them around.
+ */
+ if (!first_attempt && imbalance >= useful_load_threshold) {
+ rotate_irqs_among_cpus(useful_load_threshold);
+ return;
+ }
+ goto not_worth_the_effort;
+ }
+
+ first_attempt = 0; /* heaviest search */
+ max_cpu_irq = tmp_cpu_irq; /* load */
+ max_loaded = tmp_loaded; /* processor */
+ imbalance = (max_cpu_irq - min_cpu_irq) / 2;
+
+ Dprintk("max_loaded cpu = %d\n", max_loaded);
+ Dprintk("min_loaded cpu = %d\n", min_loaded);
+ Dprintk("max_cpu_irq load = %ld\n", max_cpu_irq);
+ Dprintk("min_cpu_irq load = %ld\n", min_cpu_irq);
+ Dprintk("load imbalance = %lu\n", imbalance);
+
+ /* if imbalance is less than approx 10% of max load, then
+ * observe diminishing returns action. - quit
+ */
+ if (imbalance < (max_cpu_irq >> 3)) {
+ Dprintk("Imbalance too trivial\n");
+ goto not_worth_the_effort;
+ }
+
+tryanotherirq:
+ /* if we select an IRQ to move that can't go where we want, then
+ * see if there is another one to try.
+ */
+ move_this_load = 0;
+ selected_irq = -1;
+ for (j = 0; j < NR_IRQS; j++) {
+ /* Is this an active IRQ? */
+ if (!irq_desc[j].action)
+ continue;
+ if (imbalance <= IRQ_DELTA(max_loaded,j))
+ continue;
+ /* Try to find the IRQ that is closest to the imbalance
+ * without going over.
+ */
+ if (move_this_load < IRQ_DELTA(max_loaded,j)) {
+ move_this_load = IRQ_DELTA(max_loaded,j);
+ selected_irq = j;
+ }
+ }
+ if (selected_irq == -1) {
+ goto tryanothercpu;
+ }
+
+ imbalance = move_this_load;
+
+ /* For physical_balance case, we accumlated both load
+ * values in the one of the siblings cpu_irq[],
+ * to use the same code for physical and logical processors
+ * as much as possible.
+ *
+ * NOTE: the cpu_irq[] array holds the sum of the load for
+ * sibling A and sibling B in the slot for the lowest numbered
+ * sibling (A), _AND_ the load for sibling B in the slot for
+ * the higher numbered sibling.
+ *
+ * We seek the least loaded sibling by making the comparison
+ * (A+B)/2 vs B
+ */
+ load = CPU_IRQ(min_loaded) >> 1;
+ for_each_cpu_mask(j, cpu_sibling_map[min_loaded]) {
+ if (load > CPU_IRQ(j)) {
+ /* This won't change cpu_sibling_map[min_loaded] */
+ load = CPU_IRQ(j);
+ min_loaded = j;
+ }
+ }
+
+ cpus_and(allowed_mask,
+ cpu_online_map,
+ balance_irq_affinity[selected_irq]);
+ target_cpu_mask = cpumask_of_cpu(min_loaded);
+ cpus_and(tmp, target_cpu_mask, allowed_mask);
+
+ if (!cpus_empty(tmp)) {
+
+ Dprintk("irq = %d moved to cpu = %d\n",
+ selected_irq, min_loaded);
+ /* mark for change destination */
+ set_pending_irq(selected_irq, cpumask_of_cpu(min_loaded));
+
+ /* Since we made a change, come back sooner to
+ * check for more variation.
+ */
+ balanced_irq_interval = max((long)MIN_BALANCED_IRQ_INTERVAL,
+ balanced_irq_interval - BALANCED_IRQ_LESS_DELTA);
+ return;
+ }
+ goto tryanotherirq;
+
+not_worth_the_effort:
+ /*
+ * if we did not find an IRQ to move, then adjust the time interval
+ * upward
+ */
+ balanced_irq_interval = min((long)MAX_BALANCED_IRQ_INTERVAL,
+ balanced_irq_interval + BALANCED_IRQ_MORE_DELTA);
+ Dprintk("IRQ worth rotating not found\n");
+ return;
+}
+
+static int balanced_irq(void *unused)
+{
+ int i;
+ unsigned long prev_balance_time = jiffies;
+ long time_remaining = balanced_irq_interval;
+
+ daemonize("kirqd");
+
+ /* push everything to CPU 0 to give us a starting point. */
+ for (i = 0 ; i < NR_IRQS ; i++) {
+ irq_desc[i].pending_mask = cpumask_of_cpu(0);
+ set_pending_irq(i, cpumask_of_cpu(0));
+ }
+
+ for ( ; ; ) {
+ time_remaining = schedule_timeout_interruptible(time_remaining);
+ try_to_freeze();
+ if (time_after(jiffies,
+ prev_balance_time+balanced_irq_interval)) {
+ preempt_disable();
+ do_irq_balance();
+ prev_balance_time = jiffies;
+ time_remaining = balanced_irq_interval;
+ preempt_enable();
+ }
+ }
+ return 0;
+}
+
+static int __init balanced_irq_init(void)
+{
+ int i;
+ struct cpuinfo_x86 *c;
+ cpumask_t tmp;
+
+ cpus_shift_right(tmp, cpu_online_map, 2);
+ c = &boot_cpu_data;
+ /* When not overwritten by the command line ask subarchitecture. */
+ if (irqbalance_disabled == IRQBALANCE_CHECK_ARCH)
+ irqbalance_disabled = NO_BALANCE_IRQ;
+ if (irqbalance_disabled)
+ return 0;
+
+ /* disable irqbalance completely if there is only one processor online */
+ if (num_online_cpus() < 2) {
+ irqbalance_disabled = 1;
+ return 0;
+ }
+ /*
+ * Enable physical balance only if more than 1 physical processor
+ * is present
+ */
+ if (smp_num_siblings > 1 && !cpus_empty(tmp))
+ physical_balance = 1;
+
+ for_each_online_cpu(i) {
+ irq_cpu_data[i].irq_delta = kmalloc(sizeof(unsigned long) * NR_IRQS, GFP_KERNEL);
+ irq_cpu_data[i].last_irq = kmalloc(sizeof(unsigned long) * NR_IRQS, GFP_KERNEL);
+ if (irq_cpu_data[i].irq_delta == NULL || irq_cpu_data[i].last_irq == NULL) {
+ printk(KERN_ERR "balanced_irq_init: out of memory");
+ goto failed;
+ }
+ memset(irq_cpu_data[i].irq_delta,0,sizeof(unsigned long) * NR_IRQS);
+ memset(irq_cpu_data[i].last_irq,0,sizeof(unsigned long) * NR_IRQS);
+ }
+
+ printk(KERN_INFO "Starting balanced_irq\n");
+ if (kernel_thread(balanced_irq, NULL, CLONE_KERNEL) >= 0)
+ return 0;
+ else
+ printk(KERN_ERR "balanced_irq_init: failed to spawn balanced_irq");
+failed:
+ for_each_possible_cpu(i) {
+ kfree(irq_cpu_data[i].irq_delta);
+ irq_cpu_data[i].irq_delta = NULL;
+ kfree(irq_cpu_data[i].last_irq);
+ irq_cpu_data[i].last_irq = NULL;
+ }
+ return 0;
+}
+
+int __init irqbalance_disable(char *str)
+{
+ irqbalance_disabled = 1;
+ return 1;
+}
+
+__setup("noirqbalance", irqbalance_disable);
+
+late_initcall(balanced_irq_init);
+#endif /* CONFIG_IRQBALANCE */
+#endif /* CONFIG_SMP */
+#endif
+
+#ifndef CONFIG_SMP
+void fastcall send_IPI_self(int vector)
+{
+#ifndef CONFIG_XEN
+ unsigned int cfg;
+
+ /*
+ * Wait for idle.
+ */
+ apic_wait_icr_idle();
+ cfg = APIC_DM_FIXED | APIC_DEST_SELF | vector | APIC_DEST_LOGICAL;
+ /*
+ * Send the IPI. The write to APIC_ICR fires this off.
+ */
+ apic_write_around(APIC_ICR, cfg);
+#endif
+}
+#endif /* !CONFIG_SMP */
+
+
+/*
+ * support for broken MP BIOSs, enables hand-redirection of PIRQ0-7 to
+ * specific CPU-side IRQs.
+ */
+
+#define MAX_PIRQS 8
+static int pirq_entries [MAX_PIRQS];
+static int pirqs_enabled;
+int skip_ioapic_setup;
+
+static int __init ioapic_setup(char *str)
+{
+ skip_ioapic_setup = 1;
+ return 1;
+}
+
+__setup("noapic", ioapic_setup);
+
+static int __init ioapic_pirq_setup(char *str)
+{
+ int i, max;
+ int ints[MAX_PIRQS+1];
+
+ get_options(str, ARRAY_SIZE(ints), ints);
+
+ for (i = 0; i < MAX_PIRQS; i++)
+ pirq_entries[i] = -1;
+
+ pirqs_enabled = 1;
+ apic_printk(APIC_VERBOSE, KERN_INFO
+ "PIRQ redirection, working around broken MP-BIOS.\n");
+ max = MAX_PIRQS;
+ if (ints[0] < MAX_PIRQS)
+ max = ints[0];
+
+ for (i = 0; i < max; i++) {
+ apic_printk(APIC_VERBOSE, KERN_DEBUG
+ "... PIRQ%d -> IRQ %d\n", i, ints[i+1]);
+ /*
+ * PIRQs are mapped upside down, usually.
+ */
+ pirq_entries[MAX_PIRQS-i-1] = ints[i+1];
+ }
+ return 1;
+}
+
+__setup("pirq=", ioapic_pirq_setup);
+
+/*
+ * Find the IRQ entry number of a certain pin.
+ */
+static int find_irq_entry(int apic, int pin, int type)
+{
+ int i;
+
+ for (i = 0; i < mp_irq_entries; i++)
+ if (mp_irqs[i].mpc_irqtype == type &&
+ (mp_irqs[i].mpc_dstapic == mp_ioapics[apic].mpc_apicid ||
+ mp_irqs[i].mpc_dstapic == MP_APIC_ALL) &&
+ mp_irqs[i].mpc_dstirq == pin)
+ return i;
+
+ return -1;
+}
+
+#ifndef CONFIG_XEN
+/*
+ * Find the pin to which IRQ[irq] (ISA) is connected
+ */
+static int __init find_isa_irq_pin(int irq, int type)
+{
+ int i;
+
+ for (i = 0; i < mp_irq_entries; i++) {
+ int lbus = mp_irqs[i].mpc_srcbus;
+
+ if ((mp_bus_id_to_type[lbus] == MP_BUS_ISA ||
+ mp_bus_id_to_type[lbus] == MP_BUS_EISA ||
+ mp_bus_id_to_type[lbus] == MP_BUS_MCA ||
+ mp_bus_id_to_type[lbus] == MP_BUS_NEC98
+ ) &&
+ (mp_irqs[i].mpc_irqtype == type) &&
+ (mp_irqs[i].mpc_srcbusirq == irq))
+
+ return mp_irqs[i].mpc_dstirq;
+ }
+ return -1;
+}
+
+static int __init find_isa_irq_apic(int irq, int type)
+{
+ int i;
+
+ for (i = 0; i < mp_irq_entries; i++) {
+ int lbus = mp_irqs[i].mpc_srcbus;
+
+ if ((mp_bus_id_to_type[lbus] == MP_BUS_ISA ||
+ mp_bus_id_to_type[lbus] == MP_BUS_EISA ||
+ mp_bus_id_to_type[lbus] == MP_BUS_MCA ||
+ mp_bus_id_to_type[lbus] == MP_BUS_NEC98
+ ) &&
+ (mp_irqs[i].mpc_irqtype == type) &&
+ (mp_irqs[i].mpc_srcbusirq == irq))
+ break;
+ }
+ if (i < mp_irq_entries) {
+ int apic;
+ for(apic = 0; apic < nr_ioapics; apic++) {
+ if (mp_ioapics[apic].mpc_apicid == mp_irqs[i].mpc_dstapic)
+ return apic;
+ }
+ }
+
+ return -1;
+}
+#endif
+
+/*
+ * Find a specific PCI IRQ entry.
+ * Not an __init, possibly needed by modules
+ */
+static int pin_2_irq(int idx, int apic, int pin);
+
+int IO_APIC_get_PCI_irq_vector(int bus, int slot, int pin)
+{
+ int apic, i, best_guess = -1;
+
+ apic_printk(APIC_DEBUG, "querying PCI -> IRQ mapping bus:%d, "
+ "slot:%d, pin:%d.\n", bus, slot, pin);
+ if (mp_bus_id_to_pci_bus[bus] == -1) {
+ printk(KERN_WARNING "PCI BIOS passed nonexistent PCI bus %d!\n", bus);
+ return -1;
+ }
+ for (i = 0; i < mp_irq_entries; i++) {
+ int lbus = mp_irqs[i].mpc_srcbus;
+
+ for (apic = 0; apic < nr_ioapics; apic++)
+ if (mp_ioapics[apic].mpc_apicid == mp_irqs[i].mpc_dstapic ||
+ mp_irqs[i].mpc_dstapic == MP_APIC_ALL)
+ break;
+
+ if ((mp_bus_id_to_type[lbus] == MP_BUS_PCI) &&
+ !mp_irqs[i].mpc_irqtype &&
+ (bus == lbus) &&
+ (slot == ((mp_irqs[i].mpc_srcbusirq >> 2) & 0x1f))) {
+ int irq = pin_2_irq(i,apic,mp_irqs[i].mpc_dstirq);
+
+ if (!(apic || IO_APIC_IRQ(irq)))
+ continue;
+
+ if (pin == (mp_irqs[i].mpc_srcbusirq & 3))
+ return irq;
+ /*
+ * Use the first all-but-pin matching entry as a
+ * best-guess fuzzy result for broken mptables.
+ */
+ if (best_guess < 0)
+ best_guess = irq;
+ }
+ }
+ return best_guess;
+}
+EXPORT_SYMBOL(IO_APIC_get_PCI_irq_vector);
+
+/*
+ * This function currently is only a helper for the i386 smp boot process where
+ * we need to reprogram the ioredtbls to cater for the cpus which have come online
+ * so mask in all cases should simply be TARGET_CPUS
+ */
+#ifdef CONFIG_SMP
+#ifndef CONFIG_XEN
+void __init setup_ioapic_dest(void)
+{
+ int pin, ioapic, irq, irq_entry;
+
+ if (skip_ioapic_setup == 1)
+ return;
+
+ for (ioapic = 0; ioapic < nr_ioapics; ioapic++) {
+ for (pin = 0; pin < nr_ioapic_registers[ioapic]; pin++) {
+ irq_entry = find_irq_entry(ioapic, pin, mp_INT);
+ if (irq_entry == -1)
+ continue;
+ irq = pin_2_irq(irq_entry, ioapic, pin);
+ set_ioapic_affinity_irq(irq, TARGET_CPUS);
+ }
+
+ }
+}
+#endif /* !CONFIG_XEN */
+#endif
+
+/*
+ * EISA Edge/Level control register, ELCR
+ */
+static int EISA_ELCR(unsigned int irq)
+{
+ if (irq < 16) {
+ unsigned int port = 0x4d0 + (irq >> 3);
+ return (inb(port) >> (irq & 7)) & 1;
+ }
+ apic_printk(APIC_VERBOSE, KERN_INFO
+ "Broken MPtable reports ISA irq %d\n", irq);
+ return 0;
+}
+
+/* EISA interrupts are always polarity zero and can be edge or level
+ * trigger depending on the ELCR value. If an interrupt is listed as
+ * EISA conforming in the MP table, that means its trigger type must
+ * be read in from the ELCR */
+
+#define default_EISA_trigger(idx) (EISA_ELCR(mp_irqs[idx].mpc_srcbusirq))
+#define default_EISA_polarity(idx) (0)
+
+/* ISA interrupts are always polarity zero edge triggered,
+ * when listed as conforming in the MP table. */
+
+#define default_ISA_trigger(idx) (0)
+#define default_ISA_polarity(idx) (0)
+
+/* PCI interrupts are always polarity one level triggered,
+ * when listed as conforming in the MP table. */
+
+#define default_PCI_trigger(idx) (1)
+#define default_PCI_polarity(idx) (1)
+
+/* MCA interrupts are always polarity zero level triggered,
+ * when listed as conforming in the MP table. */
+
+#define default_MCA_trigger(idx) (1)
+#define default_MCA_polarity(idx) (0)
+
+/* NEC98 interrupts are always polarity zero edge triggered,
+ * when listed as conforming in the MP table. */
+
+#define default_NEC98_trigger(idx) (0)
+#define default_NEC98_polarity(idx) (0)
+
+static int __init MPBIOS_polarity(int idx)
+{
+ int bus = mp_irqs[idx].mpc_srcbus;
+ int polarity;
+
+ /*
+ * Determine IRQ line polarity (high active or low active):
+ */
+ switch (mp_irqs[idx].mpc_irqflag & 3)
+ {
+ case 0: /* conforms, ie. bus-type dependent polarity */
+ {
+ switch (mp_bus_id_to_type[bus])
+ {
+ case MP_BUS_ISA: /* ISA pin */
+ {
+ polarity = default_ISA_polarity(idx);
+ break;
+ }
+ case MP_BUS_EISA: /* EISA pin */
+ {
+ polarity = default_EISA_polarity(idx);
+ break;
+ }
+ case MP_BUS_PCI: /* PCI pin */
+ {
+ polarity = default_PCI_polarity(idx);
+ break;
+ }
+ case MP_BUS_MCA: /* MCA pin */
+ {
+ polarity = default_MCA_polarity(idx);
+ break;
+ }
+ case MP_BUS_NEC98: /* NEC 98 pin */
+ {
+ polarity = default_NEC98_polarity(idx);
+ break;
+ }
+ default:
+ {
+ printk(KERN_WARNING "broken BIOS!!\n");
+ polarity = 1;
+ break;
+ }
+ }
+ break;
+ }
+ case 1: /* high active */
+ {
+ polarity = 0;
+ break;
+ }
+ case 2: /* reserved */
+ {
+ printk(KERN_WARNING "broken BIOS!!\n");
+ polarity = 1;
+ break;
+ }
+ case 3: /* low active */
+ {
+ polarity = 1;
+ break;
+ }
+ default: /* invalid */
+ {
+ printk(KERN_WARNING "broken BIOS!!\n");
+ polarity = 1;
+ break;
+ }
+ }
+ return polarity;
+}
+
+static int MPBIOS_trigger(int idx)
+{
+ int bus = mp_irqs[idx].mpc_srcbus;
+ int trigger;
+
+ /*
+ * Determine IRQ trigger mode (edge or level sensitive):
+ */
+ switch ((mp_irqs[idx].mpc_irqflag>>2) & 3)
+ {
+ case 0: /* conforms, ie. bus-type dependent */
+ {
+ switch (mp_bus_id_to_type[bus])
+ {
+ case MP_BUS_ISA: /* ISA pin */
+ {
+ trigger = default_ISA_trigger(idx);
+ break;
+ }
+ case MP_BUS_EISA: /* EISA pin */
+ {
+ trigger = default_EISA_trigger(idx);
+ break;
+ }
+ case MP_BUS_PCI: /* PCI pin */
+ {
+ trigger = default_PCI_trigger(idx);
+ break;
+ }
+ case MP_BUS_MCA: /* MCA pin */
+ {
+ trigger = default_MCA_trigger(idx);
+ break;
+ }
+ case MP_BUS_NEC98: /* NEC 98 pin */
+ {
+ trigger = default_NEC98_trigger(idx);
+ break;
+ }
+ default:
+ {
+ printk(KERN_WARNING "broken BIOS!!\n");
+ trigger = 1;
+ break;
+ }
+ }
+ break;
+ }
+ case 1: /* edge */
+ {
+ trigger = 0;
+ break;
+ }
+ case 2: /* reserved */
+ {
+ printk(KERN_WARNING "broken BIOS!!\n");
+ trigger = 1;
+ break;
+ }
+ case 3: /* level */
+ {
+ trigger = 1;
+ break;
+ }
+ default: /* invalid */
+ {
+ printk(KERN_WARNING "broken BIOS!!\n");
+ trigger = 0;
+ break;
+ }
+ }
+ return trigger;
+}
+
+static inline int irq_polarity(int idx)
+{
+ return MPBIOS_polarity(idx);
+}
+
+static inline int irq_trigger(int idx)
+{
+ return MPBIOS_trigger(idx);
+}
+
+static int pin_2_irq(int idx, int apic, int pin)
+{
+ int irq, i;
+ int bus = mp_irqs[idx].mpc_srcbus;
+
+ /*
+ * Debugging check, we are in big trouble if this message pops up!
+ */
+ if (mp_irqs[idx].mpc_dstirq != pin)
+ printk(KERN_ERR "broken BIOS or MPTABLE parser, ayiee!!\n");
+
+ switch (mp_bus_id_to_type[bus])
+ {
+ case MP_BUS_ISA: /* ISA pin */
+ case MP_BUS_EISA:
+ case MP_BUS_MCA:
+ case MP_BUS_NEC98:
+ {
+ irq = mp_irqs[idx].mpc_srcbusirq;
+ break;
+ }
+ case MP_BUS_PCI: /* PCI pin */
+ {
+ /*
+ * PCI IRQs are mapped in order
+ */
+ i = irq = 0;
+ while (i < apic)
+ irq += nr_ioapic_registers[i++];
+ irq += pin;
+
+ /*
+ * For MPS mode, so far only needed by ES7000 platform
+ */
+ if (ioapic_renumber_irq)
+ irq = ioapic_renumber_irq(apic, irq);
+
+ break;
+ }
+ default:
+ {
+ printk(KERN_ERR "unknown bus type %d.\n",bus);
+ irq = 0;
+ break;
+ }
+ }
+
+ /*
+ * PCI IRQ command line redirection. Yes, limits are hardcoded.
+ */
+ if ((pin >= 16) && (pin <= 23)) {
+ if (pirq_entries[pin-16] != -1) {
+ if (!pirq_entries[pin-16]) {
+ apic_printk(APIC_VERBOSE, KERN_DEBUG
+ "disabling PIRQ%d\n", pin-16);
+ } else {
+ irq = pirq_entries[pin-16];
+ apic_printk(APIC_VERBOSE, KERN_DEBUG
+ "using PIRQ%d -> IRQ %d\n",
+ pin-16, irq);
+ }
+ }
+ }
+ return irq;
+}
+
+static inline int IO_APIC_irq_trigger(int irq)
+{
+ int apic, idx, pin;
+
+ for (apic = 0; apic < nr_ioapics; apic++) {
+ for (pin = 0; pin < nr_ioapic_registers[apic]; pin++) {
+ idx = find_irq_entry(apic,pin,mp_INT);
+ if ((idx != -1) && (irq == pin_2_irq(idx,apic,pin)))
+ return irq_trigger(idx);
+ }
+ }
+ /*
+ * nonexistent IRQs are edge default
+ */
+ return 0;
+}
+
+/* irq_vectors is indexed by the sum of all RTEs in all I/O APICs. */
+u8 irq_vector[NR_IRQ_VECTORS] __read_mostly; /* = { FIRST_DEVICE_VECTOR , 0 }; */
+
+int assign_irq_vector(int irq)
+{
+ unsigned long flags;
+ int vector;
+ struct physdev_irq irq_op;
+
+ BUG_ON(irq != AUTO_ASSIGN && (unsigned)irq >= NR_IRQ_VECTORS);
+
+ if (irq < PIRQ_BASE || irq - PIRQ_BASE >= NR_PIRQS)
+ return -EINVAL;
+
+ spin_lock_irqsave(&vector_lock, flags);
+
+ if (irq != AUTO_ASSIGN && IO_APIC_VECTOR(irq) > 0) {
+ spin_unlock_irqrestore(&vector_lock, flags);
+ return IO_APIC_VECTOR(irq);
+ }
+
+ irq_op.irq = irq;
+ if (HYPERVISOR_physdev_op(PHYSDEVOP_alloc_irq_vector, &irq_op)) {
+ spin_unlock_irqrestore(&vector_lock, flags);
+ return -ENOSPC;
+ }
+
+ vector = irq_op.vector;
+ vector_irq[vector] = irq;
+ if (irq != AUTO_ASSIGN)
+ IO_APIC_VECTOR(irq) = vector;
+
+ spin_unlock_irqrestore(&vector_lock, flags);
+
+ return vector;
+}
+
+#ifndef CONFIG_XEN
+static struct hw_interrupt_type ioapic_level_type;
+static struct hw_interrupt_type ioapic_edge_type;
+
+#define IOAPIC_AUTO -1
+#define IOAPIC_EDGE 0
+#define IOAPIC_LEVEL 1
+
+static void ioapic_register_intr(int irq, int vector, unsigned long trigger)
+{
+ unsigned idx;
+
+ idx = use_pci_vector() && !platform_legacy_irq(irq) ? vector : irq;
+
+ if ((trigger == IOAPIC_AUTO && IO_APIC_irq_trigger(irq)) ||
+ trigger == IOAPIC_LEVEL)
+ irq_desc[idx].chip = &ioapic_level_type;
+ else
+ irq_desc[idx].chip = &ioapic_edge_type;
+ set_intr_gate(vector, interrupt[idx]);
+}
+#else
+#define ioapic_register_intr(irq, vector, trigger) evtchn_register_pirq(irq)
+#endif
+
+static void __init setup_IO_APIC_irqs(void)
+{
+ struct IO_APIC_route_entry entry;
+ int apic, pin, idx, irq, first_notcon = 1, vector;
+ unsigned long flags;
+
+ apic_printk(APIC_VERBOSE, KERN_DEBUG "init IO_APIC IRQs\n");
+
+ for (apic = 0; apic < nr_ioapics; apic++) {
+ for (pin = 0; pin < nr_ioapic_registers[apic]; pin++) {
+
+ /*
+ * add it to the IO-APIC irq-routing table:
+ */
+ memset(&entry,0,sizeof(entry));
+
+ entry.delivery_mode = INT_DELIVERY_MODE;
+ entry.dest_mode = INT_DEST_MODE;
+ entry.mask = 0; /* enable IRQ */
+ entry.dest.logical.logical_dest =
+ cpu_mask_to_apicid(TARGET_CPUS);
+
+ idx = find_irq_entry(apic,pin,mp_INT);
+ if (idx == -1) {
+ if (first_notcon) {
+ apic_printk(APIC_VERBOSE, KERN_DEBUG
+ " IO-APIC (apicid-pin) %d-%d",
+ mp_ioapics[apic].mpc_apicid,
+ pin);
+ first_notcon = 0;
+ } else
+ apic_printk(APIC_VERBOSE, ", %d-%d",
+ mp_ioapics[apic].mpc_apicid, pin);
+ continue;
+ }
+
+ entry.trigger = irq_trigger(idx);
+ entry.polarity = irq_polarity(idx);
+
+ if (irq_trigger(idx)) {
+ entry.trigger = 1;
+ entry.mask = 1;
+ }
+
+ irq = pin_2_irq(idx, apic, pin);
+ /*
+ * skip adding the timer int on secondary nodes, which causes
+ * a small but painful rift in the time-space continuum
+ */
+ if (multi_timer_check(apic, irq))
+ continue;
+ else
+ add_pin_to_irq(irq, apic, pin);
+
+ if (/*!apic &&*/ !IO_APIC_IRQ(irq))
+ continue;
+
+ if (IO_APIC_IRQ(irq)) {
+ vector = assign_irq_vector(irq);
+ entry.vector = vector;
+ ioapic_register_intr(irq, vector, IOAPIC_AUTO);
+
+ if (!apic && (irq < 16))
+ disable_8259A_irq(irq);
+ }
+ spin_lock_irqsave(&ioapic_lock, flags);
+ io_apic_write(apic, 0x11+2*pin, *(((int *)&entry)+1));
+ io_apic_write(apic, 0x10+2*pin, *(((int *)&entry)+0));
+ set_native_irq_info(irq, TARGET_CPUS);
+ spin_unlock_irqrestore(&ioapic_lock, flags);
+ }
+ }
+
+ if (!first_notcon)
+ apic_printk(APIC_VERBOSE, " not connected.\n");
+}
+
+/*
+ * Set up the 8259A-master output pin:
+ */
+#ifndef CONFIG_XEN
+static void __init setup_ExtINT_IRQ0_pin(unsigned int apic, unsigned int pin, int vector)
+{
+ struct IO_APIC_route_entry entry;
+ unsigned long flags;
+
+ memset(&entry,0,sizeof(entry));
+
+ disable_8259A_irq(0);
+
+ /* mask LVT0 */
+ apic_write_around(APIC_LVT0, APIC_LVT_MASKED | APIC_DM_EXTINT);
+
+ /*
+ * We use logical delivery to get the timer IRQ
+ * to the first CPU.
+ */
+ entry.dest_mode = INT_DEST_MODE;
+ entry.mask = 0; /* unmask IRQ now */
+ entry.dest.logical.logical_dest = cpu_mask_to_apicid(TARGET_CPUS);
+ entry.delivery_mode = INT_DELIVERY_MODE;
+ entry.polarity = 0;
+ entry.trigger = 0;
+ entry.vector = vector;
+
+ /*
+ * The timer IRQ doesn't have to know that behind the
+ * scene we have a 8259A-master in AEOI mode ...
+ */
+ irq_desc[0].chip = &ioapic_edge_type;
+
+ /*
+ * Add it to the IO-APIC irq-routing table:
+ */
+ spin_lock_irqsave(&ioapic_lock, flags);
+ io_apic_write(apic, 0x11+2*pin, *(((int *)&entry)+1));
+ io_apic_write(apic, 0x10+2*pin, *(((int *)&entry)+0));
+ spin_unlock_irqrestore(&ioapic_lock, flags);
+
+ enable_8259A_irq(0);
+}
+
+static inline void UNEXPECTED_IO_APIC(void)
+{
+}
+
+void __init print_IO_APIC(void)
+{
+ int apic, i;
+ union IO_APIC_reg_00 reg_00;
+ union IO_APIC_reg_01 reg_01;
+ union IO_APIC_reg_02 reg_02;
+ union IO_APIC_reg_03 reg_03;
+ unsigned long flags;
+
+ if (apic_verbosity == APIC_QUIET)
+ return;
+
+ printk(KERN_DEBUG "number of MP IRQ sources: %d.\n", mp_irq_entries);
+ for (i = 0; i < nr_ioapics; i++)
+ printk(KERN_DEBUG "number of IO-APIC #%d registers: %d.\n",
+ mp_ioapics[i].mpc_apicid, nr_ioapic_registers[i]);
+
+ /*
+ * We are a bit conservative about what we expect. We have to
+ * know about every hardware change ASAP.
+ */
+ printk(KERN_INFO "testing the IO APIC.......................\n");
+
+ for (apic = 0; apic < nr_ioapics; apic++) {
+
+ spin_lock_irqsave(&ioapic_lock, flags);
+ reg_00.raw = io_apic_read(apic, 0);
+ reg_01.raw = io_apic_read(apic, 1);
+ if (reg_01.bits.version >= 0x10)
+ reg_02.raw = io_apic_read(apic, 2);
+ if (reg_01.bits.version >= 0x20)
+ reg_03.raw = io_apic_read(apic, 3);
+ spin_unlock_irqrestore(&ioapic_lock, flags);
+
+ printk(KERN_DEBUG "IO APIC #%d......\n", mp_ioapics[apic].mpc_apicid);
+ printk(KERN_DEBUG ".... register #00: %08X\n", reg_00.raw);
+ printk(KERN_DEBUG "....... : physical APIC id: %02X\n", reg_00.bits.ID);
+ printk(KERN_DEBUG "....... : Delivery Type: %X\n", reg_00.bits.delivery_type);
+ printk(KERN_DEBUG "....... : LTS : %X\n", reg_00.bits.LTS);
+ if (reg_00.bits.ID >= get_physical_broadcast())
+ UNEXPECTED_IO_APIC();
+ if (reg_00.bits.__reserved_1 || reg_00.bits.__reserved_2)
+ UNEXPECTED_IO_APIC();
+
+ printk(KERN_DEBUG ".... register #01: %08X\n", reg_01.raw);
+ printk(KERN_DEBUG "....... : max redirection entries: %04X\n", reg_01.bits.entries);
+ if ( (reg_01.bits.entries != 0x0f) && /* older (Neptune) boards */
+ (reg_01.bits.entries != 0x17) && /* typical ISA+PCI boards */
+ (reg_01.bits.entries != 0x1b) && /* Compaq Proliant boards */
+ (reg_01.bits.entries != 0x1f) && /* dual Xeon boards */
+ (reg_01.bits.entries != 0x22) && /* bigger Xeon boards */
+ (reg_01.bits.entries != 0x2E) &&
+ (reg_01.bits.entries != 0x3F)
+ )
+ UNEXPECTED_IO_APIC();
+
+ printk(KERN_DEBUG "....... : PRQ implemented: %X\n", reg_01.bits.PRQ);
+ printk(KERN_DEBUG "....... : IO APIC version: %04X\n", reg_01.bits.version);
+ if ( (reg_01.bits.version != 0x01) && /* 82489DX IO-APICs */
+ (reg_01.bits.version != 0x10) && /* oldest IO-APICs */
+ (reg_01.bits.version != 0x11) && /* Pentium/Pro IO-APICs */
+ (reg_01.bits.version != 0x13) && /* Xeon IO-APICs */
+ (reg_01.bits.version != 0x20) /* Intel P64H (82806 AA) */
+ )
+ UNEXPECTED_IO_APIC();
+ if (reg_01.bits.__reserved_1 || reg_01.bits.__reserved_2)
+ UNEXPECTED_IO_APIC();
+
+ /*
+ * Some Intel chipsets with IO APIC VERSION of 0x1? don't have reg_02,
+ * but the value of reg_02 is read as the previous read register
+ * value, so ignore it if reg_02 == reg_01.
+ */
+ if (reg_01.bits.version >= 0x10 && reg_02.raw != reg_01.raw) {
+ printk(KERN_DEBUG ".... register #02: %08X\n", reg_02.raw);
+ printk(KERN_DEBUG "....... : arbitration: %02X\n", reg_02.bits.arbitration);
+ if (reg_02.bits.__reserved_1 || reg_02.bits.__reserved_2)
+ UNEXPECTED_IO_APIC();
+ }
+
+ /*
+ * Some Intel chipsets with IO APIC VERSION of 0x2? don't have reg_02
+ * or reg_03, but the value of reg_0[23] is read as the previous read
+ * register value, so ignore it if reg_03 == reg_0[12].
+ */
+ if (reg_01.bits.version >= 0x20 && reg_03.raw != reg_02.raw &&
+ reg_03.raw != reg_01.raw) {
+ printk(KERN_DEBUG ".... register #03: %08X\n", reg_03.raw);
+ printk(KERN_DEBUG "....... : Boot DT : %X\n", reg_03.bits.boot_DT);
+ if (reg_03.bits.__reserved_1)
+ UNEXPECTED_IO_APIC();
+ }
+
+ printk(KERN_DEBUG ".... IRQ redirection table:\n");
+
+ printk(KERN_DEBUG " NR Log Phy Mask Trig IRR Pol"
+ " Stat Dest Deli Vect: \n");
+
+ for (i = 0; i <= reg_01.bits.entries; i++) {
+ struct IO_APIC_route_entry entry;
+
+ spin_lock_irqsave(&ioapic_lock, flags);
+ *(((int *)&entry)+0) = io_apic_read(apic, 0x10+i*2);
+ *(((int *)&entry)+1) = io_apic_read(apic, 0x11+i*2);
+ spin_unlock_irqrestore(&ioapic_lock, flags);
+
+ printk(KERN_DEBUG " %02x %03X %02X ",
+ i,
+ entry.dest.logical.logical_dest,
+ entry.dest.physical.physical_dest
+ );
+
+ printk("%1d %1d %1d %1d %1d %1d %1d %02X\n",
+ entry.mask,
+ entry.trigger,
+ entry.irr,
+ entry.polarity,
+ entry.delivery_status,
+ entry.dest_mode,
+ entry.delivery_mode,
+ entry.vector
+ );
+ }
+ }
+ if (use_pci_vector())
+ printk(KERN_INFO "Using vector-based indexing\n");
+ printk(KERN_DEBUG "IRQ to pin mappings:\n");
+ for (i = 0; i < NR_IRQS; i++) {
+ struct irq_pin_list *entry = irq_2_pin + i;
+ if (entry->pin < 0)
+ continue;
+ if (use_pci_vector() && !platform_legacy_irq(i))
+ printk(KERN_DEBUG "IRQ%d ", IO_APIC_VECTOR(i));
+ else
+ printk(KERN_DEBUG "IRQ%d ", i);
+ for (;;) {
+ printk("-> %d:%d", entry->apic, entry->pin);
+ if (!entry->next)
+ break;
+ entry = irq_2_pin + entry->next;
+ }
+ printk("\n");
+ }
+
+ printk(KERN_INFO ".................................... done.\n");
+
+ return;
+}
+
+static void print_APIC_bitfield (int base)
+{
+ unsigned int v;
+ int i, j;
+
+ if (apic_verbosity == APIC_QUIET)
+ return;
+
+ printk(KERN_DEBUG "0123456789abcdef0123456789abcdef\n" KERN_DEBUG);
+ for (i = 0; i < 8; i++) {
+ v = apic_read(base + i*0x10);
+ for (j = 0; j < 32; j++) {
+ if (v & (1<<j))
+ printk("1");
+ else
+ printk("0");
+ }
+ printk("\n");
+ }
+}
+
+void /*__init*/ print_local_APIC(void * dummy)
+{
+ unsigned int v, ver, maxlvt;
+
+ if (apic_verbosity == APIC_QUIET)
+ return;
+
+ printk("\n" KERN_DEBUG "printing local APIC contents on CPU#%d/%d:\n",
+ smp_processor_id(), hard_smp_processor_id());
+ v = apic_read(APIC_ID);
+ printk(KERN_INFO "... APIC ID: %08x (%01x)\n", v, GET_APIC_ID(v));
+ v = apic_read(APIC_LVR);
+ printk(KERN_INFO "... APIC VERSION: %08x\n", v);
+ ver = GET_APIC_VERSION(v);
+ maxlvt = get_maxlvt();
+
+ v = apic_read(APIC_TASKPRI);
+ printk(KERN_DEBUG "... APIC TASKPRI: %08x (%02x)\n", v, v & APIC_TPRI_MASK);
+
+ if (APIC_INTEGRATED(ver)) { /* !82489DX */
+ v = apic_read(APIC_ARBPRI);
+ printk(KERN_DEBUG "... APIC ARBPRI: %08x (%02x)\n", v,
+ v & APIC_ARBPRI_MASK);
+ v = apic_read(APIC_PROCPRI);
+ printk(KERN_DEBUG "... APIC PROCPRI: %08x\n", v);
+ }
+
+ v = apic_read(APIC_EOI);
+ printk(KERN_DEBUG "... APIC EOI: %08x\n", v);
+ v = apic_read(APIC_RRR);
+ printk(KERN_DEBUG "... APIC RRR: %08x\n", v);
+ v = apic_read(APIC_LDR);
+ printk(KERN_DEBUG "... APIC LDR: %08x\n", v);
+ v = apic_read(APIC_DFR);
+ printk(KERN_DEBUG "... APIC DFR: %08x\n", v);
+ v = apic_read(APIC_SPIV);
+ printk(KERN_DEBUG "... APIC SPIV: %08x\n", v);
+
+ printk(KERN_DEBUG "... APIC ISR field:\n");
+ print_APIC_bitfield(APIC_ISR);
+ printk(KERN_DEBUG "... APIC TMR field:\n");
+ print_APIC_bitfield(APIC_TMR);
+ printk(KERN_DEBUG "... APIC IRR field:\n");
+ print_APIC_bitfield(APIC_IRR);
+
+ if (APIC_INTEGRATED(ver)) { /* !82489DX */
+ if (maxlvt > 3) /* Due to the Pentium erratum 3AP. */
+ apic_write(APIC_ESR, 0);
+ v = apic_read(APIC_ESR);
+ printk(KERN_DEBUG "... APIC ESR: %08x\n", v);
+ }
+
+ v = apic_read(APIC_ICR);
+ printk(KERN_DEBUG "... APIC ICR: %08x\n", v);
+ v = apic_read(APIC_ICR2);
+ printk(KERN_DEBUG "... APIC ICR2: %08x\n", v);
+
+ v = apic_read(APIC_LVTT);
+ printk(KERN_DEBUG "... APIC LVTT: %08x\n", v);
+
+ if (maxlvt > 3) { /* PC is LVT#4. */
+ v = apic_read(APIC_LVTPC);
+ printk(KERN_DEBUG "... APIC LVTPC: %08x\n", v);
+ }
+ v = apic_read(APIC_LVT0);
+ printk(KERN_DEBUG "... APIC LVT0: %08x\n", v);
+ v = apic_read(APIC_LVT1);
+ printk(KERN_DEBUG "... APIC LVT1: %08x\n", v);
+
+ if (maxlvt > 2) { /* ERR is LVT#3. */
+ v = apic_read(APIC_LVTERR);
+ printk(KERN_DEBUG "... APIC LVTERR: %08x\n", v);
+ }
+
+ v = apic_read(APIC_TMICT);
+ printk(KERN_DEBUG "... APIC TMICT: %08x\n", v);
+ v = apic_read(APIC_TMCCT);
+ printk(KERN_DEBUG "... APIC TMCCT: %08x\n", v);
+ v = apic_read(APIC_TDCR);
+ printk(KERN_DEBUG "... APIC TDCR: %08x\n", v);
+ printk("\n");
+}
+
+void print_all_local_APICs (void)
+{
+ on_each_cpu(print_local_APIC, NULL, 1, 1);
+}
+
+void /*__init*/ print_PIC(void)
+{
+ unsigned int v;
+ unsigned long flags;
+
+ if (apic_verbosity == APIC_QUIET)
+ return;
+
+ printk(KERN_DEBUG "\nprinting PIC contents\n");
+
+ spin_lock_irqsave(&i8259A_lock, flags);
+
+ v = inb(0xa1) << 8 | inb(0x21);
+ printk(KERN_DEBUG "... PIC IMR: %04x\n", v);
+
+ v = inb(0xa0) << 8 | inb(0x20);
+ printk(KERN_DEBUG "... PIC IRR: %04x\n", v);
+
+ outb(0x0b,0xa0);
+ outb(0x0b,0x20);
+ v = inb(0xa0) << 8 | inb(0x20);
+ outb(0x0a,0xa0);
+ outb(0x0a,0x20);
+
+ spin_unlock_irqrestore(&i8259A_lock, flags);
+
+ printk(KERN_DEBUG "... PIC ISR: %04x\n", v);
+
+ v = inb(0x4d1) << 8 | inb(0x4d0);
+ printk(KERN_DEBUG "... PIC ELCR: %04x\n", v);
+}
+#endif /* !CONFIG_XEN */
+
+static void __init enable_IO_APIC(void)
+{
+ union IO_APIC_reg_01 reg_01;
+#ifndef CONFIG_XEN
+ int i8259_apic, i8259_pin;
+#endif
+ int i, apic;
+ unsigned long flags;
+
+ for (i = 0; i < PIN_MAP_SIZE; i++) {
+ irq_2_pin[i].pin = -1;
+ irq_2_pin[i].next = 0;
+ }
+ if (!pirqs_enabled)
+ for (i = 0; i < MAX_PIRQS; i++)
+ pirq_entries[i] = -1;
+
+ /*
+ * The number of IO-APIC IRQ registers (== #pins):
+ */
+ for (apic = 0; apic < nr_ioapics; apic++) {
+ spin_lock_irqsave(&ioapic_lock, flags);
+ reg_01.raw = io_apic_read(apic, 1);
+ spin_unlock_irqrestore(&ioapic_lock, flags);
+ nr_ioapic_registers[apic] = reg_01.bits.entries+1;
+ }
+#ifndef CONFIG_XEN
+ for(apic = 0; apic < nr_ioapics; apic++) {
+ int pin;
+ /* See if any of the pins is in ExtINT mode */
+ for (pin = 0; pin < nr_ioapic_registers[apic]; pin++) {
+ struct IO_APIC_route_entry entry;
+ spin_lock_irqsave(&ioapic_lock, flags);
+ *(((int *)&entry) + 0) = io_apic_read(apic, 0x10 + 2 * pin);
+ *(((int *)&entry) + 1) = io_apic_read(apic, 0x11 + 2 * pin);
+ spin_unlock_irqrestore(&ioapic_lock, flags);
+
+
+ /* If the interrupt line is enabled and in ExtInt mode
+ * I have found the pin where the i8259 is connected.
+ */
+ if ((entry.mask == 0) && (entry.delivery_mode == dest_ExtINT)) {
+ ioapic_i8259.apic = apic;
+ ioapic_i8259.pin = pin;
+ goto found_i8259;
+ }
+ }
+ }
+ found_i8259:
+ /* Look to see what if the MP table has reported the ExtINT */
+ /* If we could not find the appropriate pin by looking at the ioapic
+ * the i8259 probably is not connected the ioapic but give the
+ * mptable a chance anyway.
+ */
+ i8259_pin = find_isa_irq_pin(0, mp_ExtINT);
+ i8259_apic = find_isa_irq_apic(0, mp_ExtINT);
+ /* Trust the MP table if nothing is setup in the hardware */
+ if ((ioapic_i8259.pin == -1) && (i8259_pin >= 0)) {
+ printk(KERN_WARNING "ExtINT not setup in hardware but reported by MP table\n");
+ ioapic_i8259.pin = i8259_pin;
+ ioapic_i8259.apic = i8259_apic;
+ }
+ /* Complain if the MP table and the hardware disagree */
+ if (((ioapic_i8259.apic != i8259_apic) || (ioapic_i8259.pin != i8259_pin)) &&
+ (i8259_pin >= 0) && (ioapic_i8259.pin >= 0))
+ {
+ printk(KERN_WARNING "ExtINT in hardware and MP table differ\n");
+ }
+#endif
+
+ /*
+ * Do not trust the IO-APIC being empty at bootup
+ */
+ clear_IO_APIC();
+}
+
+/*
+ * Not an __init, needed by the reboot code
+ */
+void disable_IO_APIC(void)
+{
+ /*
+ * Clear the IO-APIC before rebooting:
+ */
+ clear_IO_APIC();
+
+#ifndef CONFIG_XEN
+ /*
+ * If the i8259 is routed through an IOAPIC
+ * Put that IOAPIC in virtual wire mode
+ * so legacy interrupts can be delivered.
+ */
+ if (ioapic_i8259.pin != -1) {
+ struct IO_APIC_route_entry entry;
+ unsigned long flags;
+
+ memset(&entry, 0, sizeof(entry));
+ entry.mask = 0; /* Enabled */
+ entry.trigger = 0; /* Edge */
+ entry.irr = 0;
+ entry.polarity = 0; /* High */
+ entry.delivery_status = 0;
+ entry.dest_mode = 0; /* Physical */
+ entry.delivery_mode = dest_ExtINT; /* ExtInt */
+ entry.vector = 0;
+ entry.dest.physical.physical_dest =
+ GET_APIC_ID(apic_read(APIC_ID));
+
+ /*
+ * Add it to the IO-APIC irq-routing table:
+ */
+ spin_lock_irqsave(&ioapic_lock, flags);
+ io_apic_write(ioapic_i8259.apic, 0x11+2*ioapic_i8259.pin,
+ *(((int *)&entry)+1));
+ io_apic_write(ioapic_i8259.apic, 0x10+2*ioapic_i8259.pin,
+ *(((int *)&entry)+0));
+ spin_unlock_irqrestore(&ioapic_lock, flags);
+ }
+ disconnect_bsp_APIC(ioapic_i8259.pin != -1);
+#endif
+}
+
+/*
+ * function to set the IO-APIC physical IDs based on the
+ * values stored in the MPC table.
+ *
+ * by Matt Domsch <Matt_Domsch@dell.com> Tue Dec 21 12:25:05 CST 1999
+ */
+
+#if !defined(CONFIG_XEN) && !defined(CONFIG_X86_NUMAQ)
+static void __init setup_ioapic_ids_from_mpc(void)
+{
+ union IO_APIC_reg_00 reg_00;
+ physid_mask_t phys_id_present_map;
+ int apic;
+ int i;
+ unsigned char old_id;
+ unsigned long flags;
+
+ /*
+ * Don't check I/O APIC IDs for xAPIC systems. They have
+ * no meaning without the serial APIC bus.
+ */
+ if (!(boot_cpu_data.x86_vendor == X86_VENDOR_INTEL)
+ || APIC_XAPIC(apic_version[boot_cpu_physical_apicid]))
+ return;
+ /*
+ * This is broken; anything with a real cpu count has to
+ * circumvent this idiocy regardless.
+ */
+ phys_id_present_map = ioapic_phys_id_map(phys_cpu_present_map);
+
+ /*
+ * Set the IOAPIC ID to the value stored in the MPC table.
+ */
+ for (apic = 0; apic < nr_ioapics; apic++) {
+
+ /* Read the register 0 value */
+ spin_lock_irqsave(&ioapic_lock, flags);
+ reg_00.raw = io_apic_read(apic, 0);
+ spin_unlock_irqrestore(&ioapic_lock, flags);
+
+ old_id = mp_ioapics[apic].mpc_apicid;
+
+ if (mp_ioapics[apic].mpc_apicid >= get_physical_broadcast()) {
+ printk(KERN_ERR "BIOS bug, IO-APIC#%d ID is %d in the MPC table!...\n",
+ apic, mp_ioapics[apic].mpc_apicid);
+ printk(KERN_ERR "... fixing up to %d. (tell your hw vendor)\n",
+ reg_00.bits.ID);
+ mp_ioapics[apic].mpc_apicid = reg_00.bits.ID;
+ }
+
+ /*
+ * Sanity check, is the ID really free? Every APIC in a
+ * system must have a unique ID or we get lots of nice
+ * 'stuck on smp_invalidate_needed IPI wait' messages.
+ */
+ if (check_apicid_used(phys_id_present_map,
+ mp_ioapics[apic].mpc_apicid)) {
+ printk(KERN_ERR "BIOS bug, IO-APIC#%d ID %d is already used!...\n",
+ apic, mp_ioapics[apic].mpc_apicid);
+ for (i = 0; i < get_physical_broadcast(); i++)
+ if (!physid_isset(i, phys_id_present_map))
+ break;
+ if (i >= get_physical_broadcast())
+ panic("Max APIC ID exceeded!\n");
+ printk(KERN_ERR "... fixing up to %d. (tell your hw vendor)\n",
+ i);
+ physid_set(i, phys_id_present_map);
+ mp_ioapics[apic].mpc_apicid = i;
+ } else {
+ physid_mask_t tmp;
+ tmp = apicid_to_cpu_present(mp_ioapics[apic].mpc_apicid);
+ apic_printk(APIC_VERBOSE, "Setting %d in the "
+ "phys_id_present_map\n",
+ mp_ioapics[apic].mpc_apicid);
+ physids_or(phys_id_present_map, phys_id_present_map, tmp);
+ }
+
+
+ /*
+ * We need to adjust the IRQ routing table
+ * if the ID changed.
+ */
+ if (old_id != mp_ioapics[apic].mpc_apicid)
+ for (i = 0; i < mp_irq_entries; i++)
+ if (mp_irqs[i].mpc_dstapic == old_id)
+ mp_irqs[i].mpc_dstapic
+ = mp_ioapics[apic].mpc_apicid;
+
+ /*
+ * Read the right value from the MPC table and
+ * write it into the ID register.
+ */
+ apic_printk(APIC_VERBOSE, KERN_INFO
+ "...changing IO-APIC physical APIC ID to %d ...",
+ mp_ioapics[apic].mpc_apicid);
+
+ reg_00.bits.ID = mp_ioapics[apic].mpc_apicid;
+ spin_lock_irqsave(&ioapic_lock, flags);
+ io_apic_write(apic, 0, reg_00.raw);
+ spin_unlock_irqrestore(&ioapic_lock, flags);
+
+ /*
+ * Sanity check
+ */
+ spin_lock_irqsave(&ioapic_lock, flags);
+ reg_00.raw = io_apic_read(apic, 0);
+ spin_unlock_irqrestore(&ioapic_lock, flags);
+ if (reg_00.bits.ID != mp_ioapics[apic].mpc_apicid)
+ printk("could not set ID!\n");
+ else
+ apic_printk(APIC_VERBOSE, " ok.\n");
+ }
+}
+#else
+static void __init setup_ioapic_ids_from_mpc(void) { }
+#endif
+
+#ifndef CONFIG_XEN
+/*
+ * There is a nasty bug in some older SMP boards, their mptable lies
+ * about the timer IRQ. We do the following to work around the situation:
+ *
+ * - timer IRQ defaults to IO-APIC IRQ
+ * - if this function detects that timer IRQs are defunct, then we fall
+ * back to ISA timer IRQs
+ */
+static int __init timer_irq_works(void)
+{
+ unsigned long t1 = jiffies;
+
+ local_irq_enable();
+ /* Let ten ticks pass... */
+ mdelay((10 * 1000) / HZ);
+
+ /*
+ * Expect a few ticks at least, to be sure some possible
+ * glue logic does not lock up after one or two first
+ * ticks in a non-ExtINT mode. Also the local APIC
+ * might have cached one ExtINT interrupt. Finally, at
+ * least one tick may be lost due to delays.
+ */
+ if (jiffies - t1 > 4)
+ return 1;
+
+ return 0;
+}
+
+/*
+ * In the SMP+IOAPIC case it might happen that there are an unspecified
+ * number of pending IRQ events unhandled. These cases are very rare,
+ * so we 'resend' these IRQs via IPIs, to the same CPU. It's much
+ * better to do it this way as thus we do not have to be aware of
+ * 'pending' interrupts in the IRQ path, except at this point.
+ */
+/*
+ * Edge triggered needs to resend any interrupt
+ * that was delayed but this is now handled in the device
+ * independent code.
+ */
+
+/*
+ * Starting up a edge-triggered IO-APIC interrupt is
+ * nasty - we need to make sure that we get the edge.
+ * If it is already asserted for some reason, we need
+ * return 1 to indicate that is was pending.
+ *
+ * This is not complete - we should be able to fake
+ * an edge even if it isn't on the 8259A...
+ */
+static unsigned int startup_edge_ioapic_irq(unsigned int irq)
+{
+ int was_pending = 0;
+ unsigned long flags;
+
+ spin_lock_irqsave(&ioapic_lock, flags);
+ if (irq < 16) {
+ disable_8259A_irq(irq);
+ if (i8259A_irq_pending(irq))
+ was_pending = 1;
+ }
+ __unmask_IO_APIC_irq(irq);
+ spin_unlock_irqrestore(&ioapic_lock, flags);
+
+ return was_pending;
+}
+
+/*
+ * Once we have recorded IRQ_PENDING already, we can mask the
+ * interrupt for real. This prevents IRQ storms from unhandled
+ * devices.
+ */
+static void ack_edge_ioapic_irq(unsigned int irq)
+{
+ move_irq(irq);
+ if ((irq_desc[irq].status & (IRQ_PENDING | IRQ_DISABLED))
+ == (IRQ_PENDING | IRQ_DISABLED))
+ mask_IO_APIC_irq(irq);
+ ack_APIC_irq();
+}
+
+/*
+ * Level triggered interrupts can just be masked,
+ * and shutting down and starting up the interrupt
+ * is the same as enabling and disabling them -- except
+ * with a startup need to return a "was pending" value.
+ *
+ * Level triggered interrupts are special because we
+ * do not touch any IO-APIC register while handling
+ * them. We ack the APIC in the end-IRQ handler, not
+ * in the start-IRQ-handler. Protection against reentrance
+ * from the same interrupt is still provided, both by the
+ * generic IRQ layer and by the fact that an unacked local
+ * APIC does not accept IRQs.
+ */
+static unsigned int startup_level_ioapic_irq (unsigned int irq)
+{
+ unmask_IO_APIC_irq(irq);
+
+ return 0; /* don't check for pending */
+}
+
+static void end_level_ioapic_irq (unsigned int irq)
+{
+ unsigned long v;
+ int i;
+
+ move_irq(irq);
+/*
+ * It appears there is an erratum which affects at least version 0x11
+ * of I/O APIC (that's the 82093AA and cores integrated into various
+ * chipsets). Under certain conditions a level-triggered interrupt is
+ * erroneously delivered as edge-triggered one but the respective IRR
+ * bit gets set nevertheless. As a result the I/O unit expects an EOI
+ * message but it will never arrive and further interrupts are blocked
+ * from the source. The exact reason is so far unknown, but the
+ * phenomenon was observed when two consecutive interrupt requests
+ * from a given source get delivered to the same CPU and the source is
+ * temporarily disabled in between.
+ *
+ * A workaround is to simulate an EOI message manually. We achieve it
+ * by setting the trigger mode to edge and then to level when the edge
+ * trigger mode gets detected in the TMR of a local APIC for a
+ * level-triggered interrupt. We mask the source for the time of the
+ * operation to prevent an edge-triggered interrupt escaping meanwhile.
+ * The idea is from Manfred Spraul. --macro
+ */
+ i = IO_APIC_VECTOR(irq);
+
+ v = apic_read(APIC_TMR + ((i & ~0x1f) >> 1));
+
+ ack_APIC_irq();
+
+ if (!(v & (1 << (i & 0x1f)))) {
+ atomic_inc(&irq_mis_count);
+ spin_lock(&ioapic_lock);
+ __mask_and_edge_IO_APIC_irq(irq);
+ __unmask_and_level_IO_APIC_irq(irq);
+ spin_unlock(&ioapic_lock);
+ }
+}
+
+#ifdef CONFIG_PCI_MSI
+static unsigned int startup_edge_ioapic_vector(unsigned int vector)
+{
+ int irq = vector_to_irq(vector);
+
+ return startup_edge_ioapic_irq(irq);
+}
+
+static void ack_edge_ioapic_vector(unsigned int vector)
+{
+ int irq = vector_to_irq(vector);
+
+ move_native_irq(vector);
+ ack_edge_ioapic_irq(irq);
+}
+
+static unsigned int startup_level_ioapic_vector (unsigned int vector)
+{
+ int irq = vector_to_irq(vector);
+
+ return startup_level_ioapic_irq (irq);
+}
+
+static void end_level_ioapic_vector (unsigned int vector)
+{
+ int irq = vector_to_irq(vector);
+
+ move_native_irq(vector);
+ end_level_ioapic_irq(irq);
+}
+
+static void mask_IO_APIC_vector (unsigned int vector)
+{
+ int irq = vector_to_irq(vector);
+
+ mask_IO_APIC_irq(irq);
+}
+
+static void unmask_IO_APIC_vector (unsigned int vector)
+{
+ int irq = vector_to_irq(vector);
+
+ unmask_IO_APIC_irq(irq);
+}
+
+#ifdef CONFIG_SMP
+static void set_ioapic_affinity_vector (unsigned int vector,
+ cpumask_t cpu_mask)
+{
+ int irq = vector_to_irq(vector);
+
+ set_native_irq_info(vector, cpu_mask);
+ set_ioapic_affinity_irq(irq, cpu_mask);
+}
+#endif
+#endif
+
+static int ioapic_retrigger(unsigned int irq)
+{
+ send_IPI_self(IO_APIC_VECTOR(irq));
+
+ return 1;
+}
+
+/*
+ * Level and edge triggered IO-APIC interrupts need different handling,
+ * so we use two separate IRQ descriptors. Edge triggered IRQs can be
+ * handled with the level-triggered descriptor, but that one has slightly
+ * more overhead. Level-triggered interrupts cannot be handled with the
+ * edge-triggered handler, without risking IRQ storms and other ugly
+ * races.
+ */
+static struct hw_interrupt_type ioapic_edge_type __read_mostly = {
+ .typename = "IO-APIC-edge",
+ .startup = startup_edge_ioapic,
+ .shutdown = shutdown_edge_ioapic,
+ .enable = enable_edge_ioapic,
+ .disable = disable_edge_ioapic,
+ .ack = ack_edge_ioapic,
+ .end = end_edge_ioapic,
+#ifdef CONFIG_SMP
+ .set_affinity = set_ioapic_affinity,
+#endif
+ .retrigger = ioapic_retrigger,
+};
+
+static struct hw_interrupt_type ioapic_level_type __read_mostly = {
+ .typename = "IO-APIC-level",
+ .startup = startup_level_ioapic,
+ .shutdown = shutdown_level_ioapic,
+ .enable = enable_level_ioapic,
+ .disable = disable_level_ioapic,
+ .ack = mask_and_ack_level_ioapic,
+ .end = end_level_ioapic,
+#ifdef CONFIG_SMP
+ .set_affinity = set_ioapic_affinity,
+#endif
+ .retrigger = ioapic_retrigger,
+};
+#endif /* !CONFIG_XEN */
+
+static inline void init_IO_APIC_traps(void)
+{
+ int irq;
+
+ /*
+ * NOTE! The local APIC isn't very good at handling
+ * multiple interrupts at the same interrupt level.
+ * As the interrupt level is determined by taking the
+ * vector number and shifting that right by 4, we
+ * want to spread these out a bit so that they don't
+ * all fall in the same interrupt level.
+ *
+ * Also, we've got to be careful not to trash gate
+ * 0x80, because int 0x80 is hm, kind of importantish. ;)
+ */
+ for (irq = 0; irq < NR_IRQS ; irq++) {
+ int tmp = irq;
+ if (use_pci_vector()) {
+ if (!platform_legacy_irq(tmp))
+ if ((tmp = vector_to_irq(tmp)) == -1)
+ continue;
+ }
+ if (IO_APIC_IRQ(tmp) && !IO_APIC_VECTOR(tmp)) {
+ /*
+ * Hmm.. We don't have an entry for this,
+ * so default to an old-fashioned 8259
+ * interrupt if we can..
+ */
+ if (irq < 16)
+ make_8259A_irq(irq);
+#ifndef CONFIG_XEN
+ else
+ /* Strange. Oh, well.. */
+ irq_desc[irq].chip = &no_irq_type;
+#endif
+ }
+ }
+}
+
+#ifndef CONFIG_XEN
+static void enable_lapic_irq (unsigned int irq)
+{
+ unsigned long v;
+
+ v = apic_read(APIC_LVT0);
+ apic_write_around(APIC_LVT0, v & ~APIC_LVT_MASKED);
+}
+
+static void disable_lapic_irq (unsigned int irq)
+{
+ unsigned long v;
+
+ v = apic_read(APIC_LVT0);
+ apic_write_around(APIC_LVT0, v | APIC_LVT_MASKED);
+}
+
+static void ack_lapic_irq (unsigned int irq)
+{
+ ack_APIC_irq();
+}
+
+static void end_lapic_irq (unsigned int i) { /* nothing */ }
+
+static struct hw_interrupt_type lapic_irq_type __read_mostly = {
+ .typename = "local-APIC-edge",
+ .startup = NULL, /* startup_irq() not used for IRQ0 */
+ .shutdown = NULL, /* shutdown_irq() not used for IRQ0 */
+ .enable = enable_lapic_irq,
+ .disable = disable_lapic_irq,
+ .ack = ack_lapic_irq,
+ .end = end_lapic_irq
+};
+
+static void setup_nmi (void)
+{
+ /*
+ * Dirty trick to enable the NMI watchdog ...
+ * We put the 8259A master into AEOI mode and
+ * unmask on all local APICs LVT0 as NMI.
+ *
+ * The idea to use the 8259A in AEOI mode ('8259A Virtual Wire')
+ * is from Maciej W. Rozycki - so we do not have to EOI from
+ * the NMI handler or the timer interrupt.
+ */
+ apic_printk(APIC_VERBOSE, KERN_INFO "activating NMI Watchdog ...");
+
+ on_each_cpu(enable_NMI_through_LVT0, NULL, 1, 1);
+
+ apic_printk(APIC_VERBOSE, " done.\n");
+}
+
+/*
+ * This looks a bit hackish but it's about the only one way of sending
+ * a few INTA cycles to 8259As and any associated glue logic. ICR does
+ * not support the ExtINT mode, unfortunately. We need to send these
+ * cycles as some i82489DX-based boards have glue logic that keeps the
+ * 8259A interrupt line asserted until INTA. --macro
+ */
+static inline void unlock_ExtINT_logic(void)
+{
+ int apic, pin, i;
+ struct IO_APIC_route_entry entry0, entry1;
+ unsigned char save_control, save_freq_select;
+ unsigned long flags;
+
+ pin = find_isa_irq_pin(8, mp_INT);
+ apic = find_isa_irq_apic(8, mp_INT);
+ if (pin == -1)
+ return;
+
+ spin_lock_irqsave(&ioapic_lock, flags);
+ *(((int *)&entry0) + 1) = io_apic_read(apic, 0x11 + 2 * pin);
+ *(((int *)&entry0) + 0) = io_apic_read(apic, 0x10 + 2 * pin);
+ spin_unlock_irqrestore(&ioapic_lock, flags);
+ clear_IO_APIC_pin(apic, pin);
+
+ memset(&entry1, 0, sizeof(entry1));
+
+ entry1.dest_mode = 0; /* physical delivery */
+ entry1.mask = 0; /* unmask IRQ now */
+ entry1.dest.physical.physical_dest = hard_smp_processor_id();
+ entry1.delivery_mode = dest_ExtINT;
+ entry1.polarity = entry0.polarity;
+ entry1.trigger = 0;
+ entry1.vector = 0;
+
+ spin_lock_irqsave(&ioapic_lock, flags);
+ io_apic_write(apic, 0x11 + 2 * pin, *(((int *)&entry1) + 1));
+ io_apic_write(apic, 0x10 + 2 * pin, *(((int *)&entry1) + 0));
+ spin_unlock_irqrestore(&ioapic_lock, flags);
+
+ save_control = CMOS_READ(RTC_CONTROL);
+ save_freq_select = CMOS_READ(RTC_FREQ_SELECT);
+ CMOS_WRITE((save_freq_select & ~RTC_RATE_SELECT) | 0x6,
+ RTC_FREQ_SELECT);
+ CMOS_WRITE(save_control | RTC_PIE, RTC_CONTROL);
+
+ i = 100;
+ while (i-- > 0) {
+ mdelay(10);
+ if ((CMOS_READ(RTC_INTR_FLAGS) & RTC_PF) == RTC_PF)
+ i -= 10;
+ }
+
+ CMOS_WRITE(save_control, RTC_CONTROL);
+ CMOS_WRITE(save_freq_select, RTC_FREQ_SELECT);
+ clear_IO_APIC_pin(apic, pin);
+
+ spin_lock_irqsave(&ioapic_lock, flags);
+ io_apic_write(apic, 0x11 + 2 * pin, *(((int *)&entry0) + 1));
+ io_apic_write(apic, 0x10 + 2 * pin, *(((int *)&entry0) + 0));
+ spin_unlock_irqrestore(&ioapic_lock, flags);
+}
+
+int timer_uses_ioapic_pin_0;
+
+/*
+ * This code may look a bit paranoid, but it's supposed to cooperate with
+ * a wide range of boards and BIOS bugs. Fortunately only the timer IRQ
+ * is so screwy. Thanks to Brian Perkins for testing/hacking this beast
+ * fanatically on his truly buggy board.
+ */
+static inline void check_timer(void)
+{
+ int apic1, pin1, apic2, pin2;
+ int vector;
+
+ /*
+ * get/set the timer IRQ vector:
+ */
+ disable_8259A_irq(0);
+ vector = assign_irq_vector(0);
+ set_intr_gate(vector, interrupt[0]);
+
+ /*
+ * Subtle, code in do_timer_interrupt() expects an AEOI
+ * mode for the 8259A whenever interrupts are routed
+ * through I/O APICs. Also IRQ0 has to be enabled in
+ * the 8259A which implies the virtual wire has to be
+ * disabled in the local APIC.
+ */
+ apic_write_around(APIC_LVT0, APIC_LVT_MASKED | APIC_DM_EXTINT);
+ init_8259A(1);
+ timer_ack = 1;
+ if (timer_over_8254 > 0)
+ enable_8259A_irq(0);
+
+ pin1 = find_isa_irq_pin(0, mp_INT);
+ apic1 = find_isa_irq_apic(0, mp_INT);
+ pin2 = ioapic_i8259.pin;
+ apic2 = ioapic_i8259.apic;
+
+ if (pin1 == 0)
+ timer_uses_ioapic_pin_0 = 1;
+
+ printk(KERN_INFO "..TIMER: vector=0x%02X apic1=%d pin1=%d apic2=%d pin2=%d\n",
+ vector, apic1, pin1, apic2, pin2);
+
+ if (pin1 != -1) {
+ /*
+ * Ok, does IRQ0 through the IOAPIC work?
+ */
+ unmask_IO_APIC_irq(0);
+ if (timer_irq_works()) {
+ if (nmi_watchdog == NMI_IO_APIC) {
+ disable_8259A_irq(0);
+ setup_nmi();
+ enable_8259A_irq(0);
+ }
+ if (disable_timer_pin_1 > 0)
+ clear_IO_APIC_pin(0, pin1);
+ return;
+ }
+ clear_IO_APIC_pin(apic1, pin1);
+ printk(KERN_ERR "..MP-BIOS bug: 8254 timer not connected to "
+ "IO-APIC\n");
+ }
+
+ printk(KERN_INFO "...trying to set up timer (IRQ0) through the 8259A ... ");
+ if (pin2 != -1) {
+ printk("\n..... (found pin %d) ...", pin2);
+ /*
+ * legacy devices should be connected to IO APIC #0
+ */
+ setup_ExtINT_IRQ0_pin(apic2, pin2, vector);
+ if (timer_irq_works()) {
+ printk("works.\n");
+ if (pin1 != -1)
+ replace_pin_at_irq(0, apic1, pin1, apic2, pin2);
+ else
+ add_pin_to_irq(0, apic2, pin2);
+ if (nmi_watchdog == NMI_IO_APIC) {
+ setup_nmi();
+ }
+ return;
+ }
+ /*
+ * Cleanup, just in case ...
+ */
+ clear_IO_APIC_pin(apic2, pin2);
+ }
+ printk(" failed.\n");
+
+ if (nmi_watchdog == NMI_IO_APIC) {
+ printk(KERN_WARNING "timer doesn't work through the IO-APIC - disabling NMI Watchdog!\n");
+ nmi_watchdog = 0;
+ }
+
+ printk(KERN_INFO "...trying to set up timer as Virtual Wire IRQ...");
+
+ disable_8259A_irq(0);
+ irq_desc[0].chip = &lapic_irq_type;
+ apic_write_around(APIC_LVT0, APIC_DM_FIXED | vector); /* Fixed mode */
+ enable_8259A_irq(0);
+
+ if (timer_irq_works()) {
+ printk(" works.\n");
+ return;
+ }
+ apic_write_around(APIC_LVT0, APIC_LVT_MASKED | APIC_DM_FIXED | vector);
+ printk(" failed.\n");
+
+ printk(KERN_INFO "...trying to set up timer as ExtINT IRQ...");
+
+ timer_ack = 0;
+ init_8259A(0);
+ make_8259A_irq(0);
+ apic_write_around(APIC_LVT0, APIC_DM_EXTINT);
+
+ unlock_ExtINT_logic();
+
+ if (timer_irq_works()) {
+ printk(" works.\n");
+ return;
+ }
+ printk(" failed :(.\n");
+ panic("IO-APIC + timer doesn't work! Boot with apic=debug and send a "
+ "report. Then try booting with the 'noapic' option");
+}
+#else
+int timer_uses_ioapic_pin_0 = 0;
+#define check_timer() ((void)0)
+#endif
+
+/*
+ *
+ * IRQ's that are handled by the PIC in the MPS IOAPIC case.
+ * - IRQ2 is the cascade IRQ, and cannot be a io-apic IRQ.
+ * Linux doesn't really care, as it's not actually used
+ * for any interrupt handling anyway.
+ */
+#define PIC_IRQS (1 << PIC_CASCADE_IR)
+
+void __init setup_IO_APIC(void)
+{
+ enable_IO_APIC();
+
+ if (acpi_ioapic)
+ io_apic_irqs = ~0; /* all IRQs go through IOAPIC */
+ else
+ io_apic_irqs = ~PIC_IRQS;
+
+ printk("ENABLING IO-APIC IRQs\n");
+
+ /*
+ * Set up IO-APIC IRQ routing.
+ */
+ if (!acpi_ioapic)
+ setup_ioapic_ids_from_mpc();
+#ifndef CONFIG_XEN
+ sync_Arb_IDs();
+#endif
+ setup_IO_APIC_irqs();
+ init_IO_APIC_traps();
+ check_timer();
+ if (!acpi_ioapic)
+ print_IO_APIC();
+}
+
+static int __init setup_disable_8254_timer(char *s)
+{
+ timer_over_8254 = -1;
+ return 1;
+}
+static int __init setup_enable_8254_timer(char *s)
+{
+ timer_over_8254 = 2;
+ return 1;
+}
+
+__setup("disable_8254_timer", setup_disable_8254_timer);
+__setup("enable_8254_timer", setup_enable_8254_timer);
+
+/*
+ * Called after all the initialization is done. If we didnt find any
+ * APIC bugs then we can allow the modify fast path
+ */
+
+static int __init io_apic_bug_finalize(void)
+{
+ if(sis_apic_bug == -1)
+ sis_apic_bug = 0;
+ if (is_initial_xendomain()) {
+ struct xen_platform_op op = { .cmd = XENPF_platform_quirk };
+ op.u.platform_quirk.quirk_id = sis_apic_bug ?
+ QUIRK_IOAPIC_BAD_REGSEL : QUIRK_IOAPIC_GOOD_REGSEL;
+ VOID(HYPERVISOR_platform_op(&op));
+ }
+ return 0;
+}
+
+late_initcall(io_apic_bug_finalize);
+
+#ifndef CONFIG_XEN
+
+struct sysfs_ioapic_data {
+ struct sys_device dev;
+ struct IO_APIC_route_entry entry[0];
+};
+static struct sysfs_ioapic_data * mp_ioapic_data[MAX_IO_APICS];
+
+static int ioapic_suspend(struct sys_device *dev, pm_message_t state)
+{
+ struct IO_APIC_route_entry *entry;
+ struct sysfs_ioapic_data *data;
+ unsigned long flags;
+ int i;
+
+ data = container_of(dev, struct sysfs_ioapic_data, dev);
+ entry = data->entry;
+ spin_lock_irqsave(&ioapic_lock, flags);
+ for (i = 0; i < nr_ioapic_registers[dev->id]; i ++, entry ++ ) {
+ *(((int *)entry) + 1) = io_apic_read(dev->id, 0x11 + 2 * i);
+ *(((int *)entry) + 0) = io_apic_read(dev->id, 0x10 + 2 * i);
+ }
+ spin_unlock_irqrestore(&ioapic_lock, flags);
+
+ return 0;
+}
+
+static int ioapic_resume(struct sys_device *dev)
+{
+ struct IO_APIC_route_entry *entry;
+ struct sysfs_ioapic_data *data;
+ unsigned long flags;
+ union IO_APIC_reg_00 reg_00;
+ int i;
+
+ data = container_of(dev, struct sysfs_ioapic_data, dev);
+ entry = data->entry;
+
+ spin_lock_irqsave(&ioapic_lock, flags);
+ reg_00.raw = io_apic_read(dev->id, 0);
+ if (reg_00.bits.ID != mp_ioapics[dev->id].mpc_apicid) {
+ reg_00.bits.ID = mp_ioapics[dev->id].mpc_apicid;
+ io_apic_write(dev->id, 0, reg_00.raw);
+ }
+ for (i = 0; i < nr_ioapic_registers[dev->id]; i ++, entry ++ ) {
+ io_apic_write(dev->id, 0x11+2*i, *(((int *)entry)+1));
+ io_apic_write(dev->id, 0x10+2*i, *(((int *)entry)+0));
+ }
+ spin_unlock_irqrestore(&ioapic_lock, flags);
+
+ return 0;
+}
+
+static struct sysdev_class ioapic_sysdev_class = {
+ set_kset_name("ioapic"),
+ .suspend = ioapic_suspend,
+ .resume = ioapic_resume,
+};
+
+static int __init ioapic_init_sysfs(void)
+{
+ struct sys_device * dev;
+ int i, size, error = 0;
+
+ error = sysdev_class_register(&ioapic_sysdev_class);
+ if (error)
+ return error;
+
+ for (i = 0; i < nr_ioapics; i++ ) {
+ size = sizeof(struct sys_device) + nr_ioapic_registers[i]
+ * sizeof(struct IO_APIC_route_entry);
+ mp_ioapic_data[i] = kmalloc(size, GFP_KERNEL);
+ if (!mp_ioapic_data[i]) {
+ printk(KERN_ERR "Can't suspend/resume IOAPIC %d\n", i);
+ continue;
+ }
+ memset(mp_ioapic_data[i], 0, size);
+ dev = &mp_ioapic_data[i]->dev;
+ dev->id = i;
+ dev->cls = &ioapic_sysdev_class;
+ error = sysdev_register(dev);
+ if (error) {
+ kfree(mp_ioapic_data[i]);
+ mp_ioapic_data[i] = NULL;
+ printk(KERN_ERR "Can't suspend/resume IOAPIC %d\n", i);
+ continue;
+ }
+ }
+
+ return 0;
+}
+
+device_initcall(ioapic_init_sysfs);
+
+#endif /* CONFIG_XEN */
+
+/* --------------------------------------------------------------------------
+ ACPI-based IOAPIC Configuration
+ -------------------------------------------------------------------------- */
+
+#ifdef CONFIG_ACPI
+
+int __init io_apic_get_unique_id (int ioapic, int apic_id)
+{
+#ifndef CONFIG_XEN
+ union IO_APIC_reg_00 reg_00;
+ static physid_mask_t apic_id_map = PHYSID_MASK_NONE;
+ physid_mask_t tmp;
+ unsigned long flags;
+ int i = 0;
+
+ /*
+ * The P4 platform supports up to 256 APIC IDs on two separate APIC
+ * buses (one for LAPICs, one for IOAPICs), where predecessors only
+ * supports up to 16 on one shared APIC bus.
+ *
+ * TBD: Expand LAPIC/IOAPIC support on P4-class systems to take full
+ * advantage of new APIC bus architecture.
+ */
+
+ if (physids_empty(apic_id_map))
+ apic_id_map = ioapic_phys_id_map(phys_cpu_present_map);
+
+ spin_lock_irqsave(&ioapic_lock, flags);
+ reg_00.raw = io_apic_read(ioapic, 0);
+ spin_unlock_irqrestore(&ioapic_lock, flags);
+
+ if (apic_id >= get_physical_broadcast()) {
+ printk(KERN_WARNING "IOAPIC[%d]: Invalid apic_id %d, trying "
+ "%d\n", ioapic, apic_id, reg_00.bits.ID);
+ apic_id = reg_00.bits.ID;
+ }
+
+ /*
+ * Every APIC in a system must have a unique ID or we get lots of nice
+ * 'stuck on smp_invalidate_needed IPI wait' messages.
+ */
+ if (check_apicid_used(apic_id_map, apic_id)) {
+
+ for (i = 0; i < get_physical_broadcast(); i++) {
+ if (!check_apicid_used(apic_id_map, i))
+ break;
+ }
+
+ if (i == get_physical_broadcast())
+ panic("Max apic_id exceeded!\n");
+
+ printk(KERN_WARNING "IOAPIC[%d]: apic_id %d already used, "
+ "trying %d\n", ioapic, apic_id, i);
+
+ apic_id = i;
+ }
+
+ tmp = apicid_to_cpu_present(apic_id);
+ physids_or(apic_id_map, apic_id_map, tmp);
+
+ if (reg_00.bits.ID != apic_id) {
+ reg_00.bits.ID = apic_id;
+
+ spin_lock_irqsave(&ioapic_lock, flags);
+ io_apic_write(ioapic, 0, reg_00.raw);
+ reg_00.raw = io_apic_read(ioapic, 0);
+ spin_unlock_irqrestore(&ioapic_lock, flags);
+
+ /* Sanity check */
+ if (reg_00.bits.ID != apic_id) {
+ printk("IOAPIC[%d]: Unable to change apic_id!\n", ioapic);
+ return -1;
+ }
+ }
+
+ apic_printk(APIC_VERBOSE, KERN_INFO
+ "IOAPIC[%d]: Assigned apic_id %d\n", ioapic, apic_id);
+#endif /* !CONFIG_XEN */
+
+ return apic_id;
+}
+
+
+int __init io_apic_get_version (int ioapic)
+{
+ union IO_APIC_reg_01 reg_01;
+ unsigned long flags;
+
+ spin_lock_irqsave(&ioapic_lock, flags);
+ reg_01.raw = io_apic_read(ioapic, 1);
+ spin_unlock_irqrestore(&ioapic_lock, flags);
+
+ return reg_01.bits.version;
+}
+
+
+int __init io_apic_get_redir_entries (int ioapic)
+{
+ union IO_APIC_reg_01 reg_01;
+ unsigned long flags;
+
+ spin_lock_irqsave(&ioapic_lock, flags);
+ reg_01.raw = io_apic_read(ioapic, 1);
+ spin_unlock_irqrestore(&ioapic_lock, flags);
+
+ return reg_01.bits.entries;
+}
+
+
+int io_apic_set_pci_routing (int ioapic, int pin, int irq, int edge_level, int active_high_low)
+{
+ struct IO_APIC_route_entry entry;
+ unsigned long flags;
+
+ if (!IO_APIC_IRQ(irq)) {
+ printk(KERN_ERR "IOAPIC[%d]: Invalid reference to IRQ 0\n",
+ ioapic);
+ return -EINVAL;
+ }
+
+ /*
+ * Generate a PCI IRQ routing entry and program the IOAPIC accordingly.
+ * Note that we mask (disable) IRQs now -- these get enabled when the
+ * corresponding device driver registers for this IRQ.
+ */
+
+ memset(&entry,0,sizeof(entry));
+
+ entry.delivery_mode = INT_DELIVERY_MODE;
+ entry.dest_mode = INT_DEST_MODE;
+ entry.dest.logical.logical_dest = cpu_mask_to_apicid(TARGET_CPUS);
+ entry.trigger = edge_level;
+ entry.polarity = active_high_low;
+ entry.mask = 1;
+
+ /*
+ * IRQs < 16 are already in the irq_2_pin[] map
+ */
+ if (irq >= 16)
+ add_pin_to_irq(irq, ioapic, pin);
+
+ entry.vector = assign_irq_vector(irq);
+
+ apic_printk(APIC_DEBUG, KERN_DEBUG "IOAPIC[%d]: Set PCI routing entry "
+ "(%d-%d -> 0x%x -> IRQ %d Mode:%i Active:%i)\n", ioapic,
+ mp_ioapics[ioapic].mpc_apicid, pin, entry.vector, irq,
+ edge_level, active_high_low);
+
+ ioapic_register_intr(irq, entry.vector, edge_level);
+
+ if (!ioapic && (irq < 16))
+ disable_8259A_irq(irq);
+
+ spin_lock_irqsave(&ioapic_lock, flags);
+ io_apic_write(ioapic, 0x11+2*pin, *(((int *)&entry)+1));
+ io_apic_write(ioapic, 0x10+2*pin, *(((int *)&entry)+0));
+ set_native_irq_info(use_pci_vector() ? entry.vector : irq, TARGET_CPUS);
+ spin_unlock_irqrestore(&ioapic_lock, flags);
+
+ return 0;
+}
+
+#endif /* CONFIG_ACPI */
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/kernel/ioport_32-xen.c 2008-01-28 12:24:19.000000000 +0100
@@ -0,0 +1,123 @@
+/*
+ * linux/arch/i386/kernel/ioport.c
+ *
+ * This contains the io-permission bitmap code - written by obz, with changes
+ * by Linus.
+ */
+
+#include <linux/sched.h>
+#include <linux/kernel.h>
+#include <linux/capability.h>
+#include <linux/errno.h>
+#include <linux/types.h>
+#include <linux/ioport.h>
+#include <linux/smp.h>
+#include <linux/smp_lock.h>
+#include <linux/stddef.h>
+#include <linux/slab.h>
+#include <linux/thread_info.h>
+#include <xen/interface/physdev.h>
+
+/* Set EXTENT bits starting at BASE in BITMAP to value TURN_ON. */
+static void set_bitmap(unsigned long *bitmap, unsigned int base, unsigned int extent, int new_value)
+{
+ unsigned long mask;
+ unsigned long *bitmap_base = bitmap + (base / BITS_PER_LONG);
+ unsigned int low_index = base & (BITS_PER_LONG-1);
+ int length = low_index + extent;
+
+ if (low_index != 0) {
+ mask = (~0UL << low_index);
+ if (length < BITS_PER_LONG)
+ mask &= ~(~0UL << length);
+ if (new_value)
+ *bitmap_base++ |= mask;
+ else
+ *bitmap_base++ &= ~mask;
+ length -= BITS_PER_LONG;
+ }
+
+ mask = (new_value ? ~0UL : 0UL);
+ while (length >= BITS_PER_LONG) {
+ *bitmap_base++ = mask;
+ length -= BITS_PER_LONG;
+ }
+
+ if (length > 0) {
+ mask = ~(~0UL << length);
+ if (new_value)
+ *bitmap_base++ |= mask;
+ else
+ *bitmap_base++ &= ~mask;
+ }
+}
+
+
+/*
+ * this changes the io permissions bitmap in the current task.
+ */
+asmlinkage long sys_ioperm(unsigned long from, unsigned long num, int turn_on)
+{
+ struct thread_struct * t = &current->thread;
+ unsigned long *bitmap;
+ struct physdev_set_iobitmap set_iobitmap;
+
+ if ((from + num <= from) || (from + num > IO_BITMAP_BITS))
+ return -EINVAL;
+ if (turn_on && !capable(CAP_SYS_RAWIO))
+ return -EPERM;
+
+ /*
+ * If it's the first ioperm() call in this thread's lifetime, set the
+ * IO bitmap up. ioperm() is much less timing critical than clone(),
+ * this is why we delay this operation until now:
+ */
+ if (!t->io_bitmap_ptr) {
+ bitmap = kmalloc(IO_BITMAP_BYTES, GFP_KERNEL);
+ if (!bitmap)
+ return -ENOMEM;
+
+ memset(bitmap, 0xff, IO_BITMAP_BYTES);
+ t->io_bitmap_ptr = bitmap;
+ set_thread_flag(TIF_IO_BITMAP);
+
+ set_xen_guest_handle(set_iobitmap.bitmap, (char *)bitmap);
+ set_iobitmap.nr_ports = IO_BITMAP_BITS;
+ WARN_ON(HYPERVISOR_physdev_op(PHYSDEVOP_set_iobitmap,
+ &set_iobitmap));
+ }
+
+ set_bitmap(t->io_bitmap_ptr, from, num, !turn_on);
+
+ return 0;
+}
+
+/*
+ * sys_iopl has to be used when you want to access the IO ports
+ * beyond the 0x3ff range: to get the full 65536 ports bitmapped
+ * you'd need 8kB of bitmaps/process, which is a bit excessive.
+ *
+ * Here we just change the eflags value on the stack: we allow
+ * only the super-user to do it. This depends on the stack-layout
+ * on system-call entry - see also fork() and the signal handling
+ * code.
+ */
+
+asmlinkage long sys_iopl(unsigned long unused)
+{
+ volatile struct pt_regs * regs = (struct pt_regs *) &unused;
+ unsigned int level = regs->ebx;
+ struct thread_struct *t = &current->thread;
+ unsigned int old = (t->iopl >> 12) & 3;
+
+ if (level > 3)
+ return -EINVAL;
+ /* Trying to gain more privileges? */
+ if (level > old) {
+ if (!capable(CAP_SYS_RAWIO))
+ return -EPERM;
+ }
+ t->iopl = level << 12;
+ set_iopl_mask(t->iopl);
+ return 0;
+}
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/kernel/ldt_32-xen.c 2007-06-12 13:12:48.000000000 +0200
@@ -0,0 +1,270 @@
+/*
+ * linux/kernel/ldt.c
+ *
+ * Copyright (C) 1992 Krishna Balasubramanian and Linus Torvalds
+ * Copyright (C) 1999 Ingo Molnar <mingo@redhat.com>
+ */
+
+#include <linux/errno.h>
+#include <linux/sched.h>
+#include <linux/string.h>
+#include <linux/mm.h>
+#include <linux/smp.h>
+#include <linux/smp_lock.h>
+#include <linux/vmalloc.h>
+#include <linux/slab.h>
+
+#include <asm/uaccess.h>
+#include <asm/system.h>
+#include <asm/ldt.h>
+#include <asm/desc.h>
+#include <asm/mmu_context.h>
+
+#ifdef CONFIG_SMP /* avoids "defined but not used" warnig */
+static void flush_ldt(void *null)
+{
+ if (current->active_mm)
+ load_LDT(&current->active_mm->context);
+}
+#endif
+
+static int alloc_ldt(mm_context_t *pc, int mincount, int reload)
+{
+ void *oldldt;
+ void *newldt;
+ int oldsize;
+
+ if (mincount <= pc->size)
+ return 0;
+ oldsize = pc->size;
+ mincount = (mincount+511)&(~511);
+ if (mincount*LDT_ENTRY_SIZE > PAGE_SIZE)
+ newldt = vmalloc(mincount*LDT_ENTRY_SIZE);
+ else
+ newldt = kmalloc(mincount*LDT_ENTRY_SIZE, GFP_KERNEL);
+
+ if (!newldt)
+ return -ENOMEM;
+
+ if (oldsize)
+ memcpy(newldt, pc->ldt, oldsize*LDT_ENTRY_SIZE);
+ oldldt = pc->ldt;
+ memset(newldt+oldsize*LDT_ENTRY_SIZE, 0, (mincount-oldsize)*LDT_ENTRY_SIZE);
+ pc->ldt = newldt;
+ wmb();
+ pc->size = mincount;
+ wmb();
+
+ if (reload) {
+#ifdef CONFIG_SMP
+ cpumask_t mask;
+ preempt_disable();
+#endif
+ make_pages_readonly(
+ pc->ldt,
+ (pc->size * LDT_ENTRY_SIZE) / PAGE_SIZE,
+ XENFEAT_writable_descriptor_tables);
+ load_LDT(pc);
+#ifdef CONFIG_SMP
+ mask = cpumask_of_cpu(smp_processor_id());
+ if (!cpus_equal(current->mm->cpu_vm_mask, mask))
+ smp_call_function(flush_ldt, NULL, 1, 1);
+ preempt_enable();
+#endif
+ }
+ if (oldsize) {
+ make_pages_writable(
+ oldldt,
+ (oldsize * LDT_ENTRY_SIZE) / PAGE_SIZE,
+ XENFEAT_writable_descriptor_tables);
+ if (oldsize*LDT_ENTRY_SIZE > PAGE_SIZE)
+ vfree(oldldt);
+ else
+ kfree(oldldt);
+ }
+ return 0;
+}
+
+static inline int copy_ldt(mm_context_t *new, mm_context_t *old)
+{
+ int err = alloc_ldt(new, old->size, 0);
+ if (err < 0)
+ return err;
+ memcpy(new->ldt, old->ldt, old->size*LDT_ENTRY_SIZE);
+ make_pages_readonly(
+ new->ldt,
+ (new->size * LDT_ENTRY_SIZE) / PAGE_SIZE,
+ XENFEAT_writable_descriptor_tables);
+ return 0;
+}
+
+/*
+ * we do not have to muck with descriptors here, that is
+ * done in switch_mm() as needed.
+ */
+int init_new_context(struct task_struct *tsk, struct mm_struct *mm)
+{
+ struct mm_struct * old_mm;
+ int retval = 0;
+
+ init_MUTEX(&mm->context.sem);
+ mm->context.size = 0;
+ mm->context.has_foreign_mappings = 0;
+ old_mm = current->mm;
+ if (old_mm && old_mm->context.size > 0) {
+ down(&old_mm->context.sem);
+ retval = copy_ldt(&mm->context, &old_mm->context);
+ up(&old_mm->context.sem);
+ }
+ return retval;
+}
+
+/*
+ * No need to lock the MM as we are the last user
+ */
+void destroy_context(struct mm_struct *mm)
+{
+ if (mm->context.size) {
+ if (mm == current->active_mm)
+ clear_LDT();
+ make_pages_writable(
+ mm->context.ldt,
+ (mm->context.size * LDT_ENTRY_SIZE) / PAGE_SIZE,
+ XENFEAT_writable_descriptor_tables);
+ if (mm->context.size*LDT_ENTRY_SIZE > PAGE_SIZE)
+ vfree(mm->context.ldt);
+ else
+ kfree(mm->context.ldt);
+ mm->context.size = 0;
+ }
+}
+
+static int read_ldt(void __user * ptr, unsigned long bytecount)
+{
+ int err;
+ unsigned long size;
+ struct mm_struct * mm = current->mm;
+
+ if (!mm->context.size)
+ return 0;
+ if (bytecount > LDT_ENTRY_SIZE*LDT_ENTRIES)
+ bytecount = LDT_ENTRY_SIZE*LDT_ENTRIES;
+
+ down(&mm->context.sem);
+ size = mm->context.size*LDT_ENTRY_SIZE;
+ if (size > bytecount)
+ size = bytecount;
+
+ err = 0;
+ if (copy_to_user(ptr, mm->context.ldt, size))
+ err = -EFAULT;
+ up(&mm->context.sem);
+ if (err < 0)
+ goto error_return;
+ if (size != bytecount) {
+ /* zero-fill the rest */
+ if (clear_user(ptr+size, bytecount-size) != 0) {
+ err = -EFAULT;
+ goto error_return;
+ }
+ }
+ return bytecount;
+error_return:
+ return err;
+}
+
+static int read_default_ldt(void __user * ptr, unsigned long bytecount)
+{
+ int err;
+ unsigned long size;
+ void *address;
+
+ err = 0;
+ address = &default_ldt[0];
+ size = 5*sizeof(struct desc_struct);
+ if (size > bytecount)
+ size = bytecount;
+
+ err = size;
+ if (copy_to_user(ptr, address, size))
+ err = -EFAULT;
+
+ return err;
+}
+
+static int write_ldt(void __user * ptr, unsigned long bytecount, int oldmode)
+{
+ struct mm_struct * mm = current->mm;
+ __u32 entry_1, entry_2;
+ int error;
+ struct user_desc ldt_info;
+
+ error = -EINVAL;
+ if (bytecount != sizeof(ldt_info))
+ goto out;
+ error = -EFAULT;
+ if (copy_from_user(&ldt_info, ptr, sizeof(ldt_info)))
+ goto out;
+
+ error = -EINVAL;
+ if (ldt_info.entry_number >= LDT_ENTRIES)
+ goto out;
+ if (ldt_info.contents == 3) {
+ if (oldmode)
+ goto out;
+ if (ldt_info.seg_not_present == 0)
+ goto out;
+ }
+
+ down(&mm->context.sem);
+ if (ldt_info.entry_number >= mm->context.size) {
+ error = alloc_ldt(&current->mm->context, ldt_info.entry_number+1, 1);
+ if (error < 0)
+ goto out_unlock;
+ }
+
+ /* Allow LDTs to be cleared by the user. */
+ if (ldt_info.base_addr == 0 && ldt_info.limit == 0) {
+ if (oldmode || LDT_empty(&ldt_info)) {
+ entry_1 = 0;
+ entry_2 = 0;
+ goto install;
+ }
+ }
+
+ entry_1 = LDT_entry_a(&ldt_info);
+ entry_2 = LDT_entry_b(&ldt_info);
+ if (oldmode)
+ entry_2 &= ~(1 << 20);
+
+ /* Install the new entry ... */
+install:
+ error = write_ldt_entry(mm->context.ldt, ldt_info.entry_number,
+ entry_1, entry_2);
+
+out_unlock:
+ up(&mm->context.sem);
+out:
+ return error;
+}
+
+asmlinkage int sys_modify_ldt(int func, void __user *ptr, unsigned long bytecount)
+{
+ int ret = -ENOSYS;
+
+ switch (func) {
+ case 0:
+ ret = read_ldt(ptr, bytecount);
+ break;
+ case 1:
+ ret = write_ldt(ptr, bytecount, 1);
+ break;
+ case 2:
+ ret = read_default_ldt(ptr, bytecount);
+ break;
+ case 0x11:
+ ret = write_ldt(ptr, bytecount, 0);
+ break;
+ }
+ return ret;
+}
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/kernel/microcode-xen.c 2007-06-12 13:12:48.000000000 +0200
@@ -0,0 +1,144 @@
+/*
+ * Intel CPU Microcode Update Driver for Linux
+ *
+ * Copyright (C) 2000-2004 Tigran Aivazian
+ *
+ * This driver allows to upgrade microcode on Intel processors
+ * belonging to IA-32 family - PentiumPro, Pentium II,
+ * Pentium III, Xeon, Pentium 4, etc.
+ *
+ * Reference: Section 8.10 of Volume III, Intel Pentium 4 Manual,
+ * Order Number 245472 or free download from:
+ *
+ * http://developer.intel.com/design/pentium4/manuals/245472.htm
+ *
+ * For more information, go to http://www.urbanmyth.org/microcode
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ */
+
+//#define DEBUG /* pr_debug */
+#include <linux/capability.h>
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/sched.h>
+#include <linux/cpumask.h>
+#include <linux/module.h>
+#include <linux/slab.h>
+#include <linux/vmalloc.h>
+#include <linux/miscdevice.h>
+#include <linux/spinlock.h>
+#include <linux/mm.h>
+#include <linux/mutex.h>
+#include <linux/syscalls.h>
+
+#include <asm/msr.h>
+#include <asm/uaccess.h>
+#include <asm/processor.h>
+
+MODULE_DESCRIPTION("Intel CPU (IA-32) Microcode Update Driver");
+MODULE_AUTHOR("Tigran Aivazian <tigran@veritas.com>");
+MODULE_LICENSE("GPL");
+
+static int verbose;
+module_param(verbose, int, 0644);
+
+#define MICROCODE_VERSION "1.14a-xen"
+
+#define DEFAULT_UCODE_DATASIZE (2000) /* 2000 bytes */
+#define MC_HEADER_SIZE (sizeof (microcode_header_t)) /* 48 bytes */
+#define DEFAULT_UCODE_TOTALSIZE (DEFAULT_UCODE_DATASIZE + MC_HEADER_SIZE) /* 2048 bytes */
+
+/* no concurrent ->write()s are allowed on /dev/cpu/microcode */
+static DEFINE_MUTEX(microcode_mutex);
+
+static int microcode_open (struct inode *unused1, struct file *unused2)
+{
+ return capable(CAP_SYS_RAWIO) ? 0 : -EPERM;
+}
+
+
+static int do_microcode_update (const void __user *ubuf, size_t len)
+{
+ int err;
+ void *kbuf;
+
+ kbuf = vmalloc(len);
+ if (!kbuf)
+ return -ENOMEM;
+
+ if (copy_from_user(kbuf, ubuf, len) == 0) {
+ struct xen_platform_op op;
+
+ op.cmd = XENPF_microcode_update;
+ set_xen_guest_handle(op.u.microcode.data, kbuf);
+ op.u.microcode.length = len;
+ err = HYPERVISOR_platform_op(&op);
+ } else
+ err = -EFAULT;
+
+ vfree(kbuf);
+
+ return err;
+}
+
+static ssize_t microcode_write (struct file *file, const char __user *buf, size_t len, loff_t *ppos)
+{
+ ssize_t ret;
+
+ if (len < MC_HEADER_SIZE) {
+ printk(KERN_ERR "microcode: not enough data\n");
+ return -EINVAL;
+ }
+
+ mutex_lock(&microcode_mutex);
+
+ ret = do_microcode_update(buf, len);
+ if (!ret)
+ ret = (ssize_t)len;
+
+ mutex_unlock(&microcode_mutex);
+
+ return ret;
+}
+
+static struct file_operations microcode_fops = {
+ .owner = THIS_MODULE,
+ .write = microcode_write,
+ .open = microcode_open,
+};
+
+static struct miscdevice microcode_dev = {
+ .minor = MICROCODE_MINOR,
+ .name = "microcode",
+ .fops = &microcode_fops,
+};
+
+static int __init microcode_init (void)
+{
+ int error;
+
+ error = misc_register(&microcode_dev);
+ if (error) {
+ printk(KERN_ERR
+ "microcode: can't misc_register on minor=%d\n",
+ MICROCODE_MINOR);
+ return error;
+ }
+
+ printk(KERN_INFO
+ "IA-32 Microcode Update Driver: v" MICROCODE_VERSION " <tigran@veritas.com>\n");
+ return 0;
+}
+
+static void __exit microcode_exit (void)
+{
+ misc_deregister(&microcode_dev);
+}
+
+module_init(microcode_init)
+module_exit(microcode_exit)
+MODULE_ALIAS_MISCDEV(MICROCODE_MINOR);
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/kernel/mpparse_32-xen.c 2007-06-12 13:12:48.000000000 +0200
@@ -0,0 +1,1185 @@
+/*
+ * Intel Multiprocessor Specification 1.1 and 1.4
+ * compliant MP-table parsing routines.
+ *
+ * (c) 1995 Alan Cox, Building #3 <alan@redhat.com>
+ * (c) 1998, 1999, 2000 Ingo Molnar <mingo@redhat.com>
+ *
+ * Fixes
+ * Erich Boleyn : MP v1.4 and additional changes.
+ * Alan Cox : Added EBDA scanning
+ * Ingo Molnar : various cleanups and rewrites
+ * Maciej W. Rozycki: Bits for default MP configurations
+ * Paul Diefenbaugh: Added full ACPI support
+ */
+
+#include <linux/mm.h>
+#include <linux/init.h>
+#include <linux/acpi.h>
+#include <linux/delay.h>
+#include <linux/bootmem.h>
+#include <linux/smp_lock.h>
+#include <linux/kernel_stat.h>
+#include <linux/mc146818rtc.h>
+#include <linux/bitops.h>
+
+#include <asm/smp.h>
+#include <asm/acpi.h>
+#include <asm/mtrr.h>
+#include <asm/mpspec.h>
+#include <asm/io_apic.h>
+
+#include <mach_apic.h>
+#include <mach_mpparse.h>
+#include <bios_ebda.h>
+
+/* Have we found an MP table */
+int smp_found_config;
+unsigned int __initdata maxcpus = NR_CPUS;
+
+/*
+ * Various Linux-internal data structures created from the
+ * MP-table.
+ */
+int apic_version [MAX_APICS];
+int mp_bus_id_to_type [MAX_MP_BUSSES];
+int mp_bus_id_to_node [MAX_MP_BUSSES];
+int mp_bus_id_to_local [MAX_MP_BUSSES];
+int quad_local_to_mp_bus_id [NR_CPUS/4][4];
+int mp_bus_id_to_pci_bus [MAX_MP_BUSSES] = { [0 ... MAX_MP_BUSSES-1] = -1 };
+static int mp_current_pci_id;
+
+/* I/O APIC entries */
+struct mpc_config_ioapic mp_ioapics[MAX_IO_APICS];
+
+/* # of MP IRQ source entries */
+struct mpc_config_intsrc mp_irqs[MAX_IRQ_SOURCES];
+
+/* MP IRQ source entries */
+int mp_irq_entries;
+
+int nr_ioapics;
+
+int pic_mode;
+unsigned long mp_lapic_addr;
+
+unsigned int def_to_bigsmp = 0;
+
+/* Processor that is doing the boot up */
+unsigned int boot_cpu_physical_apicid = -1U;
+/* Internal processor count */
+static unsigned int __devinitdata num_processors;
+
+/* Bitmask of physically existing CPUs */
+physid_mask_t phys_cpu_present_map;
+
+u8 bios_cpu_apicid[NR_CPUS] = { [0 ... NR_CPUS-1] = BAD_APICID };
+
+/*
+ * Intel MP BIOS table parsing routines:
+ */
+
+
+/*
+ * Checksum an MP configuration block.
+ */
+
+static int __init mpf_checksum(unsigned char *mp, int len)
+{
+ int sum = 0;
+
+ while (len--)
+ sum += *mp++;
+
+ return sum & 0xFF;
+}
+
+/*
+ * Have to match translation table entries to main table entries by counter
+ * hence the mpc_record variable .... can't see a less disgusting way of
+ * doing this ....
+ */
+
+static int mpc_record;
+static struct mpc_config_translation *translation_table[MAX_MPC_ENTRY] __initdata;
+
+#ifndef CONFIG_XEN
+static void __devinit MP_processor_info (struct mpc_config_processor *m)
+{
+ int ver, apicid;
+ physid_mask_t phys_cpu;
+
+ if (!(m->mpc_cpuflag & CPU_ENABLED))
+ return;
+
+ apicid = mpc_apic_id(m, translation_table[mpc_record]);
+
+ if (m->mpc_featureflag&(1<<0))
+ Dprintk(" Floating point unit present.\n");
+ if (m->mpc_featureflag&(1<<7))
+ Dprintk(" Machine Exception supported.\n");
+ if (m->mpc_featureflag&(1<<8))
+ Dprintk(" 64 bit compare & exchange supported.\n");
+ if (m->mpc_featureflag&(1<<9))
+ Dprintk(" Internal APIC present.\n");
+ if (m->mpc_featureflag&(1<<11))
+ Dprintk(" SEP present.\n");
+ if (m->mpc_featureflag&(1<<12))
+ Dprintk(" MTRR present.\n");
+ if (m->mpc_featureflag&(1<<13))
+ Dprintk(" PGE present.\n");
+ if (m->mpc_featureflag&(1<<14))
+ Dprintk(" MCA present.\n");
+ if (m->mpc_featureflag&(1<<15))
+ Dprintk(" CMOV present.\n");
+ if (m->mpc_featureflag&(1<<16))
+ Dprintk(" PAT present.\n");
+ if (m->mpc_featureflag&(1<<17))
+ Dprintk(" PSE present.\n");
+ if (m->mpc_featureflag&(1<<18))
+ Dprintk(" PSN present.\n");
+ if (m->mpc_featureflag&(1<<19))
+ Dprintk(" Cache Line Flush Instruction present.\n");
+ /* 20 Reserved */
+ if (m->mpc_featureflag&(1<<21))
+ Dprintk(" Debug Trace and EMON Store present.\n");
+ if (m->mpc_featureflag&(1<<22))
+ Dprintk(" ACPI Thermal Throttle Registers present.\n");
+ if (m->mpc_featureflag&(1<<23))
+ Dprintk(" MMX present.\n");
+ if (m->mpc_featureflag&(1<<24))
+ Dprintk(" FXSR present.\n");
+ if (m->mpc_featureflag&(1<<25))
+ Dprintk(" XMM present.\n");
+ if (m->mpc_featureflag&(1<<26))
+ Dprintk(" Willamette New Instructions present.\n");
+ if (m->mpc_featureflag&(1<<27))
+ Dprintk(" Self Snoop present.\n");
+ if (m->mpc_featureflag&(1<<28))
+ Dprintk(" HT present.\n");
+ if (m->mpc_featureflag&(1<<29))
+ Dprintk(" Thermal Monitor present.\n");
+ /* 30, 31 Reserved */
+
+
+ if (m->mpc_cpuflag & CPU_BOOTPROCESSOR) {
+ Dprintk(" Bootup CPU\n");
+ boot_cpu_physical_apicid = m->mpc_apicid;
+ }
+
+ ver = m->mpc_apicver;
+
+ /*
+ * Validate version
+ */
+ if (ver == 0x0) {
+ printk(KERN_WARNING "BIOS bug, APIC version is 0 for CPU#%d! "
+ "fixing up to 0x10. (tell your hw vendor)\n",
+ m->mpc_apicid);
+ ver = 0x10;
+ }
+ apic_version[m->mpc_apicid] = ver;
+
+ phys_cpu = apicid_to_cpu_present(apicid);
+ physids_or(phys_cpu_present_map, phys_cpu_present_map, phys_cpu);
+
+ if (num_processors >= NR_CPUS) {
+ printk(KERN_WARNING "WARNING: NR_CPUS limit of %i reached."
+ " Processor ignored.\n", NR_CPUS);
+ return;
+ }
+
+ if (num_processors >= maxcpus) {
+ printk(KERN_WARNING "WARNING: maxcpus limit of %i reached."
+ " Processor ignored.\n", maxcpus);
+ return;
+ }
+
+ cpu_set(num_processors, cpu_possible_map);
+ num_processors++;
+
+ /*
+ * Would be preferable to switch to bigsmp when CONFIG_HOTPLUG_CPU=y
+ * but we need to work other dependencies like SMP_SUSPEND etc
+ * before this can be done without some confusion.
+ * if (CPU_HOTPLUG_ENABLED || num_processors > 8)
+ * - Ashok Raj <ashok.raj@intel.com>
+ */
+ if (num_processors > 8) {
+ switch (boot_cpu_data.x86_vendor) {
+ case X86_VENDOR_INTEL:
+ if (!APIC_XAPIC(ver)) {
+ def_to_bigsmp = 0;
+ break;
+ }
+ /* If P4 and above fall through */
+ case X86_VENDOR_AMD:
+ def_to_bigsmp = 1;
+ }
+ }
+ bios_cpu_apicid[num_processors - 1] = m->mpc_apicid;
+}
+#else
+void __init MP_processor_info (struct mpc_config_processor *m)
+{
+ num_processors++;
+}
+#endif /* CONFIG_XEN */
+
+static void __init MP_bus_info (struct mpc_config_bus *m)
+{
+ char str[7];
+
+ memcpy(str, m->mpc_bustype, 6);
+ str[6] = 0;
+
+ mpc_oem_bus_info(m, str, translation_table[mpc_record]);
+
+ if (m->mpc_busid >= MAX_MP_BUSSES) {
+ printk(KERN_WARNING "MP table busid value (%d) for bustype %s "
+ " is too large, max. supported is %d\n",
+ m->mpc_busid, str, MAX_MP_BUSSES - 1);
+ return;
+ }
+
+ if (strncmp(str, BUSTYPE_ISA, sizeof(BUSTYPE_ISA)-1) == 0) {
+ mp_bus_id_to_type[m->mpc_busid] = MP_BUS_ISA;
+ } else if (strncmp(str, BUSTYPE_EISA, sizeof(BUSTYPE_EISA)-1) == 0) {
+ mp_bus_id_to_type[m->mpc_busid] = MP_BUS_EISA;
+ } else if (strncmp(str, BUSTYPE_PCI, sizeof(BUSTYPE_PCI)-1) == 0) {
+ mpc_oem_pci_bus(m, translation_table[mpc_record]);
+ mp_bus_id_to_type[m->mpc_busid] = MP_BUS_PCI;
+ mp_bus_id_to_pci_bus[m->mpc_busid] = mp_current_pci_id;
+ mp_current_pci_id++;
+ } else if (strncmp(str, BUSTYPE_MCA, sizeof(BUSTYPE_MCA)-1) == 0) {
+ mp_bus_id_to_type[m->mpc_busid] = MP_BUS_MCA;
+ } else if (strncmp(str, BUSTYPE_NEC98, sizeof(BUSTYPE_NEC98)-1) == 0) {
+ mp_bus_id_to_type[m->mpc_busid] = MP_BUS_NEC98;
+ } else {
+ printk(KERN_WARNING "Unknown bustype %s - ignoring\n", str);
+ }
+}
+
+static void __init MP_ioapic_info (struct mpc_config_ioapic *m)
+{
+ if (!(m->mpc_flags & MPC_APIC_USABLE))
+ return;
+
+ printk(KERN_INFO "I/O APIC #%d Version %d at 0x%lX.\n",
+ m->mpc_apicid, m->mpc_apicver, m->mpc_apicaddr);
+ if (nr_ioapics >= MAX_IO_APICS) {
+ printk(KERN_CRIT "Max # of I/O APICs (%d) exceeded (found %d).\n",
+ MAX_IO_APICS, nr_ioapics);
+ panic("Recompile kernel with bigger MAX_IO_APICS!.\n");
+ }
+ if (!m->mpc_apicaddr) {
+ printk(KERN_ERR "WARNING: bogus zero I/O APIC address"
+ " found in MP table, skipping!\n");
+ return;
+ }
+ mp_ioapics[nr_ioapics] = *m;
+ nr_ioapics++;
+}
+
+static void __init MP_intsrc_info (struct mpc_config_intsrc *m)
+{
+ mp_irqs [mp_irq_entries] = *m;
+ Dprintk("Int: type %d, pol %d, trig %d, bus %d,"
+ " IRQ %02x, APIC ID %x, APIC INT %02x\n",
+ m->mpc_irqtype, m->mpc_irqflag & 3,
+ (m->mpc_irqflag >> 2) & 3, m->mpc_srcbus,
+ m->mpc_srcbusirq, m->mpc_dstapic, m->mpc_dstirq);
+ if (++mp_irq_entries == MAX_IRQ_SOURCES)
+ panic("Max # of irq sources exceeded!!\n");
+}
+
+static void __init MP_lintsrc_info (struct mpc_config_lintsrc *m)
+{
+ Dprintk("Lint: type %d, pol %d, trig %d, bus %d,"
+ " IRQ %02x, APIC ID %x, APIC LINT %02x\n",
+ m->mpc_irqtype, m->mpc_irqflag & 3,
+ (m->mpc_irqflag >> 2) &3, m->mpc_srcbusid,
+ m->mpc_srcbusirq, m->mpc_destapic, m->mpc_destapiclint);
+ /*
+ * Well it seems all SMP boards in existence
+ * use ExtINT/LVT1 == LINT0 and
+ * NMI/LVT2 == LINT1 - the following check
+ * will show us if this assumptions is false.
+ * Until then we do not have to add baggage.
+ */
+ if ((m->mpc_irqtype == mp_ExtINT) &&
+ (m->mpc_destapiclint != 0))
+ BUG();
+ if ((m->mpc_irqtype == mp_NMI) &&
+ (m->mpc_destapiclint != 1))
+ BUG();
+}
+
+#ifdef CONFIG_X86_NUMAQ
+static void __init MP_translation_info (struct mpc_config_translation *m)
+{
+ printk(KERN_INFO "Translation: record %d, type %d, quad %d, global %d, local %d\n", mpc_record, m->trans_type, m->trans_quad, m->trans_global, m->trans_local);
+
+ if (mpc_record >= MAX_MPC_ENTRY)
+ printk(KERN_ERR "MAX_MPC_ENTRY exceeded!\n");
+ else
+ translation_table[mpc_record] = m; /* stash this for later */
+ if (m->trans_quad < MAX_NUMNODES && !node_online(m->trans_quad))
+ node_set_online(m->trans_quad);
+}
+
+/*
+ * Read/parse the MPC oem tables
+ */
+
+static void __init smp_read_mpc_oem(struct mp_config_oemtable *oemtable, \
+ unsigned short oemsize)
+{
+ int count = sizeof (*oemtable); /* the header size */
+ unsigned char *oemptr = ((unsigned char *)oemtable)+count;
+
+ mpc_record = 0;
+ printk(KERN_INFO "Found an OEM MPC table at %8p - parsing it ... \n", oemtable);
+ if (memcmp(oemtable->oem_signature,MPC_OEM_SIGNATURE,4))
+ {
+ printk(KERN_WARNING "SMP mpc oemtable: bad signature [%c%c%c%c]!\n",
+ oemtable->oem_signature[0],
+ oemtable->oem_signature[1],
+ oemtable->oem_signature[2],
+ oemtable->oem_signature[3]);
+ return;
+ }
+ if (mpf_checksum((unsigned char *)oemtable,oemtable->oem_length))
+ {
+ printk(KERN_WARNING "SMP oem mptable: checksum error!\n");
+ return;
+ }
+ while (count < oemtable->oem_length) {
+ switch (*oemptr) {
+ case MP_TRANSLATION:
+ {
+ struct mpc_config_translation *m=
+ (struct mpc_config_translation *)oemptr;
+ MP_translation_info(m);
+ oemptr += sizeof(*m);
+ count += sizeof(*m);
+ ++mpc_record;
+ break;
+ }
+ default:
+ {
+ printk(KERN_WARNING "Unrecognised OEM table entry type! - %d\n", (int) *oemptr);
+ return;
+ }
+ }
+ }
+}
+
+static inline void mps_oem_check(struct mp_config_table *mpc, char *oem,
+ char *productid)
+{
+ if (strncmp(oem, "IBM NUMA", 8))
+ printk("Warning! May not be a NUMA-Q system!\n");
+ if (mpc->mpc_oemptr)
+ smp_read_mpc_oem((struct mp_config_oemtable *) mpc->mpc_oemptr,
+ mpc->mpc_oemsize);
+}
+#endif /* CONFIG_X86_NUMAQ */
+
+/*
+ * Read/parse the MPC
+ */
+
+static int __init smp_read_mpc(struct mp_config_table *mpc)
+{
+ char str[16];
+ char oem[10];
+ int count=sizeof(*mpc);
+ unsigned char *mpt=((unsigned char *)mpc)+count;
+
+ if (memcmp(mpc->mpc_signature,MPC_SIGNATURE,4)) {
+ printk(KERN_ERR "SMP mptable: bad signature [0x%x]!\n",
+ *(u32 *)mpc->mpc_signature);
+ return 0;
+ }
+ if (mpf_checksum((unsigned char *)mpc,mpc->mpc_length)) {
+ printk(KERN_ERR "SMP mptable: checksum error!\n");
+ return 0;
+ }
+ if (mpc->mpc_spec!=0x01 && mpc->mpc_spec!=0x04) {
+ printk(KERN_ERR "SMP mptable: bad table version (%d)!!\n",
+ mpc->mpc_spec);
+ return 0;
+ }
+ if (!mpc->mpc_lapic) {
+ printk(KERN_ERR "SMP mptable: null local APIC address!\n");
+ return 0;
+ }
+ memcpy(oem,mpc->mpc_oem,8);
+ oem[8]=0;
+ printk(KERN_INFO "OEM ID: %s ",oem);
+
+ memcpy(str,mpc->mpc_productid,12);
+ str[12]=0;
+ printk("Product ID: %s ",str);
+
+ mps_oem_check(mpc, oem, str);
+
+ printk("APIC at: 0x%lX\n",mpc->mpc_lapic);
+
+ /*
+ * Save the local APIC address (it might be non-default) -- but only
+ * if we're not using ACPI.
+ */
+ if (!acpi_lapic)
+ mp_lapic_addr = mpc->mpc_lapic;
+
+ /*
+ * Now process the configuration blocks.
+ */
+ mpc_record = 0;
+ while (count < mpc->mpc_length) {
+ switch(*mpt) {
+ case MP_PROCESSOR:
+ {
+ struct mpc_config_processor *m=
+ (struct mpc_config_processor *)mpt;
+ /* ACPI may have already provided this data */
+ if (!acpi_lapic)
+ MP_processor_info(m);
+ mpt += sizeof(*m);
+ count += sizeof(*m);
+ break;
+ }
+ case MP_BUS:
+ {
+ struct mpc_config_bus *m=
+ (struct mpc_config_bus *)mpt;
+ MP_bus_info(m);
+ mpt += sizeof(*m);
+ count += sizeof(*m);
+ break;
+ }
+ case MP_IOAPIC:
+ {
+ struct mpc_config_ioapic *m=
+ (struct mpc_config_ioapic *)mpt;
+ MP_ioapic_info(m);
+ mpt+=sizeof(*m);
+ count+=sizeof(*m);
+ break;
+ }
+ case MP_INTSRC:
+ {
+ struct mpc_config_intsrc *m=
+ (struct mpc_config_intsrc *)mpt;
+
+ MP_intsrc_info(m);
+ mpt+=sizeof(*m);
+ count+=sizeof(*m);
+ break;
+ }
+ case MP_LINTSRC:
+ {
+ struct mpc_config_lintsrc *m=
+ (struct mpc_config_lintsrc *)mpt;
+ MP_lintsrc_info(m);
+ mpt+=sizeof(*m);
+ count+=sizeof(*m);
+ break;
+ }
+ default:
+ {
+ count = mpc->mpc_length;
+ break;
+ }
+ }
+ ++mpc_record;
+ }
+ clustered_apic_check();
+ if (!num_processors)
+ printk(KERN_ERR "SMP mptable: no processors registered!\n");
+ return num_processors;
+}
+
+static int __init ELCR_trigger(unsigned int irq)
+{
+ unsigned int port;
+
+ port = 0x4d0 + (irq >> 3);
+ return (inb(port) >> (irq & 7)) & 1;
+}
+
+static void __init construct_default_ioirq_mptable(int mpc_default_type)
+{
+ struct mpc_config_intsrc intsrc;
+ int i;
+ int ELCR_fallback = 0;
+
+ intsrc.mpc_type = MP_INTSRC;
+ intsrc.mpc_irqflag = 0; /* conforming */
+ intsrc.mpc_srcbus = 0;
+ intsrc.mpc_dstapic = mp_ioapics[0].mpc_apicid;
+
+ intsrc.mpc_irqtype = mp_INT;
+
+ /*
+ * If true, we have an ISA/PCI system with no IRQ entries
+ * in the MP table. To prevent the PCI interrupts from being set up
+ * incorrectly, we try to use the ELCR. The sanity check to see if
+ * there is good ELCR data is very simple - IRQ0, 1, 2 and 13 can
+ * never be level sensitive, so we simply see if the ELCR agrees.
+ * If it does, we assume it's valid.
+ */
+ if (mpc_default_type == 5) {
+ printk(KERN_INFO "ISA/PCI bus type with no IRQ information... falling back to ELCR\n");
+
+ if (ELCR_trigger(0) || ELCR_trigger(1) || ELCR_trigger(2) || ELCR_trigger(13))
+ printk(KERN_WARNING "ELCR contains invalid data... not using ELCR\n");
+ else {
+ printk(KERN_INFO "Using ELCR to identify PCI interrupts\n");
+ ELCR_fallback = 1;
+ }
+ }
+
+ for (i = 0; i < 16; i++) {
+ switch (mpc_default_type) {
+ case 2:
+ if (i == 0 || i == 13)
+ continue; /* IRQ0 & IRQ13 not connected */
+ /* fall through */
+ default:
+ if (i == 2)
+ continue; /* IRQ2 is never connected */
+ }
+
+ if (ELCR_fallback) {
+ /*
+ * If the ELCR indicates a level-sensitive interrupt, we
+ * copy that information over to the MP table in the
+ * irqflag field (level sensitive, active high polarity).
+ */
+ if (ELCR_trigger(i))
+ intsrc.mpc_irqflag = 13;
+ else
+ intsrc.mpc_irqflag = 0;
+ }
+
+ intsrc.mpc_srcbusirq = i;
+ intsrc.mpc_dstirq = i ? i : 2; /* IRQ0 to INTIN2 */
+ MP_intsrc_info(&intsrc);
+ }
+
+ intsrc.mpc_irqtype = mp_ExtINT;
+ intsrc.mpc_srcbusirq = 0;
+ intsrc.mpc_dstirq = 0; /* 8259A to INTIN0 */
+ MP_intsrc_info(&intsrc);
+}
+
+static inline void __init construct_default_ISA_mptable(int mpc_default_type)
+{
+ struct mpc_config_processor processor;
+ struct mpc_config_bus bus;
+ struct mpc_config_ioapic ioapic;
+ struct mpc_config_lintsrc lintsrc;
+ int linttypes[2] = { mp_ExtINT, mp_NMI };
+ int i;
+
+ /*
+ * local APIC has default address
+ */
+ mp_lapic_addr = APIC_DEFAULT_PHYS_BASE;
+
+ /*
+ * 2 CPUs, numbered 0 & 1.
+ */
+ processor.mpc_type = MP_PROCESSOR;
+ /* Either an integrated APIC or a discrete 82489DX. */
+ processor.mpc_apicver = mpc_default_type > 4 ? 0x10 : 0x01;
+ processor.mpc_cpuflag = CPU_ENABLED;
+ processor.mpc_cpufeature = (boot_cpu_data.x86 << 8) |
+ (boot_cpu_data.x86_model << 4) |
+ boot_cpu_data.x86_mask;
+ processor.mpc_featureflag = boot_cpu_data.x86_capability[0];
+ processor.mpc_reserved[0] = 0;
+ processor.mpc_reserved[1] = 0;
+ for (i = 0; i < 2; i++) {
+ processor.mpc_apicid = i;
+ MP_processor_info(&processor);
+ }
+
+ bus.mpc_type = MP_BUS;
+ bus.mpc_busid = 0;
+ switch (mpc_default_type) {
+ default:
+ printk("???\n");
+ printk(KERN_ERR "Unknown standard configuration %d\n",
+ mpc_default_type);
+ /* fall through */
+ case 1:
+ case 5:
+ memcpy(bus.mpc_bustype, "ISA ", 6);
+ break;
+ case 2:
+ case 6:
+ case 3:
+ memcpy(bus.mpc_bustype, "EISA ", 6);
+ break;
+ case 4:
+ case 7:
+ memcpy(bus.mpc_bustype, "MCA ", 6);
+ }
+ MP_bus_info(&bus);
+ if (mpc_default_type > 4) {
+ bus.mpc_busid = 1;
+ memcpy(bus.mpc_bustype, "PCI ", 6);
+ MP_bus_info(&bus);
+ }
+
+ ioapic.mpc_type = MP_IOAPIC;
+ ioapic.mpc_apicid = 2;
+ ioapic.mpc_apicver = mpc_default_type > 4 ? 0x10 : 0x01;
+ ioapic.mpc_flags = MPC_APIC_USABLE;
+ ioapic.mpc_apicaddr = 0xFEC00000;
+ MP_ioapic_info(&ioapic);
+
+ /*
+ * We set up most of the low 16 IO-APIC pins according to MPS rules.
+ */
+ construct_default_ioirq_mptable(mpc_default_type);
+
+ lintsrc.mpc_type = MP_LINTSRC;
+ lintsrc.mpc_irqflag = 0; /* conforming */
+ lintsrc.mpc_srcbusid = 0;
+ lintsrc.mpc_srcbusirq = 0;
+ lintsrc.mpc_destapic = MP_APIC_ALL;
+ for (i = 0; i < 2; i++) {
+ lintsrc.mpc_irqtype = linttypes[i];
+ lintsrc.mpc_destapiclint = i;
+ MP_lintsrc_info(&lintsrc);
+ }
+}
+
+static struct intel_mp_floating *mpf_found;
+
+/*
+ * Scan the memory blocks for an SMP configuration block.
+ */
+void __init get_smp_config (void)
+{
+ struct intel_mp_floating *mpf = mpf_found;
+
+ /*
+ * ACPI supports both logical (e.g. Hyper-Threading) and physical
+ * processors, where MPS only supports physical.
+ */
+ if (acpi_lapic && acpi_ioapic) {
+ printk(KERN_INFO "Using ACPI (MADT) for SMP configuration information\n");
+ return;
+ }
+ else if (acpi_lapic)
+ printk(KERN_INFO "Using ACPI for processor (LAPIC) configuration information\n");
+
+ printk(KERN_INFO "Intel MultiProcessor Specification v1.%d\n", mpf->mpf_specification);
+ if (mpf->mpf_feature2 & (1<<7)) {
+ printk(KERN_INFO " IMCR and PIC compatibility mode.\n");
+ pic_mode = 1;
+ } else {
+ printk(KERN_INFO " Virtual Wire compatibility mode.\n");
+ pic_mode = 0;
+ }
+
+ /*
+ * Now see if we need to read further.
+ */
+ if (mpf->mpf_feature1 != 0) {
+
+ printk(KERN_INFO "Default MP configuration #%d\n", mpf->mpf_feature1);
+ construct_default_ISA_mptable(mpf->mpf_feature1);
+
+ } else if (mpf->mpf_physptr) {
+
+ /*
+ * Read the physical hardware table. Anything here will
+ * override the defaults.
+ */
+ if (!smp_read_mpc(isa_bus_to_virt(mpf->mpf_physptr))) {
+ smp_found_config = 0;
+ printk(KERN_ERR "BIOS bug, MP table errors detected!...\n");
+ printk(KERN_ERR "... disabling SMP support. (tell your hw vendor)\n");
+ return;
+ }
+ /*
+ * If there are no explicit MP IRQ entries, then we are
+ * broken. We set up most of the low 16 IO-APIC pins to
+ * ISA defaults and hope it will work.
+ */
+ if (!mp_irq_entries) {
+ struct mpc_config_bus bus;
+
+ printk(KERN_ERR "BIOS bug, no explicit IRQ entries, using default mptable. (tell your hw vendor)\n");
+
+ bus.mpc_type = MP_BUS;
+ bus.mpc_busid = 0;
+ memcpy(bus.mpc_bustype, "ISA ", 6);
+ MP_bus_info(&bus);
+
+ construct_default_ioirq_mptable(0);
+ }
+
+ } else
+ BUG();
+
+ printk(KERN_INFO "Processors: %d\n", num_processors);
+ /*
+ * Only use the first configuration found.
+ */
+}
+
+static int __init smp_scan_config (unsigned long base, unsigned long length)
+{
+ unsigned long *bp = isa_bus_to_virt(base);
+ struct intel_mp_floating *mpf;
+
+ Dprintk("Scan SMP from %p for %ld bytes.\n", bp,length);
+ if (sizeof(*mpf) != 16)
+ printk("Error: MPF size\n");
+
+ while (length > 0) {
+ mpf = (struct intel_mp_floating *)bp;
+ if ((*bp == SMP_MAGIC_IDENT) &&
+ (mpf->mpf_length == 1) &&
+ !mpf_checksum((unsigned char *)bp, 16) &&
+ ((mpf->mpf_specification == 1)
+ || (mpf->mpf_specification == 4)) ) {
+
+ smp_found_config = 1;
+#ifndef CONFIG_XEN
+ printk(KERN_INFO "found SMP MP-table at %08lx\n",
+ virt_to_phys(mpf));
+ reserve_bootmem(virt_to_phys(mpf), PAGE_SIZE);
+ if (mpf->mpf_physptr) {
+ /*
+ * We cannot access to MPC table to compute
+ * table size yet, as only few megabytes from
+ * the bottom is mapped now.
+ * PC-9800's MPC table places on the very last
+ * of physical memory; so that simply reserving
+ * PAGE_SIZE from mpg->mpf_physptr yields BUG()
+ * in reserve_bootmem.
+ */
+ unsigned long size = PAGE_SIZE;
+ unsigned long end = max_low_pfn * PAGE_SIZE;
+ if (mpf->mpf_physptr + size > end)
+ size = end - mpf->mpf_physptr;
+ reserve_bootmem(mpf->mpf_physptr, size);
+ }
+#else
+ printk(KERN_INFO "found SMP MP-table at %08lx\n",
+ ((unsigned long)bp - (unsigned long)isa_bus_to_virt(base)) + base);
+#endif
+
+ mpf_found = mpf;
+ return 1;
+ }
+ bp += 4;
+ length -= 16;
+ }
+ return 0;
+}
+
+void __init find_smp_config (void)
+{
+#ifndef CONFIG_XEN
+ unsigned int address;
+#endif
+
+ /*
+ * FIXME: Linux assumes you have 640K of base ram..
+ * this continues the error...
+ *
+ * 1) Scan the bottom 1K for a signature
+ * 2) Scan the top 1K of base RAM
+ * 3) Scan the 64K of bios
+ */
+ if (smp_scan_config(0x0,0x400) ||
+ smp_scan_config(639*0x400,0x400) ||
+ smp_scan_config(0xF0000,0x10000))
+ return;
+ /*
+ * If it is an SMP machine we should know now, unless the
+ * configuration is in an EISA/MCA bus machine with an
+ * extended bios data area.
+ *
+ * there is a real-mode segmented pointer pointing to the
+ * 4K EBDA area at 0x40E, calculate and scan it here.
+ *
+ * NOTE! There are Linux loaders that will corrupt the EBDA
+ * area, and as such this kind of SMP config may be less
+ * trustworthy, simply because the SMP table may have been
+ * stomped on during early boot. These loaders are buggy and
+ * should be fixed.
+ *
+ * MP1.4 SPEC states to only scan first 1K of 4K EBDA.
+ */
+
+#ifndef CONFIG_XEN
+ address = get_bios_ebda();
+ if (address)
+ smp_scan_config(address, 0x400);
+#endif
+}
+
+int es7000_plat;
+
+/* --------------------------------------------------------------------------
+ ACPI-based MP Configuration
+ -------------------------------------------------------------------------- */
+
+#ifdef CONFIG_ACPI
+
+void __init mp_register_lapic_address (
+ u64 address)
+{
+#ifndef CONFIG_XEN
+ mp_lapic_addr = (unsigned long) address;
+
+ set_fixmap_nocache(FIX_APIC_BASE, mp_lapic_addr);
+
+ if (boot_cpu_physical_apicid == -1U)
+ boot_cpu_physical_apicid = GET_APIC_ID(apic_read(APIC_ID));
+
+ Dprintk("Boot CPU = %d\n", boot_cpu_physical_apicid);
+#endif
+}
+
+
+void __devinit mp_register_lapic (
+ u8 id,
+ u8 enabled)
+{
+ struct mpc_config_processor processor;
+ int boot_cpu = 0;
+
+ if (MAX_APICS - id <= 0) {
+ printk(KERN_WARNING "Processor #%d invalid (max %d)\n",
+ id, MAX_APICS);
+ return;
+ }
+
+ if (id == boot_cpu_physical_apicid)
+ boot_cpu = 1;
+
+#ifndef CONFIG_XEN
+ processor.mpc_type = MP_PROCESSOR;
+ processor.mpc_apicid = id;
+ processor.mpc_apicver = GET_APIC_VERSION(apic_read(APIC_LVR));
+ processor.mpc_cpuflag = (enabled ? CPU_ENABLED : 0);
+ processor.mpc_cpuflag |= (boot_cpu ? CPU_BOOTPROCESSOR : 0);
+ processor.mpc_cpufeature = (boot_cpu_data.x86 << 8) |
+ (boot_cpu_data.x86_model << 4) | boot_cpu_data.x86_mask;
+ processor.mpc_featureflag = boot_cpu_data.x86_capability[0];
+ processor.mpc_reserved[0] = 0;
+ processor.mpc_reserved[1] = 0;
+#endif
+
+ MP_processor_info(&processor);
+}
+
+#ifdef CONFIG_X86_IO_APIC
+
+#define MP_ISA_BUS 0
+#define MP_MAX_IOAPIC_PIN 127
+
+static struct mp_ioapic_routing {
+ int apic_id;
+ int gsi_base;
+ int gsi_end;
+ u32 pin_programmed[4];
+} mp_ioapic_routing[MAX_IO_APICS];
+
+
+static int mp_find_ioapic (
+ int gsi)
+{
+ int i = 0;
+
+ /* Find the IOAPIC that manages this GSI. */
+ for (i = 0; i < nr_ioapics; i++) {
+ if ((gsi >= mp_ioapic_routing[i].gsi_base)
+ && (gsi <= mp_ioapic_routing[i].gsi_end))
+ return i;
+ }
+
+ printk(KERN_ERR "ERROR: Unable to locate IOAPIC for GSI %d\n", gsi);
+
+ return -1;
+}
+
+
+void __init mp_register_ioapic (
+ u8 id,
+ u32 address,
+ u32 gsi_base)
+{
+ int idx = 0;
+ int tmpid;
+
+ if (nr_ioapics >= MAX_IO_APICS) {
+ printk(KERN_ERR "ERROR: Max # of I/O APICs (%d) exceeded "
+ "(found %d)\n", MAX_IO_APICS, nr_ioapics);
+ panic("Recompile kernel with bigger MAX_IO_APICS!\n");
+ }
+ if (!address) {
+ printk(KERN_ERR "WARNING: Bogus (zero) I/O APIC address"
+ " found in MADT table, skipping!\n");
+ return;
+ }
+
+ idx = nr_ioapics++;
+
+ mp_ioapics[idx].mpc_type = MP_IOAPIC;
+ mp_ioapics[idx].mpc_flags = MPC_APIC_USABLE;
+ mp_ioapics[idx].mpc_apicaddr = address;
+
+#ifndef CONFIG_XEN
+ set_fixmap_nocache(FIX_IO_APIC_BASE_0 + idx, address);
+#endif
+ if ((boot_cpu_data.x86_vendor == X86_VENDOR_INTEL)
+ && !APIC_XAPIC(apic_version[boot_cpu_physical_apicid]))
+ tmpid = io_apic_get_unique_id(idx, id);
+ else
+ tmpid = id;
+ if (tmpid == -1) {
+ nr_ioapics--;
+ return;
+ }
+ mp_ioapics[idx].mpc_apicid = tmpid;
+ mp_ioapics[idx].mpc_apicver = io_apic_get_version(idx);
+
+ /*
+ * Build basic GSI lookup table to facilitate gsi->io_apic lookups
+ * and to prevent reprogramming of IOAPIC pins (PCI GSIs).
+ */
+ mp_ioapic_routing[idx].apic_id = mp_ioapics[idx].mpc_apicid;
+ mp_ioapic_routing[idx].gsi_base = gsi_base;
+ mp_ioapic_routing[idx].gsi_end = gsi_base +
+ io_apic_get_redir_entries(idx);
+
+ printk("IOAPIC[%d]: apic_id %d, version %d, address 0x%lx, "
+ "GSI %d-%d\n", idx, mp_ioapics[idx].mpc_apicid,
+ mp_ioapics[idx].mpc_apicver, mp_ioapics[idx].mpc_apicaddr,
+ mp_ioapic_routing[idx].gsi_base,
+ mp_ioapic_routing[idx].gsi_end);
+
+ return;
+}
+
+
+void __init mp_override_legacy_irq (
+ u8 bus_irq,
+ u8 polarity,
+ u8 trigger,
+ u32 gsi)
+{
+ struct mpc_config_intsrc intsrc;
+ int ioapic = -1;
+ int pin = -1;
+
+ /*
+ * Convert 'gsi' to 'ioapic.pin'.
+ */
+ ioapic = mp_find_ioapic(gsi);
+ if (ioapic < 0)
+ return;
+ pin = gsi - mp_ioapic_routing[ioapic].gsi_base;
+
+ /*
+ * TBD: This check is for faulty timer entries, where the override
+ * erroneously sets the trigger to level, resulting in a HUGE
+ * increase of timer interrupts!
+ */
+ if ((bus_irq == 0) && (trigger == 3))
+ trigger = 1;
+
+ intsrc.mpc_type = MP_INTSRC;
+ intsrc.mpc_irqtype = mp_INT;
+ intsrc.mpc_irqflag = (trigger << 2) | polarity;
+ intsrc.mpc_srcbus = MP_ISA_BUS;
+ intsrc.mpc_srcbusirq = bus_irq; /* IRQ */
+ intsrc.mpc_dstapic = mp_ioapics[ioapic].mpc_apicid; /* APIC ID */
+ intsrc.mpc_dstirq = pin; /* INTIN# */
+
+ Dprintk("Int: type %d, pol %d, trig %d, bus %d, irq %d, %d-%d\n",
+ intsrc.mpc_irqtype, intsrc.mpc_irqflag & 3,
+ (intsrc.mpc_irqflag >> 2) & 3, intsrc.mpc_srcbus,
+ intsrc.mpc_srcbusirq, intsrc.mpc_dstapic, intsrc.mpc_dstirq);
+
+ mp_irqs[mp_irq_entries] = intsrc;
+ if (++mp_irq_entries == MAX_IRQ_SOURCES)
+ panic("Max # of irq sources exceeded!\n");
+
+ return;
+}
+
+void __init mp_config_acpi_legacy_irqs (void)
+{
+ struct mpc_config_intsrc intsrc;
+ int i = 0;
+ int ioapic = -1;
+
+ /*
+ * Fabricate the legacy ISA bus (bus #31).
+ */
+ mp_bus_id_to_type[MP_ISA_BUS] = MP_BUS_ISA;
+ Dprintk("Bus #%d is ISA\n", MP_ISA_BUS);
+
+ /*
+ * Older generations of ES7000 have no legacy identity mappings
+ */
+ if (es7000_plat == 1)
+ return;
+
+ /*
+ * Locate the IOAPIC that manages the ISA IRQs (0-15).
+ */
+ ioapic = mp_find_ioapic(0);
+ if (ioapic < 0)
+ return;
+
+ intsrc.mpc_type = MP_INTSRC;
+ intsrc.mpc_irqflag = 0; /* Conforming */
+ intsrc.mpc_srcbus = MP_ISA_BUS;
+ intsrc.mpc_dstapic = mp_ioapics[ioapic].mpc_apicid;
+
+ /*
+ * Use the default configuration for the IRQs 0-15. Unless
+ * overriden by (MADT) interrupt source override entries.
+ */
+ for (i = 0; i < 16; i++) {
+ int idx;
+
+ for (idx = 0; idx < mp_irq_entries; idx++) {
+ struct mpc_config_intsrc *irq = mp_irqs + idx;
+
+ /* Do we already have a mapping for this ISA IRQ? */
+ if (irq->mpc_srcbus == MP_ISA_BUS && irq->mpc_srcbusirq == i)
+ break;
+
+ /* Do we already have a mapping for this IOAPIC pin */
+ if ((irq->mpc_dstapic == intsrc.mpc_dstapic) &&
+ (irq->mpc_dstirq == i))
+ break;
+ }
+
+ if (idx != mp_irq_entries) {
+ printk(KERN_DEBUG "ACPI: IRQ%d used by override.\n", i);
+ continue; /* IRQ already used */
+ }
+
+ intsrc.mpc_irqtype = mp_INT;
+ intsrc.mpc_srcbusirq = i; /* Identity mapped */
+ intsrc.mpc_dstirq = i;
+
+ Dprintk("Int: type %d, pol %d, trig %d, bus %d, irq %d, "
+ "%d-%d\n", intsrc.mpc_irqtype, intsrc.mpc_irqflag & 3,
+ (intsrc.mpc_irqflag >> 2) & 3, intsrc.mpc_srcbus,
+ intsrc.mpc_srcbusirq, intsrc.mpc_dstapic,
+ intsrc.mpc_dstirq);
+
+ mp_irqs[mp_irq_entries] = intsrc;
+ if (++mp_irq_entries == MAX_IRQ_SOURCES)
+ panic("Max # of irq sources exceeded!\n");
+ }
+}
+
+#define MAX_GSI_NUM 4096
+
+int mp_register_gsi (u32 gsi, int triggering, int polarity)
+{
+ int ioapic = -1;
+ int ioapic_pin = 0;
+ int idx, bit = 0;
+ static int pci_irq = 16;
+ /*
+ * Mapping between Global System Interrups, which
+ * represent all possible interrupts, and IRQs
+ * assigned to actual devices.
+ */
+ static int gsi_to_irq[MAX_GSI_NUM];
+
+ /* Don't set up the ACPI SCI because it's already set up */
+ if (acpi_fadt.sci_int == gsi)
+ return gsi;
+
+ ioapic = mp_find_ioapic(gsi);
+ if (ioapic < 0) {
+ printk(KERN_WARNING "No IOAPIC for GSI %u\n", gsi);
+ return gsi;
+ }
+
+ ioapic_pin = gsi - mp_ioapic_routing[ioapic].gsi_base;
+
+ if (ioapic_renumber_irq)
+ gsi = ioapic_renumber_irq(ioapic, gsi);
+
+ /*
+ * Avoid pin reprogramming. PRTs typically include entries
+ * with redundant pin->gsi mappings (but unique PCI devices);
+ * we only program the IOAPIC on the first.
+ */
+ bit = ioapic_pin % 32;
+ idx = (ioapic_pin < 32) ? 0 : (ioapic_pin / 32);
+ if (idx > 3) {
+ printk(KERN_ERR "Invalid reference to IOAPIC pin "
+ "%d-%d\n", mp_ioapic_routing[ioapic].apic_id,
+ ioapic_pin);
+ return gsi;
+ }
+ if ((1<<bit) & mp_ioapic_routing[ioapic].pin_programmed[idx]) {
+ Dprintk(KERN_DEBUG "Pin %d-%d already programmed\n",
+ mp_ioapic_routing[ioapic].apic_id, ioapic_pin);
+ return gsi_to_irq[gsi];
+ }
+
+ mp_ioapic_routing[ioapic].pin_programmed[idx] |= (1<<bit);
+
+ if (triggering == ACPI_LEVEL_SENSITIVE) {
+ /*
+ * For PCI devices assign IRQs in order, avoiding gaps
+ * due to unused I/O APIC pins.
+ */
+ int irq = gsi;
+ if (gsi < MAX_GSI_NUM) {
+ /*
+ * Retain the VIA chipset work-around (gsi > 15), but
+ * avoid a problem where the 8254 timer (IRQ0) is setup
+ * via an override (so it's not on pin 0 of the ioapic),
+ * and at the same time, the pin 0 interrupt is a PCI
+ * type. The gsi > 15 test could cause these two pins
+ * to be shared as IRQ0, and they are not shareable.
+ * So test for this condition, and if necessary, avoid
+ * the pin collision.
+ */
+ if (gsi > 15 || (gsi == 0 && !timer_uses_ioapic_pin_0))
+ gsi = pci_irq++;
+ /*
+ * Don't assign IRQ used by ACPI SCI
+ */
+ if (gsi == acpi_fadt.sci_int)
+ gsi = pci_irq++;
+ gsi_to_irq[irq] = gsi;
+ } else {
+ printk(KERN_ERR "GSI %u is too high\n", gsi);
+ return gsi;
+ }
+ }
+
+ io_apic_set_pci_routing(ioapic, ioapic_pin, gsi,
+ triggering == ACPI_EDGE_SENSITIVE ? 0 : 1,
+ polarity == ACPI_ACTIVE_HIGH ? 0 : 1);
+ return gsi;
+}
+
+#endif /* CONFIG_X86_IO_APIC */
+#endif /* CONFIG_ACPI */
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/kernel/pci-dma-xen.c 2009-11-06 10:23:23.000000000 +0100
@@ -0,0 +1,406 @@
+/*
+ * Dynamic DMA mapping support.
+ *
+ * On i386 there is no hardware dynamic DMA address translation,
+ * so consistent alloc/free are merely page allocation/freeing.
+ * The rest of the dynamic DMA mapping interface is implemented
+ * in asm/pci.h.
+ */
+
+#include <linux/types.h>
+#include <linux/mm.h>
+#include <linux/string.h>
+#include <linux/pci.h>
+#include <linux/module.h>
+#include <linux/version.h>
+#include <asm/io.h>
+#include <xen/balloon.h>
+#include <xen/gnttab.h>
+#include <asm/swiotlb.h>
+#include <asm/tlbflush.h>
+#include <asm-i386/mach-xen/asm/swiotlb.h>
+#include <asm-i386/mach-xen/asm/gnttab_dma.h>
+#include <asm/bug.h>
+
+#ifdef __x86_64__
+#include <asm/proto.h>
+
+int iommu_merge __read_mostly = 0;
+EXPORT_SYMBOL(iommu_merge);
+
+dma_addr_t bad_dma_address __read_mostly;
+EXPORT_SYMBOL(bad_dma_address);
+
+/* This tells the BIO block layer to assume merging. Default to off
+ because we cannot guarantee merging later. */
+int iommu_bio_merge __read_mostly = 0;
+EXPORT_SYMBOL(iommu_bio_merge);
+
+int force_iommu __read_mostly= 0;
+
+__init int iommu_setup(char *p)
+{
+ return 1;
+}
+
+void __init pci_iommu_alloc(void)
+{
+#ifdef CONFIG_SWIOTLB
+ pci_swiotlb_init();
+#endif
+}
+
+static int __init pci_iommu_init(void)
+{
+ no_iommu_init();
+ return 0;
+}
+
+/* Must execute after PCI subsystem */
+fs_initcall(pci_iommu_init);
+#endif
+
+struct dma_coherent_mem {
+ void *virt_base;
+ u32 device_base;
+ int size;
+ int flags;
+ unsigned long *bitmap;
+};
+
+#define IOMMU_BUG_ON(test) \
+do { \
+ if (unlikely(test)) { \
+ printk(KERN_ALERT "Fatal DMA error! " \
+ "Please use 'swiotlb=force'\n"); \
+ BUG(); \
+ } \
+} while (0)
+
+static int check_pages_physically_contiguous(unsigned long pfn,
+ unsigned int offset,
+ size_t length)
+{
+ unsigned long next_mfn;
+ int i;
+ int nr_pages;
+
+ next_mfn = pfn_to_mfn(pfn);
+ nr_pages = (offset + length + PAGE_SIZE-1) >> PAGE_SHIFT;
+
+ for (i = 1; i < nr_pages; i++) {
+ if (pfn_to_mfn(++pfn) != ++next_mfn)
+ return 0;
+ }
+ return 1;
+}
+
+int range_straddles_page_boundary(paddr_t p, size_t size)
+{
+ unsigned long pfn = p >> PAGE_SHIFT;
+ unsigned int offset = p & ~PAGE_MASK;
+
+ return ((offset + size > PAGE_SIZE) &&
+ !check_pages_physically_contiguous(pfn, offset, size));
+}
+
+int
+dma_map_sg(struct device *hwdev, struct scatterlist *sg, int nents,
+ enum dma_data_direction direction)
+{
+ int i, rc;
+
+ if (direction == DMA_NONE)
+ BUG();
+ WARN_ON(nents == 0 || sg[0].length == 0);
+
+ if (swiotlb) {
+ rc = swiotlb_map_sg(hwdev, sg, nents, direction);
+ } else {
+ for (i = 0; i < nents; i++ ) {
+ BUG_ON(!sg[i].page);
+ sg[i].dma_address =
+ gnttab_dma_map_page(sg[i].page) + sg[i].offset;
+ sg[i].dma_length = sg[i].length;
+ IOMMU_BUG_ON(address_needs_mapping(
+ hwdev, sg[i].dma_address));
+ IOMMU_BUG_ON(range_straddles_page_boundary(
+ page_to_pseudophys(sg[i].page) + sg[i].offset,
+ sg[i].length));
+ }
+ rc = nents;
+ }
+
+ flush_write_buffers();
+ return rc;
+}
+EXPORT_SYMBOL(dma_map_sg);
+
+void
+dma_unmap_sg(struct device *hwdev, struct scatterlist *sg, int nents,
+ enum dma_data_direction direction)
+{
+ int i;
+
+ BUG_ON(direction == DMA_NONE);
+ if (swiotlb)
+ swiotlb_unmap_sg(hwdev, sg, nents, direction);
+ else {
+ for (i = 0; i < nents; i++ )
+ gnttab_dma_unmap_page(sg[i].dma_address);
+ }
+}
+EXPORT_SYMBOL(dma_unmap_sg);
+
+#ifdef CONFIG_HIGHMEM
+dma_addr_t
+dma_map_page(struct device *dev, struct page *page, unsigned long offset,
+ size_t size, enum dma_data_direction direction)
+{
+ dma_addr_t dma_addr;
+
+ BUG_ON(direction == DMA_NONE);
+
+ if (swiotlb) {
+ dma_addr = swiotlb_map_page(
+ dev, page, offset, size, direction);
+ } else {
+ dma_addr = gnttab_dma_map_page(page) + offset;
+ IOMMU_BUG_ON(address_needs_mapping(dev, dma_addr));
+ }
+
+ return dma_addr;
+}
+EXPORT_SYMBOL(dma_map_page);
+
+void
+dma_unmap_page(struct device *dev, dma_addr_t dma_address, size_t size,
+ enum dma_data_direction direction)
+{
+ BUG_ON(direction == DMA_NONE);
+ if (swiotlb)
+ swiotlb_unmap_page(dev, dma_address, size, direction);
+ else
+ gnttab_dma_unmap_page(dma_address);
+}
+EXPORT_SYMBOL(dma_unmap_page);
+#endif /* CONFIG_HIGHMEM */
+
+int
+dma_mapping_error(dma_addr_t dma_addr)
+{
+ if (swiotlb)
+ return swiotlb_dma_mapping_error(dma_addr);
+ return 0;
+}
+EXPORT_SYMBOL(dma_mapping_error);
+
+int
+dma_supported(struct device *dev, u64 mask)
+{
+ if (swiotlb)
+ return swiotlb_dma_supported(dev, mask);
+ /*
+ * By default we'll BUG when an infeasible DMA is requested, and
+ * request swiotlb=force (see IOMMU_BUG_ON).
+ */
+ return 1;
+}
+EXPORT_SYMBOL(dma_supported);
+
+void *dma_alloc_coherent(struct device *dev, size_t size,
+ dma_addr_t *dma_handle, gfp_t gfp)
+{
+ void *ret;
+ struct dma_coherent_mem *mem = dev ? dev->dma_mem : NULL;
+ unsigned int order = get_order(size);
+ unsigned long vstart;
+ u64 mask;
+
+ /* ignore region specifiers */
+ gfp &= ~(__GFP_DMA | __GFP_HIGHMEM | __GFP_DMA32);
+
+ if (mem) {
+ int page = bitmap_find_free_region(mem->bitmap, mem->size,
+ order);
+ if (page >= 0) {
+ *dma_handle = mem->device_base + (page << PAGE_SHIFT);
+ ret = mem->virt_base + (page << PAGE_SHIFT);
+ memset(ret, 0, size);
+ return ret;
+ }
+ if (mem->flags & DMA_MEMORY_EXCLUSIVE)
+ return NULL;
+ }
+
+ vstart = __get_free_pages(gfp, order);
+ ret = (void *)vstart;
+
+ if (dev != NULL && dev->coherent_dma_mask)
+ mask = dev->coherent_dma_mask;
+ else
+ mask = 0xffffffff;
+
+ if (ret != NULL) {
+ if (xen_create_contiguous_region(vstart, order,
+ fls64(mask)) != 0) {
+ free_pages(vstart, order);
+ return NULL;
+ }
+ memset(ret, 0, size);
+ *dma_handle = virt_to_bus(ret);
+ }
+ return ret;
+}
+EXPORT_SYMBOL(dma_alloc_coherent);
+
+void dma_free_coherent(struct device *dev, size_t size,
+ void *vaddr, dma_addr_t dma_handle)
+{
+ struct dma_coherent_mem *mem = dev ? dev->dma_mem : NULL;
+ int order = get_order(size);
+
+ if (mem && vaddr >= mem->virt_base && vaddr < (mem->virt_base + (mem->size << PAGE_SHIFT))) {
+ int page = (vaddr - mem->virt_base) >> PAGE_SHIFT;
+
+ bitmap_release_region(mem->bitmap, page, order);
+ } else {
+ xen_destroy_contiguous_region((unsigned long)vaddr, order);
+ free_pages((unsigned long)vaddr, order);
+ }
+}
+EXPORT_SYMBOL(dma_free_coherent);
+
+#ifdef ARCH_HAS_DMA_DECLARE_COHERENT_MEMORY
+int dma_declare_coherent_memory(struct device *dev, dma_addr_t bus_addr,
+ dma_addr_t device_addr, size_t size, int flags)
+{
+ void __iomem *mem_base;
+ int pages = size >> PAGE_SHIFT;
+ int bitmap_size = (pages + 31)/32;
+
+ if ((flags & (DMA_MEMORY_MAP | DMA_MEMORY_IO)) == 0)
+ goto out;
+ if (!size)
+ goto out;
+ if (dev->dma_mem)
+ goto out;
+
+ /* FIXME: this routine just ignores DMA_MEMORY_INCLUDES_CHILDREN */
+
+ mem_base = ioremap(bus_addr, size);
+ if (!mem_base)
+ goto out;
+
+ dev->dma_mem = kmalloc(sizeof(struct dma_coherent_mem), GFP_KERNEL);
+ if (!dev->dma_mem)
+ goto out;
+ memset(dev->dma_mem, 0, sizeof(struct dma_coherent_mem));
+ dev->dma_mem->bitmap = kmalloc(bitmap_size, GFP_KERNEL);
+ if (!dev->dma_mem->bitmap)
+ goto free1_out;
+ memset(dev->dma_mem->bitmap, 0, bitmap_size);
+
+ dev->dma_mem->virt_base = mem_base;
+ dev->dma_mem->device_base = device_addr;
+ dev->dma_mem->size = pages;
+ dev->dma_mem->flags = flags;
+
+ if (flags & DMA_MEMORY_MAP)
+ return DMA_MEMORY_MAP;
+
+ return DMA_MEMORY_IO;
+
+ free1_out:
+ kfree(dev->dma_mem->bitmap);
+ out:
+ return 0;
+}
+EXPORT_SYMBOL(dma_declare_coherent_memory);
+
+void dma_release_declared_memory(struct device *dev)
+{
+ struct dma_coherent_mem *mem = dev->dma_mem;
+
+ if(!mem)
+ return;
+ dev->dma_mem = NULL;
+ iounmap(mem->virt_base);
+ kfree(mem->bitmap);
+ kfree(mem);
+}
+EXPORT_SYMBOL(dma_release_declared_memory);
+
+void *dma_mark_declared_memory_occupied(struct device *dev,
+ dma_addr_t device_addr, size_t size)
+{
+ struct dma_coherent_mem *mem = dev->dma_mem;
+ int pages = (size + (device_addr & ~PAGE_MASK) + PAGE_SIZE - 1) >> PAGE_SHIFT;
+ int pos, err;
+
+ if (!mem)
+ return ERR_PTR(-EINVAL);
+
+ pos = (device_addr - mem->device_base) >> PAGE_SHIFT;
+ err = bitmap_allocate_region(mem->bitmap, pos, get_order(pages));
+ if (err != 0)
+ return ERR_PTR(err);
+ return mem->virt_base + (pos << PAGE_SHIFT);
+}
+EXPORT_SYMBOL(dma_mark_declared_memory_occupied);
+#endif /* ARCH_HAS_DMA_DECLARE_COHERENT_MEMORY */
+
+dma_addr_t
+dma_map_single(struct device *dev, void *ptr, size_t size,
+ enum dma_data_direction direction)
+{
+ dma_addr_t dma;
+
+ if (direction == DMA_NONE)
+ BUG();
+ WARN_ON(size == 0);
+
+ if (swiotlb) {
+ dma = swiotlb_map_single(dev, ptr, size, direction);
+ } else {
+ dma = gnttab_dma_map_page(virt_to_page(ptr)) +
+ offset_in_page(ptr);
+ IOMMU_BUG_ON(range_straddles_page_boundary(__pa(ptr), size));
+ IOMMU_BUG_ON(address_needs_mapping(dev, dma));
+ }
+
+ flush_write_buffers();
+ return dma;
+}
+EXPORT_SYMBOL(dma_map_single);
+
+void
+dma_unmap_single(struct device *dev, dma_addr_t dma_addr, size_t size,
+ enum dma_data_direction direction)
+{
+ if (direction == DMA_NONE)
+ BUG();
+ if (swiotlb)
+ swiotlb_unmap_single(dev, dma_addr, size, direction);
+ else
+ gnttab_dma_unmap_page(dma_addr);
+}
+EXPORT_SYMBOL(dma_unmap_single);
+
+void
+dma_sync_single_for_cpu(struct device *dev, dma_addr_t dma_handle, size_t size,
+ enum dma_data_direction direction)
+{
+ if (swiotlb)
+ swiotlb_sync_single_for_cpu(dev, dma_handle, size, direction);
+}
+EXPORT_SYMBOL(dma_sync_single_for_cpu);
+
+void
+dma_sync_single_for_device(struct device *dev, dma_addr_t dma_handle, size_t size,
+ enum dma_data_direction direction)
+{
+ if (swiotlb)
+ swiotlb_sync_single_for_device(dev, dma_handle, size, direction);
+}
+EXPORT_SYMBOL(dma_sync_single_for_device);
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/kernel/process_32-xen.c 2008-07-21 11:00:32.000000000 +0200
@@ -0,0 +1,877 @@
+/*
+ * linux/arch/i386/kernel/process.c
+ *
+ * Copyright (C) 1995 Linus Torvalds
+ *
+ * Pentium III FXSR, SSE support
+ * Gareth Hughes <gareth@valinux.com>, May 2000
+ */
+
+/*
+ * This file handles the architecture-dependent parts of process handling..
+ */
+
+#include <stdarg.h>
+
+#include <linux/cpu.h>
+#include <linux/errno.h>
+#include <linux/sched.h>
+#include <linux/fs.h>
+#include <linux/kernel.h>
+#include <linux/mm.h>
+#include <linux/elfcore.h>
+#include <linux/smp.h>
+#include <linux/smp_lock.h>
+#include <linux/stddef.h>
+#include <linux/slab.h>
+#include <linux/vmalloc.h>
+#include <linux/user.h>
+#include <linux/a.out.h>
+#include <linux/interrupt.h>
+#include <linux/utsname.h>
+#include <linux/delay.h>
+#include <linux/reboot.h>
+#include <linux/init.h>
+#include <linux/mc146818rtc.h>
+#include <linux/module.h>
+#include <linux/kallsyms.h>
+#include <linux/ptrace.h>
+#include <linux/random.h>
+
+#include <asm/uaccess.h>
+#include <asm/pgtable.h>
+#include <asm/system.h>
+#include <asm/io.h>
+#include <asm/ldt.h>
+#include <asm/processor.h>
+#include <asm/i387.h>
+#include <asm/desc.h>
+#include <asm/vm86.h>
+#ifdef CONFIG_MATH_EMULATION
+#include <asm/math_emu.h>
+#endif
+
+#include <xen/interface/physdev.h>
+#include <xen/interface/vcpu.h>
+#include <xen/cpu_hotplug.h>
+
+#include <linux/err.h>
+
+#include <asm/tlbflush.h>
+#include <asm/cpu.h>
+
+asmlinkage void ret_from_fork(void) __asm__("ret_from_fork");
+
+static int hlt_counter;
+
+unsigned long boot_option_idle_override = 0;
+EXPORT_SYMBOL(boot_option_idle_override);
+
+/*
+ * Return saved PC of a blocked thread.
+ */
+unsigned long thread_saved_pc(struct task_struct *tsk)
+{
+ return ((unsigned long *)tsk->thread.esp)[3];
+}
+
+/*
+ * Powermanagement idle function, if any..
+ */
+void (*pm_idle)(void);
+EXPORT_SYMBOL(pm_idle);
+static DEFINE_PER_CPU(unsigned int, cpu_idle_state);
+
+void disable_hlt(void)
+{
+ hlt_counter++;
+}
+
+EXPORT_SYMBOL(disable_hlt);
+
+void enable_hlt(void)
+{
+ hlt_counter--;
+}
+
+EXPORT_SYMBOL(enable_hlt);
+
+/*
+ * On SMP it's slightly faster (but much more power-consuming!)
+ * to poll the ->work.need_resched flag instead of waiting for the
+ * cross-CPU IPI to arrive. Use this option with caution.
+ */
+static void poll_idle (void)
+{
+ local_irq_enable();
+
+ asm volatile(
+ "2:"
+ "testl %0, %1;"
+ "rep; nop;"
+ "je 2b;"
+ : : "i"(_TIF_NEED_RESCHED), "m" (current_thread_info()->flags));
+}
+
+static void xen_idle(void)
+{
+ local_irq_disable();
+
+ if (need_resched())
+ local_irq_enable();
+ else {
+ current_thread_info()->status &= ~TS_POLLING;
+ smp_mb__after_clear_bit();
+ safe_halt();
+ current_thread_info()->status |= TS_POLLING;
+ }
+}
+#ifdef CONFIG_APM_MODULE
+EXPORT_SYMBOL(default_idle);
+#endif
+
+#ifdef CONFIG_HOTPLUG_CPU
+extern cpumask_t cpu_initialized;
+static inline void play_dead(void)
+{
+ idle_task_exit();
+ local_irq_disable();
+ cpu_clear(smp_processor_id(), cpu_initialized);
+ preempt_enable_no_resched();
+ VOID(HYPERVISOR_vcpu_op(VCPUOP_down, smp_processor_id(), NULL));
+ cpu_bringup();
+}
+#else
+static inline void play_dead(void)
+{
+ BUG();
+}
+#endif /* CONFIG_HOTPLUG_CPU */
+
+/*
+ * The idle thread. There's no useful work to be
+ * done, so just try to conserve power and have a
+ * low exit latency (ie sit in a loop waiting for
+ * somebody to say that they'd like to reschedule)
+ */
+void cpu_idle(void)
+{
+ int cpu = smp_processor_id();
+
+ current_thread_info()->status |= TS_POLLING;
+
+ /* endless idle loop with no priority at all */
+ while (1) {
+ while (!need_resched()) {
+ void (*idle)(void);
+
+ if (__get_cpu_var(cpu_idle_state))
+ __get_cpu_var(cpu_idle_state) = 0;
+
+ rmb();
+ idle = xen_idle; /* no alternatives */
+
+ if (cpu_is_offline(cpu))
+ play_dead();
+
+ __get_cpu_var(irq_stat).idle_timestamp = jiffies;
+ idle();
+ }
+ preempt_enable_no_resched();
+ schedule();
+ preempt_disable();
+ }
+}
+
+void cpu_idle_wait(void)
+{
+ unsigned int cpu, this_cpu = get_cpu();
+ cpumask_t map;
+
+ set_cpus_allowed(current, cpumask_of_cpu(this_cpu));
+ put_cpu();
+
+ cpus_clear(map);
+ for_each_online_cpu(cpu) {
+ per_cpu(cpu_idle_state, cpu) = 1;
+ cpu_set(cpu, map);
+ }
+
+ __get_cpu_var(cpu_idle_state) = 0;
+
+ wmb();
+ do {
+ ssleep(1);
+ for_each_online_cpu(cpu) {
+ if (cpu_isset(cpu, map) && !per_cpu(cpu_idle_state, cpu))
+ cpu_clear(cpu, map);
+ }
+ cpus_and(map, map, cpu_online_map);
+ } while (!cpus_empty(map));
+}
+EXPORT_SYMBOL_GPL(cpu_idle_wait);
+
+void __devinit select_idle_routine(const struct cpuinfo_x86 *c)
+{
+}
+
+static int __init idle_setup (char *str)
+{
+ if (!strncmp(str, "poll", 4)) {
+ printk("using polling idle threads.\n");
+ pm_idle = poll_idle;
+ }
+
+ boot_option_idle_override = 1;
+ return 1;
+}
+
+__setup("idle=", idle_setup);
+
+void show_regs(struct pt_regs * regs)
+{
+ unsigned long cr0 = 0L, cr2 = 0L, cr3 = 0L, cr4 = 0L;
+
+ printk("\n");
+ printk("Pid: %d, comm: %20s\n", current->pid, current->comm);
+ printk("EIP: %04x:[<%08lx>] CPU: %d\n",0xffff & regs->xcs,regs->eip, smp_processor_id());
+ print_symbol("EIP is at %s\n", regs->eip);
+
+ if (user_mode_vm(regs))
+ printk(" ESP: %04x:%08lx",0xffff & regs->xss,regs->esp);
+ printk(" EFLAGS: %08lx %s (%s %.*s)\n",
+ regs->eflags, print_tainted(), system_utsname.release,
+ (int)strcspn(system_utsname.version, " "),
+ system_utsname.version);
+ printk("EAX: %08lx EBX: %08lx ECX: %08lx EDX: %08lx\n",
+ regs->eax,regs->ebx,regs->ecx,regs->edx);
+ printk("ESI: %08lx EDI: %08lx EBP: %08lx",
+ regs->esi, regs->edi, regs->ebp);
+ printk(" DS: %04x ES: %04x\n",
+ 0xffff & regs->xds,0xffff & regs->xes);
+
+ cr0 = read_cr0();
+ cr2 = read_cr2();
+ cr3 = read_cr3();
+ cr4 = read_cr4_safe();
+ printk("CR0: %08lx CR2: %08lx CR3: %08lx CR4: %08lx\n", cr0, cr2, cr3, cr4);
+ show_trace(NULL, regs, &regs->esp);
+}
+
+/*
+ * This gets run with %ebx containing the
+ * function to call, and %edx containing
+ * the "args".
+ */
+extern void kernel_thread_helper(void);
+__asm__(".section .text\n"
+ ".align 4\n"
+ "kernel_thread_helper:\n\t"
+ "movl %edx,%eax\n\t"
+ "pushl %edx\n\t"
+ "call *%ebx\n\t"
+ "pushl %eax\n\t"
+ "call do_exit\n"
+ ".previous");
+
+/*
+ * Create a kernel thread
+ */
+int kernel_thread(int (*fn)(void *), void * arg, unsigned long flags)
+{
+ struct pt_regs regs;
+
+ memset(&regs, 0, sizeof(regs));
+
+ regs.ebx = (unsigned long) fn;
+ regs.edx = (unsigned long) arg;
+
+ regs.xds = __USER_DS;
+ regs.xes = __USER_DS;
+ regs.orig_eax = -1;
+ regs.eip = (unsigned long) kernel_thread_helper;
+ regs.xcs = GET_KERNEL_CS();
+ regs.eflags = X86_EFLAGS_IF | X86_EFLAGS_SF | X86_EFLAGS_PF | 0x2;
+
+ /* Ok, create the new process.. */
+ return do_fork(flags | CLONE_VM | CLONE_UNTRACED, 0, &regs, 0, NULL, NULL);
+}
+EXPORT_SYMBOL(kernel_thread);
+
+/*
+ * Free current thread data structures etc..
+ */
+void exit_thread(void)
+{
+ /* The process may have allocated an io port bitmap... nuke it. */
+ if (unlikely(test_thread_flag(TIF_IO_BITMAP))) {
+ struct task_struct *tsk = current;
+ struct thread_struct *t = &tsk->thread;
+ struct physdev_set_iobitmap set_iobitmap;
+ memset(&set_iobitmap, 0, sizeof(set_iobitmap));
+ WARN_ON(HYPERVISOR_physdev_op(PHYSDEVOP_set_iobitmap,
+ &set_iobitmap));
+ kfree(t->io_bitmap_ptr);
+ t->io_bitmap_ptr = NULL;
+ clear_thread_flag(TIF_IO_BITMAP);
+ }
+}
+
+void flush_thread(void)
+{
+ struct task_struct *tsk = current;
+
+ memset(tsk->thread.debugreg, 0, sizeof(unsigned long)*8);
+ memset(tsk->thread.tls_array, 0, sizeof(tsk->thread.tls_array));
+ clear_tsk_thread_flag(tsk, TIF_DEBUG);
+ /*
+ * Forget coprocessor state..
+ */
+ clear_fpu(tsk);
+ clear_used_math();
+}
+
+void release_thread(struct task_struct *dead_task)
+{
+ BUG_ON(dead_task->mm);
+ release_vm86_irqs(dead_task);
+}
+
+/*
+ * This gets called before we allocate a new thread and copy
+ * the current task into it.
+ */
+void prepare_to_copy(struct task_struct *tsk)
+{
+ unlazy_fpu(tsk);
+}
+
+int copy_thread(int nr, unsigned long clone_flags, unsigned long esp,
+ unsigned long unused,
+ struct task_struct * p, struct pt_regs * regs)
+{
+ struct pt_regs * childregs;
+ struct task_struct *tsk;
+ int err;
+
+ childregs = task_pt_regs(p);
+ *childregs = *regs;
+ childregs->eax = 0;
+ childregs->esp = esp;
+
+ p->thread.esp = (unsigned long) childregs;
+ p->thread.esp0 = (unsigned long) (childregs+1);
+
+ p->thread.eip = (unsigned long) ret_from_fork;
+
+ savesegment(fs,p->thread.fs);
+ savesegment(gs,p->thread.gs);
+
+ tsk = current;
+ if (unlikely(test_tsk_thread_flag(tsk, TIF_IO_BITMAP))) {
+ p->thread.io_bitmap_ptr = kmalloc(IO_BITMAP_BYTES, GFP_KERNEL);
+ if (!p->thread.io_bitmap_ptr) {
+ p->thread.io_bitmap_max = 0;
+ return -ENOMEM;
+ }
+ memcpy(p->thread.io_bitmap_ptr, tsk->thread.io_bitmap_ptr,
+ IO_BITMAP_BYTES);
+ set_tsk_thread_flag(p, TIF_IO_BITMAP);
+ }
+
+ /*
+ * Set a new TLS for the child thread?
+ */
+ if (clone_flags & CLONE_SETTLS) {
+ struct desc_struct *desc;
+ struct user_desc info;
+ int idx;
+
+ err = -EFAULT;
+ if (copy_from_user(&info, (void __user *)childregs->esi, sizeof(info)))
+ goto out;
+ err = -EINVAL;
+ if (LDT_empty(&info))
+ goto out;
+
+ idx = info.entry_number;
+ if (idx < GDT_ENTRY_TLS_MIN || idx > GDT_ENTRY_TLS_MAX)
+ goto out;
+
+ desc = p->thread.tls_array + idx - GDT_ENTRY_TLS_MIN;
+ desc->a = LDT_entry_a(&info);
+ desc->b = LDT_entry_b(&info);
+ }
+
+ p->thread.iopl = current->thread.iopl;
+
+ err = 0;
+ out:
+ if (err && p->thread.io_bitmap_ptr) {
+ kfree(p->thread.io_bitmap_ptr);
+ p->thread.io_bitmap_max = 0;
+ }
+ return err;
+}
+
+/*
+ * fill in the user structure for a core dump..
+ */
+void dump_thread(struct pt_regs * regs, struct user * dump)
+{
+ int i;
+
+/* changed the size calculations - should hopefully work better. lbt */
+ dump->magic = CMAGIC;
+ dump->start_code = 0;
+ dump->start_stack = regs->esp & ~(PAGE_SIZE - 1);
+ dump->u_tsize = ((unsigned long) current->mm->end_code) >> PAGE_SHIFT;
+ dump->u_dsize = ((unsigned long) (current->mm->brk + (PAGE_SIZE-1))) >> PAGE_SHIFT;
+ dump->u_dsize -= dump->u_tsize;
+ dump->u_ssize = 0;
+ for (i = 0; i < 8; i++)
+ dump->u_debugreg[i] = current->thread.debugreg[i];
+
+ if (dump->start_stack < TASK_SIZE)
+ dump->u_ssize = ((unsigned long) (TASK_SIZE - dump->start_stack)) >> PAGE_SHIFT;
+
+ dump->regs.ebx = regs->ebx;
+ dump->regs.ecx = regs->ecx;
+ dump->regs.edx = regs->edx;
+ dump->regs.esi = regs->esi;
+ dump->regs.edi = regs->edi;
+ dump->regs.ebp = regs->ebp;
+ dump->regs.eax = regs->eax;
+ dump->regs.ds = regs->xds;
+ dump->regs.es = regs->xes;
+ savesegment(fs,dump->regs.fs);
+ savesegment(gs,dump->regs.gs);
+ dump->regs.orig_eax = regs->orig_eax;
+ dump->regs.eip = regs->eip;
+ dump->regs.cs = regs->xcs;
+ dump->regs.eflags = regs->eflags;
+ dump->regs.esp = regs->esp;
+ dump->regs.ss = regs->xss;
+
+ dump->u_fpvalid = dump_fpu (regs, &dump->i387);
+}
+EXPORT_SYMBOL(dump_thread);
+
+/*
+ * Capture the user space registers if the task is not running (in user space)
+ */
+int dump_task_regs(struct task_struct *tsk, elf_gregset_t *regs)
+{
+ struct pt_regs ptregs = *task_pt_regs(tsk);
+ ptregs.xcs &= 0xffff;
+ ptregs.xds &= 0xffff;
+ ptregs.xes &= 0xffff;
+ ptregs.xss &= 0xffff;
+
+ elf_core_copy_regs(regs, &ptregs);
+
+ return 1;
+}
+
+static noinline void __switch_to_xtra(struct task_struct *next_p)
+{
+ struct thread_struct *next;
+
+ next = &next_p->thread;
+
+ if (test_tsk_thread_flag(next_p, TIF_DEBUG)) {
+ set_debugreg(next->debugreg[0], 0);
+ set_debugreg(next->debugreg[1], 1);
+ set_debugreg(next->debugreg[2], 2);
+ set_debugreg(next->debugreg[3], 3);
+ /* no 4 and 5 */
+ set_debugreg(next->debugreg[6], 6);
+ set_debugreg(next->debugreg[7], 7);
+ }
+}
+
+/*
+ * This function selects if the context switch from prev to next
+ * has to tweak the TSC disable bit in the cr4.
+ */
+static inline void disable_tsc(struct task_struct *prev_p,
+ struct task_struct *next_p)
+{
+ struct thread_info *prev, *next;
+
+ /*
+ * gcc should eliminate the ->thread_info dereference if
+ * has_secure_computing returns 0 at compile time (SECCOMP=n).
+ */
+ prev = task_thread_info(prev_p);
+ next = task_thread_info(next_p);
+
+ if (has_secure_computing(prev) || has_secure_computing(next)) {
+ /* slow path here */
+ if (has_secure_computing(prev) &&
+ !has_secure_computing(next)) {
+ write_cr4(read_cr4() & ~X86_CR4_TSD);
+ } else if (!has_secure_computing(prev) &&
+ has_secure_computing(next))
+ write_cr4(read_cr4() | X86_CR4_TSD);
+ }
+}
+
+/*
+ * switch_to(x,yn) should switch tasks from x to y.
+ *
+ * We fsave/fwait so that an exception goes off at the right time
+ * (as a call from the fsave or fwait in effect) rather than to
+ * the wrong process. Lazy FP saving no longer makes any sense
+ * with modern CPU's, and this simplifies a lot of things (SMP
+ * and UP become the same).
+ *
+ * NOTE! We used to use the x86 hardware context switching. The
+ * reason for not using it any more becomes apparent when you
+ * try to recover gracefully from saved state that is no longer
+ * valid (stale segment register values in particular). With the
+ * hardware task-switch, there is no way to fix up bad state in
+ * a reasonable manner.
+ *
+ * The fact that Intel documents the hardware task-switching to
+ * be slow is a fairly red herring - this code is not noticeably
+ * faster. However, there _is_ some room for improvement here,
+ * so the performance issues may eventually be a valid point.
+ * More important, however, is the fact that this allows us much
+ * more flexibility.
+ *
+ * The return value (in %eax) will be the "prev" task after
+ * the task-switch, and shows up in ret_from_fork in entry.S,
+ * for example.
+ */
+struct task_struct fastcall * __switch_to(struct task_struct *prev_p, struct task_struct *next_p)
+{
+ struct thread_struct *prev = &prev_p->thread,
+ *next = &next_p->thread;
+ int cpu = smp_processor_id();
+#ifndef CONFIG_X86_NO_TSS
+ struct tss_struct *tss = &per_cpu(init_tss, cpu);
+#endif
+#if CONFIG_XEN_COMPAT > 0x030002
+ struct physdev_set_iopl iopl_op;
+ struct physdev_set_iobitmap iobmp_op;
+#else
+ struct physdev_op _pdo[2], *pdo = _pdo;
+#define iopl_op pdo->u.set_iopl
+#define iobmp_op pdo->u.set_iobitmap
+#endif
+ multicall_entry_t _mcl[8], *mcl = _mcl;
+
+ /* XEN NOTE: FS/GS saved in switch_mm(), not here. */
+
+ /*
+ * This is basically '__unlazy_fpu', except that we queue a
+ * multicall to indicate FPU task switch, rather than
+ * synchronously trapping to Xen.
+ */
+ if (prev_p->thread_info->status & TS_USEDFPU) {
+ __save_init_fpu(prev_p); /* _not_ save_init_fpu() */
+ mcl->op = __HYPERVISOR_fpu_taskswitch;
+ mcl->args[0] = 1;
+ mcl++;
+ }
+#if 0 /* lazy fpu sanity check */
+ else BUG_ON(!(read_cr0() & 8));
+#endif
+
+ /*
+ * Reload esp0.
+ * This is load_esp0(tss, next) with a multicall.
+ */
+ mcl->op = __HYPERVISOR_stack_switch;
+ mcl->args[0] = __KERNEL_DS;
+ mcl->args[1] = next->esp0;
+ mcl++;
+
+ /*
+ * Load the per-thread Thread-Local Storage descriptor.
+ * This is load_TLS(next, cpu) with multicalls.
+ */
+#define C(i) do { \
+ if (unlikely(next->tls_array[i].a != prev->tls_array[i].a || \
+ next->tls_array[i].b != prev->tls_array[i].b)) { \
+ mcl->op = __HYPERVISOR_update_descriptor; \
+ *(u64 *)&mcl->args[0] = virt_to_machine( \
+ &get_cpu_gdt_table(cpu)[GDT_ENTRY_TLS_MIN + i]);\
+ *(u64 *)&mcl->args[2] = *(u64 *)&next->tls_array[i]; \
+ mcl++; \
+ } \
+} while (0)
+ C(0); C(1); C(2);
+#undef C
+
+ if (unlikely(prev->iopl != next->iopl)) {
+ iopl_op.iopl = (next->iopl == 0) ? 1 : (next->iopl >> 12) & 3;
+#if CONFIG_XEN_COMPAT > 0x030002
+ mcl->op = __HYPERVISOR_physdev_op;
+ mcl->args[0] = PHYSDEVOP_set_iopl;
+ mcl->args[1] = (unsigned long)&iopl_op;
+#else
+ mcl->op = __HYPERVISOR_physdev_op_compat;
+ pdo->cmd = PHYSDEVOP_set_iopl;
+ mcl->args[0] = (unsigned long)pdo++;
+#endif
+ mcl++;
+ }
+
+ if (unlikely(prev->io_bitmap_ptr || next->io_bitmap_ptr)) {
+ set_xen_guest_handle(iobmp_op.bitmap,
+ (char *)next->io_bitmap_ptr);
+ iobmp_op.nr_ports = next->io_bitmap_ptr ? IO_BITMAP_BITS : 0;
+#if CONFIG_XEN_COMPAT > 0x030002
+ mcl->op = __HYPERVISOR_physdev_op;
+ mcl->args[0] = PHYSDEVOP_set_iobitmap;
+ mcl->args[1] = (unsigned long)&iobmp_op;
+#else
+ mcl->op = __HYPERVISOR_physdev_op_compat;
+ pdo->cmd = PHYSDEVOP_set_iobitmap;
+ mcl->args[0] = (unsigned long)pdo++;
+#endif
+ mcl++;
+ }
+
+#if CONFIG_XEN_COMPAT <= 0x030002
+ BUG_ON(pdo > _pdo + ARRAY_SIZE(_pdo));
+#endif
+ BUG_ON(mcl > _mcl + ARRAY_SIZE(_mcl));
+ if (unlikely(HYPERVISOR_multicall_check(_mcl, mcl - _mcl, NULL)))
+ BUG();
+
+ /*
+ * Restore %fs and %gs if needed.
+ *
+ * Glibc normally makes %fs be zero, and %gs is one of
+ * the TLS segments.
+ */
+ if (unlikely(next->fs))
+ loadsegment(fs, next->fs);
+
+ if (next->gs)
+ loadsegment(gs, next->gs);
+
+ /*
+ * Now maybe handle debug registers
+ */
+ if (unlikely(task_thread_info(next_p)->flags & _TIF_WORK_CTXSW))
+ __switch_to_xtra(next_p);
+
+ disable_tsc(prev_p, next_p);
+
+ return prev_p;
+}
+
+asmlinkage int sys_fork(struct pt_regs regs)
+{
+ return do_fork(SIGCHLD, regs.esp, &regs, 0, NULL, NULL);
+}
+
+asmlinkage int sys_clone(struct pt_regs regs)
+{
+ unsigned long clone_flags;
+ unsigned long newsp;
+ int __user *parent_tidptr, *child_tidptr;
+
+ clone_flags = regs.ebx;
+ newsp = regs.ecx;
+ parent_tidptr = (int __user *)regs.edx;
+ child_tidptr = (int __user *)regs.edi;
+ if (!newsp)
+ newsp = regs.esp;
+ return do_fork(clone_flags, newsp, &regs, 0, parent_tidptr, child_tidptr);
+}
+
+/*
+ * This is trivial, and on the face of it looks like it
+ * could equally well be done in user mode.
+ *
+ * Not so, for quite unobvious reasons - register pressure.
+ * In user mode vfork() cannot have a stack frame, and if
+ * done by calling the "clone()" system call directly, you
+ * do not have enough call-clobbered registers to hold all
+ * the information you need.
+ */
+asmlinkage int sys_vfork(struct pt_regs regs)
+{
+ return do_fork(CLONE_VFORK | CLONE_VM | SIGCHLD, regs.esp, &regs, 0, NULL, NULL);
+}
+
+/*
+ * sys_execve() executes a new program.
+ */
+asmlinkage int sys_execve(struct pt_regs regs)
+{
+ int error;
+ char * filename;
+
+ filename = getname((char __user *) regs.ebx);
+ error = PTR_ERR(filename);
+ if (IS_ERR(filename))
+ goto out;
+ error = do_execve(filename,
+ (char __user * __user *) regs.ecx,
+ (char __user * __user *) regs.edx,
+ &regs);
+ if (error == 0) {
+ task_lock(current);
+ current->ptrace &= ~PT_DTRACE;
+ task_unlock(current);
+ /* Make sure we don't return using sysenter.. */
+ set_thread_flag(TIF_IRET);
+ }
+ putname(filename);
+out:
+ return error;
+}
+
+#define top_esp (THREAD_SIZE - sizeof(unsigned long))
+#define top_ebp (THREAD_SIZE - 2*sizeof(unsigned long))
+
+unsigned long get_wchan(struct task_struct *p)
+{
+ unsigned long ebp, esp, eip;
+ unsigned long stack_page;
+ int count = 0;
+ if (!p || p == current || p->state == TASK_RUNNING)
+ return 0;
+ stack_page = (unsigned long)task_stack_page(p);
+ esp = p->thread.esp;
+ if (!stack_page || esp < stack_page || esp > top_esp+stack_page)
+ return 0;
+ /* include/asm-i386/system.h:switch_to() pushes ebp last. */
+ ebp = *(unsigned long *) esp;
+ do {
+ if (ebp < stack_page || ebp > top_ebp+stack_page)
+ return 0;
+ eip = *(unsigned long *) (ebp+4);
+ if (!in_sched_functions(eip))
+ return eip;
+ ebp = *(unsigned long *) ebp;
+ } while (count++ < 16);
+ return 0;
+}
+
+/*
+ * sys_alloc_thread_area: get a yet unused TLS descriptor index.
+ */
+static int get_free_idx(void)
+{
+ struct thread_struct *t = &current->thread;
+ int idx;
+
+ for (idx = 0; idx < GDT_ENTRY_TLS_ENTRIES; idx++)
+ if (desc_empty(t->tls_array + idx))
+ return idx + GDT_ENTRY_TLS_MIN;
+ return -ESRCH;
+}
+
+/*
+ * Set a given TLS descriptor:
+ */
+asmlinkage int sys_set_thread_area(struct user_desc __user *u_info)
+{
+ struct thread_struct *t = &current->thread;
+ struct user_desc info;
+ struct desc_struct *desc;
+ int cpu, idx;
+
+ if (copy_from_user(&info, u_info, sizeof(info)))
+ return -EFAULT;
+ idx = info.entry_number;
+
+ /*
+ * index -1 means the kernel should try to find and
+ * allocate an empty descriptor:
+ */
+ if (idx == -1) {
+ idx = get_free_idx();
+ if (idx < 0)
+ return idx;
+ if (put_user(idx, &u_info->entry_number))
+ return -EFAULT;
+ }
+
+ if (idx < GDT_ENTRY_TLS_MIN || idx > GDT_ENTRY_TLS_MAX)
+ return -EINVAL;
+
+ desc = t->tls_array + idx - GDT_ENTRY_TLS_MIN;
+
+ /*
+ * We must not get preempted while modifying the TLS.
+ */
+ cpu = get_cpu();
+
+ if (LDT_empty(&info)) {
+ desc->a = 0;
+ desc->b = 0;
+ } else {
+ desc->a = LDT_entry_a(&info);
+ desc->b = LDT_entry_b(&info);
+ }
+ load_TLS(t, cpu);
+
+ put_cpu();
+
+ return 0;
+}
+
+/*
+ * Get the current Thread-Local Storage area:
+ */
+
+#define GET_BASE(desc) ( \
+ (((desc)->a >> 16) & 0x0000ffff) | \
+ (((desc)->b << 16) & 0x00ff0000) | \
+ ( (desc)->b & 0xff000000) )
+
+#define GET_LIMIT(desc) ( \
+ ((desc)->a & 0x0ffff) | \
+ ((desc)->b & 0xf0000) )
+
+#define GET_32BIT(desc) (((desc)->b >> 22) & 1)
+#define GET_CONTENTS(desc) (((desc)->b >> 10) & 3)
+#define GET_WRITABLE(desc) (((desc)->b >> 9) & 1)
+#define GET_LIMIT_PAGES(desc) (((desc)->b >> 23) & 1)
+#define GET_PRESENT(desc) (((desc)->b >> 15) & 1)
+#define GET_USEABLE(desc) (((desc)->b >> 20) & 1)
+
+asmlinkage int sys_get_thread_area(struct user_desc __user *u_info)
+{
+ struct user_desc info;
+ struct desc_struct *desc;
+ int idx;
+
+ if (get_user(idx, &u_info->entry_number))
+ return -EFAULT;
+ if (idx < GDT_ENTRY_TLS_MIN || idx > GDT_ENTRY_TLS_MAX)
+ return -EINVAL;
+
+ memset(&info, 0, sizeof(info));
+
+ desc = current->thread.tls_array + idx - GDT_ENTRY_TLS_MIN;
+
+ info.entry_number = idx;
+ info.base_addr = GET_BASE(desc);
+ info.limit = GET_LIMIT(desc);
+ info.seg_32bit = GET_32BIT(desc);
+ info.contents = GET_CONTENTS(desc);
+ info.read_exec_only = !GET_WRITABLE(desc);
+ info.limit_in_pages = GET_LIMIT_PAGES(desc);
+ info.seg_not_present = !GET_PRESENT(desc);
+ info.useable = GET_USEABLE(desc);
+
+ if (copy_to_user(u_info, &info, sizeof(info)))
+ return -EFAULT;
+ return 0;
+}
+
+unsigned long arch_align_stack(unsigned long sp)
+{
+ if (randomize_va_space)
+ sp -= get_random_int() % 8192;
+ return sp & ~0xf;
+}
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/kernel/setup_32-xen.c 2008-04-22 15:41:51.000000000 +0200
@@ -0,0 +1,1919 @@
+/*
+ * linux/arch/i386/kernel/setup.c
+ *
+ * Copyright (C) 1995 Linus Torvalds
+ *
+ * Support of BIGMEM added by Gerhard Wichert, Siemens AG, July 1999
+ *
+ * Memory region support
+ * David Parsons <orc@pell.chi.il.us>, July-August 1999
+ *
+ * Added E820 sanitization routine (removes overlapping memory regions);
+ * Brian Moyle <bmoyle@mvista.com>, February 2001
+ *
+ * Moved CPU detection code to cpu/${cpu}.c
+ * Patrick Mochel <mochel@osdl.org>, March 2002
+ *
+ * Provisions for empty E820 memory regions (reported by certain BIOSes).
+ * Alex Achenbach <xela@slit.de>, December 2002.
+ *
+ */
+
+/*
+ * This file handles the architecture-dependent parts of initialization
+ */
+
+#include <linux/sched.h>
+#include <linux/mm.h>
+#include <linux/mmzone.h>
+#include <linux/screen_info.h>
+#include <linux/ioport.h>
+#include <linux/acpi.h>
+#include <linux/apm_bios.h>
+#include <linux/initrd.h>
+#include <linux/bootmem.h>
+#include <linux/seq_file.h>
+#include <linux/platform_device.h>
+#include <linux/console.h>
+#include <linux/mca.h>
+#include <linux/root_dev.h>
+#include <linux/highmem.h>
+#include <linux/module.h>
+#include <linux/efi.h>
+#include <linux/init.h>
+#include <linux/edd.h>
+#include <linux/nodemask.h>
+#include <linux/kernel.h>
+#include <linux/percpu.h>
+#include <linux/notifier.h>
+#include <linux/kexec.h>
+#include <linux/crash_dump.h>
+#include <linux/dmi.h>
+#include <linux/pfn.h>
+
+#include <video/edid.h>
+
+#include <asm/apic.h>
+#include <asm/e820.h>
+#include <asm/mpspec.h>
+#include <asm/setup.h>
+#include <asm/arch_hooks.h>
+#include <asm/sections.h>
+#include <asm/io_apic.h>
+#include <asm/ist.h>
+#include <asm/io.h>
+#include <asm/hypervisor.h>
+#include <xen/interface/physdev.h>
+#include <xen/interface/memory.h>
+#include <xen/features.h>
+#include <xen/firmware.h>
+#include <xen/xencons.h>
+#include <setup_arch.h>
+#include <bios_ebda.h>
+
+#ifdef CONFIG_XEN
+#include <xen/interface/kexec.h>
+#endif
+
+/* Forward Declaration. */
+void __init find_max_pfn(void);
+
+static int xen_panic_event(struct notifier_block *, unsigned long, void *);
+static struct notifier_block xen_panic_block = {
+ xen_panic_event, NULL, 0 /* try to go last */
+};
+
+extern char hypercall_page[PAGE_SIZE];
+EXPORT_SYMBOL(hypercall_page);
+
+int disable_pse __devinitdata = 0;
+
+/*
+ * Machine setup..
+ */
+
+#ifdef CONFIG_EFI
+int efi_enabled = 0;
+EXPORT_SYMBOL(efi_enabled);
+#endif
+
+/* cpu data as detected by the assembly code in head.S */
+struct cpuinfo_x86 new_cpu_data __initdata = { 0, 0, 0, 0, -1, 1, 0, 0, -1 };
+/* common cpu data for all cpus */
+struct cpuinfo_x86 boot_cpu_data __read_mostly = { 0, 0, 0, 0, -1, 1, 0, 0, -1 };
+EXPORT_SYMBOL(boot_cpu_data);
+
+unsigned long mmu_cr4_features;
+
+#ifdef CONFIG_ACPI
+ int acpi_disabled = 0;
+#else
+ int acpi_disabled = 1;
+#endif
+EXPORT_SYMBOL(acpi_disabled);
+
+#ifdef CONFIG_ACPI
+int __initdata acpi_force = 0;
+extern acpi_interrupt_flags acpi_sci_flags;
+#endif
+
+/* for MCA, but anyone else can use it if they want */
+unsigned int machine_id;
+#ifdef CONFIG_MCA
+EXPORT_SYMBOL(machine_id);
+#endif
+unsigned int machine_submodel_id;
+unsigned int BIOS_revision;
+unsigned int mca_pentium_flag;
+
+/* For PCI or other memory-mapped resources */
+unsigned long pci_mem_start = 0x10000000;
+#ifdef CONFIG_PCI
+EXPORT_SYMBOL(pci_mem_start);
+#endif
+
+/* Boot loader ID as an integer, for the benefit of proc_dointvec */
+int bootloader_type;
+
+/* user-defined highmem size */
+static unsigned int highmem_pages = -1;
+
+/*
+ * Setup options
+ */
+struct drive_info_struct { char dummy[32]; } drive_info;
+#if defined(CONFIG_BLK_DEV_IDE) || defined(CONFIG_BLK_DEV_HD) || \
+ defined(CONFIG_BLK_DEV_IDE_MODULE) || defined(CONFIG_BLK_DEV_HD_MODULE)
+EXPORT_SYMBOL(drive_info);
+#endif
+struct screen_info screen_info;
+EXPORT_SYMBOL(screen_info);
+struct apm_info apm_info;
+EXPORT_SYMBOL(apm_info);
+struct sys_desc_table_struct {
+ unsigned short length;
+ unsigned char table[0];
+};
+struct edid_info edid_info;
+EXPORT_SYMBOL_GPL(edid_info);
+#ifndef CONFIG_XEN
+#define copy_edid() (edid_info = EDID_INFO)
+#endif
+struct ist_info ist_info;
+#if defined(CONFIG_X86_SPEEDSTEP_SMI) || \
+ defined(CONFIG_X86_SPEEDSTEP_SMI_MODULE)
+EXPORT_SYMBOL(ist_info);
+#endif
+struct e820map e820;
+#ifdef CONFIG_XEN
+struct e820map machine_e820;
+#endif
+
+extern void early_cpu_init(void);
+extern void generic_apic_probe(char *);
+extern int root_mountflags;
+
+unsigned long saved_videomode;
+
+#define RAMDISK_IMAGE_START_MASK 0x07FF
+#define RAMDISK_PROMPT_FLAG 0x8000
+#define RAMDISK_LOAD_FLAG 0x4000
+
+static char command_line[COMMAND_LINE_SIZE];
+
+unsigned char __initdata boot_params[PARAM_SIZE];
+
+static struct resource data_resource = {
+ .name = "Kernel data",
+ .start = 0,
+ .end = 0,
+ .flags = IORESOURCE_BUSY | IORESOURCE_MEM
+};
+
+static struct resource code_resource = {
+ .name = "Kernel code",
+ .start = 0,
+ .end = 0,
+ .flags = IORESOURCE_BUSY | IORESOURCE_MEM
+};
+
+static struct resource system_rom_resource = {
+ .name = "System ROM",
+ .start = 0xf0000,
+ .end = 0xfffff,
+ .flags = IORESOURCE_BUSY | IORESOURCE_READONLY | IORESOURCE_MEM
+};
+
+static struct resource extension_rom_resource = {
+ .name = "Extension ROM",
+ .start = 0xe0000,
+ .end = 0xeffff,
+ .flags = IORESOURCE_BUSY | IORESOURCE_READONLY | IORESOURCE_MEM
+};
+
+static struct resource adapter_rom_resources[] = { {
+ .name = "Adapter ROM",
+ .start = 0xc8000,
+ .end = 0,
+ .flags = IORESOURCE_BUSY | IORESOURCE_READONLY | IORESOURCE_MEM
+}, {
+ .name = "Adapter ROM",
+ .start = 0,
+ .end = 0,
+ .flags = IORESOURCE_BUSY | IORESOURCE_READONLY | IORESOURCE_MEM
+}, {
+ .name = "Adapter ROM",
+ .start = 0,
+ .end = 0,
+ .flags = IORESOURCE_BUSY | IORESOURCE_READONLY | IORESOURCE_MEM
+}, {
+ .name = "Adapter ROM",
+ .start = 0,
+ .end = 0,
+ .flags = IORESOURCE_BUSY | IORESOURCE_READONLY | IORESOURCE_MEM
+}, {
+ .name = "Adapter ROM",
+ .start = 0,
+ .end = 0,
+ .flags = IORESOURCE_BUSY | IORESOURCE_READONLY | IORESOURCE_MEM
+}, {
+ .name = "Adapter ROM",
+ .start = 0,
+ .end = 0,
+ .flags = IORESOURCE_BUSY | IORESOURCE_READONLY | IORESOURCE_MEM
+} };
+
+#define ADAPTER_ROM_RESOURCES \
+ (sizeof adapter_rom_resources / sizeof adapter_rom_resources[0])
+
+static struct resource video_rom_resource = {
+ .name = "Video ROM",
+ .start = 0xc0000,
+ .end = 0xc7fff,
+ .flags = IORESOURCE_BUSY | IORESOURCE_READONLY | IORESOURCE_MEM
+};
+
+static struct resource video_ram_resource = {
+ .name = "Video RAM area",
+ .start = 0xa0000,
+ .end = 0xbffff,
+ .flags = IORESOURCE_BUSY | IORESOURCE_MEM
+};
+
+static struct resource standard_io_resources[] = { {
+ .name = "dma1",
+ .start = 0x0000,
+ .end = 0x001f,
+ .flags = IORESOURCE_BUSY | IORESOURCE_IO
+}, {
+ .name = "pic1",
+ .start = 0x0020,
+ .end = 0x0021,
+ .flags = IORESOURCE_BUSY | IORESOURCE_IO
+}, {
+ .name = "timer0",
+ .start = 0x0040,
+ .end = 0x0043,
+ .flags = IORESOURCE_BUSY | IORESOURCE_IO
+}, {
+ .name = "timer1",
+ .start = 0x0050,
+ .end = 0x0053,
+ .flags = IORESOURCE_BUSY | IORESOURCE_IO
+}, {
+ .name = "keyboard",
+ .start = 0x0060,
+ .end = 0x006f,
+ .flags = IORESOURCE_BUSY | IORESOURCE_IO
+}, {
+ .name = "dma page reg",
+ .start = 0x0080,
+ .end = 0x008f,
+ .flags = IORESOURCE_BUSY | IORESOURCE_IO
+}, {
+ .name = "pic2",
+ .start = 0x00a0,
+ .end = 0x00a1,
+ .flags = IORESOURCE_BUSY | IORESOURCE_IO
+}, {
+ .name = "dma2",
+ .start = 0x00c0,
+ .end = 0x00df,
+ .flags = IORESOURCE_BUSY | IORESOURCE_IO
+}, {
+ .name = "fpu",
+ .start = 0x00f0,
+ .end = 0x00ff,
+ .flags = IORESOURCE_BUSY | IORESOURCE_IO
+} };
+
+#define STANDARD_IO_RESOURCES \
+ (sizeof standard_io_resources / sizeof standard_io_resources[0])
+
+#define romsignature(x) (*(unsigned short *)(x) == 0xaa55)
+
+static int __init romchecksum(unsigned char *rom, unsigned long length)
+{
+ unsigned char *p, sum = 0;
+
+ for (p = rom; p < rom + length; p++)
+ sum += *p;
+ return sum == 0;
+}
+
+static void __init probe_roms(void)
+{
+ unsigned long start, length, upper;
+ unsigned char *rom;
+ int i;
+
+#ifdef CONFIG_XEN
+ /* Nothing to do if not running in dom0. */
+ if (!is_initial_xendomain())
+ return;
+#endif
+
+ /* video rom */
+ upper = adapter_rom_resources[0].start;
+ for (start = video_rom_resource.start; start < upper; start += 2048) {
+ rom = isa_bus_to_virt(start);
+ if (!romsignature(rom))
+ continue;
+
+ video_rom_resource.start = start;
+
+ /* 0 < length <= 0x7f * 512, historically */
+ length = rom[2] * 512;
+
+ /* if checksum okay, trust length byte */
+ if (length && romchecksum(rom, length))
+ video_rom_resource.end = start + length - 1;
+
+ request_resource(&iomem_resource, &video_rom_resource);
+ break;
+ }
+
+ start = (video_rom_resource.end + 1 + 2047) & ~2047UL;
+ if (start < upper)
+ start = upper;
+
+ /* system rom */
+ request_resource(&iomem_resource, &system_rom_resource);
+ upper = system_rom_resource.start;
+
+ /* check for extension rom (ignore length byte!) */
+ rom = isa_bus_to_virt(extension_rom_resource.start);
+ if (romsignature(rom)) {
+ length = extension_rom_resource.end - extension_rom_resource.start + 1;
+ if (romchecksum(rom, length)) {
+ request_resource(&iomem_resource, &extension_rom_resource);
+ upper = extension_rom_resource.start;
+ }
+ }
+
+ /* check for adapter roms on 2k boundaries */
+ for (i = 0; i < ADAPTER_ROM_RESOURCES && start < upper; start += 2048) {
+ rom = isa_bus_to_virt(start);
+ if (!romsignature(rom))
+ continue;
+
+ /* 0 < length <= 0x7f * 512, historically */
+ length = rom[2] * 512;
+
+ /* but accept any length that fits if checksum okay */
+ if (!length || start + length > upper || !romchecksum(rom, length))
+ continue;
+
+ adapter_rom_resources[i].start = start;
+ adapter_rom_resources[i].end = start + length - 1;
+ request_resource(&iomem_resource, &adapter_rom_resources[i]);
+
+ start = adapter_rom_resources[i++].end & ~2047UL;
+ }
+}
+
+/*
+ * Point at the empty zero page to start with. We map the real shared_info
+ * page as soon as fixmap is up and running.
+ */
+shared_info_t *HYPERVISOR_shared_info = (shared_info_t *)empty_zero_page;
+EXPORT_SYMBOL(HYPERVISOR_shared_info);
+
+unsigned long *phys_to_machine_mapping;
+unsigned long *pfn_to_mfn_frame_list_list, *pfn_to_mfn_frame_list[16];
+EXPORT_SYMBOL(phys_to_machine_mapping);
+
+/* Raw start-of-day parameters from the hypervisor. */
+start_info_t *xen_start_info;
+EXPORT_SYMBOL(xen_start_info);
+
+void __init add_memory_region(unsigned long long start,
+ unsigned long long size, int type)
+{
+ int x;
+
+ if (!efi_enabled) {
+ x = e820.nr_map;
+
+ if (x == E820MAX) {
+ printk(KERN_ERR "Ooops! Too many entries in the memory map!\n");
+ return;
+ }
+
+ e820.map[x].addr = start;
+ e820.map[x].size = size;
+ e820.map[x].type = type;
+ e820.nr_map++;
+ }
+} /* add_memory_region */
+
+static void __init limit_regions(unsigned long long size)
+{
+ unsigned long long current_addr = 0;
+ int i;
+
+ if (efi_enabled) {
+ efi_memory_desc_t *md;
+ void *p;
+
+ for (p = memmap.map, i = 0; p < memmap.map_end;
+ p += memmap.desc_size, i++) {
+ md = p;
+ current_addr = md->phys_addr + (md->num_pages << 12);
+ if (md->type == EFI_CONVENTIONAL_MEMORY) {
+ if (current_addr >= size) {
+ md->num_pages -=
+ (((current_addr-size) + PAGE_SIZE-1) >> PAGE_SHIFT);
+ memmap.nr_map = i + 1;
+ return;
+ }
+ }
+ }
+ }
+ for (i = 0; i < e820.nr_map; i++) {
+ current_addr = e820.map[i].addr + e820.map[i].size;
+ if (current_addr < size)
+ continue;
+
+ if (e820.map[i].type != E820_RAM)
+ continue;
+
+ if (e820.map[i].addr >= size) {
+ /*
+ * This region starts past the end of the
+ * requested size, skip it completely.
+ */
+ e820.nr_map = i;
+ } else {
+ e820.nr_map = i + 1;
+ e820.map[i].size -= current_addr - size;
+ }
+ return;
+ }
+#ifdef CONFIG_XEN
+ if (i==e820.nr_map && current_addr < size) {
+ /*
+ * The e820 map finished before our requested size so
+ * extend the final entry to the requested address.
+ */
+ --i;
+ if (e820.map[i].type == E820_RAM)
+ e820.map[i].size -= current_addr - size;
+ else
+ add_memory_region(current_addr, size - current_addr, E820_RAM);
+ }
+#endif
+}
+
+#define E820_DEBUG 1
+
+static void __init print_memory_map(char *who)
+{
+ int i;
+
+ for (i = 0; i < e820.nr_map; i++) {
+ printk(" %s: %016Lx - %016Lx ", who,
+ e820.map[i].addr,
+ e820.map[i].addr + e820.map[i].size);
+ switch (e820.map[i].type) {
+ case E820_RAM: printk("(usable)\n");
+ break;
+ case E820_RESERVED:
+ printk("(reserved)\n");
+ break;
+ case E820_ACPI:
+ printk("(ACPI data)\n");
+ break;
+ case E820_NVS:
+ printk("(ACPI NVS)\n");
+ break;
+ default: printk("type %lu\n", e820.map[i].type);
+ break;
+ }
+ }
+}
+
+/*
+ * Sanitize the BIOS e820 map.
+ *
+ * Some e820 responses include overlapping entries. The following
+ * replaces the original e820 map with a new one, removing overlaps.
+ *
+ */
+struct change_member {
+ struct e820entry *pbios; /* pointer to original bios entry */
+ unsigned long long addr; /* address for this change point */
+};
+static struct change_member change_point_list[2*E820MAX] __initdata;
+static struct change_member *change_point[2*E820MAX] __initdata;
+static struct e820entry *overlap_list[E820MAX] __initdata;
+static struct e820entry new_bios[E820MAX] __initdata;
+
+int __init sanitize_e820_map(struct e820entry * biosmap, char * pnr_map)
+{
+ struct change_member *change_tmp;
+ unsigned long current_type, last_type;
+ unsigned long long last_addr;
+ int chgidx, still_changing;
+ int overlap_entries;
+ int new_bios_entry;
+ int old_nr, new_nr, chg_nr;
+ int i;
+
+ /*
+ Visually we're performing the following (1,2,3,4 = memory types)...
+
+ Sample memory map (w/overlaps):
+ ____22__________________
+ ______________________4_
+ ____1111________________
+ _44_____________________
+ 11111111________________
+ ____________________33__
+ ___________44___________
+ __________33333_________
+ ______________22________
+ ___________________2222_
+ _________111111111______
+ _____________________11_
+ _________________4______
+
+ Sanitized equivalent (no overlap):
+ 1_______________________
+ _44_____________________
+ ___1____________________
+ ____22__________________
+ ______11________________
+ _________1______________
+ __________3_____________
+ ___________44___________
+ _____________33_________
+ _______________2________
+ ________________1_______
+ _________________4______
+ ___________________2____
+ ____________________33__
+ ______________________4_
+ */
+
+ /* if there's only one memory region, don't bother */
+ if (*pnr_map < 2)
+ return -1;
+
+ old_nr = *pnr_map;
+
+ /* bail out if we find any unreasonable addresses in bios map */
+ for (i=0; i<old_nr; i++)
+ if (biosmap[i].addr + biosmap[i].size < biosmap[i].addr)
+ return -1;
+
+ /* create pointers for initial change-point information (for sorting) */
+ for (i=0; i < 2*old_nr; i++)
+ change_point[i] = &change_point_list[i];
+
+ /* record all known change-points (starting and ending addresses),
+ omitting those that are for empty memory regions */
+ chgidx = 0;
+ for (i=0; i < old_nr; i++) {
+ if (biosmap[i].size != 0) {
+ change_point[chgidx]->addr = biosmap[i].addr;
+ change_point[chgidx++]->pbios = &biosmap[i];
+ change_point[chgidx]->addr = biosmap[i].addr + biosmap[i].size;
+ change_point[chgidx++]->pbios = &biosmap[i];
+ }
+ }
+ chg_nr = chgidx; /* true number of change-points */
+
+ /* sort change-point list by memory addresses (low -> high) */
+ still_changing = 1;
+ while (still_changing) {
+ still_changing = 0;
+ for (i=1; i < chg_nr; i++) {
+ /* if <current_addr> > <last_addr>, swap */
+ /* or, if current=<start_addr> & last=<end_addr>, swap */
+ if ((change_point[i]->addr < change_point[i-1]->addr) ||
+ ((change_point[i]->addr == change_point[i-1]->addr) &&
+ (change_point[i]->addr == change_point[i]->pbios->addr) &&
+ (change_point[i-1]->addr != change_point[i-1]->pbios->addr))
+ )
+ {
+ change_tmp = change_point[i];
+ change_point[i] = change_point[i-1];
+ change_point[i-1] = change_tmp;
+ still_changing=1;
+ }
+ }
+ }
+
+ /* create a new bios memory map, removing overlaps */
+ overlap_entries=0; /* number of entries in the overlap table */
+ new_bios_entry=0; /* index for creating new bios map entries */
+ last_type = 0; /* start with undefined memory type */
+ last_addr = 0; /* start with 0 as last starting address */
+ /* loop through change-points, determining affect on the new bios map */
+ for (chgidx=0; chgidx < chg_nr; chgidx++)
+ {
+ /* keep track of all overlapping bios entries */
+ if (change_point[chgidx]->addr == change_point[chgidx]->pbios->addr)
+ {
+ /* add map entry to overlap list (> 1 entry implies an overlap) */
+ overlap_list[overlap_entries++]=change_point[chgidx]->pbios;
+ }
+ else
+ {
+ /* remove entry from list (order independent, so swap with last) */
+ for (i=0; i<overlap_entries; i++)
+ {
+ if (overlap_list[i] == change_point[chgidx]->pbios)
+ overlap_list[i] = overlap_list[overlap_entries-1];
+ }
+ overlap_entries--;
+ }
+ /* if there are overlapping entries, decide which "type" to use */
+ /* (larger value takes precedence -- 1=usable, 2,3,4,4+=unusable) */
+ current_type = 0;
+ for (i=0; i<overlap_entries; i++)
+ if (overlap_list[i]->type > current_type)
+ current_type = overlap_list[i]->type;
+ /* continue building up new bios map based on this information */
+ if (current_type != last_type) {
+ if (last_type != 0) {
+ new_bios[new_bios_entry].size =
+ change_point[chgidx]->addr - last_addr;
+ /* move forward only if the new size was non-zero */
+ if (new_bios[new_bios_entry].size != 0)
+ if (++new_bios_entry >= E820MAX)
+ break; /* no more space left for new bios entries */
+ }
+ if (current_type != 0) {
+ new_bios[new_bios_entry].addr = change_point[chgidx]->addr;
+ new_bios[new_bios_entry].type = current_type;
+ last_addr=change_point[chgidx]->addr;
+ }
+ last_type = current_type;
+ }
+ }
+ new_nr = new_bios_entry; /* retain count for new bios entries */
+
+ /* copy new bios mapping into original location */
+ memcpy(biosmap, new_bios, new_nr*sizeof(struct e820entry));
+ *pnr_map = new_nr;
+
+ return 0;
+}
+
+/*
+ * Copy the BIOS e820 map into a safe place.
+ *
+ * Sanity-check it while we're at it..
+ *
+ * If we're lucky and live on a modern system, the setup code
+ * will have given us a memory map that we can use to properly
+ * set up memory. If we aren't, we'll fake a memory map.
+ *
+ * We check to see that the memory map contains at least 2 elements
+ * before we'll use it, because the detection code in setup.S may
+ * not be perfect and most every PC known to man has two memory
+ * regions: one from 0 to 640k, and one from 1mb up. (The IBM
+ * thinkpad 560x, for example, does not cooperate with the memory
+ * detection code.)
+ */
+int __init copy_e820_map(struct e820entry * biosmap, int nr_map)
+{
+#ifndef CONFIG_XEN
+ /* Only one memory region (or negative)? Ignore it */
+ if (nr_map < 2)
+ return -1;
+#else
+ BUG_ON(nr_map < 1);
+#endif
+
+ do {
+ unsigned long long start = biosmap->addr;
+ unsigned long long size = biosmap->size;
+ unsigned long long end = start + size;
+ unsigned long type = biosmap->type;
+
+ /* Overflow in 64 bits? Ignore the memory map. */
+ if (start > end)
+ return -1;
+
+#ifndef CONFIG_XEN
+ /*
+ * Some BIOSes claim RAM in the 640k - 1M region.
+ * Not right. Fix it up.
+ */
+ if (type == E820_RAM) {
+ if (start < 0x100000ULL && end > 0xA0000ULL) {
+ if (start < 0xA0000ULL)
+ add_memory_region(start, 0xA0000ULL-start, type);
+ if (end <= 0x100000ULL)
+ continue;
+ start = 0x100000ULL;
+ size = end - start;
+ }
+ }
+#endif
+ add_memory_region(start, size, type);
+ } while (biosmap++,--nr_map);
+
+#ifdef CONFIG_XEN
+ if (is_initial_xendomain()) {
+ struct xen_memory_map memmap;
+
+ memmap.nr_entries = E820MAX;
+ set_xen_guest_handle(memmap.buffer, machine_e820.map);
+
+ if (HYPERVISOR_memory_op(XENMEM_machine_memory_map, &memmap))
+ BUG();
+ machine_e820.nr_map = memmap.nr_entries;
+ } else
+ machine_e820 = e820;
+#endif
+
+ return 0;
+}
+
+#if defined(CONFIG_EDD) || defined(CONFIG_EDD_MODULE)
+struct edd edd;
+#ifdef CONFIG_EDD_MODULE
+EXPORT_SYMBOL(edd);
+#endif
+#ifndef CONFIG_XEN
+/**
+ * copy_edd() - Copy the BIOS EDD information
+ * from boot_params into a safe place.
+ *
+ */
+static inline void copy_edd(void)
+{
+ memcpy(edd.mbr_signature, EDD_MBR_SIGNATURE, sizeof(edd.mbr_signature));
+ memcpy(edd.edd_info, EDD_BUF, sizeof(edd.edd_info));
+ edd.mbr_signature_nr = EDD_MBR_SIG_NR;
+ edd.edd_info_nr = EDD_NR;
+}
+#endif
+#else
+static inline void copy_edd(void)
+{
+}
+#endif
+
+static void __init parse_cmdline_early (char ** cmdline_p)
+{
+ char c = ' ', *to = command_line, *from = saved_command_line;
+ int len = 0, max_cmdline;
+ int userdef = 0;
+
+ if ((max_cmdline = MAX_GUEST_CMDLINE) > COMMAND_LINE_SIZE)
+ max_cmdline = COMMAND_LINE_SIZE;
+ memcpy(saved_command_line, xen_start_info->cmd_line, max_cmdline);
+ /* Save unparsed command line copy for /proc/cmdline */
+ saved_command_line[max_cmdline-1] = '\0';
+
+ for (;;) {
+ if (c != ' ')
+ goto next_char;
+ /*
+ * "mem=nopentium" disables the 4MB page tables.
+ * "mem=XXX[kKmM]" defines a memory region from HIGH_MEM
+ * to <mem>, overriding the bios size.
+ * "memmap=XXX[KkmM]@XXX[KkmM]" defines a memory region from
+ * <start> to <start>+<mem>, overriding the bios size.
+ *
+ * HPA tells me bootloaders need to parse mem=, so no new
+ * option should be mem= [also see Documentation/i386/boot.txt]
+ */
+ if (!memcmp(from, "mem=", 4)) {
+ if (to != command_line)
+ to--;
+ if (!memcmp(from+4, "nopentium", 9)) {
+ from += 9+4;
+ clear_bit(X86_FEATURE_PSE, boot_cpu_data.x86_capability);
+ disable_pse = 1;
+ } else {
+ /* If the user specifies memory size, we
+ * limit the BIOS-provided memory map to
+ * that size. exactmap can be used to specify
+ * the exact map. mem=number can be used to
+ * trim the existing memory map.
+ */
+ unsigned long long mem_size;
+
+ mem_size = memparse(from+4, &from);
+ limit_regions(mem_size);
+ userdef=1;
+ }
+ }
+
+ else if (!memcmp(from, "memmap=", 7)) {
+ if (to != command_line)
+ to--;
+ if (!memcmp(from+7, "exactmap", 8)) {
+#ifdef CONFIG_CRASH_DUMP
+ /* If we are doing a crash dump, we
+ * still need to know the real mem
+ * size before original memory map is
+ * reset.
+ */
+ find_max_pfn();
+ saved_max_pfn = max_pfn;
+#endif
+ from += 8+7;
+ e820.nr_map = 0;
+ userdef = 1;
+ } else {
+ /* If the user specifies memory size, we
+ * limit the BIOS-provided memory map to
+ * that size. exactmap can be used to specify
+ * the exact map. mem=number can be used to
+ * trim the existing memory map.
+ */
+ unsigned long long start_at, mem_size;
+
+ mem_size = memparse(from+7, &from);
+ if (*from == '@') {
+ start_at = memparse(from+1, &from);
+ add_memory_region(start_at, mem_size, E820_RAM);
+ } else if (*from == '#') {
+ start_at = memparse(from+1, &from);
+ add_memory_region(start_at, mem_size, E820_ACPI);
+ } else if (*from == '$') {
+ start_at = memparse(from+1, &from);
+ add_memory_region(start_at, mem_size, E820_RESERVED);
+ } else {
+ limit_regions(mem_size);
+ userdef=1;
+ }
+ }
+ }
+
+ else if (!memcmp(from, "noexec=", 7))
+ noexec_setup(from + 7);
+
+
+#ifdef CONFIG_X86_MPPARSE
+ /*
+ * If the BIOS enumerates physical processors before logical,
+ * maxcpus=N at enumeration-time can be used to disable HT.
+ */
+ else if (!memcmp(from, "maxcpus=", 8)) {
+ extern unsigned int maxcpus;
+
+ maxcpus = simple_strtoul(from + 8, NULL, 0);
+ }
+#endif
+
+#ifdef CONFIG_ACPI
+ /* "acpi=off" disables both ACPI table parsing and interpreter */
+ else if (!memcmp(from, "acpi=off", 8)) {
+ disable_acpi();
+ }
+
+ /* acpi=force to over-ride black-list */
+ else if (!memcmp(from, "acpi=force", 10)) {
+ acpi_force = 1;
+ acpi_ht = 1;
+ acpi_disabled = 0;
+ }
+
+ /* acpi=strict disables out-of-spec workarounds */
+ else if (!memcmp(from, "acpi=strict", 11)) {
+ acpi_strict = 1;
+ }
+
+ /* Limit ACPI just to boot-time to enable HT */
+ else if (!memcmp(from, "acpi=ht", 7)) {
+ if (!acpi_force)
+ disable_acpi();
+ acpi_ht = 1;
+ }
+
+ /* "pci=noacpi" disable ACPI IRQ routing and PCI scan */
+ else if (!memcmp(from, "pci=noacpi", 10)) {
+ acpi_disable_pci();
+ }
+ /* "acpi=noirq" disables ACPI interrupt routing */
+ else if (!memcmp(from, "acpi=noirq", 10)) {
+ acpi_noirq_set();
+ }
+
+ else if (!memcmp(from, "acpi_sci=edge", 13))
+ acpi_sci_flags.trigger = 1;
+
+ else if (!memcmp(from, "acpi_sci=level", 14))
+ acpi_sci_flags.trigger = 3;
+
+ else if (!memcmp(from, "acpi_sci=high", 13))
+ acpi_sci_flags.polarity = 1;
+
+ else if (!memcmp(from, "acpi_sci=low", 12))
+ acpi_sci_flags.polarity = 3;
+
+#ifdef CONFIG_X86_IO_APIC
+ else if (!memcmp(from, "acpi_skip_timer_override", 24))
+ acpi_skip_timer_override = 1;
+
+ if (!memcmp(from, "disable_timer_pin_1", 19))
+ disable_timer_pin_1 = 1;
+ if (!memcmp(from, "enable_timer_pin_1", 18))
+ disable_timer_pin_1 = -1;
+
+ /* disable IO-APIC */
+ else if (!memcmp(from, "noapic", 6))
+ disable_ioapic_setup();
+#endif /* CONFIG_X86_IO_APIC */
+#endif /* CONFIG_ACPI */
+
+#ifdef CONFIG_X86_LOCAL_APIC
+ /* enable local APIC */
+ else if (!memcmp(from, "lapic", 5))
+ lapic_enable();
+
+ /* disable local APIC */
+ else if (!memcmp(from, "nolapic", 6))
+ lapic_disable();
+#endif /* CONFIG_X86_LOCAL_APIC */
+
+#ifdef CONFIG_KEXEC
+ /* crashkernel=size@addr specifies the location to reserve for
+ * a crash kernel. By reserving this memory we guarantee
+ * that linux never set's it up as a DMA target.
+ * Useful for holding code to do something appropriate
+ * after a kernel panic.
+ */
+ else if (!memcmp(from, "crashkernel=", 12)) {
+#ifndef CONFIG_XEN
+ unsigned long size, base;
+ size = memparse(from+12, &from);
+ if (*from == '@') {
+ base = memparse(from+1, &from);
+ /* FIXME: Do I want a sanity check
+ * to validate the memory range?
+ */
+ crashk_res.start = base;
+ crashk_res.end = base + size - 1;
+ }
+#else
+ printk("Ignoring crashkernel command line, "
+ "parameter will be supplied by xen\n");
+#endif
+ }
+#endif
+#ifdef CONFIG_PROC_VMCORE
+ /* elfcorehdr= specifies the location of elf core header
+ * stored by the crashed kernel.
+ */
+ else if (!memcmp(from, "elfcorehdr=", 11))
+ elfcorehdr_addr = memparse(from+11, &from);
+#endif
+
+ /*
+ * highmem=size forces highmem to be exactly 'size' bytes.
+ * This works even on boxes that have no highmem otherwise.
+ * This also works to reduce highmem size on bigger boxes.
+ */
+ else if (!memcmp(from, "highmem=", 8))
+ highmem_pages = memparse(from+8, &from) >> PAGE_SHIFT;
+
+ /*
+ * vmalloc=size forces the vmalloc area to be exactly 'size'
+ * bytes. This can be used to increase (or decrease) the
+ * vmalloc area - the default is 128m.
+ */
+ else if (!memcmp(from, "vmalloc=", 8))
+ __VMALLOC_RESERVE = memparse(from+8, &from);
+
+ next_char:
+ c = *(from++);
+ if (!c)
+ break;
+ if (COMMAND_LINE_SIZE <= ++len)
+ break;
+ *(to++) = c;
+ }
+ *to = '\0';
+ *cmdline_p = command_line;
+ if (userdef) {
+ printk(KERN_INFO "user-defined physical RAM map:\n");
+ print_memory_map("user");
+ }
+}
+
+/*
+ * Callback for efi_memory_walk.
+ */
+static int __init
+efi_find_max_pfn(unsigned long start, unsigned long end, void *arg)
+{
+ unsigned long *max_pfn = arg, pfn;
+
+ if (start < end) {
+ pfn = PFN_UP(end -1);
+ if (pfn > *max_pfn)
+ *max_pfn = pfn;
+ }
+ return 0;
+}
+
+static int __init
+efi_memory_present_wrapper(unsigned long start, unsigned long end, void *arg)
+{
+ memory_present(0, start, end);
+ return 0;
+}
+
+/*
+ * This function checks if any part of the range <start,end> is mapped
+ * with type.
+ */
+int
+e820_any_mapped(u64 start, u64 end, unsigned type)
+{
+ int i;
+
+#ifndef CONFIG_XEN
+ for (i = 0; i < e820.nr_map; i++) {
+ const struct e820entry *ei = &e820.map[i];
+#else
+ if (!is_initial_xendomain())
+ return 0;
+ for (i = 0; i < machine_e820.nr_map; ++i) {
+ const struct e820entry *ei = &machine_e820.map[i];
+#endif
+
+ if (type && ei->type != type)
+ continue;
+ if (ei->addr >= end || ei->addr + ei->size <= start)
+ continue;
+ return 1;
+ }
+ return 0;
+}
+EXPORT_SYMBOL_GPL(e820_any_mapped);
+
+ /*
+ * This function checks if the entire range <start,end> is mapped with type.
+ *
+ * Note: this function only works correct if the e820 table is sorted and
+ * not-overlapping, which is the case
+ */
+int __init
+e820_all_mapped(unsigned long s, unsigned long e, unsigned type)
+{
+ u64 start = s;
+ u64 end = e;
+ int i;
+
+#ifndef CONFIG_XEN
+ for (i = 0; i < e820.nr_map; i++) {
+ struct e820entry *ei = &e820.map[i];
+#else
+ if (!is_initial_xendomain())
+ return 0;
+ for (i = 0; i < machine_e820.nr_map; ++i) {
+ const struct e820entry *ei = &machine_e820.map[i];
+#endif
+ if (type && ei->type != type)
+ continue;
+ /* is the region (part) in overlap with the current region ?*/
+ if (ei->addr >= end || ei->addr + ei->size <= start)
+ continue;
+ /* if the region is at the beginning of <start,end> we move
+ * start to the end of the region since it's ok until there
+ */
+ if (ei->addr <= start)
+ start = ei->addr + ei->size;
+ /* if start is now at or beyond end, we're done, full
+ * coverage */
+ if (start >= end)
+ return 1; /* we're done */
+ }
+ return 0;
+}
+
+/*
+ * Find the highest page frame number we have available
+ */
+void __init find_max_pfn(void)
+{
+ int i;
+
+ max_pfn = 0;
+ if (efi_enabled) {
+ efi_memmap_walk(efi_find_max_pfn, &max_pfn);
+ efi_memmap_walk(efi_memory_present_wrapper, NULL);
+ return;
+ }
+
+ for (i = 0; i < e820.nr_map; i++) {
+ unsigned long start, end;
+ /* RAM? */
+ if (e820.map[i].type != E820_RAM)
+ continue;
+ start = PFN_UP(e820.map[i].addr);
+ end = PFN_DOWN(e820.map[i].addr + e820.map[i].size);
+ if (start >= end)
+ continue;
+ if (end > max_pfn)
+ max_pfn = end;
+ memory_present(0, start, end);
+ }
+}
+
+/*
+ * Determine low and high memory ranges:
+ */
+unsigned long __init find_max_low_pfn(void)
+{
+ unsigned long max_low_pfn;
+
+ max_low_pfn = max_pfn;
+ if (max_low_pfn > MAXMEM_PFN) {
+ if (highmem_pages == -1)
+ highmem_pages = max_pfn - MAXMEM_PFN;
+ if (highmem_pages + MAXMEM_PFN < max_pfn)
+ max_pfn = MAXMEM_PFN + highmem_pages;
+ if (highmem_pages + MAXMEM_PFN > max_pfn) {
+ printk("only %luMB highmem pages available, ignoring highmem size of %uMB.\n", pages_to_mb(max_pfn - MAXMEM_PFN), pages_to_mb(highmem_pages));
+ highmem_pages = 0;
+ }
+ max_low_pfn = MAXMEM_PFN;
+#ifndef CONFIG_HIGHMEM
+ /* Maximum memory usable is what is directly addressable */
+ printk(KERN_WARNING "Warning only %ldMB will be used.\n",
+ MAXMEM>>20);
+ if (max_pfn > MAX_NONPAE_PFN)
+ printk(KERN_WARNING "Use a PAE enabled kernel.\n");
+ else
+ printk(KERN_WARNING "Use a HIGHMEM enabled kernel.\n");
+ max_pfn = MAXMEM_PFN;
+#else /* !CONFIG_HIGHMEM */
+#ifndef CONFIG_X86_PAE
+ if (max_pfn > MAX_NONPAE_PFN) {
+ max_pfn = MAX_NONPAE_PFN;
+ printk(KERN_WARNING "Warning only 4GB will be used.\n");
+ printk(KERN_WARNING "Use a PAE enabled kernel.\n");
+ }
+#endif /* !CONFIG_X86_PAE */
+#endif /* !CONFIG_HIGHMEM */
+ } else {
+ if (highmem_pages == -1)
+ highmem_pages = 0;
+#ifdef CONFIG_HIGHMEM
+ if (highmem_pages >= max_pfn) {
+ printk(KERN_ERR "highmem size specified (%uMB) is bigger than pages available (%luMB)!.\n", pages_to_mb(highmem_pages), pages_to_mb(max_pfn));
+ highmem_pages = 0;
+ }
+ if (highmem_pages) {
+ if (max_low_pfn-highmem_pages < 64*1024*1024/PAGE_SIZE){
+ printk(KERN_ERR "highmem size %uMB results in smaller than 64MB lowmem, ignoring it.\n", pages_to_mb(highmem_pages));
+ highmem_pages = 0;
+ }
+ max_low_pfn -= highmem_pages;
+ }
+#else
+ if (highmem_pages)
+ printk(KERN_ERR "ignoring highmem size on non-highmem kernel!\n");
+#endif
+ }
+ return max_low_pfn;
+}
+
+/*
+ * Free all available memory for boot time allocation. Used
+ * as a callback function by efi_memory_walk()
+ */
+
+static int __init
+free_available_memory(unsigned long start, unsigned long end, void *arg)
+{
+ /* check max_low_pfn */
+ if (start >= (max_low_pfn << PAGE_SHIFT))
+ return 0;
+ if (end >= (max_low_pfn << PAGE_SHIFT))
+ end = max_low_pfn << PAGE_SHIFT;
+ if (start < end)
+ free_bootmem(start, end - start);
+
+ return 0;
+}
+/*
+ * Register fully available low RAM pages with the bootmem allocator.
+ */
+static void __init register_bootmem_low_pages(unsigned long max_low_pfn)
+{
+ int i;
+
+ if (efi_enabled) {
+ efi_memmap_walk(free_available_memory, NULL);
+ return;
+ }
+ for (i = 0; i < e820.nr_map; i++) {
+ unsigned long curr_pfn, last_pfn, size;
+ /*
+ * Reserve usable low memory
+ */
+ if (e820.map[i].type != E820_RAM)
+ continue;
+ /*
+ * We are rounding up the start address of usable memory:
+ */
+ curr_pfn = PFN_UP(e820.map[i].addr);
+ if (curr_pfn >= max_low_pfn)
+ continue;
+ /*
+ * ... and at the end of the usable range downwards:
+ */
+ last_pfn = PFN_DOWN(e820.map[i].addr + e820.map[i].size);
+
+#ifdef CONFIG_XEN
+ /*
+ * Truncate to the number of actual pages currently
+ * present.
+ */
+ if (last_pfn > xen_start_info->nr_pages)
+ last_pfn = xen_start_info->nr_pages;
+#endif
+
+ if (last_pfn > max_low_pfn)
+ last_pfn = max_low_pfn;
+
+ /*
+ * .. finally, did all the rounding and playing
+ * around just make the area go away?
+ */
+ if (last_pfn <= curr_pfn)
+ continue;
+
+ size = last_pfn - curr_pfn;
+ free_bootmem(PFN_PHYS(curr_pfn), PFN_PHYS(size));
+ }
+}
+
+#ifndef CONFIG_XEN
+/*
+ * workaround for Dell systems that neglect to reserve EBDA
+ */
+static void __init reserve_ebda_region(void)
+{
+ unsigned int addr;
+ addr = get_bios_ebda();
+ if (addr)
+ reserve_bootmem(addr, PAGE_SIZE);
+}
+#endif
+
+#ifndef CONFIG_NEED_MULTIPLE_NODES
+void __init setup_bootmem_allocator(void);
+static unsigned long __init setup_memory(void)
+{
+ /*
+ * partially used pages are not usable - thus
+ * we are rounding upwards:
+ */
+ min_low_pfn = PFN_UP(__pa(xen_start_info->pt_base)) +
+ xen_start_info->nr_pt_frames;
+
+ find_max_pfn();
+
+ max_low_pfn = find_max_low_pfn();
+
+#ifdef CONFIG_HIGHMEM
+ highstart_pfn = highend_pfn = max_pfn;
+ if (max_pfn > max_low_pfn) {
+ highstart_pfn = max_low_pfn;
+ }
+ printk(KERN_NOTICE "%ldMB HIGHMEM available.\n",
+ pages_to_mb(highend_pfn - highstart_pfn));
+#endif
+ printk(KERN_NOTICE "%ldMB LOWMEM available.\n",
+ pages_to_mb(max_low_pfn));
+
+ setup_bootmem_allocator();
+
+ return max_low_pfn;
+}
+
+void __init zone_sizes_init(void)
+{
+ unsigned long zones_size[MAX_NR_ZONES] = {0, 0, 0};
+ unsigned int max_dma, low;
+
+ max_dma = virt_to_phys((char *)MAX_DMA_ADDRESS) >> PAGE_SHIFT;
+ low = max_low_pfn;
+
+ if (low < max_dma)
+ zones_size[ZONE_DMA] = low;
+ else {
+ zones_size[ZONE_DMA] = max_dma;
+ zones_size[ZONE_NORMAL] = low - max_dma;
+#ifdef CONFIG_HIGHMEM
+ zones_size[ZONE_HIGHMEM] = highend_pfn - low;
+#endif
+ }
+ free_area_init(zones_size);
+}
+#else
+extern unsigned long __init setup_memory(void);
+extern void zone_sizes_init(void);
+#endif /* !CONFIG_NEED_MULTIPLE_NODES */
+
+void __init setup_bootmem_allocator(void)
+{
+ unsigned long bootmap_size;
+ /*
+ * Initialize the boot-time allocator (with low memory only):
+ */
+ bootmap_size = init_bootmem(min_low_pfn, max_low_pfn);
+
+ register_bootmem_low_pages(max_low_pfn);
+
+ /*
+ * Reserve the bootmem bitmap itself as well. We do this in two
+ * steps (first step was init_bootmem()) because this catches
+ * the (very unlikely) case of us accidentally initializing the
+ * bootmem allocator with an invalid RAM area.
+ */
+ reserve_bootmem(__PHYSICAL_START, (PFN_PHYS(min_low_pfn) +
+ bootmap_size + PAGE_SIZE-1) - (__PHYSICAL_START));
+
+#ifndef CONFIG_XEN
+ /*
+ * reserve physical page 0 - it's a special BIOS page on many boxes,
+ * enabling clean reboots, SMP operation, laptop functions.
+ */
+ reserve_bootmem(0, PAGE_SIZE);
+
+ /* reserve EBDA region, it's a 4K region */
+ reserve_ebda_region();
+
+ /* could be an AMD 768MPX chipset. Reserve a page before VGA to prevent
+ PCI prefetch into it (errata #56). Usually the page is reserved anyways,
+ unless you have no PS/2 mouse plugged in. */
+ if (boot_cpu_data.x86_vendor == X86_VENDOR_AMD &&
+ boot_cpu_data.x86 == 6)
+ reserve_bootmem(0xa0000 - 4096, 4096);
+
+#ifdef CONFIG_SMP
+ /*
+ * But first pinch a few for the stack/trampoline stuff
+ * FIXME: Don't need the extra page at 4K, but need to fix
+ * trampoline before removing it. (see the GDT stuff)
+ */
+ reserve_bootmem(PAGE_SIZE, PAGE_SIZE);
+#endif
+#ifdef CONFIG_ACPI_SLEEP
+ /*
+ * Reserve low memory region for sleep support.
+ */
+ acpi_reserve_bootmem();
+#endif
+#endif /* !CONFIG_XEN */
+
+#ifdef CONFIG_BLK_DEV_INITRD
+ if (xen_start_info->mod_start) {
+ if (INITRD_START + INITRD_SIZE <= (max_low_pfn << PAGE_SHIFT)) {
+ /*reserve_bootmem(INITRD_START, INITRD_SIZE);*/
+ initrd_start = INITRD_START + PAGE_OFFSET;
+ initrd_end = initrd_start+INITRD_SIZE;
+ initrd_below_start_ok = 1;
+ }
+ else {
+ printk(KERN_ERR "initrd extends beyond end of memory "
+ "(0x%08lx > 0x%08lx)\ndisabling initrd\n",
+ INITRD_START + INITRD_SIZE,
+ max_low_pfn << PAGE_SHIFT);
+ initrd_start = 0;
+ }
+ }
+#endif
+#ifdef CONFIG_KEXEC
+#ifdef CONFIG_XEN
+ xen_machine_kexec_setup_resources();
+#else
+ if (crashk_res.start != crashk_res.end)
+ reserve_bootmem(crashk_res.start,
+ crashk_res.end - crashk_res.start + 1);
+#endif
+#endif
+}
+
+/*
+ * The node 0 pgdat is initialized before all of these because
+ * it's needed for bootmem. node>0 pgdats have their virtual
+ * space allocated before the pagetables are in place to access
+ * them, so they can't be cleared then.
+ *
+ * This should all compile down to nothing when NUMA is off.
+ */
+void __init remapped_pgdat_init(void)
+{
+ int nid;
+
+ for_each_online_node(nid) {
+ if (nid != 0)
+ memset(NODE_DATA(nid), 0, sizeof(struct pglist_data));
+ }
+}
+
+/*
+ * Request address space for all standard RAM and ROM resources
+ * and also for regions reported as reserved by the e820.
+ */
+static void __init
+legacy_init_iomem_resources(struct e820entry *e820, int nr_map,
+ struct resource *code_resource,
+ struct resource *data_resource)
+{
+ int i;
+
+ probe_roms();
+
+ for (i = 0; i < nr_map; i++) {
+ struct resource *res;
+#ifndef CONFIG_RESOURCES_64BIT
+ if (e820[i].addr + e820[i].size > 0x100000000ULL)
+ continue;
+#endif
+ res = kzalloc(sizeof(struct resource), GFP_ATOMIC);
+ switch (e820[i].type) {
+ case E820_RAM: res->name = "System RAM"; break;
+ case E820_ACPI: res->name = "ACPI Tables"; break;
+ case E820_NVS: res->name = "ACPI Non-volatile Storage"; break;
+ default: res->name = "reserved";
+ }
+ res->start = e820[i].addr;
+ res->end = res->start + e820[i].size - 1;
+ res->flags = IORESOURCE_MEM | IORESOURCE_BUSY;
+ if (request_resource(&iomem_resource, res)) {
+ kfree(res);
+ continue;
+ }
+ if (e820[i].type == E820_RAM) {
+ /*
+ * We don't know which RAM region contains kernel data,
+ * so we try it repeatedly and let the resource manager
+ * test it.
+ */
+#ifndef CONFIG_XEN
+ request_resource(res, code_resource);
+ request_resource(res, data_resource);
+#endif
+#ifdef CONFIG_KEXEC
+ if (crashk_res.start != crashk_res.end)
+ request_resource(res, &crashk_res);
+#ifdef CONFIG_XEN
+ xen_machine_kexec_register_resources(res);
+#endif
+#endif
+ }
+ }
+}
+
+/*
+ * Locate a unused range of the physical address space below 4G which
+ * can be used for PCI mappings.
+ */
+static void __init
+e820_setup_gap(struct e820entry *e820, int nr_map)
+{
+ unsigned long gapstart, gapsize, round;
+ unsigned long long last;
+ int i;
+
+ /*
+ * Search for the bigest gap in the low 32 bits of the e820
+ * memory space.
+ */
+ last = 0x100000000ull;
+ gapstart = 0x10000000;
+ gapsize = 0x400000;
+ i = nr_map;
+ while (--i >= 0) {
+ unsigned long long start = e820[i].addr;
+ unsigned long long end = start + e820[i].size;
+
+ /*
+ * Since "last" is at most 4GB, we know we'll
+ * fit in 32 bits if this condition is true
+ */
+ if (last > end) {
+ unsigned long gap = last - end;
+
+ if (gap > gapsize) {
+ gapsize = gap;
+ gapstart = end;
+ }
+ }
+ if (start < last)
+ last = start;
+ }
+
+ /*
+ * See how much we want to round up: start off with
+ * rounding to the next 1MB area.
+ */
+ round = 0x100000;
+ while ((gapsize >> 4) > round)
+ round += round;
+ /* Fun with two's complement */
+ pci_mem_start = (gapstart + round) & -round;
+
+ printk("Allocating PCI resources starting at %08lx (gap: %08lx:%08lx)\n",
+ pci_mem_start, gapstart, gapsize);
+}
+
+/*
+ * Request address space for all standard resources
+ *
+ * This is called just before pcibios_init(), which is also a
+ * subsys_initcall, but is linked in later (in arch/i386/pci/common.c).
+ */
+static int __init request_standard_resources(void)
+{
+ int i;
+
+ /* Nothing to do if not running in dom0. */
+ if (!is_initial_xendomain())
+ return 0;
+
+ printk("Setting up standard PCI resources\n");
+#ifdef CONFIG_XEN
+ legacy_init_iomem_resources(machine_e820.map, machine_e820.nr_map,
+ &code_resource, &data_resource);
+#else
+ if (efi_enabled)
+ efi_initialize_iomem_resources(&code_resource, &data_resource);
+ else
+ legacy_init_iomem_resources(e820.map, e820.nr_map,
+ &code_resource, &data_resource);
+#endif
+
+ /* EFI systems may still have VGA */
+ request_resource(&iomem_resource, &video_ram_resource);
+
+ /* request I/O space for devices used on all i[345]86 PCs */
+ for (i = 0; i < STANDARD_IO_RESOURCES; i++)
+ request_resource(&ioport_resource, &standard_io_resources[i]);
+ return 0;
+}
+
+subsys_initcall(request_standard_resources);
+
+static void __init register_memory(void)
+{
+#ifdef CONFIG_XEN
+ if (is_initial_xendomain())
+ e820_setup_gap(machine_e820.map, machine_e820.nr_map);
+ else
+#endif
+ e820_setup_gap(e820.map, e820.nr_map);
+}
+
+#ifdef CONFIG_MCA
+static void set_mca_bus(int x)
+{
+ MCA_bus = x;
+}
+#else
+static void set_mca_bus(int x) { }
+#endif
+
+/*
+ * Determine if we were loaded by an EFI loader. If so, then we have also been
+ * passed the efi memmap, systab, etc., so we should use these data structures
+ * for initialization. Note, the efi init code path is determined by the
+ * global efi_enabled. This allows the same kernel image to be used on existing
+ * systems (with a traditional BIOS) as well as on EFI systems.
+ */
+void __init setup_arch(char **cmdline_p)
+{
+ int i, j, k, fpp;
+ struct physdev_set_iopl set_iopl;
+ unsigned long max_low_pfn;
+ unsigned long p2m_pages;
+
+ /* Force a quick death if the kernel panics (not domain 0). */
+ extern int panic_timeout;
+ if (!panic_timeout && !is_initial_xendomain())
+ panic_timeout = 1;
+
+ /* Register a call for panic conditions. */
+ atomic_notifier_chain_register(&panic_notifier_list, &xen_panic_block);
+
+ WARN_ON(HYPERVISOR_vm_assist(VMASST_CMD_enable,
+ VMASST_TYPE_4gb_segments));
+ WARN_ON(HYPERVISOR_vm_assist(VMASST_CMD_enable,
+ VMASST_TYPE_writable_pagetables));
+
+ memcpy(&boot_cpu_data, &new_cpu_data, sizeof(new_cpu_data));
+ pre_setup_arch_hook();
+ early_cpu_init();
+#ifdef CONFIG_SMP
+ prefill_possible_map();
+#endif
+
+ /*
+ * FIXME: This isn't an official loader_type right
+ * now but does currently work with elilo.
+ * If we were configured as an EFI kernel, check to make
+ * sure that we were loaded correctly from elilo and that
+ * the system table is valid. If not, then initialize normally.
+ */
+#ifdef CONFIG_EFI
+ if ((LOADER_TYPE == 0x50) && EFI_SYSTAB)
+ efi_enabled = 1;
+#endif
+
+ /* This must be initialized to UNNAMED_MAJOR for ipconfig to work
+ properly. Setting ROOT_DEV to default to /dev/ram0 breaks initrd.
+ */
+ ROOT_DEV = MKDEV(UNNAMED_MAJOR,0);
+ drive_info = DRIVE_INFO;
+ screen_info = SCREEN_INFO;
+ copy_edid();
+ apm_info.bios = APM_BIOS_INFO;
+ ist_info = IST_INFO;
+ saved_videomode = VIDEO_MODE;
+ if( SYS_DESC_TABLE.length != 0 ) {
+ set_mca_bus(SYS_DESC_TABLE.table[3] & 0x2);
+ machine_id = SYS_DESC_TABLE.table[0];
+ machine_submodel_id = SYS_DESC_TABLE.table[1];
+ BIOS_revision = SYS_DESC_TABLE.table[2];
+ }
+ bootloader_type = LOADER_TYPE;
+
+ if (is_initial_xendomain()) {
+ const struct dom0_vga_console_info *info =
+ (void *)((char *)xen_start_info +
+ xen_start_info->console.dom0.info_off);
+
+ dom0_init_screen_info(info,
+ xen_start_info->console.dom0.info_size);
+ xen_start_info->console.domU.mfn = 0;
+ xen_start_info->console.domU.evtchn = 0;
+ } else
+ screen_info.orig_video_isVGA = 0;
+
+#ifdef CONFIG_BLK_DEV_RAM
+ rd_image_start = RAMDISK_FLAGS & RAMDISK_IMAGE_START_MASK;
+ rd_prompt = ((RAMDISK_FLAGS & RAMDISK_PROMPT_FLAG) != 0);
+ rd_doload = ((RAMDISK_FLAGS & RAMDISK_LOAD_FLAG) != 0);
+#endif
+
+ ARCH_SETUP
+ if (efi_enabled)
+ efi_init();
+ else {
+ printk(KERN_INFO "BIOS-provided physical RAM map:\n");
+ print_memory_map(machine_specific_memory_setup());
+ }
+
+ copy_edd();
+
+ if (!MOUNT_ROOT_RDONLY)
+ root_mountflags &= ~MS_RDONLY;
+ init_mm.start_code = (unsigned long) _text;
+ init_mm.end_code = (unsigned long) _etext;
+ init_mm.end_data = (unsigned long) _edata;
+ init_mm.brk = (PFN_UP(__pa(xen_start_info->pt_base)) +
+ xen_start_info->nr_pt_frames) << PAGE_SHIFT;
+
+ code_resource.start = virt_to_phys(_text);
+ code_resource.end = virt_to_phys(_etext)-1;
+ data_resource.start = virt_to_phys(_etext);
+ data_resource.end = virt_to_phys(_edata)-1;
+
+ parse_cmdline_early(cmdline_p);
+
+#ifdef CONFIG_EARLY_PRINTK
+ {
+ char *s = strstr(*cmdline_p, "earlyprintk=");
+ if (s) {
+ setup_early_printk(strchr(s, '=') + 1);
+ printk("early console enabled\n");
+ }
+ }
+#endif
+
+ max_low_pfn = setup_memory();
+
+ /*
+ * NOTE: before this point _nobody_ is allowed to allocate
+ * any memory using the bootmem allocator. Although the
+ * alloctor is now initialised only the first 8Mb of the kernel
+ * virtual address space has been mapped. All allocations before
+ * paging_init() has completed must use the alloc_bootmem_low_pages()
+ * variant (which allocates DMA'able memory) and care must be taken
+ * not to exceed the 8Mb limit.
+ */
+
+#ifdef CONFIG_SMP
+ smp_alloc_memory(); /* AP processor realmode stacks in low memory*/
+#endif
+ paging_init();
+ remapped_pgdat_init();
+ sparse_init();
+ zone_sizes_init();
+
+#ifdef CONFIG_X86_FIND_SMP_CONFIG
+ /*
+ * Find and reserve possible boot-time SMP configuration:
+ */
+ find_smp_config();
+#endif
+
+ p2m_pages = max_pfn;
+ if (xen_start_info->nr_pages > max_pfn) {
+ /*
+ * the max_pfn was shrunk (probably by mem= or highmem=
+ * kernel parameter); shrink reservation with the HV
+ */
+ struct xen_memory_reservation reservation = {
+ .address_bits = 0,
+ .extent_order = 0,
+ .domid = DOMID_SELF
+ };
+ unsigned int difference;
+ int ret;
+
+ difference = xen_start_info->nr_pages - max_pfn;
+
+ set_xen_guest_handle(reservation.extent_start,
+ ((unsigned long *)xen_start_info->mfn_list) + max_pfn);
+ reservation.nr_extents = difference;
+ ret = HYPERVISOR_memory_op(XENMEM_decrease_reservation,
+ &reservation);
+ BUG_ON (ret != difference);
+ }
+ else if (max_pfn > xen_start_info->nr_pages)
+ p2m_pages = xen_start_info->nr_pages;
+
+ /* Make sure we have a correctly sized P->M table. */
+ if (!xen_feature(XENFEAT_auto_translated_physmap)) {
+ phys_to_machine_mapping = alloc_bootmem_low_pages(
+ max_pfn * sizeof(unsigned long));
+ memset(phys_to_machine_mapping, ~0,
+ max_pfn * sizeof(unsigned long));
+ memcpy(phys_to_machine_mapping,
+ (unsigned long *)xen_start_info->mfn_list,
+ p2m_pages * sizeof(unsigned long));
+ free_bootmem(
+ __pa(xen_start_info->mfn_list),
+ PFN_PHYS(PFN_UP(xen_start_info->nr_pages *
+ sizeof(unsigned long))));
+
+ /*
+ * Initialise the list of the frames that specify the list of
+ * frames that make up the p2m table. Used by save/restore
+ */
+ pfn_to_mfn_frame_list_list = alloc_bootmem_low_pages(PAGE_SIZE);
+
+ fpp = PAGE_SIZE/sizeof(unsigned long);
+ for (i=0, j=0, k=-1; i< max_pfn; i+=fpp, j++) {
+ if ((j % fpp) == 0) {
+ k++;
+ BUG_ON(k>=16);
+ pfn_to_mfn_frame_list[k] =
+ alloc_bootmem_low_pages(PAGE_SIZE);
+ pfn_to_mfn_frame_list_list[k] =
+ virt_to_mfn(pfn_to_mfn_frame_list[k]);
+ j=0;
+ }
+ pfn_to_mfn_frame_list[k][j] =
+ virt_to_mfn(&phys_to_machine_mapping[i]);
+ }
+ HYPERVISOR_shared_info->arch.max_pfn = max_pfn;
+ HYPERVISOR_shared_info->arch.pfn_to_mfn_frame_list_list =
+ virt_to_mfn(pfn_to_mfn_frame_list_list);
+ }
+
+ /* Mark all ISA DMA channels in-use - using them wouldn't work. */
+ for (i = 0; i < MAX_DMA_CHANNELS; ++i)
+ if (i != 4 && request_dma(i, "xen") != 0)
+ BUG();
+
+ /*
+ * NOTE: at this point the bootmem allocator is fully available.
+ */
+
+ if (is_initial_xendomain())
+ dmi_scan_machine();
+
+#ifdef CONFIG_X86_GENERICARCH
+ generic_apic_probe(*cmdline_p);
+#endif
+ if (efi_enabled)
+ efi_map_memmap();
+
+ set_iopl.iopl = 1;
+ WARN_ON(HYPERVISOR_physdev_op(PHYSDEVOP_set_iopl, &set_iopl));
+
+#ifdef CONFIG_ACPI
+ if (!is_initial_xendomain()) {
+ printk(KERN_INFO "ACPI in unprivileged domain disabled\n");
+ acpi_disabled = 1;
+ acpi_ht = 0;
+ }
+
+ /*
+ * Parse the ACPI tables for possible boot-time SMP configuration.
+ */
+ acpi_boot_table_init();
+#endif
+
+#ifdef CONFIG_X86_IO_APIC
+ check_acpi_pci(); /* Checks more than just ACPI actually */
+#endif
+
+#ifdef CONFIG_ACPI
+ acpi_boot_init();
+
+#if defined(CONFIG_SMP) && defined(CONFIG_X86_PC)
+ if (def_to_bigsmp)
+ printk(KERN_WARNING "More than 8 CPUs detected and "
+ "CONFIG_X86_PC cannot handle it.\nUse "
+ "CONFIG_X86_GENERICARCH or CONFIG_X86_BIGSMP.\n");
+#endif
+#endif
+#ifdef CONFIG_X86_LOCAL_APIC
+ if (smp_found_config)
+ get_smp_config();
+#endif
+
+ register_memory();
+
+ if (is_initial_xendomain()) {
+#ifdef CONFIG_VT
+#if defined(CONFIG_VGA_CONSOLE)
+ if (!efi_enabled ||
+ (efi_mem_type(0xa0000) != EFI_CONVENTIONAL_MEMORY))
+ conswitchp = &vga_con;
+#elif defined(CONFIG_DUMMY_CONSOLE)
+ conswitchp = &dummy_con;
+#endif
+#endif
+ } else {
+#if defined(CONFIG_VT) && defined(CONFIG_DUMMY_CONSOLE)
+ conswitchp = &dummy_con;
+#endif
+ }
+ tsc_init();
+}
+
+static int
+xen_panic_event(struct notifier_block *this, unsigned long event, void *ptr)
+{
+ HYPERVISOR_shutdown(SHUTDOWN_crash);
+ /* we're never actually going to get here... */
+ return NOTIFY_DONE;
+}
+
+static __init int add_pcspkr(void)
+{
+ struct platform_device *pd;
+ int ret;
+
+ if (!is_initial_xendomain())
+ return 0;
+
+ pd = platform_device_alloc("pcspkr", -1);
+ if (!pd)
+ return -ENOMEM;
+
+ ret = platform_device_add(pd);
+ if (ret)
+ platform_device_put(pd);
+
+ return ret;
+}
+device_initcall(add_pcspkr);
+
+/*
+ * Local Variables:
+ * mode:c
+ * c-file-style:"k&r"
+ * c-basic-offset:8
+ * End:
+ */
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/kernel/smp_32-xen.c 2007-12-10 08:47:31.000000000 +0100
@@ -0,0 +1,605 @@
+/*
+ * Intel SMP support routines.
+ *
+ * (c) 1995 Alan Cox, Building #3 <alan@redhat.com>
+ * (c) 1998-99, 2000 Ingo Molnar <mingo@redhat.com>
+ *
+ * This code is released under the GNU General Public License version 2 or
+ * later.
+ */
+
+#include <linux/init.h>
+
+#include <linux/mm.h>
+#include <linux/delay.h>
+#include <linux/spinlock.h>
+#include <linux/smp_lock.h>
+#include <linux/kernel_stat.h>
+#include <linux/mc146818rtc.h>
+#include <linux/cache.h>
+#include <linux/interrupt.h>
+#include <linux/cpu.h>
+#include <linux/module.h>
+
+#include <asm/mtrr.h>
+#include <asm/tlbflush.h>
+#if 0
+#include <mach_apic.h>
+#endif
+#include <xen/evtchn.h>
+
+/*
+ * Some notes on x86 processor bugs affecting SMP operation:
+ *
+ * Pentium, Pentium Pro, II, III (and all CPUs) have bugs.
+ * The Linux implications for SMP are handled as follows:
+ *
+ * Pentium III / [Xeon]
+ * None of the E1AP-E3AP errata are visible to the user.
+ *
+ * E1AP. see PII A1AP
+ * E2AP. see PII A2AP
+ * E3AP. see PII A3AP
+ *
+ * Pentium II / [Xeon]
+ * None of the A1AP-A3AP errata are visible to the user.
+ *
+ * A1AP. see PPro 1AP
+ * A2AP. see PPro 2AP
+ * A3AP. see PPro 7AP
+ *
+ * Pentium Pro
+ * None of 1AP-9AP errata are visible to the normal user,
+ * except occasional delivery of 'spurious interrupt' as trap #15.
+ * This is very rare and a non-problem.
+ *
+ * 1AP. Linux maps APIC as non-cacheable
+ * 2AP. worked around in hardware
+ * 3AP. fixed in C0 and above steppings microcode update.
+ * Linux does not use excessive STARTUP_IPIs.
+ * 4AP. worked around in hardware
+ * 5AP. symmetric IO mode (normal Linux operation) not affected.
+ * 'noapic' mode has vector 0xf filled out properly.
+ * 6AP. 'noapic' mode might be affected - fixed in later steppings
+ * 7AP. We do not assume writes to the LVT deassering IRQs
+ * 8AP. We do not enable low power mode (deep sleep) during MP bootup
+ * 9AP. We do not use mixed mode
+ *
+ * Pentium
+ * There is a marginal case where REP MOVS on 100MHz SMP
+ * machines with B stepping processors can fail. XXX should provide
+ * an L1cache=Writethrough or L1cache=off option.
+ *
+ * B stepping CPUs may hang. There are hardware work arounds
+ * for this. We warn about it in case your board doesn't have the work
+ * arounds. Basically thats so I can tell anyone with a B stepping
+ * CPU and SMP problems "tough".
+ *
+ * Specific items [From Pentium Processor Specification Update]
+ *
+ * 1AP. Linux doesn't use remote read
+ * 2AP. Linux doesn't trust APIC errors
+ * 3AP. We work around this
+ * 4AP. Linux never generated 3 interrupts of the same priority
+ * to cause a lost local interrupt.
+ * 5AP. Remote read is never used
+ * 6AP. not affected - worked around in hardware
+ * 7AP. not affected - worked around in hardware
+ * 8AP. worked around in hardware - we get explicit CS errors if not
+ * 9AP. only 'noapic' mode affected. Might generate spurious
+ * interrupts, we log only the first one and count the
+ * rest silently.
+ * 10AP. not affected - worked around in hardware
+ * 11AP. Linux reads the APIC between writes to avoid this, as per
+ * the documentation. Make sure you preserve this as it affects
+ * the C stepping chips too.
+ * 12AP. not affected - worked around in hardware
+ * 13AP. not affected - worked around in hardware
+ * 14AP. we always deassert INIT during bootup
+ * 15AP. not affected - worked around in hardware
+ * 16AP. not affected - worked around in hardware
+ * 17AP. not affected - worked around in hardware
+ * 18AP. not affected - worked around in hardware
+ * 19AP. not affected - worked around in BIOS
+ *
+ * If this sounds worrying believe me these bugs are either ___RARE___,
+ * or are signal timing bugs worked around in hardware and there's
+ * about nothing of note with C stepping upwards.
+ */
+
+DEFINE_PER_CPU(struct tlb_state, cpu_tlbstate) ____cacheline_aligned = { &init_mm, 0, };
+
+/*
+ * the following functions deal with sending IPIs between CPUs.
+ *
+ * We use 'broadcast', CPU->CPU IPIs and self-IPIs too.
+ */
+
+static inline int __prepare_ICR (unsigned int shortcut, int vector)
+{
+ unsigned int icr = shortcut | APIC_DEST_LOGICAL;
+
+ switch (vector) {
+ default:
+ icr |= APIC_DM_FIXED | vector;
+ break;
+ case NMI_VECTOR:
+ icr |= APIC_DM_NMI;
+ break;
+ }
+ return icr;
+}
+
+static inline int __prepare_ICR2 (unsigned int mask)
+{
+ return SET_APIC_DEST_FIELD(mask);
+}
+
+DECLARE_PER_CPU(int, ipi_to_irq[NR_IPIS]);
+
+static inline void __send_IPI_one(unsigned int cpu, int vector)
+{
+ int irq = per_cpu(ipi_to_irq, cpu)[vector];
+ BUG_ON(irq < 0);
+ notify_remote_via_irq(irq);
+}
+
+void __send_IPI_shortcut(unsigned int shortcut, int vector)
+{
+ int cpu;
+
+ switch (shortcut) {
+ case APIC_DEST_SELF:
+ __send_IPI_one(smp_processor_id(), vector);
+ break;
+ case APIC_DEST_ALLBUT:
+ for (cpu = 0; cpu < NR_CPUS; ++cpu) {
+ if (cpu == smp_processor_id())
+ continue;
+ if (cpu_isset(cpu, cpu_online_map)) {
+ __send_IPI_one(cpu, vector);
+ }
+ }
+ break;
+ default:
+ printk("XXXXXX __send_IPI_shortcut %08x vector %d\n", shortcut,
+ vector);
+ break;
+ }
+}
+
+void fastcall send_IPI_self(int vector)
+{
+ __send_IPI_shortcut(APIC_DEST_SELF, vector);
+}
+
+/*
+ * This is only used on smaller machines.
+ */
+void send_IPI_mask_bitmask(cpumask_t mask, int vector)
+{
+ unsigned long flags;
+ unsigned int cpu;
+
+ local_irq_save(flags);
+ WARN_ON(cpus_addr(mask)[0] & ~cpus_addr(cpu_online_map)[0]);
+
+ for (cpu = 0; cpu < NR_CPUS; ++cpu) {
+ if (cpu_isset(cpu, mask)) {
+ __send_IPI_one(cpu, vector);
+ }
+ }
+
+ local_irq_restore(flags);
+}
+
+void send_IPI_mask_sequence(cpumask_t mask, int vector)
+{
+
+ send_IPI_mask_bitmask(mask, vector);
+}
+
+#include <mach_ipi.h> /* must come after the send_IPI functions above for inlining */
+
+#if 0 /* XEN */
+/*
+ * Smarter SMP flushing macros.
+ * c/o Linus Torvalds.
+ *
+ * These mean you can really definitely utterly forget about
+ * writing to user space from interrupts. (Its not allowed anyway).
+ *
+ * Optimizations Manfred Spraul <manfred@colorfullife.com>
+ */
+
+static cpumask_t flush_cpumask;
+static struct mm_struct * flush_mm;
+static unsigned long flush_va;
+static DEFINE_SPINLOCK(tlbstate_lock);
+#define FLUSH_ALL 0xffffffff
+
+/*
+ * We cannot call mmdrop() because we are in interrupt context,
+ * instead update mm->cpu_vm_mask.
+ *
+ * We need to reload %cr3 since the page tables may be going
+ * away from under us..
+ */
+static inline void leave_mm (unsigned long cpu)
+{
+ if (per_cpu(cpu_tlbstate, cpu).state == TLBSTATE_OK)
+ BUG();
+ cpu_clear(cpu, per_cpu(cpu_tlbstate, cpu).active_mm->cpu_vm_mask);
+ load_cr3(swapper_pg_dir);
+}
+
+/*
+ *
+ * The flush IPI assumes that a thread switch happens in this order:
+ * [cpu0: the cpu that switches]
+ * 1) switch_mm() either 1a) or 1b)
+ * 1a) thread switch to a different mm
+ * 1a1) cpu_clear(cpu, old_mm->cpu_vm_mask);
+ * Stop ipi delivery for the old mm. This is not synchronized with
+ * the other cpus, but smp_invalidate_interrupt ignore flush ipis
+ * for the wrong mm, and in the worst case we perform a superflous
+ * tlb flush.
+ * 1a2) set cpu_tlbstate to TLBSTATE_OK
+ * Now the smp_invalidate_interrupt won't call leave_mm if cpu0
+ * was in lazy tlb mode.
+ * 1a3) update cpu_tlbstate[].active_mm
+ * Now cpu0 accepts tlb flushes for the new mm.
+ * 1a4) cpu_set(cpu, new_mm->cpu_vm_mask);
+ * Now the other cpus will send tlb flush ipis.
+ * 1a4) change cr3.
+ * 1b) thread switch without mm change
+ * cpu_tlbstate[].active_mm is correct, cpu0 already handles
+ * flush ipis.
+ * 1b1) set cpu_tlbstate to TLBSTATE_OK
+ * 1b2) test_and_set the cpu bit in cpu_vm_mask.
+ * Atomically set the bit [other cpus will start sending flush ipis],
+ * and test the bit.
+ * 1b3) if the bit was 0: leave_mm was called, flush the tlb.
+ * 2) switch %%esp, ie current
+ *
+ * The interrupt must handle 2 special cases:
+ * - cr3 is changed before %%esp, ie. it cannot use current->{active_,}mm.
+ * - the cpu performs speculative tlb reads, i.e. even if the cpu only
+ * runs in kernel space, the cpu could load tlb entries for user space
+ * pages.
+ *
+ * The good news is that cpu_tlbstate is local to each cpu, no
+ * write/read ordering problems.
+ */
+
+/*
+ * TLB flush IPI:
+ *
+ * 1) Flush the tlb entries if the cpu uses the mm that's being flushed.
+ * 2) Leave the mm if we are in the lazy tlb mode.
+ */
+
+irqreturn_t smp_invalidate_interrupt(int irq, void *dev_id,
+ struct pt_regs *regs)
+{
+ unsigned long cpu;
+
+ cpu = get_cpu();
+
+ if (!cpu_isset(cpu, flush_cpumask))
+ goto out;
+ /*
+ * This was a BUG() but until someone can quote me the
+ * line from the intel manual that guarantees an IPI to
+ * multiple CPUs is retried _only_ on the erroring CPUs
+ * its staying as a return
+ *
+ * BUG();
+ */
+
+ if (flush_mm == per_cpu(cpu_tlbstate, cpu).active_mm) {
+ if (per_cpu(cpu_tlbstate, cpu).state == TLBSTATE_OK) {
+ if (flush_va == FLUSH_ALL)
+ local_flush_tlb();
+ else
+ __flush_tlb_one(flush_va);
+ } else
+ leave_mm(cpu);
+ }
+ smp_mb__before_clear_bit();
+ cpu_clear(cpu, flush_cpumask);
+ smp_mb__after_clear_bit();
+out:
+ put_cpu_no_resched();
+
+ return IRQ_HANDLED;
+}
+
+static void flush_tlb_others(cpumask_t cpumask, struct mm_struct *mm,
+ unsigned long va)
+{
+ /*
+ * A couple of (to be removed) sanity checks:
+ *
+ * - current CPU must not be in mask
+ * - mask must exist :)
+ */
+ BUG_ON(cpus_empty(cpumask));
+ BUG_ON(cpu_isset(smp_processor_id(), cpumask));
+ BUG_ON(!mm);
+
+ /* If a CPU which we ran on has gone down, OK. */
+ cpus_and(cpumask, cpumask, cpu_online_map);
+ if (cpus_empty(cpumask))
+ return;
+
+ /*
+ * i'm not happy about this global shared spinlock in the
+ * MM hot path, but we'll see how contended it is.
+ * Temporarily this turns IRQs off, so that lockups are
+ * detected by the NMI watchdog.
+ */
+ spin_lock(&tlbstate_lock);
+
+ flush_mm = mm;
+ flush_va = va;
+#if NR_CPUS <= BITS_PER_LONG
+ atomic_set_mask(cpumask, &flush_cpumask);
+#else
+ {
+ int k;
+ unsigned long *flush_mask = (unsigned long *)&flush_cpumask;
+ unsigned long *cpu_mask = (unsigned long *)&cpumask;
+ for (k = 0; k < BITS_TO_LONGS(NR_CPUS); ++k)
+ atomic_set_mask(cpu_mask[k], &flush_mask[k]);
+ }
+#endif
+ /*
+ * We have to send the IPI only to
+ * CPUs affected.
+ */
+ send_IPI_mask(cpumask, INVALIDATE_TLB_VECTOR);
+
+ while (!cpus_empty(flush_cpumask))
+ /* nothing. lockup detection does not belong here */
+ mb();
+
+ flush_mm = NULL;
+ flush_va = 0;
+ spin_unlock(&tlbstate_lock);
+}
+
+void flush_tlb_current_task(void)
+{
+ struct mm_struct *mm = current->mm;
+ cpumask_t cpu_mask;
+
+ preempt_disable();
+ cpu_mask = mm->cpu_vm_mask;
+ cpu_clear(smp_processor_id(), cpu_mask);
+
+ local_flush_tlb();
+ if (!cpus_empty(cpu_mask))
+ flush_tlb_others(cpu_mask, mm, FLUSH_ALL);
+ preempt_enable();
+}
+
+void flush_tlb_mm (struct mm_struct * mm)
+{
+ cpumask_t cpu_mask;
+
+ preempt_disable();
+ cpu_mask = mm->cpu_vm_mask;
+ cpu_clear(smp_processor_id(), cpu_mask);
+
+ if (current->active_mm == mm) {
+ if (current->mm)
+ local_flush_tlb();
+ else
+ leave_mm(smp_processor_id());
+ }
+ if (!cpus_empty(cpu_mask))
+ flush_tlb_others(cpu_mask, mm, FLUSH_ALL);
+
+ preempt_enable();
+}
+
+void flush_tlb_page(struct vm_area_struct * vma, unsigned long va)
+{
+ struct mm_struct *mm = vma->vm_mm;
+ cpumask_t cpu_mask;
+
+ preempt_disable();
+ cpu_mask = mm->cpu_vm_mask;
+ cpu_clear(smp_processor_id(), cpu_mask);
+
+ if (current->active_mm == mm) {
+ if(current->mm)
+ __flush_tlb_one(va);
+ else
+ leave_mm(smp_processor_id());
+ }
+
+ if (!cpus_empty(cpu_mask))
+ flush_tlb_others(cpu_mask, mm, va);
+
+ preempt_enable();
+}
+EXPORT_SYMBOL(flush_tlb_page);
+
+static void do_flush_tlb_all(void* info)
+{
+ unsigned long cpu = smp_processor_id();
+
+ __flush_tlb_all();
+ if (per_cpu(cpu_tlbstate, cpu).state == TLBSTATE_LAZY)
+ leave_mm(cpu);
+}
+
+void flush_tlb_all(void)
+{
+ on_each_cpu(do_flush_tlb_all, NULL, 1, 1);
+}
+
+#endif /* XEN */
+
+/*
+ * this function sends a 'reschedule' IPI to another CPU.
+ * it goes straight through and wastes no time serializing
+ * anything. Worst case is that we lose a reschedule ...
+ */
+void smp_send_reschedule(int cpu)
+{
+ WARN_ON(cpu_is_offline(cpu));
+ send_IPI_mask(cpumask_of_cpu(cpu), RESCHEDULE_VECTOR);
+}
+
+/*
+ * Structure and data for smp_call_function(). This is designed to minimise
+ * static memory requirements. It also looks cleaner.
+ */
+static DEFINE_SPINLOCK(call_lock);
+
+struct call_data_struct {
+ void (*func) (void *info);
+ void *info;
+ atomic_t started;
+ atomic_t finished;
+ int wait;
+};
+
+void lock_ipi_call_lock(void)
+{
+ spin_lock_irq(&call_lock);
+}
+
+void unlock_ipi_call_lock(void)
+{
+ spin_unlock_irq(&call_lock);
+}
+
+static struct call_data_struct *call_data;
+
+/**
+ * smp_call_function(): Run a function on all other CPUs.
+ * @func: The function to run. This must be fast and non-blocking.
+ * @info: An arbitrary pointer to pass to the function.
+ * @nonatomic: currently unused.
+ * @wait: If true, wait (atomically) until function has completed on other CPUs.
+ *
+ * Returns 0 on success, else a negative status code. Does not return until
+ * remote CPUs are nearly ready to execute <<func>> or are or have executed.
+ *
+ * You must not call this function with disabled interrupts or from a
+ * hardware interrupt handler or from a bottom half handler.
+ */
+int smp_call_function (void (*func) (void *info), void *info, int nonatomic,
+ int wait)
+{
+ struct call_data_struct data;
+ int cpus;
+
+ /* Holding any lock stops cpus from going down. */
+ spin_lock(&call_lock);
+ cpus = num_online_cpus() - 1;
+ if (!cpus) {
+ spin_unlock(&call_lock);
+ return 0;
+ }
+
+ /* Can deadlock when called with interrupts disabled */
+ WARN_ON(irqs_disabled());
+
+ data.func = func;
+ data.info = info;
+ atomic_set(&data.started, 0);
+ data.wait = wait;
+ if (wait)
+ atomic_set(&data.finished, 0);
+
+ call_data = &data;
+ mb();
+
+ /* Send a message to all other CPUs and wait for them to respond */
+ send_IPI_allbutself(CALL_FUNCTION_VECTOR);
+
+ /* Wait for response */
+ while (atomic_read(&data.started) != cpus)
+ cpu_relax();
+
+ if (wait)
+ while (atomic_read(&data.finished) != cpus)
+ cpu_relax();
+ spin_unlock(&call_lock);
+
+ return 0;
+}
+EXPORT_SYMBOL(smp_call_function);
+
+static void stop_this_cpu (void * dummy)
+{
+ /*
+ * Remove this CPU:
+ */
+ cpu_clear(smp_processor_id(), cpu_online_map);
+ local_irq_disable();
+ disable_all_local_evtchn();
+ if (cpu_data[smp_processor_id()].hlt_works_ok)
+ for(;;) halt();
+ for (;;);
+}
+
+/*
+ * this function calls the 'stop' function on all other CPUs in the system.
+ */
+
+void smp_send_stop(void)
+{
+ smp_call_function(stop_this_cpu, NULL, 1, 0);
+
+ local_irq_disable();
+ disable_all_local_evtchn();
+ local_irq_enable();
+}
+
+/*
+ * Reschedule call back. Nothing to do,
+ * all the work is done automatically when
+ * we return from the interrupt.
+ */
+irqreturn_t smp_reschedule_interrupt(int irq, void *dev_id,
+ struct pt_regs *regs)
+{
+
+ return IRQ_HANDLED;
+}
+
+#include <linux/kallsyms.h>
+irqreturn_t smp_call_function_interrupt(int irq, void *dev_id,
+ struct pt_regs *regs)
+{
+ void (*func) (void *info) = call_data->func;
+ void *info = call_data->info;
+ int wait = call_data->wait;
+
+ /*
+ * Notify initiating CPU that I've grabbed the data and am
+ * about to execute the function
+ */
+ mb();
+ atomic_inc(&call_data->started);
+ /*
+ * At this point the info structure may be out of scope unless wait==1
+ */
+ irq_enter();
+ (*func)(info);
+ irq_exit();
+
+ if (wait) {
+ mb();
+ atomic_inc(&call_data->finished);
+ }
+
+ return IRQ_HANDLED;
+}
+
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/kernel/time-xen.c 2010-08-31 09:24:21.000000000 +0200
@@ -0,0 +1,1242 @@
+/*
+ * linux/arch/i386/kernel/time.c
+ *
+ * Copyright (C) 1991, 1992, 1995 Linus Torvalds
+ *
+ * This file contains the PC-specific time handling details:
+ * reading the RTC at bootup, etc..
+ * 1994-07-02 Alan Modra
+ * fixed set_rtc_mmss, fixed time.year for >= 2000, new mktime
+ * 1995-03-26 Markus Kuhn
+ * fixed 500 ms bug at call to set_rtc_mmss, fixed DS12887
+ * precision CMOS clock update
+ * 1996-05-03 Ingo Molnar
+ * fixed time warps in do_[slow|fast]_gettimeoffset()
+ * 1997-09-10 Updated NTP code according to technical memorandum Jan '96
+ * "A Kernel Model for Precision Timekeeping" by Dave Mills
+ * 1998-09-05 (Various)
+ * More robust do_fast_gettimeoffset() algorithm implemented
+ * (works with APM, Cyrix 6x86MX and Centaur C6),
+ * monotonic gettimeofday() with fast_get_timeoffset(),
+ * drift-proof precision TSC calibration on boot
+ * (C. Scott Ananian <cananian@alumni.princeton.edu>, Andrew D.
+ * Balsa <andrebalsa@altern.org>, Philip Gladstone <philip@raptor.com>;
+ * ported from 2.0.35 Jumbo-9 by Michael Krause <m.krause@tu-harburg.de>).
+ * 1998-12-16 Andrea Arcangeli
+ * Fixed Jumbo-9 code in 2.1.131: do_gettimeofday was missing 1 jiffy
+ * because was not accounting lost_ticks.
+ * 1998-12-24 Copyright (C) 1998 Andrea Arcangeli
+ * Fixed a xtime SMP race (we need the xtime_lock rw spinlock to
+ * serialize accesses to xtime/lost_ticks).
+ */
+
+#include <linux/errno.h>
+#include <linux/sched.h>
+#include <linux/kernel.h>
+#include <linux/param.h>
+#include <linux/string.h>
+#include <linux/mm.h>
+#include <linux/interrupt.h>
+#include <linux/time.h>
+#include <linux/delay.h>
+#include <linux/init.h>
+#include <linux/smp.h>
+#include <linux/module.h>
+#include <linux/sysdev.h>
+#include <linux/bcd.h>
+#include <linux/efi.h>
+#include <linux/mca.h>
+#include <linux/sysctl.h>
+#include <linux/percpu.h>
+#include <linux/kernel_stat.h>
+#include <linux/posix-timers.h>
+#include <linux/cpufreq.h>
+
+#include <asm/io.h>
+#include <asm/smp.h>
+#include <asm/irq.h>
+#include <asm/msr.h>
+#include <asm/delay.h>
+#include <asm/mpspec.h>
+#include <asm/uaccess.h>
+#include <asm/processor.h>
+#include <asm/timer.h>
+#include <asm/sections.h>
+
+#include "mach_time.h"
+
+#include <linux/timex.h>
+
+#include <asm/hpet.h>
+
+#include <asm/arch_hooks.h>
+
+#include <xen/evtchn.h>
+#include <xen/interface/vcpu.h>
+
+#if defined (__i386__)
+#include <asm/i8259.h>
+#endif
+
+int pit_latch_buggy; /* extern */
+
+#if defined(__x86_64__)
+unsigned long vxtime_hz = PIT_TICK_RATE;
+struct vxtime_data __vxtime __section_vxtime; /* for vsyscalls */
+volatile unsigned long __jiffies __section_jiffies = INITIAL_JIFFIES;
+unsigned long __wall_jiffies __section_wall_jiffies = INITIAL_JIFFIES;
+struct timespec __xtime __section_xtime;
+struct timezone __sys_tz __section_sys_tz;
+#endif
+
+unsigned int cpu_khz; /* Detected as we calibrate the TSC */
+EXPORT_SYMBOL(cpu_khz);
+
+extern unsigned long wall_jiffies;
+
+DEFINE_SPINLOCK(rtc_lock);
+EXPORT_SYMBOL(rtc_lock);
+
+extern struct init_timer_opts timer_tsc_init;
+extern struct timer_opts timer_tsc;
+#define timer_none timer_tsc
+
+/* These are peridically updated in shared_info, and then copied here. */
+struct shadow_time_info {
+ u64 tsc_timestamp; /* TSC at last update of time vals. */
+ u64 system_timestamp; /* Time, in nanosecs, since boot. */
+ u32 tsc_to_nsec_mul;
+ u32 tsc_to_usec_mul;
+ int tsc_shift;
+ u32 version;
+};
+static DEFINE_PER_CPU(struct shadow_time_info, shadow_time);
+static struct timespec shadow_tv;
+static u32 shadow_tv_version;
+
+/* Keep track of last time we did processing/updating of jiffies and xtime. */
+static u64 processed_system_time; /* System time (ns) at last processing. */
+static DEFINE_PER_CPU(u64, processed_system_time);
+
+/* How much CPU time was spent blocked and how much was 'stolen'? */
+static DEFINE_PER_CPU(u64, processed_stolen_time);
+static DEFINE_PER_CPU(u64, processed_blocked_time);
+
+/* Current runstate of each CPU (updated automatically by the hypervisor). */
+static DEFINE_PER_CPU(struct vcpu_runstate_info, runstate);
+
+/* Must be signed, as it's compared with s64 quantities which can be -ve. */
+#define NS_PER_TICK (1000000000LL/HZ)
+
+static void __clock_was_set(void *unused)
+{
+ clock_was_set();
+}
+static DECLARE_WORK(clock_was_set_work, __clock_was_set, NULL);
+
+/*
+ * GCC 4.3 can turn loops over an induction variable into division. We do
+ * not support arbitrary 64-bit division, and so must break the induction.
+ */
+#define clobber_induction_variable(v) asm ( "" : "+r" (v) )
+
+static inline void __normalize_time(time_t *sec, s64 *nsec)
+{
+ while (*nsec >= NSEC_PER_SEC) {
+ clobber_induction_variable(*nsec);
+ (*nsec) -= NSEC_PER_SEC;
+ (*sec)++;
+ }
+ while (*nsec < 0) {
+ clobber_induction_variable(*nsec);
+ (*nsec) += NSEC_PER_SEC;
+ (*sec)--;
+ }
+}
+
+/* Does this guest OS track Xen time, or set its wall clock independently? */
+static int independent_wallclock = 0;
+static int __init __independent_wallclock(char *str)
+{
+ independent_wallclock = 1;
+ return 1;
+}
+__setup("independent_wallclock", __independent_wallclock);
+
+/* Permitted clock jitter, in nsecs, beyond which a warning will be printed. */
+static unsigned long permitted_clock_jitter = 10000000UL; /* 10ms */
+static int __init __permitted_clock_jitter(char *str)
+{
+ permitted_clock_jitter = simple_strtoul(str, NULL, 0);
+ return 1;
+}
+__setup("permitted_clock_jitter=", __permitted_clock_jitter);
+
+#if 0
+static void delay_tsc(unsigned long loops)
+{
+ unsigned long bclock, now;
+
+ rdtscl(bclock);
+ do {
+ rep_nop();
+ rdtscl(now);
+ } while ((now - bclock) < loops);
+}
+
+struct timer_opts timer_tsc = {
+ .name = "tsc",
+ .delay = delay_tsc,
+};
+#endif
+
+/*
+ * Scale a 64-bit delta by scaling and multiplying by a 32-bit fraction,
+ * yielding a 64-bit result.
+ */
+static inline u64 scale_delta(u64 delta, u32 mul_frac, int shift)
+{
+ u64 product;
+#ifdef __i386__
+ u32 tmp1, tmp2;
+#endif
+
+ if (shift < 0)
+ delta >>= -shift;
+ else
+ delta <<= shift;
+
+#ifdef __i386__
+ __asm__ (
+ "mul %5 ; "
+ "mov %4,%%eax ; "
+ "mov %%edx,%4 ; "
+ "mul %5 ; "
+ "xor %5,%5 ; "
+ "add %4,%%eax ; "
+ "adc %5,%%edx ; "
+ : "=A" (product), "=r" (tmp1), "=r" (tmp2)
+ : "a" ((u32)delta), "1" ((u32)(delta >> 32)), "2" (mul_frac) );
+#else
+ __asm__ (
+ "mul %%rdx ; shrd $32,%%rdx,%%rax"
+ : "=a" (product) : "0" (delta), "d" ((u64)mul_frac) );
+#endif
+
+ return product;
+}
+
+#if 0 /* defined (__i386__) */
+int read_current_timer(unsigned long *timer_val)
+{
+ rdtscl(*timer_val);
+ return 0;
+}
+#endif
+
+void init_cpu_khz(void)
+{
+ u64 __cpu_khz = 1000000ULL << 32;
+ struct vcpu_time_info *info = &vcpu_info(0)->time;
+ do_div(__cpu_khz, info->tsc_to_system_mul);
+ if (info->tsc_shift < 0)
+ cpu_khz = __cpu_khz << -info->tsc_shift;
+ else
+ cpu_khz = __cpu_khz >> info->tsc_shift;
+}
+
+static u64 get_nsec_offset(struct shadow_time_info *shadow)
+{
+ u64 now, delta;
+ rdtscll(now);
+ delta = now - shadow->tsc_timestamp;
+ return scale_delta(delta, shadow->tsc_to_nsec_mul, shadow->tsc_shift);
+}
+
+static unsigned long get_usec_offset(struct shadow_time_info *shadow)
+{
+ u64 now, delta;
+ rdtscll(now);
+ delta = now - shadow->tsc_timestamp;
+ return scale_delta(delta, shadow->tsc_to_usec_mul, shadow->tsc_shift);
+}
+
+static void __update_wallclock(time_t sec, long nsec)
+{
+ long wtm_nsec, xtime_nsec;
+ time_t wtm_sec, xtime_sec;
+ u64 tmp, wc_nsec;
+
+ /* Adjust wall-clock time base based on wall_jiffies ticks. */
+ wc_nsec = processed_system_time;
+ wc_nsec += sec * (u64)NSEC_PER_SEC;
+ wc_nsec += nsec;
+ wc_nsec -= (jiffies - wall_jiffies) * (u64)NS_PER_TICK;
+
+ /* Split wallclock base into seconds and nanoseconds. */
+ tmp = wc_nsec;
+ xtime_nsec = do_div(tmp, 1000000000);
+ xtime_sec = (time_t)tmp;
+
+ wtm_sec = wall_to_monotonic.tv_sec + (xtime.tv_sec - xtime_sec);
+ wtm_nsec = wall_to_monotonic.tv_nsec + (xtime.tv_nsec - xtime_nsec);
+
+ set_normalized_timespec(&xtime, xtime_sec, xtime_nsec);
+ set_normalized_timespec(&wall_to_monotonic, wtm_sec, wtm_nsec);
+}
+
+static void update_wallclock(void)
+{
+ shared_info_t *s = HYPERVISOR_shared_info;
+
+ do {
+ shadow_tv_version = s->wc_version;
+ rmb();
+ shadow_tv.tv_sec = s->wc_sec;
+ shadow_tv.tv_nsec = s->wc_nsec;
+ rmb();
+ } while ((s->wc_version & 1) | (shadow_tv_version ^ s->wc_version));
+
+ if (!independent_wallclock)
+ __update_wallclock(shadow_tv.tv_sec, shadow_tv.tv_nsec);
+}
+
+/*
+ * Reads a consistent set of time-base values from Xen, into a shadow data
+ * area.
+ */
+static void get_time_values_from_xen(unsigned int cpu)
+{
+ struct vcpu_time_info *src;
+ struct shadow_time_info *dst;
+ unsigned long flags;
+ u32 pre_version, post_version;
+
+ src = &vcpu_info(cpu)->time;
+ dst = &per_cpu(shadow_time, cpu);
+
+ local_irq_save(flags);
+
+ do {
+ pre_version = dst->version = src->version;
+ rmb();
+ dst->tsc_timestamp = src->tsc_timestamp;
+ dst->system_timestamp = src->system_time;
+ dst->tsc_to_nsec_mul = src->tsc_to_system_mul;
+ dst->tsc_shift = src->tsc_shift;
+ rmb();
+ post_version = src->version;
+ } while ((pre_version & 1) | (pre_version ^ post_version));
+
+ dst->tsc_to_usec_mul = dst->tsc_to_nsec_mul / 1000;
+
+ local_irq_restore(flags);
+}
+
+static inline int time_values_up_to_date(unsigned int cpu)
+{
+ struct vcpu_time_info *src;
+ struct shadow_time_info *dst;
+
+ src = &vcpu_info(cpu)->time;
+ dst = &per_cpu(shadow_time, cpu);
+
+ rmb();
+ return (dst->version == src->version);
+}
+
+/*
+ * This is a special lock that is owned by the CPU and holds the index
+ * register we are working with. It is required for NMI access to the
+ * CMOS/RTC registers. See include/asm-i386/mc146818rtc.h for details.
+ */
+volatile unsigned long cmos_lock = 0;
+EXPORT_SYMBOL(cmos_lock);
+
+/* Routines for accessing the CMOS RAM/RTC. */
+unsigned char rtc_cmos_read(unsigned char addr)
+{
+ unsigned char val;
+ lock_cmos_prefix(addr);
+ outb_p(addr, RTC_PORT(0));
+ val = inb_p(RTC_PORT(1));
+ lock_cmos_suffix(addr);
+ return val;
+}
+EXPORT_SYMBOL(rtc_cmos_read);
+
+void rtc_cmos_write(unsigned char val, unsigned char addr)
+{
+ lock_cmos_prefix(addr);
+ outb_p(addr, RTC_PORT(0));
+ outb_p(val, RTC_PORT(1));
+ lock_cmos_suffix(addr);
+}
+EXPORT_SYMBOL(rtc_cmos_write);
+
+static struct {
+ spinlock_t lock;
+ struct timeval tv;
+ u32 version;
+} monotonic = { .lock = SPIN_LOCK_UNLOCKED };
+
+/*
+ * This version of gettimeofday has microsecond resolution
+ * and better than microsecond precision on fast x86 machines with TSC.
+ */
+void do_gettimeofday(struct timeval *tv)
+{
+ unsigned long seq;
+ unsigned long usec, sec;
+ unsigned long flags;
+ s64 nsec;
+ unsigned int cpu;
+ struct shadow_time_info *shadow;
+ u32 local_time_version, monotonic_version;
+
+ cpu = get_cpu();
+ shadow = &per_cpu(shadow_time, cpu);
+
+ do {
+ unsigned long lost;
+
+ local_time_version = shadow->version;
+ seq = read_seqbegin(&xtime_lock);
+
+ usec = get_usec_offset(shadow);
+ lost = jiffies - wall_jiffies;
+
+ if (unlikely(lost))
+ usec += lost * (USEC_PER_SEC / HZ);
+
+ sec = xtime.tv_sec;
+ usec += (xtime.tv_nsec / NSEC_PER_USEC);
+
+ nsec = shadow->system_timestamp - processed_system_time;
+ __normalize_time(&sec, &nsec);
+ usec += (long)nsec / NSEC_PER_USEC;
+
+ monotonic_version = monotonic.version;
+
+ if (unlikely(!time_values_up_to_date(cpu))) {
+ /*
+ * We may have blocked for a long time,
+ * rendering our calculations invalid
+ * (e.g. the time delta may have
+ * overflowed). Detect that and recalculate
+ * with fresh values.
+ */
+ get_time_values_from_xen(cpu);
+ continue;
+ }
+ } while (read_seqretry(&xtime_lock, seq) ||
+ (local_time_version != shadow->version));
+
+ put_cpu();
+
+ while (usec >= USEC_PER_SEC) {
+ usec -= USEC_PER_SEC;
+ sec++;
+ }
+
+ spin_lock_irqsave(&monotonic.lock, flags);
+ if (unlikely(sec < monotonic.tv.tv_sec) ||
+ (sec == monotonic.tv.tv_sec && usec <= monotonic.tv.tv_usec)) {
+ sec = monotonic.tv.tv_sec;
+ usec = monotonic.tv.tv_usec;
+ } else if (likely(monotonic_version == monotonic.version)) {
+ monotonic.tv.tv_sec = sec;
+ monotonic.tv.tv_usec = usec;
+ }
+ spin_unlock_irqrestore(&monotonic.lock, flags);
+
+ tv->tv_sec = sec;
+ tv->tv_usec = usec;
+}
+
+EXPORT_SYMBOL(do_gettimeofday);
+
+/* Reset monotonic gettimeofday() timeval. */
+static inline void monotonic_reset(void)
+{
+ spin_lock(&monotonic.lock);
+ monotonic.tv.tv_sec = 0;
+ monotonic.tv.tv_usec = 0;
+ ++monotonic.version;
+ spin_unlock(&monotonic.lock);
+}
+
+int do_settimeofday(struct timespec *tv)
+{
+ time_t sec;
+ s64 nsec;
+ unsigned int cpu;
+ struct shadow_time_info *shadow;
+ struct xen_platform_op op;
+
+ if (unlikely(!tv)) {
+ monotonic_reset();
+ return 0;
+ }
+
+ if ((unsigned long)tv->tv_nsec >= NSEC_PER_SEC)
+ return -EINVAL;
+
+ if (!is_initial_xendomain() && !independent_wallclock)
+ return -EPERM;
+
+ cpu = get_cpu();
+ shadow = &per_cpu(shadow_time, cpu);
+
+ write_seqlock_irq(&xtime_lock);
+
+ /*
+ * Ensure we don't get blocked for a long time so that our time delta
+ * overflows. If that were to happen then our shadow time values would
+ * be stale, so we can retry with fresh ones.
+ */
+ for (;;) {
+ nsec = tv->tv_nsec - get_nsec_offset(shadow);
+ if (time_values_up_to_date(cpu))
+ break;
+ get_time_values_from_xen(cpu);
+ }
+ sec = tv->tv_sec;
+ __normalize_time(&sec, &nsec);
+
+ if (is_initial_xendomain() && !independent_wallclock) {
+ op.cmd = XENPF_settime;
+ op.u.settime.secs = sec;
+ op.u.settime.nsecs = nsec;
+ op.u.settime.system_time = shadow->system_timestamp;
+ WARN_ON(HYPERVISOR_platform_op(&op));
+ update_wallclock();
+ } else if (independent_wallclock) {
+ nsec -= shadow->system_timestamp;
+ __normalize_time(&sec, &nsec);
+ __update_wallclock(sec, nsec);
+ }
+ ntp_clear();
+
+ monotonic_reset();
+
+ write_sequnlock_irq(&xtime_lock);
+
+ put_cpu();
+
+ clock_was_set();
+ return 0;
+}
+
+EXPORT_SYMBOL(do_settimeofday);
+
+static void sync_xen_wallclock(unsigned long dummy);
+static DEFINE_TIMER(sync_xen_wallclock_timer, sync_xen_wallclock, 0, 0);
+static void sync_xen_wallclock(unsigned long dummy)
+{
+ time_t sec;
+ s64 nsec;
+ struct xen_platform_op op;
+
+ if (!ntp_synced() || independent_wallclock || !is_initial_xendomain())
+ return;
+
+ write_seqlock_irq(&xtime_lock);
+
+ sec = xtime.tv_sec;
+ nsec = xtime.tv_nsec + ((jiffies - wall_jiffies) * (u64)NS_PER_TICK);
+ __normalize_time(&sec, &nsec);
+
+ op.cmd = XENPF_settime;
+ op.u.settime.secs = sec;
+ op.u.settime.nsecs = nsec;
+ op.u.settime.system_time = processed_system_time;
+ WARN_ON(HYPERVISOR_platform_op(&op));
+
+ update_wallclock();
+
+ write_sequnlock_irq(&xtime_lock);
+
+ /* Once per minute. */
+ mod_timer(&sync_xen_wallclock_timer, jiffies + 60*HZ);
+}
+
+static int set_rtc_mmss(unsigned long nowtime)
+{
+ int retval;
+ unsigned long flags;
+
+ if (independent_wallclock || !is_initial_xendomain())
+ return 0;
+
+ /* gets recalled with irq locally disabled */
+ /* XXX - does irqsave resolve this? -johnstul */
+ spin_lock_irqsave(&rtc_lock, flags);
+ if (efi_enabled)
+ retval = efi_set_rtc_mmss(nowtime);
+ else
+ retval = mach_set_rtc_mmss(nowtime);
+ spin_unlock_irqrestore(&rtc_lock, flags);
+
+ return retval;
+}
+
+/* monotonic_clock(): returns # of nanoseconds passed since time_init()
+ * Note: This function is required to return accurate
+ * time even in the absence of multiple timer ticks.
+ */
+unsigned long long monotonic_clock(void)
+{
+ unsigned int cpu = get_cpu();
+ struct shadow_time_info *shadow = &per_cpu(shadow_time, cpu);
+ u64 time;
+ u32 local_time_version;
+
+ do {
+ local_time_version = shadow->version;
+ barrier();
+ time = shadow->system_timestamp + get_nsec_offset(shadow);
+ if (!time_values_up_to_date(cpu))
+ get_time_values_from_xen(cpu);
+ barrier();
+ } while (local_time_version != shadow->version);
+
+ put_cpu();
+
+ return time;
+}
+EXPORT_SYMBOL(monotonic_clock);
+
+#ifdef __x86_64__
+unsigned long long sched_clock(void)
+{
+ return monotonic_clock();
+}
+#endif
+
+#if defined(CONFIG_SMP) && defined(CONFIG_FRAME_POINTER)
+unsigned long profile_pc(struct pt_regs *regs)
+{
+ unsigned long pc = instruction_pointer(regs);
+
+#ifdef __x86_64__
+ /* Assume the lock function has either no stack frame or only a single word.
+ This checks if the address on the stack looks like a kernel text address.
+ There is a small window for false hits, but in that case the tick
+ is just accounted to the spinlock function.
+ Better would be to write these functions in assembler again
+ and check exactly. */
+ if (!user_mode_vm(regs) && in_lock_functions(pc)) {
+ char *v = *(char **)regs->rsp;
+ if ((v >= _stext && v <= _etext) ||
+ (v >= _sinittext && v <= _einittext) ||
+ (v >= (char *)MODULES_VADDR && v <= (char *)MODULES_END))
+ return (unsigned long)v;
+ return ((unsigned long *)regs->rsp)[1];
+ }
+#else
+ if (!user_mode_vm(regs) && in_lock_functions(pc))
+ return *(unsigned long *)(regs->ebp + 4);
+#endif
+
+ return pc;
+}
+EXPORT_SYMBOL(profile_pc);
+#endif
+
+/*
+ * This is the same as the above, except we _also_ save the current
+ * Time Stamp Counter value at the time of the timer interrupt, so that
+ * we later on can estimate the time of day more exactly.
+ */
+irqreturn_t timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
+{
+ s64 delta, delta_cpu, stolen, blocked;
+ u64 sched_time;
+ unsigned int i, cpu = smp_processor_id();
+ int schedule_clock_was_set_work = 0;
+ struct shadow_time_info *shadow = &per_cpu(shadow_time, cpu);
+ struct vcpu_runstate_info *runstate = &per_cpu(runstate, cpu);
+
+ /*
+ * Here we are in the timer irq handler. We just have irqs locally
+ * disabled but we don't know if the timer_bh is running on the other
+ * CPU. We need to avoid to SMP race with it. NOTE: we don' t need
+ * the irq version of write_lock because as just said we have irq
+ * locally disabled. -arca
+ */
+ write_seqlock(&xtime_lock);
+
+ do {
+ get_time_values_from_xen(cpu);
+
+ /* Obtain a consistent snapshot of elapsed wallclock cycles. */
+ delta = delta_cpu =
+ shadow->system_timestamp + get_nsec_offset(shadow);
+ delta -= processed_system_time;
+ delta_cpu -= per_cpu(processed_system_time, cpu);
+
+ /*
+ * Obtain a consistent snapshot of stolen/blocked cycles. We
+ * can use state_entry_time to detect if we get preempted here.
+ */
+ do {
+ sched_time = runstate->state_entry_time;
+ barrier();
+ stolen = runstate->time[RUNSTATE_runnable] +
+ runstate->time[RUNSTATE_offline] -
+ per_cpu(processed_stolen_time, cpu);
+ blocked = runstate->time[RUNSTATE_blocked] -
+ per_cpu(processed_blocked_time, cpu);
+ barrier();
+ } while (sched_time != runstate->state_entry_time);
+ } while (!time_values_up_to_date(cpu));
+
+ if ((unlikely(delta < -(s64)permitted_clock_jitter) ||
+ unlikely(delta_cpu < -(s64)permitted_clock_jitter))
+ && printk_ratelimit()) {
+ printk("Timer ISR/%u: Time went backwards: "
+ "delta=%lld delta_cpu=%lld shadow=%lld "
+ "off=%lld processed=%lld cpu_processed=%lld\n",
+ cpu, delta, delta_cpu, shadow->system_timestamp,
+ (s64)get_nsec_offset(shadow),
+ processed_system_time,
+ per_cpu(processed_system_time, cpu));
+ for (i = 0; i < num_online_cpus(); i++)
+ printk(" %d: %lld\n", i,
+ per_cpu(processed_system_time, i));
+ }
+
+ /* System-wide jiffy work. */
+ while (delta >= NS_PER_TICK) {
+ delta -= NS_PER_TICK;
+ processed_system_time += NS_PER_TICK;
+ do_timer(regs);
+ }
+
+ if (shadow_tv_version != HYPERVISOR_shared_info->wc_version) {
+ update_wallclock();
+ schedule_clock_was_set_work = 1;
+ }
+
+ write_sequnlock(&xtime_lock);
+
+ if (schedule_clock_was_set_work && keventd_up())
+ schedule_work(&clock_was_set_work);
+
+ /*
+ * Account stolen ticks.
+ * HACK: Passing NULL to account_steal_time()
+ * ensures that the ticks are accounted as stolen.
+ */
+ if ((stolen > 0) && (delta_cpu > 0)) {
+ delta_cpu -= stolen;
+ if (unlikely(delta_cpu < 0))
+ stolen += delta_cpu; /* clamp local-time progress */
+ do_div(stolen, NS_PER_TICK);
+ per_cpu(processed_stolen_time, cpu) += stolen * NS_PER_TICK;
+ per_cpu(processed_system_time, cpu) += stolen * NS_PER_TICK;
+ account_steal_time(NULL, (cputime_t)stolen);
+ }
+
+ /*
+ * Account blocked ticks.
+ * HACK: Passing idle_task to account_steal_time()
+ * ensures that the ticks are accounted as idle/wait.
+ */
+ if ((blocked > 0) && (delta_cpu > 0)) {
+ delta_cpu -= blocked;
+ if (unlikely(delta_cpu < 0))
+ blocked += delta_cpu; /* clamp local-time progress */
+ do_div(blocked, NS_PER_TICK);
+ per_cpu(processed_blocked_time, cpu) += blocked * NS_PER_TICK;
+ per_cpu(processed_system_time, cpu) += blocked * NS_PER_TICK;
+ account_steal_time(idle_task(cpu), (cputime_t)blocked);
+ }
+
+ /* Account user/system ticks. */
+ if (delta_cpu > 0) {
+ do_div(delta_cpu, NS_PER_TICK);
+ per_cpu(processed_system_time, cpu) += delta_cpu * NS_PER_TICK;
+ if (user_mode_vm(regs))
+ account_user_time(current, (cputime_t)delta_cpu);
+ else
+ account_system_time(current, HARDIRQ_OFFSET,
+ (cputime_t)delta_cpu);
+ }
+
+ /* Offlined for more than a few seconds? Avoid lockup warnings. */
+ if (stolen > 5*HZ)
+ touch_softlockup_watchdog();
+
+ /* Local timer processing (see update_process_times()). */
+ run_local_timers();
+ if (rcu_pending(cpu))
+ rcu_check_callbacks(cpu, user_mode_vm(regs));
+ scheduler_tick();
+ run_posix_cpu_timers(current);
+ profile_tick(CPU_PROFILING, regs);
+
+ return IRQ_HANDLED;
+}
+
+static void init_missing_ticks_accounting(unsigned int cpu)
+{
+ struct vcpu_register_runstate_memory_area area;
+ struct vcpu_runstate_info *runstate = &per_cpu(runstate, cpu);
+ int rc;
+
+ memset(runstate, 0, sizeof(*runstate));
+
+ area.addr.v = runstate;
+ rc = HYPERVISOR_vcpu_op(VCPUOP_register_runstate_memory_area, cpu, &area);
+ WARN_ON(rc && rc != -ENOSYS);
+
+ per_cpu(processed_blocked_time, cpu) =
+ runstate->time[RUNSTATE_blocked];
+ per_cpu(processed_stolen_time, cpu) =
+ runstate->time[RUNSTATE_runnable] +
+ runstate->time[RUNSTATE_offline];
+}
+
+/* not static: needed by APM */
+unsigned long get_cmos_time(void)
+{
+ unsigned long retval;
+ unsigned long flags;
+
+ spin_lock_irqsave(&rtc_lock, flags);
+
+ if (efi_enabled)
+ retval = efi_get_time();
+ else
+ retval = mach_get_cmos_time();
+
+ spin_unlock_irqrestore(&rtc_lock, flags);
+
+ return retval;
+}
+EXPORT_SYMBOL(get_cmos_time);
+
+static void sync_cmos_clock(unsigned long dummy);
+
+static DEFINE_TIMER(sync_cmos_timer, sync_cmos_clock, 0, 0);
+
+static void sync_cmos_clock(unsigned long dummy)
+{
+ struct timeval now, next;
+ int fail = 1;
+
+ /*
+ * If we have an externally synchronized Linux clock, then update
+ * CMOS clock accordingly every ~11 minutes. Set_rtc_mmss() has to be
+ * called as close as possible to 500 ms before the new second starts.
+ * This code is run on a timer. If the clock is set, that timer
+ * may not expire at the correct time. Thus, we adjust...
+ */
+ if (!ntp_synced())
+ /*
+ * Not synced, exit, do not restart a timer (if one is
+ * running, let it run out).
+ */
+ return;
+
+ do_gettimeofday(&now);
+ if (now.tv_usec >= USEC_AFTER - ((unsigned) TICK_SIZE) / 2 &&
+ now.tv_usec <= USEC_BEFORE + ((unsigned) TICK_SIZE) / 2)
+ fail = set_rtc_mmss(now.tv_sec);
+
+ next.tv_usec = USEC_AFTER - now.tv_usec;
+ if (next.tv_usec <= 0)
+ next.tv_usec += USEC_PER_SEC;
+
+ if (!fail)
+ next.tv_sec = 659;
+ else
+ next.tv_sec = 0;
+
+ if (next.tv_usec >= USEC_PER_SEC) {
+ next.tv_sec++;
+ next.tv_usec -= USEC_PER_SEC;
+ }
+ mod_timer(&sync_cmos_timer, jiffies + timeval_to_jiffies(&next));
+}
+
+void notify_arch_cmos_timer(void)
+{
+ mod_timer(&sync_cmos_timer, jiffies + 1);
+ mod_timer(&sync_xen_wallclock_timer, jiffies + 1);
+}
+
+static int timer_resume(struct sys_device *dev)
+{
+ extern void time_resume(void);
+ time_resume();
+ return 0;
+}
+
+static struct sysdev_class timer_sysclass = {
+ .resume = timer_resume,
+ set_kset_name("timer"),
+};
+
+
+/* XXX this driverfs stuff should probably go elsewhere later -john */
+static struct sys_device device_timer = {
+ .id = 0,
+ .cls = &timer_sysclass,
+};
+
+static int time_init_device(void)
+{
+ int error = sysdev_class_register(&timer_sysclass);
+ if (!error)
+ error = sysdev_register(&device_timer);
+ return error;
+}
+
+device_initcall(time_init_device);
+
+#ifdef CONFIG_HPET_TIMER
+extern void (*late_time_init)(void);
+/* Duplicate of time_init() below, with hpet_enable part added */
+static void __init hpet_time_init(void)
+{
+ xtime.tv_sec = get_cmos_time();
+ xtime.tv_nsec = (INITIAL_JIFFIES % HZ) * (NSEC_PER_SEC / HZ);
+ set_normalized_timespec(&wall_to_monotonic,
+ -xtime.tv_sec, -xtime.tv_nsec);
+
+ if ((hpet_enable() >= 0) && hpet_use_timer) {
+ printk("Using HPET for base-timer\n");
+ }
+
+ time_init_hook();
+}
+#endif
+
+/* Dynamically-mapped IRQ. */
+DEFINE_PER_CPU(int, timer_irq);
+
+extern void (*late_time_init)(void);
+static void setup_cpu0_timer_irq(void)
+{
+ per_cpu(timer_irq, 0) =
+ bind_virq_to_irqhandler(
+ VIRQ_TIMER,
+ 0,
+ timer_interrupt,
+ SA_INTERRUPT,
+ "timer0",
+ NULL);
+ BUG_ON(per_cpu(timer_irq, 0) < 0);
+}
+
+static struct vcpu_set_periodic_timer xen_set_periodic_tick = {
+ .period_ns = NS_PER_TICK
+};
+
+void __init time_init(void)
+{
+#ifdef CONFIG_HPET_TIMER
+ if (is_hpet_capable()) {
+ /*
+ * HPET initialization needs to do memory-mapped io. So, let
+ * us do a late initialization after mem_init().
+ */
+ late_time_init = hpet_time_init;
+ return;
+ }
+#endif
+
+ switch (HYPERVISOR_vcpu_op(VCPUOP_set_periodic_timer, 0,
+ &xen_set_periodic_tick)) {
+ case 0:
+#if CONFIG_XEN_COMPAT <= 0x030004
+ case -ENOSYS:
+#endif
+ break;
+ default:
+ BUG();
+ }
+
+ get_time_values_from_xen(0);
+
+ processed_system_time = per_cpu(shadow_time, 0).system_timestamp;
+ per_cpu(processed_system_time, 0) = processed_system_time;
+ init_missing_ticks_accounting(0);
+
+ update_wallclock();
+
+ init_cpu_khz();
+ printk(KERN_INFO "Xen reported: %u.%03u MHz processor.\n",
+ cpu_khz / 1000, cpu_khz % 1000);
+
+#if defined(__x86_64__)
+ vxtime.mode = VXTIME_TSC;
+ vxtime.quot = (1000000L << 32) / vxtime_hz;
+ vxtime.tsc_quot = (1000L << 32) / cpu_khz;
+ sync_core();
+ rdtscll(vxtime.last_tsc);
+#endif
+
+ /* Cannot request_irq() until kmem is initialised. */
+ late_time_init = setup_cpu0_timer_irq;
+}
+
+/* Convert jiffies to system time. */
+u64 jiffies_to_st(unsigned long j)
+{
+ unsigned long seq;
+ long delta;
+ u64 st;
+
+ do {
+ seq = read_seqbegin(&xtime_lock);
+ delta = j - jiffies;
+ if (delta < 1) {
+ /* Triggers in some wrap-around cases, but that's okay:
+ * we just end up with a shorter timeout. */
+ st = processed_system_time + NS_PER_TICK;
+ } else if (((unsigned long)delta >> (BITS_PER_LONG-3)) != 0) {
+ /* Very long timeout means there is no pending timer.
+ * We indicate this to Xen by passing zero timeout. */
+ st = 0;
+ } else {
+ st = processed_system_time + delta * (u64)NS_PER_TICK;
+ }
+ } while (read_seqretry(&xtime_lock, seq));
+
+ return st;
+}
+EXPORT_SYMBOL(jiffies_to_st);
+
+/*
+ * stop_hz_timer / start_hz_timer - enter/exit 'tickless mode' on an idle cpu
+ * These functions are based on implementations from arch/s390/kernel/time.c
+ */
+static void stop_hz_timer(void)
+{
+ struct vcpu_set_singleshot_timer singleshot;
+ unsigned int cpu = smp_processor_id();
+ unsigned long j;
+ int rc;
+
+ cpu_set(cpu, nohz_cpu_mask);
+
+ /* See matching smp_mb in rcu_start_batch in rcupdate.c. These mbs */
+ /* ensure that if __rcu_pending (nested in rcu_needs_cpu) fetches a */
+ /* value of rcp->cur that matches rdp->quiescbatch and allows us to */
+ /* stop the hz timer then the cpumasks created for subsequent values */
+ /* of cur in rcu_start_batch are guaranteed to pick up the updated */
+ /* nohz_cpu_mask and so will not depend on this cpu. */
+
+ smp_mb();
+
+ /* Leave ourselves in tick mode if rcu or softirq or timer pending. */
+ if (rcu_needs_cpu(cpu) || local_softirq_pending() ||
+ (j = next_timer_interrupt(), time_before_eq(j, jiffies))) {
+ cpu_clear(cpu, nohz_cpu_mask);
+ j = jiffies + 1;
+ }
+
+ singleshot.timeout_abs_ns = jiffies_to_st(j);
+ if (!singleshot.timeout_abs_ns)
+ return;
+ singleshot.timeout_abs_ns += NS_PER_TICK / 2;
+ singleshot.flags = 0;
+ rc = HYPERVISOR_vcpu_op(VCPUOP_set_singleshot_timer, cpu, &singleshot);
+#if CONFIG_XEN_COMPAT <= 0x030004
+ if (rc) {
+ BUG_ON(rc != -ENOSYS);
+ rc = HYPERVISOR_set_timer_op(singleshot.timeout_abs_ns);
+ }
+#endif
+ BUG_ON(rc);
+}
+
+static void start_hz_timer(void)
+{
+ unsigned int cpu = smp_processor_id();
+ int rc = HYPERVISOR_vcpu_op(VCPUOP_stop_singleshot_timer, cpu, NULL);
+
+#if CONFIG_XEN_COMPAT <= 0x030004
+ if (rc) {
+ BUG_ON(rc != -ENOSYS);
+ rc = HYPERVISOR_set_timer_op(0);
+ }
+#endif
+ BUG_ON(rc);
+ cpu_clear(cpu, nohz_cpu_mask);
+}
+
+void raw_safe_halt(void)
+{
+ stop_hz_timer();
+ /* Blocking includes an implicit local_irq_enable(). */
+ HYPERVISOR_block();
+ start_hz_timer();
+}
+EXPORT_SYMBOL(raw_safe_halt);
+
+void halt(void)
+{
+ if (irqs_disabled())
+ VOID(HYPERVISOR_vcpu_op(VCPUOP_down, smp_processor_id(), NULL));
+}
+EXPORT_SYMBOL(halt);
+
+/* No locking required. Interrupts are disabled on all CPUs. */
+void time_resume(void)
+{
+ unsigned int cpu;
+
+ init_cpu_khz();
+
+ for_each_online_cpu(cpu) {
+ switch (HYPERVISOR_vcpu_op(VCPUOP_set_periodic_timer, cpu,
+ &xen_set_periodic_tick)) {
+ case 0:
+#if CONFIG_XEN_COMPAT <= 0x030004
+ case -ENOSYS:
+#endif
+ break;
+ default:
+ BUG();
+ }
+ get_time_values_from_xen(cpu);
+ per_cpu(processed_system_time, cpu) =
+ per_cpu(shadow_time, 0).system_timestamp;
+ init_missing_ticks_accounting(cpu);
+ }
+
+ processed_system_time = per_cpu(shadow_time, 0).system_timestamp;
+
+ update_wallclock();
+}
+
+#ifdef CONFIG_SMP
+static char timer_name[NR_CPUS][15];
+
+int __cpuinit local_setup_timer(unsigned int cpu)
+{
+ int seq, irq;
+
+ BUG_ON(cpu == 0);
+
+ switch (HYPERVISOR_vcpu_op(VCPUOP_set_periodic_timer, cpu,
+ &xen_set_periodic_tick)) {
+ case 0:
+#if CONFIG_XEN_COMPAT <= 0x030004
+ case -ENOSYS:
+#endif
+ break;
+ default:
+ BUG();
+ }
+
+ do {
+ seq = read_seqbegin(&xtime_lock);
+ /* Use cpu0 timestamp: cpu's shadow is not initialised yet. */
+ per_cpu(processed_system_time, cpu) =
+ per_cpu(shadow_time, 0).system_timestamp;
+ init_missing_ticks_accounting(cpu);
+ } while (read_seqretry(&xtime_lock, seq));
+
+ sprintf(timer_name[cpu], "timer%u", cpu);
+ irq = bind_virq_to_irqhandler(VIRQ_TIMER,
+ cpu,
+ timer_interrupt,
+ SA_INTERRUPT,
+ timer_name[cpu],
+ NULL);
+ if (irq < 0)
+ return irq;
+ per_cpu(timer_irq, cpu) = irq;
+
+ return 0;
+}
+
+void __cpuexit local_teardown_timer(unsigned int cpu)
+{
+ BUG_ON(cpu == 0);
+ unbind_from_irqhandler(per_cpu(timer_irq, cpu), NULL);
+}
+#endif
+
+#ifdef CONFIG_CPU_FREQ
+static int time_cpufreq_notifier(struct notifier_block *nb, unsigned long val,
+ void *data)
+{
+ struct cpufreq_freqs *freq = data;
+ struct xen_platform_op op;
+
+ if (cpu_has(&cpu_data[freq->cpu], X86_FEATURE_CONSTANT_TSC))
+ return 0;
+
+ if (val == CPUFREQ_PRECHANGE)
+ return 0;
+
+ op.cmd = XENPF_change_freq;
+ op.u.change_freq.flags = 0;
+ op.u.change_freq.cpu = freq->cpu;
+ op.u.change_freq.freq = (u64)freq->new * 1000;
+ WARN_ON(HYPERVISOR_platform_op(&op));
+
+ return 0;
+}
+
+static struct notifier_block time_cpufreq_notifier_block = {
+ .notifier_call = time_cpufreq_notifier
+};
+
+static int __init cpufreq_time_setup(void)
+{
+ if (!cpufreq_register_notifier(&time_cpufreq_notifier_block,
+ CPUFREQ_TRANSITION_NOTIFIER)) {
+ printk(KERN_ERR "failed to set up cpufreq notifier\n");
+ return -ENODEV;
+ }
+ return 0;
+}
+
+core_initcall(cpufreq_time_setup);
+#endif
+
+/*
+ * /proc/sys/xen: This really belongs in another file. It can stay here for
+ * now however.
+ */
+static ctl_table xen_subtable[] = {
+ {
+ .ctl_name = 1,
+ .procname = "independent_wallclock",
+ .data = &independent_wallclock,
+ .maxlen = sizeof(independent_wallclock),
+ .mode = 0644,
+ .proc_handler = proc_dointvec
+ },
+ {
+ .ctl_name = 2,
+ .procname = "permitted_clock_jitter",
+ .data = &permitted_clock_jitter,
+ .maxlen = sizeof(permitted_clock_jitter),
+ .mode = 0644,
+ .proc_handler = proc_doulongvec_minmax
+ },
+ { 0 }
+};
+static ctl_table xen_table[] = {
+ {
+ .ctl_name = 123,
+ .procname = "xen",
+ .mode = 0555,
+ .child = xen_subtable},
+ { 0 }
+};
+static int __init xen_sysctl_init(void)
+{
+ (void)register_sysctl_table(xen_table, 0);
+ return 0;
+}
+__initcall(xen_sysctl_init);
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/kernel/traps_32-xen.c 2008-04-02 12:34:02.000000000 +0200
@@ -0,0 +1,1190 @@
+/*
+ * linux/arch/i386/traps.c
+ *
+ * Copyright (C) 1991, 1992 Linus Torvalds
+ *
+ * Pentium III FXSR, SSE support
+ * Gareth Hughes <gareth@valinux.com>, May 2000
+ */
+
+/*
+ * 'Traps.c' handles hardware traps and faults after we have saved some
+ * state in 'asm.s'.
+ */
+#include <linux/sched.h>
+#include <linux/kernel.h>
+#include <linux/string.h>
+#include <linux/errno.h>
+#include <linux/timer.h>
+#include <linux/mm.h>
+#include <linux/init.h>
+#include <linux/delay.h>
+#include <linux/spinlock.h>
+#include <linux/interrupt.h>
+#include <linux/highmem.h>
+#include <linux/kallsyms.h>
+#include <linux/ptrace.h>
+#include <linux/utsname.h>
+#include <linux/kprobes.h>
+#include <linux/kexec.h>
+#include <linux/unwind.h>
+
+#ifdef CONFIG_EISA
+#include <linux/ioport.h>
+#include <linux/eisa.h>
+#endif
+
+#ifdef CONFIG_MCA
+#include <linux/mca.h>
+#endif
+
+#include <asm/processor.h>
+#include <asm/system.h>
+#include <asm/uaccess.h>
+#include <asm/io.h>
+#include <asm/atomic.h>
+#include <asm/debugreg.h>
+#include <asm/desc.h>
+#include <asm/i387.h>
+#include <asm/nmi.h>
+#include <asm/unwind.h>
+#include <asm/smp.h>
+#include <asm/arch_hooks.h>
+#include <asm/kdebug.h>
+
+#include <linux/module.h>
+
+#include "mach_traps.h"
+
+asmlinkage int system_call(void);
+
+struct desc_struct default_ldt[] = { { 0, 0 }, { 0, 0 }, { 0, 0 },
+ { 0, 0 }, { 0, 0 } };
+
+/* Do we ignore FPU interrupts ? */
+char ignore_fpu_irq = 0;
+
+#ifndef CONFIG_X86_NO_IDT
+/*
+ * The IDT has to be page-aligned to simplify the Pentium
+ * F0 0F bug workaround.. We have a special link segment
+ * for this.
+ */
+struct desc_struct idt_table[256] __attribute__((__section__(".data.idt"))) = { {0, 0}, };
+#endif
+
+asmlinkage void divide_error(void);
+asmlinkage void debug(void);
+asmlinkage void nmi(void);
+asmlinkage void int3(void);
+asmlinkage void overflow(void);
+asmlinkage void bounds(void);
+asmlinkage void invalid_op(void);
+asmlinkage void device_not_available(void);
+asmlinkage void coprocessor_segment_overrun(void);
+asmlinkage void invalid_TSS(void);
+asmlinkage void segment_not_present(void);
+asmlinkage void stack_segment(void);
+asmlinkage void general_protection(void);
+asmlinkage void page_fault(void);
+asmlinkage void coprocessor_error(void);
+asmlinkage void simd_coprocessor_error(void);
+asmlinkage void alignment_check(void);
+#ifndef CONFIG_XEN
+asmlinkage void spurious_interrupt_bug(void);
+#else
+asmlinkage void fixup_4gb_segment(void);
+#endif
+asmlinkage void machine_check(void);
+
+static int kstack_depth_to_print = 24;
+#ifdef CONFIG_STACK_UNWIND
+static int call_trace = 1;
+#else
+#define call_trace (-1)
+#endif
+ATOMIC_NOTIFIER_HEAD(i386die_chain);
+
+int register_die_notifier(struct notifier_block *nb)
+{
+ vmalloc_sync_all();
+ return atomic_notifier_chain_register(&i386die_chain, nb);
+}
+EXPORT_SYMBOL(register_die_notifier); /* used modular by kdb */
+
+int unregister_die_notifier(struct notifier_block *nb)
+{
+ return atomic_notifier_chain_unregister(&i386die_chain, nb);
+}
+EXPORT_SYMBOL(unregister_die_notifier); /* used modular by kdb */
+
+static inline int valid_stack_ptr(struct thread_info *tinfo, void *p)
+{
+ return p > (void *)tinfo &&
+ p < (void *)tinfo + THREAD_SIZE - 3;
+}
+
+/*
+ * Print one address/symbol entries per line.
+ */
+static inline void print_addr_and_symbol(unsigned long addr, char *log_lvl)
+{
+ printk(" [<%08lx>] ", addr);
+
+ print_symbol("%s\n", addr);
+}
+
+static inline unsigned long print_context_stack(struct thread_info *tinfo,
+ unsigned long *stack, unsigned long ebp,
+ char *log_lvl)
+{
+ unsigned long addr;
+
+#ifdef CONFIG_FRAME_POINTER
+ while (valid_stack_ptr(tinfo, (void *)ebp)) {
+ addr = *(unsigned long *)(ebp + 4);
+ print_addr_and_symbol(addr, log_lvl);
+ /*
+ * break out of recursive entries (such as
+ * end_of_stack_stop_unwind_function):
+ */
+ if (ebp == *(unsigned long *)ebp)
+ break;
+ ebp = *(unsigned long *)ebp;
+ }
+#else
+ while (valid_stack_ptr(tinfo, stack)) {
+ addr = *stack++;
+ if (__kernel_text_address(addr))
+ print_addr_and_symbol(addr, log_lvl);
+ }
+#endif
+ return ebp;
+}
+
+static asmlinkage int
+show_trace_unwind(struct unwind_frame_info *info, void *log_lvl)
+{
+ int n = 0;
+
+ while (unwind(info) == 0 && UNW_PC(info)) {
+ n++;
+ print_addr_and_symbol(UNW_PC(info), log_lvl);
+ if (arch_unw_user_mode(info))
+ break;
+ }
+ return n;
+}
+
+static void show_trace_log_lvl(struct task_struct *task, struct pt_regs *regs,
+ unsigned long *stack, char *log_lvl)
+{
+ unsigned long ebp;
+
+ if (!task)
+ task = current;
+
+ if (call_trace >= 0) {
+ int unw_ret = 0;
+ struct unwind_frame_info info;
+
+ if (regs) {
+ if (unwind_init_frame_info(&info, task, regs) == 0)
+ unw_ret = show_trace_unwind(&info, log_lvl);
+ } else if (task == current)
+ unw_ret = unwind_init_running(&info, show_trace_unwind, log_lvl);
+ else {
+ if (unwind_init_blocked(&info, task) == 0)
+ unw_ret = show_trace_unwind(&info, log_lvl);
+ }
+ if (unw_ret > 0) {
+ if (call_trace == 1 && !arch_unw_user_mode(&info)) {
+ print_symbol("DWARF2 unwinder stuck at %s\n",
+ UNW_PC(&info));
+ if (UNW_SP(&info) >= PAGE_OFFSET) {
+ printk("Leftover inexact backtrace:\n");
+ stack = (void *)UNW_SP(&info);
+ } else
+ printk("Full inexact backtrace again:\n");
+ } else if (call_trace >= 1)
+ return;
+ else
+ printk("Full inexact backtrace again:\n");
+ } else
+ printk("Inexact backtrace:\n");
+ }
+
+ if (task == current) {
+ /* Grab ebp right from our regs */
+ asm ("movl %%ebp, %0" : "=r" (ebp) : );
+ } else {
+ /* ebp is the last reg pushed by switch_to */
+ ebp = *(unsigned long *) task->thread.esp;
+ }
+
+ while (1) {
+ struct thread_info *context;
+ context = (struct thread_info *)
+ ((unsigned long)stack & (~(THREAD_SIZE - 1)));
+ ebp = print_context_stack(context, stack, ebp, log_lvl);
+ stack = (unsigned long*)context->previous_esp;
+ if (!stack)
+ break;
+ printk("%s =======================\n", log_lvl);
+ }
+}
+
+void show_trace(struct task_struct *task, struct pt_regs *regs, unsigned long * stack)
+{
+ show_trace_log_lvl(task, regs, stack, "");
+}
+
+static void show_stack_log_lvl(struct task_struct *task, struct pt_regs *regs,
+ unsigned long *esp, char *log_lvl)
+{
+ unsigned long *stack;
+ int i;
+
+ if (esp == NULL) {
+ if (task)
+ esp = (unsigned long*)task->thread.esp;
+ else
+ esp = (unsigned long *)&esp;
+ }
+
+ stack = esp;
+ for(i = 0; i < kstack_depth_to_print; i++) {
+ if (kstack_end(stack))
+ break;
+ if (i && ((i % 8) == 0))
+ printk("\n%s ", log_lvl);
+ printk("%08lx ", *stack++);
+ }
+ printk("\n%sCall Trace:\n", log_lvl);
+ show_trace_log_lvl(task, regs, esp, log_lvl);
+}
+
+void show_stack(struct task_struct *task, unsigned long *esp)
+{
+ printk(" ");
+ show_stack_log_lvl(task, NULL, esp, "");
+}
+
+/*
+ * The architecture-independent dump_stack generator
+ */
+void dump_stack(void)
+{
+ unsigned long stack;
+
+ show_trace(current, NULL, &stack);
+}
+
+EXPORT_SYMBOL(dump_stack);
+
+void show_registers(struct pt_regs *regs)
+{
+ int i;
+ int in_kernel = 1;
+ unsigned long esp;
+ unsigned short ss;
+
+ esp = (unsigned long) (&regs->esp);
+ savesegment(ss, ss);
+ if (user_mode_vm(regs)) {
+ in_kernel = 0;
+ esp = regs->esp;
+ ss = regs->xss & 0xffff;
+ }
+ print_modules();
+ printk(KERN_EMERG "CPU: %d\nEIP: %04x:[<%08lx>] %s VLI\n"
+ "EFLAGS: %08lx (%s %.*s) \n",
+ smp_processor_id(), 0xffff & regs->xcs, regs->eip,
+ print_tainted(), regs->eflags, system_utsname.release,
+ (int)strcspn(system_utsname.version, " "),
+ system_utsname.version);
+ print_symbol(KERN_EMERG "EIP is at %s\n", regs->eip);
+ printk(KERN_EMERG "eax: %08lx ebx: %08lx ecx: %08lx edx: %08lx\n",
+ regs->eax, regs->ebx, regs->ecx, regs->edx);
+ printk(KERN_EMERG "esi: %08lx edi: %08lx ebp: %08lx esp: %08lx\n",
+ regs->esi, regs->edi, regs->ebp, esp);
+ printk(KERN_EMERG "ds: %04x es: %04x ss: %04x\n",
+ regs->xds & 0xffff, regs->xes & 0xffff, ss);
+ printk(KERN_EMERG "Process %.*s (pid: %d, ti=%p task=%p task.ti=%p)",
+ TASK_COMM_LEN, current->comm, current->pid,
+ current_thread_info(), current, current->thread_info);
+ /*
+ * When in-kernel, we also print out the stack and code at the
+ * time of the fault..
+ */
+ if (in_kernel) {
+ u8 __user *eip;
+
+ printk("\n" KERN_EMERG "Stack: ");
+ show_stack_log_lvl(NULL, regs, (unsigned long *)esp, KERN_EMERG);
+
+ printk(KERN_EMERG "Code: ");
+
+ eip = (u8 __user *)regs->eip - 43;
+ for (i = 0; i < 64; i++, eip++) {
+ unsigned char c;
+
+ if (eip < (u8 __user *)PAGE_OFFSET || __get_user(c, eip)) {
+ printk(" Bad EIP value.");
+ break;
+ }
+ if (eip == (u8 __user *)regs->eip)
+ printk("<%02x> ", c);
+ else
+ printk("%02x ", c);
+ }
+ }
+ printk("\n");
+}
+
+static void handle_BUG(struct pt_regs *regs)
+{
+ unsigned long eip = regs->eip;
+ unsigned short ud2;
+
+ if (eip < PAGE_OFFSET)
+ return;
+ if (__get_user(ud2, (unsigned short __user *)eip))
+ return;
+ if (ud2 != 0x0b0f)
+ return;
+
+ printk(KERN_EMERG "------------[ cut here ]------------\n");
+
+#ifdef CONFIG_DEBUG_BUGVERBOSE
+ do {
+ unsigned short line;
+ char *file;
+ char c;
+
+ if (__get_user(line, (unsigned short __user *)(eip + 2)))
+ break;
+ if (__get_user(file, (char * __user *)(eip + 4)) ||
+ (unsigned long)file < PAGE_OFFSET || __get_user(c, file))
+ file = "<bad filename>";
+
+ printk(KERN_EMERG "kernel BUG at %s:%d!\n", file, line);
+ return;
+ } while (0);
+#endif
+ printk(KERN_EMERG "Kernel BUG at [verbose debug info unavailable]\n");
+}
+
+/* This is gone through when something in the kernel
+ * has done something bad and is about to be terminated.
+*/
+void die(const char * str, struct pt_regs * regs, long err)
+{
+ static struct {
+ spinlock_t lock;
+ u32 lock_owner;
+ int lock_owner_depth;
+ } die = {
+ .lock = SPIN_LOCK_UNLOCKED,
+ .lock_owner = -1,
+ .lock_owner_depth = 0
+ };
+ static int die_counter;
+ unsigned long flags;
+
+ oops_enter();
+
+ if (die.lock_owner != raw_smp_processor_id()) {
+ console_verbose();
+ spin_lock_irqsave(&die.lock, flags);
+ die.lock_owner = smp_processor_id();
+ die.lock_owner_depth = 0;
+ bust_spinlocks(1);
+ }
+ else
+ local_save_flags(flags);
+
+ if (++die.lock_owner_depth < 3) {
+ int nl = 0;
+ unsigned long esp;
+ unsigned short ss;
+
+ handle_BUG(regs);
+ printk(KERN_EMERG "%s: %04lx [#%d]\n", str, err & 0xffff, ++die_counter);
+#ifdef CONFIG_PREEMPT
+ printk(KERN_EMERG "PREEMPT ");
+ nl = 1;
+#endif
+#ifdef CONFIG_SMP
+ if (!nl)
+ printk(KERN_EMERG);
+ printk("SMP ");
+ nl = 1;
+#endif
+#ifdef CONFIG_DEBUG_PAGEALLOC
+ if (!nl)
+ printk(KERN_EMERG);
+ printk("DEBUG_PAGEALLOC");
+ nl = 1;
+#endif
+ if (nl)
+ printk("\n");
+ if (notify_die(DIE_OOPS, str, regs, err,
+ current->thread.trap_no, SIGSEGV) !=
+ NOTIFY_STOP) {
+ show_registers(regs);
+ /* Executive summary in case the oops scrolled away */
+ esp = (unsigned long) (&regs->esp);
+ savesegment(ss, ss);
+ if (user_mode(regs)) {
+ esp = regs->esp;
+ ss = regs->xss & 0xffff;
+ }
+ printk(KERN_EMERG "EIP: [<%08lx>] ", regs->eip);
+ print_symbol("%s", regs->eip);
+ printk(" SS:ESP %04x:%08lx\n", ss, esp);
+ }
+ else
+ regs = NULL;
+ } else
+ printk(KERN_EMERG "Recursive die() failure, output suppressed\n");
+
+ bust_spinlocks(0);
+ die.lock_owner = -1;
+ spin_unlock_irqrestore(&die.lock, flags);
+
+ if (!regs)
+ return;
+
+ if (kexec_should_crash(current))
+ crash_kexec(regs);
+
+ if (in_interrupt())
+ panic("Fatal exception in interrupt");
+
+ if (panic_on_oops)
+ panic("Fatal exception");
+
+ oops_exit();
+ do_exit(SIGSEGV);
+}
+
+static inline void die_if_kernel(const char * str, struct pt_regs * regs, long err)
+{
+ if (!user_mode_vm(regs))
+ die(str, regs, err);
+}
+
+static void __kprobes do_trap(int trapnr, int signr, char *str, int vm86,
+ struct pt_regs * regs, long error_code,
+ siginfo_t *info)
+{
+ struct task_struct *tsk = current;
+ tsk->thread.error_code = error_code;
+ tsk->thread.trap_no = trapnr;
+
+ if (regs->eflags & VM_MASK) {
+ if (vm86)
+ goto vm86_trap;
+ goto trap_signal;
+ }
+
+ if (!user_mode(regs))
+ goto kernel_trap;
+
+ trap_signal: {
+ if (info)
+ force_sig_info(signr, info, tsk);
+ else
+ force_sig(signr, tsk);
+ return;
+ }
+
+ kernel_trap: {
+ if (!fixup_exception(regs))
+ die(str, regs, error_code);
+ return;
+ }
+
+ vm86_trap: {
+ int ret = handle_vm86_trap((struct kernel_vm86_regs *) regs, error_code, trapnr);
+ if (ret) goto trap_signal;
+ return;
+ }
+}
+
+#define DO_ERROR(trapnr, signr, str, name) \
+fastcall void do_##name(struct pt_regs * regs, long error_code) \
+{ \
+ if (notify_die(DIE_TRAP, str, regs, error_code, trapnr, signr) \
+ == NOTIFY_STOP) \
+ return; \
+ do_trap(trapnr, signr, str, 0, regs, error_code, NULL); \
+}
+
+#define DO_ERROR_INFO(trapnr, signr, str, name, sicode, siaddr) \
+fastcall void do_##name(struct pt_regs * regs, long error_code) \
+{ \
+ siginfo_t info; \
+ info.si_signo = signr; \
+ info.si_errno = 0; \
+ info.si_code = sicode; \
+ info.si_addr = (void __user *)siaddr; \
+ if (notify_die(DIE_TRAP, str, regs, error_code, trapnr, signr) \
+ == NOTIFY_STOP) \
+ return; \
+ do_trap(trapnr, signr, str, 0, regs, error_code, &info); \
+}
+
+#define DO_VM86_ERROR(trapnr, signr, str, name) \
+fastcall void do_##name(struct pt_regs * regs, long error_code) \
+{ \
+ if (notify_die(DIE_TRAP, str, regs, error_code, trapnr, signr) \
+ == NOTIFY_STOP) \
+ return; \
+ do_trap(trapnr, signr, str, 1, regs, error_code, NULL); \
+}
+
+#define DO_VM86_ERROR_INFO(trapnr, signr, str, name, sicode, siaddr) \
+fastcall void do_##name(struct pt_regs * regs, long error_code) \
+{ \
+ siginfo_t info; \
+ info.si_signo = signr; \
+ info.si_errno = 0; \
+ info.si_code = sicode; \
+ info.si_addr = (void __user *)siaddr; \
+ if (notify_die(DIE_TRAP, str, regs, error_code, trapnr, signr) \
+ == NOTIFY_STOP) \
+ return; \
+ do_trap(trapnr, signr, str, 1, regs, error_code, &info); \
+}
+
+DO_VM86_ERROR_INFO( 0, SIGFPE, "divide error", divide_error, FPE_INTDIV, regs->eip)
+#ifndef CONFIG_KPROBES
+DO_VM86_ERROR( 3, SIGTRAP, "int3", int3)
+#endif
+DO_VM86_ERROR( 4, SIGSEGV, "overflow", overflow)
+DO_VM86_ERROR( 5, SIGSEGV, "bounds", bounds)
+DO_ERROR_INFO( 6, SIGILL, "invalid opcode", invalid_op, ILL_ILLOPN, regs->eip)
+DO_ERROR( 9, SIGFPE, "coprocessor segment overrun", coprocessor_segment_overrun)
+DO_ERROR(10, SIGSEGV, "invalid TSS", invalid_TSS)
+DO_ERROR(11, SIGBUS, "segment not present", segment_not_present)
+DO_ERROR(12, SIGBUS, "stack segment", stack_segment)
+DO_ERROR_INFO(17, SIGBUS, "alignment check", alignment_check, BUS_ADRALN, 0)
+DO_ERROR_INFO(32, SIGSEGV, "iret exception", iret_error, ILL_BADSTK, 0)
+
+fastcall void __kprobes do_general_protection(struct pt_regs * regs,
+ long error_code)
+{
+ current->thread.error_code = error_code;
+ current->thread.trap_no = 13;
+
+ if (regs->eflags & VM_MASK)
+ goto gp_in_vm86;
+
+ if (!user_mode(regs))
+ goto gp_in_kernel;
+
+ current->thread.error_code = error_code;
+ current->thread.trap_no = 13;
+ force_sig(SIGSEGV, current);
+ return;
+
+gp_in_vm86:
+ local_irq_enable();
+ handle_vm86_fault((struct kernel_vm86_regs *) regs, error_code);
+ return;
+
+gp_in_kernel:
+ if (!fixup_exception(regs)) {
+ if (notify_die(DIE_GPF, "general protection fault", regs,
+ error_code, 13, SIGSEGV) == NOTIFY_STOP)
+ return;
+ die("general protection fault", regs, error_code);
+ }
+}
+
+static void mem_parity_error(unsigned char reason, struct pt_regs * regs)
+{
+ printk(KERN_EMERG "Uhhuh. NMI received. Dazed and confused, but trying "
+ "to continue\n");
+ printk(KERN_EMERG "You probably have a hardware problem with your RAM "
+ "chips\n");
+
+ /* Clear and disable the memory parity error line. */
+ clear_mem_error(reason);
+}
+
+static void io_check_error(unsigned char reason, struct pt_regs * regs)
+{
+ printk(KERN_EMERG "NMI: IOCK error (debug interrupt?)\n");
+ show_registers(regs);
+
+ /* Re-enable the IOCK line, wait for a few seconds */
+ clear_io_check_error(reason);
+}
+
+static void unknown_nmi_error(unsigned char reason, struct pt_regs * regs)
+{
+#ifdef CONFIG_MCA
+ /* Might actually be able to figure out what the guilty party
+ * is. */
+ if( MCA_bus ) {
+ mca_handle_nmi();
+ return;
+ }
+#endif
+ printk("Uhhuh. NMI received for unknown reason %02x on CPU %d.\n",
+ reason, smp_processor_id());
+ printk("Dazed and confused, but trying to continue\n");
+ printk("Do you have a strange power saving mode enabled?\n");
+}
+
+static DEFINE_SPINLOCK(nmi_print_lock);
+
+void die_nmi (struct pt_regs *regs, const char *msg)
+{
+ if (notify_die(DIE_NMIWATCHDOG, msg, regs, 0, 2, SIGINT) ==
+ NOTIFY_STOP)
+ return;
+
+ spin_lock(&nmi_print_lock);
+ /*
+ * We are in trouble anyway, lets at least try
+ * to get a message out.
+ */
+ bust_spinlocks(1);
+ printk(KERN_EMERG "%s", msg);
+ printk(" on CPU%d, eip %08lx, registers:\n",
+ smp_processor_id(), regs->eip);
+ show_registers(regs);
+ printk(KERN_EMERG "console shuts up ...\n");
+ console_silent();
+ spin_unlock(&nmi_print_lock);
+ bust_spinlocks(0);
+
+ /* If we are in kernel we are probably nested up pretty bad
+ * and might aswell get out now while we still can.
+ */
+ if (!user_mode_vm(regs)) {
+ current->thread.trap_no = 2;
+ crash_kexec(regs);
+ }
+
+ do_exit(SIGSEGV);
+}
+
+static void default_do_nmi(struct pt_regs * regs)
+{
+ unsigned char reason = 0;
+
+ /* Only the BSP gets external NMIs from the system. */
+ if (!smp_processor_id())
+ reason = get_nmi_reason();
+
+ if (!(reason & 0xc0)) {
+ if (notify_die(DIE_NMI_IPI, "nmi_ipi", regs, reason, 2, SIGINT)
+ == NOTIFY_STOP)
+ return;
+#ifdef CONFIG_X86_LOCAL_APIC
+ /*
+ * Ok, so this is none of the documented NMI sources,
+ * so it must be the NMI watchdog.
+ */
+ if (nmi_watchdog) {
+ nmi_watchdog_tick(regs);
+ return;
+ }
+#endif
+ unknown_nmi_error(reason, regs);
+ return;
+ }
+ if (notify_die(DIE_NMI, "nmi", regs, reason, 2, SIGINT) == NOTIFY_STOP)
+ return;
+ if (reason & 0x80)
+ mem_parity_error(reason, regs);
+ if (reason & 0x40)
+ io_check_error(reason, regs);
+ /*
+ * Reassert NMI in case it became active meanwhile
+ * as it's edge-triggered.
+ */
+ reassert_nmi();
+}
+
+static int dummy_nmi_callback(struct pt_regs * regs, int cpu)
+{
+ return 0;
+}
+
+static nmi_callback_t nmi_callback = dummy_nmi_callback;
+
+fastcall void do_nmi(struct pt_regs * regs, long error_code)
+{
+ int cpu;
+
+ nmi_enter();
+
+ cpu = smp_processor_id();
+
+ ++nmi_count(cpu);
+
+ if (!rcu_dereference(nmi_callback)(regs, cpu))
+ default_do_nmi(regs);
+
+ nmi_exit();
+}
+
+void set_nmi_callback(nmi_callback_t callback)
+{
+ vmalloc_sync_all();
+ rcu_assign_pointer(nmi_callback, callback);
+}
+EXPORT_SYMBOL_GPL(set_nmi_callback);
+
+void unset_nmi_callback(void)
+{
+ nmi_callback = dummy_nmi_callback;
+}
+EXPORT_SYMBOL_GPL(unset_nmi_callback);
+
+#ifdef CONFIG_KPROBES
+fastcall void __kprobes do_int3(struct pt_regs *regs, long error_code)
+{
+ if (notify_die(DIE_INT3, "int3", regs, error_code, 3, SIGTRAP)
+ == NOTIFY_STOP)
+ return;
+ /* This is an interrupt gate, because kprobes wants interrupts
+ disabled. Normal trap handlers don't. */
+ restore_interrupts(regs);
+ do_trap(3, SIGTRAP, "int3", 1, regs, error_code, NULL);
+}
+#endif
+
+/*
+ * Our handling of the processor debug registers is non-trivial.
+ * We do not clear them on entry and exit from the kernel. Therefore
+ * it is possible to get a watchpoint trap here from inside the kernel.
+ * However, the code in ./ptrace.c has ensured that the user can
+ * only set watchpoints on userspace addresses. Therefore the in-kernel
+ * watchpoint trap can only occur in code which is reading/writing
+ * from user space. Such code must not hold kernel locks (since it
+ * can equally take a page fault), therefore it is safe to call
+ * force_sig_info even though that claims and releases locks.
+ *
+ * Code in ./signal.c ensures that the debug control register
+ * is restored before we deliver any signal, and therefore that
+ * user code runs with the correct debug control register even though
+ * we clear it here.
+ *
+ * Being careful here means that we don't have to be as careful in a
+ * lot of more complicated places (task switching can be a bit lazy
+ * about restoring all the debug state, and ptrace doesn't have to
+ * find every occurrence of the TF bit that could be saved away even
+ * by user code)
+ */
+fastcall void __kprobes do_debug(struct pt_regs * regs, long error_code)
+{
+ unsigned int condition;
+ struct task_struct *tsk = current;
+
+ get_debugreg(condition, 6);
+
+ if (notify_die(DIE_DEBUG, "debug", regs, condition, error_code,
+ SIGTRAP) == NOTIFY_STOP)
+ return;
+ /* It's safe to allow irq's after DR6 has been saved */
+ if (regs->eflags & X86_EFLAGS_IF)
+ local_irq_enable();
+
+ /* Mask out spurious debug traps due to lazy DR7 setting */
+ if (condition & (DR_TRAP0|DR_TRAP1|DR_TRAP2|DR_TRAP3)) {
+ if (!tsk->thread.debugreg[7])
+ goto clear_dr7;
+ }
+
+ if (regs->eflags & VM_MASK)
+ goto debug_vm86;
+
+ /* Save debug status register where ptrace can see it */
+ tsk->thread.debugreg[6] = condition;
+
+ /*
+ * Single-stepping through TF: make sure we ignore any events in
+ * kernel space (but re-enable TF when returning to user mode).
+ */
+ if (condition & DR_STEP) {
+ /*
+ * We already checked v86 mode above, so we can
+ * check for kernel mode by just checking the CPL
+ * of CS.
+ */
+ if (!user_mode(regs))
+ goto clear_TF_reenable;
+ }
+
+ /* Ok, finally something we can handle */
+ send_sigtrap(tsk, regs, error_code);
+
+ /* Disable additional traps. They'll be re-enabled when
+ * the signal is delivered.
+ */
+clear_dr7:
+ set_debugreg(0, 7);
+ return;
+
+debug_vm86:
+ handle_vm86_trap((struct kernel_vm86_regs *) regs, error_code, 1);
+ return;
+
+clear_TF_reenable:
+ set_tsk_thread_flag(tsk, TIF_SINGLESTEP);
+ regs->eflags &= ~TF_MASK;
+ return;
+}
+
+/*
+ * Note that we play around with the 'TS' bit in an attempt to get
+ * the correct behaviour even in the presence of the asynchronous
+ * IRQ13 behaviour
+ */
+void math_error(void __user *eip)
+{
+ struct task_struct * task;
+ siginfo_t info;
+ unsigned short cwd, swd;
+
+ /*
+ * Save the info for the exception handler and clear the error.
+ */
+ task = current;
+ save_init_fpu(task);
+ task->thread.trap_no = 16;
+ task->thread.error_code = 0;
+ info.si_signo = SIGFPE;
+ info.si_errno = 0;
+ info.si_code = __SI_FAULT;
+ info.si_addr = eip;
+ /*
+ * (~cwd & swd) will mask out exceptions that are not set to unmasked
+ * status. 0x3f is the exception bits in these regs, 0x200 is the
+ * C1 reg you need in case of a stack fault, 0x040 is the stack
+ * fault bit. We should only be taking one exception at a time,
+ * so if this combination doesn't produce any single exception,
+ * then we have a bad program that isn't syncronizing its FPU usage
+ * and it will suffer the consequences since we won't be able to
+ * fully reproduce the context of the exception
+ */
+ cwd = get_fpu_cwd(task);
+ swd = get_fpu_swd(task);
+ switch (swd & ~cwd & 0x3f) {
+ case 0x000: /* No unmasked exception */
+ return;
+ default: /* Multiple exceptions */
+ break;
+ case 0x001: /* Invalid Op */
+ /*
+ * swd & 0x240 == 0x040: Stack Underflow
+ * swd & 0x240 == 0x240: Stack Overflow
+ * User must clear the SF bit (0x40) if set
+ */
+ info.si_code = FPE_FLTINV;
+ break;
+ case 0x002: /* Denormalize */
+ case 0x010: /* Underflow */
+ info.si_code = FPE_FLTUND;
+ break;
+ case 0x004: /* Zero Divide */
+ info.si_code = FPE_FLTDIV;
+ break;
+ case 0x008: /* Overflow */
+ info.si_code = FPE_FLTOVF;
+ break;
+ case 0x020: /* Precision */
+ info.si_code = FPE_FLTRES;
+ break;
+ }
+ force_sig_info(SIGFPE, &info, task);
+}
+
+fastcall void do_coprocessor_error(struct pt_regs * regs, long error_code)
+{
+ ignore_fpu_irq = 1;
+ math_error((void __user *)regs->eip);
+}
+
+static void simd_math_error(void __user *eip)
+{
+ struct task_struct * task;
+ siginfo_t info;
+ unsigned short mxcsr;
+
+ /*
+ * Save the info for the exception handler and clear the error.
+ */
+ task = current;
+ save_init_fpu(task);
+ task->thread.trap_no = 19;
+ task->thread.error_code = 0;
+ info.si_signo = SIGFPE;
+ info.si_errno = 0;
+ info.si_code = __SI_FAULT;
+ info.si_addr = eip;
+ /*
+ * The SIMD FPU exceptions are handled a little differently, as there
+ * is only a single status/control register. Thus, to determine which
+ * unmasked exception was caught we must mask the exception mask bits
+ * at 0x1f80, and then use these to mask the exception bits at 0x3f.
+ */
+ mxcsr = get_fpu_mxcsr(task);
+ switch (~((mxcsr & 0x1f80) >> 7) & (mxcsr & 0x3f)) {
+ case 0x000:
+ default:
+ break;
+ case 0x001: /* Invalid Op */
+ info.si_code = FPE_FLTINV;
+ break;
+ case 0x002: /* Denormalize */
+ case 0x010: /* Underflow */
+ info.si_code = FPE_FLTUND;
+ break;
+ case 0x004: /* Zero Divide */
+ info.si_code = FPE_FLTDIV;
+ break;
+ case 0x008: /* Overflow */
+ info.si_code = FPE_FLTOVF;
+ break;
+ case 0x020: /* Precision */
+ info.si_code = FPE_FLTRES;
+ break;
+ }
+ force_sig_info(SIGFPE, &info, task);
+}
+
+fastcall void do_simd_coprocessor_error(struct pt_regs * regs,
+ long error_code)
+{
+ if (cpu_has_xmm) {
+ /* Handle SIMD FPU exceptions on PIII+ processors. */
+ ignore_fpu_irq = 1;
+ simd_math_error((void __user *)regs->eip);
+ } else {
+ /*
+ * Handle strange cache flush from user space exception
+ * in all other cases. This is undocumented behaviour.
+ */
+ if (regs->eflags & VM_MASK) {
+ handle_vm86_fault((struct kernel_vm86_regs *)regs,
+ error_code);
+ return;
+ }
+ current->thread.trap_no = 19;
+ current->thread.error_code = error_code;
+ die_if_kernel("cache flush denied", regs, error_code);
+ force_sig(SIGSEGV, current);
+ }
+}
+
+#ifndef CONFIG_XEN
+fastcall void do_spurious_interrupt_bug(struct pt_regs * regs,
+ long error_code)
+{
+#if 0
+ /* No need to warn about this any longer. */
+ printk("Ignoring P6 Local APIC Spurious Interrupt Bug...\n");
+#endif
+}
+
+fastcall void setup_x86_bogus_stack(unsigned char * stk)
+{
+ unsigned long *switch16_ptr, *switch32_ptr;
+ struct pt_regs *regs;
+ unsigned long stack_top, stack_bot;
+ unsigned short iret_frame16_off;
+ int cpu = smp_processor_id();
+ /* reserve the space on 32bit stack for the magic switch16 pointer */
+ memmove(stk, stk + 8, sizeof(struct pt_regs));
+ switch16_ptr = (unsigned long *)(stk + sizeof(struct pt_regs));
+ regs = (struct pt_regs *)stk;
+ /* now the switch32 on 16bit stack */
+ stack_bot = (unsigned long)&per_cpu(cpu_16bit_stack, cpu);
+ stack_top = stack_bot + CPU_16BIT_STACK_SIZE;
+ switch32_ptr = (unsigned long *)(stack_top - 8);
+ iret_frame16_off = CPU_16BIT_STACK_SIZE - 8 - 20;
+ /* copy iret frame on 16bit stack */
+ memcpy((void *)(stack_bot + iret_frame16_off), &regs->eip, 20);
+ /* fill in the switch pointers */
+ switch16_ptr[0] = (regs->esp & 0xffff0000) | iret_frame16_off;
+ switch16_ptr[1] = __ESPFIX_SS;
+ switch32_ptr[0] = (unsigned long)stk + sizeof(struct pt_regs) +
+ 8 - CPU_16BIT_STACK_SIZE;
+ switch32_ptr[1] = __KERNEL_DS;
+}
+
+fastcall unsigned char * fixup_x86_bogus_stack(unsigned short sp)
+{
+ unsigned long *switch32_ptr;
+ unsigned char *stack16, *stack32;
+ unsigned long stack_top, stack_bot;
+ int len;
+ int cpu = smp_processor_id();
+ stack_bot = (unsigned long)&per_cpu(cpu_16bit_stack, cpu);
+ stack_top = stack_bot + CPU_16BIT_STACK_SIZE;
+ switch32_ptr = (unsigned long *)(stack_top - 8);
+ /* copy the data from 16bit stack to 32bit stack */
+ len = CPU_16BIT_STACK_SIZE - 8 - sp;
+ stack16 = (unsigned char *)(stack_bot + sp);
+ stack32 = (unsigned char *)
+ (switch32_ptr[0] + CPU_16BIT_STACK_SIZE - 8 - len);
+ memcpy(stack32, stack16, len);
+ return stack32;
+}
+#endif
+
+/*
+ * 'math_state_restore()' saves the current math information in the
+ * old math state array, and gets the new ones from the current task
+ *
+ * Careful.. There are problems with IBM-designed IRQ13 behaviour.
+ * Don't touch unless you *really* know how it works.
+ *
+ * Must be called with kernel preemption disabled (in this case,
+ * local interrupts are disabled at the call-site in entry.S).
+ */
+asmlinkage void math_state_restore(struct pt_regs regs)
+{
+ struct thread_info *thread = current_thread_info();
+ struct task_struct *tsk = thread->task;
+
+ /* NB. 'clts' is done for us by Xen during virtual trap. */
+ if (!tsk_used_math(tsk))
+ init_fpu(tsk);
+ restore_fpu(tsk);
+ thread->status |= TS_USEDFPU; /* So we fnsave on switch_to() */
+}
+
+#ifndef CONFIG_MATH_EMULATION
+
+asmlinkage void math_emulate(long arg)
+{
+ printk(KERN_EMERG "math-emulation not enabled and no coprocessor found.\n");
+ printk(KERN_EMERG "killing %s.\n",current->comm);
+ force_sig(SIGFPE,current);
+ schedule();
+}
+
+#endif /* CONFIG_MATH_EMULATION */
+
+#ifdef CONFIG_X86_F00F_BUG
+void __init trap_init_f00f_bug(void)
+{
+ __set_fixmap(FIX_F00F_IDT, __pa(&idt_table), PAGE_KERNEL_RO);
+
+ /*
+ * Update the IDT descriptor and reload the IDT so that
+ * it uses the read-only mapped virtual address.
+ */
+ idt_descr.address = fix_to_virt(FIX_F00F_IDT);
+ load_idt(&idt_descr);
+}
+#endif
+
+
+/*
+ * NB. All these are "trap gates" (i.e. events_mask isn't set) except
+ * for those that specify <dpl>|4 in the second field.
+ */
+static trap_info_t __cpuinitdata trap_table[] = {
+ { 0, 0, __KERNEL_CS, (unsigned long)divide_error },
+ { 1, 0|4, __KERNEL_CS, (unsigned long)debug },
+ { 3, 3|4, __KERNEL_CS, (unsigned long)int3 },
+ { 4, 3, __KERNEL_CS, (unsigned long)overflow },
+ { 5, 0, __KERNEL_CS, (unsigned long)bounds },
+ { 6, 0, __KERNEL_CS, (unsigned long)invalid_op },
+ { 7, 0|4, __KERNEL_CS, (unsigned long)device_not_available },
+ { 9, 0, __KERNEL_CS, (unsigned long)coprocessor_segment_overrun },
+ { 10, 0, __KERNEL_CS, (unsigned long)invalid_TSS },
+ { 11, 0, __KERNEL_CS, (unsigned long)segment_not_present },
+ { 12, 0, __KERNEL_CS, (unsigned long)stack_segment },
+ { 13, 0, __KERNEL_CS, (unsigned long)general_protection },
+ { 14, 0|4, __KERNEL_CS, (unsigned long)page_fault },
+ { 15, 0, __KERNEL_CS, (unsigned long)fixup_4gb_segment },
+ { 16, 0, __KERNEL_CS, (unsigned long)coprocessor_error },
+ { 17, 0, __KERNEL_CS, (unsigned long)alignment_check },
+#ifdef CONFIG_X86_MCE
+ { 18, 0, __KERNEL_CS, (unsigned long)machine_check },
+#endif
+ { 19, 0, __KERNEL_CS, (unsigned long)simd_coprocessor_error },
+ { SYSCALL_VECTOR, 3, __KERNEL_CS, (unsigned long)system_call },
+ { 0, 0, 0, 0 }
+};
+
+void __init trap_init(void)
+{
+ int ret;
+
+ ret = HYPERVISOR_set_trap_table(trap_table);
+ if (ret)
+ printk("HYPERVISOR_set_trap_table failed: error %d\n", ret);
+
+ if (cpu_has_fxsr) {
+ /*
+ * Verify that the FXSAVE/FXRSTOR data will be 16-byte aligned.
+ * Generates a compile-time "error: zero width for bit-field" if
+ * the alignment is wrong.
+ */
+ struct fxsrAlignAssert {
+ int _:!(offsetof(struct task_struct,
+ thread.i387.fxsave) & 15);
+ };
+
+ printk(KERN_INFO "Enabling fast FPU save and restore... ");
+ set_in_cr4(X86_CR4_OSFXSR);
+ printk("done.\n");
+ }
+ if (cpu_has_xmm) {
+ printk(KERN_INFO "Enabling unmasked SIMD FPU exception "
+ "support... ");
+ set_in_cr4(X86_CR4_OSXMMEXCPT);
+ printk("done.\n");
+ }
+
+ /*
+ * Should be a barrier for any external CPU state.
+ */
+ cpu_init();
+}
+
+void __cpuinit smp_trap_init(trap_info_t *trap_ctxt)
+{
+ const trap_info_t *t = trap_table;
+
+ for (t = trap_table; t->address; t++) {
+ trap_ctxt[t->vector].flags = t->flags;
+ trap_ctxt[t->vector].cs = t->cs;
+ trap_ctxt[t->vector].address = t->address;
+ }
+}
+
+static int __init kstack_setup(char *s)
+{
+ kstack_depth_to_print = simple_strtoul(s, NULL, 0);
+ return 1;
+}
+__setup("kstack=", kstack_setup);
+
+#ifdef CONFIG_STACK_UNWIND
+static int __init call_trace_setup(char *s)
+{
+ if (strcmp(s, "old") == 0)
+ call_trace = -1;
+ else if (strcmp(s, "both") == 0)
+ call_trace = 0;
+ else if (strcmp(s, "newfallback") == 0)
+ call_trace = 1;
+ else if (strcmp(s, "new") == 2)
+ call_trace = 2;
+ return 1;
+}
+__setup("call_trace=", call_trace_setup);
+#endif
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/mach-xen/Makefile 2007-06-12 13:12:48.000000000 +0200
@@ -0,0 +1,5 @@
+#
+# Makefile for the linux kernel.
+#
+
+obj-y := setup.o
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/mach-xen/setup.c 2008-04-02 12:34:02.000000000 +0200
@@ -0,0 +1,158 @@
+/*
+ * Machine specific setup for generic
+ */
+
+#include <linux/mm.h>
+#include <linux/smp.h>
+#include <linux/init.h>
+#include <linux/interrupt.h>
+#include <linux/module.h>
+#include <asm/acpi.h>
+#include <asm/arch_hooks.h>
+#include <asm/e820.h>
+#include <asm/setup.h>
+#include <asm/fixmap.h>
+
+#include <xen/interface/callback.h>
+#include <xen/interface/memory.h>
+
+#ifdef CONFIG_HOTPLUG_CPU
+#define DEFAULT_SEND_IPI (1)
+#else
+#define DEFAULT_SEND_IPI (0)
+#endif
+
+int no_broadcast=DEFAULT_SEND_IPI;
+
+static __init int no_ipi_broadcast(char *str)
+{
+ get_option(&str, &no_broadcast);
+ printk ("Using %s mode\n", no_broadcast ? "No IPI Broadcast" :
+ "IPI Broadcast");
+ return 1;
+}
+
+__setup("no_ipi_broadcast", no_ipi_broadcast);
+
+static int __init print_ipi_mode(void)
+{
+ printk ("Using IPI %s mode\n", no_broadcast ? "No-Shortcut" :
+ "Shortcut");
+ return 0;
+}
+
+late_initcall(print_ipi_mode);
+
+/**
+ * machine_specific_memory_setup - Hook for machine specific memory setup.
+ *
+ * Description:
+ * This is included late in kernel/setup.c so that it can make
+ * use of all of the static functions.
+ **/
+
+char * __init machine_specific_memory_setup(void)
+{
+ int rc;
+ struct xen_memory_map memmap;
+ /*
+ * This is rather large for a stack variable but this early in
+ * the boot process we know we have plenty slack space.
+ */
+ struct e820entry map[E820MAX];
+
+ memmap.nr_entries = E820MAX;
+ set_xen_guest_handle(memmap.buffer, map);
+
+ rc = HYPERVISOR_memory_op(XENMEM_memory_map, &memmap);
+ if ( rc == -ENOSYS ) {
+ memmap.nr_entries = 1;
+ map[0].addr = 0ULL;
+ map[0].size = PFN_PHYS((unsigned long long)xen_start_info->nr_pages);
+ /* 8MB slack (to balance backend allocations). */
+ map[0].size += 8ULL << 20;
+ map[0].type = E820_RAM;
+ rc = 0;
+ }
+ BUG_ON(rc);
+
+ sanitize_e820_map(map, (char *)&memmap.nr_entries);
+
+ BUG_ON(copy_e820_map(map, (char)memmap.nr_entries) < 0);
+
+ return "Xen";
+}
+
+
+extern void hypervisor_callback(void);
+extern void failsafe_callback(void);
+extern void nmi(void);
+
+unsigned long *machine_to_phys_mapping = (void *)MACH2PHYS_VIRT_START;
+EXPORT_SYMBOL(machine_to_phys_mapping);
+unsigned int machine_to_phys_order;
+EXPORT_SYMBOL(machine_to_phys_order);
+
+void __init pre_setup_arch_hook(void)
+{
+ struct xen_machphys_mapping mapping;
+ unsigned long machine_to_phys_nr_ents;
+ struct xen_platform_parameters pp;
+
+ init_mm.pgd = swapper_pg_dir = (pgd_t *)xen_start_info->pt_base;
+
+ setup_xen_features();
+
+ if (HYPERVISOR_xen_version(XENVER_platform_parameters, &pp) == 0)
+ set_fixaddr_top(pp.virt_start);
+
+ if (HYPERVISOR_memory_op(XENMEM_machphys_mapping, &mapping) == 0) {
+ machine_to_phys_mapping = (unsigned long *)mapping.v_start;
+ machine_to_phys_nr_ents = mapping.max_mfn + 1;
+ } else
+ machine_to_phys_nr_ents = MACH2PHYS_NR_ENTRIES;
+ machine_to_phys_order = fls(machine_to_phys_nr_ents - 1);
+
+ if (!xen_feature(XENFEAT_auto_translated_physmap))
+ phys_to_machine_mapping =
+ (unsigned long *)xen_start_info->mfn_list;
+}
+
+void __init machine_specific_arch_setup(void)
+{
+ int ret;
+ static struct callback_register __initdata event = {
+ .type = CALLBACKTYPE_event,
+ .address = { __KERNEL_CS, (unsigned long)hypervisor_callback },
+ };
+ static struct callback_register __initdata failsafe = {
+ .type = CALLBACKTYPE_failsafe,
+ .address = { __KERNEL_CS, (unsigned long)failsafe_callback },
+ };
+ static struct callback_register __initdata nmi_cb = {
+ .type = CALLBACKTYPE_nmi,
+ .address = { __KERNEL_CS, (unsigned long)nmi },
+ };
+
+ ret = HYPERVISOR_callback_op(CALLBACKOP_register, &event);
+ if (ret == 0)
+ ret = HYPERVISOR_callback_op(CALLBACKOP_register, &failsafe);
+#if CONFIG_XEN_COMPAT <= 0x030002
+ if (ret == -ENOSYS)
+ ret = HYPERVISOR_set_callbacks(
+ event.address.cs, event.address.eip,
+ failsafe.address.cs, failsafe.address.eip);
+#endif
+ BUG_ON(ret);
+
+ ret = HYPERVISOR_callback_op(CALLBACKOP_register, &nmi_cb);
+#if CONFIG_XEN_COMPAT <= 0x030002
+ if (ret == -ENOSYS) {
+ static struct xennmi_callback __initdata cb = {
+ .handler_address = (unsigned long)nmi
+ };
+
+ HYPERVISOR_nmi_op(XENNMI_register_callback, &cb);
+ }
+#endif
+}
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/lib/scrub.c 2008-02-08 12:30:51.000000000 +0100
@@ -0,0 +1,21 @@
+#include <asm/cpufeature.h>
+#include <asm/page.h>
+#include <asm/processor.h>
+
+void scrub_pages(void *v, unsigned int count)
+{
+ if (likely(cpu_has_xmm2)) {
+ unsigned long n = count * (PAGE_SIZE / sizeof(long) / 4);
+
+ for (; n--; v += sizeof(long) * 4)
+ asm("movnti %1,(%0)\n\t"
+ "movnti %1,%c2(%0)\n\t"
+ "movnti %1,2*%c2(%0)\n\t"
+ "movnti %1,3*%c2(%0)\n\t"
+ : : "r" (v), "r" (0L), "i" (sizeof(long))
+ : "memory");
+ asm volatile("sfence" : : : "memory");
+ } else
+ for (; count--; v += PAGE_SIZE)
+ clear_page(v);
+}
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/mm/fault_32-xen.c 2010-09-23 15:39:04.000000000 +0200
@@ -0,0 +1,792 @@
+/*
+ * linux/arch/i386/mm/fault.c
+ *
+ * Copyright (C) 1995 Linus Torvalds
+ */
+
+#include <linux/signal.h>
+#include <linux/sched.h>
+#include <linux/kernel.h>
+#include <linux/errno.h>
+#include <linux/string.h>
+#include <linux/types.h>
+#include <linux/ptrace.h>
+#include <linux/mman.h>
+#include <linux/mm.h>
+#include <linux/smp.h>
+#include <linux/smp_lock.h>
+#include <linux/interrupt.h>
+#include <linux/init.h>
+#include <linux/tty.h>
+#include <linux/vt_kern.h> /* For unblank_screen() */
+#include <linux/highmem.h>
+#include <linux/module.h>
+#include <linux/kprobes.h>
+
+#include <asm/system.h>
+#include <asm/uaccess.h>
+#include <asm/desc.h>
+#include <asm/kdebug.h>
+
+extern void die(const char *,struct pt_regs *,long);
+
+#ifdef CONFIG_KPROBES
+ATOMIC_NOTIFIER_HEAD(notify_page_fault_chain);
+int register_page_fault_notifier(struct notifier_block *nb)
+{
+ vmalloc_sync_all();
+ return atomic_notifier_chain_register(&notify_page_fault_chain, nb);
+}
+
+int unregister_page_fault_notifier(struct notifier_block *nb)
+{
+ return atomic_notifier_chain_unregister(&notify_page_fault_chain, nb);
+}
+
+static inline int notify_page_fault(enum die_val val, const char *str,
+ struct pt_regs *regs, long err, int trap, int sig)
+{
+ struct die_args args = {
+ .regs = regs,
+ .str = str,
+ .err = err,
+ .trapnr = trap,
+ .signr = sig
+ };
+ return atomic_notifier_call_chain(&notify_page_fault_chain, val, &args);
+}
+#else
+static inline int notify_page_fault(enum die_val val, const char *str,
+ struct pt_regs *regs, long err, int trap, int sig)
+{
+ return NOTIFY_DONE;
+}
+#endif
+
+
+/*
+ * Unlock any spinlocks which will prevent us from getting the
+ * message out
+ */
+void bust_spinlocks(int yes)
+{
+ int loglevel_save = console_loglevel;
+
+ if (yes) {
+ oops_in_progress = 1;
+ return;
+ }
+#ifdef CONFIG_VT
+ unblank_screen();
+#endif
+ oops_in_progress = 0;
+ /*
+ * OK, the message is on the console. Now we call printk()
+ * without oops_in_progress set so that printk will give klogd
+ * a poke. Hold onto your hats...
+ */
+ console_loglevel = 15; /* NMI oopser may have shut the console up */
+ printk(" ");
+ console_loglevel = loglevel_save;
+}
+
+/*
+ * Return EIP plus the CS segment base. The segment limit is also
+ * adjusted, clamped to the kernel/user address space (whichever is
+ * appropriate), and returned in *eip_limit.
+ *
+ * The segment is checked, because it might have been changed by another
+ * task between the original faulting instruction and here.
+ *
+ * If CS is no longer a valid code segment, or if EIP is beyond the
+ * limit, or if it is a kernel address when CS is not a kernel segment,
+ * then the returned value will be greater than *eip_limit.
+ *
+ * This is slow, but is very rarely executed.
+ */
+static inline unsigned long get_segment_eip(struct pt_regs *regs,
+ unsigned long *eip_limit)
+{
+ unsigned long eip = regs->eip;
+ unsigned seg = regs->xcs & 0xffff;
+ u32 seg_ar, seg_limit, base, *desc;
+
+ /* Unlikely, but must come before segment checks. */
+ if (unlikely(regs->eflags & VM_MASK)) {
+ base = seg << 4;
+ *eip_limit = base + 0xffff;
+ return base + (eip & 0xffff);
+ }
+
+ /* The standard kernel/user address space limit. */
+ *eip_limit = (seg & 2) ? USER_DS.seg : KERNEL_DS.seg;
+
+ /* By far the most common cases. */
+ if (likely(seg == __USER_CS || seg == GET_KERNEL_CS()))
+ return eip;
+
+ /* Check the segment exists, is within the current LDT/GDT size,
+ that kernel/user (ring 0..3) has the appropriate privilege,
+ that it's a code segment, and get the limit. */
+ __asm__ ("larl %3,%0; lsll %3,%1"
+ : "=&r" (seg_ar), "=r" (seg_limit) : "0" (0), "rm" (seg));
+ if ((~seg_ar & 0x9800) || eip > seg_limit) {
+ *eip_limit = 0;
+ return 1; /* So that returned eip > *eip_limit. */
+ }
+
+ /* Get the GDT/LDT descriptor base.
+ When you look for races in this code remember that
+ LDT and other horrors are only used in user space. */
+ if (seg & (1<<2)) {
+ /* Must lock the LDT while reading it. */
+ down(&current->mm->context.sem);
+ desc = current->mm->context.ldt;
+ desc = (void *)desc + (seg & ~7);
+ } else {
+ /* Must disable preemption while reading the GDT. */
+ desc = (u32 *)get_cpu_gdt_table(get_cpu());
+ desc = (void *)desc + (seg & ~7);
+ }
+
+ /* Decode the code segment base from the descriptor */
+ base = get_desc_base((unsigned long *)desc);
+
+ if (seg & (1<<2)) {
+ up(&current->mm->context.sem);
+ } else
+ put_cpu();
+
+ /* Adjust EIP and segment limit, and clamp at the kernel limit.
+ It's legitimate for segments to wrap at 0xffffffff. */
+ seg_limit += base;
+ if (seg_limit < *eip_limit && seg_limit >= base)
+ *eip_limit = seg_limit;
+ return eip + base;
+}
+
+/*
+ * Sometimes AMD Athlon/Opteron CPUs report invalid exceptions on prefetch.
+ * Check that here and ignore it.
+ */
+static int __is_prefetch(struct pt_regs *regs, unsigned long addr)
+{
+ unsigned long limit;
+ unsigned long instr = get_segment_eip (regs, &limit);
+ int scan_more = 1;
+ int prefetch = 0;
+ int i;
+
+ for (i = 0; scan_more && i < 15; i++) {
+ unsigned char opcode;
+ unsigned char instr_hi;
+ unsigned char instr_lo;
+
+ if (instr > limit)
+ break;
+ if (__get_user(opcode, (unsigned char __user *) instr))
+ break;
+
+ instr_hi = opcode & 0xf0;
+ instr_lo = opcode & 0x0f;
+ instr++;
+
+ switch (instr_hi) {
+ case 0x20:
+ case 0x30:
+ /* Values 0x26,0x2E,0x36,0x3E are valid x86 prefixes. */
+ scan_more = ((instr_lo & 7) == 0x6);
+ break;
+
+ case 0x60:
+ /* 0x64 thru 0x67 are valid prefixes in all modes. */
+ scan_more = (instr_lo & 0xC) == 0x4;
+ break;
+ case 0xF0:
+ /* 0xF0, 0xF2, and 0xF3 are valid prefixes */
+ scan_more = !instr_lo || (instr_lo>>1) == 1;
+ break;
+ case 0x00:
+ /* Prefetch instruction is 0x0F0D or 0x0F18 */
+ scan_more = 0;
+ if (instr > limit)
+ break;
+ if (__get_user(opcode, (unsigned char __user *) instr))
+ break;
+ prefetch = (instr_lo == 0xF) &&
+ (opcode == 0x0D || opcode == 0x18);
+ break;
+ default:
+ scan_more = 0;
+ break;
+ }
+ }
+ return prefetch;
+}
+
+static inline int is_prefetch(struct pt_regs *regs, unsigned long addr,
+ unsigned long error_code)
+{
+ if (unlikely(boot_cpu_data.x86_vendor == X86_VENDOR_AMD &&
+ boot_cpu_data.x86 >= 6)) {
+ /* Catch an obscure case of prefetch inside an NX page. */
+ if (nx_enabled && (error_code & 16))
+ return 0;
+ return __is_prefetch(regs, addr);
+ }
+ return 0;
+}
+
+static noinline void force_sig_info_fault(int si_signo, int si_code,
+ unsigned long address, struct task_struct *tsk)
+{
+ siginfo_t info;
+
+ info.si_signo = si_signo;
+ info.si_errno = 0;
+ info.si_code = si_code;
+ info.si_addr = (void __user *)address;
+ force_sig_info(si_signo, &info, tsk);
+}
+
+fastcall void do_invalid_op(struct pt_regs *, unsigned long);
+
+#ifdef CONFIG_X86_PAE
+static void dump_fault_path(unsigned long address)
+{
+ unsigned long *p, page;
+ unsigned long mfn;
+
+ page = read_cr3();
+ p = (unsigned long *)__va(page);
+ p += (address >> 30) * 2;
+ printk(KERN_ALERT "%08lx -> *pde = %08lx:%08lx\n", page, p[1], p[0]);
+ if (p[0] & _PAGE_PRESENT) {
+ mfn = (p[0] >> PAGE_SHIFT) | (p[1] << 20);
+ page = mfn_to_pfn(mfn) << PAGE_SHIFT;
+ p = (unsigned long *)__va(page);
+ address &= 0x3fffffff;
+ p += (address >> 21) * 2;
+ printk(KERN_ALERT "%08lx -> *pme = %08lx:%08lx\n",
+ page, p[1], p[0]);
+ mfn = (p[0] >> PAGE_SHIFT) | (p[1] << 20);
+#ifdef CONFIG_HIGHPTE
+ if (mfn_to_pfn(mfn) >= highstart_pfn)
+ return;
+#endif
+ if (p[0] & _PAGE_PRESENT) {
+ page = mfn_to_pfn(mfn) << PAGE_SHIFT;
+ p = (unsigned long *) __va(page);
+ address &= 0x001fffff;
+ p += (address >> 12) * 2;
+ printk(KERN_ALERT "%08lx -> *pte = %08lx:%08lx\n",
+ page, p[1], p[0]);
+ }
+ }
+}
+#else
+static void dump_fault_path(unsigned long address)
+{
+ unsigned long page;
+
+ page = read_cr3();
+ page = ((unsigned long *) __va(page))[address >> 22];
+ if (oops_may_print())
+ printk(KERN_ALERT "*pde = ma %08lx pa %08lx\n", page,
+ machine_to_phys(page));
+ /*
+ * We must not directly access the pte in the highpte
+ * case if the page table is located in highmem.
+ * And lets rather not kmap-atomic the pte, just in case
+ * it's allocated already.
+ */
+#ifdef CONFIG_HIGHPTE
+ if ((page >> PAGE_SHIFT) >= highstart_pfn)
+ return;
+#endif
+ if ((page & 1) && oops_may_print()) {
+ page &= PAGE_MASK;
+ address &= 0x003ff000;
+ page = machine_to_phys(page);
+ page = ((unsigned long *) __va(page))[address >> PAGE_SHIFT];
+ printk(KERN_ALERT "*pte = ma %08lx pa %08lx\n", page,
+ machine_to_phys(page));
+ }
+}
+#endif
+
+static int spurious_fault(struct pt_regs *regs,
+ unsigned long address,
+ unsigned long error_code)
+{
+ pgd_t *pgd;
+ pud_t *pud;
+ pmd_t *pmd;
+ pte_t *pte;
+
+ /* Reserved-bit violation or user access to kernel space? */
+ if (error_code & 0x0c)
+ return 0;
+
+ pgd = init_mm.pgd + pgd_index(address);
+ if (!pgd_present(*pgd))
+ return 0;
+
+ pud = pud_offset(pgd, address);
+ if (!pud_present(*pud))
+ return 0;
+
+ pmd = pmd_offset(pud, address);
+ if (!pmd_present(*pmd))
+ return 0;
+
+ pte = pte_offset_kernel(pmd, address);
+ if (!pte_present(*pte))
+ return 0;
+ if ((error_code & 0x02) && !pte_write(*pte))
+ return 0;
+#ifdef CONFIG_X86_PAE
+ if ((error_code & 0x10) && (__pte_val(*pte) & _PAGE_NX))
+ return 0;
+#endif
+
+ return 1;
+}
+
+static inline pmd_t *vmalloc_sync_one(pgd_t *pgd, unsigned long address)
+{
+ unsigned index = pgd_index(address);
+ pgd_t *pgd_k;
+ pud_t *pud, *pud_k;
+ pmd_t *pmd, *pmd_k;
+
+ pgd += index;
+ pgd_k = init_mm.pgd + index;
+
+ if (!pgd_present(*pgd_k))
+ return NULL;
+
+ /*
+ * set_pgd(pgd, *pgd_k); here would be useless on PAE
+ * and redundant with the set_pmd() on non-PAE. As would
+ * set_pud.
+ */
+
+ pud = pud_offset(pgd, address);
+ pud_k = pud_offset(pgd_k, address);
+ if (!pud_present(*pud_k))
+ return NULL;
+
+ pmd = pmd_offset(pud, address);
+ pmd_k = pmd_offset(pud_k, address);
+ if (!pmd_present(*pmd_k))
+ return NULL;
+ if (!pmd_present(*pmd))
+#if CONFIG_XEN_COMPAT > 0x030002
+ set_pmd(pmd, *pmd_k);
+#else
+ /*
+ * When running on older Xen we must launder *pmd_k through
+ * pmd_val() to ensure that _PAGE_PRESENT is correctly set.
+ */
+ set_pmd(pmd, __pmd(pmd_val(*pmd_k)));
+#endif
+ else
+ BUG_ON(pmd_page(*pmd) != pmd_page(*pmd_k));
+ return pmd_k;
+}
+
+/*
+ * Handle a fault on the vmalloc or module mapping area
+ *
+ * This assumes no large pages in there.
+ */
+static inline int vmalloc_fault(unsigned long address)
+{
+ unsigned long pgd_paddr;
+ pmd_t *pmd_k;
+ pte_t *pte_k;
+ /*
+ * Synchronize this task's top level page-table
+ * with the 'reference' page table.
+ *
+ * Do _not_ use "current" here. We might be inside
+ * an interrupt in the middle of a task switch..
+ */
+ pgd_paddr = read_cr3();
+ pmd_k = vmalloc_sync_one(__va(pgd_paddr), address);
+ if (!pmd_k)
+ return -1;
+ pte_k = pte_offset_kernel(pmd_k, address);
+ if (!pte_present(*pte_k))
+ return -1;
+ return 0;
+}
+
+/*
+ * This routine handles page faults. It determines the address,
+ * and the problem, and then passes it off to one of the appropriate
+ * routines.
+ *
+ * error_code:
+ * bit 0 == 0 means no page found, 1 means protection fault
+ * bit 1 == 0 means read, 1 means write
+ * bit 2 == 0 means kernel, 1 means user-mode
+ * bit 3 == 1 means use of reserved bit detected
+ * bit 4 == 1 means fault was an instruction fetch
+ */
+fastcall void __kprobes do_page_fault(struct pt_regs *regs,
+ unsigned long error_code)
+{
+ struct task_struct *tsk;
+ struct mm_struct *mm;
+ struct vm_area_struct * vma;
+ unsigned long address;
+ int write, si_code;
+
+ /* get the address */
+ address = read_cr2();
+
+ /* Set the "privileged fault" bit to something sane. */
+ error_code &= ~4;
+ error_code |= (regs->xcs & 2) << 1;
+ if (regs->eflags & X86_EFLAGS_VM)
+ error_code |= 4;
+
+ tsk = current;
+
+ si_code = SEGV_MAPERR;
+
+ /*
+ * We fault-in kernel-space virtual memory on-demand. The
+ * 'reference' page table is init_mm.pgd.
+ *
+ * NOTE! We MUST NOT take any locks for this case. We may
+ * be in an interrupt or a critical region, and should
+ * only copy the information from the master page table,
+ * nothing more.
+ *
+ * This verifies that the fault happens in kernel space
+ * (error_code & 4) == 0, and that the fault was not a
+ * protection error (error_code & 9) == 0.
+ */
+ if (unlikely(address >= TASK_SIZE)) {
+#ifdef CONFIG_XEN
+ /* Faults in hypervisor area can never be patched up. */
+ if (address >= hypervisor_virt_start)
+ goto bad_area_nosemaphore;
+#endif
+ if (!(error_code & 0x0000000d) && vmalloc_fault(address) >= 0)
+ return;
+ /* Can take a spurious fault if mapping changes R/O -> R/W. */
+ if (spurious_fault(regs, address, error_code))
+ return;
+ if (notify_page_fault(DIE_PAGE_FAULT, "page fault", regs, error_code, 14,
+ SIGSEGV) == NOTIFY_STOP)
+ return;
+ /*
+ * Don't take the mm semaphore here. If we fixup a prefetch
+ * fault we could otherwise deadlock.
+ */
+ goto bad_area_nosemaphore;
+ }
+
+ if (notify_page_fault(DIE_PAGE_FAULT, "page fault", regs, error_code, 14,
+ SIGSEGV) == NOTIFY_STOP)
+ return;
+
+ /* It's safe to allow irq's after cr2 has been saved and the vmalloc
+ fault has been handled. */
+ if (regs->eflags & (X86_EFLAGS_IF|VM_MASK))
+ local_irq_enable();
+
+ mm = tsk->mm;
+
+ /*
+ * If we're in an interrupt, have no user context or are running in an
+ * atomic region then we must not take the fault..
+ */
+ if (in_atomic() || !mm)
+ goto bad_area_nosemaphore;
+
+ /* When running in the kernel we expect faults to occur only to
+ * addresses in user space. All other faults represent errors in the
+ * kernel and should generate an OOPS. Unfortunatly, in the case of an
+ * erroneous fault occurring in a code path which already holds mmap_sem
+ * we will deadlock attempting to validate the fault against the
+ * address space. Luckily the kernel only validly references user
+ * space from well defined areas of code, which are listed in the
+ * exceptions table.
+ *
+ * As the vast majority of faults will be valid we will only perform
+ * the source reference check when there is a possibilty of a deadlock.
+ * Attempt to lock the address space, if we cannot we then validate the
+ * source. If this is invalid we can skip the address space check,
+ * thus avoiding the deadlock.
+ */
+ if (!down_read_trylock(&mm->mmap_sem)) {
+ if ((error_code & 4) == 0 &&
+ !search_exception_tables(regs->eip))
+ goto bad_area_nosemaphore;
+ down_read(&mm->mmap_sem);
+ }
+
+ vma = find_vma(mm, address);
+ if (!vma)
+ goto bad_area;
+ if (vma->vm_start <= address)
+ goto good_area;
+ if (!(vma->vm_flags & VM_GROWSDOWN))
+ goto bad_area;
+ if (error_code & 4) {
+ /*
+ * Accessing the stack below %esp is always a bug.
+ * The large cushion allows instructions like enter
+ * and pusha to work. ("enter $65535,$31" pushes
+ * 32 pointers and then decrements %esp by 65535.)
+ */
+ if (address + 65536 + 32 * sizeof(unsigned long) < regs->esp)
+ goto bad_area;
+ }
+ if (expand_stack(vma, address))
+ goto bad_area;
+/*
+ * Ok, we have a good vm_area for this memory access, so
+ * we can handle it..
+ */
+good_area:
+ si_code = SEGV_ACCERR;
+ write = 0;
+ switch (error_code & 3) {
+ default: /* 3: write, present */
+#ifdef TEST_VERIFY_AREA
+ if (regs->cs == GET_KERNEL_CS())
+ printk("WP fault at %08lx\n", regs->eip);
+#endif
+ /* fall through */
+ case 2: /* write, not present */
+ if (!(vma->vm_flags & VM_WRITE))
+ goto bad_area;
+ write++;
+ break;
+ case 1: /* read, present */
+ goto bad_area;
+ case 0: /* read, not present */
+ if (!(vma->vm_flags & (VM_READ | VM_EXEC)))
+ goto bad_area;
+ }
+
+ survive:
+ /*
+ * If for any reason at all we couldn't handle the fault,
+ * make sure we exit gracefully rather than endlessly redo
+ * the fault.
+ */
+ switch (handle_mm_fault(mm, vma, address, write)) {
+ case VM_FAULT_MINOR:
+ tsk->min_flt++;
+ break;
+ case VM_FAULT_MAJOR:
+ tsk->maj_flt++;
+ break;
+ case VM_FAULT_SIGBUS:
+ goto do_sigbus;
+ case VM_FAULT_OOM:
+ goto out_of_memory;
+ default:
+ BUG();
+ }
+
+ /*
+ * Did it hit the DOS screen memory VA from vm86 mode?
+ */
+ if (regs->eflags & VM_MASK) {
+ unsigned long bit = (address - 0xA0000) >> PAGE_SHIFT;
+ if (bit < 32)
+ tsk->thread.screen_bitmap |= 1 << bit;
+ }
+ up_read(&mm->mmap_sem);
+ return;
+
+/*
+ * Something tried to access memory that isn't in our memory map..
+ * Fix it, but check if it's kernel or user first..
+ */
+bad_area:
+ up_read(&mm->mmap_sem);
+
+bad_area_nosemaphore:
+ /* User mode accesses just cause a SIGSEGV */
+ if (error_code & 4) {
+ /*
+ * Valid to do another page fault here because this one came
+ * from user space.
+ */
+ if (is_prefetch(regs, address, error_code))
+ return;
+
+ tsk->thread.cr2 = address;
+ /* Kernel addresses are always protection faults */
+ tsk->thread.error_code = error_code | (address >= TASK_SIZE);
+ tsk->thread.trap_no = 14;
+ force_sig_info_fault(SIGSEGV, si_code, address, tsk);
+ return;
+ }
+
+#ifdef CONFIG_X86_F00F_BUG
+ /*
+ * Pentium F0 0F C7 C8 bug workaround.
+ */
+ if (boot_cpu_data.f00f_bug) {
+ unsigned long nr;
+
+ nr = (address - idt_descr.address) >> 3;
+
+ if (nr == 6) {
+ do_invalid_op(regs, 0);
+ return;
+ }
+ }
+#endif
+
+no_context:
+ /* Are we prepared to handle this kernel fault? */
+ if (fixup_exception(regs))
+ return;
+
+ /*
+ * Valid to do another page fault here, because if this fault
+ * had been triggered by is_prefetch fixup_exception would have
+ * handled it.
+ */
+ if (is_prefetch(regs, address, error_code))
+ return;
+
+/*
+ * Oops. The kernel tried to access some bad page. We'll have to
+ * terminate things with extreme prejudice.
+ */
+
+ bust_spinlocks(1);
+
+ if (oops_may_print()) {
+ #ifdef CONFIG_X86_PAE
+ if (error_code & 16) {
+ pte_t *pte = lookup_address(address);
+
+ if (pte && pte_present(*pte) && !pte_exec_kernel(*pte))
+ printk(KERN_CRIT "kernel tried to execute "
+ "NX-protected page - exploit attempt? "
+ "(uid: %d)\n", current->uid);
+ }
+ #endif
+ if (address < PAGE_SIZE)
+ printk(KERN_ALERT "BUG: unable to handle kernel NULL "
+ "pointer dereference");
+ else
+ printk(KERN_ALERT "BUG: unable to handle kernel paging"
+ " request");
+ printk(" at virtual address %08lx\n",address);
+ printk(KERN_ALERT " printing eip:\n");
+ printk("%08lx\n", regs->eip);
+ }
+ dump_fault_path(address);
+ tsk->thread.cr2 = address;
+ tsk->thread.trap_no = 14;
+ tsk->thread.error_code = error_code;
+ die("Oops", regs, error_code);
+ bust_spinlocks(0);
+ do_exit(SIGKILL);
+
+/*
+ * We ran out of memory, or some other thing happened to us that made
+ * us unable to handle the page fault gracefully.
+ */
+out_of_memory:
+ up_read(&mm->mmap_sem);
+ if (tsk->pid == 1) {
+ yield();
+ down_read(&mm->mmap_sem);
+ goto survive;
+ }
+ printk("VM: killing process %s\n", tsk->comm);
+ if (error_code & 4)
+ do_exit(SIGKILL);
+ goto no_context;
+
+do_sigbus:
+ up_read(&mm->mmap_sem);
+
+ /* Kernel mode? Handle exceptions or die */
+ if (!(error_code & 4))
+ goto no_context;
+
+ /* User space => ok to do another page fault */
+ if (is_prefetch(regs, address, error_code))
+ return;
+
+ tsk->thread.cr2 = address;
+ tsk->thread.error_code = error_code;
+ tsk->thread.trap_no = 14;
+ force_sig_info_fault(SIGBUS, BUS_ADRERR, address, tsk);
+}
+
+#if !HAVE_SHARED_KERNEL_PMD
+void vmalloc_sync_all(void)
+{
+ /*
+ * Note that races in the updates of insync and start aren't
+ * problematic: insync can only get set bits added, and updates to
+ * start are only improving performance (without affecting correctness
+ * if undone).
+ * XEN: To work on PAE, we need to iterate over PMDs rather than PGDs.
+ * This change works just fine with 2-level paging too.
+ */
+#define sync_index(a) ((a) >> PMD_SHIFT)
+ static DECLARE_BITMAP(insync, PTRS_PER_PGD*PTRS_PER_PMD);
+ static unsigned long start = TASK_SIZE;
+ unsigned long address;
+
+ BUILD_BUG_ON(TASK_SIZE & ~PGDIR_MASK);
+ for (address = start;
+ address >= TASK_SIZE && address < hypervisor_virt_start;
+ address += 1UL << PMD_SHIFT) {
+ if (!test_bit(sync_index(address), insync)) {
+ unsigned long flags;
+ struct page *page;
+
+ spin_lock_irqsave(&pgd_lock, flags);
+ /* XEN: failure path assumes non-empty pgd_list. */
+ if (unlikely(!pgd_list)) {
+ spin_unlock_irqrestore(&pgd_lock, flags);
+ return;
+ }
+ for (page = pgd_list; page; page =
+ (struct page *)page->index) {
+ spinlock_t *lock = page->mapping
+ ? &((struct mm_struct *)page->mapping)
+ ->page_table_lock
+ : NULL;
+ pmd_t *pmd;
+
+ if (lock)
+ spin_lock(lock);
+ pmd = vmalloc_sync_one(page_address(page),
+ address);
+ if (lock)
+ spin_unlock(lock);
+
+ if (!pmd) {
+ BUG_ON(page != pgd_list);
+ break;
+ }
+ }
+ spin_unlock_irqrestore(&pgd_lock, flags);
+ if (!page)
+ set_bit(sync_index(address), insync);
+ }
+ if (address == start && test_bit(sync_index(address), insync))
+ start = address + (1UL << PMD_SHIFT);
+ }
+}
+#endif
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/mm/highmem_32-xen.c 2008-10-29 09:55:56.000000000 +0100
@@ -0,0 +1,183 @@
+#include <linux/highmem.h>
+#include <linux/module.h>
+
+void *kmap(struct page *page)
+{
+ might_sleep();
+ if (!PageHighMem(page))
+ return page_address(page);
+ return kmap_high(page);
+}
+
+void kunmap(struct page *page)
+{
+ if (in_interrupt())
+ BUG();
+ if (!PageHighMem(page))
+ return;
+ kunmap_high(page);
+}
+
+/*
+ * kmap_atomic/kunmap_atomic is significantly faster than kmap/kunmap because
+ * no global lock is needed and because the kmap code must perform a global TLB
+ * invalidation when the kmap pool wraps.
+ *
+ * However when holding an atomic kmap is is not legal to sleep, so atomic
+ * kmaps are appropriate for short, tight code paths only.
+ */
+static void *__kmap_atomic(struct page *page, enum km_type type, pgprot_t prot)
+{
+ enum fixed_addresses idx;
+ unsigned long vaddr;
+
+ /* even !CONFIG_PREEMPT needs this, for in_atomic in do_page_fault */
+ inc_preempt_count();
+ if (!PageHighMem(page))
+ return page_address(page);
+
+ idx = type + KM_TYPE_NR*smp_processor_id();
+ vaddr = __fix_to_virt(FIX_KMAP_BEGIN + idx);
+#ifdef CONFIG_DEBUG_HIGHMEM
+ if (!pte_none(*(kmap_pte-idx)))
+ BUG();
+#endif
+ set_pte_at_sync(&init_mm, vaddr, kmap_pte-idx, mk_pte(page, prot));
+
+ return (void*) vaddr;
+}
+
+void *kmap_atomic(struct page *page, enum km_type type)
+{
+ return __kmap_atomic(page, type, kmap_prot);
+}
+
+/* Same as kmap_atomic but with PAGE_KERNEL_RO page protection. */
+void *kmap_atomic_pte(struct page *page, enum km_type type)
+{
+ return __kmap_atomic(page, type,
+ test_bit(PG_pinned, &page->flags)
+ ? PAGE_KERNEL_RO : kmap_prot);
+}
+
+void kunmap_atomic(void *kvaddr, enum km_type type)
+{
+#if defined(CONFIG_DEBUG_HIGHMEM) || defined(CONFIG_XEN)
+ unsigned long vaddr = (unsigned long) kvaddr & PAGE_MASK;
+ enum fixed_addresses idx = type + KM_TYPE_NR*smp_processor_id();
+
+ if (vaddr < FIXADDR_START) { // FIXME
+ dec_preempt_count();
+ preempt_check_resched();
+ return;
+ }
+#endif
+
+#if defined(CONFIG_DEBUG_HIGHMEM)
+ if (vaddr != __fix_to_virt(FIX_KMAP_BEGIN+idx))
+ BUG();
+
+ /*
+ * force other mappings to Oops if they'll try to access
+ * this pte without first remap it
+ */
+ pte_clear(&init_mm, vaddr, kmap_pte-idx);
+ __flush_tlb_one(vaddr);
+#elif defined(CONFIG_XEN)
+ /*
+ * We must ensure there are no dangling pagetable references when
+ * returning memory to Xen (decrease_reservation).
+ * XXX TODO: We could make this faster by only zapping when
+ * kmap_flush_unused is called but that is trickier and more invasive.
+ */
+ pte_clear(&init_mm, vaddr, kmap_pte-idx);
+#endif
+
+ dec_preempt_count();
+ preempt_check_resched();
+}
+
+/* This is the same as kmap_atomic() but can map memory that doesn't
+ * have a struct page associated with it.
+ */
+void *kmap_atomic_pfn(unsigned long pfn, enum km_type type)
+{
+ enum fixed_addresses idx;
+ unsigned long vaddr;
+
+ inc_preempt_count();
+
+ idx = type + KM_TYPE_NR*smp_processor_id();
+ vaddr = __fix_to_virt(FIX_KMAP_BEGIN + idx);
+ set_pte(kmap_pte-idx, pfn_pte(pfn, kmap_prot));
+ __flush_tlb_one(vaddr);
+
+ return (void*) vaddr;
+}
+
+struct page *kmap_atomic_to_page(void *ptr)
+{
+ unsigned long idx, vaddr = (unsigned long)ptr;
+ pte_t *pte;
+
+ if (vaddr < FIXADDR_START)
+ return virt_to_page(ptr);
+
+ idx = virt_to_fix(vaddr);
+ pte = kmap_pte - (idx - FIX_KMAP_BEGIN);
+ return pte_page(*pte);
+}
+
+void clear_highpage(struct page *page)
+{
+ void *kaddr;
+
+ if (likely(xen_feature(XENFEAT_highmem_assist))
+ && PageHighMem(page)) {
+ struct mmuext_op meo;
+
+ meo.cmd = MMUEXT_CLEAR_PAGE;
+ meo.arg1.mfn = pfn_to_mfn(page_to_pfn(page));
+ if (HYPERVISOR_mmuext_op(&meo, 1, NULL, DOMID_SELF) == 0)
+ return;
+ }
+
+ kaddr = kmap_atomic(page, KM_USER0);
+ clear_page(kaddr);
+ kunmap_atomic(kaddr, KM_USER0);
+}
+
+void copy_highpage(struct page *to, struct page *from)
+{
+ void *vfrom, *vto;
+
+ if (likely(xen_feature(XENFEAT_highmem_assist))
+ && (PageHighMem(from) || PageHighMem(to))) {
+ unsigned long from_pfn = page_to_pfn(from);
+ unsigned long to_pfn = page_to_pfn(to);
+ struct mmuext_op meo;
+
+ meo.cmd = MMUEXT_COPY_PAGE;
+ meo.arg1.mfn = pfn_to_mfn(to_pfn);
+ meo.arg2.src_mfn = pfn_to_mfn(from_pfn);
+ if (mfn_to_pfn(meo.arg2.src_mfn) == from_pfn
+ && mfn_to_pfn(meo.arg1.mfn) == to_pfn
+ && HYPERVISOR_mmuext_op(&meo, 1, NULL, DOMID_SELF) == 0)
+ return;
+ }
+
+ vfrom = kmap_atomic(from, KM_USER0);
+ vto = kmap_atomic(to, KM_USER1);
+ copy_page(vto, vfrom);
+ kunmap_atomic(vfrom, KM_USER0);
+ kunmap_atomic(vto, KM_USER1);
+}
+
+EXPORT_SYMBOL(kmap);
+EXPORT_SYMBOL(kunmap);
+EXPORT_SYMBOL(kmap_atomic);
+EXPORT_SYMBOL(kmap_atomic_pte);
+EXPORT_SYMBOL(kunmap_atomic);
+EXPORT_SYMBOL(kmap_atomic_to_page);
+EXPORT_SYMBOL(clear_highpage);
+EXPORT_SYMBOL(copy_highpage);
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/mm/hypervisor.c 2009-06-09 15:01:37.000000000 +0200
@@ -0,0 +1,579 @@
+/******************************************************************************
+ * mm/hypervisor.c
+ *
+ * Update page tables via the hypervisor.
+ *
+ * Copyright (c) 2002-2004, K A Fraser
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License version 2
+ * as published by the Free Software Foundation; or, when distributed
+ * separately from the Linux kernel or incorporated into other
+ * software packages, subject to the following license:
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this source file (the "Software"), to deal in the Software without
+ * restriction, including without limitation the rights to use, copy, modify,
+ * merge, publish, distribute, sublicense, and/or sell copies of the Software,
+ * and to permit persons to whom the Software is furnished to do so, subject to
+ * the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
+ * IN THE SOFTWARE.
+ */
+
+#include <linux/sched.h>
+#include <linux/mm.h>
+#include <linux/vmalloc.h>
+#include <asm/page.h>
+#include <asm/pgtable.h>
+#include <asm/hypervisor.h>
+#include <xen/balloon.h>
+#include <xen/features.h>
+#include <xen/interface/memory.h>
+#include <linux/module.h>
+#include <linux/percpu.h>
+#include <asm/tlbflush.h>
+#include <linux/highmem.h>
+
+void xen_l1_entry_update(pte_t *ptr, pte_t val)
+{
+ mmu_update_t u;
+ u.ptr = ptep_to_machine(ptr);
+ u.val = __pte_val(val);
+ BUG_ON(HYPERVISOR_mmu_update(&u, 1, NULL, DOMID_SELF) < 0);
+}
+EXPORT_SYMBOL_GPL(xen_l1_entry_update);
+
+void xen_l2_entry_update(pmd_t *ptr, pmd_t val)
+{
+ mmu_update_t u;
+ u.ptr = virt_to_machine(ptr);
+ u.val = __pmd_val(val);
+ BUG_ON(HYPERVISOR_mmu_update(&u, 1, NULL, DOMID_SELF) < 0);
+}
+
+#if defined(CONFIG_X86_PAE) || defined(CONFIG_X86_64)
+void xen_l3_entry_update(pud_t *ptr, pud_t val)
+{
+ mmu_update_t u;
+ u.ptr = virt_to_machine(ptr);
+ u.val = __pud_val(val);
+ BUG_ON(HYPERVISOR_mmu_update(&u, 1, NULL, DOMID_SELF) < 0);
+}
+#endif
+
+#ifdef CONFIG_X86_64
+void xen_l4_entry_update(pgd_t *ptr, pgd_t val)
+{
+ mmu_update_t u;
+ u.ptr = virt_to_machine(ptr);
+ u.val = __pgd_val(val);
+ BUG_ON(HYPERVISOR_mmu_update(&u, 1, NULL, DOMID_SELF) < 0);
+}
+#endif /* CONFIG_X86_64 */
+
+void xen_pt_switch(unsigned long ptr)
+{
+ struct mmuext_op op;
+ op.cmd = MMUEXT_NEW_BASEPTR;
+ op.arg1.mfn = pfn_to_mfn(ptr >> PAGE_SHIFT);
+ BUG_ON(HYPERVISOR_mmuext_op(&op, 1, NULL, DOMID_SELF) < 0);
+}
+
+void xen_new_user_pt(unsigned long ptr)
+{
+ struct mmuext_op op;
+ op.cmd = MMUEXT_NEW_USER_BASEPTR;
+ op.arg1.mfn = pfn_to_mfn(ptr >> PAGE_SHIFT);
+ BUG_ON(HYPERVISOR_mmuext_op(&op, 1, NULL, DOMID_SELF) < 0);
+}
+
+void xen_tlb_flush(void)
+{
+ struct mmuext_op op;
+ op.cmd = MMUEXT_TLB_FLUSH_LOCAL;
+ BUG_ON(HYPERVISOR_mmuext_op(&op, 1, NULL, DOMID_SELF) < 0);
+}
+EXPORT_SYMBOL(xen_tlb_flush);
+
+void xen_invlpg(unsigned long ptr)
+{
+ struct mmuext_op op;
+ op.cmd = MMUEXT_INVLPG_LOCAL;
+ op.arg1.linear_addr = ptr & PAGE_MASK;
+ BUG_ON(HYPERVISOR_mmuext_op(&op, 1, NULL, DOMID_SELF) < 0);
+}
+EXPORT_SYMBOL(xen_invlpg);
+
+#ifdef CONFIG_SMP
+
+void xen_tlb_flush_all(void)
+{
+ struct mmuext_op op;
+ op.cmd = MMUEXT_TLB_FLUSH_ALL;
+ BUG_ON(HYPERVISOR_mmuext_op(&op, 1, NULL, DOMID_SELF) < 0);
+}
+EXPORT_SYMBOL_GPL(xen_tlb_flush_all);
+
+void xen_tlb_flush_mask(cpumask_t *mask)
+{
+ struct mmuext_op op;
+ if ( cpus_empty(*mask) )
+ return;
+ op.cmd = MMUEXT_TLB_FLUSH_MULTI;
+ set_xen_guest_handle(op.arg2.vcpumask, mask->bits);
+ BUG_ON(HYPERVISOR_mmuext_op(&op, 1, NULL, DOMID_SELF) < 0);
+}
+EXPORT_SYMBOL_GPL(xen_tlb_flush_mask);
+
+void xen_invlpg_all(unsigned long ptr)
+{
+ struct mmuext_op op;
+ op.cmd = MMUEXT_INVLPG_ALL;
+ op.arg1.linear_addr = ptr & PAGE_MASK;
+ BUG_ON(HYPERVISOR_mmuext_op(&op, 1, NULL, DOMID_SELF) < 0);
+}
+EXPORT_SYMBOL_GPL(xen_invlpg_all);
+
+void xen_invlpg_mask(cpumask_t *mask, unsigned long ptr)
+{
+ struct mmuext_op op;
+ if ( cpus_empty(*mask) )
+ return;
+ op.cmd = MMUEXT_INVLPG_MULTI;
+ op.arg1.linear_addr = ptr & PAGE_MASK;
+ set_xen_guest_handle(op.arg2.vcpumask, mask->bits);
+ BUG_ON(HYPERVISOR_mmuext_op(&op, 1, NULL, DOMID_SELF) < 0);
+}
+EXPORT_SYMBOL_GPL(xen_invlpg_mask);
+
+#endif /* CONFIG_SMP */
+
+void xen_pgd_pin(unsigned long ptr)
+{
+ struct mmuext_op op;
+#ifdef CONFIG_X86_64
+ op.cmd = MMUEXT_PIN_L4_TABLE;
+#elif defined(CONFIG_X86_PAE)
+ op.cmd = MMUEXT_PIN_L3_TABLE;
+#else
+ op.cmd = MMUEXT_PIN_L2_TABLE;
+#endif
+ op.arg1.mfn = pfn_to_mfn(ptr >> PAGE_SHIFT);
+ BUG_ON(HYPERVISOR_mmuext_op(&op, 1, NULL, DOMID_SELF) < 0);
+}
+
+void xen_pgd_unpin(unsigned long ptr)
+{
+ struct mmuext_op op;
+ op.cmd = MMUEXT_UNPIN_TABLE;
+ op.arg1.mfn = pfn_to_mfn(ptr >> PAGE_SHIFT);
+ BUG_ON(HYPERVISOR_mmuext_op(&op, 1, NULL, DOMID_SELF) < 0);
+}
+
+void xen_set_ldt(const void *ptr, unsigned int ents)
+{
+ struct mmuext_op op;
+ op.cmd = MMUEXT_SET_LDT;
+ op.arg1.linear_addr = (unsigned long)ptr;
+ op.arg2.nr_ents = ents;
+ BUG_ON(HYPERVISOR_mmuext_op(&op, 1, NULL, DOMID_SELF) < 0);
+}
+
+/* Protected by balloon_lock. */
+#define MAX_CONTIG_ORDER 9 /* 2MB */
+static unsigned long discontig_frames[1<<MAX_CONTIG_ORDER];
+static unsigned long limited_frames[1<<MAX_CONTIG_ORDER];
+static multicall_entry_t cr_mcl[1<<MAX_CONTIG_ORDER];
+
+/* Ensure multi-page extents are contiguous in machine memory. */
+int xen_create_contiguous_region(
+ unsigned long vstart, unsigned int order, unsigned int address_bits)
+{
+ unsigned long *in_frames = discontig_frames, out_frame;
+ unsigned long frame, flags;
+ unsigned int i;
+ int rc, success;
+ struct xen_memory_exchange exchange = {
+ .in = {
+ .nr_extents = 1UL << order,
+ .extent_order = 0,
+ .domid = DOMID_SELF
+ },
+ .out = {
+ .nr_extents = 1,
+ .extent_order = order,
+ .address_bits = address_bits,
+ .domid = DOMID_SELF
+ }
+ };
+
+ /*
+ * Currently an auto-translated guest will not perform I/O, nor will
+ * it require PAE page directories below 4GB. Therefore any calls to
+ * this function are redundant and can be ignored.
+ */
+ if (xen_feature(XENFEAT_auto_translated_physmap))
+ return 0;
+
+ if (unlikely(order > MAX_CONTIG_ORDER))
+ return -ENOMEM;
+
+ set_xen_guest_handle(exchange.in.extent_start, in_frames);
+ set_xen_guest_handle(exchange.out.extent_start, &out_frame);
+
+ scrub_pages((void *)vstart, 1 << order);
+
+ balloon_lock(flags);
+
+ /* 1. Zap current PTEs, remembering MFNs. */
+ for (i = 0; i < (1U<<order); i++) {
+ in_frames[i] = pfn_to_mfn((__pa(vstart) >> PAGE_SHIFT) + i);
+ MULTI_update_va_mapping(cr_mcl + i, vstart + (i*PAGE_SIZE),
+ __pte_ma(0), 0);
+ set_phys_to_machine((__pa(vstart)>>PAGE_SHIFT)+i,
+ INVALID_P2M_ENTRY);
+ }
+ if (HYPERVISOR_multicall_check(cr_mcl, i, NULL))
+ BUG();
+
+ /* 2. Get a new contiguous memory extent. */
+ out_frame = __pa(vstart) >> PAGE_SHIFT;
+ rc = HYPERVISOR_memory_op(XENMEM_exchange, &exchange);
+ success = (exchange.nr_exchanged == (1UL << order));
+ BUG_ON(!success && ((exchange.nr_exchanged != 0) || (rc == 0)));
+ BUG_ON(success && (rc != 0));
+#if CONFIG_XEN_COMPAT <= 0x030002
+ if (unlikely(rc == -ENOSYS)) {
+ /* Compatibility when XENMEM_exchange is unsupported. */
+ if (HYPERVISOR_memory_op(XENMEM_decrease_reservation,
+ &exchange.in) != (1UL << order))
+ BUG();
+ success = (HYPERVISOR_memory_op(XENMEM_populate_physmap,
+ &exchange.out) == 1);
+ if (!success) {
+ /* Couldn't get special memory: fall back to normal. */
+ for (i = 0; i < (1U<<order); i++)
+ in_frames[i] = (__pa(vstart)>>PAGE_SHIFT) + i;
+ if (HYPERVISOR_memory_op(XENMEM_populate_physmap,
+ &exchange.in) != (1UL<<order))
+ BUG();
+ }
+ }
+#endif
+
+ /* 3. Map the new extent in place of old pages. */
+ for (i = 0; i < (1U<<order); i++) {
+ frame = success ? (out_frame + i) : in_frames[i];
+ MULTI_update_va_mapping(cr_mcl + i, vstart + (i*PAGE_SIZE),
+ pfn_pte_ma(frame, PAGE_KERNEL), 0);
+ set_phys_to_machine((__pa(vstart)>>PAGE_SHIFT)+i, frame);
+ }
+
+ cr_mcl[i - 1].args[MULTI_UVMFLAGS_INDEX] = order
+ ? UVMF_TLB_FLUSH|UVMF_ALL
+ : UVMF_INVLPG|UVMF_ALL;
+ if (HYPERVISOR_multicall_check(cr_mcl, i, NULL))
+ BUG();
+
+ balloon_unlock(flags);
+
+ return success ? 0 : -ENOMEM;
+}
+EXPORT_SYMBOL_GPL(xen_create_contiguous_region);
+
+void xen_destroy_contiguous_region(unsigned long vstart, unsigned int order)
+{
+ unsigned long *out_frames = discontig_frames, in_frame;
+ unsigned long frame, flags;
+ unsigned int i;
+ int rc, success;
+ struct xen_memory_exchange exchange = {
+ .in = {
+ .nr_extents = 1,
+ .extent_order = order,
+ .domid = DOMID_SELF
+ },
+ .out = {
+ .nr_extents = 1UL << order,
+ .extent_order = 0,
+ .domid = DOMID_SELF
+ }
+ };
+
+ if (xen_feature(XENFEAT_auto_translated_physmap))
+ return;
+
+ if (unlikely(order > MAX_CONTIG_ORDER))
+ return;
+
+ set_xen_guest_handle(exchange.in.extent_start, &in_frame);
+ set_xen_guest_handle(exchange.out.extent_start, out_frames);
+
+ scrub_pages((void *)vstart, 1 << order);
+
+ balloon_lock(flags);
+
+ /* 1. Find start MFN of contiguous extent. */
+ in_frame = pfn_to_mfn(__pa(vstart) >> PAGE_SHIFT);
+
+ /* 2. Zap current PTEs. */
+ for (i = 0; i < (1U<<order); i++) {
+ MULTI_update_va_mapping(cr_mcl + i, vstart + (i*PAGE_SIZE),
+ __pte_ma(0), 0);
+ set_phys_to_machine((__pa(vstart)>>PAGE_SHIFT)+i,
+ INVALID_P2M_ENTRY);
+ out_frames[i] = (__pa(vstart) >> PAGE_SHIFT) + i;
+ }
+ if (HYPERVISOR_multicall_check(cr_mcl, i, NULL))
+ BUG();
+
+ /* 3. Do the exchange for non-contiguous MFNs. */
+ rc = HYPERVISOR_memory_op(XENMEM_exchange, &exchange);
+ success = (exchange.nr_exchanged == 1);
+ BUG_ON(!success && ((exchange.nr_exchanged != 0) || (rc == 0)));
+ BUG_ON(success && (rc != 0));
+#if CONFIG_XEN_COMPAT <= 0x030002
+ if (unlikely(rc == -ENOSYS)) {
+ /* Compatibility when XENMEM_exchange is unsupported. */
+ if (HYPERVISOR_memory_op(XENMEM_decrease_reservation,
+ &exchange.in) != 1)
+ BUG();
+ if (HYPERVISOR_memory_op(XENMEM_populate_physmap,
+ &exchange.out) != (1UL << order))
+ BUG();
+ success = 1;
+ }
+#endif
+
+ /* 4. Map new pages in place of old pages. */
+ for (i = 0; i < (1U<<order); i++) {
+ frame = success ? out_frames[i] : (in_frame + i);
+ MULTI_update_va_mapping(cr_mcl + i, vstart + (i*PAGE_SIZE),
+ pfn_pte_ma(frame, PAGE_KERNEL), 0);
+ set_phys_to_machine((__pa(vstart)>>PAGE_SHIFT)+i, frame);
+ }
+
+ cr_mcl[i - 1].args[MULTI_UVMFLAGS_INDEX] = order
+ ? UVMF_TLB_FLUSH|UVMF_ALL
+ : UVMF_INVLPG|UVMF_ALL;
+ if (HYPERVISOR_multicall_check(cr_mcl, i, NULL))
+ BUG();
+
+ balloon_unlock(flags);
+}
+EXPORT_SYMBOL_GPL(xen_destroy_contiguous_region);
+
+static void undo_limit_pages(struct page *pages, unsigned int order)
+{
+ BUG_ON(xen_feature(XENFEAT_auto_translated_physmap));
+ BUG_ON(order > MAX_CONTIG_ORDER);
+ xen_limit_pages_to_max_mfn(pages, order, 0);
+ ClearPageForeign(pages);
+ __free_pages(pages, order);
+}
+
+int xen_limit_pages_to_max_mfn(
+ struct page *pages, unsigned int order, unsigned int address_bits)
+{
+ unsigned long flags, frame;
+ unsigned long *in_frames = discontig_frames, *out_frames = limited_frames;
+ struct page *page;
+ unsigned int i, n, nr_mcl;
+ int rc, success;
+ DECLARE_BITMAP(limit_map, 1 << MAX_CONTIG_ORDER);
+
+ struct xen_memory_exchange exchange = {
+ .in = {
+ .extent_order = 0,
+ .domid = DOMID_SELF
+ },
+ .out = {
+ .extent_order = 0,
+ .address_bits = address_bits,
+ .domid = DOMID_SELF
+ }
+ };
+
+ if (xen_feature(XENFEAT_auto_translated_physmap))
+ return 0;
+
+ if (unlikely(order > MAX_CONTIG_ORDER))
+ return -ENOMEM;
+
+ if (address_bits) {
+ if (address_bits < PAGE_SHIFT)
+ return -EINVAL;
+ bitmap_zero(limit_map, 1U << order);
+ } else if (order) {
+ BUILD_BUG_ON(sizeof(pages->index) != sizeof(*limit_map));
+ for (i = 0; i < BITS_TO_LONGS(1U << order); ++i)
+ limit_map[i] = pages[i + 1].index;
+ } else
+ __set_bit(0, limit_map);
+
+ set_xen_guest_handle(exchange.in.extent_start, in_frames);
+ set_xen_guest_handle(exchange.out.extent_start, out_frames);
+
+ /* 0. Scrub the pages. */
+ for (i = 0, n = 0; i < 1U<<order ; i++) {
+ page = &pages[i];
+ if (address_bits) {
+ if (!(pfn_to_mfn(page_to_pfn(page)) >> (address_bits - PAGE_SHIFT)))
+ continue;
+ __set_bit(i, limit_map);
+ }
+
+ if (!PageHighMem(page))
+ scrub_pages(page_address(page), 1);
+#ifdef CONFIG_XEN_SCRUB_PAGES
+ else {
+ scrub_pages(kmap(page), 1);
+ kunmap(page);
+ ++n;
+ }
+#endif
+ }
+ if (bitmap_empty(limit_map, 1U << order))
+ return 0;
+
+ if (n)
+ kmap_flush_unused();
+
+ balloon_lock(flags);
+
+ /* 1. Zap current PTEs (if any), remembering MFNs. */
+ for (i = 0, n = 0, nr_mcl = 0; i < (1U<<order); i++) {
+ if(!test_bit(i, limit_map))
+ continue;
+ page = &pages[i];
+
+ out_frames[n] = page_to_pfn(page);
+ in_frames[n] = pfn_to_mfn(out_frames[n]);
+
+ if (!PageHighMem(page))
+ MULTI_update_va_mapping(cr_mcl + nr_mcl++,
+ (unsigned long)page_address(page),
+ __pte_ma(0), 0);
+
+ set_phys_to_machine(out_frames[n], INVALID_P2M_ENTRY);
+ ++n;
+ }
+ if (nr_mcl && HYPERVISOR_multicall_check(cr_mcl, nr_mcl, NULL))
+ BUG();
+
+ /* 2. Get new memory below the required limit. */
+ exchange.in.nr_extents = n;
+ exchange.out.nr_extents = n;
+ rc = HYPERVISOR_memory_op(XENMEM_exchange, &exchange);
+ success = (exchange.nr_exchanged == n);
+ BUG_ON(!success && ((exchange.nr_exchanged != 0) || (rc == 0)));
+ BUG_ON(success && (rc != 0));
+#if CONFIG_XEN_COMPAT <= 0x030002
+ if (unlikely(rc == -ENOSYS)) {
+ /* Compatibility when XENMEM_exchange is unsupported. */
+ if (HYPERVISOR_memory_op(XENMEM_decrease_reservation,
+ &exchange.in) != n)
+ BUG();
+ if (HYPERVISOR_memory_op(XENMEM_populate_physmap,
+ &exchange.out) != n)
+ BUG();
+ success = 1;
+ }
+#endif
+
+ /* 3. Map the new pages in place of old pages. */
+ for (i = 0, n = 0, nr_mcl = 0; i < (1U<<order); i++) {
+ if(!test_bit(i, limit_map))
+ continue;
+ page = &pages[i];
+
+ frame = success ? out_frames[n] : in_frames[n];
+
+ if (!PageHighMem(page))
+ MULTI_update_va_mapping(cr_mcl + nr_mcl++,
+ (unsigned long)page_address(page),
+ pfn_pte_ma(frame, PAGE_KERNEL), 0);
+
+ set_phys_to_machine(page_to_pfn(page), frame);
+ ++n;
+ }
+ if (nr_mcl) {
+ cr_mcl[nr_mcl - 1].args[MULTI_UVMFLAGS_INDEX] = order
+ ? UVMF_TLB_FLUSH|UVMF_ALL
+ : UVMF_INVLPG|UVMF_ALL;
+ if (HYPERVISOR_multicall_check(cr_mcl, nr_mcl, NULL))
+ BUG();
+ }
+
+ balloon_unlock(flags);
+
+ if (!success)
+ return -ENOMEM;
+
+ if (address_bits) {
+ if (order) {
+ BUILD_BUG_ON(sizeof(*limit_map) != sizeof(pages->index));
+ for (i = 0; i < BITS_TO_LONGS(1U << order); ++i)
+ pages[i + 1].index = limit_map[i];
+ }
+ SetPageForeign(pages, undo_limit_pages);
+ }
+
+ return 0;
+}
+EXPORT_SYMBOL_GPL(xen_limit_pages_to_max_mfn);
+
+#ifdef __i386__
+int write_ldt_entry(void *ldt, int entry, __u32 entry_a, __u32 entry_b)
+{
+ __u32 *lp = (__u32 *)((char *)ldt + entry * 8);
+ maddr_t mach_lp = arbitrary_virt_to_machine(lp);
+ return HYPERVISOR_update_descriptor(
+ mach_lp, (u64)entry_a | ((u64)entry_b<<32));
+}
+#endif
+
+#define MAX_BATCHED_FULL_PTES 32
+
+int xen_change_pte_range(struct mm_struct *mm, pmd_t *pmd,
+ unsigned long addr, unsigned long end, pgprot_t newprot)
+{
+ int rc = 0, i = 0;
+ mmu_update_t u[MAX_BATCHED_FULL_PTES];
+ pte_t *pte;
+ spinlock_t *ptl;
+
+ if (!xen_feature(XENFEAT_mmu_pt_update_preserve_ad))
+ return 0;
+
+ pte = pte_offset_map_lock(mm, pmd, addr, &ptl);
+ do {
+ if (pte_present(*pte)) {
+ u[i].ptr = (__pmd_val(*pmd) & PHYSICAL_PAGE_MASK)
+ | ((unsigned long)pte & ~PAGE_MASK)
+ | MMU_PT_UPDATE_PRESERVE_AD;
+ u[i].val = __pte_val(pte_modify(*pte, newprot));
+ if (++i == MAX_BATCHED_FULL_PTES) {
+ if ((rc = HYPERVISOR_mmu_update(
+ &u[0], i, NULL, DOMID_SELF)) != 0)
+ break;
+ i = 0;
+ }
+ }
+ } while (pte++, addr += PAGE_SIZE, addr != end);
+ if (i)
+ rc = HYPERVISOR_mmu_update( &u[0], i, NULL, DOMID_SELF);
+ pte_unmap_unlock(pte - 1, ptl);
+ BUG_ON(rc && rc != -ENOSYS);
+ return !rc;
+}
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/mm/init_32-xen.c 2008-10-29 09:55:56.000000000 +0100
@@ -0,0 +1,840 @@
+/*
+ * linux/arch/i386/mm/init.c
+ *
+ * Copyright (C) 1995 Linus Torvalds
+ *
+ * Support of BIGMEM added by Gerhard Wichert, Siemens AG, July 1999
+ */
+
+#include <linux/module.h>
+#include <linux/signal.h>
+#include <linux/sched.h>
+#include <linux/kernel.h>
+#include <linux/errno.h>
+#include <linux/string.h>
+#include <linux/types.h>
+#include <linux/ptrace.h>
+#include <linux/mman.h>
+#include <linux/mm.h>
+#include <linux/hugetlb.h>
+#include <linux/swap.h>
+#include <linux/smp.h>
+#include <linux/init.h>
+#include <linux/highmem.h>
+#include <linux/pagemap.h>
+#include <linux/poison.h>
+#include <linux/bootmem.h>
+#include <linux/slab.h>
+#include <linux/proc_fs.h>
+#include <linux/efi.h>
+#include <linux/memory_hotplug.h>
+#include <linux/initrd.h>
+#include <linux/cpumask.h>
+#include <linux/dma-mapping.h>
+#include <linux/scatterlist.h>
+
+#include <asm/processor.h>
+#include <asm/system.h>
+#include <asm/uaccess.h>
+#include <asm/pgtable.h>
+#include <asm/dma.h>
+#include <asm/fixmap.h>
+#include <asm/e820.h>
+#include <asm/apic.h>
+#include <asm/tlb.h>
+#include <asm/tlbflush.h>
+#include <asm/sections.h>
+#include <asm/hypervisor.h>
+#include <asm/swiotlb.h>
+
+unsigned int __VMALLOC_RESERVE = 128 << 20;
+
+DEFINE_PER_CPU(struct mmu_gather, mmu_gathers);
+unsigned long highstart_pfn, highend_pfn;
+
+static int noinline do_test_wp_bit(void);
+
+/*
+ * Creates a middle page table and puts a pointer to it in the
+ * given global directory entry. This only returns the gd entry
+ * in non-PAE compilation mode, since the middle layer is folded.
+ */
+static pmd_t * __init one_md_table_init(pgd_t *pgd)
+{
+ pud_t *pud;
+ pmd_t *pmd_table;
+
+#ifdef CONFIG_X86_PAE
+ pmd_table = (pmd_t *) alloc_bootmem_low_pages(PAGE_SIZE);
+ make_lowmem_page_readonly(pmd_table, XENFEAT_writable_page_tables);
+ set_pgd(pgd, __pgd(__pa(pmd_table) | _PAGE_PRESENT));
+ pud = pud_offset(pgd, 0);
+ if (pmd_table != pmd_offset(pud, 0))
+ BUG();
+#else
+ pud = pud_offset(pgd, 0);
+ pmd_table = pmd_offset(pud, 0);
+#endif
+
+ return pmd_table;
+}
+
+/*
+ * Create a page table and place a pointer to it in a middle page
+ * directory entry.
+ */
+static pte_t * __init one_page_table_init(pmd_t *pmd)
+{
+ if (pmd_none(*pmd)) {
+ pte_t *page_table = (pte_t *) alloc_bootmem_low_pages(PAGE_SIZE);
+ make_lowmem_page_readonly(page_table,
+ XENFEAT_writable_page_tables);
+ set_pmd(pmd, __pmd(__pa(page_table) | _PAGE_TABLE));
+ if (page_table != pte_offset_kernel(pmd, 0))
+ BUG();
+
+ return page_table;
+ }
+
+ return pte_offset_kernel(pmd, 0);
+}
+
+/*
+ * This function initializes a certain range of kernel virtual memory
+ * with new bootmem page tables, everywhere page tables are missing in
+ * the given range.
+ */
+
+/*
+ * NOTE: The pagetables are allocated contiguous on the physical space
+ * so we can cache the place of the first one and move around without
+ * checking the pgd every time.
+ */
+static void __init page_table_range_init (unsigned long start, unsigned long end, pgd_t *pgd_base)
+{
+ pgd_t *pgd;
+ pud_t *pud;
+ pmd_t *pmd;
+ int pgd_idx, pmd_idx;
+ unsigned long vaddr;
+
+ vaddr = start;
+ pgd_idx = pgd_index(vaddr);
+ pmd_idx = pmd_index(vaddr);
+ pgd = pgd_base + pgd_idx;
+
+ for ( ; (pgd_idx < PTRS_PER_PGD) && (vaddr != end); pgd++, pgd_idx++) {
+ if (pgd_none(*pgd))
+ one_md_table_init(pgd);
+ pud = pud_offset(pgd, vaddr);
+ pmd = pmd_offset(pud, vaddr);
+ for (; (pmd_idx < PTRS_PER_PMD) && (vaddr != end); pmd++, pmd_idx++) {
+ if (vaddr < hypervisor_virt_start && pmd_none(*pmd))
+ one_page_table_init(pmd);
+
+ vaddr += PMD_SIZE;
+ }
+ pmd_idx = 0;
+ }
+}
+
+static inline int is_kernel_text(unsigned long addr)
+{
+ if (addr >= PAGE_OFFSET && addr <= (unsigned long)__init_end)
+ return 1;
+ return 0;
+}
+
+/*
+ * This maps the physical memory to kernel virtual address space, a total
+ * of max_low_pfn pages, by creating page tables starting from address
+ * PAGE_OFFSET.
+ */
+static void __init kernel_physical_mapping_init(pgd_t *pgd_base)
+{
+ unsigned long pfn;
+ pgd_t *pgd;
+ pmd_t *pmd;
+ pte_t *pte;
+ int pgd_idx, pmd_idx, pte_ofs;
+
+ unsigned long max_ram_pfn = xen_start_info->nr_pages;
+ if (max_ram_pfn > max_low_pfn)
+ max_ram_pfn = max_low_pfn;
+
+ pgd_idx = pgd_index(PAGE_OFFSET);
+ pgd = pgd_base + pgd_idx;
+ pfn = 0;
+ pmd_idx = pmd_index(PAGE_OFFSET);
+ pte_ofs = pte_index(PAGE_OFFSET);
+
+ for (; pgd_idx < PTRS_PER_PGD; pgd++, pgd_idx++) {
+#ifdef CONFIG_XEN
+ /*
+ * Native linux hasn't PAE-paging enabled yet at this
+ * point. When running as xen domain we are in PAE
+ * mode already, thus we can't simply hook a empty
+ * pmd. That would kill the mappings we are currently
+ * using ...
+ */
+ pmd = pmd_offset(pud_offset(pgd, PAGE_OFFSET), PAGE_OFFSET);
+#else
+ pmd = one_md_table_init(pgd);
+#endif
+ if (pfn >= max_low_pfn)
+ continue;
+ pmd += pmd_idx;
+ for (; pmd_idx < PTRS_PER_PMD && pfn < max_low_pfn; pmd++, pmd_idx++) {
+ unsigned int address = pfn * PAGE_SIZE + PAGE_OFFSET;
+ if (address >= hypervisor_virt_start)
+ continue;
+
+ /* Map with big pages if possible, otherwise create normal page tables. */
+ if (cpu_has_pse) {
+ unsigned int address2 = (pfn + PTRS_PER_PTE - 1) * PAGE_SIZE + PAGE_OFFSET + PAGE_SIZE-1;
+
+ if (is_kernel_text(address) || is_kernel_text(address2))
+ set_pmd(pmd, pfn_pmd(pfn, PAGE_KERNEL_LARGE_EXEC));
+ else
+ set_pmd(pmd, pfn_pmd(pfn, PAGE_KERNEL_LARGE));
+ pfn += PTRS_PER_PTE;
+ } else {
+ pte = one_page_table_init(pmd);
+
+ pte += pte_ofs;
+ for (; pte_ofs < PTRS_PER_PTE && pfn < max_low_pfn; pte++, pfn++, pte_ofs++) {
+ /* XEN: Only map initial RAM allocation. */
+ if ((pfn >= max_ram_pfn) || pte_present(*pte))
+ continue;
+ if (is_kernel_text(address))
+ set_pte(pte, pfn_pte(pfn, PAGE_KERNEL_EXEC));
+ else
+ set_pte(pte, pfn_pte(pfn, PAGE_KERNEL));
+ }
+ pte_ofs = 0;
+ }
+ }
+ pmd_idx = 0;
+ }
+}
+
+#ifndef CONFIG_XEN
+
+static inline int page_kills_ppro(unsigned long pagenr)
+{
+ if (pagenr >= 0x70000 && pagenr <= 0x7003F)
+ return 1;
+ return 0;
+}
+
+#else
+
+#define page_kills_ppro(p) 0
+
+#endif
+
+extern int is_available_memory(efi_memory_desc_t *);
+
+int page_is_ram(unsigned long pagenr)
+{
+ int i;
+ unsigned long addr, end;
+
+ if (efi_enabled) {
+ efi_memory_desc_t *md;
+ void *p;
+
+ for (p = memmap.map; p < memmap.map_end; p += memmap.desc_size) {
+ md = p;
+ if (!is_available_memory(md))
+ continue;
+ addr = (md->phys_addr+PAGE_SIZE-1) >> PAGE_SHIFT;
+ end = (md->phys_addr + (md->num_pages << EFI_PAGE_SHIFT)) >> PAGE_SHIFT;
+
+ if ((pagenr >= addr) && (pagenr < end))
+ return 1;
+ }
+ return 0;
+ }
+
+ for (i = 0; i < e820.nr_map; i++) {
+
+ if (e820.map[i].type != E820_RAM) /* not usable memory */
+ continue;
+ /*
+ * !!!FIXME!!! Some BIOSen report areas as RAM that
+ * are not. Notably the 640->1Mb area. We need a sanity
+ * check here.
+ */
+ addr = (e820.map[i].addr+PAGE_SIZE-1) >> PAGE_SHIFT;
+ end = (e820.map[i].addr+e820.map[i].size) >> PAGE_SHIFT;
+ if ((pagenr >= addr) && (pagenr < end))
+ return 1;
+ }
+ return 0;
+}
+
+#ifdef CONFIG_HIGHMEM
+pte_t *kmap_pte;
+pgprot_t kmap_prot;
+
+#define kmap_get_fixmap_pte(vaddr) \
+ pte_offset_kernel(pmd_offset(pud_offset(pgd_offset_k(vaddr), vaddr), (vaddr)), (vaddr))
+
+static void __init kmap_init(void)
+{
+ unsigned long kmap_vstart;
+
+ /* cache the first kmap pte */
+ kmap_vstart = __fix_to_virt(FIX_KMAP_BEGIN);
+ kmap_pte = kmap_get_fixmap_pte(kmap_vstart);
+
+ kmap_prot = PAGE_KERNEL;
+}
+
+static void __init permanent_kmaps_init(pgd_t *pgd_base)
+{
+ pgd_t *pgd;
+ pud_t *pud;
+ pmd_t *pmd;
+ pte_t *pte;
+ unsigned long vaddr;
+
+ vaddr = PKMAP_BASE;
+ page_table_range_init(vaddr, vaddr + PAGE_SIZE*LAST_PKMAP, pgd_base);
+
+ pgd = swapper_pg_dir + pgd_index(vaddr);
+ pud = pud_offset(pgd, vaddr);
+ pmd = pmd_offset(pud, vaddr);
+ pte = pte_offset_kernel(pmd, vaddr);
+ pkmap_page_table = pte;
+}
+
+static void __meminit free_new_highpage(struct page *page, int pfn)
+{
+ init_page_count(page);
+ if (pfn < xen_start_info->nr_pages)
+ __free_page(page);
+ totalhigh_pages++;
+}
+
+void __init add_one_highpage_init(struct page *page, int pfn, int bad_ppro)
+{
+ if (page_is_ram(pfn) && !(bad_ppro && page_kills_ppro(pfn))) {
+ ClearPageReserved(page);
+ free_new_highpage(page, pfn);
+ } else
+ SetPageReserved(page);
+}
+
+static int add_one_highpage_hotplug(struct page *page, unsigned long pfn)
+{
+ free_new_highpage(page, pfn);
+ totalram_pages++;
+#ifdef CONFIG_FLATMEM
+ max_mapnr = max(pfn, max_mapnr);
+#endif
+ num_physpages++;
+ return 0;
+}
+
+/*
+ * Not currently handling the NUMA case.
+ * Assuming single node and all memory that
+ * has been added dynamically that would be
+ * onlined here is in HIGHMEM
+ */
+void online_page(struct page *page)
+{
+ ClearPageReserved(page);
+ add_one_highpage_hotplug(page, page_to_pfn(page));
+}
+
+
+#ifdef CONFIG_NUMA
+extern void set_highmem_pages_init(int);
+#else
+static void __init set_highmem_pages_init(int bad_ppro)
+{
+ int pfn;
+ for (pfn = highstart_pfn; pfn < highend_pfn; pfn++)
+ add_one_highpage_init(pfn_to_page(pfn), pfn, bad_ppro);
+ totalram_pages += totalhigh_pages;
+}
+#endif /* CONFIG_FLATMEM */
+
+#else
+#define kmap_init() do { } while (0)
+#define permanent_kmaps_init(pgd_base) do { } while (0)
+#define set_highmem_pages_init(bad_ppro) do { } while (0)
+#endif /* CONFIG_HIGHMEM */
+
+unsigned long long __PAGE_KERNEL = _PAGE_KERNEL;
+EXPORT_SYMBOL(__PAGE_KERNEL);
+unsigned long long __PAGE_KERNEL_EXEC = _PAGE_KERNEL_EXEC;
+
+#ifdef CONFIG_NUMA
+extern void __init remap_numa_kva(void);
+#else
+#define remap_numa_kva() do {} while (0)
+#endif
+
+pgd_t *swapper_pg_dir;
+
+static void __init pagetable_init (void)
+{
+ unsigned long vaddr;
+ pgd_t *pgd_base = (pgd_t *)xen_start_info->pt_base;
+
+ /* Enable PSE if available */
+ if (cpu_has_pse) {
+ set_in_cr4(X86_CR4_PSE);
+ }
+
+ /* Enable PGE if available */
+ if (cpu_has_pge) {
+ set_in_cr4(X86_CR4_PGE);
+ __PAGE_KERNEL |= _PAGE_GLOBAL;
+ __PAGE_KERNEL_EXEC |= _PAGE_GLOBAL;
+ }
+
+ kernel_physical_mapping_init(pgd_base);
+ remap_numa_kva();
+
+ /*
+ * Fixed mappings, only the page table structure has to be
+ * created - mappings will be set by set_fixmap():
+ */
+ vaddr = __fix_to_virt(__end_of_fixed_addresses - 1) & PMD_MASK;
+ page_table_range_init(vaddr, hypervisor_virt_start, pgd_base);
+
+ permanent_kmaps_init(pgd_base);
+}
+
+#if defined(CONFIG_SOFTWARE_SUSPEND) || defined(CONFIG_ACPI_SLEEP)
+/*
+ * Swap suspend & friends need this for resume because things like the intel-agp
+ * driver might have split up a kernel 4MB mapping.
+ */
+char __nosavedata swsusp_pg_dir[PAGE_SIZE]
+ __attribute__ ((aligned (PAGE_SIZE)));
+
+static inline void save_pg_dir(void)
+{
+ memcpy(swsusp_pg_dir, swapper_pg_dir, PAGE_SIZE);
+}
+#else
+static inline void save_pg_dir(void)
+{
+}
+#endif
+
+void zap_low_mappings (void)
+{
+ int i;
+
+ save_pg_dir();
+
+ /*
+ * Zap initial low-memory mappings.
+ *
+ * Note that "pgd_clear()" doesn't do it for
+ * us, because pgd_clear() is a no-op on i386.
+ */
+ for (i = 0; i < USER_PTRS_PER_PGD; i++)
+#if defined(CONFIG_X86_PAE) && !defined(CONFIG_XEN)
+ set_pgd(swapper_pg_dir+i, __pgd(1 + __pa(empty_zero_page)));
+#else
+ set_pgd(swapper_pg_dir+i, __pgd(0));
+#endif
+ flush_tlb_all();
+}
+
+static int disable_nx __initdata = 0;
+u64 __supported_pte_mask __read_mostly = ~_PAGE_NX;
+EXPORT_SYMBOL(__supported_pte_mask);
+
+/*
+ * noexec = on|off
+ *
+ * Control non executable mappings.
+ *
+ * on Enable
+ * off Disable
+ */
+void __init noexec_setup(const char *str)
+{
+ if (!strncmp(str, "on",2) && cpu_has_nx) {
+ __supported_pte_mask |= _PAGE_NX;
+ disable_nx = 0;
+ } else if (!strncmp(str,"off",3)) {
+ disable_nx = 1;
+ __supported_pte_mask &= ~_PAGE_NX;
+ }
+}
+
+int nx_enabled = 0;
+#ifdef CONFIG_X86_PAE
+
+static void __init set_nx(void)
+{
+ unsigned int v[4], l, h;
+
+ if (cpu_has_pae && (cpuid_eax(0x80000000) > 0x80000001)) {
+ cpuid(0x80000001, &v[0], &v[1], &v[2], &v[3]);
+ if ((v[3] & (1 << 20)) && !disable_nx) {
+ rdmsr(MSR_EFER, l, h);
+ l |= EFER_NX;
+ wrmsr(MSR_EFER, l, h);
+ nx_enabled = 1;
+ __supported_pte_mask |= _PAGE_NX;
+ }
+ }
+}
+
+/*
+ * Enables/disables executability of a given kernel page and
+ * returns the previous setting.
+ */
+int __init set_kernel_exec(unsigned long vaddr, int enable)
+{
+ pte_t *pte;
+ int ret = 1;
+
+ if (!nx_enabled)
+ goto out;
+
+ pte = lookup_address(vaddr);
+ BUG_ON(!pte);
+
+ if (!pte_exec_kernel(*pte))
+ ret = 0;
+
+ if (enable)
+ pte->pte_high &= ~(1 << (_PAGE_BIT_NX - 32));
+ else
+ pte->pte_high |= 1 << (_PAGE_BIT_NX - 32);
+ __flush_tlb_all();
+out:
+ return ret;
+}
+
+#endif
+
+/*
+ * paging_init() sets up the page tables - note that the first 8MB are
+ * already mapped by head.S.
+ *
+ * This routines also unmaps the page at virtual kernel address 0, so
+ * that we can trap those pesky NULL-reference errors in the kernel.
+ */
+void __init paging_init(void)
+{
+ int i;
+
+#ifdef CONFIG_X86_PAE
+ set_nx();
+ if (nx_enabled)
+ printk("NX (Execute Disable) protection: active\n");
+#endif
+
+ pagetable_init();
+
+#if defined(CONFIG_X86_PAE) && !defined(CONFIG_XEN)
+ /*
+ * We will bail out later - printk doesn't work right now so
+ * the user would just see a hanging kernel.
+ * when running as xen domain we are already in PAE mode at
+ * this point.
+ */
+ if (cpu_has_pae)
+ set_in_cr4(X86_CR4_PAE);
+#endif
+ __flush_tlb_all();
+
+ kmap_init();
+
+ /* Switch to the real shared_info page, and clear the
+ * dummy page. */
+ set_fixmap(FIX_SHARED_INFO, xen_start_info->shared_info);
+ HYPERVISOR_shared_info = (shared_info_t *)fix_to_virt(FIX_SHARED_INFO);
+ memset(empty_zero_page, 0, sizeof(empty_zero_page));
+
+ /* Setup mapping of lower 1st MB */
+ for (i = 0; i < NR_FIX_ISAMAPS; i++)
+ if (is_initial_xendomain())
+ set_fixmap(FIX_ISAMAP_BEGIN - i, i * PAGE_SIZE);
+ else
+ __set_fixmap(FIX_ISAMAP_BEGIN - i,
+ virt_to_machine(empty_zero_page),
+ PAGE_KERNEL_RO);
+}
+
+/*
+ * Test if the WP bit works in supervisor mode. It isn't supported on 386's
+ * and also on some strange 486's (NexGen etc.). All 586+'s are OK. This
+ * used to involve black magic jumps to work around some nasty CPU bugs,
+ * but fortunately the switch to using exceptions got rid of all that.
+ */
+
+static void __init test_wp_bit(void)
+{
+ printk("Checking if this processor honours the WP bit even in supervisor mode... ");
+
+ /* Any page-aligned address will do, the test is non-destructive */
+ __set_fixmap(FIX_WP_TEST, __pa(&swapper_pg_dir), PAGE_READONLY);
+ boot_cpu_data.wp_works_ok = do_test_wp_bit();
+ clear_fixmap(FIX_WP_TEST);
+
+ if (!boot_cpu_data.wp_works_ok) {
+ printk("No.\n");
+#ifdef CONFIG_X86_WP_WORKS_OK
+ panic("This kernel doesn't support CPU's with broken WP. Recompile it for a 386!");
+#endif
+ } else {
+ printk("Ok.\n");
+ }
+}
+
+static void __init set_max_mapnr_init(void)
+{
+#ifdef CONFIG_HIGHMEM
+ num_physpages = highend_pfn;
+#else
+ num_physpages = max_low_pfn;
+#endif
+#ifdef CONFIG_FLATMEM
+ max_mapnr = num_physpages;
+#endif
+}
+
+static struct kcore_list kcore_mem, kcore_vmalloc;
+
+void __init mem_init(void)
+{
+ extern int ppro_with_ram_bug(void);
+ int codesize, reservedpages, datasize, initsize;
+ int tmp;
+ int bad_ppro;
+ unsigned long pfn;
+
+#if defined(CONFIG_SWIOTLB)
+ swiotlb_init();
+#endif
+
+#ifdef CONFIG_FLATMEM
+ if (!mem_map)
+ BUG();
+#endif
+
+ bad_ppro = ppro_with_ram_bug();
+
+#ifdef CONFIG_HIGHMEM
+ /* check that fixmap and pkmap do not overlap */
+ if (PKMAP_BASE+LAST_PKMAP*PAGE_SIZE >= FIXADDR_START) {
+ printk(KERN_ERR "fixmap and kmap areas overlap - this will crash\n");
+ printk(KERN_ERR "pkstart: %lxh pkend: %lxh fixstart %lxh\n",
+ PKMAP_BASE, PKMAP_BASE+LAST_PKMAP*PAGE_SIZE, FIXADDR_START);
+ BUG();
+ }
+#endif
+
+ set_max_mapnr_init();
+
+#ifdef CONFIG_HIGHMEM
+ high_memory = (void *) __va(highstart_pfn * PAGE_SIZE - 1) + 1;
+#else
+ high_memory = (void *) __va(max_low_pfn * PAGE_SIZE - 1) + 1;
+#endif
+ printk("vmalloc area: %lx-%lx, maxmem %lx\n",
+ VMALLOC_START,VMALLOC_END,MAXMEM);
+ BUG_ON(VMALLOC_START > VMALLOC_END);
+
+ /* this will put all low memory onto the freelists */
+ totalram_pages += free_all_bootmem();
+ /* XEN: init and count low-mem pages outside initial allocation. */
+ for (pfn = xen_start_info->nr_pages; pfn < max_low_pfn; pfn++) {
+ ClearPageReserved(pfn_to_page(pfn));
+ init_page_count(pfn_to_page(pfn));
+ totalram_pages++;
+ }
+
+ reservedpages = 0;
+ for (tmp = 0; tmp < max_low_pfn; tmp++)
+ /*
+ * Only count reserved RAM pages
+ */
+ if (page_is_ram(tmp) && PageReserved(pfn_to_page(tmp)))
+ reservedpages++;
+
+ set_highmem_pages_init(bad_ppro);
+
+ codesize = (unsigned long) &_etext - (unsigned long) &_text;
+ datasize = (unsigned long) &_edata - (unsigned long) &_etext;
+ initsize = (unsigned long) &__init_end - (unsigned long) &__init_begin;
+
+ kclist_add(&kcore_mem, __va(0), max_low_pfn << PAGE_SHIFT);
+ kclist_add(&kcore_vmalloc, (void *)VMALLOC_START,
+ VMALLOC_END-VMALLOC_START);
+
+ printk(KERN_INFO "Memory: %luk/%luk available (%dk kernel code, %dk reserved, %dk data, %dk init, %ldk highmem)\n",
+ (unsigned long) nr_free_pages() << (PAGE_SHIFT-10),
+ num_physpages << (PAGE_SHIFT-10),
+ codesize >> 10,
+ reservedpages << (PAGE_SHIFT-10),
+ datasize >> 10,
+ initsize >> 10,
+ (unsigned long) (totalhigh_pages << (PAGE_SHIFT-10))
+ );
+
+#ifdef CONFIG_X86_PAE
+ if (!cpu_has_pae)
+ panic("cannot execute a PAE-enabled kernel on a PAE-less CPU!");
+#endif
+ if (boot_cpu_data.wp_works_ok < 0)
+ test_wp_bit();
+
+ /*
+ * Subtle. SMP is doing it's boot stuff late (because it has to
+ * fork idle threads) - but it also needs low mappings for the
+ * protected-mode entry to work. We zap these entries only after
+ * the WP-bit has been tested.
+ */
+#ifndef CONFIG_SMP
+ zap_low_mappings();
+#endif
+
+ set_bit(PG_pinned, &virt_to_page(init_mm.pgd)->flags);
+}
+
+/*
+ * this is for the non-NUMA, single node SMP system case.
+ * Specifically, in the case of x86, we will always add
+ * memory to the highmem for now.
+ */
+#ifdef CONFIG_MEMORY_HOTPLUG
+#ifndef CONFIG_NEED_MULTIPLE_NODES
+int arch_add_memory(int nid, u64 start, u64 size)
+{
+ struct pglist_data *pgdata = &contig_page_data;
+ struct zone *zone = pgdata->node_zones + MAX_NR_ZONES-1;
+ unsigned long start_pfn = start >> PAGE_SHIFT;
+ unsigned long nr_pages = size >> PAGE_SHIFT;
+
+ return __add_pages(zone, start_pfn, nr_pages);
+}
+
+int remove_memory(u64 start, u64 size)
+{
+ return -EINVAL;
+}
+#endif
+#endif
+
+kmem_cache_t *pgd_cache;
+kmem_cache_t *pmd_cache;
+
+void __init pgtable_cache_init(void)
+{
+ if (PTRS_PER_PMD > 1) {
+ pmd_cache = kmem_cache_create("pmd",
+ PTRS_PER_PMD*sizeof(pmd_t),
+ PTRS_PER_PMD*sizeof(pmd_t),
+ 0,
+ pmd_ctor,
+ NULL);
+ if (!pmd_cache)
+ panic("pgtable_cache_init(): cannot create pmd cache");
+ }
+ pgd_cache = kmem_cache_create("pgd",
+#ifndef CONFIG_XEN
+ PTRS_PER_PGD*sizeof(pgd_t),
+ PTRS_PER_PGD*sizeof(pgd_t),
+#else
+ PAGE_SIZE,
+ PAGE_SIZE,
+#endif
+ 0,
+ pgd_ctor,
+ PTRS_PER_PMD == 1 ? pgd_dtor : NULL);
+ if (!pgd_cache)
+ panic("pgtable_cache_init(): Cannot create pgd cache");
+}
+
+/*
+ * This function cannot be __init, since exceptions don't work in that
+ * section. Put this after the callers, so that it cannot be inlined.
+ */
+static int noinline do_test_wp_bit(void)
+{
+ char tmp_reg;
+ int flag;
+
+ __asm__ __volatile__(
+ " movb %0,%1 \n"
+ "1: movb %1,%0 \n"
+ " xorl %2,%2 \n"
+ "2: \n"
+ ".section __ex_table,\"a\"\n"
+ " .align 4 \n"
+ " .long 1b,2b \n"
+ ".previous \n"
+ :"=m" (*(char *)fix_to_virt(FIX_WP_TEST)),
+ "=q" (tmp_reg),
+ "=r" (flag)
+ :"2" (1)
+ :"memory");
+
+ return flag;
+}
+
+#ifdef CONFIG_DEBUG_RODATA
+
+void mark_rodata_ro(void)
+{
+ unsigned long addr = (unsigned long)__start_rodata;
+
+ for (; addr < (unsigned long)__end_rodata; addr += PAGE_SIZE)
+ change_page_attr(virt_to_page(addr), 1, PAGE_KERNEL_RO);
+
+ printk("Write protecting the kernel read-only data: %uk\n",
+ (__end_rodata - __start_rodata) >> 10);
+
+ /*
+ * change_page_attr() requires a global_flush_tlb() call after it.
+ * We do this after the printk so that if something went wrong in the
+ * change, the printk gets out at least to give a better debug hint
+ * of who is the culprit.
+ */
+ global_flush_tlb();
+}
+#endif
+
+void free_init_pages(char *what, unsigned long begin, unsigned long end)
+{
+ unsigned long addr;
+
+ for (addr = begin; addr < end; addr += PAGE_SIZE) {
+ ClearPageReserved(virt_to_page(addr));
+ init_page_count(virt_to_page(addr));
+ memset((void *)addr, POISON_FREE_INITMEM, PAGE_SIZE);
+ free_page(addr);
+ totalram_pages++;
+ }
+ printk(KERN_INFO "Freeing %s: %ldk freed\n", what, (end - begin) >> 10);
+}
+
+void free_initmem(void)
+{
+ free_init_pages("unused kernel memory",
+ (unsigned long)(&__init_begin),
+ (unsigned long)(&__init_end));
+}
+
+#ifdef CONFIG_BLK_DEV_INITRD
+void free_initrd_mem(unsigned long start, unsigned long end)
+{
+ free_init_pages("initrd memory", start, end);
+}
+#endif
+
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/mm/ioremap-xen.c 2010-09-16 13:31:46.000000000 +0200
@@ -0,0 +1,439 @@
+/*
+ * arch/i386/mm/ioremap.c
+ *
+ * Re-map IO memory to kernel address space so that we can access it.
+ * This is needed for high PCI addresses that aren't mapped in the
+ * 640k-1MB IO memory area on PC's
+ *
+ * (C) Copyright 1995 1996 Linus Torvalds
+ */
+
+#include <linux/vmalloc.h>
+#include <linux/init.h>
+#include <linux/slab.h>
+#include <linux/module.h>
+#include <asm/io.h>
+#include <asm/fixmap.h>
+#include <asm/cacheflush.h>
+#include <asm/tlbflush.h>
+#include <asm/pgtable.h>
+#include <asm/pgalloc.h>
+
+#define ISA_START_ADDRESS 0x0
+#define ISA_END_ADDRESS 0x100000
+
+static int direct_remap_area_pte_fn(pte_t *pte,
+ struct page *pmd_page,
+ unsigned long address,
+ void *data)
+{
+ mmu_update_t **v = (mmu_update_t **)data;
+
+ BUG_ON(!pte_none(*pte));
+
+ (*v)->ptr = ((u64)pfn_to_mfn(page_to_pfn(pmd_page)) <<
+ PAGE_SHIFT) | ((unsigned long)pte & ~PAGE_MASK);
+ (*v)++;
+
+ return 0;
+}
+
+static int __direct_remap_pfn_range(struct mm_struct *mm,
+ unsigned long address,
+ unsigned long mfn,
+ unsigned long size,
+ pgprot_t prot,
+ domid_t domid)
+{
+ int rc = 0;
+ unsigned long i, start_address;
+ mmu_update_t *u, *v, *w;
+
+ u = v = w = (mmu_update_t *)__get_free_page(GFP_KERNEL|__GFP_REPEAT);
+ if (u == NULL)
+ return -ENOMEM;
+
+ start_address = address;
+
+ flush_cache_all();
+
+ for (i = 0; i < size; i += PAGE_SIZE) {
+ if ((v - u) == (PAGE_SIZE / sizeof(mmu_update_t))) {
+ /* Flush a full batch after filling in the PTE ptrs. */
+ rc = apply_to_page_range(mm, start_address,
+ address - start_address,
+ direct_remap_area_pte_fn, &w);
+ if (rc)
+ goto out;
+ rc = HYPERVISOR_mmu_update(u, v - u, NULL, domid);
+ if (rc < 0)
+ goto out;
+ v = w = u;
+ start_address = address;
+ }
+
+ /*
+ * Fill in the machine address: PTE ptr is done later by
+ * apply_to_page_range().
+ */
+ v->val = __pte_val(pfn_pte_ma(mfn, prot)) | _PAGE_IO;
+
+ mfn++;
+ address += PAGE_SIZE;
+ v++;
+ }
+
+ if (v != u) {
+ /* Final batch. */
+ rc = apply_to_page_range(mm, start_address,
+ address - start_address,
+ direct_remap_area_pte_fn, &w);
+ if (rc)
+ goto out;
+ rc = HYPERVISOR_mmu_update(u, v - u, NULL, domid);
+ }
+
+ out:
+ flush_tlb_all();
+
+ free_page((unsigned long)u);
+
+ return rc;
+}
+
+int direct_remap_pfn_range(struct vm_area_struct *vma,
+ unsigned long address,
+ unsigned long mfn,
+ unsigned long size,
+ pgprot_t prot,
+ domid_t domid)
+{
+ if (xen_feature(XENFEAT_auto_translated_physmap))
+ return remap_pfn_range(vma, address, mfn, size, prot);
+
+ if (domid == DOMID_SELF)
+ return -EINVAL;
+
+ vma->vm_flags |= VM_IO | VM_RESERVED;
+
+ vma->vm_mm->context.has_foreign_mappings = 1;
+
+ return __direct_remap_pfn_range(
+ vma->vm_mm, address, mfn, size, prot, domid);
+}
+EXPORT_SYMBOL(direct_remap_pfn_range);
+
+int direct_kernel_remap_pfn_range(unsigned long address,
+ unsigned long mfn,
+ unsigned long size,
+ pgprot_t prot,
+ domid_t domid)
+{
+ return __direct_remap_pfn_range(
+ &init_mm, address, mfn, size, prot, domid);
+}
+EXPORT_SYMBOL(direct_kernel_remap_pfn_range);
+
+static int lookup_pte_fn(
+ pte_t *pte, struct page *pmd_page, unsigned long addr, void *data)
+{
+ uint64_t *ptep = (uint64_t *)data;
+ if (ptep)
+ *ptep = ((uint64_t)pfn_to_mfn(page_to_pfn(pmd_page)) <<
+ PAGE_SHIFT) | ((unsigned long)pte & ~PAGE_MASK);
+ return 0;
+}
+
+int create_lookup_pte_addr(struct mm_struct *mm,
+ unsigned long address,
+ uint64_t *ptep)
+{
+ return apply_to_page_range(mm, address, PAGE_SIZE,
+ lookup_pte_fn, ptep);
+}
+
+EXPORT_SYMBOL(create_lookup_pte_addr);
+
+static int noop_fn(
+ pte_t *pte, struct page *pmd_page, unsigned long addr, void *data)
+{
+ return 0;
+}
+
+int touch_pte_range(struct mm_struct *mm,
+ unsigned long address,
+ unsigned long size)
+{
+ return apply_to_page_range(mm, address, size, noop_fn, NULL);
+}
+
+EXPORT_SYMBOL(touch_pte_range);
+
+/*
+ * Does @address reside within a non-highmem page that is local to this virtual
+ * machine (i.e., not an I/O page, nor a memory page belonging to another VM).
+ * See the comment that accompanies mfn_to_local_pfn() in page.h to understand
+ * why this works.
+ */
+static inline int is_local_lowmem(unsigned long address)
+{
+ extern unsigned long max_low_pfn;
+ return (mfn_to_local_pfn(address >> PAGE_SHIFT) < max_low_pfn);
+}
+
+/*
+ * Generic mapping function (not visible outside):
+ */
+
+/*
+ * Remap an arbitrary physical address space into the kernel virtual
+ * address space. Needed when the kernel wants to access high addresses
+ * directly.
+ *
+ * NOTE! We need to allow non-page-aligned mappings too: we will obviously
+ * have to convert them into an offset in a page-aligned mapping, but the
+ * caller shouldn't need to know that small detail.
+ */
+void __iomem * __ioremap(unsigned long phys_addr, unsigned long size, unsigned long flags)
+{
+ void __iomem * addr;
+ struct vm_struct * area;
+ unsigned long offset, last_addr;
+ domid_t domid = DOMID_IO;
+
+ /* Don't allow wraparound or zero size */
+ last_addr = phys_addr + size - 1;
+ if (!size || last_addr < phys_addr)
+ return NULL;
+
+ /*
+ * Don't remap the low PCI/ISA area, it's always mapped..
+ */
+ if (is_initial_xendomain() &&
+ phys_addr >= ISA_START_ADDRESS && last_addr < ISA_END_ADDRESS)
+ return (void __iomem *) isa_bus_to_virt(phys_addr);
+
+ /*
+ * Don't allow anybody to remap normal RAM that we're using..
+ */
+ if (is_local_lowmem(phys_addr)) {
+ char *t_addr, *t_end;
+ struct page *page;
+
+ t_addr = bus_to_virt(phys_addr);
+ t_end = t_addr + (size - 1);
+
+ for(page = virt_to_page(t_addr); page <= virt_to_page(t_end); page++)
+ if(!PageReserved(page))
+ return NULL;
+
+ domid = DOMID_SELF;
+ }
+
+ /*
+ * Mappings have to be page-aligned
+ */
+ offset = phys_addr & ~PAGE_MASK;
+ phys_addr &= PAGE_MASK;
+ size = PAGE_ALIGN(last_addr+1) - phys_addr;
+
+ /*
+ * Ok, go for it..
+ */
+ area = get_vm_area(size, VM_IOREMAP | (flags << 20));
+ if (!area)
+ return NULL;
+ area->phys_addr = phys_addr;
+ addr = (void __iomem *) area->addr;
+ flags |= _KERNPG_TABLE;
+ if (__direct_remap_pfn_range(&init_mm, (unsigned long)addr,
+ phys_addr>>PAGE_SHIFT,
+ size, __pgprot(flags), domid)) {
+ vunmap((void __force *) addr);
+ return NULL;
+ }
+ return (void __iomem *) (offset + (char __iomem *)addr);
+}
+EXPORT_SYMBOL(__ioremap);
+
+/**
+ * ioremap_nocache - map bus memory into CPU space
+ * @offset: bus address of the memory
+ * @size: size of the resource to map
+ *
+ * ioremap_nocache performs a platform specific sequence of operations to
+ * make bus memory CPU accessible via the readb/readw/readl/writeb/
+ * writew/writel functions and the other mmio helpers. The returned
+ * address is not guaranteed to be usable directly as a virtual
+ * address.
+ *
+ * This version of ioremap ensures that the memory is marked uncachable
+ * on the CPU as well as honouring existing caching rules from things like
+ * the PCI bus. Note that there are other caches and buffers on many
+ * busses. In particular driver authors should read up on PCI writes
+ *
+ * It's useful if some control registers are in such an area and
+ * write combining or read caching is not desirable:
+ *
+ * Must be freed with iounmap.
+ */
+
+void __iomem *ioremap_nocache (unsigned long phys_addr, unsigned long size)
+{
+ unsigned long last_addr;
+ void __iomem *p = __ioremap(phys_addr, size, _PAGE_PCD);
+ if (!p)
+ return p;
+
+ /* Guaranteed to be > phys_addr, as per __ioremap() */
+ last_addr = phys_addr + size - 1;
+
+ if (is_local_lowmem(last_addr)) {
+ struct page *ppage = virt_to_page(bus_to_virt(phys_addr));
+ unsigned long npages;
+
+ phys_addr &= PAGE_MASK;
+
+ /* This might overflow and become zero.. */
+ last_addr = PAGE_ALIGN(last_addr);
+
+ /* .. but that's ok, because modulo-2**n arithmetic will make
+ * the page-aligned "last - first" come out right.
+ */
+ npages = (last_addr - phys_addr) >> PAGE_SHIFT;
+
+ if (change_page_attr(ppage, npages, PAGE_KERNEL_NOCACHE) < 0) {
+ iounmap(p);
+ p = NULL;
+ }
+ global_flush_tlb();
+ }
+
+ return p;
+}
+EXPORT_SYMBOL(ioremap_nocache);
+
+/**
+ * iounmap - Free a IO remapping
+ * @addr: virtual address from ioremap_*
+ *
+ * Caller must ensure there is only one unmapping for the same pointer.
+ */
+void iounmap(volatile void __iomem *addr)
+{
+ struct vm_struct *p, *o;
+
+ if ((void __force *)addr <= high_memory)
+ return;
+
+ /*
+ * __ioremap special-cases the PCI/ISA range by not instantiating a
+ * vm_area and by simply returning an address into the kernel mapping
+ * of ISA space. So handle that here.
+ */
+ if ((unsigned long) addr >= fix_to_virt(FIX_ISAMAP_BEGIN))
+ return;
+
+ addr = (volatile void __iomem *)(PAGE_MASK & (unsigned long __force)addr);
+
+ /* Use the vm area unlocked, assuming the caller
+ ensures there isn't another iounmap for the same address
+ in parallel. Reuse of the virtual address is prevented by
+ leaving it in the global lists until we're done with it.
+ cpa takes care of the direct mappings. */
+ read_lock(&vmlist_lock);
+ for (p = vmlist; p; p = p->next) {
+ if (p->addr == addr)
+ break;
+ }
+ read_unlock(&vmlist_lock);
+
+ if (!p) {
+ printk("iounmap: bad address %p\n", addr);
+ dump_stack();
+ return;
+ }
+
+ /* Reset the direct mapping. Can block */
+ if ((p->flags >> 20) && is_local_lowmem(p->phys_addr)) {
+ /* p->size includes the guard page, but cpa doesn't like that */
+ change_page_attr(virt_to_page(bus_to_virt(p->phys_addr)),
+ (p->size - PAGE_SIZE) >> PAGE_SHIFT,
+ PAGE_KERNEL);
+ global_flush_tlb();
+ }
+
+ /* Finally remove it */
+ o = remove_vm_area((void *)addr);
+ BUG_ON(p != o || o == NULL);
+ kfree(p);
+}
+EXPORT_SYMBOL(iounmap);
+
+void __init *bt_ioremap(unsigned long phys_addr, unsigned long size)
+{
+ unsigned long offset, last_addr;
+ unsigned int nrpages;
+ enum fixed_addresses idx;
+
+ /* Don't allow wraparound or zero size */
+ last_addr = phys_addr + size - 1;
+ if (!size || last_addr < phys_addr)
+ return NULL;
+
+ /*
+ * Don't remap the low PCI/ISA area, it's always mapped..
+ */
+ if (is_initial_xendomain() &&
+ phys_addr >= ISA_START_ADDRESS && last_addr < ISA_END_ADDRESS)
+ return isa_bus_to_virt(phys_addr);
+
+ /*
+ * Mappings have to be page-aligned
+ */
+ offset = phys_addr & ~PAGE_MASK;
+ phys_addr &= PAGE_MASK;
+ size = PAGE_ALIGN(last_addr) - phys_addr;
+
+ /*
+ * Mappings have to fit in the FIX_BTMAP area.
+ */
+ nrpages = size >> PAGE_SHIFT;
+ if (nrpages > NR_FIX_BTMAPS)
+ return NULL;
+
+ /*
+ * Ok, go for it..
+ */
+ idx = FIX_BTMAP_BEGIN;
+ while (nrpages > 0) {
+ set_fixmap(idx, phys_addr);
+ phys_addr += PAGE_SIZE;
+ --idx;
+ --nrpages;
+ }
+ return (void*) (offset + fix_to_virt(FIX_BTMAP_BEGIN));
+}
+
+void __init bt_iounmap(void *addr, unsigned long size)
+{
+ unsigned long virt_addr;
+ unsigned long offset;
+ unsigned int nrpages;
+ enum fixed_addresses idx;
+
+ virt_addr = (unsigned long)addr;
+ if (virt_addr < fix_to_virt(FIX_BTMAP_BEGIN))
+ return;
+ if (virt_addr >= fix_to_virt(FIX_ISAMAP_BEGIN))
+ return;
+ offset = virt_addr & ~PAGE_MASK;
+ nrpages = PAGE_ALIGN(offset + size - 1) >> PAGE_SHIFT;
+
+ idx = FIX_BTMAP_BEGIN;
+ while (nrpages > 0) {
+ clear_fixmap(idx);
+ --idx;
+ --nrpages;
+ }
+}
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/mm/pgtable_32-xen.c 2010-09-23 15:39:04.000000000 +0200
@@ -0,0 +1,738 @@
+/*
+ * linux/arch/i386/mm/pgtable.c
+ */
+
+#include <linux/sched.h>
+#include <linux/kernel.h>
+#include <linux/errno.h>
+#include <linux/mm.h>
+#include <linux/swap.h>
+#include <linux/smp.h>
+#include <linux/highmem.h>
+#include <linux/slab.h>
+#include <linux/pagemap.h>
+#include <linux/spinlock.h>
+#include <linux/module.h>
+
+#include <asm/system.h>
+#include <asm/pgtable.h>
+#include <asm/pgalloc.h>
+#include <asm/fixmap.h>
+#include <asm/e820.h>
+#include <asm/tlb.h>
+#include <asm/tlbflush.h>
+#include <asm/io.h>
+#include <asm/mmu_context.h>
+
+#include <xen/features.h>
+#include <asm/hypervisor.h>
+
+static void pgd_test_and_unpin(pgd_t *pgd);
+
+void show_mem(void)
+{
+ int total = 0, reserved = 0;
+ int shared = 0, cached = 0;
+ int highmem = 0;
+ struct page *page;
+ pg_data_t *pgdat;
+ unsigned long i;
+ unsigned long flags;
+
+ printk(KERN_INFO "Mem-info:\n");
+ show_free_areas();
+ printk(KERN_INFO "Free swap: %6ldkB\n", nr_swap_pages<<(PAGE_SHIFT-10));
+ for_each_online_pgdat(pgdat) {
+ pgdat_resize_lock(pgdat, &flags);
+ for (i = 0; i < pgdat->node_spanned_pages; ++i) {
+ page = pgdat_page_nr(pgdat, i);
+ total++;
+ if (PageHighMem(page))
+ highmem++;
+ if (PageReserved(page))
+ reserved++;
+ else if (PageSwapCache(page))
+ cached++;
+ else if (page_count(page))
+ shared += page_count(page) - 1;
+ }
+ pgdat_resize_unlock(pgdat, &flags);
+ }
+ printk(KERN_INFO "%d pages of RAM\n", total);
+ printk(KERN_INFO "%d pages of HIGHMEM\n", highmem);
+ printk(KERN_INFO "%d reserved pages\n", reserved);
+ printk(KERN_INFO "%d pages shared\n", shared);
+ printk(KERN_INFO "%d pages swap cached\n", cached);
+
+ printk(KERN_INFO "%lu pages dirty\n", global_page_state(NR_FILE_DIRTY));
+ printk(KERN_INFO "%lu pages writeback\n",
+ global_page_state(NR_WRITEBACK));
+ printk(KERN_INFO "%lu pages mapped\n", global_page_state(NR_FILE_MAPPED));
+ printk(KERN_INFO "%lu pages slab\n", global_page_state(NR_SLAB));
+ printk(KERN_INFO "%lu pages pagetables\n",
+ global_page_state(NR_PAGETABLE));
+}
+
+/*
+ * Associate a large virtual page frame with a given physical page frame
+ * and protection flags for that frame. pfn is for the base of the page,
+ * vaddr is what the page gets mapped to - both must be properly aligned.
+ * The pmd must already be instantiated. Assumes PAE mode.
+ */
+void set_pmd_pfn(unsigned long vaddr, unsigned long pfn, pgprot_t flags)
+{
+ pgd_t *pgd;
+ pud_t *pud;
+ pmd_t *pmd;
+
+ if (vaddr & (PMD_SIZE-1)) { /* vaddr is misaligned */
+ printk(KERN_WARNING "set_pmd_pfn: vaddr misaligned\n");
+ return; /* BUG(); */
+ }
+ if (pfn & (PTRS_PER_PTE-1)) { /* pfn is misaligned */
+ printk(KERN_WARNING "set_pmd_pfn: pfn misaligned\n");
+ return; /* BUG(); */
+ }
+ pgd = swapper_pg_dir + pgd_index(vaddr);
+ if (pgd_none(*pgd)) {
+ printk(KERN_WARNING "set_pmd_pfn: pgd_none\n");
+ return; /* BUG(); */
+ }
+ pud = pud_offset(pgd, vaddr);
+ pmd = pmd_offset(pud, vaddr);
+ set_pmd(pmd, pfn_pmd(pfn, flags));
+ /*
+ * It's enough to flush this one mapping.
+ * (PGE mappings get flushed as well)
+ */
+ __flush_tlb_one(vaddr);
+}
+
+static int nr_fixmaps = 0;
+unsigned long hypervisor_virt_start = HYPERVISOR_VIRT_START;
+unsigned long __FIXADDR_TOP = (HYPERVISOR_VIRT_START - 2 * PAGE_SIZE);
+EXPORT_SYMBOL(__FIXADDR_TOP);
+
+void __init set_fixaddr_top(unsigned long top)
+{
+ BUG_ON(nr_fixmaps > 0);
+ hypervisor_virt_start = top;
+ __FIXADDR_TOP = hypervisor_virt_start - 2 * PAGE_SIZE;
+}
+
+void __set_fixmap (enum fixed_addresses idx, maddr_t phys, pgprot_t flags)
+{
+ unsigned long address = __fix_to_virt(idx);
+ pte_t pte;
+
+ if (idx >= __end_of_fixed_addresses) {
+ BUG();
+ return;
+ }
+ switch (idx) {
+ case FIX_WP_TEST:
+ case FIX_VDSO:
+ pte = pfn_pte(phys >> PAGE_SHIFT, flags);
+ break;
+ default:
+ pte = pfn_pte_ma(phys >> PAGE_SHIFT, flags);
+ break;
+ }
+ if (HYPERVISOR_update_va_mapping(address, pte,
+ UVMF_INVLPG|UVMF_ALL))
+ BUG();
+ nr_fixmaps++;
+}
+
+pte_t *pte_alloc_one_kernel(struct mm_struct *mm, unsigned long address)
+{
+ pte_t *pte = (pte_t *)__get_free_page(GFP_KERNEL|__GFP_REPEAT|__GFP_ZERO);
+ if (pte)
+ make_lowmem_page_readonly(pte, XENFEAT_writable_page_tables);
+ return pte;
+}
+
+static void _pte_free(struct page *page, unsigned int order)
+{
+ BUG_ON(order);
+ pte_free(page);
+}
+
+struct page *pte_alloc_one(struct mm_struct *mm, unsigned long address)
+{
+ struct page *pte;
+
+#ifdef CONFIG_HIGHPTE
+ pte = alloc_pages(GFP_KERNEL|__GFP_HIGHMEM|__GFP_REPEAT|__GFP_ZERO, 0);
+#else
+ pte = alloc_pages(GFP_KERNEL|__GFP_REPEAT|__GFP_ZERO, 0);
+#endif
+ if (pte) {
+ SetPageForeign(pte, _pte_free);
+ init_page_count(pte);
+ }
+ return pte;
+}
+
+void pte_free(struct page *pte)
+{
+ unsigned long pfn = page_to_pfn(pte);
+
+ if (!PageHighMem(pte)) {
+ unsigned long va = (unsigned long)__va(pfn << PAGE_SHIFT);
+
+ if (!pte_write(*virt_to_ptep(va)))
+ if (HYPERVISOR_update_va_mapping(
+ va, pfn_pte(pfn, PAGE_KERNEL), 0))
+ BUG();
+ } else
+ clear_bit(PG_pinned, &pte->flags);
+
+ ClearPageForeign(pte);
+ init_page_count(pte);
+
+ __free_page(pte);
+}
+
+void pmd_ctor(void *pmd, kmem_cache_t *cache, unsigned long flags)
+{
+ memset(pmd, 0, PTRS_PER_PMD*sizeof(pmd_t));
+}
+
+/*
+ * List of all pgd's needed for non-PAE so it can invalidate entries
+ * in both cached and uncached pgd's; not needed for PAE since the
+ * kernel pmd is shared. If PAE were not to share the pmd a similar
+ * tactic would be needed. This is essentially codepath-based locking
+ * against pageattr.c; it is the unique case in which a valid change
+ * of kernel pagetables can't be lazily synchronized by vmalloc faults.
+ * vmalloc faults work because attached pagetables are never freed.
+ * The locking scheme was chosen on the basis of manfred's
+ * recommendations and having no core impact whatsoever.
+ * -- wli
+ */
+DEFINE_SPINLOCK(pgd_lock);
+struct page *pgd_list;
+
+static inline void pgd_list_add(pgd_t *pgd)
+{
+ struct page *page = virt_to_page(pgd);
+ page->index = (unsigned long)pgd_list;
+ if (pgd_list)
+ set_page_private(pgd_list, (unsigned long)&page->index);
+ pgd_list = page;
+ set_page_private(page, (unsigned long)&pgd_list);
+}
+
+static inline void pgd_list_del(pgd_t *pgd)
+{
+ struct page *next, **pprev, *page = virt_to_page(pgd);
+ next = (struct page *)page->index;
+ pprev = (struct page **)page_private(page);
+ *pprev = next;
+ if (next)
+ set_page_private(next, (unsigned long)pprev);
+ page->mapping = NULL;
+}
+
+void pgd_ctor(void *pgd, kmem_cache_t *cache, unsigned long unused)
+{
+ unsigned long flags;
+
+ if (PTRS_PER_PMD > 1) {
+ if (HAVE_SHARED_KERNEL_PMD)
+ clone_pgd_range((pgd_t *)pgd + USER_PTRS_PER_PGD,
+ swapper_pg_dir + USER_PTRS_PER_PGD,
+ KERNEL_PGD_PTRS);
+ } else {
+ spin_lock_irqsave(&pgd_lock, flags);
+ clone_pgd_range((pgd_t *)pgd + USER_PTRS_PER_PGD,
+ swapper_pg_dir + USER_PTRS_PER_PGD,
+ KERNEL_PGD_PTRS);
+ memset(pgd, 0, USER_PTRS_PER_PGD*sizeof(pgd_t));
+ pgd_list_add(pgd);
+ spin_unlock_irqrestore(&pgd_lock, flags);
+ }
+}
+
+/* never called when PTRS_PER_PMD > 1 */
+void pgd_dtor(void *pgd, kmem_cache_t *cache, unsigned long unused)
+{
+ unsigned long flags; /* can be called from interrupt context */
+
+ spin_lock_irqsave(&pgd_lock, flags);
+ pgd_list_del(pgd);
+ spin_unlock_irqrestore(&pgd_lock, flags);
+
+ pgd_test_and_unpin(pgd);
+}
+
+pgd_t *pgd_alloc(struct mm_struct *mm)
+{
+ int i;
+ pgd_t *pgd = kmem_cache_alloc(pgd_cache, GFP_KERNEL);
+ pmd_t **pmd;
+ unsigned long flags;
+
+ if (!pgd)
+ return NULL;
+
+ pgd_test_and_unpin(pgd);
+
+ /* Store a back link for vmalloc_sync_all(). */
+ virt_to_page(pgd)->mapping = (void *)mm;
+
+ if (PTRS_PER_PMD == 1)
+ return pgd;
+
+ if (HAVE_SHARED_KERNEL_PMD) {
+ for (i = 0; i < USER_PTRS_PER_PGD; ++i) {
+ pmd_t *pmd = kmem_cache_alloc(pmd_cache, GFP_KERNEL);
+ if (!pmd)
+ goto out_oom;
+ set_pgd(&pgd[i], __pgd(1 + __pa(pmd)));
+ }
+ return pgd;
+ }
+
+ /*
+ * We can race save/restore (if we sleep during a GFP_KERNEL memory
+ * allocation). We therefore store virtual addresses of pmds as they
+ * do not change across save/restore, and poke the machine addresses
+ * into the pgdir under the pgd_lock.
+ */
+ pmd = kmalloc(PTRS_PER_PGD * sizeof(pmd_t *), GFP_KERNEL);
+ if (!pmd) {
+ kmem_cache_free(pgd_cache, pgd);
+ return NULL;
+ }
+
+ /* Allocate pmds, remember virtual addresses. */
+ for (i = 0; i < PTRS_PER_PGD; ++i) {
+ pmd[i] = kmem_cache_alloc(pmd_cache, GFP_KERNEL);
+ if (!pmd[i])
+ goto out_oom;
+ }
+
+ spin_lock_irqsave(&pgd_lock, flags);
+
+ /* Protect against save/restore: move below 4GB under pgd_lock. */
+ if (!xen_feature(XENFEAT_pae_pgdir_above_4gb)) {
+ int rc = xen_create_contiguous_region(
+ (unsigned long)pgd, 0, 32);
+ if (rc) {
+ spin_unlock_irqrestore(&pgd_lock, flags);
+ goto out_oom;
+ }
+ }
+
+ /* Copy kernel pmd contents and write-protect the new pmds. */
+ for (i = USER_PTRS_PER_PGD; i < PTRS_PER_PGD; i++) {
+ unsigned long v = (unsigned long)i << PGDIR_SHIFT;
+ pgd_t *kpgd = pgd_offset_k(v);
+ pud_t *kpud = pud_offset(kpgd, v);
+ pmd_t *kpmd = pmd_offset(kpud, v);
+ memcpy(pmd[i], kpmd, PAGE_SIZE);
+ make_lowmem_page_readonly(
+ pmd[i], XENFEAT_writable_page_tables);
+ }
+
+ /* It is safe to poke machine addresses of pmds under the pmd_lock. */
+ for (i = 0; i < PTRS_PER_PGD; i++)
+ set_pgd(&pgd[i], __pgd(1 + __pa(pmd[i])));
+
+ /* Ensure this pgd gets picked up and pinned on save/restore. */
+ pgd_list_add(pgd);
+
+ spin_unlock_irqrestore(&pgd_lock, flags);
+
+ kfree(pmd);
+
+ return pgd;
+
+out_oom:
+ if (HAVE_SHARED_KERNEL_PMD) {
+ for (i--; i >= 0; i--)
+ kmem_cache_free(pmd_cache,
+ (void *)__va(pgd_val(pgd[i])-1));
+ } else {
+ for (i--; i >= 0; i--)
+ kmem_cache_free(pmd_cache, pmd[i]);
+ kfree(pmd);
+ }
+ kmem_cache_free(pgd_cache, pgd);
+ return NULL;
+}
+
+void pgd_free(pgd_t *pgd)
+{
+ int i;
+
+ /*
+ * After this the pgd should not be pinned for the duration of this
+ * function's execution. We should never sleep and thus never race:
+ * 1. User pmds will not become write-protected under our feet due
+ * to a concurrent mm_pin_all().
+ * 2. The machine addresses in PGD entries will not become invalid
+ * due to a concurrent save/restore.
+ */
+ pgd_test_and_unpin(pgd);
+
+ /* in the PAE case user pgd entries are overwritten before usage */
+ if (PTRS_PER_PMD > 1) {
+ for (i = 0; i < USER_PTRS_PER_PGD; ++i) {
+ pmd_t *pmd = (void *)__va(pgd_val(pgd[i])-1);
+ kmem_cache_free(pmd_cache, pmd);
+ }
+
+ if (!HAVE_SHARED_KERNEL_PMD) {
+ unsigned long flags;
+ spin_lock_irqsave(&pgd_lock, flags);
+ pgd_list_del(pgd);
+ spin_unlock_irqrestore(&pgd_lock, flags);
+
+ for (i = USER_PTRS_PER_PGD; i < PTRS_PER_PGD; i++) {
+ pmd_t *pmd = (void *)__va(pgd_val(pgd[i])-1);
+ make_lowmem_page_writable(
+ pmd, XENFEAT_writable_page_tables);
+ memset(pmd, 0, PTRS_PER_PMD*sizeof(pmd_t));
+ kmem_cache_free(pmd_cache, pmd);
+ }
+
+ if (!xen_feature(XENFEAT_pae_pgdir_above_4gb))
+ xen_destroy_contiguous_region(
+ (unsigned long)pgd, 0);
+ }
+ }
+
+ /* in the non-PAE case, free_pgtables() clears user pgd entries */
+ kmem_cache_free(pgd_cache, pgd);
+}
+
+void make_lowmem_page_readonly(void *va, unsigned int feature)
+{
+ pte_t *pte;
+ int rc;
+
+ if (xen_feature(feature))
+ return;
+
+ pte = virt_to_ptep(va);
+ rc = HYPERVISOR_update_va_mapping(
+ (unsigned long)va, pte_wrprotect(*pte), 0);
+ BUG_ON(rc);
+}
+
+void make_lowmem_page_writable(void *va, unsigned int feature)
+{
+ pte_t *pte;
+ int rc;
+
+ if (xen_feature(feature))
+ return;
+
+ pte = virt_to_ptep(va);
+ rc = HYPERVISOR_update_va_mapping(
+ (unsigned long)va, pte_mkwrite(*pte), 0);
+ BUG_ON(rc);
+}
+
+void make_page_readonly(void *va, unsigned int feature)
+{
+ pte_t *pte;
+ int rc;
+
+ if (xen_feature(feature))
+ return;
+
+ pte = virt_to_ptep(va);
+ rc = HYPERVISOR_update_va_mapping(
+ (unsigned long)va, pte_wrprotect(*pte), 0);
+ if (rc) /* fallback? */
+ xen_l1_entry_update(pte, pte_wrprotect(*pte));
+ if ((unsigned long)va >= (unsigned long)high_memory) {
+ unsigned long pfn = pte_pfn(*pte);
+#ifdef CONFIG_HIGHMEM
+ if (pfn >= highstart_pfn)
+ kmap_flush_unused(); /* flush stale writable kmaps */
+ else
+#endif
+ make_lowmem_page_readonly(
+ phys_to_virt(pfn << PAGE_SHIFT), feature);
+ }
+}
+
+void make_page_writable(void *va, unsigned int feature)
+{
+ pte_t *pte;
+ int rc;
+
+ if (xen_feature(feature))
+ return;
+
+ pte = virt_to_ptep(va);
+ rc = HYPERVISOR_update_va_mapping(
+ (unsigned long)va, pte_mkwrite(*pte), 0);
+ if (rc) /* fallback? */
+ xen_l1_entry_update(pte, pte_mkwrite(*pte));
+ if ((unsigned long)va >= (unsigned long)high_memory) {
+ unsigned long pfn = pte_pfn(*pte);
+#ifdef CONFIG_HIGHMEM
+ if (pfn < highstart_pfn)
+#endif
+ make_lowmem_page_writable(
+ phys_to_virt(pfn << PAGE_SHIFT), feature);
+ }
+}
+
+void make_pages_readonly(void *va, unsigned int nr, unsigned int feature)
+{
+ if (xen_feature(feature))
+ return;
+
+ while (nr-- != 0) {
+ make_page_readonly(va, feature);
+ va = (void *)((unsigned long)va + PAGE_SIZE);
+ }
+}
+
+void make_pages_writable(void *va, unsigned int nr, unsigned int feature)
+{
+ if (xen_feature(feature))
+ return;
+
+ while (nr-- != 0) {
+ make_page_writable(va, feature);
+ va = (void *)((unsigned long)va + PAGE_SIZE);
+ }
+}
+
+static void _pin_lock(struct mm_struct *mm, int lock) {
+ if (lock)
+ spin_lock(&mm->page_table_lock);
+#if NR_CPUS >= CONFIG_SPLIT_PTLOCK_CPUS
+ /* While mm->page_table_lock protects us against insertions and
+ * removals of higher level page table pages, it doesn't protect
+ * against updates of pte-s. Such updates, however, require the
+ * pte pages to be in consistent state (unpinned+writable or
+ * pinned+readonly). The pinning and attribute changes, however
+ * cannot be done atomically, which is why such updates must be
+ * prevented from happening concurrently.
+ * Note that no pte lock can ever elsewhere be acquired nesting
+ * with an already acquired one in the same mm, or with the mm's
+ * page_table_lock already acquired, as that would break in the
+ * non-split case (where all these are actually resolving to the
+ * one page_table_lock). Thus acquiring all of them here is not
+ * going to result in dead locks, and the order of acquires
+ * doesn't matter.
+ */
+ {
+ pgd_t *pgd = mm->pgd;
+ unsigned g;
+
+ for (g = 0; g < USER_PTRS_PER_PGD; g++, pgd++) {
+ pud_t *pud;
+ unsigned u;
+
+ if (pgd_none(*pgd))
+ continue;
+ pud = pud_offset(pgd, 0);
+ for (u = 0; u < PTRS_PER_PUD; u++, pud++) {
+ pmd_t *pmd;
+ unsigned m;
+
+ if (pud_none(*pud))
+ continue;
+ pmd = pmd_offset(pud, 0);
+ for (m = 0; m < PTRS_PER_PMD; m++, pmd++) {
+ spinlock_t *ptl;
+
+ if (pmd_none(*pmd))
+ continue;
+ ptl = pte_lockptr(0, pmd);
+ if (lock)
+ spin_lock(ptl);
+ else
+ spin_unlock(ptl);
+ }
+ }
+ }
+ }
+#endif
+ if (!lock)
+ spin_unlock(&mm->page_table_lock);
+}
+#define pin_lock(mm) _pin_lock(mm, 1)
+#define pin_unlock(mm) _pin_lock(mm, 0)
+
+#define PIN_BATCH 4
+static DEFINE_PER_CPU(multicall_entry_t[PIN_BATCH], pb_mcl);
+
+static inline unsigned int pgd_walk_set_prot(struct page *page, pgprot_t flags,
+ unsigned int cpu, unsigned seq)
+{
+ unsigned long pfn = page_to_pfn(page);
+
+ if (PageHighMem(page)) {
+ if (pgprot_val(flags) & _PAGE_RW)
+ clear_bit(PG_pinned, &page->flags);
+ else
+ set_bit(PG_pinned, &page->flags);
+ } else {
+ MULTI_update_va_mapping(per_cpu(pb_mcl, cpu) + seq,
+ (unsigned long)__va(pfn << PAGE_SHIFT),
+ pfn_pte(pfn, flags), 0);
+ if (unlikely(++seq == PIN_BATCH)) {
+ if (unlikely(HYPERVISOR_multicall_check(per_cpu(pb_mcl, cpu),
+ PIN_BATCH, NULL)))
+ BUG();
+ seq = 0;
+ }
+ }
+
+ return seq;
+}
+
+static void pgd_walk(pgd_t *pgd_base, pgprot_t flags)
+{
+ pgd_t *pgd = pgd_base;
+ pud_t *pud;
+ pmd_t *pmd;
+ int g, u, m;
+ unsigned int cpu, seq;
+
+ if (xen_feature(XENFEAT_auto_translated_physmap))
+ return;
+
+ cpu = get_cpu();
+
+ for (g = 0, seq = 0; g < USER_PTRS_PER_PGD; g++, pgd++) {
+ if (pgd_none(*pgd))
+ continue;
+ pud = pud_offset(pgd, 0);
+ if (PTRS_PER_PUD > 1) /* not folded */
+ seq = pgd_walk_set_prot(virt_to_page(pud),flags,cpu,seq);
+ for (u = 0; u < PTRS_PER_PUD; u++, pud++) {
+ if (pud_none(*pud))
+ continue;
+ pmd = pmd_offset(pud, 0);
+ if (PTRS_PER_PMD > 1) /* not folded */
+ seq = pgd_walk_set_prot(virt_to_page(pmd),flags,cpu,seq);
+ for (m = 0; m < PTRS_PER_PMD; m++, pmd++) {
+ if (pmd_none(*pmd))
+ continue;
+ seq = pgd_walk_set_prot(pmd_page(*pmd),flags,cpu,seq);
+ }
+ }
+ }
+
+ if (likely(seq != 0)) {
+ MULTI_update_va_mapping(per_cpu(pb_mcl, cpu) + seq,
+ (unsigned long)pgd_base,
+ pfn_pte(virt_to_phys(pgd_base)>>PAGE_SHIFT, flags),
+ UVMF_TLB_FLUSH);
+ if (unlikely(HYPERVISOR_multicall_check(per_cpu(pb_mcl, cpu),
+ seq + 1, NULL)))
+ BUG();
+ } else if(HYPERVISOR_update_va_mapping((unsigned long)pgd_base,
+ pfn_pte(virt_to_phys(pgd_base)>>PAGE_SHIFT, flags),
+ UVMF_TLB_FLUSH))
+ BUG();
+
+ put_cpu();
+}
+
+static void __pgd_pin(pgd_t *pgd)
+{
+ pgd_walk(pgd, PAGE_KERNEL_RO);
+ kmap_flush_unused();
+ xen_pgd_pin(__pa(pgd));
+ set_bit(PG_pinned, &virt_to_page(pgd)->flags);
+}
+
+static void __pgd_unpin(pgd_t *pgd)
+{
+ xen_pgd_unpin(__pa(pgd));
+ pgd_walk(pgd, PAGE_KERNEL);
+ clear_bit(PG_pinned, &virt_to_page(pgd)->flags);
+}
+
+static void pgd_test_and_unpin(pgd_t *pgd)
+{
+ if (test_bit(PG_pinned, &virt_to_page(pgd)->flags))
+ __pgd_unpin(pgd);
+}
+
+void mm_pin(struct mm_struct *mm)
+{
+ if (xen_feature(XENFEAT_writable_page_tables))
+ return;
+ pin_lock(mm);
+ __pgd_pin(mm->pgd);
+ pin_unlock(mm);
+}
+
+void mm_unpin(struct mm_struct *mm)
+{
+ if (xen_feature(XENFEAT_writable_page_tables))
+ return;
+ pin_lock(mm);
+ __pgd_unpin(mm->pgd);
+ pin_unlock(mm);
+}
+
+void mm_pin_all(void)
+{
+ struct page *page;
+ unsigned long flags;
+
+ if (xen_feature(XENFEAT_writable_page_tables))
+ return;
+
+ /*
+ * Allow uninterrupted access to the pgd_list. Also protects
+ * __pgd_pin() by disabling preemption.
+ * All other CPUs must be at a safe point (e.g., in stop_machine
+ * or offlined entirely).
+ */
+ spin_lock_irqsave(&pgd_lock, flags);
+ for (page = pgd_list; page; page = (struct page *)page->index) {
+ if (!test_bit(PG_pinned, &page->flags))
+ __pgd_pin((pgd_t *)page_address(page));
+ }
+ spin_unlock_irqrestore(&pgd_lock, flags);
+}
+
+void _arch_dup_mmap(struct mm_struct *mm)
+{
+ if (!test_bit(PG_pinned, &virt_to_page(mm->pgd)->flags))
+ mm_pin(mm);
+}
+
+void _arch_exit_mmap(struct mm_struct *mm)
+{
+ struct task_struct *tsk = current;
+
+ task_lock(tsk);
+
+ /*
+ * We aggressively remove defunct pgd from cr3. We execute unmap_vmas()
+ * *much* faster this way, as no tlb flushes means bigger wrpt batches.
+ */
+ if (tsk->active_mm == mm) {
+ tsk->active_mm = &init_mm;
+ atomic_inc(&init_mm.mm_count);
+
+ switch_mm(mm, &init_mm, tsk);
+
+ atomic_dec(&mm->mm_count);
+ BUG_ON(atomic_read(&mm->mm_count) == 0);
+ }
+
+ task_unlock(tsk);
+
+ if (test_bit(PG_pinned, &virt_to_page(mm->pgd)->flags) &&
+ (atomic_read(&mm->mm_count) == 1) &&
+ !mm->context.has_foreign_mappings)
+ mm_unpin(mm);
+}
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/oprofile/xenoprof.c 2008-01-28 12:24:19.000000000 +0100
@@ -0,0 +1,179 @@
+/**
+ * @file xenoprof.c
+ *
+ * @remark Copyright 2002 OProfile authors
+ * @remark Read the file COPYING
+ *
+ * @author John Levon <levon@movementarian.org>
+ *
+ * Modified by Aravind Menon and Jose Renato Santos for Xen
+ * These modifications are:
+ * Copyright (C) 2005 Hewlett-Packard Co.
+ *
+ * x86-specific part
+ * Copyright (c) 2006 Isaku Yamahata <yamahata at valinux co jp>
+ * VA Linux Systems Japan K.K.
+ */
+
+#include <linux/init.h>
+#include <linux/oprofile.h>
+#include <linux/sched.h>
+#include <asm/pgtable.h>
+
+#include <xen/driver_util.h>
+#include <xen/interface/xen.h>
+#include <xen/interface/xenoprof.h>
+#include <xen/xenoprof.h>
+#include "op_counter.h"
+
+static unsigned int num_events = 0;
+
+void __init xenoprof_arch_init_counter(struct xenoprof_init *init)
+{
+ num_events = init->num_events;
+ /* just in case - make sure we do not overflow event list
+ (i.e. counter_config list) */
+ if (num_events > OP_MAX_COUNTER) {
+ num_events = OP_MAX_COUNTER;
+ init->num_events = num_events;
+ }
+}
+
+void xenoprof_arch_counter(void)
+{
+ int i;
+ struct xenoprof_counter counter;
+
+ for (i=0; i<num_events; i++) {
+ counter.ind = i;
+ counter.count = (uint64_t)counter_config[i].count;
+ counter.enabled = (uint32_t)counter_config[i].enabled;
+ counter.event = (uint32_t)counter_config[i].event;
+ counter.kernel = (uint32_t)counter_config[i].kernel;
+ counter.user = (uint32_t)counter_config[i].user;
+ counter.unit_mask = (uint64_t)counter_config[i].unit_mask;
+ WARN_ON(HYPERVISOR_xenoprof_op(XENOPROF_counter,
+ &counter));
+ }
+}
+
+void xenoprof_arch_start(void)
+{
+ /* nothing */
+}
+
+void xenoprof_arch_stop(void)
+{
+ /* nothing */
+}
+
+void xenoprof_arch_unmap_shared_buffer(struct xenoprof_shared_buffer * sbuf)
+{
+ if (sbuf->buffer) {
+ vunmap(sbuf->buffer);
+ sbuf->buffer = NULL;
+ }
+}
+
+int xenoprof_arch_map_shared_buffer(struct xenoprof_get_buffer * get_buffer,
+ struct xenoprof_shared_buffer * sbuf)
+{
+ int npages, ret;
+ struct vm_struct *area;
+
+ sbuf->buffer = NULL;
+ if ( (ret = HYPERVISOR_xenoprof_op(XENOPROF_get_buffer, get_buffer)) )
+ return ret;
+
+ npages = (get_buffer->bufsize * get_buffer->nbuf - 1) / PAGE_SIZE + 1;
+
+ area = alloc_vm_area(npages * PAGE_SIZE);
+ if (area == NULL)
+ return -ENOMEM;
+
+ if ( (ret = direct_kernel_remap_pfn_range(
+ (unsigned long)area->addr,
+ get_buffer->buf_gmaddr >> PAGE_SHIFT,
+ npages * PAGE_SIZE, __pgprot(_KERNPG_TABLE),
+ DOMID_SELF)) ) {
+ vunmap(area->addr);
+ return ret;
+ }
+
+ sbuf->buffer = area->addr;
+ return ret;
+}
+
+int xenoprof_arch_set_passive(struct xenoprof_passive * pdomain,
+ struct xenoprof_shared_buffer * sbuf)
+{
+ int ret;
+ int npages;
+ struct vm_struct *area;
+ pgprot_t prot = __pgprot(_KERNPG_TABLE);
+
+ sbuf->buffer = NULL;
+ ret = HYPERVISOR_xenoprof_op(XENOPROF_set_passive, pdomain);
+ if (ret)
+ goto out;
+
+ npages = (pdomain->bufsize * pdomain->nbuf - 1) / PAGE_SIZE + 1;
+
+ area = alloc_vm_area(npages * PAGE_SIZE);
+ if (area == NULL) {
+ ret = -ENOMEM;
+ goto out;
+ }
+
+ ret = direct_kernel_remap_pfn_range(
+ (unsigned long)area->addr,
+ pdomain->buf_gmaddr >> PAGE_SHIFT,
+ npages * PAGE_SIZE, prot, DOMID_SELF);
+ if (ret) {
+ vunmap(area->addr);
+ goto out;
+ }
+ sbuf->buffer = area->addr;
+
+out:
+ return ret;
+}
+
+struct op_counter_config counter_config[OP_MAX_COUNTER];
+
+int xenoprof_create_files(struct super_block * sb, struct dentry * root)
+{
+ unsigned int i;
+
+ for (i = 0; i < num_events; ++i) {
+ struct dentry * dir;
+ char buf[2];
+
+ snprintf(buf, 2, "%d", i);
+ dir = oprofilefs_mkdir(sb, root, buf);
+ oprofilefs_create_ulong(sb, dir, "enabled",
+ &counter_config[i].enabled);
+ oprofilefs_create_ulong(sb, dir, "event",
+ &counter_config[i].event);
+ oprofilefs_create_ulong(sb, dir, "count",
+ &counter_config[i].count);
+ oprofilefs_create_ulong(sb, dir, "unit_mask",
+ &counter_config[i].unit_mask);
+ oprofilefs_create_ulong(sb, dir, "kernel",
+ &counter_config[i].kernel);
+ oprofilefs_create_ulong(sb, dir, "user",
+ &counter_config[i].user);
+ }
+
+ return 0;
+}
+
+int __init oprofile_arch_init(struct oprofile_operations * ops)
+{
+ return xenoprofile_init(ops);
+}
+
+void oprofile_arch_exit(void)
+{
+ xenoprofile_exit();
+}
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/pci/irq-xen.c 2008-03-06 08:54:32.000000000 +0100
@@ -0,0 +1,1211 @@
+/*
+ * Low-Level PCI Support for PC -- Routing of Interrupts
+ *
+ * (c) 1999--2000 Martin Mares <mj@ucw.cz>
+ */
+
+#include <linux/types.h>
+#include <linux/kernel.h>
+#include <linux/pci.h>
+#include <linux/init.h>
+#include <linux/slab.h>
+#include <linux/interrupt.h>
+#include <linux/dmi.h>
+#include <asm/io.h>
+#include <asm/smp.h>
+#include <asm/io_apic.h>
+#include <linux/irq.h>
+#include <linux/acpi.h>
+
+#include "pci.h"
+
+#define PIRQ_SIGNATURE (('$' << 0) + ('P' << 8) + ('I' << 16) + ('R' << 24))
+#define PIRQ_VERSION 0x0100
+
+static int broken_hp_bios_irq9;
+static int acer_tm360_irqrouting;
+
+static struct irq_routing_table *pirq_table;
+
+static int pirq_enable_irq(struct pci_dev *dev);
+
+/*
+ * Never use: 0, 1, 2 (timer, keyboard, and cascade)
+ * Avoid using: 13, 14 and 15 (FP error and IDE).
+ * Penalize: 3, 4, 6, 7, 12 (known ISA uses: serial, floppy, parallel and mouse)
+ */
+unsigned int pcibios_irq_mask = 0xfff8;
+
+static int pirq_penalty[16] = {
+ 1000000, 1000000, 1000000, 1000, 1000, 0, 1000, 1000,
+ 0, 0, 0, 0, 1000, 100000, 100000, 100000
+};
+
+struct irq_router {
+ char *name;
+ u16 vendor, device;
+ int (*get)(struct pci_dev *router, struct pci_dev *dev, int pirq);
+ int (*set)(struct pci_dev *router, struct pci_dev *dev, int pirq, int new);
+};
+
+struct irq_router_handler {
+ u16 vendor;
+ int (*probe)(struct irq_router *r, struct pci_dev *router, u16 device);
+};
+
+int (*pcibios_enable_irq)(struct pci_dev *dev) = NULL;
+void (*pcibios_disable_irq)(struct pci_dev *dev) = NULL;
+
+/*
+ * Check passed address for the PCI IRQ Routing Table signature
+ * and perform checksum verification.
+ */
+
+static inline struct irq_routing_table * pirq_check_routing_table(u8 *addr)
+{
+ struct irq_routing_table *rt;
+ int i;
+ u8 sum;
+
+ rt = (struct irq_routing_table *) addr;
+ if (rt->signature != PIRQ_SIGNATURE ||
+ rt->version != PIRQ_VERSION ||
+ rt->size % 16 ||
+ rt->size < sizeof(struct irq_routing_table))
+ return NULL;
+ sum = 0;
+ for (i=0; i < rt->size; i++)
+ sum += addr[i];
+ if (!sum) {
+ DBG(KERN_DEBUG "PCI: Interrupt Routing Table found at 0x%p\n", rt);
+ return rt;
+ }
+ return NULL;
+}
+
+
+
+/*
+ * Search 0xf0000 -- 0xfffff for the PCI IRQ Routing Table.
+ */
+
+static struct irq_routing_table * __init pirq_find_routing_table(void)
+{
+ u8 *addr;
+ struct irq_routing_table *rt;
+
+#ifdef CONFIG_XEN
+ if (!is_initial_xendomain())
+ return NULL;
+#endif
+ if (pirq_table_addr) {
+ rt = pirq_check_routing_table((u8 *) isa_bus_to_virt(pirq_table_addr));
+ if (rt)
+ return rt;
+ printk(KERN_WARNING "PCI: PIRQ table NOT found at pirqaddr\n");
+ }
+ for(addr = (u8 *) isa_bus_to_virt(0xf0000); addr < (u8 *) isa_bus_to_virt(0x100000); addr += 16) {
+ rt = pirq_check_routing_table(addr);
+ if (rt)
+ return rt;
+ }
+ return NULL;
+}
+
+/*
+ * If we have a IRQ routing table, use it to search for peer host
+ * bridges. It's a gross hack, but since there are no other known
+ * ways how to get a list of buses, we have to go this way.
+ */
+
+static void __init pirq_peer_trick(void)
+{
+ struct irq_routing_table *rt = pirq_table;
+ u8 busmap[256];
+ int i;
+ struct irq_info *e;
+
+ memset(busmap, 0, sizeof(busmap));
+ for(i=0; i < (rt->size - sizeof(struct irq_routing_table)) / sizeof(struct irq_info); i++) {
+ e = &rt->slots[i];
+#ifdef DEBUG
+ {
+ int j;
+ DBG(KERN_DEBUG "%02x:%02x slot=%02x", e->bus, e->devfn/8, e->slot);
+ for(j=0; j<4; j++)
+ DBG(" %d:%02x/%04x", j, e->irq[j].link, e->irq[j].bitmap);
+ DBG("\n");
+ }
+#endif
+ busmap[e->bus] = 1;
+ }
+ for(i = 1; i < 256; i++) {
+ if (!busmap[i] || pci_find_bus(0, i))
+ continue;
+ if (pci_scan_bus(i, &pci_root_ops, NULL))
+ printk(KERN_INFO "PCI: Discovered primary peer bus %02x [IRQ]\n", i);
+ }
+ pcibios_last_bus = -1;
+}
+
+/*
+ * Code for querying and setting of IRQ routes on various interrupt routers.
+ */
+
+void eisa_set_level_irq(unsigned int irq)
+{
+ unsigned char mask = 1 << (irq & 7);
+ unsigned int port = 0x4d0 + (irq >> 3);
+ unsigned char val;
+ static u16 eisa_irq_mask;
+
+ if (irq >= 16 || (1 << irq) & eisa_irq_mask)
+ return;
+
+ eisa_irq_mask |= (1 << irq);
+ printk(KERN_DEBUG "PCI: setting IRQ %u as level-triggered\n", irq);
+ val = inb(port);
+ if (!(val & mask)) {
+ DBG(KERN_DEBUG " -> edge");
+ outb(val | mask, port);
+ }
+}
+
+/*
+ * Common IRQ routing practice: nybbles in config space,
+ * offset by some magic constant.
+ */
+static unsigned int read_config_nybble(struct pci_dev *router, unsigned offset, unsigned nr)
+{
+ u8 x;
+ unsigned reg = offset + (nr >> 1);
+
+ pci_read_config_byte(router, reg, &x);
+ return (nr & 1) ? (x >> 4) : (x & 0xf);
+}
+
+static void write_config_nybble(struct pci_dev *router, unsigned offset, unsigned nr, unsigned int val)
+{
+ u8 x;
+ unsigned reg = offset + (nr >> 1);
+
+ pci_read_config_byte(router, reg, &x);
+ x = (nr & 1) ? ((x & 0x0f) | (val << 4)) : ((x & 0xf0) | val);
+ pci_write_config_byte(router, reg, x);
+}
+
+/*
+ * ALI pirq entries are damn ugly, and completely undocumented.
+ * This has been figured out from pirq tables, and it's not a pretty
+ * picture.
+ */
+static int pirq_ali_get(struct pci_dev *router, struct pci_dev *dev, int pirq)
+{
+ static const unsigned char irqmap[16] = { 0, 9, 3, 10, 4, 5, 7, 6, 1, 11, 0, 12, 0, 14, 0, 15 };
+
+ return irqmap[read_config_nybble(router, 0x48, pirq-1)];
+}
+
+static int pirq_ali_set(struct pci_dev *router, struct pci_dev *dev, int pirq, int irq)
+{
+ static const unsigned char irqmap[16] = { 0, 8, 0, 2, 4, 5, 7, 6, 0, 1, 3, 9, 11, 0, 13, 15 };
+ unsigned int val = irqmap[irq];
+
+ if (val) {
+ write_config_nybble(router, 0x48, pirq-1, val);
+ return 1;
+ }
+ return 0;
+}
+
+/*
+ * The Intel PIIX4 pirq rules are fairly simple: "pirq" is
+ * just a pointer to the config space.
+ */
+static int pirq_piix_get(struct pci_dev *router, struct pci_dev *dev, int pirq)
+{
+ u8 x;
+
+ pci_read_config_byte(router, pirq, &x);
+ return (x < 16) ? x : 0;
+}
+
+static int pirq_piix_set(struct pci_dev *router, struct pci_dev *dev, int pirq, int irq)
+{
+ pci_write_config_byte(router, pirq, irq);
+ return 1;
+}
+
+/*
+ * The VIA pirq rules are nibble-based, like ALI,
+ * but without the ugly irq number munging.
+ * However, PIRQD is in the upper instead of lower 4 bits.
+ */
+static int pirq_via_get(struct pci_dev *router, struct pci_dev *dev, int pirq)
+{
+ return read_config_nybble(router, 0x55, pirq == 4 ? 5 : pirq);
+}
+
+static int pirq_via_set(struct pci_dev *router, struct pci_dev *dev, int pirq, int irq)
+{
+ write_config_nybble(router, 0x55, pirq == 4 ? 5 : pirq, irq);
+ return 1;
+}
+
+/*
+ * The VIA pirq rules are nibble-based, like ALI,
+ * but without the ugly irq number munging.
+ * However, for 82C586, nibble map is different .
+ */
+static int pirq_via586_get(struct pci_dev *router, struct pci_dev *dev, int pirq)
+{
+ static const unsigned int pirqmap[5] = { 3, 2, 5, 1, 1 };
+ return read_config_nybble(router, 0x55, pirqmap[pirq-1]);
+}
+
+static int pirq_via586_set(struct pci_dev *router, struct pci_dev *dev, int pirq, int irq)
+{
+ static const unsigned int pirqmap[5] = { 3, 2, 5, 1, 1 };
+ write_config_nybble(router, 0x55, pirqmap[pirq-1], irq);
+ return 1;
+}
+
+/*
+ * ITE 8330G pirq rules are nibble-based
+ * FIXME: pirqmap may be { 1, 0, 3, 2 },
+ * 2+3 are both mapped to irq 9 on my system
+ */
+static int pirq_ite_get(struct pci_dev *router, struct pci_dev *dev, int pirq)
+{
+ static const unsigned char pirqmap[4] = { 1, 0, 2, 3 };
+ return read_config_nybble(router,0x43, pirqmap[pirq-1]);
+}
+
+static int pirq_ite_set(struct pci_dev *router, struct pci_dev *dev, int pirq, int irq)
+{
+ static const unsigned char pirqmap[4] = { 1, 0, 2, 3 };
+ write_config_nybble(router, 0x43, pirqmap[pirq-1], irq);
+ return 1;
+}
+
+/*
+ * OPTI: high four bits are nibble pointer..
+ * I wonder what the low bits do?
+ */
+static int pirq_opti_get(struct pci_dev *router, struct pci_dev *dev, int pirq)
+{
+ return read_config_nybble(router, 0xb8, pirq >> 4);
+}
+
+static int pirq_opti_set(struct pci_dev *router, struct pci_dev *dev, int pirq, int irq)
+{
+ write_config_nybble(router, 0xb8, pirq >> 4, irq);
+ return 1;
+}
+
+/*
+ * Cyrix: nibble offset 0x5C
+ * 0x5C bits 7:4 is INTB bits 3:0 is INTA
+ * 0x5D bits 7:4 is INTD bits 3:0 is INTC
+ */
+static int pirq_cyrix_get(struct pci_dev *router, struct pci_dev *dev, int pirq)
+{
+ return read_config_nybble(router, 0x5C, (pirq-1)^1);
+}
+
+static int pirq_cyrix_set(struct pci_dev *router, struct pci_dev *dev, int pirq, int irq)
+{
+ write_config_nybble(router, 0x5C, (pirq-1)^1, irq);
+ return 1;
+}
+
+/*
+ * PIRQ routing for SiS 85C503 router used in several SiS chipsets.
+ * We have to deal with the following issues here:
+ * - vendors have different ideas about the meaning of link values
+ * - some onboard devices (integrated in the chipset) have special
+ * links and are thus routed differently (i.e. not via PCI INTA-INTD)
+ * - different revision of the router have a different layout for
+ * the routing registers, particularly for the onchip devices
+ *
+ * For all routing registers the common thing is we have one byte
+ * per routeable link which is defined as:
+ * bit 7 IRQ mapping enabled (0) or disabled (1)
+ * bits [6:4] reserved (sometimes used for onchip devices)
+ * bits [3:0] IRQ to map to
+ * allowed: 3-7, 9-12, 14-15
+ * reserved: 0, 1, 2, 8, 13
+ *
+ * The config-space registers located at 0x41/0x42/0x43/0x44 are
+ * always used to route the normal PCI INT A/B/C/D respectively.
+ * Apparently there are systems implementing PCI routing table using
+ * link values 0x01-0x04 and others using 0x41-0x44 for PCI INTA..D.
+ * We try our best to handle both link mappings.
+ *
+ * Currently (2003-05-21) it appears most SiS chipsets follow the
+ * definition of routing registers from the SiS-5595 southbridge.
+ * According to the SiS 5595 datasheets the revision id's of the
+ * router (ISA-bridge) should be 0x01 or 0xb0.
+ *
+ * Furthermore we've also seen lspci dumps with revision 0x00 and 0xb1.
+ * Looks like these are used in a number of SiS 5xx/6xx/7xx chipsets.
+ * They seem to work with the current routing code. However there is
+ * some concern because of the two USB-OHCI HCs (original SiS 5595
+ * had only one). YMMV.
+ *
+ * Onchip routing for router rev-id 0x01/0xb0 and probably 0x00/0xb1:
+ *
+ * 0x61: IDEIRQ:
+ * bits [6:5] must be written 01
+ * bit 4 channel-select primary (0), secondary (1)
+ *
+ * 0x62: USBIRQ:
+ * bit 6 OHCI function disabled (0), enabled (1)
+ *
+ * 0x6a: ACPI/SCI IRQ: bits 4-6 reserved
+ *
+ * 0x7e: Data Acq. Module IRQ - bits 4-6 reserved
+ *
+ * We support USBIRQ (in addition to INTA-INTD) and keep the
+ * IDE, ACPI and DAQ routing untouched as set by the BIOS.
+ *
+ * Currently the only reported exception is the new SiS 65x chipset
+ * which includes the SiS 69x southbridge. Here we have the 85C503
+ * router revision 0x04 and there are changes in the register layout
+ * mostly related to the different USB HCs with USB 2.0 support.
+ *
+ * Onchip routing for router rev-id 0x04 (try-and-error observation)
+ *
+ * 0x60/0x61/0x62/0x63: 1xEHCI and 3xOHCI (companion) USB-HCs
+ * bit 6-4 are probably unused, not like 5595
+ */
+
+#define PIRQ_SIS_IRQ_MASK 0x0f
+#define PIRQ_SIS_IRQ_DISABLE 0x80
+#define PIRQ_SIS_USB_ENABLE 0x40
+
+static int pirq_sis_get(struct pci_dev *router, struct pci_dev *dev, int pirq)
+{
+ u8 x;
+ int reg;
+
+ reg = pirq;
+ if (reg >= 0x01 && reg <= 0x04)
+ reg += 0x40;
+ pci_read_config_byte(router, reg, &x);
+ return (x & PIRQ_SIS_IRQ_DISABLE) ? 0 : (x & PIRQ_SIS_IRQ_MASK);
+}
+
+static int pirq_sis_set(struct pci_dev *router, struct pci_dev *dev, int pirq, int irq)
+{
+ u8 x;
+ int reg;
+
+ reg = pirq;
+ if (reg >= 0x01 && reg <= 0x04)
+ reg += 0x40;
+ pci_read_config_byte(router, reg, &x);
+ x &= ~(PIRQ_SIS_IRQ_MASK | PIRQ_SIS_IRQ_DISABLE);
+ x |= irq ? irq: PIRQ_SIS_IRQ_DISABLE;
+ pci_write_config_byte(router, reg, x);
+ return 1;
+}
+
+
+/*
+ * VLSI: nibble offset 0x74 - educated guess due to routing table and
+ * config space of VLSI 82C534 PCI-bridge/router (1004:0102)
+ * Tested on HP OmniBook 800 covering PIRQ 1, 2, 4, 8 for onboard
+ * devices, PIRQ 3 for non-pci(!) soundchip and (untested) PIRQ 6
+ * for the busbridge to the docking station.
+ */
+
+static int pirq_vlsi_get(struct pci_dev *router, struct pci_dev *dev, int pirq)
+{
+ if (pirq > 8) {
+ printk(KERN_INFO "VLSI router pirq escape (%d)\n", pirq);
+ return 0;
+ }
+ return read_config_nybble(router, 0x74, pirq-1);
+}
+
+static int pirq_vlsi_set(struct pci_dev *router, struct pci_dev *dev, int pirq, int irq)
+{
+ if (pirq > 8) {
+ printk(KERN_INFO "VLSI router pirq escape (%d)\n", pirq);
+ return 0;
+ }
+ write_config_nybble(router, 0x74, pirq-1, irq);
+ return 1;
+}
+
+/*
+ * ServerWorks: PCI interrupts mapped to system IRQ lines through Index
+ * and Redirect I/O registers (0x0c00 and 0x0c01). The Index register
+ * format is (PCIIRQ## | 0x10), e.g.: PCIIRQ10=0x1a. The Redirect
+ * register is a straight binary coding of desired PIC IRQ (low nibble).
+ *
+ * The 'link' value in the PIRQ table is already in the correct format
+ * for the Index register. There are some special index values:
+ * 0x00 for ACPI (SCI), 0x01 for USB, 0x02 for IDE0, 0x04 for IDE1,
+ * and 0x03 for SMBus.
+ */
+static int pirq_serverworks_get(struct pci_dev *router, struct pci_dev *dev, int pirq)
+{
+ outb_p(pirq, 0xc00);
+ return inb(0xc01) & 0xf;
+}
+
+static int pirq_serverworks_set(struct pci_dev *router, struct pci_dev *dev, int pirq, int irq)
+{
+ outb_p(pirq, 0xc00);
+ outb_p(irq, 0xc01);
+ return 1;
+}
+
+/* Support for AMD756 PCI IRQ Routing
+ * Jhon H. Caicedo <jhcaiced@osso.org.co>
+ * Jun/21/2001 0.2.0 Release, fixed to use "nybble" functions... (jhcaiced)
+ * Jun/19/2001 Alpha Release 0.1.0 (jhcaiced)
+ * The AMD756 pirq rules are nibble-based
+ * offset 0x56 0-3 PIRQA 4-7 PIRQB
+ * offset 0x57 0-3 PIRQC 4-7 PIRQD
+ */
+static int pirq_amd756_get(struct pci_dev *router, struct pci_dev *dev, int pirq)
+{
+ u8 irq;
+ irq = 0;
+ if (pirq <= 4)
+ {
+ irq = read_config_nybble(router, 0x56, pirq - 1);
+ }
+ printk(KERN_INFO "AMD756: dev %04x:%04x, router pirq : %d get irq : %2d\n",
+ dev->vendor, dev->device, pirq, irq);
+ return irq;
+}
+
+static int pirq_amd756_set(struct pci_dev *router, struct pci_dev *dev, int pirq, int irq)
+{
+ printk(KERN_INFO "AMD756: dev %04x:%04x, router pirq : %d SET irq : %2d\n",
+ dev->vendor, dev->device, pirq, irq);
+ if (pirq <= 4)
+ {
+ write_config_nybble(router, 0x56, pirq - 1, irq);
+ }
+ return 1;
+}
+
+#ifdef CONFIG_PCI_BIOS
+
+static int pirq_bios_set(struct pci_dev *router, struct pci_dev *dev, int pirq, int irq)
+{
+ struct pci_dev *bridge;
+ int pin = pci_get_interrupt_pin(dev, &bridge);
+ return pcibios_set_irq_routing(bridge, pin, irq);
+}
+
+#endif
+
+static __init int intel_router_probe(struct irq_router *r, struct pci_dev *router, u16 device)
+{
+ static struct pci_device_id __initdata pirq_440gx[] = {
+ { PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82443GX_0) },
+ { PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82443GX_2) },
+ { },
+ };
+
+ /* 440GX has a proprietary PIRQ router -- don't use it */
+ if (pci_dev_present(pirq_440gx))
+ return 0;
+
+ switch(device)
+ {
+ case PCI_DEVICE_ID_INTEL_82371FB_0:
+ case PCI_DEVICE_ID_INTEL_82371SB_0:
+ case PCI_DEVICE_ID_INTEL_82371AB_0:
+ case PCI_DEVICE_ID_INTEL_82371MX:
+ case PCI_DEVICE_ID_INTEL_82443MX_0:
+ case PCI_DEVICE_ID_INTEL_82801AA_0:
+ case PCI_DEVICE_ID_INTEL_82801AB_0:
+ case PCI_DEVICE_ID_INTEL_82801BA_0:
+ case PCI_DEVICE_ID_INTEL_82801BA_10:
+ case PCI_DEVICE_ID_INTEL_82801CA_0:
+ case PCI_DEVICE_ID_INTEL_82801CA_12:
+ case PCI_DEVICE_ID_INTEL_82801DB_0:
+ case PCI_DEVICE_ID_INTEL_82801E_0:
+ case PCI_DEVICE_ID_INTEL_82801EB_0:
+ case PCI_DEVICE_ID_INTEL_ESB_1:
+ case PCI_DEVICE_ID_INTEL_ICH6_0:
+ case PCI_DEVICE_ID_INTEL_ICH6_1:
+ case PCI_DEVICE_ID_INTEL_ICH7_0:
+ case PCI_DEVICE_ID_INTEL_ICH7_1:
+ case PCI_DEVICE_ID_INTEL_ICH7_30:
+ case PCI_DEVICE_ID_INTEL_ICH7_31:
+ case PCI_DEVICE_ID_INTEL_ESB2_0:
+ case PCI_DEVICE_ID_INTEL_ICH8_0:
+ case PCI_DEVICE_ID_INTEL_ICH8_1:
+ case PCI_DEVICE_ID_INTEL_ICH8_2:
+ case PCI_DEVICE_ID_INTEL_ICH8_3:
+ case PCI_DEVICE_ID_INTEL_ICH8_4:
+ case PCI_DEVICE_ID_INTEL_ICH9_0:
+ case PCI_DEVICE_ID_INTEL_ICH9_1:
+ case PCI_DEVICE_ID_INTEL_ICH9_2:
+ case PCI_DEVICE_ID_INTEL_ICH9_3:
+ case PCI_DEVICE_ID_INTEL_ICH9_4:
+ case PCI_DEVICE_ID_INTEL_ICH9_5:
+ r->name = "PIIX/ICH";
+ r->get = pirq_piix_get;
+ r->set = pirq_piix_set;
+ return 1;
+ }
+ return 0;
+}
+
+static __init int via_router_probe(struct irq_router *r,
+ struct pci_dev *router, u16 device)
+{
+ /* FIXME: We should move some of the quirk fixup stuff here */
+
+ /*
+ * work arounds for some buggy BIOSes
+ */
+ if (device == PCI_DEVICE_ID_VIA_82C586_0) {
+ switch(router->device) {
+ case PCI_DEVICE_ID_VIA_82C686:
+ /*
+ * Asus k7m bios wrongly reports 82C686A
+ * as 586-compatible
+ */
+ device = PCI_DEVICE_ID_VIA_82C686;
+ break;
+ case PCI_DEVICE_ID_VIA_8235:
+ /**
+ * Asus a7v-x bios wrongly reports 8235
+ * as 586-compatible
+ */
+ device = PCI_DEVICE_ID_VIA_8235;
+ break;
+ }
+ }
+
+ switch(device) {
+ case PCI_DEVICE_ID_VIA_82C586_0:
+ r->name = "VIA";
+ r->get = pirq_via586_get;
+ r->set = pirq_via586_set;
+ return 1;
+ case PCI_DEVICE_ID_VIA_82C596:
+ case PCI_DEVICE_ID_VIA_82C686:
+ case PCI_DEVICE_ID_VIA_8231:
+ case PCI_DEVICE_ID_VIA_8233A:
+ case PCI_DEVICE_ID_VIA_8235:
+ case PCI_DEVICE_ID_VIA_8237:
+ /* FIXME: add new ones for 8233/5 */
+ r->name = "VIA";
+ r->get = pirq_via_get;
+ r->set = pirq_via_set;
+ return 1;
+ }
+ return 0;
+}
+
+static __init int vlsi_router_probe(struct irq_router *r, struct pci_dev *router, u16 device)
+{
+ switch(device)
+ {
+ case PCI_DEVICE_ID_VLSI_82C534:
+ r->name = "VLSI 82C534";
+ r->get = pirq_vlsi_get;
+ r->set = pirq_vlsi_set;
+ return 1;
+ }
+ return 0;
+}
+
+
+static __init int serverworks_router_probe(struct irq_router *r, struct pci_dev *router, u16 device)
+{
+ switch(device)
+ {
+ case PCI_DEVICE_ID_SERVERWORKS_OSB4:
+ case PCI_DEVICE_ID_SERVERWORKS_CSB5:
+ r->name = "ServerWorks";
+ r->get = pirq_serverworks_get;
+ r->set = pirq_serverworks_set;
+ return 1;
+ }
+ return 0;
+}
+
+static __init int sis_router_probe(struct irq_router *r, struct pci_dev *router, u16 device)
+{
+ if (device != PCI_DEVICE_ID_SI_503)
+ return 0;
+
+ r->name = "SIS";
+ r->get = pirq_sis_get;
+ r->set = pirq_sis_set;
+ return 1;
+}
+
+static __init int cyrix_router_probe(struct irq_router *r, struct pci_dev *router, u16 device)
+{
+ switch(device)
+ {
+ case PCI_DEVICE_ID_CYRIX_5520:
+ r->name = "NatSemi";
+ r->get = pirq_cyrix_get;
+ r->set = pirq_cyrix_set;
+ return 1;
+ }
+ return 0;
+}
+
+static __init int opti_router_probe(struct irq_router *r, struct pci_dev *router, u16 device)
+{
+ switch(device)
+ {
+ case PCI_DEVICE_ID_OPTI_82C700:
+ r->name = "OPTI";
+ r->get = pirq_opti_get;
+ r->set = pirq_opti_set;
+ return 1;
+ }
+ return 0;
+}
+
+static __init int ite_router_probe(struct irq_router *r, struct pci_dev *router, u16 device)
+{
+ switch(device)
+ {
+ case PCI_DEVICE_ID_ITE_IT8330G_0:
+ r->name = "ITE";
+ r->get = pirq_ite_get;
+ r->set = pirq_ite_set;
+ return 1;
+ }
+ return 0;
+}
+
+static __init int ali_router_probe(struct irq_router *r, struct pci_dev *router, u16 device)
+{
+ switch(device)
+ {
+ case PCI_DEVICE_ID_AL_M1533:
+ case PCI_DEVICE_ID_AL_M1563:
+ printk(KERN_DEBUG "PCI: Using ALI IRQ Router\n");
+ r->name = "ALI";
+ r->get = pirq_ali_get;
+ r->set = pirq_ali_set;
+ return 1;
+ }
+ return 0;
+}
+
+static __init int amd_router_probe(struct irq_router *r, struct pci_dev *router, u16 device)
+{
+ switch(device)
+ {
+ case PCI_DEVICE_ID_AMD_VIPER_740B:
+ r->name = "AMD756";
+ break;
+ case PCI_DEVICE_ID_AMD_VIPER_7413:
+ r->name = "AMD766";
+ break;
+ case PCI_DEVICE_ID_AMD_VIPER_7443:
+ r->name = "AMD768";
+ break;
+ default:
+ return 0;
+ }
+ r->get = pirq_amd756_get;
+ r->set = pirq_amd756_set;
+ return 1;
+}
+
+static __initdata struct irq_router_handler pirq_routers[] = {
+ { PCI_VENDOR_ID_INTEL, intel_router_probe },
+ { PCI_VENDOR_ID_AL, ali_router_probe },
+ { PCI_VENDOR_ID_ITE, ite_router_probe },
+ { PCI_VENDOR_ID_VIA, via_router_probe },
+ { PCI_VENDOR_ID_OPTI, opti_router_probe },
+ { PCI_VENDOR_ID_SI, sis_router_probe },
+ { PCI_VENDOR_ID_CYRIX, cyrix_router_probe },
+ { PCI_VENDOR_ID_VLSI, vlsi_router_probe },
+ { PCI_VENDOR_ID_SERVERWORKS, serverworks_router_probe },
+ { PCI_VENDOR_ID_AMD, amd_router_probe },
+ /* Someone with docs needs to add the ATI Radeon IGP */
+ { 0, NULL }
+};
+static struct irq_router pirq_router;
+static struct pci_dev *pirq_router_dev;
+
+
+/*
+ * FIXME: should we have an option to say "generic for
+ * chipset" ?
+ */
+
+static void __init pirq_find_router(struct irq_router *r)
+{
+ struct irq_routing_table *rt = pirq_table;
+ struct irq_router_handler *h;
+
+#ifdef CONFIG_PCI_BIOS
+ if (!rt->signature) {
+ printk(KERN_INFO "PCI: Using BIOS for IRQ routing\n");
+ r->set = pirq_bios_set;
+ r->name = "BIOS";
+ return;
+ }
+#endif
+
+ /* Default unless a driver reloads it */
+ r->name = "default";
+ r->get = NULL;
+ r->set = NULL;
+
+ DBG(KERN_DEBUG "PCI: Attempting to find IRQ router for %04x:%04x\n",
+ rt->rtr_vendor, rt->rtr_device);
+
+ pirq_router_dev = pci_find_slot(rt->rtr_bus, rt->rtr_devfn);
+ if (!pirq_router_dev) {
+ DBG(KERN_DEBUG "PCI: Interrupt router not found at "
+ "%02x:%02x\n", rt->rtr_bus, rt->rtr_devfn);
+ return;
+ }
+
+ for( h = pirq_routers; h->vendor; h++) {
+ /* First look for a router match */
+ if (rt->rtr_vendor == h->vendor && h->probe(r, pirq_router_dev, rt->rtr_device))
+ break;
+ /* Fall back to a device match */
+ if (pirq_router_dev->vendor == h->vendor && h->probe(r, pirq_router_dev, pirq_router_dev->device))
+ break;
+ }
+ printk(KERN_INFO "PCI: Using IRQ router %s [%04x/%04x] at %s\n",
+ pirq_router.name,
+ pirq_router_dev->vendor,
+ pirq_router_dev->device,
+ pci_name(pirq_router_dev));
+}
+
+static struct irq_info *pirq_get_info(struct pci_dev *dev)
+{
+ struct irq_routing_table *rt = pirq_table;
+ int entries = (rt->size - sizeof(struct irq_routing_table)) / sizeof(struct irq_info);
+ struct irq_info *info;
+
+ for (info = rt->slots; entries--; info++)
+ if (info->bus == dev->bus->number && PCI_SLOT(info->devfn) == PCI_SLOT(dev->devfn))
+ return info;
+ return NULL;
+}
+
+static int pcibios_lookup_irq(struct pci_dev *dev, int assign)
+{
+ u8 pin;
+ struct irq_info *info;
+ int i, pirq, newirq;
+ int irq = 0;
+ u32 mask;
+ struct irq_router *r = &pirq_router;
+ struct pci_dev *dev2 = NULL;
+ char *msg = NULL;
+
+ /* Find IRQ pin */
+ pci_read_config_byte(dev, PCI_INTERRUPT_PIN, &pin);
+ if (!pin) {
+ DBG(KERN_DEBUG " -> no interrupt pin\n");
+ return 0;
+ }
+ pin = pin - 1;
+
+ /* Find IRQ routing entry */
+
+ if (!pirq_table)
+ return 0;
+
+ DBG(KERN_DEBUG "IRQ for %s[%c]", pci_name(dev), 'A' + pin);
+ info = pirq_get_info(dev);
+ if (!info) {
+ DBG(" -> not found in routing table\n" KERN_DEBUG);
+ return 0;
+ }
+ pirq = info->irq[pin].link;
+ mask = info->irq[pin].bitmap;
+ if (!pirq) {
+ DBG(" -> not routed\n" KERN_DEBUG);
+ return 0;
+ }
+ DBG(" -> PIRQ %02x, mask %04x, excl %04x", pirq, mask, pirq_table->exclusive_irqs);
+ mask &= pcibios_irq_mask;
+
+ /* Work around broken HP Pavilion Notebooks which assign USB to
+ IRQ 9 even though it is actually wired to IRQ 11 */
+
+ if (broken_hp_bios_irq9 && pirq == 0x59 && dev->irq == 9) {
+ dev->irq = 11;
+ pci_write_config_byte(dev, PCI_INTERRUPT_LINE, 11);
+ r->set(pirq_router_dev, dev, pirq, 11);
+ }
+
+ /* same for Acer Travelmate 360, but with CB and irq 11 -> 10 */
+ if (acer_tm360_irqrouting && dev->irq == 11 && dev->vendor == PCI_VENDOR_ID_O2) {
+ pirq = 0x68;
+ mask = 0x400;
+ dev->irq = r->get(pirq_router_dev, dev, pirq);
+ pci_write_config_byte(dev, PCI_INTERRUPT_LINE, dev->irq);
+ }
+
+ /*
+ * Find the best IRQ to assign: use the one
+ * reported by the device if possible.
+ */
+ newirq = dev->irq;
+ if (newirq && !((1 << newirq) & mask)) {
+ if ( pci_probe & PCI_USE_PIRQ_MASK) newirq = 0;
+ else printk("\n" KERN_WARNING
+ "PCI: IRQ %i for device %s doesn't match PIRQ mask "
+ "- try pci=usepirqmask\n" KERN_DEBUG, newirq,
+ pci_name(dev));
+ }
+ if (!newirq && assign) {
+ for (i = 0; i < 16; i++) {
+ if (!(mask & (1 << i)))
+ continue;
+ if (pirq_penalty[i] < pirq_penalty[newirq] && can_request_irq(i, IRQF_SHARED))
+ newirq = i;
+ }
+ }
+ DBG(" -> newirq=%d", newirq);
+
+ /* Check if it is hardcoded */
+ if ((pirq & 0xf0) == 0xf0) {
+ irq = pirq & 0xf;
+ DBG(" -> hardcoded IRQ %d\n", irq);
+ msg = "Hardcoded";
+ } else if ( r->get && (irq = r->get(pirq_router_dev, dev, pirq)) && \
+ ((!(pci_probe & PCI_USE_PIRQ_MASK)) || ((1 << irq) & mask)) ) {
+ DBG(" -> got IRQ %d\n", irq);
+ msg = "Found";
+ eisa_set_level_irq(irq);
+ } else if (newirq && r->set && (dev->class >> 8) != PCI_CLASS_DISPLAY_VGA) {
+ DBG(" -> assigning IRQ %d", newirq);
+ if (r->set(pirq_router_dev, dev, pirq, newirq)) {
+ eisa_set_level_irq(newirq);
+ DBG(" ... OK\n");
+ msg = "Assigned";
+ irq = newirq;
+ }
+ }
+
+ if (!irq) {
+ DBG(" ... failed\n");
+ if (newirq && mask == (1 << newirq)) {
+ msg = "Guessed";
+ irq = newirq;
+ } else
+ return 0;
+ }
+ printk(KERN_INFO "PCI: %s IRQ %d for device %s\n", msg, irq, pci_name(dev));
+
+ /* Update IRQ for all devices with the same pirq value */
+ while ((dev2 = pci_get_device(PCI_ANY_ID, PCI_ANY_ID, dev2)) != NULL) {
+ pci_read_config_byte(dev2, PCI_INTERRUPT_PIN, &pin);
+ if (!pin)
+ continue;
+ pin--;
+ info = pirq_get_info(dev2);
+ if (!info)
+ continue;
+ if (info->irq[pin].link == pirq) {
+ /* We refuse to override the dev->irq information. Give a warning! */
+ if ( dev2->irq && dev2->irq != irq && \
+ (!(pci_probe & PCI_USE_PIRQ_MASK) || \
+ ((1 << dev2->irq) & mask)) ) {
+#ifndef CONFIG_PCI_MSI
+ printk(KERN_INFO "IRQ routing conflict for %s, have irq %d, want irq %d\n",
+ pci_name(dev2), dev2->irq, irq);
+#endif
+ continue;
+ }
+ dev2->irq = irq;
+ pirq_penalty[irq]++;
+ if (dev != dev2)
+ printk(KERN_INFO "PCI: Sharing IRQ %d with %s\n", irq, pci_name(dev2));
+ }
+ }
+ return 1;
+}
+
+static void __init pcibios_fixup_irqs(void)
+{
+ struct pci_dev *dev = NULL;
+ u8 pin;
+
+ DBG(KERN_DEBUG "PCI: IRQ fixup\n");
+ while ((dev = pci_get_device(PCI_ANY_ID, PCI_ANY_ID, dev)) != NULL) {
+ /*
+ * If the BIOS has set an out of range IRQ number, just ignore it.
+ * Also keep track of which IRQ's are already in use.
+ */
+ if (dev->irq >= 16) {
+ DBG(KERN_DEBUG "%s: ignoring bogus IRQ %d\n", pci_name(dev), dev->irq);
+ dev->irq = 0;
+ }
+ /* If the IRQ is already assigned to a PCI device, ignore its ISA use penalty */
+ if (pirq_penalty[dev->irq] >= 100 && pirq_penalty[dev->irq] < 100000)
+ pirq_penalty[dev->irq] = 0;
+ pirq_penalty[dev->irq]++;
+ }
+
+ dev = NULL;
+ while ((dev = pci_get_device(PCI_ANY_ID, PCI_ANY_ID, dev)) != NULL) {
+ pci_read_config_byte(dev, PCI_INTERRUPT_PIN, &pin);
+#ifdef CONFIG_X86_IO_APIC
+ /*
+ * Recalculate IRQ numbers if we use the I/O APIC.
+ */
+ if (io_apic_assign_pci_irqs)
+ {
+ int irq;
+
+ if (pin) {
+ pin--; /* interrupt pins are numbered starting from 1 */
+ irq = IO_APIC_get_PCI_irq_vector(dev->bus->number, PCI_SLOT(dev->devfn), pin);
+ /*
+ * Busses behind bridges are typically not listed in the MP-table.
+ * In this case we have to look up the IRQ based on the parent bus,
+ * parent slot, and pin number. The SMP code detects such bridged
+ * busses itself so we should get into this branch reliably.
+ */
+ if (irq < 0 && dev->bus->parent) { /* go back to the bridge */
+ struct pci_dev * bridge = dev->bus->self;
+
+ pin = (pin + PCI_SLOT(dev->devfn)) % 4;
+ irq = IO_APIC_get_PCI_irq_vector(bridge->bus->number,
+ PCI_SLOT(bridge->devfn), pin);
+ if (irq >= 0)
+ printk(KERN_WARNING "PCI: using PPB %s[%c] to get irq %d\n",
+ pci_name(bridge), 'A' + pin, irq);
+ }
+ if (irq >= 0) {
+ if (use_pci_vector() &&
+ !platform_legacy_irq(irq))
+ irq = IO_APIC_VECTOR(irq);
+
+ printk(KERN_INFO "PCI->APIC IRQ transform: %s[%c] -> IRQ %d\n",
+ pci_name(dev), 'A' + pin, irq);
+ dev->irq = irq;
+ }
+ }
+ }
+#endif
+ /*
+ * Still no IRQ? Try to lookup one...
+ */
+ if (pin && !dev->irq)
+ pcibios_lookup_irq(dev, 0);
+ }
+}
+
+/*
+ * Work around broken HP Pavilion Notebooks which assign USB to
+ * IRQ 9 even though it is actually wired to IRQ 11
+ */
+static int __init fix_broken_hp_bios_irq9(struct dmi_system_id *d)
+{
+ if (!broken_hp_bios_irq9) {
+ broken_hp_bios_irq9 = 1;
+ printk(KERN_INFO "%s detected - fixing broken IRQ routing\n", d->ident);
+ }
+ return 0;
+}
+
+/*
+ * Work around broken Acer TravelMate 360 Notebooks which assign
+ * Cardbus to IRQ 11 even though it is actually wired to IRQ 10
+ */
+static int __init fix_acer_tm360_irqrouting(struct dmi_system_id *d)
+{
+ if (!acer_tm360_irqrouting) {
+ acer_tm360_irqrouting = 1;
+ printk(KERN_INFO "%s detected - fixing broken IRQ routing\n", d->ident);
+ }
+ return 0;
+}
+
+static struct dmi_system_id __initdata pciirq_dmi_table[] = {
+ {
+ .callback = fix_broken_hp_bios_irq9,
+ .ident = "HP Pavilion N5400 Series Laptop",
+ .matches = {
+ DMI_MATCH(DMI_SYS_VENDOR, "Hewlett-Packard"),
+ DMI_MATCH(DMI_BIOS_VERSION, "GE.M1.03"),
+ DMI_MATCH(DMI_PRODUCT_VERSION, "HP Pavilion Notebook Model GE"),
+ DMI_MATCH(DMI_BOARD_VERSION, "OmniBook N32N-736"),
+ },
+ },
+ {
+ .callback = fix_acer_tm360_irqrouting,
+ .ident = "Acer TravelMate 36x Laptop",
+ .matches = {
+ DMI_MATCH(DMI_SYS_VENDOR, "Acer"),
+ DMI_MATCH(DMI_PRODUCT_NAME, "TravelMate 360"),
+ },
+ },
+ { }
+};
+
+static int __init pcibios_irq_init(void)
+{
+ DBG(KERN_DEBUG "PCI: IRQ init\n");
+
+ if (pcibios_enable_irq || raw_pci_ops == NULL)
+ return 0;
+
+ dmi_check_system(pciirq_dmi_table);
+
+ pirq_table = pirq_find_routing_table();
+
+#ifdef CONFIG_PCI_BIOS
+ if (!pirq_table && (pci_probe & PCI_BIOS_IRQ_SCAN))
+ pirq_table = pcibios_get_irq_routing_table();
+#endif
+ if (pirq_table) {
+ pirq_peer_trick();
+ pirq_find_router(&pirq_router);
+ if (pirq_table->exclusive_irqs) {
+ int i;
+ for (i=0; i<16; i++)
+ if (!(pirq_table->exclusive_irqs & (1 << i)))
+ pirq_penalty[i] += 100;
+ }
+ /* If we're using the I/O APIC, avoid using the PCI IRQ routing table */
+ if (io_apic_assign_pci_irqs)
+ pirq_table = NULL;
+ }
+
+ pcibios_enable_irq = pirq_enable_irq;
+
+ pcibios_fixup_irqs();
+ return 0;
+}
+
+subsys_initcall(pcibios_irq_init);
+
+
+static void pirq_penalize_isa_irq(int irq, int active)
+{
+ /*
+ * If any ISAPnP device reports an IRQ in its list of possible
+ * IRQ's, we try to avoid assigning it to PCI devices.
+ */
+ if (irq < 16) {
+ if (active)
+ pirq_penalty[irq] += 1000;
+ else
+ pirq_penalty[irq] += 100;
+ }
+}
+
+void pcibios_penalize_isa_irq(int irq, int active)
+{
+#ifdef CONFIG_ACPI
+ if (!acpi_noirq)
+ acpi_penalize_isa_irq(irq, active);
+ else
+#endif
+ pirq_penalize_isa_irq(irq, active);
+}
+
+static int pirq_enable_irq(struct pci_dev *dev)
+{
+ u8 pin;
+ struct pci_dev *temp_dev;
+
+ pci_read_config_byte(dev, PCI_INTERRUPT_PIN, &pin);
+ if (pin && !pcibios_lookup_irq(dev, 1) && !dev->irq) {
+ char *msg = "";
+
+ pin--; /* interrupt pins are numbered starting from 1 */
+
+ if (io_apic_assign_pci_irqs) {
+ int irq;
+
+ irq = IO_APIC_get_PCI_irq_vector(dev->bus->number, PCI_SLOT(dev->devfn), pin);
+ /*
+ * Busses behind bridges are typically not listed in the MP-table.
+ * In this case we have to look up the IRQ based on the parent bus,
+ * parent slot, and pin number. The SMP code detects such bridged
+ * busses itself so we should get into this branch reliably.
+ */
+ temp_dev = dev;
+ while (irq < 0 && dev->bus->parent) { /* go back to the bridge */
+ struct pci_dev * bridge = dev->bus->self;
+
+ pin = (pin + PCI_SLOT(dev->devfn)) % 4;
+ irq = IO_APIC_get_PCI_irq_vector(bridge->bus->number,
+ PCI_SLOT(bridge->devfn), pin);
+ if (irq >= 0)
+ printk(KERN_WARNING "PCI: using PPB %s[%c] to get irq %d\n",
+ pci_name(bridge), 'A' + pin, irq);
+ dev = bridge;
+ }
+ dev = temp_dev;
+ if (irq >= 0) {
+#ifdef CONFIG_PCI_MSI
+ if (!platform_legacy_irq(irq))
+ irq = IO_APIC_VECTOR(irq);
+#endif
+ printk(KERN_INFO "PCI->APIC IRQ transform: %s[%c] -> IRQ %d\n",
+ pci_name(dev), 'A' + pin, irq);
+ dev->irq = irq;
+ return 0;
+ } else
+ msg = " Probably buggy MP table.";
+ } else if (pci_probe & PCI_BIOS_IRQ_SCAN)
+ msg = "";
+ else
+ msg = " Please try using pci=biosirq.";
+
+ /* With IDE legacy devices the IRQ lookup failure is not a problem.. */
+ if (dev->class >> 8 == PCI_CLASS_STORAGE_IDE && !(dev->class & 0x5))
+ return 0;
+
+ printk(KERN_WARNING "PCI: No IRQ known for interrupt pin %c of device %s.%s\n",
+ 'A' + pin, pci_name(dev), msg);
+ }
+ return 0;
+}
+
+int pci_vector_resources(int last, int nr_released)
+{
+ int count = nr_released;
+
+ int next = last;
+ int offset = (last % 8);
+
+ while (next < FIRST_SYSTEM_VECTOR) {
+ next += 8;
+#ifdef CONFIG_X86_64
+ if (next == IA32_SYSCALL_VECTOR)
+ continue;
+#else
+ if (next == SYSCALL_VECTOR)
+ continue;
+#endif
+ count++;
+ if (next >= FIRST_SYSTEM_VECTOR) {
+ if (offset%8) {
+ next = FIRST_DEVICE_VECTOR + offset;
+ offset++;
+ continue;
+ }
+ count--;
+ }
+ }
+
+ return count;
+}
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/pci/pcifront.c 2009-03-18 10:39:31.000000000 +0100
@@ -0,0 +1,57 @@
+/*
+ * PCI Frontend Stub - puts some "dummy" functions in to the Linux x86 PCI core
+ * to support the Xen PCI Frontend's operation
+ *
+ * Author: Ryan Wilson <hap9@epoch.ncsc.mil>
+ */
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/pci.h>
+#include <asm/acpi.h>
+#include <xen/evtchn.h>
+#include "pci.h"
+
+static int pcifront_enable_irq(struct pci_dev *dev)
+{
+ u8 irq;
+ pci_read_config_byte(dev, PCI_INTERRUPT_LINE, &irq);
+ evtchn_register_pirq(irq);
+ dev->irq = irq;
+
+ return 0;
+}
+
+extern u8 pci_cache_line_size;
+
+static int __init pcifront_x86_stub_init(void)
+{
+ struct cpuinfo_x86 *c = &boot_cpu_data;
+
+ /* Only install our method if we haven't found real hardware already */
+ if (raw_pci_ops)
+ return 0;
+
+ printk(KERN_INFO "PCI: setting up Xen PCI frontend stub\n");
+
+ /* Copied from arch/i386/pci/common.c */
+ pci_cache_line_size = 32 >> 2;
+ if (c->x86 >= 6 && c->x86_vendor == X86_VENDOR_AMD)
+ pci_cache_line_size = 64 >> 2; /* K7 & K8 */
+ else if (c->x86 > 6 && c->x86_vendor == X86_VENDOR_INTEL)
+ pci_cache_line_size = 128 >> 2; /* P4 */
+
+ /* On x86, we need to disable the normal IRQ routing table and
+ * just ask the backend
+ */
+ pcibios_enable_irq = pcifront_enable_irq;
+ pcibios_disable_irq = NULL;
+
+#ifdef CONFIG_ACPI
+ /* Keep ACPI out of the picture */
+ acpi_noirq = 1;
+#endif
+
+ return 0;
+}
+
+arch_initcall(pcifront_x86_stub_init);
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/ia32/ia32entry-xen.S 2008-04-02 12:34:02.000000000 +0200
@@ -0,0 +1,666 @@
+/*
+ * Compatibility mode system call entry point for x86-64.
+ *
+ * Copyright 2000-2002 Andi Kleen, SuSE Labs.
+ */
+
+#include <asm/dwarf2.h>
+#include <asm/calling.h>
+#include <asm/asm-offsets.h>
+#include <asm/current.h>
+#include <asm/errno.h>
+#include <asm/ia32_unistd.h>
+#include <asm/thread_info.h>
+#include <asm/segment.h>
+#include <asm/vsyscall32.h>
+#include <asm/irqflags.h>
+#include <linux/linkage.h>
+
+#define IA32_NR_syscalls ((ia32_syscall_end - ia32_sys_call_table)/8)
+
+ .macro IA32_ARG_FIXUP noebp=0
+ movl %edi,%r8d
+ .if \noebp
+ .else
+ movl %ebp,%r9d
+ .endif
+ xchg %ecx,%esi
+ movl %ebx,%edi
+ movl %edx,%edx /* zero extension */
+ .endm
+
+ /* clobbers %eax */
+ .macro CLEAR_RREGS
+ xorl %eax,%eax
+ movq %rax,R11(%rsp)
+ movq %rax,R10(%rsp)
+ movq %rax,R9(%rsp)
+ movq %rax,R8(%rsp)
+ .endm
+
+ .macro LOAD_ARGS32 offset
+ movl \offset(%rsp),%r11d
+ movl \offset+8(%rsp),%r10d
+ movl \offset+16(%rsp),%r9d
+ movl \offset+24(%rsp),%r8d
+ movl \offset+40(%rsp),%ecx
+ movl \offset+48(%rsp),%edx
+ movl \offset+56(%rsp),%esi
+ movl \offset+64(%rsp),%edi
+ movl \offset+72(%rsp),%eax
+ .endm
+
+ .macro CFI_STARTPROC32 simple
+ CFI_STARTPROC \simple
+ CFI_UNDEFINED r8
+ CFI_UNDEFINED r9
+ CFI_UNDEFINED r10
+ CFI_UNDEFINED r11
+ CFI_UNDEFINED r12
+ CFI_UNDEFINED r13
+ CFI_UNDEFINED r14
+ CFI_UNDEFINED r15
+ .endm
+
+/*
+ * 32bit SYSENTER instruction entry.
+ *
+ * Arguments:
+ * %eax System call number.
+ * %ebx Arg1
+ * %ecx Arg2
+ * %edx Arg3
+ * %esi Arg4
+ * %edi Arg5
+ * %ebp user stack
+ * 0(%ebp) Arg6
+ *
+ * Interrupts on.
+ *
+ * This is purely a fast path. For anything complicated we use the int 0x80
+ * path below. Set up a complete hardware stack frame to share code
+ * with the int 0x80 path.
+ */
+ENTRY(ia32_sysenter_target)
+ CFI_STARTPROC32 simple
+ CFI_DEF_CFA rsp,SS+8-RIP+16
+ /*CFI_REL_OFFSET ss,SS-RIP+16*/
+ CFI_REL_OFFSET rsp,RSP-RIP+16
+ /*CFI_REL_OFFSET rflags,EFLAGS-RIP+16*/
+ /*CFI_REL_OFFSET cs,CS-RIP+16*/
+ CFI_REL_OFFSET rip,RIP-RIP+16
+ CFI_REL_OFFSET r11,8
+ CFI_REL_OFFSET rcx,0
+ movq 8(%rsp),%r11
+ CFI_RESTORE r11
+ popq %rcx
+ CFI_ADJUST_CFA_OFFSET -8
+ CFI_RESTORE rcx
+ movl %ebp,%ebp /* zero extension */
+ movl %eax,%eax
+ movl $__USER32_DS,40(%rsp)
+ movq %rbp,32(%rsp)
+ movl $__USER32_CS,16(%rsp)
+ movl $VSYSCALL32_SYSEXIT,8(%rsp)
+ movq %rax,(%rsp)
+ cld
+ SAVE_ARGS 0,0,0
+ /* no need to do an access_ok check here because rbp has been
+ 32bit zero extended */
+1: movl (%rbp),%r9d
+ .section __ex_table,"a"
+ .quad 1b,ia32_badarg
+ .previous
+ GET_THREAD_INFO(%r10)
+ orl $TS_COMPAT,threadinfo_status(%r10)
+ testl $(_TIF_SYSCALL_TRACE|_TIF_SYSCALL_AUDIT|_TIF_SECCOMP),threadinfo_flags(%r10)
+ jnz sysenter_tracesys
+sysenter_do_call:
+ cmpl $(IA32_NR_syscalls-1),%eax
+ ja ia32_badsys
+ IA32_ARG_FIXUP 1
+ call *ia32_sys_call_table(,%rax,8)
+ movq %rax,RAX-ARGOFFSET(%rsp)
+ jmp int_ret_from_sys_call
+
+sysenter_tracesys:
+ SAVE_REST
+ CLEAR_RREGS
+ movq $-ENOSYS,RAX(%rsp) /* really needed? */
+ movq %rsp,%rdi /* &pt_regs -> arg1 */
+ call syscall_trace_enter
+ LOAD_ARGS32 ARGOFFSET /* reload args from stack in case ptrace changed it */
+ RESTORE_REST
+ movl %ebp, %ebp
+ /* no need to do an access_ok check here because rbp has been
+ 32bit zero extended */
+1: movl (%rbp),%r9d
+ .section __ex_table,"a"
+ .quad 1b,ia32_badarg
+ .previous
+ jmp sysenter_do_call
+ CFI_ENDPROC
+ENDPROC(ia32_sysenter_target)
+
+/*
+ * 32bit SYSCALL instruction entry.
+ *
+ * Arguments:
+ * %eax System call number.
+ * %ebx Arg1
+ * %ecx return EIP
+ * %edx Arg3
+ * %esi Arg4
+ * %edi Arg5
+ * %ebp Arg2 [note: not saved in the stack frame, should not be touched]
+ * %esp user stack
+ * 0(%esp) Arg6
+ *
+ * Interrupts on.
+ *
+ * This is purely a fast path. For anything complicated we use the int 0x80
+ * path below. Set up a complete hardware stack frame to share code
+ * with the int 0x80 path.
+ */
+ENTRY(ia32_cstar_target)
+ CFI_STARTPROC32 simple
+ CFI_DEF_CFA rsp,SS+8-RIP+16
+ /*CFI_REL_OFFSET ss,SS-RIP+16*/
+ CFI_REL_OFFSET rsp,RSP-RIP+16
+ /*CFI_REL_OFFSET rflags,EFLAGS-RIP+16*/
+ /*CFI_REL_OFFSET cs,CS-RIP+16*/
+ CFI_REL_OFFSET rip,RIP-RIP+16
+ movl %eax,%eax /* zero extension */
+ movl RSP-RIP+16(%rsp),%r8d
+ SAVE_ARGS -8,1,1
+ movq %rax,ORIG_RAX-ARGOFFSET(%rsp)
+ movq %rbp,RCX-ARGOFFSET(%rsp) /* this lies slightly to ptrace */
+ movl %ebp,%ecx
+ movl $__USER32_CS,CS-ARGOFFSET(%rsp)
+ movl $__USER32_DS,SS-ARGOFFSET(%rsp)
+ /* no need to do an access_ok check here because r8 has been
+ 32bit zero extended */
+ /* hardware stack frame is complete now */
+1: movl (%r8),%r9d
+ .section __ex_table,"a"
+ .quad 1b,ia32_badarg
+ .previous
+ GET_THREAD_INFO(%r10)
+ orl $TS_COMPAT,threadinfo_status(%r10)
+ testl $(_TIF_SYSCALL_TRACE|_TIF_SYSCALL_AUDIT|_TIF_SECCOMP),threadinfo_flags(%r10)
+ jnz cstar_tracesys
+cstar_do_call:
+ cmpl $IA32_NR_syscalls-1,%eax
+ ja ia32_badsys
+ IA32_ARG_FIXUP 1
+ call *ia32_sys_call_table(,%rax,8)
+ movq %rax,RAX-ARGOFFSET(%rsp)
+ jmp int_ret_from_sys_call
+
+cstar_tracesys:
+ SAVE_REST
+ CLEAR_RREGS
+ movq $-ENOSYS,RAX(%rsp) /* really needed? */
+ movq %rsp,%rdi /* &pt_regs -> arg1 */
+ call syscall_trace_enter
+ LOAD_ARGS32 ARGOFFSET /* reload args from stack in case ptrace changed it */
+ RESTORE_REST
+ movl RSP-ARGOFFSET(%rsp), %r8d
+ /* no need to do an access_ok check here because r8 has been
+ 32bit zero extended */
+1: movl (%r8),%r9d
+ .section __ex_table,"a"
+ .quad 1b,ia32_badarg
+ .previous
+ jmp cstar_do_call
+END(ia32_cstar_target)
+
+ia32_badarg:
+ movq $-EFAULT,%rax
+ jmp ia32_sysret
+ CFI_ENDPROC
+
+/*
+ * Emulated IA32 system calls via int 0x80.
+ *
+ * Arguments:
+ * %eax System call number.
+ * %ebx Arg1
+ * %ecx Arg2
+ * %edx Arg3
+ * %esi Arg4
+ * %edi Arg5
+ * %ebp Arg6 [note: not saved in the stack frame, should not be touched]
+ *
+ * Notes:
+ * Uses the same stack frame as the x86-64 version.
+ * All registers except %eax must be saved (but ptrace may violate that)
+ * Arguments are zero extended. For system calls that want sign extension and
+ * take long arguments a wrapper is needed. Most calls can just be called
+ * directly.
+ * Assumes it is only called from user space and entered with interrupts on.
+ */
+
+ENTRY(ia32_syscall)
+ CFI_STARTPROC simple
+ CFI_DEF_CFA rsp,SS+8-RIP+16
+ /*CFI_REL_OFFSET ss,SS-RIP+16*/
+ CFI_REL_OFFSET rsp,RSP-RIP+16
+ /*CFI_REL_OFFSET rflags,EFLAGS-RIP+16*/
+ /*CFI_REL_OFFSET cs,CS-RIP+16*/
+ CFI_REL_OFFSET rip,RIP-RIP+16
+ CFI_REL_OFFSET r11,8
+ CFI_REL_OFFSET rcx,0
+ movq 8(%rsp),%r11
+ CFI_RESTORE r11
+ popq %rcx
+ CFI_ADJUST_CFA_OFFSET -8
+ CFI_RESTORE rcx
+ movl %eax,%eax
+ movq %rax,(%rsp)
+ cld
+ /* note the registers are not zero extended to the sf.
+ this could be a problem. */
+ SAVE_ARGS 0,0,1
+ GET_THREAD_INFO(%r10)
+ orl $TS_COMPAT,threadinfo_status(%r10)
+ testl $(_TIF_SYSCALL_TRACE|_TIF_SYSCALL_AUDIT|_TIF_SECCOMP),threadinfo_flags(%r10)
+ jnz ia32_tracesys
+ia32_do_syscall:
+ cmpl $(IA32_NR_syscalls-1),%eax
+ ja ia32_badsys
+ IA32_ARG_FIXUP
+ call *ia32_sys_call_table(,%rax,8) # xxx: rip relative
+ia32_sysret:
+ movq %rax,RAX-ARGOFFSET(%rsp)
+ jmp int_ret_from_sys_call
+
+ia32_tracesys:
+ SAVE_REST
+ movq $-ENOSYS,RAX(%rsp) /* really needed? */
+ movq %rsp,%rdi /* &pt_regs -> arg1 */
+ call syscall_trace_enter
+ LOAD_ARGS32 ARGOFFSET /* reload args from stack in case ptrace changed it */
+ RESTORE_REST
+ jmp ia32_do_syscall
+END(ia32_syscall)
+
+ia32_badsys:
+ movq $0,ORIG_RAX-ARGOFFSET(%rsp)
+ movq $-ENOSYS,RAX-ARGOFFSET(%rsp)
+ jmp int_ret_from_sys_call
+
+quiet_ni_syscall:
+ movq $-ENOSYS,%rax
+ ret
+ CFI_ENDPROC
+
+ .macro PTREGSCALL label, func, arg
+ .globl \label
+\label:
+ leaq \func(%rip),%rax
+ leaq -ARGOFFSET+8(%rsp),\arg /* 8 for return address */
+ jmp ia32_ptregs_common
+ .endm
+
+ CFI_STARTPROC32
+
+ PTREGSCALL stub32_rt_sigreturn, sys32_rt_sigreturn, %rdi
+ PTREGSCALL stub32_sigreturn, sys32_sigreturn, %rdi
+ PTREGSCALL stub32_sigaltstack, sys32_sigaltstack, %rdx
+ PTREGSCALL stub32_sigsuspend, sys32_sigsuspend, %rcx
+ PTREGSCALL stub32_execve, sys32_execve, %rcx
+ PTREGSCALL stub32_fork, sys_fork, %rdi
+ PTREGSCALL stub32_clone, sys32_clone, %rdx
+ PTREGSCALL stub32_vfork, sys_vfork, %rdi
+ PTREGSCALL stub32_iopl, sys_iopl, %rsi
+ PTREGSCALL stub32_rt_sigsuspend, sys_rt_sigsuspend, %rdx
+
+ENTRY(ia32_ptregs_common)
+ popq %r11
+ CFI_ENDPROC
+ CFI_STARTPROC32 simple
+ CFI_DEF_CFA rsp,SS+8-ARGOFFSET
+ CFI_REL_OFFSET rax,RAX-ARGOFFSET
+ CFI_REL_OFFSET rcx,RCX-ARGOFFSET
+ CFI_REL_OFFSET rdx,RDX-ARGOFFSET
+ CFI_REL_OFFSET rsi,RSI-ARGOFFSET
+ CFI_REL_OFFSET rdi,RDI-ARGOFFSET
+ CFI_REL_OFFSET rip,RIP-ARGOFFSET
+/* CFI_REL_OFFSET cs,CS-ARGOFFSET*/
+/* CFI_REL_OFFSET rflags,EFLAGS-ARGOFFSET*/
+ CFI_REL_OFFSET rsp,RSP-ARGOFFSET
+/* CFI_REL_OFFSET ss,SS-ARGOFFSET*/
+ SAVE_REST
+ call *%rax
+ RESTORE_REST
+ jmp ia32_sysret /* misbalances the return cache */
+ CFI_ENDPROC
+END(ia32_ptregs_common)
+
+ .section .rodata,"a"
+ .align 8
+ia32_sys_call_table:
+ .quad sys_restart_syscall
+ .quad sys_exit
+ .quad stub32_fork
+ .quad sys_read
+ .quad sys_write
+ .quad compat_sys_open /* 5 */
+ .quad sys_close
+ .quad sys32_waitpid
+ .quad sys_creat
+ .quad sys_link
+ .quad sys_unlink /* 10 */
+ .quad stub32_execve
+ .quad sys_chdir
+ .quad compat_sys_time
+ .quad sys_mknod
+ .quad sys_chmod /* 15 */
+ .quad sys_lchown16
+ .quad quiet_ni_syscall /* old break syscall holder */
+ .quad sys_stat
+ .quad sys32_lseek
+ .quad sys_getpid /* 20 */
+ .quad compat_sys_mount /* mount */
+ .quad sys_oldumount /* old_umount */
+ .quad sys_setuid16
+ .quad sys_getuid16
+ .quad compat_sys_stime /* stime */ /* 25 */
+ .quad sys32_ptrace /* ptrace */
+ .quad sys_alarm
+ .quad sys_fstat /* (old)fstat */
+ .quad sys_pause
+ .quad compat_sys_utime /* 30 */
+ .quad quiet_ni_syscall /* old stty syscall holder */
+ .quad quiet_ni_syscall /* old gtty syscall holder */
+ .quad sys_access
+ .quad sys_nice
+ .quad quiet_ni_syscall /* 35 */ /* old ftime syscall holder */
+ .quad sys_sync
+ .quad sys32_kill
+ .quad sys_rename
+ .quad sys_mkdir
+ .quad sys_rmdir /* 40 */
+ .quad sys_dup
+ .quad sys32_pipe
+ .quad compat_sys_times
+ .quad quiet_ni_syscall /* old prof syscall holder */
+ .quad sys_brk /* 45 */
+ .quad sys_setgid16
+ .quad sys_getgid16
+ .quad sys_signal
+ .quad sys_geteuid16
+ .quad sys_getegid16 /* 50 */
+ .quad sys_acct
+ .quad sys_umount /* new_umount */
+ .quad quiet_ni_syscall /* old lock syscall holder */
+ .quad compat_sys_ioctl
+ .quad compat_sys_fcntl64 /* 55 */
+ .quad quiet_ni_syscall /* old mpx syscall holder */
+ .quad sys_setpgid
+ .quad quiet_ni_syscall /* old ulimit syscall holder */
+ .quad sys32_olduname
+ .quad sys_umask /* 60 */
+ .quad sys_chroot
+ .quad sys32_ustat
+ .quad sys_dup2
+ .quad sys_getppid
+ .quad sys_getpgrp /* 65 */
+ .quad sys_setsid
+ .quad sys32_sigaction
+ .quad sys_sgetmask
+ .quad sys_ssetmask
+ .quad sys_setreuid16 /* 70 */
+ .quad sys_setregid16
+ .quad stub32_sigsuspend
+ .quad compat_sys_sigpending
+ .quad sys_sethostname
+ .quad compat_sys_setrlimit /* 75 */
+ .quad compat_sys_old_getrlimit /* old_getrlimit */
+ .quad compat_sys_getrusage
+ .quad sys32_gettimeofday
+ .quad sys32_settimeofday
+ .quad sys_getgroups16 /* 80 */
+ .quad sys_setgroups16
+ .quad sys32_old_select
+ .quad sys_symlink
+ .quad sys_lstat
+ .quad sys_readlink /* 85 */
+#ifdef CONFIG_IA32_AOUT
+ .quad sys_uselib
+#else
+ .quad quiet_ni_syscall
+#endif
+ .quad sys_swapon
+ .quad sys_reboot
+ .quad compat_sys_old_readdir
+ .quad sys32_mmap /* 90 */
+ .quad sys_munmap
+ .quad sys_truncate
+ .quad sys_ftruncate
+ .quad sys_fchmod
+ .quad sys_fchown16 /* 95 */
+ .quad sys_getpriority
+ .quad sys_setpriority
+ .quad quiet_ni_syscall /* old profil syscall holder */
+ .quad compat_sys_statfs
+ .quad compat_sys_fstatfs /* 100 */
+ .quad sys_ioperm
+ .quad compat_sys_socketcall
+ .quad sys_syslog
+ .quad compat_sys_setitimer
+ .quad compat_sys_getitimer /* 105 */
+ .quad compat_sys_newstat
+ .quad compat_sys_newlstat
+ .quad compat_sys_newfstat
+ .quad sys32_uname
+ .quad stub32_iopl /* 110 */
+ .quad sys_vhangup
+ .quad quiet_ni_syscall /* old "idle" system call */
+ .quad sys32_vm86_warning /* vm86old */
+ .quad compat_sys_wait4
+ .quad sys_swapoff /* 115 */
+ .quad sys32_sysinfo
+ .quad sys32_ipc
+ .quad sys_fsync
+ .quad stub32_sigreturn
+ .quad stub32_clone /* 120 */
+ .quad sys_setdomainname
+ .quad sys_uname
+ .quad sys_modify_ldt
+ .quad compat_sys_adjtimex
+ .quad sys32_mprotect /* 125 */
+ .quad compat_sys_sigprocmask
+ .quad quiet_ni_syscall /* create_module */
+ .quad sys_init_module
+ .quad sys_delete_module
+ .quad quiet_ni_syscall /* 130 get_kernel_syms */
+ .quad sys_quotactl
+ .quad sys_getpgid
+ .quad sys_fchdir
+ .quad quiet_ni_syscall /* bdflush */
+ .quad sys_sysfs /* 135 */
+ .quad sys_personality
+ .quad quiet_ni_syscall /* for afs_syscall */
+ .quad sys_setfsuid16
+ .quad sys_setfsgid16
+ .quad sys_llseek /* 140 */
+ .quad compat_sys_getdents
+ .quad compat_sys_select
+ .quad sys_flock
+ .quad sys_msync
+ .quad compat_sys_readv /* 145 */
+ .quad compat_sys_writev
+ .quad sys_getsid
+ .quad sys_fdatasync
+ .quad sys32_sysctl /* sysctl */
+ .quad sys_mlock /* 150 */
+ .quad sys_munlock
+ .quad sys_mlockall
+ .quad sys_munlockall
+ .quad sys_sched_setparam
+ .quad sys_sched_getparam /* 155 */
+ .quad sys_sched_setscheduler
+ .quad sys_sched_getscheduler
+ .quad sys_sched_yield
+ .quad sys_sched_get_priority_max
+ .quad sys_sched_get_priority_min /* 160 */
+ .quad sys_sched_rr_get_interval
+ .quad compat_sys_nanosleep
+ .quad sys_mremap
+ .quad sys_setresuid16
+ .quad sys_getresuid16 /* 165 */
+ .quad sys32_vm86_warning /* vm86 */
+ .quad quiet_ni_syscall /* query_module */
+ .quad sys_poll
+ .quad compat_sys_nfsservctl
+ .quad sys_setresgid16 /* 170 */
+ .quad sys_getresgid16
+ .quad sys_prctl
+ .quad stub32_rt_sigreturn
+ .quad sys32_rt_sigaction
+ .quad sys32_rt_sigprocmask /* 175 */
+ .quad sys32_rt_sigpending
+ .quad compat_sys_rt_sigtimedwait
+ .quad sys32_rt_sigqueueinfo
+ .quad stub32_rt_sigsuspend
+ .quad sys32_pread /* 180 */
+ .quad sys32_pwrite
+ .quad sys_chown16
+ .quad sys_getcwd
+ .quad sys_capget
+ .quad sys_capset
+ .quad stub32_sigaltstack
+ .quad sys32_sendfile
+ .quad quiet_ni_syscall /* streams1 */
+ .quad quiet_ni_syscall /* streams2 */
+ .quad stub32_vfork /* 190 */
+ .quad compat_sys_getrlimit
+ .quad sys32_mmap2
+ .quad sys32_truncate64
+ .quad sys32_ftruncate64
+ .quad sys32_stat64 /* 195 */
+ .quad sys32_lstat64
+ .quad sys32_fstat64
+ .quad sys_lchown
+ .quad sys_getuid
+ .quad sys_getgid /* 200 */
+ .quad sys_geteuid
+ .quad sys_getegid
+ .quad sys_setreuid
+ .quad sys_setregid
+ .quad sys_getgroups /* 205 */
+ .quad sys_setgroups
+ .quad sys_fchown
+ .quad sys_setresuid
+ .quad sys_getresuid
+ .quad sys_setresgid /* 210 */
+ .quad sys_getresgid
+ .quad sys_chown
+ .quad sys_setuid
+ .quad sys_setgid
+ .quad sys_setfsuid /* 215 */
+ .quad sys_setfsgid
+ .quad sys_pivot_root
+ .quad sys_mincore
+ .quad sys_madvise
+ .quad compat_sys_getdents64 /* 220 getdents64 */
+ .quad compat_sys_fcntl64
+ .quad quiet_ni_syscall /* tux */
+ .quad quiet_ni_syscall /* security */
+ .quad sys_gettid
+ .quad sys_readahead /* 225 */
+ .quad sys_setxattr
+ .quad sys_lsetxattr
+ .quad sys_fsetxattr
+ .quad sys_getxattr
+ .quad sys_lgetxattr /* 230 */
+ .quad sys_fgetxattr
+ .quad sys_listxattr
+ .quad sys_llistxattr
+ .quad sys_flistxattr
+ .quad sys_removexattr /* 235 */
+ .quad sys_lremovexattr
+ .quad sys_fremovexattr
+ .quad sys_tkill
+ .quad sys_sendfile64
+ .quad compat_sys_futex /* 240 */
+ .quad compat_sys_sched_setaffinity
+ .quad compat_sys_sched_getaffinity
+ .quad sys32_set_thread_area
+ .quad sys32_get_thread_area
+ .quad compat_sys_io_setup /* 245 */
+ .quad sys_io_destroy
+ .quad compat_sys_io_getevents
+ .quad compat_sys_io_submit
+ .quad sys_io_cancel
+ .quad sys_fadvise64 /* 250 */
+ .quad quiet_ni_syscall /* free_huge_pages */
+ .quad sys_exit_group
+ .quad sys32_lookup_dcookie
+ .quad sys_epoll_create
+ .quad sys_epoll_ctl /* 255 */
+ .quad sys_epoll_wait
+ .quad sys_remap_file_pages
+ .quad sys_set_tid_address
+ .quad compat_sys_timer_create
+ .quad compat_sys_timer_settime /* 260 */
+ .quad compat_sys_timer_gettime
+ .quad sys_timer_getoverrun
+ .quad sys_timer_delete
+ .quad compat_sys_clock_settime
+ .quad compat_sys_clock_gettime /* 265 */
+ .quad compat_sys_clock_getres
+ .quad compat_sys_clock_nanosleep
+ .quad compat_sys_statfs64
+ .quad compat_sys_fstatfs64
+ .quad sys_tgkill /* 270 */
+ .quad compat_sys_utimes
+ .quad sys32_fadvise64_64
+ .quad quiet_ni_syscall /* sys_vserver */
+ .quad sys_mbind
+ .quad compat_sys_get_mempolicy /* 275 */
+ .quad sys_set_mempolicy
+ .quad compat_sys_mq_open
+ .quad sys_mq_unlink
+ .quad compat_sys_mq_timedsend
+ .quad compat_sys_mq_timedreceive /* 280 */
+ .quad compat_sys_mq_notify
+ .quad compat_sys_mq_getsetattr
+ .quad compat_sys_kexec_load /* reserved for kexec */
+ .quad compat_sys_waitid
+ .quad quiet_ni_syscall /* 285: sys_altroot */
+ .quad sys_add_key
+ .quad sys_request_key
+ .quad sys_keyctl
+ .quad sys_ioprio_set
+ .quad sys_ioprio_get /* 290 */
+ .quad sys_inotify_init
+ .quad sys_inotify_add_watch
+ .quad sys_inotify_rm_watch
+ .quad sys_migrate_pages
+ .quad compat_sys_openat /* 295 */
+ .quad sys_mkdirat
+ .quad sys_mknodat
+ .quad sys_fchownat
+ .quad compat_sys_futimesat
+ .quad sys32_fstatat /* 300 */
+ .quad sys_unlinkat
+ .quad sys_renameat
+ .quad sys_linkat
+ .quad sys_symlinkat
+ .quad sys_readlinkat /* 305 */
+ .quad sys_fchmodat
+ .quad sys_faccessat
+ .quad quiet_ni_syscall /* pselect6 for now */
+ .quad quiet_ni_syscall /* ppoll for now */
+ .quad sys_unshare /* 310 */
+ .quad compat_sys_set_robust_list
+ .quad compat_sys_get_robust_list
+ .quad sys_splice
+ .quad sys_sync_file_range
+ .quad sys_tee
+ .quad compat_sys_vmsplice
+ .quad compat_sys_move_pages
+ia32_syscall_end:
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/kernel/acpi/sleep_64-xen.c 2008-04-15 09:29:41.000000000 +0200
@@ -0,0 +1,146 @@
+/*
+ * acpi.c - Architecture-Specific Low-Level ACPI Support
+ *
+ * Copyright (C) 2001, 2002 Paul Diefenbaugh <paul.s.diefenbaugh@intel.com>
+ * Copyright (C) 2001 Jun Nakajima <jun.nakajima@intel.com>
+ * Copyright (C) 2001 Patrick Mochel <mochel@osdl.org>
+ * Copyright (C) 2002 Andi Kleen, SuSE Labs (x86-64 port)
+ * Copyright (C) 2003 Pavel Machek, SuSE Labs
+ *
+ * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ *
+ * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+ */
+
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/types.h>
+#include <linux/stddef.h>
+#include <linux/slab.h>
+#include <linux/pci.h>
+#include <linux/bootmem.h>
+#include <linux/acpi.h>
+#include <linux/cpumask.h>
+
+#include <asm/mpspec.h>
+#include <asm/io.h>
+#include <asm/apic.h>
+#include <asm/apicdef.h>
+#include <asm/page.h>
+#include <asm/pgtable.h>
+#include <asm/pgalloc.h>
+#include <asm/io_apic.h>
+#include <asm/proto.h>
+#include <asm/tlbflush.h>
+
+/* --------------------------------------------------------------------------
+ Low-Level Sleep Support
+ -------------------------------------------------------------------------- */
+
+#ifdef CONFIG_ACPI_SLEEP
+
+#ifndef CONFIG_ACPI_PV_SLEEP
+/* address in low memory of the wakeup routine. */
+unsigned long acpi_wakeup_address = 0;
+unsigned long acpi_video_flags;
+extern char wakeup_start, wakeup_end;
+
+extern unsigned long FASTCALL(acpi_copy_wakeup_routine(unsigned long));
+
+static pgd_t low_ptr;
+
+static void init_low_mapping(void)
+{
+ pgd_t *slot0 = pgd_offset(current->mm, 0UL);
+ low_ptr = *slot0;
+ set_pgd(slot0, *pgd_offset(current->mm, PAGE_OFFSET));
+ WARN_ON(num_online_cpus() != 1);
+ local_flush_tlb();
+}
+#endif
+
+/**
+ * acpi_save_state_mem - save kernel state
+ *
+ * Create an identity mapped page table and copy the wakeup routine to
+ * low memory.
+ */
+int acpi_save_state_mem(void)
+{
+#ifndef CONFIG_ACPI_PV_SLEEP
+ init_low_mapping();
+
+ memcpy((void *)acpi_wakeup_address, &wakeup_start,
+ &wakeup_end - &wakeup_start);
+ acpi_copy_wakeup_routine(acpi_wakeup_address);
+#endif
+ return 0;
+}
+
+/*
+ * acpi_restore_state
+ */
+void acpi_restore_state_mem(void)
+{
+#ifndef CONFIG_ACPI_PV_SLEEP
+ set_pgd(pgd_offset(current->mm, 0UL), low_ptr);
+ local_flush_tlb();
+#endif
+}
+
+/**
+ * acpi_reserve_bootmem - do _very_ early ACPI initialisation
+ *
+ * We allocate a page in low memory for the wakeup
+ * routine for when we come back from a sleep state. The
+ * runtime allocator allows specification of <16M pages, but not
+ * <1M pages.
+ */
+void __init acpi_reserve_bootmem(void)
+{
+#ifndef CONFIG_ACPI_PV_SLEEP
+ acpi_wakeup_address = (unsigned long)alloc_bootmem_low(PAGE_SIZE);
+ if ((&wakeup_end - &wakeup_start) > PAGE_SIZE)
+ printk(KERN_CRIT
+ "ACPI: Wakeup code way too big, will crash on attempt to suspend\n");
+#endif
+}
+
+#ifndef CONFIG_ACPI_PV_SLEEP
+static int __init acpi_sleep_setup(char *str)
+{
+ while ((str != NULL) && (*str != '\0')) {
+ if (strncmp(str, "s3_bios", 7) == 0)
+ acpi_video_flags = 1;
+ if (strncmp(str, "s3_mode", 7) == 0)
+ acpi_video_flags |= 2;
+ str = strchr(str, ',');
+ if (str != NULL)
+ str += strspn(str, ", \t");
+ }
+
+ return 1;
+}
+
+__setup("acpi_sleep=", acpi_sleep_setup);
+#endif /* CONFIG_ACPI_PV_SLEEP */
+
+#endif /*CONFIG_ACPI_SLEEP */
+
+void acpi_pci_link_exit(void)
+{
+}
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/kernel/e820_64-xen.c 2009-12-04 08:45:56.000000000 +0100
@@ -0,0 +1,800 @@
+/*
+ * Handle the memory map.
+ * The functions here do the job until bootmem takes over.
+ *
+ * Getting sanitize_e820_map() in sync with i386 version by applying change:
+ * - Provisions for empty E820 memory regions (reported by certain BIOSes).
+ * Alex Achenbach <xela@slit.de>, December 2002.
+ * Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
+ *
+ */
+#include <linux/kernel.h>
+#include <linux/types.h>
+#include <linux/init.h>
+#include <linux/bootmem.h>
+#include <linux/ioport.h>
+#include <linux/string.h>
+#include <linux/kexec.h>
+#include <linux/module.h>
+
+#include <asm/pgtable.h>
+#include <asm/page.h>
+#include <asm/e820.h>
+#include <asm/proto.h>
+#include <asm/bootsetup.h>
+#include <asm/sections.h>
+#include <xen/interface/memory.h>
+
+/*
+ * PFN of last memory page.
+ */
+unsigned long end_pfn;
+EXPORT_SYMBOL(end_pfn);
+
+#ifndef CONFIG_XEN
+/*
+ * end_pfn only includes RAM, while end_pfn_map includes all e820 entries.
+ * The direct mapping extends to end_pfn_map, so that we can directly access
+ * apertures, ACPI and other tables without having to play with fixmaps.
+ */
+unsigned long end_pfn_map;
+#endif
+
+/*
+ * Last pfn which the user wants to use.
+ */
+unsigned long end_user_pfn = MAXMEM>>PAGE_SHIFT;
+
+extern struct resource code_resource, data_resource;
+
+#ifdef CONFIG_XEN
+extern struct e820map machine_e820;
+#endif
+
+/* Check for some hardcoded bad areas that early boot is not allowed to touch */
+static inline int bad_addr(unsigned long *addrp, unsigned long size)
+{
+ unsigned long addr = *addrp, last = addr + size;
+
+#ifndef CONFIG_XEN
+ /* various gunk below that needed for SMP startup */
+ if (addr < 0x8000) {
+ *addrp = 0x8000;
+ return 1;
+ }
+
+ /* direct mapping tables of the kernel */
+ if (last >= table_start<<PAGE_SHIFT && addr < table_end<<PAGE_SHIFT) {
+ *addrp = table_end << PAGE_SHIFT;
+ return 1;
+ }
+
+ /* initrd */
+#ifdef CONFIG_BLK_DEV_INITRD
+ if (LOADER_TYPE && INITRD_START && last >= INITRD_START &&
+ addr < INITRD_START+INITRD_SIZE) {
+ *addrp = INITRD_START + INITRD_SIZE;
+ return 1;
+ }
+#endif
+ /* kernel code + 640k memory hole (later should not be needed, but
+ be paranoid for now) */
+ if (last >= 640*1024 && addr < 1024*1024) {
+ *addrp = 1024*1024;
+ return 1;
+ }
+ if (last >= __pa_symbol(&_text) && last < __pa_symbol(&_end)) {
+ *addrp = __pa_symbol(&_end);
+ return 1;
+ }
+
+ if (last >= ebda_addr && addr < ebda_addr + ebda_size) {
+ *addrp = ebda_addr + ebda_size;
+ return 1;
+ }
+
+ /* XXX ramdisk image here? */
+#else
+ if (last < (table_end<<PAGE_SHIFT)) {
+ *addrp = table_end << PAGE_SHIFT;
+ return 1;
+ }
+#endif
+ return 0;
+}
+
+/*
+ * This function checks if any part of the range <start,end> is mapped
+ * with type.
+ */
+int e820_any_mapped(unsigned long start, unsigned long end, unsigned type)
+{
+ int i;
+
+#ifndef CONFIG_XEN
+ for (i = 0; i < e820.nr_map; i++) {
+ struct e820entry *ei = &e820.map[i];
+#else
+ if (!is_initial_xendomain())
+ return 0;
+ for (i = 0; i < machine_e820.nr_map; i++) {
+ const struct e820entry *ei = &machine_e820.map[i];
+#endif
+
+ if (type && ei->type != type)
+ continue;
+ if (ei->addr >= end || ei->addr + ei->size <= start)
+ continue;
+ return 1;
+ }
+ return 0;
+}
+EXPORT_SYMBOL_GPL(e820_any_mapped);
+
+/*
+ * This function checks if the entire range <start,end> is mapped with type.
+ *
+ * Note: this function only works correct if the e820 table is sorted and
+ * not-overlapping, which is the case
+ */
+int __init e820_all_mapped(unsigned long start, unsigned long end, unsigned type)
+{
+ int i;
+
+#ifndef CONFIG_XEN
+ for (i = 0; i < e820.nr_map; i++) {
+ struct e820entry *ei = &e820.map[i];
+#else
+ if (!is_initial_xendomain())
+ return 0;
+ for (i = 0; i < machine_e820.nr_map; i++) {
+ const struct e820entry *ei = &machine_e820.map[i];
+#endif
+
+ if (type && ei->type != type)
+ continue;
+ /* is the region (part) in overlap with the current region ?*/
+ if (ei->addr >= end || ei->addr + ei->size <= start)
+ continue;
+
+ /* if the region is at the beginning of <start,end> we move
+ * start to the end of the region since it's ok until there
+ */
+ if (ei->addr <= start)
+ start = ei->addr + ei->size;
+ /* if start is now at or beyond end, we're done, full coverage */
+ if (start >= end)
+ return 1; /* we're done */
+ }
+ return 0;
+}
+
+/*
+ * Find a free area in a specific range.
+ */
+unsigned long __init find_e820_area(unsigned long start, unsigned long end, unsigned size)
+{
+ int i;
+ for (i = 0; i < e820.nr_map; i++) {
+ struct e820entry *ei = &e820.map[i];
+ unsigned long addr = ei->addr, last;
+ if (ei->type != E820_RAM)
+ continue;
+ if (addr < start)
+ addr = start;
+ if (addr > ei->addr + ei->size)
+ continue;
+ while (bad_addr(&addr, size) && addr+size <= ei->addr+ei->size)
+ ;
+ last = addr + size;
+ if (last > ei->addr + ei->size)
+ continue;
+ if (last > end)
+ continue;
+ return addr;
+ }
+ return -1UL;
+}
+
+/*
+ * Free bootmem based on the e820 table for a node.
+ */
+void __init e820_bootmem_free(pg_data_t *pgdat, unsigned long start,unsigned long end)
+{
+ int i;
+ for (i = 0; i < e820.nr_map; i++) {
+ struct e820entry *ei = &e820.map[i];
+ unsigned long last, addr;
+
+ if (ei->type != E820_RAM ||
+ ei->addr+ei->size <= start ||
+ ei->addr >= end)
+ continue;
+
+ addr = round_up(ei->addr, PAGE_SIZE);
+ if (addr < start)
+ addr = start;
+
+ last = round_down(ei->addr + ei->size, PAGE_SIZE);
+ if (last >= end)
+ last = end;
+
+ if (last > addr && last-addr >= PAGE_SIZE)
+ free_bootmem_node(pgdat, addr, last-addr);
+ }
+}
+
+/*
+ * Find the highest page frame number we have available
+ */
+unsigned long __init e820_end_of_ram(void)
+{
+ int i;
+ unsigned long end_pfn = 0;
+
+ for (i = 0; i < e820.nr_map; i++) {
+ struct e820entry *ei = &e820.map[i];
+ unsigned long start, end;
+
+ start = round_up(ei->addr, PAGE_SIZE);
+ end = round_down(ei->addr + ei->size, PAGE_SIZE);
+ if (start >= end)
+ continue;
+ if (ei->type == E820_RAM) {
+ if (end > end_pfn<<PAGE_SHIFT)
+ end_pfn = end>>PAGE_SHIFT;
+ } else {
+ if (end > end_pfn_map<<PAGE_SHIFT)
+ end_pfn_map = end>>PAGE_SHIFT;
+ }
+ }
+
+ if (end_pfn > end_pfn_map)
+ end_pfn_map = end_pfn;
+ if (end_pfn_map > MAXMEM>>PAGE_SHIFT)
+ end_pfn_map = MAXMEM>>PAGE_SHIFT;
+ if (end_pfn > end_user_pfn)
+ end_pfn = end_user_pfn;
+ if (end_pfn > end_pfn_map)
+ end_pfn = end_pfn_map;
+
+ return end_pfn;
+}
+
+/*
+ * Compute how much memory is missing in a range.
+ * Unlike the other functions in this file the arguments are in page numbers.
+ */
+unsigned long __init
+e820_hole_size(unsigned long start_pfn, unsigned long end_pfn)
+{
+ unsigned long ram = 0;
+ unsigned long start = start_pfn << PAGE_SHIFT;
+ unsigned long end = end_pfn << PAGE_SHIFT;
+ int i;
+ for (i = 0; i < e820.nr_map; i++) {
+ struct e820entry *ei = &e820.map[i];
+ unsigned long last, addr;
+
+ if (ei->type != E820_RAM ||
+ ei->addr+ei->size <= start ||
+ ei->addr >= end)
+ continue;
+
+ addr = round_up(ei->addr, PAGE_SIZE);
+ if (addr < start)
+ addr = start;
+
+ last = round_down(ei->addr + ei->size, PAGE_SIZE);
+ if (last >= end)
+ last = end;
+
+ if (last > addr)
+ ram += last - addr;
+ }
+ return ((end - start) - ram) >> PAGE_SHIFT;
+}
+
+/*
+ * Mark e820 reserved areas as busy for the resource manager.
+ */
+void __init e820_reserve_resources(struct e820entry *e820, int nr_map)
+{
+ int i;
+ for (i = 0; i < nr_map; i++) {
+ struct resource *res;
+ res = alloc_bootmem_low(sizeof(struct resource));
+ switch (e820[i].type) {
+ case E820_RAM: res->name = "System RAM"; break;
+ case E820_ACPI: res->name = "ACPI Tables"; break;
+ case E820_NVS: res->name = "ACPI Non-volatile Storage"; break;
+ default: res->name = "reserved";
+ }
+ res->start = e820[i].addr;
+ res->end = res->start + e820[i].size - 1;
+ res->flags = IORESOURCE_MEM | IORESOURCE_BUSY;
+ request_resource(&iomem_resource, res);
+ if (e820[i].type == E820_RAM) {
+ /*
+ * We don't know which RAM region contains kernel data,
+ * so we try it repeatedly and let the resource manager
+ * test it.
+ */
+#ifndef CONFIG_XEN
+ request_resource(res, &code_resource);
+ request_resource(res, &data_resource);
+#endif
+#ifdef CONFIG_KEXEC
+ if (crashk_res.start != crashk_res.end)
+ request_resource(res, &crashk_res);
+#ifdef CONFIG_XEN
+ xen_machine_kexec_register_resources(res);
+#endif
+#endif
+ }
+ }
+}
+
+/*
+ * Add a memory region to the kernel e820 map.
+ */
+void __init add_memory_region(unsigned long start, unsigned long size, int type)
+{
+ int x = e820.nr_map;
+
+ if (x == E820MAX) {
+ printk(KERN_ERR "Ooops! Too many entries in the memory map!\n");
+ return;
+ }
+
+ e820.map[x].addr = start;
+ e820.map[x].size = size;
+ e820.map[x].type = type;
+ e820.nr_map++;
+}
+
+void __init e820_print_map(char *who)
+{
+ int i;
+
+ for (i = 0; i < e820.nr_map; i++) {
+ printk(" %s: %016Lx - %016Lx ", who,
+ (unsigned long long) e820.map[i].addr,
+ (unsigned long long) (e820.map[i].addr + e820.map[i].size));
+ switch (e820.map[i].type) {
+ case E820_RAM: printk("(usable)\n");
+ break;
+ case E820_RESERVED:
+ printk("(reserved)\n");
+ break;
+ case E820_ACPI:
+ printk("(ACPI data)\n");
+ break;
+ case E820_NVS:
+ printk("(ACPI NVS)\n");
+ break;
+ default: printk("type %u\n", e820.map[i].type);
+ break;
+ }
+ }
+}
+
+/*
+ * Sanitize the BIOS e820 map.
+ *
+ * Some e820 responses include overlapping entries. The following
+ * replaces the original e820 map with a new one, removing overlaps.
+ *
+ */
+static int __init sanitize_e820_map(struct e820entry * biosmap, char * pnr_map)
+{
+ struct change_member {
+ struct e820entry *pbios; /* pointer to original bios entry */
+ unsigned long long addr; /* address for this change point */
+ };
+ static struct change_member change_point_list[2*E820MAX] __initdata;
+ static struct change_member *change_point[2*E820MAX] __initdata;
+ static struct e820entry *overlap_list[E820MAX] __initdata;
+ static struct e820entry new_bios[E820MAX] __initdata;
+ struct change_member *change_tmp;
+ unsigned long current_type, last_type;
+ unsigned long long last_addr;
+ int chgidx, still_changing;
+ int overlap_entries;
+ int new_bios_entry;
+ int old_nr, new_nr, chg_nr;
+ int i;
+
+ /*
+ Visually we're performing the following (1,2,3,4 = memory types)...
+
+ Sample memory map (w/overlaps):
+ ____22__________________
+ ______________________4_
+ ____1111________________
+ _44_____________________
+ 11111111________________
+ ____________________33__
+ ___________44___________
+ __________33333_________
+ ______________22________
+ ___________________2222_
+ _________111111111______
+ _____________________11_
+ _________________4______
+
+ Sanitized equivalent (no overlap):
+ 1_______________________
+ _44_____________________
+ ___1____________________
+ ____22__________________
+ ______11________________
+ _________1______________
+ __________3_____________
+ ___________44___________
+ _____________33_________
+ _______________2________
+ ________________1_______
+ _________________4______
+ ___________________2____
+ ____________________33__
+ ______________________4_
+ */
+
+ /* if there's only one memory region, don't bother */
+ if (*pnr_map < 2)
+ return -1;
+
+ old_nr = *pnr_map;
+
+ /* bail out if we find any unreasonable addresses in bios map */
+ for (i=0; i<old_nr; i++)
+ if (biosmap[i].addr + biosmap[i].size < biosmap[i].addr)
+ return -1;
+
+ /* create pointers for initial change-point information (for sorting) */
+ for (i=0; i < 2*old_nr; i++)
+ change_point[i] = &change_point_list[i];
+
+ /* record all known change-points (starting and ending addresses),
+ omitting those that are for empty memory regions */
+ chgidx = 0;
+ for (i=0; i < old_nr; i++) {
+ if (biosmap[i].size != 0) {
+ change_point[chgidx]->addr = biosmap[i].addr;
+ change_point[chgidx++]->pbios = &biosmap[i];
+ change_point[chgidx]->addr = biosmap[i].addr + biosmap[i].size;
+ change_point[chgidx++]->pbios = &biosmap[i];
+ }
+ }
+ chg_nr = chgidx;
+
+ /* sort change-point list by memory addresses (low -> high) */
+ still_changing = 1;
+ while (still_changing) {
+ still_changing = 0;
+ for (i=1; i < chg_nr; i++) {
+ /* if <current_addr> > <last_addr>, swap */
+ /* or, if current=<start_addr> & last=<end_addr>, swap */
+ if ((change_point[i]->addr < change_point[i-1]->addr) ||
+ ((change_point[i]->addr == change_point[i-1]->addr) &&
+ (change_point[i]->addr == change_point[i]->pbios->addr) &&
+ (change_point[i-1]->addr != change_point[i-1]->pbios->addr))
+ )
+ {
+ change_tmp = change_point[i];
+ change_point[i] = change_point[i-1];
+ change_point[i-1] = change_tmp;
+ still_changing=1;
+ }
+ }
+ }
+
+ /* create a new bios memory map, removing overlaps */
+ overlap_entries=0; /* number of entries in the overlap table */
+ new_bios_entry=0; /* index for creating new bios map entries */
+ last_type = 0; /* start with undefined memory type */
+ last_addr = 0; /* start with 0 as last starting address */
+ /* loop through change-points, determining affect on the new bios map */
+ for (chgidx=0; chgidx < chg_nr; chgidx++)
+ {
+ /* keep track of all overlapping bios entries */
+ if (change_point[chgidx]->addr == change_point[chgidx]->pbios->addr)
+ {
+ /* add map entry to overlap list (> 1 entry implies an overlap) */
+ overlap_list[overlap_entries++]=change_point[chgidx]->pbios;
+ }
+ else
+ {
+ /* remove entry from list (order independent, so swap with last) */
+ for (i=0; i<overlap_entries; i++)
+ {
+ if (overlap_list[i] == change_point[chgidx]->pbios)
+ overlap_list[i] = overlap_list[overlap_entries-1];
+ }
+ overlap_entries--;
+ }
+ /* if there are overlapping entries, decide which "type" to use */
+ /* (larger value takes precedence -- 1=usable, 2,3,4,4+=unusable) */
+ current_type = 0;
+ for (i=0; i<overlap_entries; i++)
+ if (overlap_list[i]->type > current_type)
+ current_type = overlap_list[i]->type;
+ /* continue building up new bios map based on this information */
+ if (current_type != last_type) {
+ if (last_type != 0) {
+ new_bios[new_bios_entry].size =
+ change_point[chgidx]->addr - last_addr;
+ /* move forward only if the new size was non-zero */
+ if (new_bios[new_bios_entry].size != 0)
+ if (++new_bios_entry >= E820MAX)
+ break; /* no more space left for new bios entries */
+ }
+ if (current_type != 0) {
+ new_bios[new_bios_entry].addr = change_point[chgidx]->addr;
+ new_bios[new_bios_entry].type = current_type;
+ last_addr=change_point[chgidx]->addr;
+ }
+ last_type = current_type;
+ }
+ }
+ new_nr = new_bios_entry; /* retain count for new bios entries */
+
+ /* copy new bios mapping into original location */
+ memcpy(biosmap, new_bios, new_nr*sizeof(struct e820entry));
+ *pnr_map = new_nr;
+
+ return 0;
+}
+
+/*
+ * Copy the BIOS e820 map into a safe place.
+ *
+ * Sanity-check it while we're at it..
+ *
+ * If we're lucky and live on a modern system, the setup code
+ * will have given us a memory map that we can use to properly
+ * set up memory. If we aren't, we'll fake a memory map.
+ *
+ * We check to see that the memory map contains at least 2 elements
+ * before we'll use it, because the detection code in setup.S may
+ * not be perfect and most every PC known to man has two memory
+ * regions: one from 0 to 640k, and one from 1mb up. (The IBM
+ * thinkpad 560x, for example, does not cooperate with the memory
+ * detection code.)
+ */
+static int __init copy_e820_map(struct e820entry * biosmap, int nr_map)
+{
+#ifndef CONFIG_XEN
+ /* Only one memory region (or negative)? Ignore it */
+ if (nr_map < 2)
+ return -1;
+#else
+ BUG_ON(nr_map < 1);
+#endif
+
+ do {
+ unsigned long start = biosmap->addr;
+ unsigned long size = biosmap->size;
+ unsigned long end = start + size;
+ unsigned long type = biosmap->type;
+
+ /* Overflow in 64 bits? Ignore the memory map. */
+ if (start > end)
+ return -1;
+
+#ifndef CONFIG_XEN
+ /*
+ * Some BIOSes claim RAM in the 640k - 1M region.
+ * Not right. Fix it up.
+ *
+ * This should be removed on Hammer which is supposed to not
+ * have non e820 covered ISA mappings there, but I had some strange
+ * problems so it stays for now. -AK
+ */
+ if (type == E820_RAM) {
+ if (start < 0x100000ULL && end > 0xA0000ULL) {
+ if (start < 0xA0000ULL)
+ add_memory_region(start, 0xA0000ULL-start, type);
+ if (end <= 0x100000ULL)
+ continue;
+ start = 0x100000ULL;
+ size = end - start;
+ }
+ }
+#endif
+
+ add_memory_region(start, size, type);
+ } while (biosmap++,--nr_map);
+
+#ifdef CONFIG_XEN
+ if (is_initial_xendomain()) {
+ struct xen_memory_map memmap;
+
+ memmap.nr_entries = E820MAX;
+ set_xen_guest_handle(memmap.buffer, machine_e820.map);
+
+ if (HYPERVISOR_memory_op(XENMEM_machine_memory_map, &memmap))
+ BUG();
+ machine_e820.nr_map = memmap.nr_entries;
+ } else
+ machine_e820 = e820;
+#endif
+
+ return 0;
+}
+
+#ifndef CONFIG_XEN
+void __init setup_memory_region(void)
+{
+ char *who = "BIOS-e820";
+
+ /*
+ * Try to copy the BIOS-supplied E820-map.
+ *
+ * Otherwise fake a memory map; one section from 0k->640k,
+ * the next section from 1mb->appropriate_mem_k
+ */
+ sanitize_e820_map(E820_MAP, &E820_MAP_NR);
+ if (copy_e820_map(E820_MAP, E820_MAP_NR) < 0) {
+ unsigned long mem_size;
+
+ /* compare results from other methods and take the greater */
+ if (ALT_MEM_K < EXT_MEM_K) {
+ mem_size = EXT_MEM_K;
+ who = "BIOS-88";
+ } else {
+ mem_size = ALT_MEM_K;
+ who = "BIOS-e801";
+ }
+
+ e820.nr_map = 0;
+ add_memory_region(0, LOWMEMSIZE(), E820_RAM);
+ add_memory_region(HIGH_MEMORY, mem_size << 10, E820_RAM);
+ }
+ printk(KERN_INFO "BIOS-provided physical RAM map:\n");
+ e820_print_map(who);
+}
+
+#else /* CONFIG_XEN */
+
+void __init setup_memory_region(void)
+{
+ int rc;
+ struct xen_memory_map memmap;
+ /*
+ * This is rather large for a stack variable but this early in
+ * the boot process we know we have plenty slack space.
+ */
+ struct e820entry map[E820MAX];
+
+ memmap.nr_entries = E820MAX;
+ set_xen_guest_handle(memmap.buffer, map);
+
+ rc = HYPERVISOR_memory_op(XENMEM_memory_map, &memmap);
+ if ( rc == -ENOSYS ) {
+ memmap.nr_entries = 1;
+ map[0].addr = 0ULL;
+ map[0].size = xen_start_info->nr_pages << PAGE_SHIFT;
+ /* 8MB slack (to balance backend allocations). */
+ map[0].size += 8 << 20;
+ map[0].type = E820_RAM;
+ rc = 0;
+ }
+ BUG_ON(rc);
+
+ sanitize_e820_map(map, (char *)&memmap.nr_entries);
+
+ BUG_ON(copy_e820_map(map, (char)memmap.nr_entries) < 0);
+
+ printk(KERN_INFO "BIOS-provided physical RAM map:\n");
+ e820_print_map("Xen");
+}
+#endif
+
+void __init parse_memopt(char *p, char **from)
+{
+ int i;
+ unsigned long current_end;
+ unsigned long end;
+
+ end_user_pfn = memparse(p, from);
+ end_user_pfn >>= PAGE_SHIFT;
+
+ end = end_user_pfn<<PAGE_SHIFT;
+ i = e820.nr_map-1;
+ current_end = e820.map[i].addr + e820.map[i].size;
+
+ if (current_end < end) {
+ /*
+ * The e820 map ends before our requested size so
+ * extend the final entry to the requested address.
+ */
+ if (e820.map[i].type == E820_RAM)
+ e820.map[i].size = end - e820.map[i].addr;
+ else
+ add_memory_region(current_end, end - current_end, E820_RAM);
+ }
+}
+
+void __init parse_memmapopt(char *p, char **from)
+{
+ unsigned long long start_at, mem_size;
+
+ mem_size = memparse(p, from);
+ p = *from;
+ if (*p == '@') {
+ start_at = memparse(p+1, from);
+ add_memory_region(start_at, mem_size, E820_RAM);
+ } else if (*p == '#') {
+ start_at = memparse(p+1, from);
+ add_memory_region(start_at, mem_size, E820_ACPI);
+ } else if (*p == '$') {
+ start_at = memparse(p+1, from);
+ add_memory_region(start_at, mem_size, E820_RESERVED);
+ } else {
+ end_user_pfn = (mem_size >> PAGE_SHIFT);
+ }
+ p = *from;
+}
+
+unsigned long pci_mem_start = 0xaeedbabe;
+EXPORT_SYMBOL(pci_mem_start);
+
+/*
+ * Search for the biggest gap in the low 32 bits of the e820
+ * memory space. We pass this space to PCI to assign MMIO resources
+ * for hotplug or unconfigured devices in.
+ * Hopefully the BIOS let enough space left.
+ */
+__init void e820_setup_gap(struct e820entry *e820, int nr_map)
+{
+ unsigned long gapstart, gapsize, round;
+ unsigned long last;
+ int i;
+ int found = 0;
+
+ last = 0x100000000ull;
+ gapstart = 0x10000000;
+ gapsize = 0x400000;
+ i = nr_map;
+ while (--i >= 0) {
+ unsigned long long start = e820[i].addr;
+ unsigned long long end = start + e820[i].size;
+
+ /*
+ * Since "last" is at most 4GB, we know we'll
+ * fit in 32 bits if this condition is true
+ */
+ if (last > end) {
+ unsigned long gap = last - end;
+
+ if (gap > gapsize) {
+ gapsize = gap;
+ gapstart = end;
+ found = 1;
+ }
+ }
+ if (start < last)
+ last = start;
+ }
+
+ if (!found) {
+ gapstart = (end_pfn << PAGE_SHIFT) + 1024*1024;
+ printk(KERN_ERR "PCI: Warning: Cannot find a gap in the 32bit address range\n"
+ KERN_ERR "PCI: Unassigned devices with 32bit resource registers may break!\n");
+ }
+
+ /*
+ * See how much we want to round up: start off with
+ * rounding to the next 1MB area.
+ */
+ round = 0x100000;
+ while ((gapsize >> 4) > round)
+ round += round;
+ /* Fun with two's complement */
+ pci_mem_start = (gapstart + round) & -round;
+
+ printk(KERN_INFO "Allocating PCI resources starting at %lx (gap: %lx:%lx)\n",
+ pci_mem_start, gapstart, gapsize);
+}
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/kernel/early_printk-xen.c 2007-06-12 13:13:01.000000000 +0200
@@ -0,0 +1,302 @@
+#include <linux/console.h>
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/string.h>
+#include <linux/screen_info.h>
+#include <asm/io.h>
+#include <asm/processor.h>
+#include <asm/fcntl.h>
+
+/* Simple VGA output */
+
+#ifdef __i386__
+#include <asm/setup.h>
+#define VGABASE (__ISA_IO_base + 0xb8000)
+#else
+#include <asm/bootsetup.h>
+#define VGABASE ((void __iomem *)0xffffffff800b8000UL)
+#endif
+
+#ifndef CONFIG_XEN
+static int max_ypos = 25, max_xpos = 80;
+static int current_ypos = 25, current_xpos = 0;
+
+static void early_vga_write(struct console *con, const char *str, unsigned n)
+{
+ char c;
+ int i, k, j;
+
+ while ((c = *str++) != '\0' && n-- > 0) {
+ if (current_ypos >= max_ypos) {
+ /* scroll 1 line up */
+ for (k = 1, j = 0; k < max_ypos; k++, j++) {
+ for (i = 0; i < max_xpos; i++) {
+ writew(readw(VGABASE+2*(max_xpos*k+i)),
+ VGABASE + 2*(max_xpos*j + i));
+ }
+ }
+ for (i = 0; i < max_xpos; i++)
+ writew(0x720, VGABASE + 2*(max_xpos*j + i));
+ current_ypos = max_ypos-1;
+ }
+ if (c == '\n') {
+ current_xpos = 0;
+ current_ypos++;
+ } else if (c != '\r') {
+ writew(((0x7 << 8) | (unsigned short) c),
+ VGABASE + 2*(max_xpos*current_ypos +
+ current_xpos++));
+ if (current_xpos >= max_xpos) {
+ current_xpos = 0;
+ current_ypos++;
+ }
+ }
+ }
+}
+
+static struct console early_vga_console = {
+ .name = "earlyvga",
+ .write = early_vga_write,
+ .flags = CON_PRINTBUFFER,
+ .index = -1,
+};
+
+/* Serial functions loosely based on a similar package from Klaus P. Gerlicher */
+
+static int early_serial_base = 0x3f8; /* ttyS0 */
+
+#define XMTRDY 0x20
+
+#define DLAB 0x80
+
+#define TXR 0 /* Transmit register (WRITE) */
+#define RXR 0 /* Receive register (READ) */
+#define IER 1 /* Interrupt Enable */
+#define IIR 2 /* Interrupt ID */
+#define FCR 2 /* FIFO control */
+#define LCR 3 /* Line control */
+#define MCR 4 /* Modem control */
+#define LSR 5 /* Line Status */
+#define MSR 6 /* Modem Status */
+#define DLL 0 /* Divisor Latch Low */
+#define DLH 1 /* Divisor latch High */
+
+static int early_serial_putc(unsigned char ch)
+{
+ unsigned timeout = 0xffff;
+ while ((inb(early_serial_base + LSR) & XMTRDY) == 0 && --timeout)
+ cpu_relax();
+ outb(ch, early_serial_base + TXR);
+ return timeout ? 0 : -1;
+}
+
+static void early_serial_write(struct console *con, const char *s, unsigned n)
+{
+ while (*s && n-- > 0) {
+ early_serial_putc(*s);
+ if (*s == '\n')
+ early_serial_putc('\r');
+ s++;
+ }
+}
+
+#define DEFAULT_BAUD 9600
+
+static __init void early_serial_init(char *s)
+{
+ unsigned char c;
+ unsigned divisor;
+ unsigned baud = DEFAULT_BAUD;
+ char *e;
+
+ if (*s == ',')
+ ++s;
+
+ if (*s) {
+ unsigned port;
+ if (!strncmp(s,"0x",2)) {
+ early_serial_base = simple_strtoul(s, &e, 16);
+ } else {
+ static int bases[] = { 0x3f8, 0x2f8 };
+
+ if (!strncmp(s,"ttyS",4))
+ s += 4;
+ port = simple_strtoul(s, &e, 10);
+ if (port > 1 || s == e)
+ port = 0;
+ early_serial_base = bases[port];
+ }
+ s += strcspn(s, ",");
+ if (*s == ',')
+ s++;
+ }
+
+ outb(0x3, early_serial_base + LCR); /* 8n1 */
+ outb(0, early_serial_base + IER); /* no interrupt */
+ outb(0, early_serial_base + FCR); /* no fifo */
+ outb(0x3, early_serial_base + MCR); /* DTR + RTS */
+
+ if (*s) {
+ baud = simple_strtoul(s, &e, 0);
+ if (baud == 0 || s == e)
+ baud = DEFAULT_BAUD;
+ }
+
+ divisor = 115200 / baud;
+ c = inb(early_serial_base + LCR);
+ outb(c | DLAB, early_serial_base + LCR);
+ outb(divisor & 0xff, early_serial_base + DLL);
+ outb((divisor >> 8) & 0xff, early_serial_base + DLH);
+ outb(c & ~DLAB, early_serial_base + LCR);
+}
+
+#else /* CONFIG_XEN */
+
+static void
+early_serial_write(struct console *con, const char *s, unsigned count)
+{
+ int n;
+
+ while (count > 0) {
+ n = HYPERVISOR_console_io(CONSOLEIO_write, count, (char *)s);
+ if (n <= 0)
+ break;
+ count -= n;
+ s += n;
+ }
+}
+
+static __init void early_serial_init(char *s)
+{
+}
+
+/*
+ * No early VGA console on Xen, as we do not have convenient ISA-space
+ * mappings. Someone should fix this for domain 0. For now, use fake serial.
+ */
+#define early_vga_console early_serial_console
+
+#endif
+
+static struct console early_serial_console = {
+ .name = "earlyser",
+ .write = early_serial_write,
+ .flags = CON_PRINTBUFFER,
+ .index = -1,
+};
+
+/* Console interface to a host file on AMD's SimNow! */
+
+static int simnow_fd;
+
+enum {
+ MAGIC1 = 0xBACCD00A,
+ MAGIC2 = 0xCA110000,
+ XOPEN = 5,
+ XWRITE = 4,
+};
+
+static noinline long simnow(long cmd, long a, long b, long c)
+{
+ long ret;
+ asm volatile("cpuid" :
+ "=a" (ret) :
+ "b" (a), "c" (b), "d" (c), "0" (MAGIC1), "D" (cmd + MAGIC2));
+ return ret;
+}
+
+void __init simnow_init(char *str)
+{
+ char *fn = "klog";
+ if (*str == '=')
+ fn = ++str;
+ /* error ignored */
+ simnow_fd = simnow(XOPEN, (unsigned long)fn, O_WRONLY|O_APPEND|O_CREAT, 0644);
+}
+
+static void simnow_write(struct console *con, const char *s, unsigned n)
+{
+ simnow(XWRITE, simnow_fd, (unsigned long)s, n);
+}
+
+static struct console simnow_console = {
+ .name = "simnow",
+ .write = simnow_write,
+ .flags = CON_PRINTBUFFER,
+ .index = -1,
+};
+
+/* Direct interface for emergencies */
+struct console *early_console = &early_vga_console;
+static int early_console_initialized = 0;
+
+void early_printk(const char *fmt, ...)
+{
+ char buf[512];
+ int n;
+ va_list ap;
+
+ va_start(ap,fmt);
+ n = vscnprintf(buf,512,fmt,ap);
+ early_console->write(early_console,buf,n);
+ va_end(ap);
+}
+
+static int __initdata keep_early;
+
+int __init setup_early_printk(char *opt)
+{
+ char *space;
+ char buf[256];
+
+ if (early_console_initialized)
+ return 1;
+
+ strlcpy(buf,opt,sizeof(buf));
+ space = strchr(buf, ' ');
+ if (space)
+ *space = 0;
+
+ if (strstr(buf,"keep"))
+ keep_early = 1;
+
+ if (!strncmp(buf, "serial", 6)) {
+ early_serial_init(buf + 6);
+ early_console = &early_serial_console;
+ } else if (!strncmp(buf, "ttyS", 4)) {
+ early_serial_init(buf);
+ early_console = &early_serial_console;
+ } else if (!strncmp(buf, "vga", 3)
+#ifndef CONFIG_XEN
+ && SCREEN_INFO.orig_video_isVGA == 1) {
+ max_xpos = SCREEN_INFO.orig_video_cols;
+ max_ypos = SCREEN_INFO.orig_video_lines;
+ current_ypos = SCREEN_INFO.orig_y;
+#else
+ || !strncmp(buf, "xen", 3)) {
+#endif
+ early_console = &early_vga_console;
+ } else if (!strncmp(buf, "simnow", 6)) {
+ simnow_init(buf + 6);
+ early_console = &simnow_console;
+ keep_early = 1;
+ }
+ early_console_initialized = 1;
+ register_console(early_console);
+ return 0;
+}
+
+void __init disable_early_printk(void)
+{
+ if (!early_console_initialized || !early_console)
+ return;
+ if (!keep_early) {
+ printk("disabling early console\n");
+ unregister_console(early_console);
+ early_console_initialized = 0;
+ } else {
+ printk("keeping early console\n");
+ }
+}
+
+__setup("earlyprintk=", setup_early_printk);
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/kernel/entry_64-xen.S 2009-06-23 09:28:21.000000000 +0200
@@ -0,0 +1,1317 @@
+/*
+ * linux/arch/x86_64/entry.S
+ *
+ * Copyright (C) 1991, 1992 Linus Torvalds
+ * Copyright (C) 2000, 2001, 2002 Andi Kleen SuSE Labs
+ * Copyright (C) 2000 Pavel Machek <pavel@suse.cz>
+ *
+ * $Id$
+ *
+ * Jun Nakajima <jun.nakajima@intel.com>
+ * Asit Mallick <asit.k.mallick@intel.com>
+ * Modified for Xen
+ */
+
+/*
+ * entry.S contains the system-call and fault low-level handling routines.
+ *
+ * NOTE: This code handles signal-recognition, which happens every time
+ * after an interrupt and after each system call.
+ *
+ * Normal syscalls and interrupts don't save a full stack frame, this is
+ * only done for syscall tracing, signals or fork/exec et.al.
+ *
+ * A note on terminology:
+ * - top of stack: Architecture defined interrupt frame from SS to RIP
+ * at the top of the kernel process stack.
+ * - partial stack frame: partially saved registers upto R11.
+ * - full stack frame: Like partial stack frame, but all register saved.
+ *
+ * TODO:
+ * - schedule it carefully for the final hardware.
+ */
+
+#define ASSEMBLY 1
+#include <linux/linkage.h>
+#include <asm/segment.h>
+#include <asm/smp.h>
+#include <asm/cache.h>
+#include <asm/errno.h>
+#include <asm/dwarf2.h>
+#include <asm/calling.h>
+#include <asm/asm-offsets.h>
+#include <asm/msr.h>
+#include <asm/unistd.h>
+#include <asm/thread_info.h>
+#include <asm/hw_irq.h>
+#include <asm/page.h>
+#include <asm/irqflags.h>
+#include <asm/errno.h>
+#include <xen/interface/arch-x86_64.h>
+#include <xen/interface/features.h>
+
+#include "xen_entry.S"
+
+ .code64
+
+#ifndef CONFIG_PREEMPT
+#define retint_kernel retint_restore_args
+#endif
+
+
+.macro TRACE_IRQS_IRETQ offset=ARGOFFSET
+#ifdef CONFIG_TRACE_IRQFLAGS
+ bt $9,EFLAGS-\offset(%rsp) /* interrupts off? */
+ jnc 1f
+ TRACE_IRQS_ON
+1:
+#endif
+.endm
+
+NMI_MASK = 0x80000000
+
+/*
+ * C code is not supposed to know about undefined top of stack. Every time
+ * a C function with an pt_regs argument is called from the SYSCALL based
+ * fast path FIXUP_TOP_OF_STACK is needed.
+ * RESTORE_TOP_OF_STACK syncs the syscall state after any possible ptregs
+ * manipulation.
+ */
+
+ /* %rsp:at FRAMEEND */
+ .macro FIXUP_TOP_OF_STACK tmp
+ movq $__USER_CS,CS(%rsp)
+ movq $-1,RCX(%rsp)
+ .endm
+
+ .macro RESTORE_TOP_OF_STACK tmp,offset=0
+ .endm
+
+ .macro FAKE_STACK_FRAME child_rip
+ /* push in order ss, rsp, eflags, cs, rip */
+ xorl %eax, %eax
+ pushq %rax /* ss */
+ CFI_ADJUST_CFA_OFFSET 8
+ /*CFI_REL_OFFSET ss,0*/
+ pushq %rax /* rsp */
+ CFI_ADJUST_CFA_OFFSET 8
+ CFI_REL_OFFSET rsp,0
+ pushq $(1<<9) /* eflags - interrupts on */
+ CFI_ADJUST_CFA_OFFSET 8
+ /*CFI_REL_OFFSET rflags,0*/
+ pushq $__KERNEL_CS /* cs */
+ CFI_ADJUST_CFA_OFFSET 8
+ /*CFI_REL_OFFSET cs,0*/
+ pushq \child_rip /* rip */
+ CFI_ADJUST_CFA_OFFSET 8
+ CFI_REL_OFFSET rip,0
+ pushq %rax /* orig rax */
+ CFI_ADJUST_CFA_OFFSET 8
+ .endm
+
+ .macro UNFAKE_STACK_FRAME
+ addq $8*6, %rsp
+ CFI_ADJUST_CFA_OFFSET -(6*8)
+ .endm
+
+ .macro CFI_DEFAULT_STACK start=1,adj=0
+ .if \start
+ CFI_STARTPROC simple
+ CFI_DEF_CFA rsp,SS+8 - \adj*ARGOFFSET
+ .else
+ CFI_DEF_CFA_OFFSET SS+8 - \adj*ARGOFFSET
+ .endif
+ .if \adj == 0
+ CFI_REL_OFFSET r15,R15
+ CFI_REL_OFFSET r14,R14
+ CFI_REL_OFFSET r13,R13
+ CFI_REL_OFFSET r12,R12
+ CFI_REL_OFFSET rbp,RBP
+ CFI_REL_OFFSET rbx,RBX
+ .endif
+ CFI_REL_OFFSET r11,R11 - \adj*ARGOFFSET
+ CFI_REL_OFFSET r10,R10 - \adj*ARGOFFSET
+ CFI_REL_OFFSET r9,R9 - \adj*ARGOFFSET
+ CFI_REL_OFFSET r8,R8 - \adj*ARGOFFSET
+ CFI_REL_OFFSET rax,RAX - \adj*ARGOFFSET
+ CFI_REL_OFFSET rcx,RCX - \adj*ARGOFFSET
+ CFI_REL_OFFSET rdx,RDX - \adj*ARGOFFSET
+ CFI_REL_OFFSET rsi,RSI - \adj*ARGOFFSET
+ CFI_REL_OFFSET rdi,RDI - \adj*ARGOFFSET
+ CFI_REL_OFFSET rip,RIP - \adj*ARGOFFSET
+ /*CFI_REL_OFFSET cs,CS - \adj*ARGOFFSET*/
+ /*CFI_REL_OFFSET rflags,EFLAGS - \adj*ARGOFFSET*/
+ CFI_REL_OFFSET rsp,RSP - \adj*ARGOFFSET
+ /*CFI_REL_OFFSET ss,SS - \adj*ARGOFFSET*/
+ .endm
+
+ /*
+ * Must be consistent with the definition in arch-x86/xen-x86_64.h:
+ * struct iret_context {
+ * u64 rax, r11, rcx, flags, rip, cs, rflags, rsp, ss;
+ * };
+ * with rax, r11, and rcx being taken care of in the hypercall stub.
+ */
+ .macro HYPERVISOR_IRET flag
+ testb $3,1*8(%rsp)
+ jnz 2f
+ testl $NMI_MASK,2*8(%rsp)
+ jnz 2f
+
+ cmpb $0,(xen_features+XENFEAT_supervisor_mode_kernel)(%rip)
+ jne 1f
+
+ /* Direct iret to kernel space. Correct CS and SS. */
+ orl $3,1*8(%rsp)
+ orl $3,4*8(%rsp)
+1: iretq
+
+2: /* Slow iret via hypervisor. */
+ andl $~NMI_MASK, 2*8(%rsp)
+ pushq $\flag
+ jmp hypercall_page + (__HYPERVISOR_iret * 32)
+ .endm
+
+/*
+ * A newly forked process directly context switches into this.
+ */
+/* rdi: prev */
+ENTRY(ret_from_fork)
+ CFI_DEFAULT_STACK
+ push kernel_eflags(%rip)
+ CFI_ADJUST_CFA_OFFSET 4
+ popf # reset kernel eflags
+ CFI_ADJUST_CFA_OFFSET -4
+ call schedule_tail
+ GET_THREAD_INFO(%rcx)
+ testl $(_TIF_SYSCALL_TRACE|_TIF_SYSCALL_AUDIT),threadinfo_flags(%rcx)
+ jnz rff_trace
+rff_action:
+ RESTORE_REST
+ testl $3,CS-ARGOFFSET(%rsp) # from kernel_thread?
+ je int_ret_from_sys_call
+ testl $_TIF_IA32,threadinfo_flags(%rcx)
+ jnz int_ret_from_sys_call
+ RESTORE_TOP_OF_STACK %rdi,ARGOFFSET
+ jmp ret_from_sys_call
+rff_trace:
+ movq %rsp,%rdi
+ call syscall_trace_leave
+ GET_THREAD_INFO(%rcx)
+ jmp rff_action
+ CFI_ENDPROC
+END(ret_from_fork)
+
+/*
+ * initial frame state for interrupts and exceptions
+ */
+ .macro _frame ref
+ CFI_STARTPROC simple
+ CFI_DEF_CFA rsp,SS+8-\ref
+ /*CFI_REL_OFFSET ss,SS-\ref*/
+ CFI_REL_OFFSET rsp,RSP-\ref
+ /*CFI_REL_OFFSET rflags,EFLAGS-\ref*/
+ /*CFI_REL_OFFSET cs,CS-\ref*/
+ CFI_REL_OFFSET rip,RIP-\ref
+ .endm
+
+/*
+ * System call entry. Upto 6 arguments in registers are supported.
+ *
+ * SYSCALL does not save anything on the stack and does not change the
+ * stack pointer.
+ */
+
+/*
+ * Register setup:
+ * rax system call number
+ * rdi arg0
+ * rcx return address for syscall/sysret, C arg3
+ * rsi arg1
+ * rdx arg2
+ * r10 arg3 (--> moved to rcx for C)
+ * r8 arg4
+ * r9 arg5
+ * r11 eflags for syscall/sysret, temporary for C
+ * r12-r15,rbp,rbx saved by C code, not touched.
+ *
+ * Interrupts are enabled on entry.
+ * Only called from user space.
+ *
+ * XXX if we had a free scratch register we could save the RSP into the stack frame
+ * and report it properly in ps. Unfortunately we haven't.
+ *
+ * When user can change the frames always force IRET. That is because
+ * it deals with uncanonical addresses better. SYSRET has trouble
+ * with them due to bugs in both AMD and Intel CPUs.
+ */
+
+ENTRY(system_call)
+ _frame (RIP-0x10)
+ SAVE_ARGS -8,0
+ movq %rax,ORIG_RAX-ARGOFFSET(%rsp)
+ GET_THREAD_INFO(%rcx)
+ testl $(_TIF_SYSCALL_TRACE|_TIF_SYSCALL_AUDIT|_TIF_SECCOMP),threadinfo_flags(%rcx)
+ CFI_REMEMBER_STATE
+ jnz tracesys
+ cmpq $__NR_syscall_max,%rax
+ ja badsys
+ movq %r10,%rcx
+ call *sys_call_table(,%rax,8) # XXX: rip relative
+ movq %rax,RAX-ARGOFFSET(%rsp)
+/*
+ * Syscall return path ending with SYSRET (fast path)
+ * Has incomplete stack frame and undefined top of stack.
+ */
+ .globl ret_from_sys_call
+ret_from_sys_call:
+ movl $_TIF_ALLWORK_MASK,%edi
+ /* edi: flagmask */
+sysret_check:
+ GET_THREAD_INFO(%rcx)
+ XEN_BLOCK_EVENTS(%rsi)
+ TRACE_IRQS_OFF
+ movl threadinfo_flags(%rcx),%edx
+ andl %edi,%edx
+ CFI_REMEMBER_STATE
+ jnz sysret_careful
+ /*
+ * sysretq will re-enable interrupts:
+ */
+ TRACE_IRQS_ON
+ XEN_UNBLOCK_EVENTS(%rsi)
+ RESTORE_ARGS 0,8,0
+ HYPERVISOR_IRET VGCF_IN_SYSCALL
+
+ /* Handle reschedules */
+ /* edx: work, edi: workmask */
+sysret_careful:
+ CFI_RESTORE_STATE
+ bt $TIF_NEED_RESCHED,%edx
+ jnc sysret_signal
+ TRACE_IRQS_ON
+ XEN_UNBLOCK_EVENTS(%rsi)
+ pushq %rdi
+ CFI_ADJUST_CFA_OFFSET 8
+ call schedule
+ popq %rdi
+ CFI_ADJUST_CFA_OFFSET -8
+ jmp sysret_check
+
+ /* Handle a signal */
+sysret_signal:
+ TRACE_IRQS_ON
+/* sti */
+ XEN_UNBLOCK_EVENTS(%rsi)
+ testl $(_TIF_SIGPENDING|_TIF_NOTIFY_RESUME|_TIF_SINGLESTEP),%edx
+ jz 1f
+
+ /* Really a signal */
+ /* edx: work flags (arg3) */
+ leaq do_notify_resume(%rip),%rax
+ leaq -ARGOFFSET(%rsp),%rdi # &pt_regs -> arg1
+ xorl %esi,%esi # oldset -> arg2
+ call ptregscall_common
+1: movl $_TIF_NEED_RESCHED,%edi
+ /* Use IRET because user could have changed frame. This
+ works because ptregscall_common has called FIXUP_TOP_OF_STACK. */
+ XEN_BLOCK_EVENTS(%rsi)
+ TRACE_IRQS_OFF
+ jmp int_with_check
+
+badsys:
+ movq $-ENOSYS,RAX-ARGOFFSET(%rsp)
+ jmp ret_from_sys_call
+
+ /* Do syscall tracing */
+tracesys:
+ CFI_RESTORE_STATE
+ SAVE_REST
+ movq $-ENOSYS,RAX(%rsp)
+ FIXUP_TOP_OF_STACK %rdi
+ movq %rsp,%rdi
+ call syscall_trace_enter
+ LOAD_ARGS ARGOFFSET /* reload args from stack in case ptrace changed it */
+ RESTORE_REST
+ cmpq $__NR_syscall_max,%rax
+ ja 1f
+ movq %r10,%rcx /* fixup for C */
+ call *sys_call_table(,%rax,8)
+1: movq %rax,RAX-ARGOFFSET(%rsp)
+ /* Use IRET because user could have changed frame */
+ jmp int_ret_from_sys_call
+ CFI_ENDPROC
+END(system_call)
+
+/*
+ * Syscall return path ending with IRET.
+ * Has correct top of stack, but partial stack frame.
+ */
+ENTRY(int_ret_from_sys_call)
+ CFI_STARTPROC simple
+ CFI_DEF_CFA rsp,SS+8-ARGOFFSET
+ /*CFI_REL_OFFSET ss,SS-ARGOFFSET*/
+ CFI_REL_OFFSET rsp,RSP-ARGOFFSET
+ /*CFI_REL_OFFSET rflags,EFLAGS-ARGOFFSET*/
+ /*CFI_REL_OFFSET cs,CS-ARGOFFSET*/
+ CFI_REL_OFFSET rip,RIP-ARGOFFSET
+ CFI_REL_OFFSET rdx,RDX-ARGOFFSET
+ CFI_REL_OFFSET rcx,RCX-ARGOFFSET
+ CFI_REL_OFFSET rax,RAX-ARGOFFSET
+ CFI_REL_OFFSET rdi,RDI-ARGOFFSET
+ CFI_REL_OFFSET rsi,RSI-ARGOFFSET
+ CFI_REL_OFFSET r8,R8-ARGOFFSET
+ CFI_REL_OFFSET r9,R9-ARGOFFSET
+ CFI_REL_OFFSET r10,R10-ARGOFFSET
+ CFI_REL_OFFSET r11,R11-ARGOFFSET
+ XEN_BLOCK_EVENTS(%rsi)
+ TRACE_IRQS_OFF
+ testb $3,CS-ARGOFFSET(%rsp)
+ jnz 1f
+ /* Need to set the proper %ss (not NULL) for ring 3 iretq */
+ movl $__KERNEL_DS,SS-ARGOFFSET(%rsp)
+ jmp retint_restore_args # retrun from ring3 kernel
+1:
+ movl $_TIF_ALLWORK_MASK,%edi
+ /* edi: mask to check */
+int_with_check:
+ GET_THREAD_INFO(%rcx)
+ movl threadinfo_flags(%rcx),%edx
+ andl %edi,%edx
+ jnz int_careful
+ andl $~TS_COMPAT,threadinfo_status(%rcx)
+ jmp retint_restore_args
+
+ /* Either reschedule or signal or syscall exit tracking needed. */
+ /* First do a reschedule test. */
+ /* edx: work, edi: workmask */
+int_careful:
+ bt $TIF_NEED_RESCHED,%edx
+ jnc int_very_careful
+ TRACE_IRQS_ON
+/* sti */
+ XEN_UNBLOCK_EVENTS(%rsi)
+ pushq %rdi
+ CFI_ADJUST_CFA_OFFSET 8
+ call schedule
+ popq %rdi
+ CFI_ADJUST_CFA_OFFSET -8
+ XEN_BLOCK_EVENTS(%rsi)
+ TRACE_IRQS_OFF
+ jmp int_with_check
+
+ /* handle signals and tracing -- both require a full stack frame */
+int_very_careful:
+ TRACE_IRQS_ON
+/* sti */
+ XEN_UNBLOCK_EVENTS(%rsi)
+ SAVE_REST
+ /* Check for syscall exit trace */
+ testl $(_TIF_SYSCALL_TRACE|_TIF_SYSCALL_AUDIT|_TIF_SINGLESTEP),%edx
+ jz int_signal
+ pushq %rdi
+ CFI_ADJUST_CFA_OFFSET 8
+ leaq 8(%rsp),%rdi # &ptregs -> arg1
+ call syscall_trace_leave
+ popq %rdi
+ CFI_ADJUST_CFA_OFFSET -8
+ andl $~(_TIF_SYSCALL_TRACE|_TIF_SYSCALL_AUDIT|_TIF_SINGLESTEP),%edi
+ XEN_BLOCK_EVENTS(%rsi)
+ TRACE_IRQS_OFF
+ jmp int_restore_rest
+
+int_signal:
+ testl $(_TIF_NOTIFY_RESUME|_TIF_SIGPENDING|_TIF_SINGLESTEP),%edx
+ jz 1f
+ movq %rsp,%rdi # &ptregs -> arg1
+ xorl %esi,%esi # oldset -> arg2
+ call do_notify_resume
+1: movl $_TIF_NEED_RESCHED,%edi
+int_restore_rest:
+ RESTORE_REST
+ XEN_BLOCK_EVENTS(%rsi)
+ TRACE_IRQS_OFF
+ jmp int_with_check
+ CFI_ENDPROC
+END(int_ret_from_sys_call)
+
+/*
+ * Certain special system calls that need to save a complete full stack frame.
+ */
+
+ .macro PTREGSCALL label,func,arg
+ .globl \label
+\label:
+ leaq \func(%rip),%rax
+ leaq -ARGOFFSET+8(%rsp),\arg /* 8 for return address */
+ jmp ptregscall_common
+END(\label)
+ .endm
+
+ CFI_STARTPROC
+
+ PTREGSCALL stub_clone, sys_clone, %r8
+ PTREGSCALL stub_fork, sys_fork, %rdi
+ PTREGSCALL stub_vfork, sys_vfork, %rdi
+ PTREGSCALL stub_rt_sigsuspend, sys_rt_sigsuspend, %rdx
+ PTREGSCALL stub_sigaltstack, sys_sigaltstack, %rdx
+ PTREGSCALL stub_iopl, sys_iopl, %rsi
+
+ENTRY(ptregscall_common)
+ popq %r11
+ CFI_ADJUST_CFA_OFFSET -8
+ CFI_REGISTER rip, r11
+ SAVE_REST
+ movq %r11, %r15
+ CFI_REGISTER rip, r15
+ FIXUP_TOP_OF_STACK %r11
+ call *%rax
+ RESTORE_TOP_OF_STACK %r11
+ movq %r15, %r11
+ CFI_REGISTER rip, r11
+ RESTORE_REST
+ pushq %r11
+ CFI_ADJUST_CFA_OFFSET 8
+ CFI_REL_OFFSET rip, 0
+ ret
+ CFI_ENDPROC
+END(ptregscall_common)
+
+ENTRY(stub_execve)
+ CFI_STARTPROC
+ popq %r11
+ CFI_ADJUST_CFA_OFFSET -8
+ CFI_REGISTER rip, r11
+ SAVE_REST
+ FIXUP_TOP_OF_STACK %r11
+ call sys_execve
+ RESTORE_TOP_OF_STACK %r11
+ movq %rax,RAX(%rsp)
+ RESTORE_REST
+ jmp int_ret_from_sys_call
+ CFI_ENDPROC
+END(stub_execve)
+
+/*
+ * sigreturn is special because it needs to restore all registers on return.
+ * This cannot be done with SYSRET, so use the IRET return path instead.
+ */
+ENTRY(stub_rt_sigreturn)
+ CFI_STARTPROC
+ addq $8, %rsp
+ CFI_ADJUST_CFA_OFFSET -8
+ SAVE_REST
+ movq %rsp,%rdi
+ FIXUP_TOP_OF_STACK %r11
+ call sys_rt_sigreturn
+ movq %rax,RAX(%rsp) # fixme, this could be done at the higher layer
+ RESTORE_REST
+ jmp int_ret_from_sys_call
+ CFI_ENDPROC
+END(stub_rt_sigreturn)
+
+/* initial frame state for interrupts (and exceptions without error code) */
+#define INTR_FRAME _frame (RIP-0x10); \
+ CFI_REL_OFFSET rcx,0; \
+ CFI_REL_OFFSET r11,8
+
+/* initial frame state for exceptions with error code (and interrupts with
+ vector already pushed) */
+#define XCPT_FRAME _frame (RIP-0x18); \
+ CFI_REL_OFFSET rcx,0; \
+ CFI_REL_OFFSET r11,8
+
+/*
+ * Interrupt exit.
+ *
+ */
+
+retint_check:
+ CFI_DEFAULT_STACK adj=1
+ movl threadinfo_flags(%rcx),%edx
+ andl %edi,%edx
+ CFI_REMEMBER_STATE
+ jnz retint_careful
+retint_restore_args:
+ movl EFLAGS-REST_SKIP(%rsp), %eax
+ shr $9, %eax # EAX[0] == IRET_EFLAGS.IF
+ XEN_GET_VCPU_INFO(%rsi)
+ andb evtchn_upcall_mask(%rsi),%al
+ andb $1,%al # EAX[0] == IRET_EFLAGS.IF & event_mask
+ jnz restore_all_enable_events # != 0 => enable event delivery
+ XEN_PUT_VCPU_INFO(%rsi)
+
+ RESTORE_ARGS 0,8,0
+ HYPERVISOR_IRET 0
+
+ /* edi: workmask, edx: work */
+retint_careful:
+ CFI_RESTORE_STATE
+ bt $TIF_NEED_RESCHED,%edx
+ jnc retint_signal
+ TRACE_IRQS_ON
+ XEN_UNBLOCK_EVENTS(%rsi)
+/* sti */
+ pushq %rdi
+ CFI_ADJUST_CFA_OFFSET 8
+ call schedule
+ popq %rdi
+ CFI_ADJUST_CFA_OFFSET -8
+ GET_THREAD_INFO(%rcx)
+ XEN_BLOCK_EVENTS(%rsi)
+/* cli */
+ TRACE_IRQS_OFF
+ jmp retint_check
+
+retint_signal:
+ testl $(_TIF_SIGPENDING|_TIF_NOTIFY_RESUME|_TIF_SINGLESTEP),%edx
+ jz retint_restore_args
+ TRACE_IRQS_ON
+ XEN_UNBLOCK_EVENTS(%rsi)
+ SAVE_REST
+ movq $-1,ORIG_RAX(%rsp)
+ xorl %esi,%esi # oldset
+ movq %rsp,%rdi # &pt_regs
+ call do_notify_resume
+ RESTORE_REST
+ XEN_BLOCK_EVENTS(%rsi)
+ TRACE_IRQS_OFF
+ movl $_TIF_NEED_RESCHED,%edi
+ GET_THREAD_INFO(%rcx)
+ jmp retint_check
+
+#ifdef CONFIG_PREEMPT
+ /* Returning to kernel space. Check if we need preemption */
+ /* rcx: threadinfo. interrupts off. */
+ .p2align
+retint_kernel:
+ cmpl $0,threadinfo_preempt_count(%rcx)
+ jnz retint_restore_args
+ bt $TIF_NEED_RESCHED,threadinfo_flags(%rcx)
+ jnc retint_restore_args
+ bt $9,EFLAGS-ARGOFFSET(%rsp) /* interrupts off? */
+ jnc retint_restore_args
+ call preempt_schedule_irq
+ jmp retint_kernel /* check again */
+#endif
+
+ CFI_ENDPROC
+END(retint_check)
+
+#ifndef CONFIG_XEN
+/*
+ * APIC interrupts.
+ */
+ .macro apicinterrupt num,func
+ INTR_FRAME
+ pushq $~(\num)
+ CFI_ADJUST_CFA_OFFSET 8
+ interrupt \func
+ jmp error_entry
+ CFI_ENDPROC
+ .endm
+
+ENTRY(thermal_interrupt)
+ apicinterrupt THERMAL_APIC_VECTOR,smp_thermal_interrupt
+END(thermal_interrupt)
+
+ENTRY(threshold_interrupt)
+ apicinterrupt THRESHOLD_APIC_VECTOR,mce_threshold_interrupt
+END(threshold_interrupt)
+
+#ifdef CONFIG_SMP
+ENTRY(reschedule_interrupt)
+ apicinterrupt RESCHEDULE_VECTOR,smp_reschedule_interrupt
+END(reschedule_interrupt)
+
+ .macro INVALIDATE_ENTRY num
+ENTRY(invalidate_interrupt\num)
+ apicinterrupt INVALIDATE_TLB_VECTOR_START+\num,smp_invalidate_interrupt
+END(invalidate_interrupt\num)
+ .endm
+
+ INVALIDATE_ENTRY 0
+ INVALIDATE_ENTRY 1
+ INVALIDATE_ENTRY 2
+ INVALIDATE_ENTRY 3
+ INVALIDATE_ENTRY 4
+ INVALIDATE_ENTRY 5
+ INVALIDATE_ENTRY 6
+ INVALIDATE_ENTRY 7
+
+ENTRY(call_function_interrupt)
+ apicinterrupt CALL_FUNCTION_VECTOR,smp_call_function_interrupt
+END(call_function_interrupt)
+#endif
+
+#ifdef CONFIG_X86_LOCAL_APIC
+ENTRY(apic_timer_interrupt)
+ apicinterrupt LOCAL_TIMER_VECTOR,smp_apic_timer_interrupt
+END(apic_timer_interrupt)
+
+ENTRY(error_interrupt)
+ apicinterrupt ERROR_APIC_VECTOR,smp_error_interrupt
+END(error_interrupt)
+
+ENTRY(spurious_interrupt)
+ apicinterrupt SPURIOUS_APIC_VECTOR,smp_spurious_interrupt
+END(spurious_interrupt)
+#endif
+#endif /* !CONFIG_XEN */
+
+/*
+ * Exception entry points.
+ */
+ .macro zeroentry sym
+ INTR_FRAME
+ movq (%rsp),%rcx
+ CFI_RESTORE rcx
+ movq 8(%rsp),%r11
+ CFI_RESTORE r11
+ addq $0x10,%rsp /* skip rcx and r11 */
+ CFI_ADJUST_CFA_OFFSET -0x10
+ pushq $0 /* push error code/oldrax */
+ CFI_ADJUST_CFA_OFFSET 8
+ pushq %rax /* push real oldrax to the rdi slot */
+ CFI_ADJUST_CFA_OFFSET 8
+ CFI_REL_OFFSET rax,0
+ leaq \sym(%rip),%rax
+ jmp error_entry
+ CFI_ENDPROC
+ .endm
+
+ .macro errorentry sym
+ XCPT_FRAME
+ movq (%rsp),%rcx
+ CFI_RESTORE rcx
+ movq 8(%rsp),%r11
+ CFI_RESTORE r11
+ addq $0x10,%rsp /* rsp points to the error code */
+ CFI_ADJUST_CFA_OFFSET -0x10
+ pushq %rax
+ CFI_ADJUST_CFA_OFFSET 8
+ CFI_REL_OFFSET rax,0
+ leaq \sym(%rip),%rax
+ jmp error_entry
+ CFI_ENDPROC
+ .endm
+
+#if 0 /* not XEN */
+ /* error code is on the stack already */
+ /* handle NMI like exceptions that can happen everywhere */
+ .macro paranoidentry sym, ist=0, irqtrace=1
+ movq (%rsp),%rcx
+ movq 8(%rsp),%r11
+ addq $0x10,%rsp /* skip rcx and r11 */
+ SAVE_ALL
+ cld
+#if 0 /* not XEN */
+ movl $1,%ebx
+ movl $MSR_GS_BASE,%ecx
+ rdmsr
+ testl %edx,%edx
+ js 1f
+ swapgs
+ xorl %ebx,%ebx
+1:
+#endif
+ .if \ist
+ movq %gs:pda_data_offset, %rbp
+ .endif
+ movq %rsp,%rdi
+ movq ORIG_RAX(%rsp),%rsi
+ movq $-1,ORIG_RAX(%rsp)
+ .if \ist
+ subq $EXCEPTION_STKSZ, per_cpu__init_tss + TSS_ist + (\ist - 1) * 8(%rbp)
+ .endif
+ call \sym
+ .if \ist
+ addq $EXCEPTION_STKSZ, per_cpu__init_tss + TSS_ist + (\ist - 1) * 8(%rbp)
+ .endif
+/* cli */
+ XEN_BLOCK_EVENTS(%rsi)
+ .if \irqtrace
+ TRACE_IRQS_OFF
+ .endif
+ .endm
+
+ /*
+ * "Paranoid" exit path from exception stack.
+ * Paranoid because this is used by NMIs and cannot take
+ * any kernel state for granted.
+ * We don't do kernel preemption checks here, because only
+ * NMI should be common and it does not enable IRQs and
+ * cannot get reschedule ticks.
+ *
+ * "trace" is 0 for the NMI handler only, because irq-tracing
+ * is fundamentally NMI-unsafe. (we cannot change the soft and
+ * hard flags at once, atomically)
+ */
+ .macro paranoidexit trace=1
+ /* ebx: no swapgs flag */
+paranoid_exit\trace:
+ testl %ebx,%ebx /* swapgs needed? */
+ jnz paranoid_restore\trace
+ testl $3,CS(%rsp)
+ jnz paranoid_userspace\trace
+paranoid_swapgs\trace:
+ TRACE_IRQS_IRETQ 0
+ swapgs
+paranoid_restore\trace:
+ RESTORE_ALL 8
+ iretq
+paranoid_userspace\trace:
+ GET_THREAD_INFO(%rcx)
+ movl threadinfo_flags(%rcx),%ebx
+ andl $_TIF_WORK_MASK,%ebx
+ jz paranoid_swapgs\trace
+ movq %rsp,%rdi /* &pt_regs */
+ call sync_regs
+ movq %rax,%rsp /* switch stack for scheduling */
+ testl $_TIF_NEED_RESCHED,%ebx
+ jnz paranoid_schedule\trace
+ movl %ebx,%edx /* arg3: thread flags */
+ .if \trace
+ TRACE_IRQS_ON
+ .endif
+ sti
+ xorl %esi,%esi /* arg2: oldset */
+ movq %rsp,%rdi /* arg1: &pt_regs */
+ call do_notify_resume
+ cli
+ .if \trace
+ TRACE_IRQS_OFF
+ .endif
+ jmp paranoid_userspace\trace
+paranoid_schedule\trace:
+ .if \trace
+ TRACE_IRQS_ON
+ .endif
+ sti
+ call schedule
+ cli
+ .if \trace
+ TRACE_IRQS_OFF
+ .endif
+ jmp paranoid_userspace\trace
+ CFI_ENDPROC
+ .endm
+#endif
+
+/*
+ * Exception entry point. This expects an error code/orig_rax on the stack
+ * and the exception handler in %rax.
+ */
+ENTRY(error_entry)
+ _frame RDI
+ CFI_REL_OFFSET rax,0
+ /* rdi slot contains rax, oldrax contains error code */
+ cld
+ subq $14*8,%rsp
+ CFI_ADJUST_CFA_OFFSET (14*8)
+ movq %rsi,13*8(%rsp)
+ CFI_REL_OFFSET rsi,RSI
+ movq 14*8(%rsp),%rsi /* load rax from rdi slot */
+ CFI_REGISTER rax,rsi
+ movq %rdx,12*8(%rsp)
+ CFI_REL_OFFSET rdx,RDX
+ movq %rcx,11*8(%rsp)
+ CFI_REL_OFFSET rcx,RCX
+ movq %rsi,10*8(%rsp) /* store rax */
+ CFI_REL_OFFSET rax,RAX
+ movq %r8, 9*8(%rsp)
+ CFI_REL_OFFSET r8,R8
+ movq %r9, 8*8(%rsp)
+ CFI_REL_OFFSET r9,R9
+ movq %r10,7*8(%rsp)
+ CFI_REL_OFFSET r10,R10
+ movq %r11,6*8(%rsp)
+ CFI_REL_OFFSET r11,R11
+ movq %rbx,5*8(%rsp)
+ CFI_REL_OFFSET rbx,RBX
+ movq %rbp,4*8(%rsp)
+ CFI_REL_OFFSET rbp,RBP
+ movq %r12,3*8(%rsp)
+ CFI_REL_OFFSET r12,R12
+ movq %r13,2*8(%rsp)
+ CFI_REL_OFFSET r13,R13
+ movq %r14,1*8(%rsp)
+ CFI_REL_OFFSET r14,R14
+ movq %r15,(%rsp)
+ CFI_REL_OFFSET r15,R15
+#if 0
+ cmpl $__KERNEL_CS,CS(%rsp)
+ CFI_REMEMBER_STATE
+ je error_kernelspace
+#endif
+error_call_handler:
+ movq %rdi, RDI(%rsp)
+ CFI_REL_OFFSET rdi,RDI
+ movq %rsp,%rdi
+ movq ORIG_RAX(%rsp),%rsi # get error code
+ movq $-1,ORIG_RAX(%rsp)
+ call *%rax
+error_exit:
+ RESTORE_REST
+/* cli */
+ XEN_BLOCK_EVENTS(%rsi)
+ TRACE_IRQS_OFF
+ GET_THREAD_INFO(%rcx)
+ testb $3,CS-ARGOFFSET(%rsp)
+ jz retint_kernel
+ movl threadinfo_flags(%rcx),%edx
+ movl $_TIF_WORK_MASK,%edi
+ andl %edi,%edx
+ jnz retint_careful
+ /*
+ * The iret might restore flags:
+ */
+ TRACE_IRQS_IRETQ
+ jmp retint_restore_args
+
+#if 0
+ /*
+ * We need to re-write the logic here because we don't do iretq to
+ * to return to user mode. It's still possible that we get trap/fault
+ * in the kernel (when accessing buffers pointed to by system calls,
+ * for example).
+ *
+ */
+ CFI_RESTORE_STATE
+error_kernelspace:
+ incl %ebx
+ /* There are two places in the kernel that can potentially fault with
+ usergs. Handle them here. The exception handlers after
+ iret run with kernel gs again, so don't set the user space flag.
+ B stepping K8s sometimes report an truncated RIP for IRET
+ exceptions returning to compat mode. Check for these here too. */
+ leaq iret_label(%rip),%rbp
+ cmpq %rbp,RIP(%rsp)
+ je error_swapgs
+ movl %ebp,%ebp /* zero extend */
+ cmpq %rbp,RIP(%rsp)
+ je error_swapgs
+ cmpq $gs_change,RIP(%rsp)
+ je error_swapgs
+ jmp error_sti
+#endif
+ CFI_ENDPROC
+END(error_entry)
+
+ENTRY(hypervisor_callback)
+ zeroentry do_hypervisor_callback
+END(hypervisor_callback)
+
+/*
+ * Copied from arch/xen/i386/kernel/entry.S
+ */
+# A note on the "critical region" in our callback handler.
+# We want to avoid stacking callback handlers due to events occurring
+# during handling of the last event. To do this, we keep events disabled
+# until we've done all processing. HOWEVER, we must enable events before
+# popping the stack frame (can't be done atomically) and so it would still
+# be possible to get enough handler activations to overflow the stack.
+# Although unlikely, bugs of that kind are hard to track down, so we'd
+# like to avoid the possibility.
+# So, on entry to the handler we detect whether we interrupted an
+# existing activation in its critical region -- if so, we pop the current
+# activation and restart the handler using the previous one.
+ENTRY(do_hypervisor_callback) # do_hypervisor_callback(struct *pt_regs)
+ CFI_STARTPROC
+# Since we don't modify %rdi, evtchn_do_upall(struct *pt_regs) will
+# see the correct pointer to the pt_regs
+ movq %rdi, %rsp # we don't return, adjust the stack frame
+ CFI_ENDPROC
+ CFI_DEFAULT_STACK
+11: incl %gs:pda_irqcount
+ movq %rsp,%rbp
+ CFI_DEF_CFA_REGISTER rbp
+ cmovzq %gs:pda_irqstackptr,%rsp
+ pushq %rbp # backlink for old unwinder
+ call evtchn_do_upcall
+ popq %rsp
+ CFI_DEF_CFA_REGISTER rsp
+ decl %gs:pda_irqcount
+ jmp error_exit
+ CFI_ENDPROC
+END(do_hypervisor_callback)
+
+#ifdef CONFIG_X86_LOCAL_APIC
+KPROBE_ENTRY(nmi)
+ zeroentry do_nmi_callback
+ENTRY(do_nmi_callback)
+ CFI_STARTPROC
+ addq $8, %rsp
+ CFI_ENDPROC
+ CFI_DEFAULT_STACK
+ call do_nmi
+ orl $NMI_MASK,EFLAGS(%rsp)
+ RESTORE_REST
+ XEN_BLOCK_EVENTS(%rsi)
+ TRACE_IRQS_OFF
+ GET_THREAD_INFO(%rcx)
+ jmp retint_restore_args
+ CFI_ENDPROC
+ .previous .text
+END(nmi)
+#endif
+
+ ALIGN
+restore_all_enable_events:
+ CFI_DEFAULT_STACK adj=1
+ TRACE_IRQS_ON
+ XEN_UNBLOCK_EVENTS(%rsi) # %rsi is already set up...
+
+scrit: /**** START OF CRITICAL REGION ****/
+ XEN_TEST_PENDING(%rsi)
+ CFI_REMEMBER_STATE
+ jnz 14f # process more events if necessary...
+ XEN_PUT_VCPU_INFO(%rsi)
+ RESTORE_ARGS 0,8,0
+ HYPERVISOR_IRET 0
+
+ CFI_RESTORE_STATE
+14: XEN_LOCKED_BLOCK_EVENTS(%rsi)
+ XEN_PUT_VCPU_INFO(%rsi)
+ SAVE_REST
+ movq %rsp,%rdi # set the argument again
+ jmp 11b
+ CFI_ENDPROC
+ecrit: /**** END OF CRITICAL REGION ****/
+# At this point, unlike on x86-32, we don't do the fixup to simplify the
+# code and the stack frame is more complex on x86-64.
+# When the kernel is interrupted in the critical section, the kernel
+# will do IRET in that case, and everything will be restored at that point,
+# i.e. it just resumes from the next instruction interrupted with the same context.
+
+# Hypervisor uses this for application faults while it executes.
+# We get here for two reasons:
+# 1. Fault while reloading DS, ES, FS or GS
+# 2. Fault while executing IRET
+# Category 1 we do not need to fix up as Xen has already reloaded all segment
+# registers that could be reloaded and zeroed the others.
+# Category 2 we fix up by killing the current process. We cannot use the
+# normal Linux return path in this case because if we use the IRET hypercall
+# to pop the stack frame we end up in an infinite loop of failsafe callbacks.
+# We distinguish between categories by comparing each saved segment register
+# with its current contents: any discrepancy means we in category 1.
+ENTRY(failsafe_callback)
+ _frame (RIP-0x30)
+ CFI_REL_OFFSET rcx, 0
+ CFI_REL_OFFSET r11, 8
+ movw %ds,%cx
+ cmpw %cx,0x10(%rsp)
+ CFI_REMEMBER_STATE
+ jne 1f
+ movw %es,%cx
+ cmpw %cx,0x18(%rsp)
+ jne 1f
+ movw %fs,%cx
+ cmpw %cx,0x20(%rsp)
+ jne 1f
+ movw %gs,%cx
+ cmpw %cx,0x28(%rsp)
+ jne 1f
+ /* All segments match their saved values => Category 2 (Bad IRET). */
+ movq (%rsp),%rcx
+ CFI_RESTORE rcx
+ movq 8(%rsp),%r11
+ CFI_RESTORE r11
+ addq $0x30,%rsp
+ CFI_ADJUST_CFA_OFFSET -0x30
+ movq $11,%rdi /* SIGSEGV */
+ jmp do_exit
+ CFI_RESTORE_STATE
+1: /* Segment mismatch => Category 1 (Bad segment). Retry the IRET. */
+ movq (%rsp),%rcx
+ CFI_RESTORE rcx
+ movq 8(%rsp),%r11
+ CFI_RESTORE r11
+ addq $0x30,%rsp
+ CFI_ADJUST_CFA_OFFSET -0x30
+ pushq $0
+ CFI_ADJUST_CFA_OFFSET 8
+ SAVE_ALL
+ jmp error_exit
+ CFI_ENDPROC
+#if 0
+ .section __ex_table,"a"
+ .align 8
+ .quad gs_change,bad_gs
+ .previous
+ .section .fixup,"ax"
+ /* running with kernelgs */
+bad_gs:
+/* swapgs */ /* switch back to user gs */
+ xorl %eax,%eax
+ movl %eax,%gs
+ jmp 2b
+ .previous
+#endif
+
+/*
+ * Create a kernel thread.
+ *
+ * C extern interface:
+ * extern long kernel_thread(int (*fn)(void *), void * arg, unsigned long flags)
+ *
+ * asm input arguments:
+ * rdi: fn, rsi: arg, rdx: flags
+ */
+ENTRY(kernel_thread)
+ CFI_STARTPROC
+ FAKE_STACK_FRAME $child_rip
+ SAVE_ALL
+
+ # rdi: flags, rsi: usp, rdx: will be &pt_regs
+ movq %rdx,%rdi
+ orq kernel_thread_flags(%rip),%rdi
+ movq $-1, %rsi
+ movq %rsp, %rdx
+
+ xorl %r8d,%r8d
+ xorl %r9d,%r9d
+
+ # clone now
+ call do_fork
+ movq %rax,RAX(%rsp)
+ xorl %edi,%edi
+
+ /*
+ * It isn't worth to check for reschedule here,
+ * so internally to the x86_64 port you can rely on kernel_thread()
+ * not to reschedule the child before returning, this avoids the need
+ * of hacks for example to fork off the per-CPU idle tasks.
+ * [Hopefully no generic code relies on the reschedule -AK]
+ */
+ RESTORE_ALL
+ UNFAKE_STACK_FRAME
+ ret
+ CFI_ENDPROC
+ENDPROC(kernel_thread)
+
+child_rip:
+ pushq $0 # fake return address
+ CFI_STARTPROC
+ /*
+ * Here we are in the child and the registers are set as they were
+ * at kernel_thread() invocation in the parent.
+ */
+ movq %rdi, %rax
+ movq %rsi, %rdi
+ call *%rax
+ # exit
+ xorl %edi, %edi
+ call do_exit
+ CFI_ENDPROC
+ENDPROC(child_rip)
+
+/*
+ * execve(). This function needs to use IRET, not SYSRET, to set up all state properly.
+ *
+ * C extern interface:
+ * extern long execve(char *name, char **argv, char **envp)
+ *
+ * asm input arguments:
+ * rdi: name, rsi: argv, rdx: envp
+ *
+ * We want to fallback into:
+ * extern long sys_execve(char *name, char **argv,char **envp, struct pt_regs regs)
+ *
+ * do_sys_execve asm fallback arguments:
+ * rdi: name, rsi: argv, rdx: envp, fake frame on the stack
+ */
+ENTRY(execve)
+ CFI_STARTPROC
+ FAKE_STACK_FRAME $0
+ SAVE_ALL
+ call sys_execve
+ movq %rax, RAX(%rsp)
+ RESTORE_REST
+ testq %rax,%rax
+ jne 1f
+ jmp int_ret_from_sys_call
+1: RESTORE_ARGS
+ UNFAKE_STACK_FRAME
+ ret
+ CFI_ENDPROC
+ENDPROC(execve)
+
+KPROBE_ENTRY(page_fault)
+ errorentry do_page_fault
+END(page_fault)
+ .previous .text
+
+ENTRY(coprocessor_error)
+ zeroentry do_coprocessor_error
+END(coprocessor_error)
+
+ENTRY(simd_coprocessor_error)
+ zeroentry do_simd_coprocessor_error
+END(simd_coprocessor_error)
+
+ENTRY(device_not_available)
+ zeroentry math_state_restore
+END(device_not_available)
+
+ /* runs on exception stack */
+KPROBE_ENTRY(debug)
+/* INTR_FRAME
+ pushq $0
+ CFI_ADJUST_CFA_OFFSET 8 */
+ zeroentry do_debug
+/* paranoidexit
+ CFI_ENDPROC */
+END(debug)
+ .previous .text
+
+#if 0
+ /* runs on exception stack */
+KPROBE_ENTRY(nmi)
+ INTR_FRAME
+ pushq $-1
+ CFI_ADJUST_CFA_OFFSET 8
+ paranoidentry do_nmi, 0, 0
+#ifdef CONFIG_TRACE_IRQFLAGS
+ paranoidexit 0
+#else
+ jmp paranoid_exit1
+ CFI_ENDPROC
+#endif
+END(nmi)
+ .previous .text
+#endif
+
+KPROBE_ENTRY(int3)
+/* INTR_FRAME
+ pushq $0
+ CFI_ADJUST_CFA_OFFSET 8 */
+ zeroentry do_int3
+/* jmp paranoid_exit1
+ CFI_ENDPROC */
+END(int3)
+ .previous .text
+
+ENTRY(overflow)
+ zeroentry do_overflow
+END(overflow)
+
+ENTRY(bounds)
+ zeroentry do_bounds
+END(bounds)
+
+ENTRY(invalid_op)
+ zeroentry do_invalid_op
+END(invalid_op)
+
+ENTRY(coprocessor_segment_overrun)
+ zeroentry do_coprocessor_segment_overrun
+END(coprocessor_segment_overrun)
+
+ENTRY(reserved)
+ zeroentry do_reserved
+END(reserved)
+
+#if 0
+ /* runs on exception stack */
+ENTRY(double_fault)
+ XCPT_FRAME
+ paranoidentry do_double_fault
+ jmp paranoid_exit1
+ CFI_ENDPROC
+END(double_fault)
+#endif
+
+ENTRY(invalid_TSS)
+ errorentry do_invalid_TSS
+END(invalid_TSS)
+
+ENTRY(segment_not_present)
+ errorentry do_segment_not_present
+END(segment_not_present)
+
+ /* runs on exception stack */
+ENTRY(stack_segment)
+/* XCPT_FRAME
+ paranoidentry do_stack_segment */
+ errorentry do_stack_segment
+/* jmp paranoid_exit1
+ CFI_ENDPROC */
+END(stack_segment)
+
+KPROBE_ENTRY(general_protection)
+ errorentry do_general_protection
+END(general_protection)
+ .previous .text
+
+ENTRY(alignment_check)
+ errorentry do_alignment_check
+END(alignment_check)
+
+ENTRY(divide_error)
+ zeroentry do_divide_error
+END(divide_error)
+
+ENTRY(spurious_interrupt_bug)
+ zeroentry do_spurious_interrupt_bug
+END(spurious_interrupt_bug)
+
+#ifdef CONFIG_X86_MCE
+ /* runs on exception stack */
+KPROBE_ENTRY(machine_check)
+ zeroentry do_machine_check
+END(machine_check)
+#endif
+
+/* Call softirq on interrupt stack. Interrupts are off. */
+ENTRY(call_softirq)
+ CFI_STARTPROC
+ push %rbp
+ CFI_ADJUST_CFA_OFFSET 8
+ CFI_REL_OFFSET rbp,0
+ mov %rsp,%rbp
+ CFI_DEF_CFA_REGISTER rbp
+ incl %gs:pda_irqcount
+ cmove %gs:pda_irqstackptr,%rsp
+ push %rbp # backlink for old unwinder
+ call __do_softirq
+ leaveq
+ CFI_DEF_CFA_REGISTER rsp
+ CFI_ADJUST_CFA_OFFSET -8
+ decl %gs:pda_irqcount
+ ret
+ CFI_ENDPROC
+ENDPROC(call_softirq)
+
+#ifdef CONFIG_STACK_UNWIND
+ENTRY(arch_unwind_init_running)
+ CFI_STARTPROC
+ movq %r15, R15(%rdi)
+ movq %r14, R14(%rdi)
+ xchgq %rsi, %rdx
+ movq %r13, R13(%rdi)
+ movq %r12, R12(%rdi)
+ xorl %eax, %eax
+ movq %rbp, RBP(%rdi)
+ movq %rbx, RBX(%rdi)
+ movq (%rsp), %rcx
+ movq %rax, R11(%rdi)
+ movq %rax, R10(%rdi)
+ movq %rax, R9(%rdi)
+ movq %rax, R8(%rdi)
+ movq %rax, RAX(%rdi)
+ movq %rax, RCX(%rdi)
+ movq %rax, RDX(%rdi)
+ movq %rax, RSI(%rdi)
+ movq %rax, RDI(%rdi)
+ movq %rax, ORIG_RAX(%rdi)
+ movq %rcx, RIP(%rdi)
+ leaq 8(%rsp), %rcx
+ movq $__KERNEL_CS, CS(%rdi)
+ movq %rax, EFLAGS(%rdi)
+ movq %rcx, RSP(%rdi)
+ movq $__KERNEL_DS, SS(%rdi)
+ jmpq *%rdx
+ CFI_ENDPROC
+ENDPROC(arch_unwind_init_running)
+#endif
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/kernel/head_64-xen.S 2010-11-08 17:27:03.000000000 +0100
@@ -0,0 +1,211 @@
+/*
+ * linux/arch/x86_64/kernel/head.S -- start in 32bit and switch to 64bit
+ *
+ * Copyright (C) 2000 Andrea Arcangeli <andrea@suse.de> SuSE
+ * Copyright (C) 2000 Pavel Machek <pavel@suse.cz>
+ * Copyright (C) 2000 Karsten Keil <kkeil@suse.de>
+ * Copyright (C) 2001,2002 Andi Kleen <ak@suse.de>
+ *
+ * $Id: head.S,v 1.49 2002/03/19 17:39:25 ak Exp $
+ *
+ * Jun Nakajima <jun.nakajima@intel.com>
+ * Modified for Xen
+ */
+
+
+#include <linux/linkage.h>
+#include <linux/threads.h>
+#include <linux/init.h>
+#include <linux/elfnote.h>
+#include <asm/desc.h>
+#include <asm/segment.h>
+#include <asm/page.h>
+#include <asm/msr.h>
+#include <asm/cache.h>
+#include <asm/dwarf2.h>
+#include <xen/interface/elfnote.h>
+
+ .section .bootstrap.text, "ax", @progbits
+ .code64
+ .globl startup_64
+startup_64:
+ movq $(init_thread_union+THREAD_SIZE-8),%rsp
+
+ /* rsi is pointer to startup info structure.
+ pass it to C */
+ movq %rsi,%rdi
+ pushq $0 # fake return address
+ jmp x86_64_start_kernel
+
+#ifdef CONFIG_ACPI_SLEEP
+.org 0xf00
+ .globl pGDT32
+pGDT32:
+ .word gdt_end-cpu_gdt_table-1
+ .long cpu_gdt_table-__START_KERNEL_map
+#endif
+ENTRY(stext)
+ENTRY(_stext)
+
+ $page = 0
+#define NEXT_PAGE(name) \
+ $page = $page + 1; \
+ .org $page * 0x1000; \
+ phys_##name = $page * 0x1000 + __PHYSICAL_START; \
+ENTRY(name)
+
+NEXT_PAGE(init_level4_pgt)
+ /* This gets initialized in x86_64_start_kernel */
+ .fill 512,8,0
+NEXT_PAGE(init_level4_user_pgt)
+ /*
+ * We update two pgd entries to make kernel and user pgd consistent
+ * at pgd_populate(). It can be used for kernel modules. So we place
+ * this page here for those cases to avoid memory corruption.
+ * We also use this page to establish the initial mapping for the
+ * vsyscall area.
+ */
+ .fill 512,8,0
+
+NEXT_PAGE(level3_kernel_pgt)
+ .fill 512,8,0
+
+ /*
+ * This is used for vsyscall area mapping as we have a different
+ * level4 page table for user.
+ */
+NEXT_PAGE(level3_user_pgt)
+ .fill 512,8,0
+
+NEXT_PAGE(hypercall_page)
+ CFI_STARTPROC
+ .rept 0x1000 / 0x20
+ .skip 1 /* push %rcx */
+ CFI_ADJUST_CFA_OFFSET 8
+ CFI_REL_OFFSET rcx,0
+ .skip 2 /* push %r11 */
+ CFI_ADJUST_CFA_OFFSET 8
+ CFI_REL_OFFSET rcx,0
+ .skip 5 /* mov $#,%eax */
+ .skip 2 /* syscall */
+ .skip 2 /* pop %r11 */
+ CFI_ADJUST_CFA_OFFSET -8
+ CFI_RESTORE r11
+ .skip 1 /* pop %rcx */
+ CFI_ADJUST_CFA_OFFSET -8
+ CFI_RESTORE rcx
+ .align 0x20,0 /* ret */
+ .endr
+ CFI_ENDPROC
+
+#undef NEXT_PAGE
+
+ .data
+/* Just dummy symbol to allow compilation. Not used in sleep path */
+#ifdef CONFIG_ACPI_SLEEP
+ .align PAGE_SIZE
+ENTRY(wakeup_level4_pgt)
+ .fill 512,8,0
+#endif
+
+ .data
+
+ .align 16
+ .globl cpu_gdt_descr
+cpu_gdt_descr:
+ .word gdt_end-cpu_gdt_table-1
+gdt:
+ .quad cpu_gdt_table
+#ifdef CONFIG_SMP
+ .rept NR_CPUS-1
+ .word 0
+ .quad 0
+ .endr
+#endif
+
+/* We need valid kernel segments for data and code in long mode too
+ * IRET will check the segment types kkeil 2000/10/28
+ * Also sysret mandates a special GDT layout
+ */
+
+ .section .data.page_aligned, "aw"
+ .align PAGE_SIZE
+
+/* The TLS descriptors are currently at a different place compared to i386.
+ Hopefully nobody expects them at a fixed place (Wine?) */
+
+ENTRY(cpu_gdt_table)
+ .quad 0x0000000000000000 /* NULL descriptor */
+ .quad 0x0 /* unused */
+ .quad 0x00af9a000000ffff /* __KERNEL_CS */
+ .quad 0x00cf92000000ffff /* __KERNEL_DS */
+ .quad 0x00cffa000000ffff /* __USER32_CS */
+ .quad 0x00cff2000000ffff /* __USER_DS, __USER32_DS */
+ .quad 0x00affa000000ffff /* __USER_CS */
+ .quad 0x00cf9a000000ffff /* __KERNEL32_CS */
+ .quad 0,0 /* TSS */
+ .quad 0,0 /* LDT */
+ .quad 0,0,0 /* three TLS descriptors */
+ .quad 0 /* unused */
+gdt_end:
+ /* asm/segment.h:GDT_ENTRIES must match this */
+ /* This should be a multiple of the cache line size */
+ /* GDTs of other CPUs are now dynamically allocated */
+
+ /* zero the remaining page */
+ .fill PAGE_SIZE / 8 - GDT_ENTRIES,8,0
+
+ .section .bss.page_aligned, "aw", @nobits
+ .align PAGE_SIZE
+ENTRY(empty_zero_page)
+ .skip PAGE_SIZE
+
+#if CONFIG_XEN_COMPAT <= 0x030002
+/*
+ * __xen_guest information
+ */
+.macro utoh value
+ .if (\value) < 0 || (\value) >= 0x10
+ utoh (((\value)>>4)&0x0fffffffffffffff)
+ .endif
+ .if ((\value) & 0xf) < 10
+ .byte '0' + ((\value) & 0xf)
+ .else
+ .byte 'A' + ((\value) & 0xf) - 10
+ .endif
+.endm
+
+.section __xen_guest
+ .ascii "GUEST_OS=linux,GUEST_VER=2.6"
+ .ascii ",XEN_VER=xen-3.0"
+ .ascii ",VIRT_BASE=0x"
+ utoh __START_KERNEL_map
+ .ascii ",ELF_PADDR_OFFSET=0x"
+ utoh __START_KERNEL_map
+ .ascii ",VIRT_ENTRY=0x"
+ utoh (__START_KERNEL_map + __PHYSICAL_START)
+ .ascii ",HYPERCALL_PAGE=0x"
+ utoh (phys_hypercall_page >> PAGE_SHIFT)
+ .ascii ",FEATURES=writable_page_tables"
+ .ascii "|writable_descriptor_tables"
+ .ascii "|auto_translated_physmap"
+ .ascii "|supervisor_mode_kernel"
+ .ascii ",LOADER=generic"
+ .byte 0
+#endif /* CONFIG_XEN_COMPAT <= 0x030002 */
+
+ ELFNOTE(Xen, XEN_ELFNOTE_GUEST_OS, .asciz, "linux")
+ ELFNOTE(Xen, XEN_ELFNOTE_GUEST_VERSION, .asciz, "2.6")
+ ELFNOTE(Xen, XEN_ELFNOTE_XEN_VERSION, .asciz, "xen-3.0")
+ ELFNOTE(Xen, XEN_ELFNOTE_VIRT_BASE, .quad, __START_KERNEL_map)
+#if CONFIG_XEN_COMPAT <= 0x030002
+ ELFNOTE(Xen, XEN_ELFNOTE_PADDR_OFFSET, .quad, __START_KERNEL_map)
+#else
+ ELFNOTE(Xen, XEN_ELFNOTE_PADDR_OFFSET, .quad, 0)
+#endif
+ ELFNOTE(Xen, XEN_ELFNOTE_ENTRY, .quad, startup_64)
+ ELFNOTE(Xen, XEN_ELFNOTE_HYPERCALL_PAGE, .quad, hypercall_page)
+ ELFNOTE(Xen, XEN_ELFNOTE_L1_MFN_VALID, .quad, _PAGE_PRESENT,_PAGE_PRESENT)
+ ELFNOTE(Xen, XEN_ELFNOTE_FEATURES, .asciz, "writable_page_tables|writable_descriptor_tables|auto_translated_physmap|supervisor_mode_kernel")
+ ELFNOTE(Xen, XEN_ELFNOTE_LOADER, .asciz, "generic")
+ ELFNOTE(Xen, XEN_ELFNOTE_SUSPEND_CANCEL, .long, 1)
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/kernel/head64-xen.c 2007-06-12 13:13:01.000000000 +0200
@@ -0,0 +1,162 @@
+/*
+ * linux/arch/x86_64/kernel/head64.c -- prepare to run common code
+ *
+ * Copyright (C) 2000 Andrea Arcangeli <andrea@suse.de> SuSE
+ *
+ * Jun Nakajima <jun.nakajima@intel.com>
+ * Modified for Xen.
+ */
+
+#include <linux/init.h>
+#include <linux/linkage.h>
+#include <linux/types.h>
+#include <linux/kernel.h>
+#include <linux/string.h>
+#include <linux/percpu.h>
+#include <linux/module.h>
+
+#include <asm/processor.h>
+#include <asm/proto.h>
+#include <asm/smp.h>
+#include <asm/bootsetup.h>
+#include <asm/setup.h>
+#include <asm/desc.h>
+#include <asm/pgtable.h>
+#include <asm/sections.h>
+
+unsigned long start_pfn;
+
+/* Don't add a printk in there. printk relies on the PDA which is not initialized
+ yet. */
+#if 0
+static void __init clear_bss(void)
+{
+ memset(__bss_start, 0,
+ (unsigned long) __bss_stop - (unsigned long) __bss_start);
+}
+#endif
+
+#define NEW_CL_POINTER 0x228 /* Relative to real mode data */
+#define OLD_CL_MAGIC_ADDR 0x90020
+#define OLD_CL_MAGIC 0xA33F
+#define OLD_CL_BASE_ADDR 0x90000
+#define OLD_CL_OFFSET 0x90022
+
+extern char saved_command_line[];
+
+static void __init copy_bootdata(char *real_mode_data)
+{
+#ifndef CONFIG_XEN
+ int new_data;
+ char * command_line;
+
+ memcpy(x86_boot_params, real_mode_data, BOOT_PARAM_SIZE);
+ new_data = *(int *) (x86_boot_params + NEW_CL_POINTER);
+ if (!new_data) {
+ if (OLD_CL_MAGIC != * (u16 *) OLD_CL_MAGIC_ADDR) {
+ printk("so old bootloader that it does not support commandline?!\n");
+ return;
+ }
+ new_data = OLD_CL_BASE_ADDR + * (u16 *) OLD_CL_OFFSET;
+ printk("old bootloader convention, maybe loadlin?\n");
+ }
+ command_line = (char *) ((u64)(new_data));
+ memcpy(saved_command_line, command_line, COMMAND_LINE_SIZE);
+#else
+ int max_cmdline;
+
+ if ((max_cmdline = MAX_GUEST_CMDLINE) > COMMAND_LINE_SIZE)
+ max_cmdline = COMMAND_LINE_SIZE;
+ memcpy(saved_command_line, xen_start_info->cmd_line, max_cmdline);
+ saved_command_line[max_cmdline-1] = '\0';
+#endif
+ printk("Bootdata ok (command line is %s)\n", saved_command_line);
+}
+
+static void __init setup_boot_cpu_data(void)
+{
+ unsigned int dummy, eax;
+
+ /* get vendor info */
+ cpuid(0, (unsigned int *)&boot_cpu_data.cpuid_level,
+ (unsigned int *)&boot_cpu_data.x86_vendor_id[0],
+ (unsigned int *)&boot_cpu_data.x86_vendor_id[8],
+ (unsigned int *)&boot_cpu_data.x86_vendor_id[4]);
+
+ /* get cpu type */
+ cpuid(1, &eax, &dummy, &dummy,
+ (unsigned int *) &boot_cpu_data.x86_capability);
+ boot_cpu_data.x86 = (eax >> 8) & 0xf;
+ boot_cpu_data.x86_model = (eax >> 4) & 0xf;
+ boot_cpu_data.x86_mask = eax & 0xf;
+}
+
+#include <xen/interface/memory.h>
+unsigned long *machine_to_phys_mapping;
+EXPORT_SYMBOL(machine_to_phys_mapping);
+unsigned int machine_to_phys_order;
+EXPORT_SYMBOL(machine_to_phys_order);
+
+void __init x86_64_start_kernel(char * real_mode_data)
+{
+ struct xen_machphys_mapping mapping;
+ unsigned long machine_to_phys_nr_ents;
+ char *s;
+ int i;
+
+ setup_xen_features();
+
+ xen_start_info = (struct start_info *)real_mode_data;
+ if (!xen_feature(XENFEAT_auto_translated_physmap))
+ phys_to_machine_mapping =
+ (unsigned long *)xen_start_info->mfn_list;
+ start_pfn = (__pa(xen_start_info->pt_base) >> PAGE_SHIFT) +
+ xen_start_info->nr_pt_frames;
+
+ machine_to_phys_mapping = (unsigned long *)MACH2PHYS_VIRT_START;
+ machine_to_phys_nr_ents = MACH2PHYS_NR_ENTRIES;
+ if (HYPERVISOR_memory_op(XENMEM_machphys_mapping, &mapping) == 0) {
+ machine_to_phys_mapping = (unsigned long *)mapping.v_start;
+ machine_to_phys_nr_ents = mapping.max_mfn + 1;
+ }
+ while ((1UL << machine_to_phys_order) < machine_to_phys_nr_ents )
+ machine_to_phys_order++;
+
+#if 0
+ for (i = 0; i < 256; i++)
+ set_intr_gate(i, early_idt_handler);
+ asm volatile("lidt %0" :: "m" (idt_descr));
+#endif
+
+ /*
+ * This must be called really, really early:
+ */
+ lockdep_init();
+
+ for (i = 0; i < NR_CPUS; i++)
+ cpu_pda(i) = &boot_cpu_pda[i];
+
+ pda_init(0);
+ copy_bootdata(real_mode_data);
+#ifdef CONFIG_SMP
+ cpu_set(0, cpu_online_map);
+#endif
+ s = strstr(saved_command_line, "earlyprintk=");
+ if (s != NULL)
+ setup_early_printk(strchr(s, '=') + 1);
+#ifdef CONFIG_NUMA
+ s = strstr(saved_command_line, "numa=");
+ if (s != NULL)
+ numa_setup(s+5);
+#endif
+#ifdef CONFIG_X86_IO_APIC
+ if (strstr(saved_command_line, "disableapic"))
+ disable_apic = 1;
+#endif
+ /* You need early console to see that */
+ if (__pa_symbol(&_end) >= KERNEL_TEXT_SIZE)
+ panic("Kernel too big for kernel mapping\n");
+
+ setup_boot_cpu_data();
+ start_kernel();
+}
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/kernel/io_apic_64-xen.c 2009-03-18 10:39:31.000000000 +0100
@@ -0,0 +1,2270 @@
+/*
+ * Intel IO-APIC support for multi-Pentium hosts.
+ *
+ * Copyright (C) 1997, 1998, 1999, 2000 Ingo Molnar, Hajnalka Szabo
+ *
+ * Many thanks to Stig Venaas for trying out countless experimental
+ * patches and reporting/debugging problems patiently!
+ *
+ * (c) 1999, Multiple IO-APIC support, developed by
+ * Ken-ichi Yaku <yaku@css1.kbnes.nec.co.jp> and
+ * Hidemi Kishimoto <kisimoto@css1.kbnes.nec.co.jp>,
+ * further tested and cleaned up by Zach Brown <zab@redhat.com>
+ * and Ingo Molnar <mingo@redhat.com>
+ *
+ * Fixes
+ * Maciej W. Rozycki : Bits for genuine 82489DX APICs;
+ * thanks to Eric Gilmore
+ * and Rolf G. Tews
+ * for testing these extensively
+ * Paul Diefenbaugh : Added full ACPI support
+ */
+
+#include <linux/mm.h>
+#include <linux/interrupt.h>
+#include <linux/init.h>
+#include <linux/delay.h>
+#include <linux/sched.h>
+#include <linux/smp_lock.h>
+#include <linux/mc146818rtc.h>
+#include <linux/acpi.h>
+#include <linux/sysdev.h>
+#ifdef CONFIG_ACPI
+#include <acpi/acpi_bus.h>
+#endif
+
+#include <asm/io.h>
+#include <asm/smp.h>
+#include <asm/desc.h>
+#include <asm/proto.h>
+#include <asm/mach_apic.h>
+#include <asm/acpi.h>
+#include <asm/dma.h>
+#include <asm/nmi.h>
+
+#define __apicdebuginit __init
+
+int sis_apic_bug; /* not actually supported, dummy for compile */
+
+static int no_timer_check;
+
+int disable_timer_pin_1 __initdata;
+
+#ifndef CONFIG_XEN
+int timer_over_8254 __initdata = 0;
+
+/* Where if anywhere is the i8259 connect in external int mode */
+static struct { int pin, apic; } ioapic_i8259 = { -1, -1 };
+#endif
+
+static DEFINE_SPINLOCK(ioapic_lock);
+static DEFINE_SPINLOCK(vector_lock);
+
+/*
+ * # of IRQ routing registers
+ */
+int nr_ioapic_registers[MAX_IO_APICS];
+
+/*
+ * Rough estimation of how many shared IRQs there are, can
+ * be changed anytime.
+ */
+#define MAX_PLUS_SHARED_IRQS NR_IRQ_VECTORS
+#define PIN_MAP_SIZE (MAX_PLUS_SHARED_IRQS + NR_IRQS)
+
+/*
+ * This is performance-critical, we want to do it O(1)
+ *
+ * the indexing order of this array favors 1:1 mappings
+ * between pins and IRQs.
+ */
+
+static struct irq_pin_list {
+ short apic, pin, next;
+} irq_2_pin[PIN_MAP_SIZE];
+
+int vector_irq[NR_VECTORS] __read_mostly = { [0 ... NR_VECTORS - 1] = -1};
+#ifdef CONFIG_PCI_MSI
+#define vector_to_irq(vector) \
+ (platform_legacy_irq(vector) ? vector : vector_irq[vector])
+#else
+#define vector_to_irq(vector) (vector)
+#endif
+
+#ifdef CONFIG_XEN
+
+#include <xen/interface/xen.h>
+#include <xen/interface/physdev.h>
+#include <xen/evtchn.h>
+
+/* Fake i8259 */
+#define make_8259A_irq(_irq) (io_apic_irqs &= ~(1UL<<(_irq)))
+#define disable_8259A_irq(_irq) ((void)0)
+#define i8259A_irq_pending(_irq) (0)
+
+unsigned long io_apic_irqs;
+
+static inline unsigned int xen_io_apic_read(unsigned int apic, unsigned int reg)
+{
+ struct physdev_apic apic_op;
+ int ret;
+
+ apic_op.apic_physbase = mp_ioapics[apic].mpc_apicaddr;
+ apic_op.reg = reg;
+ ret = HYPERVISOR_physdev_op(PHYSDEVOP_apic_read, &apic_op);
+ if (ret)
+ return ret;
+ return apic_op.value;
+}
+
+static inline void xen_io_apic_write(unsigned int apic, unsigned int reg, unsigned int value)
+{
+ struct physdev_apic apic_op;
+
+ apic_op.apic_physbase = mp_ioapics[apic].mpc_apicaddr;
+ apic_op.reg = reg;
+ apic_op.value = value;
+ WARN_ON(HYPERVISOR_physdev_op(PHYSDEVOP_apic_write, &apic_op));
+}
+
+#define io_apic_read(a,r) xen_io_apic_read(a,r)
+#define io_apic_write(a,r,v) xen_io_apic_write(a,r,v)
+
+#define clear_IO_APIC() ((void)0)
+
+#else
+
+#ifdef CONFIG_SMP
+static void set_ioapic_affinity_irq(unsigned int irq, cpumask_t mask)
+{
+ unsigned long flags;
+ unsigned int dest;
+ cpumask_t tmp;
+
+ cpus_and(tmp, mask, cpu_online_map);
+ if (cpus_empty(tmp))
+ tmp = TARGET_CPUS;
+
+ cpus_and(mask, tmp, CPU_MASK_ALL);
+
+ dest = cpu_mask_to_apicid(mask);
+
+ /*
+ * Only the high 8 bits are valid.
+ */
+ dest = SET_APIC_LOGICAL_ID(dest);
+
+ spin_lock_irqsave(&ioapic_lock, flags);
+ __DO_ACTION(1, = dest, )
+ set_irq_info(irq, mask);
+ spin_unlock_irqrestore(&ioapic_lock, flags);
+}
+#endif
+
+#endif /* !CONFIG_XEN */
+
+/*
+ * The common case is 1:1 IRQ<->pin mappings. Sometimes there are
+ * shared ISA-space IRQs, so we have to support them. We are super
+ * fast in the common case, and fast for shared ISA-space IRQs.
+ */
+static void add_pin_to_irq(unsigned int irq, int apic, int pin)
+{
+ static int first_free_entry = NR_IRQS;
+ struct irq_pin_list *entry = irq_2_pin + irq;
+
+ BUG_ON(irq >= NR_IRQS);
+ while (entry->next)
+ entry = irq_2_pin + entry->next;
+
+ if (entry->pin != -1) {
+ entry->next = first_free_entry;
+ entry = irq_2_pin + entry->next;
+ if (++first_free_entry >= PIN_MAP_SIZE)
+ panic("io_apic.c: ran out of irq_2_pin entries!");
+ }
+ entry->apic = apic;
+ entry->pin = pin;
+}
+
+#ifndef CONFIG_XEN
+#define __DO_ACTION(R, ACTION, FINAL) \
+ \
+{ \
+ int pin; \
+ struct irq_pin_list *entry = irq_2_pin + irq; \
+ \
+ BUG_ON(irq >= NR_IRQS); \
+ for (;;) { \
+ unsigned int reg; \
+ pin = entry->pin; \
+ if (pin == -1) \
+ break; \
+ reg = io_apic_read(entry->apic, 0x10 + R + pin*2); \
+ reg ACTION; \
+ io_apic_modify(entry->apic, reg); \
+ if (!entry->next) \
+ break; \
+ entry = irq_2_pin + entry->next; \
+ } \
+ FINAL; \
+}
+
+#define DO_ACTION(name,R,ACTION, FINAL) \
+ \
+ static void name##_IO_APIC_irq (unsigned int irq) \
+ __DO_ACTION(R, ACTION, FINAL)
+
+DO_ACTION( __mask, 0, |= 0x00010000, io_apic_sync(entry->apic) )
+ /* mask = 1 */
+DO_ACTION( __unmask, 0, &= 0xfffeffff, )
+ /* mask = 0 */
+
+static void mask_IO_APIC_irq (unsigned int irq)
+{
+ unsigned long flags;
+
+ spin_lock_irqsave(&ioapic_lock, flags);
+ __mask_IO_APIC_irq(irq);
+ spin_unlock_irqrestore(&ioapic_lock, flags);
+}
+
+static void unmask_IO_APIC_irq (unsigned int irq)
+{
+ unsigned long flags;
+
+ spin_lock_irqsave(&ioapic_lock, flags);
+ __unmask_IO_APIC_irq(irq);
+ spin_unlock_irqrestore(&ioapic_lock, flags);
+}
+
+static void clear_IO_APIC_pin(unsigned int apic, unsigned int pin)
+{
+ struct IO_APIC_route_entry entry;
+ unsigned long flags;
+
+ /* Check delivery_mode to be sure we're not clearing an SMI pin */
+ spin_lock_irqsave(&ioapic_lock, flags);
+ *(((int*)&entry) + 0) = io_apic_read(apic, 0x10 + 2 * pin);
+ *(((int*)&entry) + 1) = io_apic_read(apic, 0x11 + 2 * pin);
+ spin_unlock_irqrestore(&ioapic_lock, flags);
+ if (entry.delivery_mode == dest_SMI)
+ return;
+ /*
+ * Disable it in the IO-APIC irq-routing table:
+ */
+ memset(&entry, 0, sizeof(entry));
+ entry.mask = 1;
+ spin_lock_irqsave(&ioapic_lock, flags);
+ io_apic_write(apic, 0x10 + 2 * pin, *(((int *)&entry) + 0));
+ io_apic_write(apic, 0x11 + 2 * pin, *(((int *)&entry) + 1));
+ spin_unlock_irqrestore(&ioapic_lock, flags);
+}
+
+static void clear_IO_APIC (void)
+{
+ int apic, pin;
+
+ for (apic = 0; apic < nr_ioapics; apic++)
+ for (pin = 0; pin < nr_ioapic_registers[apic]; pin++)
+ clear_IO_APIC_pin(apic, pin);
+}
+
+#endif /* !CONFIG_XEN */
+
+static u8 gsi_2_irq[NR_IRQ_VECTORS] = { [0 ... NR_IRQ_VECTORS-1] = 0xFF };
+
+/*
+ * support for broken MP BIOSs, enables hand-redirection of PIRQ0-7 to
+ * specific CPU-side IRQs.
+ */
+
+#define MAX_PIRQS 8
+static int pirq_entries [MAX_PIRQS];
+static int pirqs_enabled;
+int skip_ioapic_setup;
+int ioapic_force;
+
+/* dummy parsing: see setup.c */
+
+static int __init disable_ioapic_setup(char *str)
+{
+ skip_ioapic_setup = 1;
+ return 1;
+}
+
+static int __init enable_ioapic_setup(char *str)
+{
+ ioapic_force = 1;
+ skip_ioapic_setup = 0;
+ return 1;
+}
+
+__setup("noapic", disable_ioapic_setup);
+__setup("apic", enable_ioapic_setup);
+
+#ifndef CONFIG_XEN
+static int __init setup_disable_8254_timer(char *s)
+{
+ timer_over_8254 = -1;
+ return 1;
+}
+static int __init setup_enable_8254_timer(char *s)
+{
+ timer_over_8254 = 2;
+ return 1;
+}
+
+__setup("disable_8254_timer", setup_disable_8254_timer);
+__setup("enable_8254_timer", setup_enable_8254_timer);
+#endif /* !CONFIG_XEN */
+
+#include <asm/pci-direct.h>
+#include <linux/pci_ids.h>
+#include <linux/pci.h>
+
+
+#ifdef CONFIG_ACPI
+
+static int nvidia_hpet_detected __initdata;
+
+static int __init nvidia_hpet_check(unsigned long phys, unsigned long size)
+{
+ nvidia_hpet_detected = 1;
+ return 0;
+}
+#endif
+
+/* Temporary Hack. Nvidia and VIA boards currently only work with IO-APIC
+ off. Check for an Nvidia or VIA PCI bridge and turn it off.
+ Use pci direct infrastructure because this runs before the PCI subsystem.
+
+ Can be overwritten with "apic"
+
+ And another hack to disable the IOMMU on VIA chipsets.
+
+ ... and others. Really should move this somewhere else.
+
+ Kludge-O-Rama. */
+void __init check_ioapic(void)
+{
+ int num,slot,func;
+ /* Poor man's PCI discovery */
+ for (num = 0; num < 32; num++) {
+ for (slot = 0; slot < 32; slot++) {
+ for (func = 0; func < 8; func++) {
+ u32 class;
+ u32 vendor;
+ u8 type;
+ class = read_pci_config(num,slot,func,
+ PCI_CLASS_REVISION);
+ if (class == 0xffffffff)
+ break;
+
+ if ((class >> 16) != PCI_CLASS_BRIDGE_PCI)
+ continue;
+
+ vendor = read_pci_config(num, slot, func,
+ PCI_VENDOR_ID);
+ vendor &= 0xffff;
+ switch (vendor) {
+ case PCI_VENDOR_ID_VIA:
+#ifdef CONFIG_IOMMU
+ if ((end_pfn > MAX_DMA32_PFN ||
+ force_iommu) &&
+ !iommu_aperture_allowed) {
+ printk(KERN_INFO
+ "Looks like a VIA chipset. Disabling IOMMU. Override with \"iommu=allowed\"\n");
+ iommu_aperture_disabled = 1;
+ }
+#endif
+ return;
+ case PCI_VENDOR_ID_NVIDIA:
+#ifdef CONFIG_ACPI
+ /*
+ * All timer overrides on Nvidia are
+ * wrong unless HPET is enabled.
+ */
+ nvidia_hpet_detected = 0;
+ acpi_table_parse(ACPI_HPET,
+ nvidia_hpet_check);
+ if (nvidia_hpet_detected == 0) {
+ acpi_skip_timer_override = 1;
+ printk(KERN_INFO "Nvidia board "
+ "detected. Ignoring ACPI "
+ "timer override.\n");
+ }
+#endif
+ /* RED-PEN skip them on mptables too? */
+ return;
+ case PCI_VENDOR_ID_ATI:
+
+ /* This should be actually default, but
+ for 2.6.16 let's do it for ATI only where
+ it's really needed. */
+#ifndef CONFIG_XEN
+ if (timer_over_8254 == 1) {
+ timer_over_8254 = 0;
+ printk(KERN_INFO
+ "ATI board detected. Disabling timer routing over 8254.\n");
+ }
+#endif
+ return;
+ }
+
+
+ /* No multi-function device? */
+ type = read_pci_config_byte(num,slot,func,
+ PCI_HEADER_TYPE);
+ if (!(type & 0x80))
+ break;
+ }
+ }
+ }
+}
+
+static int __init ioapic_pirq_setup(char *str)
+{
+ int i, max;
+ int ints[MAX_PIRQS+1];
+
+ get_options(str, ARRAY_SIZE(ints), ints);
+
+ for (i = 0; i < MAX_PIRQS; i++)
+ pirq_entries[i] = -1;
+
+ pirqs_enabled = 1;
+ apic_printk(APIC_VERBOSE, "PIRQ redirection, working around broken MP-BIOS.\n");
+ max = MAX_PIRQS;
+ if (ints[0] < MAX_PIRQS)
+ max = ints[0];
+
+ for (i = 0; i < max; i++) {
+ apic_printk(APIC_VERBOSE, "... PIRQ%d -> IRQ %d\n", i, ints[i+1]);
+ /*
+ * PIRQs are mapped upside down, usually.
+ */
+ pirq_entries[MAX_PIRQS-i-1] = ints[i+1];
+ }
+ return 1;
+}
+
+__setup("pirq=", ioapic_pirq_setup);
+
+/*
+ * Find the IRQ entry number of a certain pin.
+ */
+static int find_irq_entry(int apic, int pin, int type)
+{
+ int i;
+
+ for (i = 0; i < mp_irq_entries; i++)
+ if (mp_irqs[i].mpc_irqtype == type &&
+ (mp_irqs[i].mpc_dstapic == mp_ioapics[apic].mpc_apicid ||
+ mp_irqs[i].mpc_dstapic == MP_APIC_ALL) &&
+ mp_irqs[i].mpc_dstirq == pin)
+ return i;
+
+ return -1;
+}
+
+#ifndef CONFIG_XEN
+/*
+ * Find the pin to which IRQ[irq] (ISA) is connected
+ */
+static int __init find_isa_irq_pin(int irq, int type)
+{
+ int i;
+
+ for (i = 0; i < mp_irq_entries; i++) {
+ int lbus = mp_irqs[i].mpc_srcbus;
+
+ if ((mp_bus_id_to_type[lbus] == MP_BUS_ISA ||
+ mp_bus_id_to_type[lbus] == MP_BUS_EISA ||
+ mp_bus_id_to_type[lbus] == MP_BUS_MCA) &&
+ (mp_irqs[i].mpc_irqtype == type) &&
+ (mp_irqs[i].mpc_srcbusirq == irq))
+
+ return mp_irqs[i].mpc_dstirq;
+ }
+ return -1;
+}
+
+static int __init find_isa_irq_apic(int irq, int type)
+{
+ int i;
+
+ for (i = 0; i < mp_irq_entries; i++) {
+ int lbus = mp_irqs[i].mpc_srcbus;
+
+ if ((mp_bus_id_to_type[lbus] == MP_BUS_ISA ||
+ mp_bus_id_to_type[lbus] == MP_BUS_EISA ||
+ mp_bus_id_to_type[lbus] == MP_BUS_MCA) &&
+ (mp_irqs[i].mpc_irqtype == type) &&
+ (mp_irqs[i].mpc_srcbusirq == irq))
+ break;
+ }
+ if (i < mp_irq_entries) {
+ int apic;
+ for(apic = 0; apic < nr_ioapics; apic++) {
+ if (mp_ioapics[apic].mpc_apicid == mp_irqs[i].mpc_dstapic)
+ return apic;
+ }
+ }
+
+ return -1;
+}
+#endif
+
+/*
+ * Find a specific PCI IRQ entry.
+ * Not an __init, possibly needed by modules
+ */
+static int pin_2_irq(int idx, int apic, int pin);
+
+int IO_APIC_get_PCI_irq_vector(int bus, int slot, int pin)
+{
+ int apic, i, best_guess = -1;
+
+ apic_printk(APIC_DEBUG, "querying PCI -> IRQ mapping bus:%d, slot:%d, pin:%d.\n",
+ bus, slot, pin);
+ if (mp_bus_id_to_pci_bus[bus] == -1) {
+ apic_printk(APIC_VERBOSE, "PCI BIOS passed nonexistent PCI bus %d!\n", bus);
+ return -1;
+ }
+ for (i = 0; i < mp_irq_entries; i++) {
+ int lbus = mp_irqs[i].mpc_srcbus;
+
+ for (apic = 0; apic < nr_ioapics; apic++)
+ if (mp_ioapics[apic].mpc_apicid == mp_irqs[i].mpc_dstapic ||
+ mp_irqs[i].mpc_dstapic == MP_APIC_ALL)
+ break;
+
+ if ((mp_bus_id_to_type[lbus] == MP_BUS_PCI) &&
+ !mp_irqs[i].mpc_irqtype &&
+ (bus == lbus) &&
+ (slot == ((mp_irqs[i].mpc_srcbusirq >> 2) & 0x1f))) {
+ int irq = pin_2_irq(i,apic,mp_irqs[i].mpc_dstirq);
+
+ if (!(apic || IO_APIC_IRQ(irq)))
+ continue;
+
+ if (pin == (mp_irqs[i].mpc_srcbusirq & 3))
+ return irq;
+ /*
+ * Use the first all-but-pin matching entry as a
+ * best-guess fuzzy result for broken mptables.
+ */
+ if (best_guess < 0)
+ best_guess = irq;
+ }
+ }
+ BUG_ON(best_guess >= NR_IRQS);
+ return best_guess;
+}
+
+/*
+ * EISA Edge/Level control register, ELCR
+ */
+static int EISA_ELCR(unsigned int irq)
+{
+ if (irq < 16) {
+ unsigned int port = 0x4d0 + (irq >> 3);
+ return (inb(port) >> (irq & 7)) & 1;
+ }
+ apic_printk(APIC_VERBOSE, "Broken MPtable reports ISA irq %d\n", irq);
+ return 0;
+}
+
+/* EISA interrupts are always polarity zero and can be edge or level
+ * trigger depending on the ELCR value. If an interrupt is listed as
+ * EISA conforming in the MP table, that means its trigger type must
+ * be read in from the ELCR */
+
+#define default_EISA_trigger(idx) (EISA_ELCR(mp_irqs[idx].mpc_srcbusirq))
+#define default_EISA_polarity(idx) (0)
+
+/* ISA interrupts are always polarity zero edge triggered,
+ * when listed as conforming in the MP table. */
+
+#define default_ISA_trigger(idx) (0)
+#define default_ISA_polarity(idx) (0)
+
+/* PCI interrupts are always polarity one level triggered,
+ * when listed as conforming in the MP table. */
+
+#define default_PCI_trigger(idx) (1)
+#define default_PCI_polarity(idx) (1)
+
+/* MCA interrupts are always polarity zero level triggered,
+ * when listed as conforming in the MP table. */
+
+#define default_MCA_trigger(idx) (1)
+#define default_MCA_polarity(idx) (0)
+
+static int __init MPBIOS_polarity(int idx)
+{
+ int bus = mp_irqs[idx].mpc_srcbus;
+ int polarity;
+
+ /*
+ * Determine IRQ line polarity (high active or low active):
+ */
+ switch (mp_irqs[idx].mpc_irqflag & 3)
+ {
+ case 0: /* conforms, ie. bus-type dependent polarity */
+ {
+ switch (mp_bus_id_to_type[bus])
+ {
+ case MP_BUS_ISA: /* ISA pin */
+ {
+ polarity = default_ISA_polarity(idx);
+ break;
+ }
+ case MP_BUS_EISA: /* EISA pin */
+ {
+ polarity = default_EISA_polarity(idx);
+ break;
+ }
+ case MP_BUS_PCI: /* PCI pin */
+ {
+ polarity = default_PCI_polarity(idx);
+ break;
+ }
+ case MP_BUS_MCA: /* MCA pin */
+ {
+ polarity = default_MCA_polarity(idx);
+ break;
+ }
+ default:
+ {
+ printk(KERN_WARNING "broken BIOS!!\n");
+ polarity = 1;
+ break;
+ }
+ }
+ break;
+ }
+ case 1: /* high active */
+ {
+ polarity = 0;
+ break;
+ }
+ case 2: /* reserved */
+ {
+ printk(KERN_WARNING "broken BIOS!!\n");
+ polarity = 1;
+ break;
+ }
+ case 3: /* low active */
+ {
+ polarity = 1;
+ break;
+ }
+ default: /* invalid */
+ {
+ printk(KERN_WARNING "broken BIOS!!\n");
+ polarity = 1;
+ break;
+ }
+ }
+ return polarity;
+}
+
+static int MPBIOS_trigger(int idx)
+{
+ int bus = mp_irqs[idx].mpc_srcbus;
+ int trigger;
+
+ /*
+ * Determine IRQ trigger mode (edge or level sensitive):
+ */
+ switch ((mp_irqs[idx].mpc_irqflag>>2) & 3)
+ {
+ case 0: /* conforms, ie. bus-type dependent */
+ {
+ switch (mp_bus_id_to_type[bus])
+ {
+ case MP_BUS_ISA: /* ISA pin */
+ {
+ trigger = default_ISA_trigger(idx);
+ break;
+ }
+ case MP_BUS_EISA: /* EISA pin */
+ {
+ trigger = default_EISA_trigger(idx);
+ break;
+ }
+ case MP_BUS_PCI: /* PCI pin */
+ {
+ trigger = default_PCI_trigger(idx);
+ break;
+ }
+ case MP_BUS_MCA: /* MCA pin */
+ {
+ trigger = default_MCA_trigger(idx);
+ break;
+ }
+ default:
+ {
+ printk(KERN_WARNING "broken BIOS!!\n");
+ trigger = 1;
+ break;
+ }
+ }
+ break;
+ }
+ case 1: /* edge */
+ {
+ trigger = 0;
+ break;
+ }
+ case 2: /* reserved */
+ {
+ printk(KERN_WARNING "broken BIOS!!\n");
+ trigger = 1;
+ break;
+ }
+ case 3: /* level */
+ {
+ trigger = 1;
+ break;
+ }
+ default: /* invalid */
+ {
+ printk(KERN_WARNING "broken BIOS!!\n");
+ trigger = 0;
+ break;
+ }
+ }
+ return trigger;
+}
+
+static inline int irq_polarity(int idx)
+{
+ return MPBIOS_polarity(idx);
+}
+
+static inline int irq_trigger(int idx)
+{
+ return MPBIOS_trigger(idx);
+}
+
+static int next_irq = 16;
+
+/*
+ * gsi_irq_sharing -- Name overload! "irq" can be either a legacy IRQ
+ * in the range 0-15, a linux IRQ in the range 0-223, or a GSI number
+ * from ACPI, which can reach 800 in large boxen.
+ *
+ * Compact the sparse GSI space into a sequential IRQ series and reuse
+ * vectors if possible.
+ */
+int gsi_irq_sharing(int gsi)
+{
+ int i, tries, vector;
+
+ BUG_ON(gsi >= NR_IRQ_VECTORS);
+
+ if (platform_legacy_irq(gsi))
+ return gsi;
+
+ if (gsi_2_irq[gsi] != 0xFF)
+ return (int)gsi_2_irq[gsi];
+
+ tries = NR_IRQS;
+ try_again:
+ vector = assign_irq_vector(gsi);
+
+ /*
+ * Sharing vectors means sharing IRQs, so scan irq_vectors for previous
+ * use of vector and if found, return that IRQ. However, we never want
+ * to share legacy IRQs, which usually have a different trigger mode
+ * than PCI.
+ */
+ for (i = 0; i < NR_IRQS; i++)
+ if (IO_APIC_VECTOR(i) == vector)
+ break;
+ if (platform_legacy_irq(i)) {
+ if (--tries >= 0) {
+ IO_APIC_VECTOR(i) = 0;
+ goto try_again;
+ }
+ panic("gsi_irq_sharing: didn't find an IRQ using vector 0x%02X for GSI %d", vector, gsi);
+ }
+ if (i < NR_IRQS) {
+ gsi_2_irq[gsi] = i;
+ printk(KERN_INFO "GSI %d sharing vector 0x%02X and IRQ %d\n",
+ gsi, vector, i);
+ return i;
+ }
+
+ i = next_irq++;
+ BUG_ON(i >= NR_IRQS);
+ gsi_2_irq[gsi] = i;
+ IO_APIC_VECTOR(i) = vector;
+ printk(KERN_INFO "GSI %d assigned vector 0x%02X and IRQ %d\n",
+ gsi, vector, i);
+ return i;
+}
+
+static int pin_2_irq(int idx, int apic, int pin)
+{
+ int irq, i;
+ int bus = mp_irqs[idx].mpc_srcbus;
+
+ /*
+ * Debugging check, we are in big trouble if this message pops up!
+ */
+ if (mp_irqs[idx].mpc_dstirq != pin)
+ printk(KERN_ERR "broken BIOS or MPTABLE parser, ayiee!!\n");
+
+ switch (mp_bus_id_to_type[bus])
+ {
+ case MP_BUS_ISA: /* ISA pin */
+ case MP_BUS_EISA:
+ case MP_BUS_MCA:
+ {
+ irq = mp_irqs[idx].mpc_srcbusirq;
+ break;
+ }
+ case MP_BUS_PCI: /* PCI pin */
+ {
+ /*
+ * PCI IRQs are mapped in order
+ */
+ i = irq = 0;
+ while (i < apic)
+ irq += nr_ioapic_registers[i++];
+ irq += pin;
+ irq = gsi_irq_sharing(irq);
+ break;
+ }
+ default:
+ {
+ printk(KERN_ERR "unknown bus type %d.\n",bus);
+ irq = 0;
+ break;
+ }
+ }
+ BUG_ON(irq >= NR_IRQS);
+
+ /*
+ * PCI IRQ command line redirection. Yes, limits are hardcoded.
+ */
+ if ((pin >= 16) && (pin <= 23)) {
+ if (pirq_entries[pin-16] != -1) {
+ if (!pirq_entries[pin-16]) {
+ apic_printk(APIC_VERBOSE, "disabling PIRQ%d\n", pin-16);
+ } else {
+ irq = pirq_entries[pin-16];
+ apic_printk(APIC_VERBOSE, "using PIRQ%d -> IRQ %d\n",
+ pin-16, irq);
+ }
+ }
+ }
+ BUG_ON(irq >= NR_IRQS);
+ return irq;
+}
+
+static inline int IO_APIC_irq_trigger(int irq)
+{
+ int apic, idx, pin;
+
+ for (apic = 0; apic < nr_ioapics; apic++) {
+ for (pin = 0; pin < nr_ioapic_registers[apic]; pin++) {
+ idx = find_irq_entry(apic,pin,mp_INT);
+ if ((idx != -1) && (irq == pin_2_irq(idx,apic,pin)))
+ return irq_trigger(idx);
+ }
+ }
+ /*
+ * nonexistent IRQs are edge default
+ */
+ return 0;
+}
+
+/* irq_vectors is indexed by the sum of all RTEs in all I/O APICs. */
+u8 irq_vector[NR_IRQ_VECTORS] __read_mostly;
+
+int assign_irq_vector(int irq)
+{
+ unsigned long flags;
+ int vector;
+ struct physdev_irq irq_op;
+
+ BUG_ON(irq != AUTO_ASSIGN && (unsigned)irq >= NR_IRQ_VECTORS);
+
+ if (irq < PIRQ_BASE || irq - PIRQ_BASE >= NR_PIRQS)
+ return -EINVAL;
+
+ spin_lock_irqsave(&vector_lock, flags);
+
+ if (irq != AUTO_ASSIGN && IO_APIC_VECTOR(irq) > 0) {
+ spin_unlock_irqrestore(&vector_lock, flags);
+ return IO_APIC_VECTOR(irq);
+ }
+
+ irq_op.irq = irq;
+ if (HYPERVISOR_physdev_op(PHYSDEVOP_alloc_irq_vector, &irq_op)) {
+ spin_unlock_irqrestore(&vector_lock, flags);
+ return -ENOSPC;
+ }
+
+ vector = irq_op.vector;
+ vector_irq[vector] = irq;
+ if (irq != AUTO_ASSIGN)
+ IO_APIC_VECTOR(irq) = vector;
+
+ spin_unlock_irqrestore(&vector_lock, flags);
+
+ return vector;
+}
+
+extern void (*interrupt[NR_IRQS])(void);
+#ifndef CONFIG_XEN
+static struct hw_interrupt_type ioapic_level_type;
+static struct hw_interrupt_type ioapic_edge_type;
+
+#define IOAPIC_AUTO -1
+#define IOAPIC_EDGE 0
+#define IOAPIC_LEVEL 1
+
+static void ioapic_register_intr(int irq, int vector, unsigned long trigger)
+{
+ unsigned idx;
+
+ idx = use_pci_vector() && !platform_legacy_irq(irq) ? vector : irq;
+
+ if ((trigger == IOAPIC_AUTO && IO_APIC_irq_trigger(irq)) ||
+ trigger == IOAPIC_LEVEL)
+ irq_desc[idx].chip = &ioapic_level_type;
+ else
+ irq_desc[idx].chip = &ioapic_edge_type;
+ set_intr_gate(vector, interrupt[idx]);
+}
+#else
+#define ioapic_register_intr(irq, vector, trigger) evtchn_register_pirq(irq)
+#endif /* !CONFIG_XEN */
+
+static void __init setup_IO_APIC_irqs(void)
+{
+ struct IO_APIC_route_entry entry;
+ int apic, pin, idx, irq, first_notcon = 1, vector;
+ unsigned long flags;
+
+ apic_printk(APIC_VERBOSE, KERN_DEBUG "init IO_APIC IRQs\n");
+
+ for (apic = 0; apic < nr_ioapics; apic++) {
+ for (pin = 0; pin < nr_ioapic_registers[apic]; pin++) {
+
+ /*
+ * add it to the IO-APIC irq-routing table:
+ */
+ memset(&entry,0,sizeof(entry));
+
+ entry.delivery_mode = INT_DELIVERY_MODE;
+ entry.dest_mode = INT_DEST_MODE;
+ entry.mask = 0; /* enable IRQ */
+ entry.dest.logical.logical_dest = cpu_mask_to_apicid(TARGET_CPUS);
+
+ idx = find_irq_entry(apic,pin,mp_INT);
+ if (idx == -1) {
+ if (first_notcon) {
+ apic_printk(APIC_VERBOSE, KERN_DEBUG " IO-APIC (apicid-pin) %d-%d", mp_ioapics[apic].mpc_apicid, pin);
+ first_notcon = 0;
+ } else
+ apic_printk(APIC_VERBOSE, ", %d-%d", mp_ioapics[apic].mpc_apicid, pin);
+ continue;
+ }
+
+ entry.trigger = irq_trigger(idx);
+ entry.polarity = irq_polarity(idx);
+
+ if (irq_trigger(idx)) {
+ entry.trigger = 1;
+ entry.mask = 1;
+ entry.dest.logical.logical_dest = cpu_mask_to_apicid(TARGET_CPUS);
+ }
+
+ irq = pin_2_irq(idx, apic, pin);
+ add_pin_to_irq(irq, apic, pin);
+
+ if (/* !apic && */ !IO_APIC_IRQ(irq))
+ continue;
+
+ if (IO_APIC_IRQ(irq)) {
+ vector = assign_irq_vector(irq);
+ entry.vector = vector;
+
+ ioapic_register_intr(irq, vector, IOAPIC_AUTO);
+ if (!apic && (irq < 16))
+ disable_8259A_irq(irq);
+ }
+ spin_lock_irqsave(&ioapic_lock, flags);
+ io_apic_write(apic, 0x11+2*pin, *(((int *)&entry)+1));
+ io_apic_write(apic, 0x10+2*pin, *(((int *)&entry)+0));
+ set_native_irq_info(irq, TARGET_CPUS);
+ spin_unlock_irqrestore(&ioapic_lock, flags);
+ }
+ }
+
+ if (!first_notcon)
+ apic_printk(APIC_VERBOSE," not connected.\n");
+}
+
+#ifndef CONFIG_XEN
+/*
+ * Set up the 8259A-master output pin as broadcast to all
+ * CPUs.
+ */
+static void __init setup_ExtINT_IRQ0_pin(unsigned int apic, unsigned int pin, int vector)
+{
+ struct IO_APIC_route_entry entry;
+ unsigned long flags;
+
+ memset(&entry,0,sizeof(entry));
+
+ disable_8259A_irq(0);
+
+ /* mask LVT0 */
+ apic_write(APIC_LVT0, APIC_LVT_MASKED | APIC_DM_EXTINT);
+
+ /*
+ * We use logical delivery to get the timer IRQ
+ * to the first CPU.
+ */
+ entry.dest_mode = INT_DEST_MODE;
+ entry.mask = 0; /* unmask IRQ now */
+ entry.dest.logical.logical_dest = cpu_mask_to_apicid(TARGET_CPUS);
+ entry.delivery_mode = INT_DELIVERY_MODE;
+ entry.polarity = 0;
+ entry.trigger = 0;
+ entry.vector = vector;
+
+ /*
+ * The timer IRQ doesn't have to know that behind the
+ * scene we have a 8259A-master in AEOI mode ...
+ */
+ irq_desc[0].chip = &ioapic_edge_type;
+
+ /*
+ * Add it to the IO-APIC irq-routing table:
+ */
+ spin_lock_irqsave(&ioapic_lock, flags);
+ io_apic_write(apic, 0x11+2*pin, *(((int *)&entry)+1));
+ io_apic_write(apic, 0x10+2*pin, *(((int *)&entry)+0));
+ spin_unlock_irqrestore(&ioapic_lock, flags);
+
+ enable_8259A_irq(0);
+}
+
+void __init UNEXPECTED_IO_APIC(void)
+{
+}
+
+void __apicdebuginit print_IO_APIC(void)
+{
+ int apic, i;
+ union IO_APIC_reg_00 reg_00;
+ union IO_APIC_reg_01 reg_01;
+ union IO_APIC_reg_02 reg_02;
+ unsigned long flags;
+
+ if (apic_verbosity == APIC_QUIET)
+ return;
+
+ printk(KERN_DEBUG "number of MP IRQ sources: %d.\n", mp_irq_entries);
+ for (i = 0; i < nr_ioapics; i++)
+ printk(KERN_DEBUG "number of IO-APIC #%d registers: %d.\n",
+ mp_ioapics[i].mpc_apicid, nr_ioapic_registers[i]);
+
+ /*
+ * We are a bit conservative about what we expect. We have to
+ * know about every hardware change ASAP.
+ */
+ printk(KERN_INFO "testing the IO APIC.......................\n");
+
+ for (apic = 0; apic < nr_ioapics; apic++) {
+
+ spin_lock_irqsave(&ioapic_lock, flags);
+ reg_00.raw = io_apic_read(apic, 0);
+ reg_01.raw = io_apic_read(apic, 1);
+ if (reg_01.bits.version >= 0x10)
+ reg_02.raw = io_apic_read(apic, 2);
+ spin_unlock_irqrestore(&ioapic_lock, flags);
+
+ printk("\n");
+ printk(KERN_DEBUG "IO APIC #%d......\n", mp_ioapics[apic].mpc_apicid);
+ printk(KERN_DEBUG ".... register #00: %08X\n", reg_00.raw);
+ printk(KERN_DEBUG "....... : physical APIC id: %02X\n", reg_00.bits.ID);
+ if (reg_00.bits.__reserved_1 || reg_00.bits.__reserved_2)
+ UNEXPECTED_IO_APIC();
+
+ printk(KERN_DEBUG ".... register #01: %08X\n", *(int *)&reg_01);
+ printk(KERN_DEBUG "....... : max redirection entries: %04X\n", reg_01.bits.entries);
+ if ( (reg_01.bits.entries != 0x0f) && /* older (Neptune) boards */
+ (reg_01.bits.entries != 0x17) && /* typical ISA+PCI boards */
+ (reg_01.bits.entries != 0x1b) && /* Compaq Proliant boards */
+ (reg_01.bits.entries != 0x1f) && /* dual Xeon boards */
+ (reg_01.bits.entries != 0x22) && /* bigger Xeon boards */
+ (reg_01.bits.entries != 0x2E) &&
+ (reg_01.bits.entries != 0x3F) &&
+ (reg_01.bits.entries != 0x03)
+ )
+ UNEXPECTED_IO_APIC();
+
+ printk(KERN_DEBUG "....... : PRQ implemented: %X\n", reg_01.bits.PRQ);
+ printk(KERN_DEBUG "....... : IO APIC version: %04X\n", reg_01.bits.version);
+ if ( (reg_01.bits.version != 0x01) && /* 82489DX IO-APICs */
+ (reg_01.bits.version != 0x02) && /* 82801BA IO-APICs (ICH2) */
+ (reg_01.bits.version != 0x10) && /* oldest IO-APICs */
+ (reg_01.bits.version != 0x11) && /* Pentium/Pro IO-APICs */
+ (reg_01.bits.version != 0x13) && /* Xeon IO-APICs */
+ (reg_01.bits.version != 0x20) /* Intel P64H (82806 AA) */
+ )
+ UNEXPECTED_IO_APIC();
+ if (reg_01.bits.__reserved_1 || reg_01.bits.__reserved_2)
+ UNEXPECTED_IO_APIC();
+
+ if (reg_01.bits.version >= 0x10) {
+ printk(KERN_DEBUG ".... register #02: %08X\n", reg_02.raw);
+ printk(KERN_DEBUG "....... : arbitration: %02X\n", reg_02.bits.arbitration);
+ if (reg_02.bits.__reserved_1 || reg_02.bits.__reserved_2)
+ UNEXPECTED_IO_APIC();
+ }
+
+ printk(KERN_DEBUG ".... IRQ redirection table:\n");
+
+ printk(KERN_DEBUG " NR Log Phy Mask Trig IRR Pol"
+ " Stat Dest Deli Vect: \n");
+
+ for (i = 0; i <= reg_01.bits.entries; i++) {
+ struct IO_APIC_route_entry entry;
+
+ spin_lock_irqsave(&ioapic_lock, flags);
+ *(((int *)&entry)+0) = io_apic_read(apic, 0x10+i*2);
+ *(((int *)&entry)+1) = io_apic_read(apic, 0x11+i*2);
+ spin_unlock_irqrestore(&ioapic_lock, flags);
+
+ printk(KERN_DEBUG " %02x %03X %02X ",
+ i,
+ entry.dest.logical.logical_dest,
+ entry.dest.physical.physical_dest
+ );
+
+ printk("%1d %1d %1d %1d %1d %1d %1d %02X\n",
+ entry.mask,
+ entry.trigger,
+ entry.irr,
+ entry.polarity,
+ entry.delivery_status,
+ entry.dest_mode,
+ entry.delivery_mode,
+ entry.vector
+ );
+ }
+ }
+ if (use_pci_vector())
+ printk(KERN_INFO "Using vector-based indexing\n");
+ printk(KERN_DEBUG "IRQ to pin mappings:\n");
+ for (i = 0; i < NR_IRQS; i++) {
+ struct irq_pin_list *entry = irq_2_pin + i;
+ if (entry->pin < 0)
+ continue;
+ if (use_pci_vector() && !platform_legacy_irq(i))
+ printk(KERN_DEBUG "IRQ%d ", IO_APIC_VECTOR(i));
+ else
+ printk(KERN_DEBUG "IRQ%d ", i);
+ for (;;) {
+ printk("-> %d:%d", entry->apic, entry->pin);
+ if (!entry->next)
+ break;
+ entry = irq_2_pin + entry->next;
+ }
+ printk("\n");
+ }
+
+ printk(KERN_INFO ".................................... done.\n");
+
+ return;
+}
+
+static __apicdebuginit void print_APIC_bitfield (int base)
+{
+ unsigned int v;
+ int i, j;
+
+ if (apic_verbosity == APIC_QUIET)
+ return;
+
+ printk(KERN_DEBUG "0123456789abcdef0123456789abcdef\n" KERN_DEBUG);
+ for (i = 0; i < 8; i++) {
+ v = apic_read(base + i*0x10);
+ for (j = 0; j < 32; j++) {
+ if (v & (1<<j))
+ printk("1");
+ else
+ printk("0");
+ }
+ printk("\n");
+ }
+}
+
+void __apicdebuginit print_local_APIC(void * dummy)
+{
+ unsigned int v, ver, maxlvt;
+
+ if (apic_verbosity == APIC_QUIET)
+ return;
+
+ printk("\n" KERN_DEBUG "printing local APIC contents on CPU#%d/%d:\n",
+ smp_processor_id(), hard_smp_processor_id());
+ v = apic_read(APIC_ID);
+ printk(KERN_INFO "... APIC ID: %08x (%01x)\n", v, GET_APIC_ID(v));
+ v = apic_read(APIC_LVR);
+ printk(KERN_INFO "... APIC VERSION: %08x\n", v);
+ ver = GET_APIC_VERSION(v);
+ maxlvt = get_maxlvt();
+
+ v = apic_read(APIC_TASKPRI);
+ printk(KERN_DEBUG "... APIC TASKPRI: %08x (%02x)\n", v, v & APIC_TPRI_MASK);
+
+ v = apic_read(APIC_ARBPRI);
+ printk(KERN_DEBUG "... APIC ARBPRI: %08x (%02x)\n", v,
+ v & APIC_ARBPRI_MASK);
+ v = apic_read(APIC_PROCPRI);
+ printk(KERN_DEBUG "... APIC PROCPRI: %08x\n", v);
+
+ v = apic_read(APIC_EOI);
+ printk(KERN_DEBUG "... APIC EOI: %08x\n", v);
+ v = apic_read(APIC_RRR);
+ printk(KERN_DEBUG "... APIC RRR: %08x\n", v);
+ v = apic_read(APIC_LDR);
+ printk(KERN_DEBUG "... APIC LDR: %08x\n", v);
+ v = apic_read(APIC_DFR);
+ printk(KERN_DEBUG "... APIC DFR: %08x\n", v);
+ v = apic_read(APIC_SPIV);
+ printk(KERN_DEBUG "... APIC SPIV: %08x\n", v);
+
+ printk(KERN_DEBUG "... APIC ISR field:\n");
+ print_APIC_bitfield(APIC_ISR);
+ printk(KERN_DEBUG "... APIC TMR field:\n");
+ print_APIC_bitfield(APIC_TMR);
+ printk(KERN_DEBUG "... APIC IRR field:\n");
+ print_APIC_bitfield(APIC_IRR);
+
+ v = apic_read(APIC_ESR);
+ printk(KERN_DEBUG "... APIC ESR: %08x\n", v);
+
+ v = apic_read(APIC_ICR);
+ printk(KERN_DEBUG "... APIC ICR: %08x\n", v);
+ v = apic_read(APIC_ICR2);
+ printk(KERN_DEBUG "... APIC ICR2: %08x\n", v);
+
+ v = apic_read(APIC_LVTT);
+ printk(KERN_DEBUG "... APIC LVTT: %08x\n", v);
+
+ if (maxlvt > 3) { /* PC is LVT#4. */
+ v = apic_read(APIC_LVTPC);
+ printk(KERN_DEBUG "... APIC LVTPC: %08x\n", v);
+ }
+ v = apic_read(APIC_LVT0);
+ printk(KERN_DEBUG "... APIC LVT0: %08x\n", v);
+ v = apic_read(APIC_LVT1);
+ printk(KERN_DEBUG "... APIC LVT1: %08x\n", v);
+
+ if (maxlvt > 2) { /* ERR is LVT#3. */
+ v = apic_read(APIC_LVTERR);
+ printk(KERN_DEBUG "... APIC LVTERR: %08x\n", v);
+ }
+
+ v = apic_read(APIC_TMICT);
+ printk(KERN_DEBUG "... APIC TMICT: %08x\n", v);
+ v = apic_read(APIC_TMCCT);
+ printk(KERN_DEBUG "... APIC TMCCT: %08x\n", v);
+ v = apic_read(APIC_TDCR);
+ printk(KERN_DEBUG "... APIC TDCR: %08x\n", v);
+ printk("\n");
+}
+
+void print_all_local_APICs (void)
+{
+ on_each_cpu(print_local_APIC, NULL, 1, 1);
+}
+
+void __apicdebuginit print_PIC(void)
+{
+ unsigned int v;
+ unsigned long flags;
+
+ if (apic_verbosity == APIC_QUIET)
+ return;
+
+ printk(KERN_DEBUG "\nprinting PIC contents\n");
+
+ spin_lock_irqsave(&i8259A_lock, flags);
+
+ v = inb(0xa1) << 8 | inb(0x21);
+ printk(KERN_DEBUG "... PIC IMR: %04x\n", v);
+
+ v = inb(0xa0) << 8 | inb(0x20);
+ printk(KERN_DEBUG "... PIC IRR: %04x\n", v);
+
+ outb(0x0b,0xa0);
+ outb(0x0b,0x20);
+ v = inb(0xa0) << 8 | inb(0x20);
+ outb(0x0a,0xa0);
+ outb(0x0a,0x20);
+
+ spin_unlock_irqrestore(&i8259A_lock, flags);
+
+ printk(KERN_DEBUG "... PIC ISR: %04x\n", v);
+
+ v = inb(0x4d1) << 8 | inb(0x4d0);
+ printk(KERN_DEBUG "... PIC ELCR: %04x\n", v);
+}
+#endif /* !CONFIG_XEN */
+
+static void __init enable_IO_APIC(void)
+{
+ union IO_APIC_reg_01 reg_01;
+#ifndef CONFIG_XEN
+ int i8259_apic, i8259_pin;
+#endif
+ int i, apic;
+ unsigned long flags;
+
+ for (i = 0; i < PIN_MAP_SIZE; i++) {
+ irq_2_pin[i].pin = -1;
+ irq_2_pin[i].next = 0;
+ }
+ if (!pirqs_enabled)
+ for (i = 0; i < MAX_PIRQS; i++)
+ pirq_entries[i] = -1;
+
+ /*
+ * The number of IO-APIC IRQ registers (== #pins):
+ */
+ for (apic = 0; apic < nr_ioapics; apic++) {
+ spin_lock_irqsave(&ioapic_lock, flags);
+ reg_01.raw = io_apic_read(apic, 1);
+ spin_unlock_irqrestore(&ioapic_lock, flags);
+ nr_ioapic_registers[apic] = reg_01.bits.entries+1;
+ }
+#ifndef CONFIG_XEN
+ for(apic = 0; apic < nr_ioapics; apic++) {
+ int pin;
+ /* See if any of the pins is in ExtINT mode */
+ for (pin = 0; pin < nr_ioapic_registers[apic]; pin++) {
+ struct IO_APIC_route_entry entry;
+ spin_lock_irqsave(&ioapic_lock, flags);
+ *(((int *)&entry) + 0) = io_apic_read(apic, 0x10 + 2 * pin);
+ *(((int *)&entry) + 1) = io_apic_read(apic, 0x11 + 2 * pin);
+ spin_unlock_irqrestore(&ioapic_lock, flags);
+
+
+ /* If the interrupt line is enabled and in ExtInt mode
+ * I have found the pin where the i8259 is connected.
+ */
+ if ((entry.mask == 0) && (entry.delivery_mode == dest_ExtINT)) {
+ ioapic_i8259.apic = apic;
+ ioapic_i8259.pin = pin;
+ goto found_i8259;
+ }
+ }
+ }
+ found_i8259:
+ /* Look to see what if the MP table has reported the ExtINT */
+ i8259_pin = find_isa_irq_pin(0, mp_ExtINT);
+ i8259_apic = find_isa_irq_apic(0, mp_ExtINT);
+ /* Trust the MP table if nothing is setup in the hardware */
+ if ((ioapic_i8259.pin == -1) && (i8259_pin >= 0)) {
+ printk(KERN_WARNING "ExtINT not setup in hardware but reported by MP table\n");
+ ioapic_i8259.pin = i8259_pin;
+ ioapic_i8259.apic = i8259_apic;
+ }
+ /* Complain if the MP table and the hardware disagree */
+ if (((ioapic_i8259.apic != i8259_apic) || (ioapic_i8259.pin != i8259_pin)) &&
+ (i8259_pin >= 0) && (ioapic_i8259.pin >= 0))
+ {
+ printk(KERN_WARNING "ExtINT in hardware and MP table differ\n");
+ }
+#endif
+
+ /*
+ * Do not trust the IO-APIC being empty at bootup
+ */
+ clear_IO_APIC();
+}
+
+/*
+ * Not an __init, needed by the reboot code
+ */
+void disable_IO_APIC(void)
+{
+ /*
+ * Clear the IO-APIC before rebooting:
+ */
+ clear_IO_APIC();
+
+#ifndef CONFIG_XEN
+ /*
+ * If the i8259 is routed through an IOAPIC
+ * Put that IOAPIC in virtual wire mode
+ * so legacy interrupts can be delivered.
+ */
+ if (ioapic_i8259.pin != -1) {
+ struct IO_APIC_route_entry entry;
+ unsigned long flags;
+
+ memset(&entry, 0, sizeof(entry));
+ entry.mask = 0; /* Enabled */
+ entry.trigger = 0; /* Edge */
+ entry.irr = 0;
+ entry.polarity = 0; /* High */
+ entry.delivery_status = 0;
+ entry.dest_mode = 0; /* Physical */
+ entry.delivery_mode = dest_ExtINT; /* ExtInt */
+ entry.vector = 0;
+ entry.dest.physical.physical_dest =
+ GET_APIC_ID(apic_read(APIC_ID));
+
+ /*
+ * Add it to the IO-APIC irq-routing table:
+ */
+ spin_lock_irqsave(&ioapic_lock, flags);
+ io_apic_write(ioapic_i8259.apic, 0x11+2*ioapic_i8259.pin,
+ *(((int *)&entry)+1));
+ io_apic_write(ioapic_i8259.apic, 0x10+2*ioapic_i8259.pin,
+ *(((int *)&entry)+0));
+ spin_unlock_irqrestore(&ioapic_lock, flags);
+ }
+
+ disconnect_bsp_APIC(ioapic_i8259.pin != -1);
+#endif
+}
+
+/*
+ * function to set the IO-APIC physical IDs based on the
+ * values stored in the MPC table.
+ *
+ * by Matt Domsch <Matt_Domsch@dell.com> Tue Dec 21 12:25:05 CST 1999
+ */
+
+#ifndef CONFIG_XEN
+static void __init setup_ioapic_ids_from_mpc (void)
+{
+ union IO_APIC_reg_00 reg_00;
+ int apic;
+ int i;
+ unsigned char old_id;
+ unsigned long flags;
+
+ /*
+ * Set the IOAPIC ID to the value stored in the MPC table.
+ */
+ for (apic = 0; apic < nr_ioapics; apic++) {
+
+ /* Read the register 0 value */
+ spin_lock_irqsave(&ioapic_lock, flags);
+ reg_00.raw = io_apic_read(apic, 0);
+ spin_unlock_irqrestore(&ioapic_lock, flags);
+
+ old_id = mp_ioapics[apic].mpc_apicid;
+
+
+ printk(KERN_INFO "Using IO-APIC %d\n", mp_ioapics[apic].mpc_apicid);
+
+
+ /*
+ * We need to adjust the IRQ routing table
+ * if the ID changed.
+ */
+ if (old_id != mp_ioapics[apic].mpc_apicid)
+ for (i = 0; i < mp_irq_entries; i++)
+ if (mp_irqs[i].mpc_dstapic == old_id)
+ mp_irqs[i].mpc_dstapic
+ = mp_ioapics[apic].mpc_apicid;
+
+ /*
+ * Read the right value from the MPC table and
+ * write it into the ID register.
+ */
+ apic_printk(APIC_VERBOSE,KERN_INFO "...changing IO-APIC physical APIC ID to %d ...",
+ mp_ioapics[apic].mpc_apicid);
+
+ reg_00.bits.ID = mp_ioapics[apic].mpc_apicid;
+ spin_lock_irqsave(&ioapic_lock, flags);
+ io_apic_write(apic, 0, reg_00.raw);
+ spin_unlock_irqrestore(&ioapic_lock, flags);
+
+ /*
+ * Sanity check
+ */
+ spin_lock_irqsave(&ioapic_lock, flags);
+ reg_00.raw = io_apic_read(apic, 0);
+ spin_unlock_irqrestore(&ioapic_lock, flags);
+ if (reg_00.bits.ID != mp_ioapics[apic].mpc_apicid)
+ printk("could not set ID!\n");
+ else
+ apic_printk(APIC_VERBOSE," ok.\n");
+ }
+}
+#else
+static void __init setup_ioapic_ids_from_mpc(void) { }
+#endif
+
+/*
+ * There is a nasty bug in some older SMP boards, their mptable lies
+ * about the timer IRQ. We do the following to work around the situation:
+ *
+ * - timer IRQ defaults to IO-APIC IRQ
+ * - if this function detects that timer IRQs are defunct, then we fall
+ * back to ISA timer IRQs
+ */
+#ifndef CONFIG_XEN
+static int __init timer_irq_works(void)
+{
+ unsigned long t1 = jiffies;
+
+ local_irq_enable();
+ /* Let ten ticks pass... */
+ mdelay((10 * 1000) / HZ);
+
+ /*
+ * Expect a few ticks at least, to be sure some possible
+ * glue logic does not lock up after one or two first
+ * ticks in a non-ExtINT mode. Also the local APIC
+ * might have cached one ExtINT interrupt. Finally, at
+ * least one tick may be lost due to delays.
+ */
+
+ /* jiffies wrap? */
+ if (jiffies - t1 > 4)
+ return 1;
+ return 0;
+}
+
+/*
+ * In the SMP+IOAPIC case it might happen that there are an unspecified
+ * number of pending IRQ events unhandled. These cases are very rare,
+ * so we 'resend' these IRQs via IPIs, to the same CPU. It's much
+ * better to do it this way as thus we do not have to be aware of
+ * 'pending' interrupts in the IRQ path, except at this point.
+ */
+/*
+ * Edge triggered needs to resend any interrupt
+ * that was delayed but this is now handled in the device
+ * independent code.
+ */
+
+/*
+ * Starting up a edge-triggered IO-APIC interrupt is
+ * nasty - we need to make sure that we get the edge.
+ * If it is already asserted for some reason, we need
+ * return 1 to indicate that is was pending.
+ *
+ * This is not complete - we should be able to fake
+ * an edge even if it isn't on the 8259A...
+ */
+
+static unsigned int startup_edge_ioapic_irq(unsigned int irq)
+{
+ int was_pending = 0;
+ unsigned long flags;
+
+ spin_lock_irqsave(&ioapic_lock, flags);
+ if (irq < 16) {
+ disable_8259A_irq(irq);
+ if (i8259A_irq_pending(irq))
+ was_pending = 1;
+ }
+ __unmask_IO_APIC_irq(irq);
+ spin_unlock_irqrestore(&ioapic_lock, flags);
+
+ return was_pending;
+}
+
+/*
+ * Once we have recorded IRQ_PENDING already, we can mask the
+ * interrupt for real. This prevents IRQ storms from unhandled
+ * devices.
+ */
+static void ack_edge_ioapic_irq(unsigned int irq)
+{
+ move_irq(irq);
+ if ((irq_desc[irq].status & (IRQ_PENDING | IRQ_DISABLED))
+ == (IRQ_PENDING | IRQ_DISABLED))
+ mask_IO_APIC_irq(irq);
+ ack_APIC_irq();
+}
+
+/*
+ * Level triggered interrupts can just be masked,
+ * and shutting down and starting up the interrupt
+ * is the same as enabling and disabling them -- except
+ * with a startup need to return a "was pending" value.
+ *
+ * Level triggered interrupts are special because we
+ * do not touch any IO-APIC register while handling
+ * them. We ack the APIC in the end-IRQ handler, not
+ * in the start-IRQ-handler. Protection against reentrance
+ * from the same interrupt is still provided, both by the
+ * generic IRQ layer and by the fact that an unacked local
+ * APIC does not accept IRQs.
+ */
+static unsigned int startup_level_ioapic_irq (unsigned int irq)
+{
+ unmask_IO_APIC_irq(irq);
+
+ return 0; /* don't check for pending */
+}
+
+static void end_level_ioapic_irq (unsigned int irq)
+{
+ move_irq(irq);
+ ack_APIC_irq();
+}
+
+#ifdef CONFIG_PCI_MSI
+static unsigned int startup_edge_ioapic_vector(unsigned int vector)
+{
+ int irq = vector_to_irq(vector);
+
+ return startup_edge_ioapic_irq(irq);
+}
+
+static void ack_edge_ioapic_vector(unsigned int vector)
+{
+ int irq = vector_to_irq(vector);
+
+ move_native_irq(vector);
+ ack_edge_ioapic_irq(irq);
+}
+
+static unsigned int startup_level_ioapic_vector (unsigned int vector)
+{
+ int irq = vector_to_irq(vector);
+
+ return startup_level_ioapic_irq (irq);
+}
+
+static void end_level_ioapic_vector (unsigned int vector)
+{
+ int irq = vector_to_irq(vector);
+
+ move_native_irq(vector);
+ end_level_ioapic_irq(irq);
+}
+
+static void mask_IO_APIC_vector (unsigned int vector)
+{
+ int irq = vector_to_irq(vector);
+
+ mask_IO_APIC_irq(irq);
+}
+
+static void unmask_IO_APIC_vector (unsigned int vector)
+{
+ int irq = vector_to_irq(vector);
+
+ unmask_IO_APIC_irq(irq);
+}
+
+#ifdef CONFIG_SMP
+static void set_ioapic_affinity_vector (unsigned int vector,
+ cpumask_t cpu_mask)
+{
+ int irq = vector_to_irq(vector);
+
+ set_native_irq_info(vector, cpu_mask);
+ set_ioapic_affinity_irq(irq, cpu_mask);
+}
+#endif // CONFIG_SMP
+#endif // CONFIG_PCI_MSI
+
+static int ioapic_retrigger(unsigned int irq)
+{
+ send_IPI_self(IO_APIC_VECTOR(irq));
+
+ return 1;
+}
+
+/*
+ * Level and edge triggered IO-APIC interrupts need different handling,
+ * so we use two separate IRQ descriptors. Edge triggered IRQs can be
+ * handled with the level-triggered descriptor, but that one has slightly
+ * more overhead. Level-triggered interrupts cannot be handled with the
+ * edge-triggered handler, without risking IRQ storms and other ugly
+ * races.
+ */
+
+static struct hw_interrupt_type ioapic_edge_type __read_mostly = {
+ .typename = "IO-APIC-edge",
+ .startup = startup_edge_ioapic,
+ .shutdown = shutdown_edge_ioapic,
+ .enable = enable_edge_ioapic,
+ .disable = disable_edge_ioapic,
+ .ack = ack_edge_ioapic,
+ .end = end_edge_ioapic,
+#ifdef CONFIG_SMP
+ .set_affinity = set_ioapic_affinity,
+#endif
+ .retrigger = ioapic_retrigger,
+};
+
+static struct hw_interrupt_type ioapic_level_type __read_mostly = {
+ .typename = "IO-APIC-level",
+ .startup = startup_level_ioapic,
+ .shutdown = shutdown_level_ioapic,
+ .enable = enable_level_ioapic,
+ .disable = disable_level_ioapic,
+ .ack = mask_and_ack_level_ioapic,
+ .end = end_level_ioapic,
+#ifdef CONFIG_SMP
+ .set_affinity = set_ioapic_affinity,
+#endif
+ .retrigger = ioapic_retrigger,
+};
+#endif /* !CONFIG_XEN */
+
+static inline void init_IO_APIC_traps(void)
+{
+ int irq;
+
+ /*
+ * NOTE! The local APIC isn't very good at handling
+ * multiple interrupts at the same interrupt level.
+ * As the interrupt level is determined by taking the
+ * vector number and shifting that right by 4, we
+ * want to spread these out a bit so that they don't
+ * all fall in the same interrupt level.
+ *
+ * Also, we've got to be careful not to trash gate
+ * 0x80, because int 0x80 is hm, kind of importantish. ;)
+ */
+ for (irq = 0; irq < NR_IRQS ; irq++) {
+ int tmp = irq;
+ if (use_pci_vector()) {
+ if (!platform_legacy_irq(tmp))
+ if ((tmp = vector_to_irq(tmp)) == -1)
+ continue;
+ }
+ if (IO_APIC_IRQ(tmp) && !IO_APIC_VECTOR(tmp)) {
+ /*
+ * Hmm.. We don't have an entry for this,
+ * so default to an old-fashioned 8259
+ * interrupt if we can..
+ */
+ if (irq < 16)
+ make_8259A_irq(irq);
+#ifndef CONFIG_XEN
+ else
+ /* Strange. Oh, well.. */
+ irq_desc[irq].chip = &no_irq_type;
+#endif
+ }
+ }
+}
+
+#ifndef CONFIG_XEN
+static void enable_lapic_irq (unsigned int irq)
+{
+ unsigned long v;
+
+ v = apic_read(APIC_LVT0);
+ apic_write(APIC_LVT0, v & ~APIC_LVT_MASKED);
+}
+
+static void disable_lapic_irq (unsigned int irq)
+{
+ unsigned long v;
+
+ v = apic_read(APIC_LVT0);
+ apic_write(APIC_LVT0, v | APIC_LVT_MASKED);
+}
+
+static void ack_lapic_irq (unsigned int irq)
+{
+ ack_APIC_irq();
+}
+
+static void end_lapic_irq (unsigned int i) { /* nothing */ }
+
+static struct hw_interrupt_type lapic_irq_type __read_mostly = {
+ .typename = "local-APIC-edge",
+ .startup = NULL, /* startup_irq() not used for IRQ0 */
+ .shutdown = NULL, /* shutdown_irq() not used for IRQ0 */
+ .enable = enable_lapic_irq,
+ .disable = disable_lapic_irq,
+ .ack = ack_lapic_irq,
+ .end = end_lapic_irq,
+};
+
+static void setup_nmi (void)
+{
+ /*
+ * Dirty trick to enable the NMI watchdog ...
+ * We put the 8259A master into AEOI mode and
+ * unmask on all local APICs LVT0 as NMI.
+ *
+ * The idea to use the 8259A in AEOI mode ('8259A Virtual Wire')
+ * is from Maciej W. Rozycki - so we do not have to EOI from
+ * the NMI handler or the timer interrupt.
+ */
+ printk(KERN_INFO "activating NMI Watchdog ...");
+
+ enable_NMI_through_LVT0(NULL);
+
+ printk(" done.\n");
+}
+
+/*
+ * This looks a bit hackish but it's about the only one way of sending
+ * a few INTA cycles to 8259As and any associated glue logic. ICR does
+ * not support the ExtINT mode, unfortunately. We need to send these
+ * cycles as some i82489DX-based boards have glue logic that keeps the
+ * 8259A interrupt line asserted until INTA. --macro
+ */
+static inline void unlock_ExtINT_logic(void)
+{
+ int apic, pin, i;
+ struct IO_APIC_route_entry entry0, entry1;
+ unsigned char save_control, save_freq_select;
+ unsigned long flags;
+
+ pin = find_isa_irq_pin(8, mp_INT);
+ apic = find_isa_irq_apic(8, mp_INT);
+ if (pin == -1)
+ return;
+
+ spin_lock_irqsave(&ioapic_lock, flags);
+ *(((int *)&entry0) + 1) = io_apic_read(apic, 0x11 + 2 * pin);
+ *(((int *)&entry0) + 0) = io_apic_read(apic, 0x10 + 2 * pin);
+ spin_unlock_irqrestore(&ioapic_lock, flags);
+ clear_IO_APIC_pin(apic, pin);
+
+ memset(&entry1, 0, sizeof(entry1));
+
+ entry1.dest_mode = 0; /* physical delivery */
+ entry1.mask = 0; /* unmask IRQ now */
+ entry1.dest.physical.physical_dest = hard_smp_processor_id();
+ entry1.delivery_mode = dest_ExtINT;
+ entry1.polarity = entry0.polarity;
+ entry1.trigger = 0;
+ entry1.vector = 0;
+
+ spin_lock_irqsave(&ioapic_lock, flags);
+ io_apic_write(apic, 0x11 + 2 * pin, *(((int *)&entry1) + 1));
+ io_apic_write(apic, 0x10 + 2 * pin, *(((int *)&entry1) + 0));
+ spin_unlock_irqrestore(&ioapic_lock, flags);
+
+ save_control = CMOS_READ(RTC_CONTROL);
+ save_freq_select = CMOS_READ(RTC_FREQ_SELECT);
+ CMOS_WRITE((save_freq_select & ~RTC_RATE_SELECT) | 0x6,
+ RTC_FREQ_SELECT);
+ CMOS_WRITE(save_control | RTC_PIE, RTC_CONTROL);
+
+ i = 100;
+ while (i-- > 0) {
+ mdelay(10);
+ if ((CMOS_READ(RTC_INTR_FLAGS) & RTC_PF) == RTC_PF)
+ i -= 10;
+ }
+
+ CMOS_WRITE(save_control, RTC_CONTROL);
+ CMOS_WRITE(save_freq_select, RTC_FREQ_SELECT);
+ clear_IO_APIC_pin(apic, pin);
+
+ spin_lock_irqsave(&ioapic_lock, flags);
+ io_apic_write(apic, 0x11 + 2 * pin, *(((int *)&entry0) + 1));
+ io_apic_write(apic, 0x10 + 2 * pin, *(((int *)&entry0) + 0));
+ spin_unlock_irqrestore(&ioapic_lock, flags);
+}
+
+int timer_uses_ioapic_pin_0;
+
+/*
+ * This code may look a bit paranoid, but it's supposed to cooperate with
+ * a wide range of boards and BIOS bugs. Fortunately only the timer IRQ
+ * is so screwy. Thanks to Brian Perkins for testing/hacking this beast
+ * fanatically on his truly buggy board.
+ *
+ * FIXME: really need to revamp this for modern platforms only.
+ */
+static inline void check_timer(void)
+{
+ int apic1, pin1, apic2, pin2;
+ int vector;
+
+ /*
+ * get/set the timer IRQ vector:
+ */
+ disable_8259A_irq(0);
+ vector = assign_irq_vector(0);
+ set_intr_gate(vector, interrupt[0]);
+
+ /*
+ * Subtle, code in do_timer_interrupt() expects an AEOI
+ * mode for the 8259A whenever interrupts are routed
+ * through I/O APICs. Also IRQ0 has to be enabled in
+ * the 8259A which implies the virtual wire has to be
+ * disabled in the local APIC.
+ */
+ apic_write(APIC_LVT0, APIC_LVT_MASKED | APIC_DM_EXTINT);
+ init_8259A(1);
+ if (timer_over_8254 > 0)
+ enable_8259A_irq(0);
+
+ pin1 = find_isa_irq_pin(0, mp_INT);
+ apic1 = find_isa_irq_apic(0, mp_INT);
+ pin2 = ioapic_i8259.pin;
+ apic2 = ioapic_i8259.apic;
+
+ if (pin1 == 0)
+ timer_uses_ioapic_pin_0 = 1;
+
+ apic_printk(APIC_VERBOSE,KERN_INFO "..TIMER: vector=0x%02X apic1=%d pin1=%d apic2=%d pin2=%d\n",
+ vector, apic1, pin1, apic2, pin2);
+
+ if (pin1 != -1) {
+ /*
+ * Ok, does IRQ0 through the IOAPIC work?
+ */
+ unmask_IO_APIC_irq(0);
+ if (!no_timer_check && timer_irq_works()) {
+ nmi_watchdog_default();
+ if (nmi_watchdog == NMI_IO_APIC) {
+ disable_8259A_irq(0);
+ setup_nmi();
+ enable_8259A_irq(0);
+ }
+ if (disable_timer_pin_1 > 0)
+ clear_IO_APIC_pin(0, pin1);
+ return;
+ }
+ clear_IO_APIC_pin(apic1, pin1);
+ apic_printk(APIC_QUIET,KERN_ERR "..MP-BIOS bug: 8254 timer not "
+ "connected to IO-APIC\n");
+ }
+
+ apic_printk(APIC_VERBOSE,KERN_INFO "...trying to set up timer (IRQ0) "
+ "through the 8259A ... ");
+ if (pin2 != -1) {
+ apic_printk(APIC_VERBOSE,"\n..... (found apic %d pin %d) ...",
+ apic2, pin2);
+ /*
+ * legacy devices should be connected to IO APIC #0
+ */
+ setup_ExtINT_IRQ0_pin(apic2, pin2, vector);
+ if (timer_irq_works()) {
+ apic_printk(APIC_VERBOSE," works.\n");
+ nmi_watchdog_default();
+ if (nmi_watchdog == NMI_IO_APIC) {
+ setup_nmi();
+ }
+ return;
+ }
+ /*
+ * Cleanup, just in case ...
+ */
+ clear_IO_APIC_pin(apic2, pin2);
+ }
+ apic_printk(APIC_VERBOSE," failed.\n");
+
+ if (nmi_watchdog == NMI_IO_APIC) {
+ printk(KERN_WARNING "timer doesn't work through the IO-APIC - disabling NMI Watchdog!\n");
+ nmi_watchdog = 0;
+ }
+
+ apic_printk(APIC_VERBOSE, KERN_INFO "...trying to set up timer as Virtual Wire IRQ...");
+
+ disable_8259A_irq(0);
+ irq_desc[0].chip = &lapic_irq_type;
+ apic_write(APIC_LVT0, APIC_DM_FIXED | vector); /* Fixed mode */
+ enable_8259A_irq(0);
+
+ if (timer_irq_works()) {
+ apic_printk(APIC_VERBOSE," works.\n");
+ return;
+ }
+ apic_write(APIC_LVT0, APIC_LVT_MASKED | APIC_DM_FIXED | vector);
+ apic_printk(APIC_VERBOSE," failed.\n");
+
+ apic_printk(APIC_VERBOSE, KERN_INFO "...trying to set up timer as ExtINT IRQ...");
+
+ init_8259A(0);
+ make_8259A_irq(0);
+ apic_write(APIC_LVT0, APIC_DM_EXTINT);
+
+ unlock_ExtINT_logic();
+
+ if (timer_irq_works()) {
+ apic_printk(APIC_VERBOSE," works.\n");
+ return;
+ }
+ apic_printk(APIC_VERBOSE," failed :(.\n");
+ panic("IO-APIC + timer doesn't work! Try using the 'noapic' kernel parameter\n");
+}
+#else
+#define check_timer() ((void)0)
+int timer_uses_ioapic_pin_0 = 0;
+#endif /* !CONFIG_XEN */
+
+static int __init notimercheck(char *s)
+{
+ no_timer_check = 1;
+ return 1;
+}
+__setup("no_timer_check", notimercheck);
+
+/*
+ *
+ * IRQ's that are handled by the PIC in the MPS IOAPIC case.
+ * - IRQ2 is the cascade IRQ, and cannot be a io-apic IRQ.
+ * Linux doesn't really care, as it's not actually used
+ * for any interrupt handling anyway.
+ */
+#define PIC_IRQS (1<<2)
+
+void __init setup_IO_APIC(void)
+{
+ enable_IO_APIC();
+
+ if (acpi_ioapic)
+ io_apic_irqs = ~0; /* all IRQs go through IOAPIC */
+ else
+ io_apic_irqs = ~PIC_IRQS;
+
+ apic_printk(APIC_VERBOSE, "ENABLING IO-APIC IRQs\n");
+
+ /*
+ * Set up the IO-APIC IRQ routing table.
+ */
+ if (!acpi_ioapic)
+ setup_ioapic_ids_from_mpc();
+#ifndef CONFIG_XEN
+ sync_Arb_IDs();
+#endif /* !CONFIG_XEN */
+ setup_IO_APIC_irqs();
+ init_IO_APIC_traps();
+ check_timer();
+ if (!acpi_ioapic)
+ print_IO_APIC();
+}
+
+#ifndef CONFIG_XEN
+
+struct sysfs_ioapic_data {
+ struct sys_device dev;
+ struct IO_APIC_route_entry entry[0];
+};
+static struct sysfs_ioapic_data * mp_ioapic_data[MAX_IO_APICS];
+
+static int ioapic_suspend(struct sys_device *dev, pm_message_t state)
+{
+ struct IO_APIC_route_entry *entry;
+ struct sysfs_ioapic_data *data;
+ unsigned long flags;
+ int i;
+
+ data = container_of(dev, struct sysfs_ioapic_data, dev);
+ entry = data->entry;
+ spin_lock_irqsave(&ioapic_lock, flags);
+ for (i = 0; i < nr_ioapic_registers[dev->id]; i ++, entry ++ ) {
+ *(((int *)entry) + 1) = io_apic_read(dev->id, 0x11 + 2 * i);
+ *(((int *)entry) + 0) = io_apic_read(dev->id, 0x10 + 2 * i);
+ }
+ spin_unlock_irqrestore(&ioapic_lock, flags);
+
+ return 0;
+}
+
+static int ioapic_resume(struct sys_device *dev)
+{
+ struct IO_APIC_route_entry *entry;
+ struct sysfs_ioapic_data *data;
+ unsigned long flags;
+ union IO_APIC_reg_00 reg_00;
+ int i;
+
+ data = container_of(dev, struct sysfs_ioapic_data, dev);
+ entry = data->entry;
+
+ spin_lock_irqsave(&ioapic_lock, flags);
+ reg_00.raw = io_apic_read(dev->id, 0);
+ if (reg_00.bits.ID != mp_ioapics[dev->id].mpc_apicid) {
+ reg_00.bits.ID = mp_ioapics[dev->id].mpc_apicid;
+ io_apic_write(dev->id, 0, reg_00.raw);
+ }
+ for (i = 0; i < nr_ioapic_registers[dev->id]; i ++, entry ++ ) {
+ io_apic_write(dev->id, 0x11+2*i, *(((int *)entry)+1));
+ io_apic_write(dev->id, 0x10+2*i, *(((int *)entry)+0));
+ }
+ spin_unlock_irqrestore(&ioapic_lock, flags);
+
+ return 0;
+}
+
+static struct sysdev_class ioapic_sysdev_class = {
+ set_kset_name("ioapic"),
+ .suspend = ioapic_suspend,
+ .resume = ioapic_resume,
+};
+
+static int __init ioapic_init_sysfs(void)
+{
+ struct sys_device * dev;
+ int i, size, error = 0;
+
+ error = sysdev_class_register(&ioapic_sysdev_class);
+ if (error)
+ return error;
+
+ for (i = 0; i < nr_ioapics; i++ ) {
+ size = sizeof(struct sys_device) + nr_ioapic_registers[i]
+ * sizeof(struct IO_APIC_route_entry);
+ mp_ioapic_data[i] = kmalloc(size, GFP_KERNEL);
+ if (!mp_ioapic_data[i]) {
+ printk(KERN_ERR "Can't suspend/resume IOAPIC %d\n", i);
+ continue;
+ }
+ memset(mp_ioapic_data[i], 0, size);
+ dev = &mp_ioapic_data[i]->dev;
+ dev->id = i;
+ dev->cls = &ioapic_sysdev_class;
+ error = sysdev_register(dev);
+ if (error) {
+ kfree(mp_ioapic_data[i]);
+ mp_ioapic_data[i] = NULL;
+ printk(KERN_ERR "Can't suspend/resume IOAPIC %d\n", i);
+ continue;
+ }
+ }
+
+ return 0;
+}
+
+device_initcall(ioapic_init_sysfs);
+
+#endif /* CONFIG_XEN */
+
+/* --------------------------------------------------------------------------
+ ACPI-based IOAPIC Configuration
+ -------------------------------------------------------------------------- */
+
+#ifdef CONFIG_ACPI
+
+#define IO_APIC_MAX_ID 0xFE
+
+int __init io_apic_get_version (int ioapic)
+{
+ union IO_APIC_reg_01 reg_01;
+ unsigned long flags;
+
+ spin_lock_irqsave(&ioapic_lock, flags);
+ reg_01.raw = io_apic_read(ioapic, 1);
+ spin_unlock_irqrestore(&ioapic_lock, flags);
+
+ return reg_01.bits.version;
+}
+
+
+int __init io_apic_get_redir_entries (int ioapic)
+{
+ union IO_APIC_reg_01 reg_01;
+ unsigned long flags;
+
+ spin_lock_irqsave(&ioapic_lock, flags);
+ reg_01.raw = io_apic_read(ioapic, 1);
+ spin_unlock_irqrestore(&ioapic_lock, flags);
+
+ return reg_01.bits.entries;
+}
+
+
+int io_apic_set_pci_routing (int ioapic, int pin, int irq, int edge_level, int active_high_low)
+{
+ struct IO_APIC_route_entry entry;
+ unsigned long flags;
+
+ if (!IO_APIC_IRQ(irq)) {
+ apic_printk(APIC_QUIET,KERN_ERR "IOAPIC[%d]: Invalid reference to IRQ 0\n",
+ ioapic);
+ return -EINVAL;
+ }
+
+ /*
+ * Generate a PCI IRQ routing entry and program the IOAPIC accordingly.
+ * Note that we mask (disable) IRQs now -- these get enabled when the
+ * corresponding device driver registers for this IRQ.
+ */
+
+ memset(&entry,0,sizeof(entry));
+
+ entry.delivery_mode = INT_DELIVERY_MODE;
+ entry.dest_mode = INT_DEST_MODE;
+ entry.dest.logical.logical_dest = cpu_mask_to_apicid(TARGET_CPUS);
+ entry.trigger = edge_level;
+ entry.polarity = active_high_low;
+ entry.mask = 1; /* Disabled (masked) */
+
+ irq = gsi_irq_sharing(irq);
+ /*
+ * IRQs < 16 are already in the irq_2_pin[] map
+ */
+ if (irq >= 16)
+ add_pin_to_irq(irq, ioapic, pin);
+
+ entry.vector = assign_irq_vector(irq);
+
+ apic_printk(APIC_VERBOSE,KERN_DEBUG "IOAPIC[%d]: Set PCI routing entry (%d-%d -> 0x%x -> "
+ "IRQ %d Mode:%i Active:%i)\n", ioapic,
+ mp_ioapics[ioapic].mpc_apicid, pin, entry.vector, irq,
+ edge_level, active_high_low);
+
+ ioapic_register_intr(irq, entry.vector, edge_level);
+
+ if (!ioapic && (irq < 16))
+ disable_8259A_irq(irq);
+
+ spin_lock_irqsave(&ioapic_lock, flags);
+ io_apic_write(ioapic, 0x11+2*pin, *(((int *)&entry)+1));
+ io_apic_write(ioapic, 0x10+2*pin, *(((int *)&entry)+0));
+ set_native_irq_info(use_pci_vector() ? entry.vector : irq, TARGET_CPUS);
+ spin_unlock_irqrestore(&ioapic_lock, flags);
+
+ return 0;
+}
+
+#endif /* CONFIG_ACPI */
+
+
+#ifndef CONFIG_XEN
+/*
+ * This function currently is only a helper for the i386 smp boot process where
+ * we need to reprogram the ioredtbls to cater for the cpus which have come online
+ * so mask in all cases should simply be TARGET_CPUS
+ */
+#ifdef CONFIG_SMP
+void __init setup_ioapic_dest(void)
+{
+ int pin, ioapic, irq, irq_entry;
+
+ if (skip_ioapic_setup == 1)
+ return;
+
+ for (ioapic = 0; ioapic < nr_ioapics; ioapic++) {
+ for (pin = 0; pin < nr_ioapic_registers[ioapic]; pin++) {
+ irq_entry = find_irq_entry(ioapic, pin, mp_INT);
+ if (irq_entry == -1)
+ continue;
+ irq = pin_2_irq(irq_entry, ioapic, pin);
+ set_ioapic_affinity_irq(irq, TARGET_CPUS);
+ }
+
+ }
+}
+#endif
+#endif /* !CONFIG_XEN */
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/kernel/ioport_64-xen.c 2008-01-28 12:24:19.000000000 +0100
@@ -0,0 +1,100 @@
+/*
+ * linux/arch/x86_64/kernel/ioport.c
+ *
+ * This contains the io-permission bitmap code - written by obz, with changes
+ * by Linus.
+ */
+
+#include <linux/sched.h>
+#include <linux/kernel.h>
+#include <linux/capability.h>
+#include <linux/errno.h>
+#include <linux/types.h>
+#include <linux/ioport.h>
+#include <linux/mm.h>
+#include <linux/smp.h>
+#include <linux/smp_lock.h>
+#include <linux/stddef.h>
+#include <linux/slab.h>
+#include <linux/thread_info.h>
+#include <xen/interface/physdev.h>
+
+/* Set EXTENT bits starting at BASE in BITMAP to value TURN_ON. */
+static void set_bitmap(unsigned long *bitmap, unsigned int base, unsigned int extent, int new_value)
+{
+ int i;
+
+ if (new_value)
+ for (i = base; i < base + extent; i++)
+ __set_bit(i, bitmap);
+ else
+ for (i = base; i < base + extent; i++)
+ clear_bit(i, bitmap);
+}
+
+/*
+ * this changes the io permissions bitmap in the current task.
+ */
+asmlinkage long sys_ioperm(unsigned long from, unsigned long num, int turn_on)
+{
+ struct thread_struct * t = &current->thread;
+ unsigned long *bitmap;
+ struct physdev_set_iobitmap set_iobitmap;
+
+ if ((from + num <= from) || (from + num > IO_BITMAP_BITS))
+ return -EINVAL;
+ if (turn_on && !capable(CAP_SYS_RAWIO))
+ return -EPERM;
+
+ /*
+ * If it's the first ioperm() call in this thread's lifetime, set the
+ * IO bitmap up. ioperm() is much less timing critical than clone(),
+ * this is why we delay this operation until now:
+ */
+ if (!t->io_bitmap_ptr) {
+ bitmap = kmalloc(IO_BITMAP_BYTES, GFP_KERNEL);
+ if (!bitmap)
+ return -ENOMEM;
+
+ memset(bitmap, 0xff, IO_BITMAP_BYTES);
+ t->io_bitmap_ptr = bitmap;
+
+ set_xen_guest_handle(set_iobitmap.bitmap, (char *)bitmap);
+ set_iobitmap.nr_ports = IO_BITMAP_BITS;
+ WARN_ON(HYPERVISOR_physdev_op(PHYSDEVOP_set_iobitmap,
+ &set_iobitmap));
+ }
+
+ set_bitmap(t->io_bitmap_ptr, from, num, !turn_on);
+
+ return 0;
+}
+
+/*
+ * sys_iopl has to be used when you want to access the IO ports
+ * beyond the 0x3ff range: to get the full 65536 ports bitmapped
+ * you'd need 8kB of bitmaps/process, which is a bit excessive.
+ *
+ */
+
+asmlinkage long sys_iopl(unsigned int new_iopl, struct pt_regs *regs)
+{
+ unsigned int old_iopl = current->thread.iopl;
+ struct physdev_set_iopl set_iopl;
+
+ if (new_iopl > 3)
+ return -EINVAL;
+
+ /* Need "raw I/O" privileges for direct port access. */
+ if ((new_iopl > old_iopl) && !capable(CAP_SYS_RAWIO))
+ return -EPERM;
+
+ /* Change our version of the privilege levels. */
+ current->thread.iopl = new_iopl;
+
+ /* Force the change at ring 0. */
+ set_iopl.iopl = (new_iopl == 0) ? 1 : new_iopl;
+ WARN_ON(HYPERVISOR_physdev_op(PHYSDEVOP_set_iopl, &set_iopl));
+
+ return 0;
+}
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/kernel/ldt_64-xen.c 2007-06-12 13:13:01.000000000 +0200
@@ -0,0 +1,282 @@
+/*
+ * linux/arch/x86_64/kernel/ldt.c
+ *
+ * Copyright (C) 1992 Krishna Balasubramanian and Linus Torvalds
+ * Copyright (C) 1999 Ingo Molnar <mingo@redhat.com>
+ * Copyright (C) 2002 Andi Kleen
+ *
+ * This handles calls from both 32bit and 64bit mode.
+ */
+
+#include <linux/errno.h>
+#include <linux/sched.h>
+#include <linux/string.h>
+#include <linux/mm.h>
+#include <linux/smp.h>
+#include <linux/smp_lock.h>
+#include <linux/vmalloc.h>
+#include <linux/slab.h>
+
+#include <asm/uaccess.h>
+#include <asm/system.h>
+#include <asm/ldt.h>
+#include <asm/desc.h>
+#include <asm/proto.h>
+#include <asm/pgalloc.h>
+
+#ifdef CONFIG_SMP /* avoids "defined but not used" warnig */
+static void flush_ldt(void *null)
+{
+ if (current->active_mm)
+ load_LDT(&current->active_mm->context);
+}
+#endif
+
+static int alloc_ldt(mm_context_t *pc, unsigned mincount, int reload)
+{
+ void *oldldt;
+ void *newldt;
+ unsigned oldsize;
+
+ if (mincount <= (unsigned)pc->size)
+ return 0;
+ oldsize = pc->size;
+ mincount = (mincount+511)&(~511);
+ if (mincount*LDT_ENTRY_SIZE > PAGE_SIZE)
+ newldt = vmalloc(mincount*LDT_ENTRY_SIZE);
+ else
+ newldt = kmalloc(mincount*LDT_ENTRY_SIZE, GFP_KERNEL);
+
+ if (!newldt)
+ return -ENOMEM;
+
+ if (oldsize)
+ memcpy(newldt, pc->ldt, oldsize*LDT_ENTRY_SIZE);
+ oldldt = pc->ldt;
+ memset(newldt+oldsize*LDT_ENTRY_SIZE, 0, (mincount-oldsize)*LDT_ENTRY_SIZE);
+ wmb();
+ pc->ldt = newldt;
+ wmb();
+ pc->size = mincount;
+ wmb();
+ if (reload) {
+#ifdef CONFIG_SMP
+ cpumask_t mask;
+
+ preempt_disable();
+#endif
+ make_pages_readonly(
+ pc->ldt,
+ (pc->size * LDT_ENTRY_SIZE) / PAGE_SIZE,
+ XENFEAT_writable_descriptor_tables);
+ load_LDT(pc);
+#ifdef CONFIG_SMP
+ mask = cpumask_of_cpu(smp_processor_id());
+ if (!cpus_equal(current->mm->cpu_vm_mask, mask))
+ smp_call_function(flush_ldt, NULL, 1, 1);
+ preempt_enable();
+#endif
+ }
+ if (oldsize) {
+ make_pages_writable(
+ oldldt,
+ (oldsize * LDT_ENTRY_SIZE) / PAGE_SIZE,
+ XENFEAT_writable_descriptor_tables);
+ if (oldsize*LDT_ENTRY_SIZE > PAGE_SIZE)
+ vfree(oldldt);
+ else
+ kfree(oldldt);
+ }
+ return 0;
+}
+
+static inline int copy_ldt(mm_context_t *new, mm_context_t *old)
+{
+ int err = alloc_ldt(new, old->size, 0);
+ if (err < 0)
+ return err;
+ memcpy(new->ldt, old->ldt, old->size*LDT_ENTRY_SIZE);
+ make_pages_readonly(
+ new->ldt,
+ (new->size * LDT_ENTRY_SIZE) / PAGE_SIZE,
+ XENFEAT_writable_descriptor_tables);
+ return 0;
+}
+
+/*
+ * we do not have to muck with descriptors here, that is
+ * done in switch_mm() as needed.
+ */
+int init_new_context(struct task_struct *tsk, struct mm_struct *mm)
+{
+ struct mm_struct * old_mm;
+ int retval = 0;
+
+ memset(&mm->context, 0, sizeof(mm->context));
+ init_MUTEX(&mm->context.sem);
+ old_mm = current->mm;
+ if (old_mm && old_mm->context.size > 0) {
+ down(&old_mm->context.sem);
+ retval = copy_ldt(&mm->context, &old_mm->context);
+ up(&old_mm->context.sem);
+ }
+ if (retval == 0) {
+ spin_lock(&mm_unpinned_lock);
+ list_add(&mm->context.unpinned, &mm_unpinned);
+ spin_unlock(&mm_unpinned_lock);
+ }
+ return retval;
+}
+
+/*
+ *
+ * Don't touch the LDT register - we're already in the next thread.
+ */
+void destroy_context(struct mm_struct *mm)
+{
+ if (mm->context.size) {
+ if (mm == current->active_mm)
+ clear_LDT();
+ make_pages_writable(
+ mm->context.ldt,
+ (mm->context.size * LDT_ENTRY_SIZE) / PAGE_SIZE,
+ XENFEAT_writable_descriptor_tables);
+ if (mm->context.size*LDT_ENTRY_SIZE > PAGE_SIZE)
+ vfree(mm->context.ldt);
+ else
+ kfree(mm->context.ldt);
+ mm->context.size = 0;
+ }
+ if (!mm->context.pinned) {
+ spin_lock(&mm_unpinned_lock);
+ list_del(&mm->context.unpinned);
+ spin_unlock(&mm_unpinned_lock);
+ }
+}
+
+static int read_ldt(void __user * ptr, unsigned long bytecount)
+{
+ int err;
+ unsigned long size;
+ struct mm_struct * mm = current->mm;
+
+ if (!mm->context.size)
+ return 0;
+ if (bytecount > LDT_ENTRY_SIZE*LDT_ENTRIES)
+ bytecount = LDT_ENTRY_SIZE*LDT_ENTRIES;
+
+ down(&mm->context.sem);
+ size = mm->context.size*LDT_ENTRY_SIZE;
+ if (size > bytecount)
+ size = bytecount;
+
+ err = 0;
+ if (copy_to_user(ptr, mm->context.ldt, size))
+ err = -EFAULT;
+ up(&mm->context.sem);
+ if (err < 0)
+ goto error_return;
+ if (size != bytecount) {
+ /* zero-fill the rest */
+ if (clear_user(ptr+size, bytecount-size) != 0) {
+ err = -EFAULT;
+ goto error_return;
+ }
+ }
+ return bytecount;
+error_return:
+ return err;
+}
+
+static int read_default_ldt(void __user * ptr, unsigned long bytecount)
+{
+ /* Arbitrary number */
+ /* x86-64 default LDT is all zeros */
+ if (bytecount > 128)
+ bytecount = 128;
+ if (clear_user(ptr, bytecount))
+ return -EFAULT;
+ return bytecount;
+}
+
+static int write_ldt(void __user * ptr, unsigned long bytecount, int oldmode)
+{
+ struct task_struct *me = current;
+ struct mm_struct * mm = me->mm;
+ __u32 entry_1, entry_2, *lp;
+ unsigned long mach_lp;
+ int error;
+ struct user_desc ldt_info;
+
+ error = -EINVAL;
+
+ if (bytecount != sizeof(ldt_info))
+ goto out;
+ error = -EFAULT;
+ if (copy_from_user(&ldt_info, ptr, bytecount))
+ goto out;
+
+ error = -EINVAL;
+ if (ldt_info.entry_number >= LDT_ENTRIES)
+ goto out;
+ if (ldt_info.contents == 3) {
+ if (oldmode)
+ goto out;
+ if (ldt_info.seg_not_present == 0)
+ goto out;
+ }
+
+ down(&mm->context.sem);
+ if (ldt_info.entry_number >= (unsigned)mm->context.size) {
+ error = alloc_ldt(&current->mm->context, ldt_info.entry_number+1, 1);
+ if (error < 0)
+ goto out_unlock;
+ }
+
+ lp = (__u32 *) ((ldt_info.entry_number << 3) + (char *) mm->context.ldt);
+ mach_lp = arbitrary_virt_to_machine(lp);
+
+ /* Allow LDTs to be cleared by the user. */
+ if (ldt_info.base_addr == 0 && ldt_info.limit == 0) {
+ if (oldmode || LDT_empty(&ldt_info)) {
+ entry_1 = 0;
+ entry_2 = 0;
+ goto install;
+ }
+ }
+
+ entry_1 = LDT_entry_a(&ldt_info);
+ entry_2 = LDT_entry_b(&ldt_info);
+ if (oldmode)
+ entry_2 &= ~(1 << 20);
+
+ /* Install the new entry ... */
+install:
+ error = HYPERVISOR_update_descriptor(mach_lp, (unsigned long)((entry_1 | (unsigned long) entry_2 << 32)));
+
+out_unlock:
+ up(&mm->context.sem);
+out:
+ return error;
+}
+
+asmlinkage int sys_modify_ldt(int func, void __user *ptr, unsigned long bytecount)
+{
+ int ret = -ENOSYS;
+
+ switch (func) {
+ case 0:
+ ret = read_ldt(ptr, bytecount);
+ break;
+ case 1:
+ ret = write_ldt(ptr, bytecount, 1);
+ break;
+ case 2:
+ ret = read_default_ldt(ptr, bytecount);
+ break;
+ case 0x11:
+ ret = write_ldt(ptr, bytecount, 0);
+ break;
+ }
+ return ret;
+}
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/kernel/mpparse_64-xen.c 2007-06-12 13:13:01.000000000 +0200
@@ -0,0 +1,1011 @@
+/*
+ * Intel Multiprocessor Specification 1.1 and 1.4
+ * compliant MP-table parsing routines.
+ *
+ * (c) 1995 Alan Cox, Building #3 <alan@redhat.com>
+ * (c) 1998, 1999, 2000 Ingo Molnar <mingo@redhat.com>
+ *
+ * Fixes
+ * Erich Boleyn : MP v1.4 and additional changes.
+ * Alan Cox : Added EBDA scanning
+ * Ingo Molnar : various cleanups and rewrites
+ * Maciej W. Rozycki: Bits for default MP configurations
+ * Paul Diefenbaugh: Added full ACPI support
+ */
+
+#include <linux/mm.h>
+#include <linux/init.h>
+#include <linux/delay.h>
+#include <linux/bootmem.h>
+#include <linux/smp_lock.h>
+#include <linux/kernel_stat.h>
+#include <linux/mc146818rtc.h>
+#include <linux/acpi.h>
+#include <linux/module.h>
+
+#include <asm/smp.h>
+#include <asm/mtrr.h>
+#include <asm/mpspec.h>
+#include <asm/pgalloc.h>
+#include <asm/io_apic.h>
+#include <asm/proto.h>
+#include <asm/acpi.h>
+
+/* Have we found an MP table */
+int smp_found_config;
+unsigned int __initdata maxcpus = NR_CPUS;
+
+int acpi_found_madt;
+
+/*
+ * Various Linux-internal data structures created from the
+ * MP-table.
+ */
+unsigned char apic_version [MAX_APICS];
+unsigned char mp_bus_id_to_type [MAX_MP_BUSSES] = { [0 ... MAX_MP_BUSSES-1] = -1 };
+int mp_bus_id_to_pci_bus [MAX_MP_BUSSES] = { [0 ... MAX_MP_BUSSES-1] = -1 };
+
+static int mp_current_pci_id = 0;
+/* I/O APIC entries */
+struct mpc_config_ioapic mp_ioapics[MAX_IO_APICS];
+
+/* # of MP IRQ source entries */
+struct mpc_config_intsrc mp_irqs[MAX_IRQ_SOURCES];
+
+/* MP IRQ source entries */
+int mp_irq_entries;
+
+int nr_ioapics;
+int pic_mode;
+unsigned long mp_lapic_addr = 0;
+
+
+
+/* Processor that is doing the boot up */
+unsigned int boot_cpu_id = -1U;
+/* Internal processor count */
+unsigned int num_processors __initdata = 0;
+
+unsigned disabled_cpus __initdata;
+
+/* Bitmask of physically existing CPUs */
+physid_mask_t phys_cpu_present_map = PHYSID_MASK_NONE;
+
+/* ACPI MADT entry parsing functions */
+#ifdef CONFIG_ACPI
+extern struct acpi_boot_flags acpi_boot;
+#ifdef CONFIG_X86_LOCAL_APIC
+extern int acpi_parse_lapic (acpi_table_entry_header *header);
+extern int acpi_parse_lapic_addr_ovr (acpi_table_entry_header *header);
+extern int acpi_parse_lapic_nmi (acpi_table_entry_header *header);
+#endif /*CONFIG_X86_LOCAL_APIC*/
+#ifdef CONFIG_X86_IO_APIC
+extern int acpi_parse_ioapic (acpi_table_entry_header *header);
+#endif /*CONFIG_X86_IO_APIC*/
+#endif /*CONFIG_ACPI*/
+
+u8 bios_cpu_apicid[NR_CPUS] = { [0 ... NR_CPUS-1] = BAD_APICID };
+
+
+/*
+ * Intel MP BIOS table parsing routines:
+ */
+
+/*
+ * Checksum an MP configuration block.
+ */
+
+static int __init mpf_checksum(unsigned char *mp, int len)
+{
+ int sum = 0;
+
+ while (len--)
+ sum += *mp++;
+
+ return sum & 0xFF;
+}
+
+#ifndef CONFIG_XEN
+static void __cpuinit MP_processor_info (struct mpc_config_processor *m)
+{
+ int cpu;
+ unsigned char ver;
+ cpumask_t tmp_map;
+
+ if (!(m->mpc_cpuflag & CPU_ENABLED)) {
+ disabled_cpus++;
+ return;
+ }
+
+ printk(KERN_INFO "Processor #%d %d:%d APIC version %d\n",
+ m->mpc_apicid,
+ (m->mpc_cpufeature & CPU_FAMILY_MASK)>>8,
+ (m->mpc_cpufeature & CPU_MODEL_MASK)>>4,
+ m->mpc_apicver);
+
+ if (m->mpc_cpuflag & CPU_BOOTPROCESSOR) {
+ Dprintk(" Bootup CPU\n");
+ boot_cpu_id = m->mpc_apicid;
+ }
+ if (num_processors >= NR_CPUS) {
+ printk(KERN_WARNING "WARNING: NR_CPUS limit of %i reached."
+ " Processor ignored.\n", NR_CPUS);
+ return;
+ }
+
+ num_processors++;
+ cpus_complement(tmp_map, cpu_present_map);
+ cpu = first_cpu(tmp_map);
+
+#if MAX_APICS < 255
+ if ((int)m->mpc_apicid > MAX_APICS) {
+ printk(KERN_ERR "Processor #%d INVALID. (Max ID: %d).\n",
+ m->mpc_apicid, MAX_APICS);
+ return;
+ }
+#endif
+ ver = m->mpc_apicver;
+
+ physid_set(m->mpc_apicid, phys_cpu_present_map);
+ /*
+ * Validate version
+ */
+ if (ver == 0x0) {
+ printk(KERN_ERR "BIOS bug, APIC version is 0 for CPU#%d! fixing up to 0x10. (tell your hw vendor)\n", m->mpc_apicid);
+ ver = 0x10;
+ }
+ apic_version[m->mpc_apicid] = ver;
+ if (m->mpc_cpuflag & CPU_BOOTPROCESSOR) {
+ /*
+ * bios_cpu_apicid is required to have processors listed
+ * in same order as logical cpu numbers. Hence the first
+ * entry is BSP, and so on.
+ */
+ cpu = 0;
+ }
+ bios_cpu_apicid[cpu] = m->mpc_apicid;
+ x86_cpu_to_apicid[cpu] = m->mpc_apicid;
+
+ cpu_set(cpu, cpu_possible_map);
+ cpu_set(cpu, cpu_present_map);
+}
+#else
+static void __cpuinit MP_processor_info (struct mpc_config_processor *m)
+{
+ num_processors++;
+}
+#endif /* CONFIG_XEN */
+
+static void __init MP_bus_info (struct mpc_config_bus *m)
+{
+ char str[7];
+
+ memcpy(str, m->mpc_bustype, 6);
+ str[6] = 0;
+ Dprintk("Bus #%d is %s\n", m->mpc_busid, str);
+
+ if (strncmp(str, "ISA", 3) == 0) {
+ mp_bus_id_to_type[m->mpc_busid] = MP_BUS_ISA;
+ } else if (strncmp(str, "EISA", 4) == 0) {
+ mp_bus_id_to_type[m->mpc_busid] = MP_BUS_EISA;
+ } else if (strncmp(str, "PCI", 3) == 0) {
+ mp_bus_id_to_type[m->mpc_busid] = MP_BUS_PCI;
+ mp_bus_id_to_pci_bus[m->mpc_busid] = mp_current_pci_id;
+ mp_current_pci_id++;
+ } else if (strncmp(str, "MCA", 3) == 0) {
+ mp_bus_id_to_type[m->mpc_busid] = MP_BUS_MCA;
+ } else {
+ printk(KERN_ERR "Unknown bustype %s\n", str);
+ }
+}
+
+static void __init MP_ioapic_info (struct mpc_config_ioapic *m)
+{
+ if (!(m->mpc_flags & MPC_APIC_USABLE))
+ return;
+
+ printk("I/O APIC #%d Version %d at 0x%X.\n",
+ m->mpc_apicid, m->mpc_apicver, m->mpc_apicaddr);
+ if (nr_ioapics >= MAX_IO_APICS) {
+ printk(KERN_ERR "Max # of I/O APICs (%d) exceeded (found %d).\n",
+ MAX_IO_APICS, nr_ioapics);
+ panic("Recompile kernel with bigger MAX_IO_APICS!.\n");
+ }
+ if (!m->mpc_apicaddr) {
+ printk(KERN_ERR "WARNING: bogus zero I/O APIC address"
+ " found in MP table, skipping!\n");
+ return;
+ }
+ mp_ioapics[nr_ioapics] = *m;
+ nr_ioapics++;
+}
+
+static void __init MP_intsrc_info (struct mpc_config_intsrc *m)
+{
+ mp_irqs [mp_irq_entries] = *m;
+ Dprintk("Int: type %d, pol %d, trig %d, bus %d,"
+ " IRQ %02x, APIC ID %x, APIC INT %02x\n",
+ m->mpc_irqtype, m->mpc_irqflag & 3,
+ (m->mpc_irqflag >> 2) & 3, m->mpc_srcbus,
+ m->mpc_srcbusirq, m->mpc_dstapic, m->mpc_dstirq);
+ if (++mp_irq_entries >= MAX_IRQ_SOURCES)
+ panic("Max # of irq sources exceeded!!\n");
+}
+
+static void __init MP_lintsrc_info (struct mpc_config_lintsrc *m)
+{
+ Dprintk("Lint: type %d, pol %d, trig %d, bus %d,"
+ " IRQ %02x, APIC ID %x, APIC LINT %02x\n",
+ m->mpc_irqtype, m->mpc_irqflag & 3,
+ (m->mpc_irqflag >> 2) &3, m->mpc_srcbusid,
+ m->mpc_srcbusirq, m->mpc_destapic, m->mpc_destapiclint);
+ /*
+ * Well it seems all SMP boards in existence
+ * use ExtINT/LVT1 == LINT0 and
+ * NMI/LVT2 == LINT1 - the following check
+ * will show us if this assumptions is false.
+ * Until then we do not have to add baggage.
+ */
+ if ((m->mpc_irqtype == mp_ExtINT) &&
+ (m->mpc_destapiclint != 0))
+ BUG();
+ if ((m->mpc_irqtype == mp_NMI) &&
+ (m->mpc_destapiclint != 1))
+ BUG();
+}
+
+/*
+ * Read/parse the MPC
+ */
+
+static int __init smp_read_mpc(struct mp_config_table *mpc)
+{
+ char str[16];
+ int count=sizeof(*mpc);
+ unsigned char *mpt=((unsigned char *)mpc)+count;
+
+ if (memcmp(mpc->mpc_signature,MPC_SIGNATURE,4)) {
+ printk("SMP mptable: bad signature [%c%c%c%c]!\n",
+ mpc->mpc_signature[0],
+ mpc->mpc_signature[1],
+ mpc->mpc_signature[2],
+ mpc->mpc_signature[3]);
+ return 0;
+ }
+ if (mpf_checksum((unsigned char *)mpc,mpc->mpc_length)) {
+ printk("SMP mptable: checksum error!\n");
+ return 0;
+ }
+ if (mpc->mpc_spec!=0x01 && mpc->mpc_spec!=0x04) {
+ printk(KERN_ERR "SMP mptable: bad table version (%d)!!\n",
+ mpc->mpc_spec);
+ return 0;
+ }
+ if (!mpc->mpc_lapic) {
+ printk(KERN_ERR "SMP mptable: null local APIC address!\n");
+ return 0;
+ }
+ memcpy(str,mpc->mpc_oem,8);
+ str[8]=0;
+ printk(KERN_INFO "OEM ID: %s ",str);
+
+ memcpy(str,mpc->mpc_productid,12);
+ str[12]=0;
+ printk("Product ID: %s ",str);
+
+ printk("APIC at: 0x%X\n",mpc->mpc_lapic);
+
+ /* save the local APIC address, it might be non-default */
+ if (!acpi_lapic)
+ mp_lapic_addr = mpc->mpc_lapic;
+
+ /*
+ * Now process the configuration blocks.
+ */
+ while (count < mpc->mpc_length) {
+ switch(*mpt) {
+ case MP_PROCESSOR:
+ {
+ struct mpc_config_processor *m=
+ (struct mpc_config_processor *)mpt;
+ if (!acpi_lapic)
+ MP_processor_info(m);
+ mpt += sizeof(*m);
+ count += sizeof(*m);
+ break;
+ }
+ case MP_BUS:
+ {
+ struct mpc_config_bus *m=
+ (struct mpc_config_bus *)mpt;
+ MP_bus_info(m);
+ mpt += sizeof(*m);
+ count += sizeof(*m);
+ break;
+ }
+ case MP_IOAPIC:
+ {
+ struct mpc_config_ioapic *m=
+ (struct mpc_config_ioapic *)mpt;
+ MP_ioapic_info(m);
+ mpt+=sizeof(*m);
+ count+=sizeof(*m);
+ break;
+ }
+ case MP_INTSRC:
+ {
+ struct mpc_config_intsrc *m=
+ (struct mpc_config_intsrc *)mpt;
+
+ MP_intsrc_info(m);
+ mpt+=sizeof(*m);
+ count+=sizeof(*m);
+ break;
+ }
+ case MP_LINTSRC:
+ {
+ struct mpc_config_lintsrc *m=
+ (struct mpc_config_lintsrc *)mpt;
+ MP_lintsrc_info(m);
+ mpt+=sizeof(*m);
+ count+=sizeof(*m);
+ break;
+ }
+ }
+ }
+ clustered_apic_check();
+ if (!num_processors)
+ printk(KERN_ERR "SMP mptable: no processors registered!\n");
+ return num_processors;
+}
+
+static int __init ELCR_trigger(unsigned int irq)
+{
+ unsigned int port;
+
+ port = 0x4d0 + (irq >> 3);
+ return (inb(port) >> (irq & 7)) & 1;
+}
+
+static void __init construct_default_ioirq_mptable(int mpc_default_type)
+{
+ struct mpc_config_intsrc intsrc;
+ int i;
+ int ELCR_fallback = 0;
+
+ intsrc.mpc_type = MP_INTSRC;
+ intsrc.mpc_irqflag = 0; /* conforming */
+ intsrc.mpc_srcbus = 0;
+ intsrc.mpc_dstapic = mp_ioapics[0].mpc_apicid;
+
+ intsrc.mpc_irqtype = mp_INT;
+
+ /*
+ * If true, we have an ISA/PCI system with no IRQ entries
+ * in the MP table. To prevent the PCI interrupts from being set up
+ * incorrectly, we try to use the ELCR. The sanity check to see if
+ * there is good ELCR data is very simple - IRQ0, 1, 2 and 13 can
+ * never be level sensitive, so we simply see if the ELCR agrees.
+ * If it does, we assume it's valid.
+ */
+ if (mpc_default_type == 5) {
+ printk(KERN_INFO "ISA/PCI bus type with no IRQ information... falling back to ELCR\n");
+
+ if (ELCR_trigger(0) || ELCR_trigger(1) || ELCR_trigger(2) || ELCR_trigger(13))
+ printk(KERN_ERR "ELCR contains invalid data... not using ELCR\n");
+ else {
+ printk(KERN_INFO "Using ELCR to identify PCI interrupts\n");
+ ELCR_fallback = 1;
+ }
+ }
+
+ for (i = 0; i < 16; i++) {
+ switch (mpc_default_type) {
+ case 2:
+ if (i == 0 || i == 13)
+ continue; /* IRQ0 & IRQ13 not connected */
+ /* fall through */
+ default:
+ if (i == 2)
+ continue; /* IRQ2 is never connected */
+ }
+
+ if (ELCR_fallback) {
+ /*
+ * If the ELCR indicates a level-sensitive interrupt, we
+ * copy that information over to the MP table in the
+ * irqflag field (level sensitive, active high polarity).
+ */
+ if (ELCR_trigger(i))
+ intsrc.mpc_irqflag = 13;
+ else
+ intsrc.mpc_irqflag = 0;
+ }
+
+ intsrc.mpc_srcbusirq = i;
+ intsrc.mpc_dstirq = i ? i : 2; /* IRQ0 to INTIN2 */
+ MP_intsrc_info(&intsrc);
+ }
+
+ intsrc.mpc_irqtype = mp_ExtINT;
+ intsrc.mpc_srcbusirq = 0;
+ intsrc.mpc_dstirq = 0; /* 8259A to INTIN0 */
+ MP_intsrc_info(&intsrc);
+}
+
+static inline void __init construct_default_ISA_mptable(int mpc_default_type)
+{
+ struct mpc_config_processor processor;
+ struct mpc_config_bus bus;
+ struct mpc_config_ioapic ioapic;
+ struct mpc_config_lintsrc lintsrc;
+ int linttypes[2] = { mp_ExtINT, mp_NMI };
+ int i;
+
+ /*
+ * local APIC has default address
+ */
+ mp_lapic_addr = APIC_DEFAULT_PHYS_BASE;
+
+ /*
+ * 2 CPUs, numbered 0 & 1.
+ */
+ processor.mpc_type = MP_PROCESSOR;
+ /* Either an integrated APIC or a discrete 82489DX. */
+ processor.mpc_apicver = mpc_default_type > 4 ? 0x10 : 0x01;
+ processor.mpc_cpuflag = CPU_ENABLED;
+ processor.mpc_cpufeature = (boot_cpu_data.x86 << 8) |
+ (boot_cpu_data.x86_model << 4) |
+ boot_cpu_data.x86_mask;
+ processor.mpc_featureflag = boot_cpu_data.x86_capability[0];
+ processor.mpc_reserved[0] = 0;
+ processor.mpc_reserved[1] = 0;
+ for (i = 0; i < 2; i++) {
+ processor.mpc_apicid = i;
+ MP_processor_info(&processor);
+ }
+
+ bus.mpc_type = MP_BUS;
+ bus.mpc_busid = 0;
+ switch (mpc_default_type) {
+ default:
+ printk(KERN_ERR "???\nUnknown standard configuration %d\n",
+ mpc_default_type);
+ /* fall through */
+ case 1:
+ case 5:
+ memcpy(bus.mpc_bustype, "ISA ", 6);
+ break;
+ case 2:
+ case 6:
+ case 3:
+ memcpy(bus.mpc_bustype, "EISA ", 6);
+ break;
+ case 4:
+ case 7:
+ memcpy(bus.mpc_bustype, "MCA ", 6);
+ }
+ MP_bus_info(&bus);
+ if (mpc_default_type > 4) {
+ bus.mpc_busid = 1;
+ memcpy(bus.mpc_bustype, "PCI ", 6);
+ MP_bus_info(&bus);
+ }
+
+ ioapic.mpc_type = MP_IOAPIC;
+ ioapic.mpc_apicid = 2;
+ ioapic.mpc_apicver = mpc_default_type > 4 ? 0x10 : 0x01;
+ ioapic.mpc_flags = MPC_APIC_USABLE;
+ ioapic.mpc_apicaddr = 0xFEC00000;
+ MP_ioapic_info(&ioapic);
+
+ /*
+ * We set up most of the low 16 IO-APIC pins according to MPS rules.
+ */
+ construct_default_ioirq_mptable(mpc_default_type);
+
+ lintsrc.mpc_type = MP_LINTSRC;
+ lintsrc.mpc_irqflag = 0; /* conforming */
+ lintsrc.mpc_srcbusid = 0;
+ lintsrc.mpc_srcbusirq = 0;
+ lintsrc.mpc_destapic = MP_APIC_ALL;
+ for (i = 0; i < 2; i++) {
+ lintsrc.mpc_irqtype = linttypes[i];
+ lintsrc.mpc_destapiclint = i;
+ MP_lintsrc_info(&lintsrc);
+ }
+}
+
+static struct intel_mp_floating *mpf_found;
+
+/*
+ * Scan the memory blocks for an SMP configuration block.
+ */
+void __init get_smp_config (void)
+{
+ struct intel_mp_floating *mpf = mpf_found;
+
+ /*
+ * ACPI supports both logical (e.g. Hyper-Threading) and physical
+ * processors, where MPS only supports physical.
+ */
+ if (acpi_lapic && acpi_ioapic) {
+ printk(KERN_INFO "Using ACPI (MADT) for SMP configuration information\n");
+ return;
+ }
+ else if (acpi_lapic)
+ printk(KERN_INFO "Using ACPI for processor (LAPIC) configuration information\n");
+
+ printk("Intel MultiProcessor Specification v1.%d\n", mpf->mpf_specification);
+ if (mpf->mpf_feature2 & (1<<7)) {
+ printk(KERN_INFO " IMCR and PIC compatibility mode.\n");
+ pic_mode = 1;
+ } else {
+ printk(KERN_INFO " Virtual Wire compatibility mode.\n");
+ pic_mode = 0;
+ }
+
+ /*
+ * Now see if we need to read further.
+ */
+ if (mpf->mpf_feature1 != 0) {
+
+ printk(KERN_INFO "Default MP configuration #%d\n", mpf->mpf_feature1);
+ construct_default_ISA_mptable(mpf->mpf_feature1);
+
+ } else if (mpf->mpf_physptr) {
+
+ /*
+ * Read the physical hardware table. Anything here will
+ * override the defaults.
+ */
+ if (!smp_read_mpc(isa_bus_to_virt(mpf->mpf_physptr))) {
+ smp_found_config = 0;
+ printk(KERN_ERR "BIOS bug, MP table errors detected!...\n");
+ printk(KERN_ERR "... disabling SMP support. (tell your hw vendor)\n");
+ return;
+ }
+ /*
+ * If there are no explicit MP IRQ entries, then we are
+ * broken. We set up most of the low 16 IO-APIC pins to
+ * ISA defaults and hope it will work.
+ */
+ if (!mp_irq_entries) {
+ struct mpc_config_bus bus;
+
+ printk(KERN_ERR "BIOS bug, no explicit IRQ entries, using default mptable. (tell your hw vendor)\n");
+
+ bus.mpc_type = MP_BUS;
+ bus.mpc_busid = 0;
+ memcpy(bus.mpc_bustype, "ISA ", 6);
+ MP_bus_info(&bus);
+
+ construct_default_ioirq_mptable(0);
+ }
+
+ } else
+ BUG();
+
+ printk(KERN_INFO "Processors: %d\n", num_processors);
+ /*
+ * Only use the first configuration found.
+ */
+}
+
+static int __init smp_scan_config (unsigned long base, unsigned long length)
+{
+ extern void __bad_mpf_size(void);
+ unsigned int *bp = isa_bus_to_virt(base);
+ struct intel_mp_floating *mpf;
+
+ Dprintk("Scan SMP from %p for %ld bytes.\n", bp,length);
+ if (sizeof(*mpf) != 16)
+ __bad_mpf_size();
+
+ while (length > 0) {
+ mpf = (struct intel_mp_floating *)bp;
+ if ((*bp == SMP_MAGIC_IDENT) &&
+ (mpf->mpf_length == 1) &&
+ !mpf_checksum((unsigned char *)bp, 16) &&
+ ((mpf->mpf_specification == 1)
+ || (mpf->mpf_specification == 4)) ) {
+
+ smp_found_config = 1;
+ mpf_found = mpf;
+ return 1;
+ }
+ bp += 4;
+ length -= 16;
+ }
+ return 0;
+}
+
+void __init find_intel_smp (void)
+{
+ unsigned int address;
+
+ /*
+ * FIXME: Linux assumes you have 640K of base ram..
+ * this continues the error...
+ *
+ * 1) Scan the bottom 1K for a signature
+ * 2) Scan the top 1K of base RAM
+ * 3) Scan the 64K of bios
+ */
+ if (smp_scan_config(0x0,0x400) ||
+ smp_scan_config(639*0x400,0x400) ||
+ smp_scan_config(0xF0000,0x10000))
+ return;
+ /*
+ * If it is an SMP machine we should know now, unless the
+ * configuration is in an EISA/MCA bus machine with an
+ * extended bios data area.
+ *
+ * there is a real-mode segmented pointer pointing to the
+ * 4K EBDA area at 0x40E, calculate and scan it here.
+ *
+ * NOTE! There are Linux loaders that will corrupt the EBDA
+ * area, and as such this kind of SMP config may be less
+ * trustworthy, simply because the SMP table may have been
+ * stomped on during early boot. These loaders are buggy and
+ * should be fixed.
+ */
+
+ address = *(unsigned short *)phys_to_virt(0x40E);
+ address <<= 4;
+ if (smp_scan_config(address, 0x1000))
+ return;
+
+ /* If we have come this far, we did not find an MP table */
+ printk(KERN_INFO "No mptable found.\n");
+}
+
+/*
+ * - Intel MP Configuration Table
+ */
+void __init find_smp_config (void)
+{
+#ifdef CONFIG_X86_LOCAL_APIC
+ find_intel_smp();
+#endif
+}
+
+
+/* --------------------------------------------------------------------------
+ ACPI-based MP Configuration
+ -------------------------------------------------------------------------- */
+
+#ifdef CONFIG_ACPI
+
+void __init mp_register_lapic_address (
+ u64 address)
+{
+#ifndef CONFIG_XEN
+ mp_lapic_addr = (unsigned long) address;
+
+ set_fixmap_nocache(FIX_APIC_BASE, mp_lapic_addr);
+
+ if (boot_cpu_id == -1U)
+ boot_cpu_id = GET_APIC_ID(apic_read(APIC_ID));
+
+ Dprintk("Boot CPU = %d\n", boot_cpu_physical_apicid);
+#endif
+}
+
+
+void __cpuinit mp_register_lapic (
+ u8 id,
+ u8 enabled)
+{
+ struct mpc_config_processor processor;
+ int boot_cpu = 0;
+
+ if (id >= MAX_APICS) {
+ printk(KERN_WARNING "Processor #%d invalid (max %d)\n",
+ id, MAX_APICS);
+ return;
+ }
+
+ if (id == boot_cpu_physical_apicid)
+ boot_cpu = 1;
+
+#ifndef CONFIG_XEN
+ processor.mpc_type = MP_PROCESSOR;
+ processor.mpc_apicid = id;
+ processor.mpc_apicver = GET_APIC_VERSION(apic_read(APIC_LVR));
+ processor.mpc_cpuflag = (enabled ? CPU_ENABLED : 0);
+ processor.mpc_cpuflag |= (boot_cpu ? CPU_BOOTPROCESSOR : 0);
+ processor.mpc_cpufeature = (boot_cpu_data.x86 << 8) |
+ (boot_cpu_data.x86_model << 4) | boot_cpu_data.x86_mask;
+ processor.mpc_featureflag = boot_cpu_data.x86_capability[0];
+ processor.mpc_reserved[0] = 0;
+ processor.mpc_reserved[1] = 0;
+#endif
+
+ MP_processor_info(&processor);
+}
+
+#ifdef CONFIG_X86_IO_APIC
+
+#define MP_ISA_BUS 0
+#define MP_MAX_IOAPIC_PIN 127
+
+static struct mp_ioapic_routing {
+ int apic_id;
+ int gsi_start;
+ int gsi_end;
+ u32 pin_programmed[4];
+} mp_ioapic_routing[MAX_IO_APICS];
+
+
+static int mp_find_ioapic (
+ int gsi)
+{
+ int i = 0;
+
+ /* Find the IOAPIC that manages this GSI. */
+ for (i = 0; i < nr_ioapics; i++) {
+ if ((gsi >= mp_ioapic_routing[i].gsi_start)
+ && (gsi <= mp_ioapic_routing[i].gsi_end))
+ return i;
+ }
+
+ printk(KERN_ERR "ERROR: Unable to locate IOAPIC for GSI %d\n", gsi);
+
+ return -1;
+}
+
+
+void __init mp_register_ioapic (
+ u8 id,
+ u32 address,
+ u32 gsi_base)
+{
+ int idx = 0;
+
+ if (nr_ioapics >= MAX_IO_APICS) {
+ printk(KERN_ERR "ERROR: Max # of I/O APICs (%d) exceeded "
+ "(found %d)\n", MAX_IO_APICS, nr_ioapics);
+ panic("Recompile kernel with bigger MAX_IO_APICS!\n");
+ }
+ if (!address) {
+ printk(KERN_ERR "WARNING: Bogus (zero) I/O APIC address"
+ " found in MADT table, skipping!\n");
+ return;
+ }
+
+ idx = nr_ioapics++;
+
+ mp_ioapics[idx].mpc_type = MP_IOAPIC;
+ mp_ioapics[idx].mpc_flags = MPC_APIC_USABLE;
+ mp_ioapics[idx].mpc_apicaddr = address;
+
+#ifndef CONFIG_XEN
+ set_fixmap_nocache(FIX_IO_APIC_BASE_0 + idx, address);
+#endif
+ mp_ioapics[idx].mpc_apicid = id;
+ mp_ioapics[idx].mpc_apicver = io_apic_get_version(idx);
+
+ /*
+ * Build basic IRQ lookup table to facilitate gsi->io_apic lookups
+ * and to prevent reprogramming of IOAPIC pins (PCI IRQs).
+ */
+ mp_ioapic_routing[idx].apic_id = mp_ioapics[idx].mpc_apicid;
+ mp_ioapic_routing[idx].gsi_start = gsi_base;
+ mp_ioapic_routing[idx].gsi_end = gsi_base +
+ io_apic_get_redir_entries(idx);
+
+ printk(KERN_INFO "IOAPIC[%d]: apic_id %d, version %d, address 0x%x, "
+ "GSI %d-%d\n", idx, mp_ioapics[idx].mpc_apicid,
+ mp_ioapics[idx].mpc_apicver, mp_ioapics[idx].mpc_apicaddr,
+ mp_ioapic_routing[idx].gsi_start,
+ mp_ioapic_routing[idx].gsi_end);
+
+ return;
+}
+
+
+void __init mp_override_legacy_irq (
+ u8 bus_irq,
+ u8 polarity,
+ u8 trigger,
+ u32 gsi)
+{
+ struct mpc_config_intsrc intsrc;
+ int ioapic = -1;
+ int pin = -1;
+
+ /*
+ * Convert 'gsi' to 'ioapic.pin'.
+ */
+ ioapic = mp_find_ioapic(gsi);
+ if (ioapic < 0)
+ return;
+ pin = gsi - mp_ioapic_routing[ioapic].gsi_start;
+
+ /*
+ * TBD: This check is for faulty timer entries, where the override
+ * erroneously sets the trigger to level, resulting in a HUGE
+ * increase of timer interrupts!
+ */
+ if ((bus_irq == 0) && (trigger == 3))
+ trigger = 1;
+
+ intsrc.mpc_type = MP_INTSRC;
+ intsrc.mpc_irqtype = mp_INT;
+ intsrc.mpc_irqflag = (trigger << 2) | polarity;
+ intsrc.mpc_srcbus = MP_ISA_BUS;
+ intsrc.mpc_srcbusirq = bus_irq; /* IRQ */
+ intsrc.mpc_dstapic = mp_ioapics[ioapic].mpc_apicid; /* APIC ID */
+ intsrc.mpc_dstirq = pin; /* INTIN# */
+
+ Dprintk("Int: type %d, pol %d, trig %d, bus %d, irq %d, %d-%d\n",
+ intsrc.mpc_irqtype, intsrc.mpc_irqflag & 3,
+ (intsrc.mpc_irqflag >> 2) & 3, intsrc.mpc_srcbus,
+ intsrc.mpc_srcbusirq, intsrc.mpc_dstapic, intsrc.mpc_dstirq);
+
+ mp_irqs[mp_irq_entries] = intsrc;
+ if (++mp_irq_entries == MAX_IRQ_SOURCES)
+ panic("Max # of irq sources exceeded!\n");
+
+ return;
+}
+
+
+void __init mp_config_acpi_legacy_irqs (void)
+{
+ struct mpc_config_intsrc intsrc;
+ int i = 0;
+ int ioapic = -1;
+
+ /*
+ * Fabricate the legacy ISA bus (bus #31).
+ */
+ mp_bus_id_to_type[MP_ISA_BUS] = MP_BUS_ISA;
+ Dprintk("Bus #%d is ISA\n", MP_ISA_BUS);
+
+ /*
+ * Locate the IOAPIC that manages the ISA IRQs (0-15).
+ */
+ ioapic = mp_find_ioapic(0);
+ if (ioapic < 0)
+ return;
+
+ intsrc.mpc_type = MP_INTSRC;
+ intsrc.mpc_irqflag = 0; /* Conforming */
+ intsrc.mpc_srcbus = MP_ISA_BUS;
+ intsrc.mpc_dstapic = mp_ioapics[ioapic].mpc_apicid;
+
+ /*
+ * Use the default configuration for the IRQs 0-15. Unless
+ * overridden by (MADT) interrupt source override entries.
+ */
+ for (i = 0; i < 16; i++) {
+ int idx;
+
+ for (idx = 0; idx < mp_irq_entries; idx++) {
+ struct mpc_config_intsrc *irq = mp_irqs + idx;
+
+ /* Do we already have a mapping for this ISA IRQ? */
+ if (irq->mpc_srcbus == MP_ISA_BUS && irq->mpc_srcbusirq == i)
+ break;
+
+ /* Do we already have a mapping for this IOAPIC pin */
+ if ((irq->mpc_dstapic == intsrc.mpc_dstapic) &&
+ (irq->mpc_dstirq == i))
+ break;
+ }
+
+ if (idx != mp_irq_entries) {
+ printk(KERN_DEBUG "ACPI: IRQ%d used by override.\n", i);
+ continue; /* IRQ already used */
+ }
+
+ intsrc.mpc_irqtype = mp_INT;
+ intsrc.mpc_srcbusirq = i; /* Identity mapped */
+ intsrc.mpc_dstirq = i;
+
+ Dprintk("Int: type %d, pol %d, trig %d, bus %d, irq %d, "
+ "%d-%d\n", intsrc.mpc_irqtype, intsrc.mpc_irqflag & 3,
+ (intsrc.mpc_irqflag >> 2) & 3, intsrc.mpc_srcbus,
+ intsrc.mpc_srcbusirq, intsrc.mpc_dstapic,
+ intsrc.mpc_dstirq);
+
+ mp_irqs[mp_irq_entries] = intsrc;
+ if (++mp_irq_entries == MAX_IRQ_SOURCES)
+ panic("Max # of irq sources exceeded!\n");
+ }
+
+ return;
+}
+
+#define MAX_GSI_NUM 4096
+
+int mp_register_gsi(u32 gsi, int triggering, int polarity)
+{
+ int ioapic = -1;
+ int ioapic_pin = 0;
+ int idx, bit = 0;
+ static int pci_irq = 16;
+ /*
+ * Mapping between Global System Interrupts, which
+ * represent all possible interrupts, to the IRQs
+ * assigned to actual devices.
+ */
+ static int gsi_to_irq[MAX_GSI_NUM];
+
+ if (acpi_irq_model != ACPI_IRQ_MODEL_IOAPIC)
+ return gsi;
+
+ /* Don't set up the ACPI SCI because it's already set up */
+ if (acpi_fadt.sci_int == gsi)
+ return gsi;
+
+ ioapic = mp_find_ioapic(gsi);
+ if (ioapic < 0) {
+ printk(KERN_WARNING "No IOAPIC for GSI %u\n", gsi);
+ return gsi;
+ }
+
+ ioapic_pin = gsi - mp_ioapic_routing[ioapic].gsi_start;
+
+ /*
+ * Avoid pin reprogramming. PRTs typically include entries
+ * with redundant pin->gsi mappings (but unique PCI devices);
+ * we only program the IOAPIC on the first.
+ */
+ bit = ioapic_pin % 32;
+ idx = (ioapic_pin < 32) ? 0 : (ioapic_pin / 32);
+ if (idx > 3) {
+ printk(KERN_ERR "Invalid reference to IOAPIC pin "
+ "%d-%d\n", mp_ioapic_routing[ioapic].apic_id,
+ ioapic_pin);
+ return gsi;
+ }
+ if ((1<<bit) & mp_ioapic_routing[ioapic].pin_programmed[idx]) {
+ Dprintk(KERN_DEBUG "Pin %d-%d already programmed\n",
+ mp_ioapic_routing[ioapic].apic_id, ioapic_pin);
+ return gsi_to_irq[gsi];
+ }
+
+ mp_ioapic_routing[ioapic].pin_programmed[idx] |= (1<<bit);
+
+ if (triggering == ACPI_LEVEL_SENSITIVE) {
+ /*
+ * For PCI devices assign IRQs in order, avoiding gaps
+ * due to unused I/O APIC pins.
+ */
+ int irq = gsi;
+ if (gsi < MAX_GSI_NUM) {
+ /*
+ * Retain the VIA chipset work-around (gsi > 15), but
+ * avoid a problem where the 8254 timer (IRQ0) is setup
+ * via an override (so it's not on pin 0 of the ioapic),
+ * and at the same time, the pin 0 interrupt is a PCI
+ * type. The gsi > 15 test could cause these two pins
+ * to be shared as IRQ0, and they are not shareable.
+ * So test for this condition, and if necessary, avoid
+ * the pin collision.
+ */
+ if (gsi > 15 || (gsi == 0 && !timer_uses_ioapic_pin_0))
+ gsi = pci_irq++;
+ /*
+ * Don't assign IRQ used by ACPI SCI
+ */
+ if (gsi == acpi_fadt.sci_int)
+ gsi = pci_irq++;
+ gsi_to_irq[irq] = gsi;
+ } else {
+ printk(KERN_ERR "GSI %u is too high\n", gsi);
+ return gsi;
+ }
+ }
+
+ io_apic_set_pci_routing(ioapic, ioapic_pin, gsi,
+ triggering == ACPI_EDGE_SENSITIVE ? 0 : 1,
+ polarity == ACPI_ACTIVE_HIGH ? 0 : 1);
+ return gsi;
+}
+
+#endif /*CONFIG_X86_IO_APIC*/
+#endif /*CONFIG_ACPI*/
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/kernel/process_64-xen.c 2008-08-07 12:44:36.000000000 +0200
@@ -0,0 +1,848 @@
+/*
+ * linux/arch/x86-64/kernel/process.c
+ *
+ * Copyright (C) 1995 Linus Torvalds
+ *
+ * Pentium III FXSR, SSE support
+ * Gareth Hughes <gareth@valinux.com>, May 2000
+ *
+ * X86-64 port
+ * Andi Kleen.
+ *
+ * CPU hotplug support - ashok.raj@intel.com
+ *
+ * Jun Nakajima <jun.nakajima@intel.com>
+ * Modified for Xen
+ */
+
+/*
+ * This file handles the architecture-dependent parts of process handling..
+ */
+
+#include <stdarg.h>
+
+#include <linux/cpu.h>
+#include <linux/errno.h>
+#include <linux/sched.h>
+#include <linux/kernel.h>
+#include <linux/mm.h>
+#include <linux/elfcore.h>
+#include <linux/smp.h>
+#include <linux/slab.h>
+#include <linux/user.h>
+#include <linux/module.h>
+#include <linux/a.out.h>
+#include <linux/interrupt.h>
+#include <linux/delay.h>
+#include <linux/ptrace.h>
+#include <linux/utsname.h>
+#include <linux/random.h>
+#include <linux/notifier.h>
+#include <linux/kprobes.h>
+
+#include <asm/uaccess.h>
+#include <asm/pgtable.h>
+#include <asm/system.h>
+#include <asm/io.h>
+#include <asm/processor.h>
+#include <asm/i387.h>
+#include <asm/mmu_context.h>
+#include <asm/pda.h>
+#include <asm/prctl.h>
+#include <asm/kdebug.h>
+#include <xen/interface/platform.h>
+#include <xen/interface/physdev.h>
+#include <xen/interface/vcpu.h>
+#include <asm/desc.h>
+#include <asm/proto.h>
+#include <asm/hardirq.h>
+#include <asm/ia32.h>
+#include <asm/idle.h>
+
+#include <xen/cpu_hotplug.h>
+
+asmlinkage extern void ret_from_fork(void);
+
+unsigned long kernel_thread_flags = CLONE_VM | CLONE_UNTRACED;
+
+unsigned long boot_option_idle_override = 0;
+EXPORT_SYMBOL(boot_option_idle_override);
+
+/*
+ * Powermanagement idle function, if any..
+ */
+void (*pm_idle)(void);
+EXPORT_SYMBOL(pm_idle);
+static DEFINE_PER_CPU(unsigned int, cpu_idle_state);
+
+static ATOMIC_NOTIFIER_HEAD(idle_notifier);
+
+void idle_notifier_register(struct notifier_block *n)
+{
+ atomic_notifier_chain_register(&idle_notifier, n);
+}
+EXPORT_SYMBOL_GPL(idle_notifier_register);
+
+void idle_notifier_unregister(struct notifier_block *n)
+{
+ atomic_notifier_chain_unregister(&idle_notifier, n);
+}
+EXPORT_SYMBOL(idle_notifier_unregister);
+
+enum idle_state { CPU_IDLE, CPU_NOT_IDLE };
+static DEFINE_PER_CPU(enum idle_state, idle_state) = CPU_NOT_IDLE;
+
+void enter_idle(void)
+{
+ __get_cpu_var(idle_state) = CPU_IDLE;
+ atomic_notifier_call_chain(&idle_notifier, IDLE_START, NULL);
+}
+
+static void __exit_idle(void)
+{
+ __get_cpu_var(idle_state) = CPU_NOT_IDLE;
+ atomic_notifier_call_chain(&idle_notifier, IDLE_END, NULL);
+}
+
+/* Called from interrupts to signify idle end */
+void exit_idle(void)
+{
+ if (current->pid | read_pda(irqcount))
+ return;
+ __exit_idle();
+}
+
+/*
+ * On SMP it's slightly faster (but much more power-consuming!)
+ * to poll the ->need_resched flag instead of waiting for the
+ * cross-CPU IPI to arrive. Use this option with caution.
+ */
+static void poll_idle (void)
+{
+ local_irq_enable();
+
+ asm volatile(
+ "2:"
+ "testl %0,%1;"
+ "rep; nop;"
+ "je 2b;"
+ : :
+ "i" (_TIF_NEED_RESCHED),
+ "m" (current_thread_info()->flags));
+}
+
+static void xen_idle(void)
+{
+ local_irq_disable();
+
+ if (need_resched())
+ local_irq_enable();
+ else {
+ current_thread_info()->status &= ~TS_POLLING;
+ smp_mb__after_clear_bit();
+ safe_halt();
+ current_thread_info()->status |= TS_POLLING;
+ }
+}
+
+#ifdef CONFIG_HOTPLUG_CPU
+static inline void play_dead(void)
+{
+ idle_task_exit();
+ local_irq_disable();
+ cpu_clear(smp_processor_id(), cpu_initialized);
+ preempt_enable_no_resched();
+ VOID(HYPERVISOR_vcpu_op(VCPUOP_down, smp_processor_id(), NULL));
+ cpu_bringup();
+}
+#else
+static inline void play_dead(void)
+{
+ BUG();
+}
+#endif /* CONFIG_HOTPLUG_CPU */
+
+/*
+ * The idle thread. There's no useful work to be
+ * done, so just try to conserve power and have a
+ * low exit latency (ie sit in a loop waiting for
+ * somebody to say that they'd like to reschedule)
+ */
+void cpu_idle (void)
+{
+ current_thread_info()->status |= TS_POLLING;
+ /* endless idle loop with no priority at all */
+ while (1) {
+ while (!need_resched()) {
+ void (*idle)(void);
+
+ if (__get_cpu_var(cpu_idle_state))
+ __get_cpu_var(cpu_idle_state) = 0;
+ rmb();
+ idle = xen_idle; /* no alternatives */
+ if (cpu_is_offline(smp_processor_id()))
+ play_dead();
+ enter_idle();
+ idle();
+ __exit_idle();
+ }
+
+ preempt_enable_no_resched();
+ schedule();
+ preempt_disable();
+ }
+}
+
+void cpu_idle_wait(void)
+{
+ unsigned int cpu, this_cpu = get_cpu();
+ cpumask_t map;
+
+ set_cpus_allowed(current, cpumask_of_cpu(this_cpu));
+ put_cpu();
+
+ cpus_clear(map);
+ for_each_online_cpu(cpu) {
+ per_cpu(cpu_idle_state, cpu) = 1;
+ cpu_set(cpu, map);
+ }
+
+ __get_cpu_var(cpu_idle_state) = 0;
+
+ wmb();
+ do {
+ ssleep(1);
+ for_each_online_cpu(cpu) {
+ if (cpu_isset(cpu, map) &&
+ !per_cpu(cpu_idle_state, cpu))
+ cpu_clear(cpu, map);
+ }
+ cpus_and(map, map, cpu_online_map);
+ } while (!cpus_empty(map));
+}
+EXPORT_SYMBOL_GPL(cpu_idle_wait);
+
+void __cpuinit select_idle_routine(const struct cpuinfo_x86 *c)
+{
+}
+
+static int __init idle_setup (char *str)
+{
+ if (!strncmp(str, "poll", 4)) {
+ printk("using polling idle threads.\n");
+ pm_idle = poll_idle;
+ }
+
+ boot_option_idle_override = 1;
+ return 1;
+}
+
+__setup("idle=", idle_setup);
+
+/* Prints also some state that isn't saved in the pt_regs */
+void __show_regs(struct pt_regs * regs)
+{
+ unsigned long fs, gs, shadowgs;
+ unsigned int fsindex,gsindex;
+ unsigned int ds,cs,es;
+
+ printk("\n");
+ print_modules();
+ printk("Pid: %d, comm: %.20s %s %s %.*s\n",
+ current->pid, current->comm, print_tainted(),
+ system_utsname.release,
+ (int)strcspn(system_utsname.version, " "),
+ system_utsname.version);
+ printk("RIP: %04lx:[<%016lx>] ", regs->cs & 0xffff, regs->rip);
+ printk_address(regs->rip);
+ printk("RSP: %04lx:%016lx EFLAGS: %08lx\n", regs->ss, regs->rsp,
+ regs->eflags);
+ printk("RAX: %016lx RBX: %016lx RCX: %016lx\n",
+ regs->rax, regs->rbx, regs->rcx);
+ printk("RDX: %016lx RSI: %016lx RDI: %016lx\n",
+ regs->rdx, regs->rsi, regs->rdi);
+ printk("RBP: %016lx R08: %016lx R09: %016lx\n",
+ regs->rbp, regs->r8, regs->r9);
+ printk("R10: %016lx R11: %016lx R12: %016lx\n",
+ regs->r10, regs->r11, regs->r12);
+ printk("R13: %016lx R14: %016lx R15: %016lx\n",
+ regs->r13, regs->r14, regs->r15);
+
+ asm("mov %%ds,%0" : "=r" (ds));
+ asm("mov %%cs,%0" : "=r" (cs));
+ asm("mov %%es,%0" : "=r" (es));
+ asm("mov %%fs,%0" : "=r" (fsindex));
+ asm("mov %%gs,%0" : "=r" (gsindex));
+
+ rdmsrl(MSR_FS_BASE, fs);
+ rdmsrl(MSR_GS_BASE, gs);
+ rdmsrl(MSR_KERNEL_GS_BASE, shadowgs);
+
+ printk("FS: %016lx(%04x) GS:%016lx(%04x) knlGS:%016lx\n",
+ fs,fsindex,gs,gsindex,shadowgs);
+ printk("CS: %04x DS: %04x ES: %04x\n", cs, ds, es);
+
+}
+
+void show_regs(struct pt_regs *regs)
+{
+ printk("CPU %d:", smp_processor_id());
+ __show_regs(regs);
+ show_trace(NULL, regs, (void *)(regs + 1));
+}
+
+/*
+ * Free current thread data structures etc..
+ */
+void exit_thread(void)
+{
+ struct task_struct *me = current;
+ struct thread_struct *t = &me->thread;
+
+ if (me->thread.io_bitmap_ptr) {
+#ifndef CONFIG_X86_NO_TSS
+ struct tss_struct *tss = &per_cpu(init_tss, get_cpu());
+#endif
+#ifdef CONFIG_XEN
+ struct physdev_set_iobitmap iobmp_op;
+ memset(&iobmp_op, 0, sizeof(iobmp_op));
+#endif
+
+ kfree(t->io_bitmap_ptr);
+ t->io_bitmap_ptr = NULL;
+ /*
+ * Careful, clear this in the TSS too:
+ */
+#ifndef CONFIG_X86_NO_TSS
+ memset(tss->io_bitmap, 0xff, t->io_bitmap_max);
+ put_cpu();
+#endif
+#ifdef CONFIG_XEN
+ WARN_ON(HYPERVISOR_physdev_op(PHYSDEVOP_set_iobitmap,
+ &iobmp_op));
+#endif
+ t->io_bitmap_max = 0;
+ }
+}
+
+void load_gs_index(unsigned gs)
+{
+ WARN_ON(HYPERVISOR_set_segment_base(SEGBASE_GS_USER_SEL, gs));
+}
+
+void flush_thread(void)
+{
+ struct task_struct *tsk = current;
+ struct thread_info *t = current_thread_info();
+
+ if (t->flags & _TIF_ABI_PENDING) {
+ t->flags ^= (_TIF_ABI_PENDING | _TIF_IA32);
+ if (t->flags & _TIF_IA32)
+ current_thread_info()->status |= TS_COMPAT;
+ }
+
+ tsk->thread.debugreg0 = 0;
+ tsk->thread.debugreg1 = 0;
+ tsk->thread.debugreg2 = 0;
+ tsk->thread.debugreg3 = 0;
+ tsk->thread.debugreg6 = 0;
+ tsk->thread.debugreg7 = 0;
+ memset(tsk->thread.tls_array, 0, sizeof(tsk->thread.tls_array));
+ /*
+ * Forget coprocessor state..
+ */
+ clear_fpu(tsk);
+ clear_used_math();
+}
+
+void release_thread(struct task_struct *dead_task)
+{
+ if (dead_task->mm) {
+ if (dead_task->mm->context.size) {
+ printk("WARNING: dead process %8s still has LDT? <%p/%d>\n",
+ dead_task->comm,
+ dead_task->mm->context.ldt,
+ dead_task->mm->context.size);
+ BUG();
+ }
+ }
+}
+
+static inline void set_32bit_tls(struct task_struct *t, int tls, u32 addr)
+{
+ struct user_desc ud = {
+ .base_addr = addr,
+ .limit = 0xfffff,
+ .seg_32bit = 1,
+ .limit_in_pages = 1,
+ .useable = 1,
+ };
+ struct n_desc_struct *desc = (void *)t->thread.tls_array;
+ desc += tls;
+ desc->a = LDT_entry_a(&ud);
+ desc->b = LDT_entry_b(&ud);
+}
+
+static inline u32 read_32bit_tls(struct task_struct *t, int tls)
+{
+ struct desc_struct *desc = (void *)t->thread.tls_array;
+ desc += tls;
+ return desc->base0 |
+ (((u32)desc->base1) << 16) |
+ (((u32)desc->base2) << 24);
+}
+
+/*
+ * This gets called before we allocate a new thread and copy
+ * the current task into it.
+ */
+void prepare_to_copy(struct task_struct *tsk)
+{
+ unlazy_fpu(tsk);
+}
+
+int copy_thread(int nr, unsigned long clone_flags, unsigned long rsp,
+ unsigned long unused,
+ struct task_struct * p, struct pt_regs * regs)
+{
+ int err;
+ struct pt_regs * childregs;
+ struct task_struct *me = current;
+
+ childregs = ((struct pt_regs *)
+ (THREAD_SIZE + task_stack_page(p))) - 1;
+ *childregs = *regs;
+
+ childregs->rax = 0;
+ childregs->rsp = rsp;
+ if (rsp == ~0UL)
+ childregs->rsp = (unsigned long)childregs;
+
+ p->thread.rsp = (unsigned long) childregs;
+ p->thread.rsp0 = (unsigned long) (childregs+1);
+ p->thread.userrsp = me->thread.userrsp;
+
+ set_tsk_thread_flag(p, TIF_FORK);
+
+ p->thread.fs = me->thread.fs;
+ p->thread.gs = me->thread.gs;
+
+ asm("mov %%gs,%0" : "=m" (p->thread.gsindex));
+ asm("mov %%fs,%0" : "=m" (p->thread.fsindex));
+ asm("mov %%es,%0" : "=m" (p->thread.es));
+ asm("mov %%ds,%0" : "=m" (p->thread.ds));
+
+ if (unlikely(me->thread.io_bitmap_ptr != NULL)) {
+ p->thread.io_bitmap_ptr = kmalloc(IO_BITMAP_BYTES, GFP_KERNEL);
+ if (!p->thread.io_bitmap_ptr) {
+ p->thread.io_bitmap_max = 0;
+ return -ENOMEM;
+ }
+ memcpy(p->thread.io_bitmap_ptr, me->thread.io_bitmap_ptr,
+ IO_BITMAP_BYTES);
+ }
+
+ /*
+ * Set a new TLS for the child thread?
+ */
+ if (clone_flags & CLONE_SETTLS) {
+#ifdef CONFIG_IA32_EMULATION
+ if (test_thread_flag(TIF_IA32))
+ err = ia32_child_tls(p, childregs);
+ else
+#endif
+ err = do_arch_prctl(p, ARCH_SET_FS, childregs->r8);
+ if (err)
+ goto out;
+ }
+ p->thread.iopl = current->thread.iopl;
+
+ err = 0;
+out:
+ if (err && p->thread.io_bitmap_ptr) {
+ kfree(p->thread.io_bitmap_ptr);
+ p->thread.io_bitmap_max = 0;
+ }
+ return err;
+}
+
+static inline void __save_init_fpu( struct task_struct *tsk )
+{
+ asm volatile( "rex64 ; fxsave %0 ; fnclex"
+ : "=m" (tsk->thread.i387.fxsave));
+ tsk->thread_info->status &= ~TS_USEDFPU;
+}
+
+/*
+ * switch_to(x,y) should switch tasks from x to y.
+ *
+ * This could still be optimized:
+ * - fold all the options into a flag word and test it with a single test.
+ * - could test fs/gs bitsliced
+ *
+ * Kprobes not supported here. Set the probe on schedule instead.
+ */
+__kprobes struct task_struct *
+__switch_to(struct task_struct *prev_p, struct task_struct *next_p)
+{
+ struct thread_struct *prev = &prev_p->thread,
+ *next = &next_p->thread;
+ int cpu = smp_processor_id();
+#ifndef CONFIG_X86_NO_TSS
+ struct tss_struct *tss = &per_cpu(init_tss, cpu);
+#endif
+#if CONFIG_XEN_COMPAT > 0x030002
+ struct physdev_set_iopl iopl_op;
+ struct physdev_set_iobitmap iobmp_op;
+#else
+ struct physdev_op _pdo[2], *pdo = _pdo;
+#define iopl_op pdo->u.set_iopl
+#define iobmp_op pdo->u.set_iobitmap
+#endif
+ multicall_entry_t _mcl[8], *mcl = _mcl;
+
+ /*
+ * This is basically '__unlazy_fpu', except that we queue a
+ * multicall to indicate FPU task switch, rather than
+ * synchronously trapping to Xen.
+ * The AMD workaround requires it to be after DS reload, or
+ * after DS has been cleared, which we do in __prepare_arch_switch.
+ */
+ if (prev_p->thread_info->status & TS_USEDFPU) {
+ __save_init_fpu(prev_p); /* _not_ save_init_fpu() */
+ mcl->op = __HYPERVISOR_fpu_taskswitch;
+ mcl->args[0] = 1;
+ mcl++;
+ }
+
+ /*
+ * Reload esp0, LDT and the page table pointer:
+ */
+ mcl->op = __HYPERVISOR_stack_switch;
+ mcl->args[0] = __KERNEL_DS;
+ mcl->args[1] = next->rsp0;
+ mcl++;
+
+ /*
+ * Load the per-thread Thread-Local Storage descriptor.
+ * This is load_TLS(next, cpu) with multicalls.
+ */
+#define C(i) do { \
+ if (unlikely(next->tls_array[i] != prev->tls_array[i])) { \
+ mcl->op = __HYPERVISOR_update_descriptor; \
+ mcl->args[0] = virt_to_machine( \
+ &cpu_gdt(cpu)[GDT_ENTRY_TLS_MIN + i]); \
+ mcl->args[1] = next->tls_array[i]; \
+ mcl++; \
+ } \
+} while (0)
+ C(0); C(1); C(2);
+#undef C
+
+ if (unlikely(prev->iopl != next->iopl)) {
+ iopl_op.iopl = (next->iopl == 0) ? 1 : next->iopl;
+#if CONFIG_XEN_COMPAT > 0x030002
+ mcl->op = __HYPERVISOR_physdev_op;
+ mcl->args[0] = PHYSDEVOP_set_iopl;
+ mcl->args[1] = (unsigned long)&iopl_op;
+#else
+ mcl->op = __HYPERVISOR_physdev_op_compat;
+ pdo->cmd = PHYSDEVOP_set_iopl;
+ mcl->args[0] = (unsigned long)pdo++;
+#endif
+ mcl++;
+ }
+
+ if (unlikely(prev->io_bitmap_ptr || next->io_bitmap_ptr)) {
+ set_xen_guest_handle(iobmp_op.bitmap,
+ (char *)next->io_bitmap_ptr);
+ iobmp_op.nr_ports = next->io_bitmap_ptr ? IO_BITMAP_BITS : 0;
+#if CONFIG_XEN_COMPAT > 0x030002
+ mcl->op = __HYPERVISOR_physdev_op;
+ mcl->args[0] = PHYSDEVOP_set_iobitmap;
+ mcl->args[1] = (unsigned long)&iobmp_op;
+#else
+ mcl->op = __HYPERVISOR_physdev_op_compat;
+ pdo->cmd = PHYSDEVOP_set_iobitmap;
+ mcl->args[0] = (unsigned long)pdo++;
+#endif
+ mcl++;
+ }
+
+#if CONFIG_XEN_COMPAT <= 0x030002
+ BUG_ON(pdo > _pdo + ARRAY_SIZE(_pdo));
+#endif
+ BUG_ON(mcl > _mcl + ARRAY_SIZE(_mcl));
+ if (unlikely(HYPERVISOR_multicall_check(_mcl, mcl - _mcl, NULL)))
+ BUG();
+
+ /*
+ * Switch DS and ES.
+ * This won't pick up thread selector changes, but I guess that is ok.
+ */
+ if (unlikely(next->es))
+ loadsegment(es, next->es);
+
+ if (unlikely(next->ds))
+ loadsegment(ds, next->ds);
+
+ /*
+ * Switch FS and GS.
+ */
+ if (unlikely(next->fsindex))
+ loadsegment(fs, next->fsindex);
+
+ if (next->fs)
+ WARN_ON(HYPERVISOR_set_segment_base(SEGBASE_FS, next->fs));
+
+ if (unlikely(next->gsindex))
+ load_gs_index(next->gsindex);
+
+ if (next->gs)
+ WARN_ON(HYPERVISOR_set_segment_base(SEGBASE_GS_USER, next->gs));
+
+ /*
+ * Switch the PDA context.
+ */
+ prev->userrsp = read_pda(oldrsp);
+ write_pda(oldrsp, next->userrsp);
+ write_pda(pcurrent, next_p);
+ write_pda(kernelstack,
+ task_stack_page(next_p) + THREAD_SIZE - PDA_STACKOFFSET);
+
+ /*
+ * Now maybe reload the debug registers
+ */
+ if (unlikely(next->debugreg7)) {
+ set_debugreg(next->debugreg0, 0);
+ set_debugreg(next->debugreg1, 1);
+ set_debugreg(next->debugreg2, 2);
+ set_debugreg(next->debugreg3, 3);
+ /* no 4 and 5 */
+ set_debugreg(next->debugreg6, 6);
+ set_debugreg(next->debugreg7, 7);
+ }
+
+ return prev_p;
+}
+
+/*
+ * sys_execve() executes a new program.
+ */
+asmlinkage
+long sys_execve(char __user *name, char __user * __user *argv,
+ char __user * __user *envp, struct pt_regs regs)
+{
+ long error;
+ char * filename;
+
+ filename = getname(name);
+ error = PTR_ERR(filename);
+ if (IS_ERR(filename))
+ return error;
+ error = do_execve(filename, argv, envp, &regs);
+ if (error == 0) {
+ task_lock(current);
+ current->ptrace &= ~PT_DTRACE;
+ task_unlock(current);
+ }
+ putname(filename);
+ return error;
+}
+
+void set_personality_64bit(void)
+{
+ /* inherit personality from parent */
+
+ /* Make sure to be in 64bit mode */
+ clear_thread_flag(TIF_IA32);
+
+ /* TBD: overwrites user setup. Should have two bits.
+ But 64bit processes have always behaved this way,
+ so it's not too bad. The main problem is just that
+ 32bit childs are affected again. */
+ current->personality &= ~READ_IMPLIES_EXEC;
+}
+
+asmlinkage long sys_fork(struct pt_regs *regs)
+{
+ return do_fork(SIGCHLD, regs->rsp, regs, 0, NULL, NULL);
+}
+
+asmlinkage long
+sys_clone(unsigned long clone_flags, unsigned long newsp,
+ void __user *parent_tid, void __user *child_tid, struct pt_regs *regs)
+{
+ if (!newsp)
+ newsp = regs->rsp;
+ return do_fork(clone_flags, newsp, regs, 0, parent_tid, child_tid);
+}
+
+/*
+ * This is trivial, and on the face of it looks like it
+ * could equally well be done in user mode.
+ *
+ * Not so, for quite unobvious reasons - register pressure.
+ * In user mode vfork() cannot have a stack frame, and if
+ * done by calling the "clone()" system call directly, you
+ * do not have enough call-clobbered registers to hold all
+ * the information you need.
+ */
+asmlinkage long sys_vfork(struct pt_regs *regs)
+{
+ return do_fork(CLONE_VFORK | CLONE_VM | SIGCHLD, regs->rsp, regs, 0,
+ NULL, NULL);
+}
+
+unsigned long get_wchan(struct task_struct *p)
+{
+ unsigned long stack;
+ u64 fp,rip;
+ int count = 0;
+
+ if (!p || p == current || p->state==TASK_RUNNING)
+ return 0;
+ stack = (unsigned long)task_stack_page(p);
+ if (p->thread.rsp < stack || p->thread.rsp > stack+THREAD_SIZE)
+ return 0;
+ fp = *(u64 *)(p->thread.rsp);
+ do {
+ if (fp < (unsigned long)stack ||
+ fp > (unsigned long)stack+THREAD_SIZE)
+ return 0;
+ rip = *(u64 *)(fp+8);
+ if (!in_sched_functions(rip))
+ return rip;
+ fp = *(u64 *)fp;
+ } while (count++ < 16);
+ return 0;
+}
+
+long do_arch_prctl(struct task_struct *task, int code, unsigned long addr)
+{
+ int ret = 0;
+ int doit = task == current;
+ int cpu;
+
+ switch (code) {
+ case ARCH_SET_GS:
+ if (addr >= TASK_SIZE_OF(task))
+ return -EPERM;
+ cpu = get_cpu();
+ /* handle small bases via the GDT because that's faster to
+ switch. */
+ if (addr <= 0xffffffff) {
+ set_32bit_tls(task, GS_TLS, addr);
+ if (doit) {
+ load_TLS(&task->thread, cpu);
+ load_gs_index(GS_TLS_SEL);
+ }
+ task->thread.gsindex = GS_TLS_SEL;
+ task->thread.gs = 0;
+ } else {
+ task->thread.gsindex = 0;
+ task->thread.gs = addr;
+ if (doit) {
+ load_gs_index(0);
+ ret = HYPERVISOR_set_segment_base(
+ SEGBASE_GS_USER, addr);
+ }
+ }
+ put_cpu();
+ break;
+ case ARCH_SET_FS:
+ /* Not strictly needed for fs, but do it for symmetry
+ with gs */
+ if (addr >= TASK_SIZE_OF(task))
+ return -EPERM;
+ cpu = get_cpu();
+ /* handle small bases via the GDT because that's faster to
+ switch. */
+ if (addr <= 0xffffffff) {
+ set_32bit_tls(task, FS_TLS, addr);
+ if (doit) {
+ load_TLS(&task->thread, cpu);
+ asm volatile("movl %0,%%fs" :: "r"(FS_TLS_SEL));
+ }
+ task->thread.fsindex = FS_TLS_SEL;
+ task->thread.fs = 0;
+ } else {
+ task->thread.fsindex = 0;
+ task->thread.fs = addr;
+ if (doit) {
+ /* set the selector to 0 to not confuse
+ __switch_to */
+ asm volatile("movl %0,%%fs" :: "r" (0));
+ ret = HYPERVISOR_set_segment_base(SEGBASE_FS,
+ addr);
+ }
+ }
+ put_cpu();
+ break;
+ case ARCH_GET_FS: {
+ unsigned long base;
+ if (task->thread.fsindex == FS_TLS_SEL)
+ base = read_32bit_tls(task, FS_TLS);
+ else if (doit)
+ rdmsrl(MSR_FS_BASE, base);
+ else
+ base = task->thread.fs;
+ ret = put_user(base, (unsigned long __user *)addr);
+ break;
+ }
+ case ARCH_GET_GS: {
+ unsigned long base;
+ unsigned gsindex;
+ if (task->thread.gsindex == GS_TLS_SEL)
+ base = read_32bit_tls(task, GS_TLS);
+ else if (doit) {
+ asm("movl %%gs,%0" : "=r" (gsindex));
+ if (gsindex)
+ rdmsrl(MSR_KERNEL_GS_BASE, base);
+ else
+ base = task->thread.gs;
+ }
+ else
+ base = task->thread.gs;
+ ret = put_user(base, (unsigned long __user *)addr);
+ break;
+ }
+
+ default:
+ ret = -EINVAL;
+ break;
+ }
+
+ return ret;
+}
+
+long sys_arch_prctl(int code, unsigned long addr)
+{
+ return do_arch_prctl(current, code, addr);
+}
+
+/*
+ * Capture the user space registers if the task is not running (in user space)
+ */
+int dump_task_regs(struct task_struct *tsk, elf_gregset_t *regs)
+{
+ struct pt_regs *pp, ptregs;
+
+ pp = task_pt_regs(tsk);
+
+ ptregs = *pp;
+ ptregs.cs &= 0xffff;
+ ptregs.ss &= 0xffff;
+
+ elf_core_copy_regs(regs, &ptregs);
+
+ boot_option_idle_override = 1;
+ return 1;
+}
+
+unsigned long arch_align_stack(unsigned long sp)
+{
+ if (randomize_va_space)
+ sp -= get_random_int() % 8192;
+ return sp & ~0xf;
+}
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/kernel/setup_64-xen.c 2010-10-05 09:58:12.000000000 +0200
@@ -0,0 +1,1656 @@
+/*
+ * linux/arch/x86-64/kernel/setup.c
+ *
+ * Copyright (C) 1995 Linus Torvalds
+ *
+ * Nov 2001 Dave Jones <davej@suse.de>
+ * Forked from i386 setup code.
+ */
+
+/*
+ * This file handles the architecture-dependent parts of initialization
+ */
+
+#include <linux/errno.h>
+#include <linux/sched.h>
+#include <linux/kernel.h>
+#include <linux/mm.h>
+#include <linux/stddef.h>
+#include <linux/unistd.h>
+#include <linux/ptrace.h>
+#include <linux/slab.h>
+#include <linux/user.h>
+#include <linux/a.out.h>
+#include <linux/screen_info.h>
+#include <linux/ioport.h>
+#include <linux/delay.h>
+#include <linux/init.h>
+#include <linux/initrd.h>
+#include <linux/highmem.h>
+#include <linux/bootmem.h>
+#include <linux/module.h>
+#include <asm/processor.h>
+#include <linux/console.h>
+#include <linux/seq_file.h>
+#include <linux/crash_dump.h>
+#include <linux/root_dev.h>
+#include <linux/pci.h>
+#include <linux/acpi.h>
+#include <linux/kallsyms.h>
+#include <linux/edd.h>
+#include <linux/mmzone.h>
+#include <linux/kexec.h>
+#include <linux/cpufreq.h>
+#include <linux/dmi.h>
+#include <linux/dma-mapping.h>
+#include <linux/ctype.h>
+
+#include <asm/mtrr.h>
+#include <asm/uaccess.h>
+#include <asm/system.h>
+#include <asm/io.h>
+#include <asm/smp.h>
+#include <asm/msr.h>
+#include <asm/desc.h>
+#include <video/edid.h>
+#include <asm/e820.h>
+#include <asm/dma.h>
+#include <asm/mpspec.h>
+#include <asm/mmu_context.h>
+#include <asm/bootsetup.h>
+#include <asm/proto.h>
+#include <asm/setup.h>
+#include <asm/mach_apic.h>
+#include <asm/numa.h>
+#include <asm/sections.h>
+#include <asm/dmi.h>
+#ifdef CONFIG_XEN
+#include <linux/percpu.h>
+#include <xen/interface/physdev.h>
+#include "setup_arch_pre.h"
+#include <asm/hypervisor.h>
+#include <xen/interface/nmi.h>
+#include <xen/features.h>
+#include <xen/firmware.h>
+#include <xen/xencons.h>
+#define PFN_UP(x) (((x) + PAGE_SIZE-1) >> PAGE_SHIFT)
+#define PFN_PHYS(x) ((x) << PAGE_SHIFT)
+#include <asm/mach-xen/setup_arch_post.h>
+#include <xen/interface/memory.h>
+
+#ifdef CONFIG_XEN
+#include <xen/interface/kexec.h>
+#endif
+
+extern unsigned long start_pfn;
+extern struct edid_info edid_info;
+
+shared_info_t *HYPERVISOR_shared_info = (shared_info_t *)empty_zero_page;
+EXPORT_SYMBOL(HYPERVISOR_shared_info);
+
+extern char hypercall_page[PAGE_SIZE];
+EXPORT_SYMBOL(hypercall_page);
+
+static int xen_panic_event(struct notifier_block *, unsigned long, void *);
+static struct notifier_block xen_panic_block = {
+ xen_panic_event, NULL, 0 /* try to go last */
+};
+
+unsigned long *phys_to_machine_mapping;
+unsigned long *pfn_to_mfn_frame_list_list, *pfn_to_mfn_frame_list[512];
+
+EXPORT_SYMBOL(phys_to_machine_mapping);
+
+DEFINE_PER_CPU(multicall_entry_t, multicall_list[8]);
+DEFINE_PER_CPU(int, nr_multicall_ents);
+
+/* Raw start-of-day parameters from the hypervisor. */
+start_info_t *xen_start_info;
+EXPORT_SYMBOL(xen_start_info);
+#endif
+
+/*
+ * Machine setup..
+ */
+
+struct cpuinfo_x86 boot_cpu_data __read_mostly;
+EXPORT_SYMBOL(boot_cpu_data);
+
+unsigned long mmu_cr4_features;
+
+int acpi_disabled;
+EXPORT_SYMBOL(acpi_disabled);
+#ifdef CONFIG_ACPI
+extern int __initdata acpi_ht;
+extern acpi_interrupt_flags acpi_sci_flags;
+int __initdata acpi_force = 0;
+#endif
+
+int acpi_numa __initdata;
+
+/* Boot loader ID as an integer, for the benefit of proc_dointvec */
+int bootloader_type;
+
+unsigned long saved_video_mode;
+
+/*
+ * Early DMI memory
+ */
+int dmi_alloc_index;
+char dmi_alloc_data[DMI_MAX_DATA];
+
+/*
+ * Setup options
+ */
+struct screen_info screen_info;
+EXPORT_SYMBOL(screen_info);
+struct sys_desc_table_struct {
+ unsigned short length;
+ unsigned char table[0];
+};
+
+struct edid_info edid_info;
+EXPORT_SYMBOL_GPL(edid_info);
+struct e820map e820;
+#ifdef CONFIG_XEN
+struct e820map machine_e820;
+#endif
+
+extern int root_mountflags;
+
+char command_line[COMMAND_LINE_SIZE];
+
+struct resource standard_io_resources[] = {
+ { .name = "dma1", .start = 0x00, .end = 0x1f,
+ .flags = IORESOURCE_BUSY | IORESOURCE_IO },
+ { .name = "pic1", .start = 0x20, .end = 0x21,
+ .flags = IORESOURCE_BUSY | IORESOURCE_IO },
+ { .name = "timer0", .start = 0x40, .end = 0x43,
+ .flags = IORESOURCE_BUSY | IORESOURCE_IO },
+ { .name = "timer1", .start = 0x50, .end = 0x53,
+ .flags = IORESOURCE_BUSY | IORESOURCE_IO },
+ { .name = "keyboard", .start = 0x60, .end = 0x6f,
+ .flags = IORESOURCE_BUSY | IORESOURCE_IO },
+ { .name = "dma page reg", .start = 0x80, .end = 0x8f,
+ .flags = IORESOURCE_BUSY | IORESOURCE_IO },
+ { .name = "pic2", .start = 0xa0, .end = 0xa1,
+ .flags = IORESOURCE_BUSY | IORESOURCE_IO },
+ { .name = "dma2", .start = 0xc0, .end = 0xdf,
+ .flags = IORESOURCE_BUSY | IORESOURCE_IO },
+ { .name = "fpu", .start = 0xf0, .end = 0xff,
+ .flags = IORESOURCE_BUSY | IORESOURCE_IO }
+};
+
+#define STANDARD_IO_RESOURCES \
+ (sizeof standard_io_resources / sizeof standard_io_resources[0])
+
+#define IORESOURCE_RAM (IORESOURCE_BUSY | IORESOURCE_MEM)
+
+struct resource data_resource = {
+ .name = "Kernel data",
+ .start = 0,
+ .end = 0,
+ .flags = IORESOURCE_RAM,
+};
+struct resource code_resource = {
+ .name = "Kernel code",
+ .start = 0,
+ .end = 0,
+ .flags = IORESOURCE_RAM,
+};
+
+#define IORESOURCE_ROM (IORESOURCE_BUSY | IORESOURCE_READONLY | IORESOURCE_MEM)
+
+static struct resource system_rom_resource = {
+ .name = "System ROM",
+ .start = 0xf0000,
+ .end = 0xfffff,
+ .flags = IORESOURCE_ROM,
+};
+
+static struct resource extension_rom_resource = {
+ .name = "Extension ROM",
+ .start = 0xe0000,
+ .end = 0xeffff,
+ .flags = IORESOURCE_ROM,
+};
+
+static struct resource adapter_rom_resources[] = {
+ { .name = "Adapter ROM", .start = 0xc8000, .end = 0,
+ .flags = IORESOURCE_ROM },
+ { .name = "Adapter ROM", .start = 0, .end = 0,
+ .flags = IORESOURCE_ROM },
+ { .name = "Adapter ROM", .start = 0, .end = 0,
+ .flags = IORESOURCE_ROM },
+ { .name = "Adapter ROM", .start = 0, .end = 0,
+ .flags = IORESOURCE_ROM },
+ { .name = "Adapter ROM", .start = 0, .end = 0,
+ .flags = IORESOURCE_ROM },
+ { .name = "Adapter ROM", .start = 0, .end = 0,
+ .flags = IORESOURCE_ROM }
+};
+
+#define ADAPTER_ROM_RESOURCES \
+ (sizeof adapter_rom_resources / sizeof adapter_rom_resources[0])
+
+static struct resource video_rom_resource = {
+ .name = "Video ROM",
+ .start = 0xc0000,
+ .end = 0xc7fff,
+ .flags = IORESOURCE_ROM,
+};
+
+static struct resource video_ram_resource = {
+ .name = "Video RAM area",
+ .start = 0xa0000,
+ .end = 0xbffff,
+ .flags = IORESOURCE_RAM,
+};
+
+#define romsignature(x) (*(unsigned short *)(x) == 0xaa55)
+
+static int __init romchecksum(unsigned char *rom, unsigned long length)
+{
+ unsigned char *p, sum = 0;
+
+ for (p = rom; p < rom + length; p++)
+ sum += *p;
+ return sum == 0;
+}
+
+static void __init probe_roms(void)
+{
+ unsigned long start, length, upper;
+ unsigned char *rom;
+ int i;
+
+#ifdef CONFIG_XEN
+ /* Nothing to do if not running in dom0. */
+ if (!is_initial_xendomain())
+ return;
+#endif
+
+ /* video rom */
+ upper = adapter_rom_resources[0].start;
+ for (start = video_rom_resource.start; start < upper; start += 2048) {
+ rom = isa_bus_to_virt(start);
+ if (!romsignature(rom))
+ continue;
+
+ video_rom_resource.start = start;
+
+ /* 0 < length <= 0x7f * 512, historically */
+ length = rom[2] * 512;
+
+ /* if checksum okay, trust length byte */
+ if (length && romchecksum(rom, length))
+ video_rom_resource.end = start + length - 1;
+
+ request_resource(&iomem_resource, &video_rom_resource);
+ break;
+ }
+
+ start = (video_rom_resource.end + 1 + 2047) & ~2047UL;
+ if (start < upper)
+ start = upper;
+
+ /* system rom */
+ request_resource(&iomem_resource, &system_rom_resource);
+ upper = system_rom_resource.start;
+
+ /* check for extension rom (ignore length byte!) */
+ rom = isa_bus_to_virt(extension_rom_resource.start);
+ if (romsignature(rom)) {
+ length = extension_rom_resource.end - extension_rom_resource.start + 1;
+ if (romchecksum(rom, length)) {
+ request_resource(&iomem_resource, &extension_rom_resource);
+ upper = extension_rom_resource.start;
+ }
+ }
+
+ /* check for adapter roms on 2k boundaries */
+ for (i = 0; i < ADAPTER_ROM_RESOURCES && start < upper; start += 2048) {
+ rom = isa_bus_to_virt(start);
+ if (!romsignature(rom))
+ continue;
+
+ /* 0 < length <= 0x7f * 512, historically */
+ length = rom[2] * 512;
+
+ /* but accept any length that fits if checksum okay */
+ if (!length || start + length > upper || !romchecksum(rom, length))
+ continue;
+
+ adapter_rom_resources[i].start = start;
+ adapter_rom_resources[i].end = start + length - 1;
+ request_resource(&iomem_resource, &adapter_rom_resources[i]);
+
+ start = adapter_rom_resources[i++].end & ~2047UL;
+ }
+}
+
+/* Check for full argument with no trailing characters */
+static int fullarg(char *p, char *arg)
+{
+ int l = strlen(arg);
+ return !memcmp(p, arg, l) && (p[l] == 0 || isspace(p[l]));
+}
+
+static __init void parse_cmdline_early (char ** cmdline_p)
+{
+ char c = ' ', *to = command_line, *from = COMMAND_LINE;
+ int len = 0;
+ int userdef = 0;
+
+ for (;;) {
+ if (c != ' ')
+ goto next_char;
+
+#ifdef CONFIG_SMP
+ /*
+ * If the BIOS enumerates physical processors before logical,
+ * maxcpus=N at enumeration-time can be used to disable HT.
+ */
+ else if (!memcmp(from, "maxcpus=", 8)) {
+ extern unsigned int maxcpus;
+
+ maxcpus = simple_strtoul(from + 8, NULL, 0);
+ }
+#endif
+#ifdef CONFIG_ACPI
+ /* "acpi=off" disables both ACPI table parsing and interpreter init */
+ if (fullarg(from,"acpi=off"))
+ disable_acpi();
+
+ if (fullarg(from, "acpi=force")) {
+ /* add later when we do DMI horrors: */
+ acpi_force = 1;
+ acpi_disabled = 0;
+ }
+
+ /* acpi=ht just means: do ACPI MADT parsing
+ at bootup, but don't enable the full ACPI interpreter */
+ if (fullarg(from, "acpi=ht")) {
+ if (!acpi_force)
+ disable_acpi();
+ acpi_ht = 1;
+ }
+ else if (fullarg(from, "pci=noacpi"))
+ acpi_disable_pci();
+ else if (fullarg(from, "acpi=noirq"))
+ acpi_noirq_set();
+
+ else if (fullarg(from, "acpi_sci=edge"))
+ acpi_sci_flags.trigger = 1;
+ else if (fullarg(from, "acpi_sci=level"))
+ acpi_sci_flags.trigger = 3;
+ else if (fullarg(from, "acpi_sci=high"))
+ acpi_sci_flags.polarity = 1;
+ else if (fullarg(from, "acpi_sci=low"))
+ acpi_sci_flags.polarity = 3;
+
+ /* acpi=strict disables out-of-spec workarounds */
+ else if (fullarg(from, "acpi=strict")) {
+ acpi_strict = 1;
+ }
+#ifdef CONFIG_X86_IO_APIC
+ else if (fullarg(from, "acpi_skip_timer_override"))
+ acpi_skip_timer_override = 1;
+#endif
+#endif
+
+#ifndef CONFIG_XEN
+ if (fullarg(from, "nolapic") || fullarg(from, "disableapic")) {
+ clear_bit(X86_FEATURE_APIC, boot_cpu_data.x86_capability);
+ disable_apic = 1;
+ }
+
+ if (fullarg(from, "noapic"))
+ skip_ioapic_setup = 1;
+
+ if (fullarg(from,"apic")) {
+ skip_ioapic_setup = 0;
+ ioapic_force = 1;
+ }
+#endif
+
+ if (!memcmp(from, "mem=", 4))
+ parse_memopt(from+4, &from);
+
+ if (!memcmp(from, "memmap=", 7)) {
+ /* exactmap option is for used defined memory */
+ if (!memcmp(from+7, "exactmap", 8)) {
+#ifdef CONFIG_CRASH_DUMP
+ /* If we are doing a crash dump, we
+ * still need to know the real mem
+ * size before original memory map is
+ * reset.
+ */
+ saved_max_pfn = e820_end_of_ram();
+#endif
+ from += 8+7;
+ end_pfn_map = 0;
+ e820.nr_map = 0;
+ userdef = 1;
+ }
+ else {
+ parse_memmapopt(from+7, &from);
+ userdef = 1;
+ }
+ }
+
+#ifdef CONFIG_NUMA
+ if (!memcmp(from, "numa=", 5))
+ numa_setup(from+5);
+#endif
+
+ if (!memcmp(from,"iommu=",6)) {
+ iommu_setup(from+6);
+ }
+
+ if (fullarg(from,"oops=panic"))
+ panic_on_oops = 1;
+
+ if (!memcmp(from, "noexec=", 7))
+ nonx_setup(from + 7);
+
+#ifdef CONFIG_KEXEC
+ /* crashkernel=size@addr specifies the location to reserve for
+ * a crash kernel. By reserving this memory we guarantee
+ * that linux never set's it up as a DMA target.
+ * Useful for holding code to do something appropriate
+ * after a kernel panic.
+ */
+ else if (!memcmp(from, "crashkernel=", 12)) {
+#ifndef CONFIG_XEN
+ unsigned long size, base;
+ size = memparse(from+12, &from);
+ if (*from == '@') {
+ base = memparse(from+1, &from);
+ /* FIXME: Do I want a sanity check
+ * to validate the memory range?
+ */
+ crashk_res.start = base;
+ crashk_res.end = base + size - 1;
+ }
+#else
+ printk("Ignoring crashkernel command line, "
+ "parameter will be supplied by xen\n");
+#endif
+ }
+#endif
+
+#ifdef CONFIG_PROC_VMCORE
+ /* elfcorehdr= specifies the location of elf core header
+ * stored by the crashed kernel. This option will be passed
+ * by kexec loader to the capture kernel.
+ */
+ else if(!memcmp(from, "elfcorehdr=", 11))
+ elfcorehdr_addr = memparse(from+11, &from);
+#endif
+
+#if defined(CONFIG_HOTPLUG_CPU) && !defined(CONFIG_XEN)
+ else if (!memcmp(from, "additional_cpus=", 16))
+ setup_additional_cpus(from+16);
+#endif
+
+ next_char:
+ c = *(from++);
+ if (!c)
+ break;
+ if (COMMAND_LINE_SIZE <= ++len)
+ break;
+ *(to++) = c;
+ }
+ if (userdef) {
+ printk(KERN_INFO "user-defined physical RAM map:\n");
+ e820_print_map("user");
+ }
+ *to = '\0';
+ *cmdline_p = command_line;
+}
+
+#ifndef CONFIG_NUMA
+static void __init
+contig_initmem_init(unsigned long start_pfn, unsigned long end_pfn)
+{
+ unsigned long bootmap_size, bootmap;
+
+ bootmap_size = bootmem_bootmap_pages(end_pfn)<<PAGE_SHIFT;
+ bootmap = find_e820_area(0, end_pfn<<PAGE_SHIFT, bootmap_size);
+ if (bootmap == -1L)
+ panic("Cannot find bootmem map of size %ld\n",bootmap_size);
+ bootmap_size = init_bootmem(bootmap >> PAGE_SHIFT, end_pfn);
+#ifdef CONFIG_XEN
+ if (xen_start_info->nr_pages < end_pfn)
+ e820_bootmem_free(NODE_DATA(0), 0,
+ xen_start_info->nr_pages<<PAGE_SHIFT);
+ else
+#endif
+ e820_bootmem_free(NODE_DATA(0), 0, end_pfn << PAGE_SHIFT);
+ reserve_bootmem(bootmap, bootmap_size);
+}
+#endif
+
+#if defined(CONFIG_EDD) || defined(CONFIG_EDD_MODULE)
+struct edd edd;
+#ifdef CONFIG_EDD_MODULE
+EXPORT_SYMBOL(edd);
+#endif
+#ifndef CONFIG_XEN
+/**
+ * copy_edd() - Copy the BIOS EDD information
+ * from boot_params into a safe place.
+ *
+ */
+static inline void copy_edd(void)
+{
+ memcpy(edd.mbr_signature, EDD_MBR_SIGNATURE, sizeof(edd.mbr_signature));
+ memcpy(edd.edd_info, EDD_BUF, sizeof(edd.edd_info));
+ edd.mbr_signature_nr = EDD_MBR_SIG_NR;
+ edd.edd_info_nr = EDD_NR;
+}
+#endif
+#else
+static inline void copy_edd(void)
+{
+}
+#endif
+
+#ifndef CONFIG_XEN
+#define EBDA_ADDR_POINTER 0x40E
+
+unsigned __initdata ebda_addr;
+unsigned __initdata ebda_size;
+
+static void discover_ebda(void)
+{
+ /*
+ * there is a real-mode segmented pointer pointing to the
+ * 4K EBDA area at 0x40E
+ */
+ ebda_addr = *(unsigned short *)EBDA_ADDR_POINTER;
+ ebda_addr <<= 4;
+
+ ebda_size = *(unsigned short *)(unsigned long)ebda_addr;
+
+ /* Round EBDA up to pages */
+ if (ebda_size == 0)
+ ebda_size = 1;
+ ebda_size <<= 10;
+ ebda_size = round_up(ebda_size + (ebda_addr & ~PAGE_MASK), PAGE_SIZE);
+ if (ebda_size > 64*1024)
+ ebda_size = 64*1024;
+}
+#else
+#define discover_ebda() ((void)0)
+#endif
+
+void __init setup_arch(char **cmdline_p)
+{
+#ifdef CONFIG_XEN
+ /* Register a call for panic conditions. */
+ atomic_notifier_chain_register(&panic_notifier_list, &xen_panic_block);
+
+ ROOT_DEV = MKDEV(RAMDISK_MAJOR,0);
+ screen_info = SCREEN_INFO;
+
+ if (is_initial_xendomain()) {
+ const struct dom0_vga_console_info *info =
+ (void *)((char *)xen_start_info +
+ xen_start_info->console.dom0.info_off);
+
+ dom0_init_screen_info(info,
+ xen_start_info->console.dom0.info_size);
+ xen_start_info->console.domU.mfn = 0;
+ xen_start_info->console.domU.evtchn = 0;
+ } else
+ screen_info.orig_video_isVGA = 0;
+
+ copy_edid();
+
+ WARN_ON(HYPERVISOR_vm_assist(VMASST_CMD_enable,
+ VMASST_TYPE_writable_pagetables));
+
+ ARCH_SETUP
+#else
+ ROOT_DEV = old_decode_dev(ORIG_ROOT_DEV);
+ screen_info = SCREEN_INFO;
+ edid_info = EDID_INFO;
+#endif /* !CONFIG_XEN */
+ saved_video_mode = SAVED_VIDEO_MODE;
+ bootloader_type = LOADER_TYPE;
+
+#ifdef CONFIG_BLK_DEV_RAM
+ rd_image_start = RAMDISK_FLAGS & RAMDISK_IMAGE_START_MASK;
+ rd_prompt = ((RAMDISK_FLAGS & RAMDISK_PROMPT_FLAG) != 0);
+ rd_doload = ((RAMDISK_FLAGS & RAMDISK_LOAD_FLAG) != 0);
+#endif
+ setup_memory_region();
+ copy_edd();
+
+ if (!MOUNT_ROOT_RDONLY)
+ root_mountflags &= ~MS_RDONLY;
+ init_mm.start_code = (unsigned long) &_text;
+ init_mm.end_code = (unsigned long) &_etext;
+ init_mm.end_data = (unsigned long) &_edata;
+ init_mm.brk = (unsigned long) &_end;
+
+ code_resource.start = virt_to_phys(&_text);
+ code_resource.end = virt_to_phys(&_etext)-1;
+ data_resource.start = virt_to_phys(&_etext);
+ data_resource.end = virt_to_phys(&_edata)-1;
+
+ parse_cmdline_early(cmdline_p);
+
+ early_identify_cpu(&boot_cpu_data);
+
+ /*
+ * partially used pages are not usable - thus
+ * we are rounding upwards:
+ */
+ end_pfn = e820_end_of_ram();
+ num_physpages = end_pfn; /* for pfn_valid */
+
+ check_efer();
+
+ discover_ebda();
+
+ init_memory_mapping(0, (end_pfn_map << PAGE_SHIFT));
+
+ if (is_initial_xendomain())
+ dmi_scan_machine();
+
+#ifdef CONFIG_ACPI_NUMA
+ /*
+ * Parse SRAT to discover nodes.
+ */
+ acpi_numa_init();
+#endif
+
+#ifdef CONFIG_NUMA
+ numa_initmem_init(0, end_pfn);
+#else
+ contig_initmem_init(0, end_pfn);
+#endif
+
+#ifdef CONFIG_XEN
+ /*
+ * Reserve kernel, physmap, start info, initial page tables, and
+ * direct mapping.
+ */
+ reserve_bootmem_generic(__pa_symbol(&_text),
+ (table_end << PAGE_SHIFT) - __pa_symbol(&_text));
+#else
+ /* Reserve direct mapping */
+ reserve_bootmem_generic(table_start << PAGE_SHIFT,
+ (table_end - table_start) << PAGE_SHIFT);
+
+ /* reserve kernel */
+ reserve_bootmem_generic(__pa_symbol(&_text),
+ __pa_symbol(&_end) - __pa_symbol(&_text));
+
+ /*
+ * reserve physical page 0 - it's a special BIOS page on many boxes,
+ * enabling clean reboots, SMP operation, laptop functions.
+ */
+ reserve_bootmem_generic(0, PAGE_SIZE);
+
+ /* reserve ebda region */
+ if (ebda_addr)
+ reserve_bootmem_generic(ebda_addr, ebda_size);
+
+#ifdef CONFIG_SMP
+ /*
+ * But first pinch a few for the stack/trampoline stuff
+ * FIXME: Don't need the extra page at 4K, but need to fix
+ * trampoline before removing it. (see the GDT stuff)
+ */
+ reserve_bootmem_generic(PAGE_SIZE, PAGE_SIZE);
+
+ /* Reserve SMP trampoline */
+ reserve_bootmem_generic(SMP_TRAMPOLINE_BASE, PAGE_SIZE);
+#endif
+#endif
+
+#ifdef CONFIG_ACPI_SLEEP
+ /*
+ * Reserve low memory region for sleep support.
+ */
+ acpi_reserve_bootmem();
+#endif
+#ifdef CONFIG_XEN
+#ifdef CONFIG_BLK_DEV_INITRD
+ if (xen_start_info->mod_start) {
+ if (INITRD_START + INITRD_SIZE <= (end_pfn << PAGE_SHIFT)) {
+ /*reserve_bootmem_generic(INITRD_START, INITRD_SIZE);*/
+ initrd_start = INITRD_START + PAGE_OFFSET;
+ initrd_end = initrd_start+INITRD_SIZE;
+ initrd_below_start_ok = 1;
+ } else {
+ printk(KERN_ERR "initrd extends beyond end of memory "
+ "(0x%08lx > 0x%08lx)\ndisabling initrd\n",
+ (unsigned long)(INITRD_START + INITRD_SIZE),
+ (unsigned long)(end_pfn << PAGE_SHIFT));
+ initrd_start = 0;
+ }
+ }
+#endif
+#else /* CONFIG_XEN */
+#ifdef CONFIG_BLK_DEV_INITRD
+ if (LOADER_TYPE && INITRD_START) {
+ if (INITRD_START + INITRD_SIZE <= (end_pfn << PAGE_SHIFT)) {
+ reserve_bootmem_generic(INITRD_START, INITRD_SIZE);
+ initrd_start =
+ INITRD_START ? INITRD_START + PAGE_OFFSET : 0;
+ initrd_end = initrd_start+INITRD_SIZE;
+ }
+ else {
+ printk(KERN_ERR "initrd extends beyond end of memory "
+ "(0x%08lx > 0x%08lx)\ndisabling initrd\n",
+ (unsigned long)(INITRD_START + INITRD_SIZE),
+ (unsigned long)(end_pfn << PAGE_SHIFT));
+ initrd_start = 0;
+ }
+ }
+#endif
+#endif /* !CONFIG_XEN */
+#ifdef CONFIG_KEXEC
+#ifdef CONFIG_XEN
+ xen_machine_kexec_setup_resources();
+#else
+ if (crashk_res.start != crashk_res.end) {
+ reserve_bootmem_generic(crashk_res.start,
+ crashk_res.end - crashk_res.start + 1);
+ }
+#endif
+#endif
+
+ paging_init();
+#ifdef CONFIG_X86_LOCAL_APIC
+ /*
+ * Find and reserve possible boot-time SMP configuration:
+ */
+ find_smp_config();
+#endif
+#ifdef CONFIG_XEN
+ {
+ int i, j, k, fpp;
+ unsigned long p2m_pages;
+
+ p2m_pages = end_pfn;
+ if (xen_start_info->nr_pages > end_pfn) {
+ /*
+ * the end_pfn was shrunk (probably by mem= or highmem=
+ * kernel parameter); shrink reservation with the HV
+ */
+ struct xen_memory_reservation reservation = {
+ .address_bits = 0,
+ .extent_order = 0,
+ .domid = DOMID_SELF
+ };
+ unsigned int difference;
+ int ret;
+
+ difference = xen_start_info->nr_pages - end_pfn;
+
+ set_xen_guest_handle(reservation.extent_start,
+ ((unsigned long *)xen_start_info->mfn_list) + end_pfn);
+ reservation.nr_extents = difference;
+ ret = HYPERVISOR_memory_op(XENMEM_decrease_reservation,
+ &reservation);
+ BUG_ON (ret != difference);
+ }
+ else if (end_pfn > xen_start_info->nr_pages)
+ p2m_pages = xen_start_info->nr_pages;
+
+ if (!xen_feature(XENFEAT_auto_translated_physmap)) {
+ /* Make sure we have a large enough P->M table. */
+ phys_to_machine_mapping = alloc_bootmem_pages(
+ end_pfn * sizeof(unsigned long));
+ memset(phys_to_machine_mapping, ~0,
+ end_pfn * sizeof(unsigned long));
+ memcpy(phys_to_machine_mapping,
+ (unsigned long *)xen_start_info->mfn_list,
+ p2m_pages * sizeof(unsigned long));
+ free_bootmem(
+ __pa(xen_start_info->mfn_list),
+ PFN_PHYS(PFN_UP(xen_start_info->nr_pages *
+ sizeof(unsigned long))));
+
+ /*
+ * Initialise the list of the frames that specify the
+ * list of frames that make up the p2m table. Used by
+ * save/restore.
+ */
+ pfn_to_mfn_frame_list_list = alloc_bootmem_pages(PAGE_SIZE);
+
+ fpp = PAGE_SIZE/sizeof(unsigned long);
+ for (i=0, j=0, k=-1; i< end_pfn; i+=fpp, j++) {
+ if ((j % fpp) == 0) {
+ k++;
+ BUG_ON(k>=fpp);
+ pfn_to_mfn_frame_list[k] =
+ alloc_bootmem_pages(PAGE_SIZE);
+ pfn_to_mfn_frame_list_list[k] =
+ virt_to_mfn(pfn_to_mfn_frame_list[k]);
+ j=0;
+ }
+ pfn_to_mfn_frame_list[k][j] =
+ virt_to_mfn(&phys_to_machine_mapping[i]);
+ }
+ HYPERVISOR_shared_info->arch.max_pfn = end_pfn;
+ HYPERVISOR_shared_info->arch.pfn_to_mfn_frame_list_list =
+ virt_to_mfn(pfn_to_mfn_frame_list_list);
+ }
+
+ /* Mark all ISA DMA channels in-use - using them wouldn't work. */
+ for (i = 0; i < MAX_DMA_CHANNELS; ++i)
+ if (i != 4 && request_dma(i, "xen") != 0)
+ BUG();
+ }
+
+ if (!is_initial_xendomain()) {
+ acpi_disabled = 1;
+#ifdef CONFIG_ACPI
+ acpi_ht = 0;
+#endif
+ }
+#endif
+
+#ifndef CONFIG_XEN
+ check_ioapic();
+#endif
+
+ zap_low_mappings(0);
+
+ /*
+ * set this early, so we dont allocate cpu0
+ * if MADT list doesnt list BSP first
+ * mpparse.c/MP_processor_info() allocates logical cpu numbers.
+ */
+ cpu_set(0, cpu_present_map);
+#ifdef CONFIG_ACPI
+ /*
+ * Initialize the ACPI boot-time table parser (gets the RSDP and SDT).
+ * Call this early for SRAT node setup.
+ */
+ acpi_boot_table_init();
+
+ /*
+ * Read APIC and some other early information from ACPI tables.
+ */
+ acpi_boot_init();
+#endif
+
+ init_cpu_to_node();
+
+#ifdef CONFIG_X86_LOCAL_APIC
+ /*
+ * get boot-time SMP configuration:
+ */
+ if (smp_found_config)
+ get_smp_config();
+#ifndef CONFIG_XEN
+ init_apic_mappings();
+#endif
+#endif
+#if defined(CONFIG_XEN) && defined(CONFIG_SMP) && !defined(CONFIG_HOTPLUG_CPU)
+ prefill_possible_map();
+#endif
+
+ /*
+ * Request address space for all standard RAM and ROM resources
+ * and also for regions reported as reserved by the e820.
+ */
+ probe_roms();
+#ifdef CONFIG_XEN
+ if (is_initial_xendomain())
+ e820_reserve_resources(machine_e820.map, machine_e820.nr_map);
+#else
+ e820_reserve_resources(e820.map, e820.nr_map);
+#endif
+
+ request_resource(&iomem_resource, &video_ram_resource);
+
+ {
+ unsigned i;
+ /* request I/O space for devices used on all i[345]86 PCs */
+ for (i = 0; i < STANDARD_IO_RESOURCES; i++)
+ request_resource(&ioport_resource, &standard_io_resources[i]);
+ }
+
+#ifdef CONFIG_XEN
+ if (is_initial_xendomain())
+ e820_setup_gap(machine_e820.map, machine_e820.nr_map);
+#else
+ e820_setup_gap(e820.map, e820.nr_map);
+#endif
+
+#ifdef CONFIG_XEN
+ {
+ struct physdev_set_iopl set_iopl;
+
+ set_iopl.iopl = 1;
+ WARN_ON(HYPERVISOR_physdev_op(PHYSDEVOP_set_iopl, &set_iopl));
+
+ if (is_initial_xendomain()) {
+#ifdef CONFIG_VT
+#if defined(CONFIG_VGA_CONSOLE)
+ conswitchp = &vga_con;
+#elif defined(CONFIG_DUMMY_CONSOLE)
+ conswitchp = &dummy_con;
+#endif
+#endif
+ } else {
+#if defined(CONFIG_VT) && defined(CONFIG_DUMMY_CONSOLE)
+ conswitchp = &dummy_con;
+#endif
+ }
+ }
+#else /* CONFIG_XEN */
+
+#ifdef CONFIG_VT
+#if defined(CONFIG_VGA_CONSOLE)
+ conswitchp = &vga_con;
+#elif defined(CONFIG_DUMMY_CONSOLE)
+ conswitchp = &dummy_con;
+#endif
+#endif
+
+#endif /* !CONFIG_XEN */
+}
+
+#ifdef CONFIG_XEN
+static int
+xen_panic_event(struct notifier_block *this, unsigned long event, void *ptr)
+{
+ HYPERVISOR_shutdown(SHUTDOWN_crash);
+ /* we're never actually going to get here... */
+ return NOTIFY_DONE;
+}
+#endif /* !CONFIG_XEN */
+
+
+static int __cpuinit get_model_name(struct cpuinfo_x86 *c)
+{
+ unsigned int *v;
+
+ if (c->extended_cpuid_level < 0x80000004)
+ return 0;
+
+ v = (unsigned int *) c->x86_model_id;
+ cpuid(0x80000002, &v[0], &v[1], &v[2], &v[3]);
+ cpuid(0x80000003, &v[4], &v[5], &v[6], &v[7]);
+ cpuid(0x80000004, &v[8], &v[9], &v[10], &v[11]);
+ c->x86_model_id[48] = 0;
+ return 1;
+}
+
+
+static void __cpuinit display_cacheinfo(struct cpuinfo_x86 *c)
+{
+ unsigned int n, dummy, eax, ebx, ecx, edx;
+
+ n = c->extended_cpuid_level;
+
+ if (n >= 0x80000005) {
+ cpuid(0x80000005, &dummy, &ebx, &ecx, &edx);
+ printk(KERN_INFO "CPU: L1 I Cache: %dK (%d bytes/line), D cache %dK (%d bytes/line)\n",
+ edx>>24, edx&0xFF, ecx>>24, ecx&0xFF);
+ c->x86_cache_size=(ecx>>24)+(edx>>24);
+ /* On K8 L1 TLB is inclusive, so don't count it */
+ c->x86_tlbsize = 0;
+ }
+
+ if (n >= 0x80000006) {
+ cpuid(0x80000006, &dummy, &ebx, &ecx, &edx);
+ ecx = cpuid_ecx(0x80000006);
+ c->x86_cache_size = ecx >> 16;
+ c->x86_tlbsize += ((ebx >> 16) & 0xfff) + (ebx & 0xfff);
+
+ printk(KERN_INFO "CPU: L2 Cache: %dK (%d bytes/line)\n",
+ c->x86_cache_size, ecx & 0xFF);
+ }
+
+ if (n >= 0x80000007)
+ cpuid(0x80000007, &dummy, &dummy, &dummy, &c->x86_power);
+ if (n >= 0x80000008) {
+ cpuid(0x80000008, &eax, &dummy, &dummy, &dummy);
+ c->x86_virt_bits = (eax >> 8) & 0xff;
+ c->x86_phys_bits = eax & 0xff;
+ }
+}
+
+#ifdef CONFIG_NUMA
+static int nearby_node(int apicid)
+{
+ int i;
+ for (i = apicid - 1; i >= 0; i--) {
+ int node = apicid_to_node[i];
+ if (node != NUMA_NO_NODE && node_online(node))
+ return node;
+ }
+ for (i = apicid + 1; i < MAX_LOCAL_APIC; i++) {
+ int node = apicid_to_node[i];
+ if (node != NUMA_NO_NODE && node_online(node))
+ return node;
+ }
+ return first_node(node_online_map); /* Shouldn't happen */
+}
+#endif
+
+/*
+ * On a AMD dual core setup the lower bits of the APIC id distingush the cores.
+ * Assumes number of cores is a power of two.
+ */
+static void __init amd_detect_cmp(struct cpuinfo_x86 *c)
+{
+#if defined(CONFIG_SMP) && !defined(CONFIG_XEN)
+ unsigned bits;
+#ifdef CONFIG_NUMA
+ int cpu = smp_processor_id();
+ int node = 0;
+ unsigned apicid = hard_smp_processor_id();
+#endif
+ unsigned ecx = cpuid_ecx(0x80000008);
+
+ c->x86_max_cores = (ecx & 0xff) + 1;
+
+ /* CPU telling us the core id bits shift? */
+ bits = (ecx >> 12) & 0xF;
+
+ /* Otherwise recompute */
+ if (bits == 0) {
+ while ((1 << bits) < c->x86_max_cores)
+ bits++;
+ }
+
+ /* Low order bits define the core id (index of core in socket) */
+ c->cpu_core_id = c->phys_proc_id & ((1 << bits)-1);
+ /* Convert the APIC ID into the socket ID */
+ c->phys_proc_id = phys_pkg_id(bits);
+
+#ifdef CONFIG_NUMA
+ node = c->phys_proc_id;
+ if (apicid_to_node[apicid] != NUMA_NO_NODE)
+ node = apicid_to_node[apicid];
+ if (!node_online(node)) {
+ /* Two possibilities here:
+ - The CPU is missing memory and no node was created.
+ In that case try picking one from a nearby CPU
+ - The APIC IDs differ from the HyperTransport node IDs
+ which the K8 northbridge parsing fills in.
+ Assume they are all increased by a constant offset,
+ but in the same order as the HT nodeids.
+ If that doesn't result in a usable node fall back to the
+ path for the previous case. */
+ int ht_nodeid = apicid - (cpu_data[0].phys_proc_id << bits);
+ if (ht_nodeid >= 0 &&
+ apicid_to_node[ht_nodeid] != NUMA_NO_NODE)
+ node = apicid_to_node[ht_nodeid];
+ /* Pick a nearby node */
+ if (!node_online(node))
+ node = nearby_node(apicid);
+ }
+ numa_set_node(cpu, node);
+
+ printk(KERN_INFO "CPU %d/%x -> Node %d\n", cpu, apicid, node);
+#endif
+#endif
+}
+
+static void __init init_amd(struct cpuinfo_x86 *c)
+{
+ unsigned level;
+
+#ifdef CONFIG_SMP
+ unsigned long value;
+
+ /*
+ * Disable TLB flush filter by setting HWCR.FFDIS on K8
+ * bit 6 of msr C001_0015
+ *
+ * Errata 63 for SH-B3 steppings
+ * Errata 122 for all steppings (F+ have it disabled by default)
+ */
+ if (c->x86 == 15) {
+ rdmsrl(MSR_K8_HWCR, value);
+ value |= 1 << 6;
+ wrmsrl(MSR_K8_HWCR, value);
+ }
+#endif
+
+ /* Bit 31 in normal CPUID used for nonstandard 3DNow ID;
+ 3DNow is IDd by bit 31 in extended CPUID (1*32+31) anyway */
+ clear_bit(0*32+31, &c->x86_capability);
+
+ /* On C+ stepping K8 rep microcode works well for copy/memset */
+ level = cpuid_eax(1);
+ if (c->x86 == 15 && ((level >= 0x0f48 && level < 0x0f50) || level >= 0x0f58))
+ set_bit(X86_FEATURE_REP_GOOD, &c->x86_capability);
+
+ /* Enable workaround for FXSAVE leak */
+ if (c->x86 >= 6)
+ set_bit(X86_FEATURE_FXSAVE_LEAK, &c->x86_capability);
+
+ level = get_model_name(c);
+ if (!level) {
+ switch (c->x86) {
+ case 15:
+ /* Should distinguish Models here, but this is only
+ a fallback anyways. */
+ strcpy(c->x86_model_id, "Hammer");
+ break;
+ }
+ }
+ display_cacheinfo(c);
+
+ /* c->x86_power is 8000_0007 edx. Bit 8 is constant TSC */
+ if (c->x86_power & (1<<8))
+ set_bit(X86_FEATURE_CONSTANT_TSC, &c->x86_capability);
+
+ /* Multi core CPU? */
+ if (c->extended_cpuid_level >= 0x80000008)
+ amd_detect_cmp(c);
+
+ /* Fix cpuid4 emulation for more */
+ num_cache_leaves = 3;
+}
+
+static void __cpuinit detect_ht(struct cpuinfo_x86 *c)
+{
+#if defined(CONFIG_SMP) && !defined(CONFIG_XEN)
+ u32 eax, ebx, ecx, edx;
+ int index_msb, core_bits;
+
+ cpuid(1, &eax, &ebx, &ecx, &edx);
+
+
+ if (!cpu_has(c, X86_FEATURE_HT))
+ return;
+ if (cpu_has(c, X86_FEATURE_CMP_LEGACY))
+ goto out;
+
+ smp_num_siblings = (ebx & 0xff0000) >> 16;
+
+ if (smp_num_siblings == 1) {
+ printk(KERN_INFO "CPU: Hyper-Threading is disabled\n");
+ } else if (smp_num_siblings > 1 ) {
+
+ if (smp_num_siblings > NR_CPUS) {
+ printk(KERN_WARNING "CPU: Unsupported number of the siblings %d", smp_num_siblings);
+ smp_num_siblings = 1;
+ return;
+ }
+
+ index_msb = get_count_order(smp_num_siblings);
+ c->phys_proc_id = phys_pkg_id(index_msb);
+
+ smp_num_siblings = smp_num_siblings / c->x86_max_cores;
+
+ index_msb = get_count_order(smp_num_siblings) ;
+
+ core_bits = get_count_order(c->x86_max_cores);
+
+ c->cpu_core_id = phys_pkg_id(index_msb) &
+ ((1 << core_bits) - 1);
+ }
+out:
+ if ((c->x86_max_cores * smp_num_siblings) > 1) {
+ printk(KERN_INFO "CPU: Physical Processor ID: %d\n", c->phys_proc_id);
+ printk(KERN_INFO "CPU: Processor Core ID: %d\n", c->cpu_core_id);
+ }
+
+#endif
+}
+
+/*
+ * find out the number of processor cores on the die
+ */
+static int __cpuinit intel_num_cpu_cores(struct cpuinfo_x86 *c)
+{
+ unsigned int eax, t;
+
+ if (c->cpuid_level < 4)
+ return 1;
+
+ cpuid_count(4, 0, &eax, &t, &t, &t);
+
+ if (eax & 0x1f)
+ return ((eax >> 26) + 1);
+ else
+ return 1;
+}
+
+static void srat_detect_node(void)
+{
+#ifdef CONFIG_NUMA
+ unsigned node;
+ int cpu = smp_processor_id();
+ int apicid = hard_smp_processor_id();
+
+ /* Don't do the funky fallback heuristics the AMD version employs
+ for now. */
+ node = apicid_to_node[apicid];
+ if (node == NUMA_NO_NODE)
+ node = first_node(node_online_map);
+ numa_set_node(cpu, node);
+
+ if (acpi_numa > 0)
+ printk(KERN_INFO "CPU %d/%x -> Node %d\n", cpu, apicid, node);
+#endif
+}
+
+static void __cpuinit init_intel(struct cpuinfo_x86 *c)
+{
+ /* Cache sizes */
+ unsigned n;
+
+ init_intel_cacheinfo(c);
+ if (c->cpuid_level > 9 ) {
+ unsigned eax = cpuid_eax(10);
+ /* Check for version and the number of counters */
+ if ((eax & 0xff) && (((eax>>8) & 0xff) > 1))
+ set_bit(X86_FEATURE_ARCH_PERFMON, &c->x86_capability);
+ }
+
+ n = c->extended_cpuid_level;
+ if (n >= 0x80000008) {
+ unsigned eax = cpuid_eax(0x80000008);
+ c->x86_virt_bits = (eax >> 8) & 0xff;
+ c->x86_phys_bits = eax & 0xff;
+ /* CPUID workaround for Intel 0F34 CPU */
+ if (c->x86_vendor == X86_VENDOR_INTEL &&
+ c->x86 == 0xF && c->x86_model == 0x3 &&
+ c->x86_mask == 0x4)
+ c->x86_phys_bits = 36;
+ }
+
+ if (c->x86 == 15)
+ c->x86_cache_alignment = c->x86_clflush_size * 2;
+ if ((c->x86 == 0xf && c->x86_model >= 0x03) ||
+ (c->x86 == 0x6 && c->x86_model >= 0x0e))
+ set_bit(X86_FEATURE_CONSTANT_TSC, &c->x86_capability);
+ set_bit(X86_FEATURE_SYNC_RDTSC, &c->x86_capability);
+ c->x86_max_cores = intel_num_cpu_cores(c);
+
+ srat_detect_node();
+}
+
+static void __cpuinit get_cpu_vendor(struct cpuinfo_x86 *c)
+{
+ char *v = c->x86_vendor_id;
+
+ if (!strcmp(v, "AuthenticAMD"))
+ c->x86_vendor = X86_VENDOR_AMD;
+ else if (!strcmp(v, "GenuineIntel"))
+ c->x86_vendor = X86_VENDOR_INTEL;
+ else
+ c->x86_vendor = X86_VENDOR_UNKNOWN;
+}
+
+struct cpu_model_info {
+ int vendor;
+ int family;
+ char *model_names[16];
+};
+
+/* Do some early cpuid on the boot CPU to get some parameter that are
+ needed before check_bugs. Everything advanced is in identify_cpu
+ below. */
+void __cpuinit early_identify_cpu(struct cpuinfo_x86 *c)
+{
+ u32 tfms;
+
+ c->loops_per_jiffy = loops_per_jiffy;
+ c->x86_cache_size = -1;
+ c->x86_vendor = X86_VENDOR_UNKNOWN;
+ c->x86_model = c->x86_mask = 0; /* So far unknown... */
+ c->x86_vendor_id[0] = '\0'; /* Unset */
+ c->x86_model_id[0] = '\0'; /* Unset */
+ c->x86_clflush_size = 64;
+ c->x86_cache_alignment = c->x86_clflush_size;
+ c->x86_max_cores = 1;
+ c->extended_cpuid_level = 0;
+ memset(&c->x86_capability, 0, sizeof c->x86_capability);
+
+ /* Get vendor name */
+ cpuid(0x00000000, (unsigned int *)&c->cpuid_level,
+ (unsigned int *)&c->x86_vendor_id[0],
+ (unsigned int *)&c->x86_vendor_id[8],
+ (unsigned int *)&c->x86_vendor_id[4]);
+
+ get_cpu_vendor(c);
+
+ /* Initialize the standard set of capabilities */
+ /* Note that the vendor-specific code below might override */
+
+ /* Intel-defined flags: level 0x00000001 */
+ if (c->cpuid_level >= 0x00000001) {
+ __u32 misc;
+ cpuid(0x00000001, &tfms, &misc, &c->x86_capability[4],
+ &c->x86_capability[0]);
+ c->x86 = (tfms >> 8) & 0xf;
+ c->x86_model = (tfms >> 4) & 0xf;
+ c->x86_mask = tfms & 0xf;
+ if (c->x86 == 0xf)
+ c->x86 += (tfms >> 20) & 0xff;
+ if (c->x86 >= 0x6)
+ c->x86_model += ((tfms >> 16) & 0xF) << 4;
+ if (c->x86_capability[0] & (1<<19))
+ c->x86_clflush_size = ((misc >> 8) & 0xff) * 8;
+ } else {
+ /* Have CPUID level 0 only - unheard of */
+ c->x86 = 4;
+ }
+
+#if defined(CONFIG_SMP) && !defined(CONFIG_XEN)
+ c->phys_proc_id = (cpuid_ebx(1) >> 24) & 0xff;
+#endif
+}
+
+/*
+ * This does the hard work of actually picking apart the CPU stuff...
+ */
+void __cpuinit identify_cpu(struct cpuinfo_x86 *c)
+{
+ int i;
+ u32 xlvl;
+
+ early_identify_cpu(c);
+
+ /* AMD-defined flags: level 0x80000001 */
+ xlvl = cpuid_eax(0x80000000);
+ c->extended_cpuid_level = xlvl;
+ if ((xlvl & 0xffff0000) == 0x80000000) {
+ if (xlvl >= 0x80000001) {
+ c->x86_capability[1] = cpuid_edx(0x80000001);
+ c->x86_capability[6] = cpuid_ecx(0x80000001);
+ }
+ if (xlvl >= 0x80000004)
+ get_model_name(c); /* Default name */
+ }
+
+ /* Transmeta-defined flags: level 0x80860001 */
+ xlvl = cpuid_eax(0x80860000);
+ if ((xlvl & 0xffff0000) == 0x80860000) {
+ /* Don't set x86_cpuid_level here for now to not confuse. */
+ if (xlvl >= 0x80860001)
+ c->x86_capability[2] = cpuid_edx(0x80860001);
+ }
+
+#ifndef CONFIG_XEN
+ c->apicid = phys_pkg_id(0);
+#endif
+
+ /*
+ * Vendor-specific initialization. In this section we
+ * canonicalize the feature flags, meaning if there are
+ * features a certain CPU supports which CPUID doesn't
+ * tell us, CPUID claiming incorrect flags, or other bugs,
+ * we handle them here.
+ *
+ * At the end of this section, c->x86_capability better
+ * indicate the features this CPU genuinely supports!
+ */
+ switch (c->x86_vendor) {
+ case X86_VENDOR_AMD:
+ init_amd(c);
+ break;
+
+ case X86_VENDOR_INTEL:
+ init_intel(c);
+ break;
+
+ case X86_VENDOR_UNKNOWN:
+ default:
+ display_cacheinfo(c);
+ break;
+ }
+
+ select_idle_routine(c);
+ detect_ht(c);
+
+ /*
+ * On SMP, boot_cpu_data holds the common feature set between
+ * all CPUs; so make sure that we indicate which features are
+ * common between the CPUs. The first time this routine gets
+ * executed, c == &boot_cpu_data.
+ */
+ if (c != &boot_cpu_data) {
+ /* AND the already accumulated flags with these */
+ for (i = 0 ; i < NCAPINTS ; i++)
+ boot_cpu_data.x86_capability[i] &= c->x86_capability[i];
+ }
+
+#ifdef CONFIG_X86_MCE
+ mcheck_init(c);
+#endif
+ if (c == &boot_cpu_data)
+ mtrr_bp_init();
+ else
+ mtrr_ap_init();
+#ifdef CONFIG_NUMA
+ numa_add_cpu(smp_processor_id());
+#endif
+}
+
+
+void __cpuinit print_cpu_info(struct cpuinfo_x86 *c)
+{
+ if (c->x86_model_id[0])
+ printk("%s", c->x86_model_id);
+
+ if (c->x86_mask || c->cpuid_level >= 0)
+ printk(" stepping %02x\n", c->x86_mask);
+ else
+ printk("\n");
+}
+
+/*
+ * Get CPU information for use by the procfs.
+ */
+
+static int show_cpuinfo(struct seq_file *m, void *v)
+{
+ struct cpuinfo_x86 *c = v;
+
+ /*
+ * These flag bits must match the definitions in <asm/cpufeature.h>.
+ * NULL means this bit is undefined or reserved; either way it doesn't
+ * have meaning as far as Linux is concerned. Note that it's important
+ * to realize there is a difference between this table and CPUID -- if
+ * applications want to get the raw CPUID data, they should access
+ * /dev/cpu/<cpu_nr>/cpuid instead.
+ */
+ static char *x86_cap_flags[] = {
+ /* Intel-defined */
+ "fpu", "vme", "de", "pse", "tsc", "msr", "pae", "mce",
+ "cx8", "apic", NULL, "sep", "mtrr", "pge", "mca", "cmov",
+ "pat", "pse36", "pn", "clflush", NULL, "dts", "acpi", "mmx",
+ "fxsr", "sse", "sse2", "ss", "ht", "tm", "ia64", NULL,
+
+ /* AMD-defined */
+ NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
+ NULL, NULL, NULL, "syscall", NULL, NULL, NULL, NULL,
+ NULL, NULL, NULL, NULL, "nx", NULL, "mmxext", NULL,
+ NULL, "fxsr_opt", NULL, "rdtscp", NULL, "lm", "3dnowext", "3dnow",
+
+ /* Transmeta-defined */
+ "recovery", "longrun", NULL, "lrti", NULL, NULL, NULL, NULL,
+ NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
+ NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
+ NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
+
+ /* Other (Linux-defined) */
+ "cxmmx", NULL, "cyrix_arr", "centaur_mcr", NULL,
+ "constant_tsc", NULL, NULL,
+ "up", NULL, NULL, NULL, NULL, NULL, NULL, NULL,
+ NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
+ NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
+
+ /* Intel-defined (#2) */
+ "pni", NULL, NULL, "monitor", "ds_cpl", "vmx", "smx", "est",
+ "tm2", NULL, "cid", NULL, NULL, "cx16", "xtpr", NULL,
+ NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
+ NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
+
+ /* VIA/Cyrix/Centaur-defined */
+ NULL, NULL, "rng", "rng_en", NULL, NULL, "ace", "ace_en",
+ NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
+ NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
+ NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
+
+ /* AMD-defined (#2) */
+ "lahf_lm", "cmp_legacy", "svm", NULL, "cr8_legacy", NULL, NULL, NULL,
+ NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
+ NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
+ NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
+ };
+ static char *x86_power_flags[] = {
+ "ts", /* temperature sensor */
+ "fid", /* frequency id control */
+ "vid", /* voltage id control */
+ "ttp", /* thermal trip */
+ "tm",
+ "stc",
+ NULL,
+ /* nothing */ /* constant_tsc - moved to flags */
+ };
+
+
+#ifdef CONFIG_SMP
+ if (!cpu_online(c-cpu_data))
+ return 0;
+#endif
+
+ seq_printf(m,"processor\t: %u\n"
+ "vendor_id\t: %s\n"
+ "cpu family\t: %d\n"
+ "model\t\t: %d\n"
+ "model name\t: %s\n",
+ (unsigned)(c-cpu_data),
+ c->x86_vendor_id[0] ? c->x86_vendor_id : "unknown",
+ c->x86,
+ (int)c->x86_model,
+ c->x86_model_id[0] ? c->x86_model_id : "unknown");
+
+ if (c->x86_mask || c->cpuid_level >= 0)
+ seq_printf(m, "stepping\t: %d\n", c->x86_mask);
+ else
+ seq_printf(m, "stepping\t: unknown\n");
+
+ if (cpu_has(c,X86_FEATURE_TSC)) {
+ unsigned int freq = cpufreq_quick_get((unsigned)(c-cpu_data));
+ if (!freq)
+ freq = cpu_khz;
+ seq_printf(m, "cpu MHz\t\t: %u.%03u\n",
+ freq / 1000, (freq % 1000));
+ }
+
+ /* Cache size */
+ if (c->x86_cache_size >= 0)
+ seq_printf(m, "cache size\t: %d KB\n", c->x86_cache_size);
+
+#if defined(CONFIG_SMP) && !defined(CONFIG_XEN)
+ if (smp_num_siblings * c->x86_max_cores > 1) {
+ int cpu = c - cpu_data;
+ seq_printf(m, "physical id\t: %d\n", c->phys_proc_id);
+ seq_printf(m, "siblings\t: %d\n", cpus_weight(cpu_core_map[cpu]));
+ seq_printf(m, "core id\t\t: %d\n", c->cpu_core_id);
+ seq_printf(m, "cpu cores\t: %d\n", c->booted_cores);
+ }
+#endif
+
+ seq_printf(m,
+ "fpu\t\t: yes\n"
+ "fpu_exception\t: yes\n"
+ "cpuid level\t: %d\n"
+ "wp\t\t: yes\n"
+ "flags\t\t:",
+ c->cpuid_level);
+
+ {
+ int i;
+ for ( i = 0 ; i < 32*NCAPINTS ; i++ )
+ if (cpu_has(c, i) && x86_cap_flags[i] != NULL)
+ seq_printf(m, " %s", x86_cap_flags[i]);
+ }
+
+ seq_printf(m, "\nbogomips\t: %lu.%02lu\n",
+ c->loops_per_jiffy/(500000/HZ),
+ (c->loops_per_jiffy/(5000/HZ)) % 100);
+
+ if (c->x86_tlbsize > 0)
+ seq_printf(m, "TLB size\t: %d 4K pages\n", c->x86_tlbsize);
+ seq_printf(m, "clflush size\t: %d\n", c->x86_clflush_size);
+ seq_printf(m, "cache_alignment\t: %d\n", c->x86_cache_alignment);
+
+ seq_printf(m, "address sizes\t: %u bits physical, %u bits virtual\n",
+ c->x86_phys_bits, c->x86_virt_bits);
+
+ seq_printf(m, "power management:");
+ {
+ unsigned i;
+ for (i = 0; i < 32; i++)
+ if (c->x86_power & (1 << i)) {
+ if (i < ARRAY_SIZE(x86_power_flags) &&
+ x86_power_flags[i])
+ seq_printf(m, "%s%s",
+ x86_power_flags[i][0]?" ":"",
+ x86_power_flags[i]);
+ else
+ seq_printf(m, " [%d]", i);
+ }
+ }
+
+ seq_printf(m, "\n\n");
+
+ return 0;
+}
+
+static void *c_start(struct seq_file *m, loff_t *pos)
+{
+ return *pos < NR_CPUS ? cpu_data + *pos : NULL;
+}
+
+static void *c_next(struct seq_file *m, void *v, loff_t *pos)
+{
+ ++*pos;
+ return c_start(m, pos);
+}
+
+static void c_stop(struct seq_file *m, void *v)
+{
+}
+
+struct seq_operations cpuinfo_op = {
+ .start =c_start,
+ .next = c_next,
+ .stop = c_stop,
+ .show = show_cpuinfo,
+};
+
+#if defined(CONFIG_INPUT_PCSPKR) || defined(CONFIG_INPUT_PCSPKR_MODULE)
+#include <linux/platform_device.h>
+static __init int add_pcspkr(void)
+{
+ struct platform_device *pd;
+ int ret;
+
+ if (!is_initial_xendomain())
+ return 0;
+
+ pd = platform_device_alloc("pcspkr", -1);
+ if (!pd)
+ return -ENOMEM;
+
+ ret = platform_device_add(pd);
+ if (ret)
+ platform_device_put(pd);
+
+ return ret;
+}
+device_initcall(add_pcspkr);
+#endif
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/kernel/setup64-xen.c 2008-01-28 12:24:19.000000000 +0100
@@ -0,0 +1,367 @@
+/*
+ * X86-64 specific CPU setup.
+ * Copyright (C) 1995 Linus Torvalds
+ * Copyright 2001, 2002, 2003 SuSE Labs / Andi Kleen.
+ * See setup.c for older changelog.
+ *
+ * Jun Nakajima <jun.nakajima@intel.com>
+ * Modified for Xen
+ *
+ */
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/sched.h>
+#include <linux/string.h>
+#include <linux/bootmem.h>
+#include <linux/bitops.h>
+#include <linux/module.h>
+#include <asm/bootsetup.h>
+#include <asm/pda.h>
+#include <asm/pgtable.h>
+#include <asm/processor.h>
+#include <asm/desc.h>
+#include <asm/atomic.h>
+#include <asm/mmu_context.h>
+#include <asm/smp.h>
+#include <asm/i387.h>
+#include <asm/percpu.h>
+#include <asm/proto.h>
+#include <asm/sections.h>
+#ifdef CONFIG_XEN
+#include <asm/hypervisor.h>
+#endif
+
+char x86_boot_params[BOOT_PARAM_SIZE] __initdata = {0,};
+
+cpumask_t cpu_initialized __cpuinitdata = CPU_MASK_NONE;
+
+struct x8664_pda *_cpu_pda[NR_CPUS] __read_mostly;
+EXPORT_SYMBOL(_cpu_pda);
+struct x8664_pda boot_cpu_pda[NR_CPUS] __cacheline_aligned;
+
+#ifndef CONFIG_X86_NO_IDT
+struct desc_ptr idt_descr = { 256 * 16 - 1, (unsigned long) idt_table };
+#endif
+
+char boot_cpu_stack[IRQSTACKSIZE] __attribute__((section(".bss.page_aligned")));
+
+unsigned long __supported_pte_mask __read_mostly = ~0UL;
+EXPORT_SYMBOL(__supported_pte_mask);
+static int do_not_nx __cpuinitdata = 0;
+
+/* noexec=on|off
+Control non executable mappings for 64bit processes.
+
+on Enable(default)
+off Disable
+*/
+int __init nonx_setup(char *str)
+{
+ if (!strncmp(str, "on", 2)) {
+ __supported_pte_mask |= _PAGE_NX;
+ do_not_nx = 0;
+ } else if (!strncmp(str, "off", 3)) {
+ do_not_nx = 1;
+ __supported_pte_mask &= ~_PAGE_NX;
+ }
+ return 1;
+}
+__setup("noexec=", nonx_setup); /* parsed early actually */
+
+int force_personality32 = 0;
+
+/* noexec32=on|off
+Control non executable heap for 32bit processes.
+To control the stack too use noexec=off
+
+on PROT_READ does not imply PROT_EXEC for 32bit processes
+off PROT_READ implies PROT_EXEC (default)
+*/
+static int __init nonx32_setup(char *str)
+{
+ if (!strcmp(str, "on"))
+ force_personality32 &= ~READ_IMPLIES_EXEC;
+ else if (!strcmp(str, "off"))
+ force_personality32 |= READ_IMPLIES_EXEC;
+ return 1;
+}
+__setup("noexec32=", nonx32_setup);
+
+/*
+ * Great future plan:
+ * Declare PDA itself and support (irqstack,tss,pgd) as per cpu data.
+ * Always point %gs to its beginning
+ */
+void __init setup_per_cpu_areas(void)
+{
+ int i;
+ unsigned long size;
+
+#ifdef CONFIG_HOTPLUG_CPU
+ prefill_possible_map();
+#endif
+
+ /* Copy section for each CPU (we discard the original) */
+ size = ALIGN(__per_cpu_end - __per_cpu_start, SMP_CACHE_BYTES);
+#ifdef CONFIG_MODULES
+ if (size < PERCPU_ENOUGH_ROOM)
+ size = PERCPU_ENOUGH_ROOM;
+#endif
+
+ for_each_cpu_mask (i, cpu_possible_map) {
+ char *ptr;
+
+ if (!NODE_DATA(cpu_to_node(i))) {
+ printk("cpu with no node %d, num_online_nodes %d\n",
+ i, num_online_nodes());
+ ptr = alloc_bootmem(size);
+ } else {
+ ptr = alloc_bootmem_node(NODE_DATA(cpu_to_node(i)), size);
+ }
+ if (!ptr)
+ panic("Cannot allocate cpu data for CPU %d\n", i);
+ cpu_pda(i)->data_offset = ptr - __per_cpu_start;
+ memcpy(ptr, __per_cpu_start, __per_cpu_end - __per_cpu_start);
+ }
+}
+
+#ifdef CONFIG_XEN
+static void switch_pt(void)
+{
+ xen_pt_switch(__pa_symbol(init_level4_pgt));
+ xen_new_user_pt(__pa_symbol(__user_pgd(init_level4_pgt)));
+}
+
+static void __cpuinit cpu_gdt_init(const struct desc_ptr *gdt_descr)
+{
+ unsigned long frames[16];
+ unsigned long va;
+ int f;
+
+ for (va = gdt_descr->address, f = 0;
+ va < gdt_descr->address + gdt_descr->size;
+ va += PAGE_SIZE, f++) {
+ frames[f] = virt_to_mfn(va);
+ make_page_readonly(
+ (void *)va, XENFEAT_writable_descriptor_tables);
+ }
+ if (HYPERVISOR_set_gdt(frames, (gdt_descr->size + 1) /
+ sizeof (struct desc_struct)))
+ BUG();
+}
+#else
+static void switch_pt(void)
+{
+ asm volatile("movq %0,%%cr3" :: "r" (__pa_symbol(&init_level4_pgt)));
+}
+
+static void __cpuinit cpu_gdt_init(const struct desc_ptr *gdt_descr)
+{
+ asm volatile("lgdt %0" :: "m" (*gdt_descr));
+ asm volatile("lidt %0" :: "m" (idt_descr));
+}
+#endif
+
+void pda_init(int cpu)
+{
+ struct x8664_pda *pda = cpu_pda(cpu);
+
+ /* Setup up data that may be needed in __get_free_pages early */
+ asm volatile("movl %0,%%fs ; movl %0,%%gs" :: "r" (0));
+#ifndef CONFIG_XEN
+ wrmsrl(MSR_GS_BASE, pda);
+#else
+ if (HYPERVISOR_set_segment_base(SEGBASE_GS_KERNEL,
+ (unsigned long)pda))
+ BUG();
+#endif
+ pda->cpunumber = cpu;
+ pda->irqcount = -1;
+ pda->kernelstack =
+ (unsigned long)stack_thread_info() - PDA_STACKOFFSET + THREAD_SIZE;
+ pda->active_mm = &init_mm;
+ pda->mmu_state = 0;
+
+ if (cpu == 0) {
+#ifdef CONFIG_XEN
+ xen_init_pt();
+#endif
+ /* others are initialized in smpboot.c */
+ pda->pcurrent = &init_task;
+ pda->irqstackptr = boot_cpu_stack;
+ } else {
+ pda->irqstackptr = (char *)
+ __get_free_pages(GFP_ATOMIC, IRQSTACK_ORDER);
+ if (!pda->irqstackptr)
+ panic("cannot allocate irqstack for cpu %d", cpu);
+ }
+
+ switch_pt();
+
+ pda->irqstackptr += IRQSTACKSIZE-64;
+}
+
+#ifndef CONFIG_X86_NO_TSS
+char boot_exception_stacks[(N_EXCEPTION_STACKS - 1) * EXCEPTION_STKSZ + DEBUG_STKSZ]
+__attribute__((section(".bss.page_aligned")));
+#endif
+
+/* May not be marked __init: used by software suspend */
+void syscall_init(void)
+{
+#ifndef CONFIG_XEN
+ /*
+ * LSTAR and STAR live in a bit strange symbiosis.
+ * They both write to the same internal register. STAR allows to set CS/DS
+ * but only a 32bit target. LSTAR sets the 64bit rip.
+ */
+ wrmsrl(MSR_STAR, ((u64)__USER32_CS)<<48 | ((u64)__KERNEL_CS)<<32);
+ wrmsrl(MSR_LSTAR, system_call);
+
+ /* Flags to clear on syscall */
+ wrmsrl(MSR_SYSCALL_MASK, EF_TF|EF_DF|EF_IE|0x3000);
+#endif
+#ifdef CONFIG_IA32_EMULATION
+ syscall32_cpu_init ();
+#endif
+}
+
+void __cpuinit check_efer(void)
+{
+ unsigned long efer;
+
+ rdmsrl(MSR_EFER, efer);
+ if (!(efer & EFER_NX) || do_not_nx) {
+ __supported_pte_mask &= ~_PAGE_NX;
+ }
+}
+
+unsigned long kernel_eflags;
+
+/*
+ * cpu_init() initializes state that is per-CPU. Some data is already
+ * initialized (naturally) in the bootstrap process, such as the GDT
+ * and IDT. We reload them nevertheless, this function acts as a
+ * 'CPU state barrier', nothing should get across.
+ * A lot of state is already set up in PDA init.
+ */
+void __cpuinit cpu_init (void)
+{
+ int cpu = stack_smp_processor_id();
+#ifndef CONFIG_X86_NO_TSS
+ struct tss_struct *t = &per_cpu(init_tss, cpu);
+ struct orig_ist *orig_ist = &per_cpu(orig_ist, cpu);
+ unsigned long v;
+ char *estacks = NULL;
+ unsigned i;
+#endif
+ struct task_struct *me;
+
+ /* CPU 0 is initialised in head64.c */
+ if (cpu != 0) {
+ pda_init(cpu);
+ zap_low_mappings(cpu);
+ }
+#ifndef CONFIG_X86_NO_TSS
+ else
+ estacks = boot_exception_stacks;
+#endif
+
+ me = current;
+
+ if (cpu_test_and_set(cpu, cpu_initialized))
+ panic("CPU#%d already initialized!\n", cpu);
+
+ printk("Initializing CPU#%d\n", cpu);
+
+ clear_in_cr4(X86_CR4_VME|X86_CR4_PVI|X86_CR4_TSD|X86_CR4_DE);
+
+ /*
+ * Initialize the per-CPU GDT with the boot GDT,
+ * and set up the GDT descriptor:
+ */
+#ifndef CONFIG_XEN
+ if (cpu)
+ memcpy(cpu_gdt(cpu), cpu_gdt_table, GDT_SIZE);
+#endif
+
+ cpu_gdt_descr[cpu].size = GDT_SIZE;
+ cpu_gdt_init(&cpu_gdt_descr[cpu]);
+
+ memset(me->thread.tls_array, 0, GDT_ENTRY_TLS_ENTRIES * 8);
+ syscall_init();
+
+ wrmsrl(MSR_FS_BASE, 0);
+ wrmsrl(MSR_KERNEL_GS_BASE, 0);
+ barrier();
+
+ check_efer();
+
+#ifndef CONFIG_X86_NO_TSS
+ /*
+ * set up and load the per-CPU TSS
+ */
+ for (v = 0; v < N_EXCEPTION_STACKS; v++) {
+ if (cpu) {
+ static const unsigned int order[N_EXCEPTION_STACKS] = {
+ [0 ... N_EXCEPTION_STACKS - 1] = EXCEPTION_STACK_ORDER,
+ [DEBUG_STACK - 1] = DEBUG_STACK_ORDER
+ };
+
+ estacks = (char *)__get_free_pages(GFP_ATOMIC, order[v]);
+ if (!estacks)
+ panic("Cannot allocate exception stack %ld %d\n",
+ v, cpu);
+ }
+ switch (v + 1) {
+#if DEBUG_STKSZ > EXCEPTION_STKSZ
+ case DEBUG_STACK:
+ cpu_pda(cpu)->debugstack = (unsigned long)estacks;
+ estacks += DEBUG_STKSZ;
+ break;
+#endif
+ default:
+ estacks += EXCEPTION_STKSZ;
+ break;
+ }
+ orig_ist->ist[v] = t->ist[v] = (unsigned long)estacks;
+ }
+
+ t->io_bitmap_base = offsetof(struct tss_struct, io_bitmap);
+ /*
+ * <= is required because the CPU will access up to
+ * 8 bits beyond the end of the IO permission bitmap.
+ */
+ for (i = 0; i <= IO_BITMAP_LONGS; i++)
+ t->io_bitmap[i] = ~0UL;
+#endif
+
+ atomic_inc(&init_mm.mm_count);
+ me->active_mm = &init_mm;
+ if (me->mm)
+ BUG();
+ enter_lazy_tlb(&init_mm, me);
+
+#ifndef CONFIG_X86_NO_TSS
+ set_tss_desc(cpu, t);
+#endif
+#ifndef CONFIG_XEN
+ load_TR_desc();
+#endif
+ load_LDT(&init_mm.context);
+
+ /*
+ * Clear all 6 debug registers:
+ */
+
+ set_debugreg(0UL, 0);
+ set_debugreg(0UL, 1);
+ set_debugreg(0UL, 2);
+ set_debugreg(0UL, 3);
+ set_debugreg(0UL, 6);
+ set_debugreg(0UL, 7);
+
+ fpu_init();
+
+ raw_local_save_flags(kernel_eflags);
+}
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/kernel/smp_64-xen.c 2008-04-02 12:34:02.000000000 +0200
@@ -0,0 +1,575 @@
+/*
+ * Intel SMP support routines.
+ *
+ * (c) 1995 Alan Cox, Building #3 <alan@redhat.com>
+ * (c) 1998-99, 2000 Ingo Molnar <mingo@redhat.com>
+ * (c) 2002,2003 Andi Kleen, SuSE Labs.
+ *
+ * This code is released under the GNU General Public License version 2 or
+ * later.
+ */
+
+#include <linux/init.h>
+
+#include <linux/mm.h>
+#include <linux/delay.h>
+#include <linux/spinlock.h>
+#include <linux/smp_lock.h>
+#include <linux/smp.h>
+#include <linux/kernel_stat.h>
+#include <linux/mc146818rtc.h>
+#include <linux/interrupt.h>
+
+#include <asm/mtrr.h>
+#include <asm/pgalloc.h>
+#include <asm/tlbflush.h>
+#include <asm/mach_apic.h>
+#include <asm/mmu_context.h>
+#include <asm/proto.h>
+#include <asm/apicdef.h>
+#include <asm/idle.h>
+#ifdef CONFIG_XEN
+#include <xen/evtchn.h>
+#endif
+
+#ifndef CONFIG_XEN
+/*
+ * Smarter SMP flushing macros.
+ * c/o Linus Torvalds.
+ *
+ * These mean you can really definitely utterly forget about
+ * writing to user space from interrupts. (Its not allowed anyway).
+ *
+ * Optimizations Manfred Spraul <manfred@colorfullife.com>
+ *
+ * More scalable flush, from Andi Kleen
+ *
+ * To avoid global state use 8 different call vectors.
+ * Each CPU uses a specific vector to trigger flushes on other
+ * CPUs. Depending on the received vector the target CPUs look into
+ * the right per cpu variable for the flush data.
+ *
+ * With more than 8 CPUs they are hashed to the 8 available
+ * vectors. The limited global vector space forces us to this right now.
+ * In future when interrupts are split into per CPU domains this could be
+ * fixed, at the cost of triggering multiple IPIs in some cases.
+ */
+
+union smp_flush_state {
+ struct {
+ cpumask_t flush_cpumask;
+ struct mm_struct *flush_mm;
+ unsigned long flush_va;
+#define FLUSH_ALL -1ULL
+ spinlock_t tlbstate_lock;
+ };
+ char pad[SMP_CACHE_BYTES];
+} ____cacheline_aligned;
+
+/* State is put into the per CPU data section, but padded
+ to a full cache line because other CPUs can access it and we don't
+ want false sharing in the per cpu data segment. */
+static DEFINE_PER_CPU(union smp_flush_state, flush_state);
+
+/*
+ * We cannot call mmdrop() because we are in interrupt context,
+ * instead update mm->cpu_vm_mask.
+ */
+static inline void leave_mm(unsigned long cpu)
+{
+ if (read_pda(mmu_state) == TLBSTATE_OK)
+ BUG();
+ cpu_clear(cpu, read_pda(active_mm)->cpu_vm_mask);
+ load_cr3(swapper_pg_dir);
+}
+
+/*
+ *
+ * The flush IPI assumes that a thread switch happens in this order:
+ * [cpu0: the cpu that switches]
+ * 1) switch_mm() either 1a) or 1b)
+ * 1a) thread switch to a different mm
+ * 1a1) cpu_clear(cpu, old_mm->cpu_vm_mask);
+ * Stop ipi delivery for the old mm. This is not synchronized with
+ * the other cpus, but smp_invalidate_interrupt ignore flush ipis
+ * for the wrong mm, and in the worst case we perform a superfluous
+ * tlb flush.
+ * 1a2) set cpu mmu_state to TLBSTATE_OK
+ * Now the smp_invalidate_interrupt won't call leave_mm if cpu0
+ * was in lazy tlb mode.
+ * 1a3) update cpu active_mm
+ * Now cpu0 accepts tlb flushes for the new mm.
+ * 1a4) cpu_set(cpu, new_mm->cpu_vm_mask);
+ * Now the other cpus will send tlb flush ipis.
+ * 1a4) change cr3.
+ * 1b) thread switch without mm change
+ * cpu active_mm is correct, cpu0 already handles
+ * flush ipis.
+ * 1b1) set cpu mmu_state to TLBSTATE_OK
+ * 1b2) test_and_set the cpu bit in cpu_vm_mask.
+ * Atomically set the bit [other cpus will start sending flush ipis],
+ * and test the bit.
+ * 1b3) if the bit was 0: leave_mm was called, flush the tlb.
+ * 2) switch %%esp, ie current
+ *
+ * The interrupt must handle 2 special cases:
+ * - cr3 is changed before %%esp, ie. it cannot use current->{active_,}mm.
+ * - the cpu performs speculative tlb reads, i.e. even if the cpu only
+ * runs in kernel space, the cpu could load tlb entries for user space
+ * pages.
+ *
+ * The good news is that cpu mmu_state is local to each cpu, no
+ * write/read ordering problems.
+ */
+
+/*
+ * TLB flush IPI:
+ *
+ * 1) Flush the tlb entries if the cpu uses the mm that's being flushed.
+ * 2) Leave the mm if we are in the lazy tlb mode.
+ *
+ * Interrupts are disabled.
+ */
+
+asmlinkage void smp_invalidate_interrupt(struct pt_regs *regs)
+{
+ int cpu;
+ int sender;
+ union smp_flush_state *f;
+
+ cpu = smp_processor_id();
+ /*
+ * orig_rax contains the negated interrupt vector.
+ * Use that to determine where the sender put the data.
+ */
+ sender = ~regs->orig_rax - INVALIDATE_TLB_VECTOR_START;
+ f = &per_cpu(flush_state, sender);
+
+ if (!cpu_isset(cpu, f->flush_cpumask))
+ goto out;
+ /*
+ * This was a BUG() but until someone can quote me the
+ * line from the intel manual that guarantees an IPI to
+ * multiple CPUs is retried _only_ on the erroring CPUs
+ * its staying as a return
+ *
+ * BUG();
+ */
+
+ if (f->flush_mm == read_pda(active_mm)) {
+ if (read_pda(mmu_state) == TLBSTATE_OK) {
+ if (f->flush_va == FLUSH_ALL)
+ local_flush_tlb();
+ else
+ __flush_tlb_one(f->flush_va);
+ } else
+ leave_mm(cpu);
+ }
+out:
+ ack_APIC_irq();
+ cpu_clear(cpu, f->flush_cpumask);
+}
+
+static void flush_tlb_others(cpumask_t cpumask, struct mm_struct *mm,
+ unsigned long va)
+{
+ int sender;
+ union smp_flush_state *f;
+
+ /* Caller has disabled preemption */
+ sender = smp_processor_id() % NUM_INVALIDATE_TLB_VECTORS;
+ f = &per_cpu(flush_state, sender);
+
+ /* Could avoid this lock when
+ num_online_cpus() <= NUM_INVALIDATE_TLB_VECTORS, but it is
+ probably not worth checking this for a cache-hot lock. */
+ spin_lock(&f->tlbstate_lock);
+
+ f->flush_mm = mm;
+ f->flush_va = va;
+ cpus_or(f->flush_cpumask, cpumask, f->flush_cpumask);
+
+ /*
+ * We have to send the IPI only to
+ * CPUs affected.
+ */
+ send_IPI_mask(cpumask, INVALIDATE_TLB_VECTOR_START + sender);
+
+ while (!cpus_empty(f->flush_cpumask))
+ cpu_relax();
+
+ f->flush_mm = NULL;
+ f->flush_va = 0;
+ spin_unlock(&f->tlbstate_lock);
+}
+
+int __cpuinit init_smp_flush(void)
+{
+ int i;
+ for_each_cpu_mask(i, cpu_possible_map) {
+ spin_lock_init(&per_cpu(flush_state, i).tlbstate_lock);
+ }
+ return 0;
+}
+
+core_initcall(init_smp_flush);
+
+void flush_tlb_current_task(void)
+{
+ struct mm_struct *mm = current->mm;
+ cpumask_t cpu_mask;
+
+ preempt_disable();
+ cpu_mask = mm->cpu_vm_mask;
+ cpu_clear(smp_processor_id(), cpu_mask);
+
+ local_flush_tlb();
+ if (!cpus_empty(cpu_mask))
+ flush_tlb_others(cpu_mask, mm, FLUSH_ALL);
+ preempt_enable();
+}
+EXPORT_SYMBOL(flush_tlb_current_task);
+
+void flush_tlb_mm (struct mm_struct * mm)
+{
+ cpumask_t cpu_mask;
+
+ preempt_disable();
+ cpu_mask = mm->cpu_vm_mask;
+ cpu_clear(smp_processor_id(), cpu_mask);
+
+ if (current->active_mm == mm) {
+ if (current->mm)
+ local_flush_tlb();
+ else
+ leave_mm(smp_processor_id());
+ }
+ if (!cpus_empty(cpu_mask))
+ flush_tlb_others(cpu_mask, mm, FLUSH_ALL);
+
+ preempt_enable();
+}
+EXPORT_SYMBOL(flush_tlb_mm);
+
+void flush_tlb_page(struct vm_area_struct * vma, unsigned long va)
+{
+ struct mm_struct *mm = vma->vm_mm;
+ cpumask_t cpu_mask;
+
+ preempt_disable();
+ cpu_mask = mm->cpu_vm_mask;
+ cpu_clear(smp_processor_id(), cpu_mask);
+
+ if (current->active_mm == mm) {
+ if(current->mm)
+ __flush_tlb_one(va);
+ else
+ leave_mm(smp_processor_id());
+ }
+
+ if (!cpus_empty(cpu_mask))
+ flush_tlb_others(cpu_mask, mm, va);
+
+ preempt_enable();
+}
+EXPORT_SYMBOL(flush_tlb_page);
+
+static void do_flush_tlb_all(void* info)
+{
+ unsigned long cpu = smp_processor_id();
+
+ __flush_tlb_all();
+ if (read_pda(mmu_state) == TLBSTATE_LAZY)
+ leave_mm(cpu);
+}
+
+void flush_tlb_all(void)
+{
+ on_each_cpu(do_flush_tlb_all, NULL, 1, 1);
+}
+#endif /* Xen */
+
+/*
+ * this function sends a 'reschedule' IPI to another CPU.
+ * it goes straight through and wastes no time serializing
+ * anything. Worst case is that we lose a reschedule ...
+ */
+
+void smp_send_reschedule(int cpu)
+{
+ send_IPI_mask(cpumask_of_cpu(cpu), RESCHEDULE_VECTOR);
+}
+
+/*
+ * Structure and data for smp_call_function(). This is designed to minimise
+ * static memory requirements. It also looks cleaner.
+ */
+static DEFINE_SPINLOCK(call_lock);
+
+struct call_data_struct {
+ void (*func) (void *info);
+ void *info;
+ atomic_t started;
+ atomic_t finished;
+ int wait;
+};
+
+static struct call_data_struct * call_data;
+
+void lock_ipi_call_lock(void)
+{
+ spin_lock_irq(&call_lock);
+}
+
+void unlock_ipi_call_lock(void)
+{
+ spin_unlock_irq(&call_lock);
+}
+
+/*
+ * this function sends a 'generic call function' IPI to one other CPU
+ * in the system.
+ *
+ * cpu is a standard Linux logical CPU number.
+ */
+static void
+__smp_call_function_single(int cpu, void (*func) (void *info), void *info,
+ int nonatomic, int wait)
+{
+ struct call_data_struct data;
+ int cpus = 1;
+
+ data.func = func;
+ data.info = info;
+ atomic_set(&data.started, 0);
+ data.wait = wait;
+ if (wait)
+ atomic_set(&data.finished, 0);
+
+ call_data = &data;
+ wmb();
+ /* Send a message to all other CPUs and wait for them to respond */
+ send_IPI_mask(cpumask_of_cpu(cpu), CALL_FUNCTION_VECTOR);
+
+ /* Wait for response */
+ while (atomic_read(&data.started) != cpus)
+ cpu_relax();
+
+ if (!wait)
+ return;
+
+ while (atomic_read(&data.finished) != cpus)
+ cpu_relax();
+}
+
+/*
+ * smp_call_function_single - Run a function on another CPU
+ * @func: The function to run. This must be fast and non-blocking.
+ * @info: An arbitrary pointer to pass to the function.
+ * @nonatomic: Currently unused.
+ * @wait: If true, wait until function has completed on other CPUs.
+ *
+ * Retrurns 0 on success, else a negative status code.
+ *
+ * Does not return until the remote CPU is nearly ready to execute <func>
+ * or is or has executed.
+ */
+
+int smp_call_function_single (int cpu, void (*func) (void *info), void *info,
+ int nonatomic, int wait)
+{
+ /* prevent preemption and reschedule on another processor */
+ int me = get_cpu();
+ if (cpu == me) {
+ WARN_ON(1);
+ put_cpu();
+ return -EBUSY;
+ }
+ spin_lock_bh(&call_lock);
+ __smp_call_function_single(cpu, func, info, nonatomic, wait);
+ spin_unlock_bh(&call_lock);
+ put_cpu();
+ return 0;
+}
+
+/*
+ * this function sends a 'generic call function' IPI to all other CPUs
+ * in the system.
+ */
+static void __smp_call_function (void (*func) (void *info), void *info,
+ int nonatomic, int wait)
+{
+ struct call_data_struct data;
+ int cpus = num_online_cpus()-1;
+
+ if (!cpus)
+ return;
+
+ data.func = func;
+ data.info = info;
+ atomic_set(&data.started, 0);
+ data.wait = wait;
+ if (wait)
+ atomic_set(&data.finished, 0);
+
+ call_data = &data;
+ wmb();
+ /* Send a message to all other CPUs and wait for them to respond */
+ send_IPI_allbutself(CALL_FUNCTION_VECTOR);
+
+ /* Wait for response */
+ while (atomic_read(&data.started) != cpus)
+ cpu_relax();
+
+ if (!wait)
+ return;
+
+ while (atomic_read(&data.finished) != cpus)
+ cpu_relax();
+}
+
+/*
+ * smp_call_function - run a function on all other CPUs.
+ * @func: The function to run. This must be fast and non-blocking.
+ * @info: An arbitrary pointer to pass to the function.
+ * @nonatomic: currently unused.
+ * @wait: If true, wait (atomically) until function has completed on other
+ * CPUs.
+ *
+ * Returns 0 on success, else a negative status code. Does not return until
+ * remote CPUs are nearly ready to execute func or are or have executed.
+ *
+ * You must not call this function with disabled interrupts or from a
+ * hardware interrupt handler or from a bottom half handler.
+ * Actually there are a few legal cases, like panic.
+ */
+int smp_call_function (void (*func) (void *info), void *info, int nonatomic,
+ int wait)
+{
+ spin_lock(&call_lock);
+ __smp_call_function(func,info,nonatomic,wait);
+ spin_unlock(&call_lock);
+ return 0;
+}
+EXPORT_SYMBOL(smp_call_function);
+
+void smp_stop_cpu(void)
+{
+ unsigned long flags;
+ /*
+ * Remove this CPU:
+ */
+ cpu_clear(smp_processor_id(), cpu_online_map);
+ local_irq_save(flags);
+ disable_all_local_evtchn();
+ local_irq_restore(flags);
+}
+
+static void smp_really_stop_cpu(void *dummy)
+{
+ smp_stop_cpu();
+ for (;;)
+ halt();
+}
+
+void smp_send_stop(void)
+{
+ int nolock = 0;
+#ifndef CONFIG_XEN
+ if (reboot_force)
+ return;
+#endif
+ /* Don't deadlock on the call lock in panic */
+ if (!spin_trylock(&call_lock)) {
+ /* ignore locking because we have panicked anyways */
+ nolock = 1;
+ }
+ __smp_call_function(smp_really_stop_cpu, NULL, 0, 0);
+ if (!nolock)
+ spin_unlock(&call_lock);
+
+ local_irq_disable();
+ disable_all_local_evtchn();
+ local_irq_enable();
+}
+
+/*
+ * Reschedule call back. Nothing to do,
+ * all the work is done automatically when
+ * we return from the interrupt.
+ */
+#ifndef CONFIG_XEN
+asmlinkage void smp_reschedule_interrupt(void)
+#else
+asmlinkage irqreturn_t smp_reschedule_interrupt(void)
+#endif
+{
+#ifndef CONFIG_XEN
+ ack_APIC_irq();
+#else
+ return IRQ_HANDLED;
+#endif
+}
+
+#ifndef CONFIG_XEN
+asmlinkage void smp_call_function_interrupt(void)
+#else
+asmlinkage irqreturn_t smp_call_function_interrupt(void)
+#endif
+{
+ void (*func) (void *info) = call_data->func;
+ void *info = call_data->info;
+ int wait = call_data->wait;
+
+#ifndef CONFIG_XEN
+ ack_APIC_irq();
+#endif
+ /*
+ * Notify initiating CPU that I've grabbed the data and am
+ * about to execute the function
+ */
+ mb();
+ atomic_inc(&call_data->started);
+ /*
+ * At this point the info structure may be out of scope unless wait==1
+ */
+ exit_idle();
+ irq_enter();
+ (*func)(info);
+ irq_exit();
+ if (wait) {
+ mb();
+ atomic_inc(&call_data->finished);
+ }
+#ifdef CONFIG_XEN
+ return IRQ_HANDLED;
+#endif
+}
+
+int safe_smp_processor_id(void)
+{
+#ifdef CONFIG_XEN
+ return smp_processor_id();
+#else
+ unsigned apicid, i;
+
+ if (disable_apic)
+ return 0;
+
+ apicid = hard_smp_processor_id();
+ if (apicid < NR_CPUS && x86_cpu_to_apicid[apicid] == apicid)
+ return apicid;
+
+ for (i = 0; i < NR_CPUS; ++i) {
+ if (x86_cpu_to_apicid[i] == apicid)
+ return i;
+ }
+
+ /* No entries in x86_cpu_to_apicid? Either no MPS|ACPI,
+ * or called too early. Either way, we must be CPU 0. */
+ if (x86_cpu_to_apicid[0] == BAD_APICID)
+ return 0;
+
+ return 0; /* Should not happen */
+#endif
+}
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/kernel/traps_64-xen.c 2008-04-02 12:34:02.000000000 +0200
@@ -0,0 +1,1173 @@
+/*
+ * linux/arch/x86-64/traps.c
+ *
+ * Copyright (C) 1991, 1992 Linus Torvalds
+ * Copyright (C) 2000, 2001, 2002 Andi Kleen, SuSE Labs
+ *
+ * Pentium III FXSR, SSE support
+ * Gareth Hughes <gareth@valinux.com>, May 2000
+ */
+
+/*
+ * 'Traps.c' handles hardware traps and faults after we have saved some
+ * state in 'entry.S'.
+ */
+#include <linux/sched.h>
+#include <linux/kernel.h>
+#include <linux/string.h>
+#include <linux/errno.h>
+#include <linux/ptrace.h>
+#include <linux/timer.h>
+#include <linux/mm.h>
+#include <linux/init.h>
+#include <linux/delay.h>
+#include <linux/spinlock.h>
+#include <linux/interrupt.h>
+#include <linux/module.h>
+#include <linux/moduleparam.h>
+#include <linux/nmi.h>
+#include <linux/kprobes.h>
+#include <linux/kexec.h>
+#include <linux/unwind.h>
+
+#include <asm/system.h>
+#include <asm/uaccess.h>
+#include <asm/io.h>
+#include <asm/atomic.h>
+#include <asm/debugreg.h>
+#include <asm/desc.h>
+#include <asm/i387.h>
+#include <asm/kdebug.h>
+#include <asm/processor.h>
+#include <asm/unwind.h>
+#include <asm/smp.h>
+#include <asm/pgalloc.h>
+#include <asm/pda.h>
+#include <asm/proto.h>
+#include <asm/nmi.h>
+
+asmlinkage void divide_error(void);
+asmlinkage void debug(void);
+asmlinkage void nmi(void);
+asmlinkage void int3(void);
+asmlinkage void overflow(void);
+asmlinkage void bounds(void);
+asmlinkage void invalid_op(void);
+asmlinkage void device_not_available(void);
+asmlinkage void double_fault(void);
+asmlinkage void coprocessor_segment_overrun(void);
+asmlinkage void invalid_TSS(void);
+asmlinkage void segment_not_present(void);
+asmlinkage void stack_segment(void);
+asmlinkage void general_protection(void);
+asmlinkage void page_fault(void);
+asmlinkage void coprocessor_error(void);
+asmlinkage void simd_coprocessor_error(void);
+asmlinkage void reserved(void);
+asmlinkage void alignment_check(void);
+asmlinkage void machine_check(void);
+asmlinkage void spurious_interrupt_bug(void);
+
+ATOMIC_NOTIFIER_HEAD(die_chain);
+EXPORT_SYMBOL(die_chain);
+
+int register_die_notifier(struct notifier_block *nb)
+{
+ vmalloc_sync_all();
+ return atomic_notifier_chain_register(&die_chain, nb);
+}
+EXPORT_SYMBOL(register_die_notifier); /* used modular by kdb */
+
+int unregister_die_notifier(struct notifier_block *nb)
+{
+ return atomic_notifier_chain_unregister(&die_chain, nb);
+}
+EXPORT_SYMBOL(unregister_die_notifier); /* used modular by kdb */
+
+static inline void conditional_sti(struct pt_regs *regs)
+{
+ if (regs->eflags & X86_EFLAGS_IF)
+ local_irq_enable();
+}
+
+static inline void preempt_conditional_sti(struct pt_regs *regs)
+{
+ preempt_disable();
+ if (regs->eflags & X86_EFLAGS_IF)
+ local_irq_enable();
+}
+
+static inline void preempt_conditional_cli(struct pt_regs *regs)
+{
+ if (regs->eflags & X86_EFLAGS_IF)
+ local_irq_disable();
+ /* Make sure to not schedule here because we could be running
+ on an exception stack. */
+ preempt_enable_no_resched();
+}
+
+static int kstack_depth_to_print = 12;
+#ifdef CONFIG_STACK_UNWIND
+static int call_trace = 1;
+#else
+#define call_trace (-1)
+#endif
+
+#ifdef CONFIG_KALLSYMS
+# include <linux/kallsyms.h>
+void printk_address(unsigned long address)
+{
+ unsigned long offset = 0, symsize;
+ const char *symname;
+ char *modname;
+ char *delim = ":";
+ char namebuf[128];
+
+ symname = kallsyms_lookup(address, &symsize, &offset,
+ &modname, namebuf);
+ if (!symname) {
+ printk(" [<%016lx>]\n", address);
+ return;
+ }
+ if (!modname)
+ modname = delim = "";
+ printk(" [<%016lx>] %s%s%s%s+0x%lx/0x%lx\n",
+ address, delim, modname, delim, symname, offset, symsize);
+}
+#else
+void printk_address(unsigned long address)
+{
+ printk(" [<%016lx>]\n", address);
+}
+#endif
+
+static unsigned long *in_exception_stack(unsigned cpu, unsigned long stack,
+ unsigned *usedp, const char **idp)
+{
+#ifndef CONFIG_X86_NO_TSS
+ static char ids[][8] = {
+ [DEBUG_STACK - 1] = "#DB",
+ [NMI_STACK - 1] = "NMI",
+ [DOUBLEFAULT_STACK - 1] = "#DF",
+ [STACKFAULT_STACK - 1] = "#SS",
+ [MCE_STACK - 1] = "#MC",
+#if DEBUG_STKSZ > EXCEPTION_STKSZ
+ [N_EXCEPTION_STACKS ... N_EXCEPTION_STACKS + DEBUG_STKSZ / EXCEPTION_STKSZ - 2] = "#DB[?]"
+#endif
+ };
+ unsigned k;
+
+ /*
+ * Iterate over all exception stacks, and figure out whether
+ * 'stack' is in one of them:
+ */
+ for (k = 0; k < N_EXCEPTION_STACKS; k++) {
+ unsigned long end;
+
+ /*
+ * set 'end' to the end of the exception stack.
+ */
+ switch (k + 1) {
+ /*
+ * TODO: this block is not needed i think, because
+ * setup64.c:cpu_init() sets up t->ist[DEBUG_STACK]
+ * properly too.
+ */
+#if DEBUG_STKSZ > EXCEPTION_STKSZ
+ case DEBUG_STACK:
+ end = cpu_pda(cpu)->debugstack + DEBUG_STKSZ;
+ break;
+#endif
+ default:
+ end = per_cpu(orig_ist, cpu).ist[k];
+ break;
+ }
+ /*
+ * Is 'stack' above this exception frame's end?
+ * If yes then skip to the next frame.
+ */
+ if (stack >= end)
+ continue;
+ /*
+ * Is 'stack' above this exception frame's start address?
+ * If yes then we found the right frame.
+ */
+ if (stack >= end - EXCEPTION_STKSZ) {
+ /*
+ * Make sure we only iterate through an exception
+ * stack once. If it comes up for the second time
+ * then there's something wrong going on - just
+ * break out and return NULL:
+ */
+ if (*usedp & (1U << k))
+ break;
+ *usedp |= 1U << k;
+ *idp = ids[k];
+ return (unsigned long *)end;
+ }
+ /*
+ * If this is a debug stack, and if it has a larger size than
+ * the usual exception stacks, then 'stack' might still
+ * be within the lower portion of the debug stack:
+ */
+#if DEBUG_STKSZ > EXCEPTION_STKSZ
+ if (k == DEBUG_STACK - 1 && stack >= end - DEBUG_STKSZ) {
+ unsigned j = N_EXCEPTION_STACKS - 1;
+
+ /*
+ * Black magic. A large debug stack is composed of
+ * multiple exception stack entries, which we
+ * iterate through now. Dont look:
+ */
+ do {
+ ++j;
+ end -= EXCEPTION_STKSZ;
+ ids[j][4] = '1' + (j - N_EXCEPTION_STACKS);
+ } while (stack < end - EXCEPTION_STKSZ);
+ if (*usedp & (1U << j))
+ break;
+ *usedp |= 1U << j;
+ *idp = ids[j];
+ return (unsigned long *)end;
+ }
+#endif
+ }
+#endif
+ return NULL;
+}
+
+static int show_trace_unwind(struct unwind_frame_info *info, void *context)
+{
+ int n = 0;
+
+ while (unwind(info) == 0 && UNW_PC(info)) {
+ n++;
+ printk_address(UNW_PC(info));
+ if (arch_unw_user_mode(info))
+ break;
+ }
+ return n;
+}
+
+/*
+ * x86-64 can have upto three kernel stacks:
+ * process stack
+ * interrupt stack
+ * severe exception (double fault, nmi, stack fault, debug, mce) hardware stack
+ */
+
+void show_trace(struct task_struct *tsk, struct pt_regs *regs, unsigned long * stack)
+{
+ const unsigned cpu = safe_smp_processor_id();
+ unsigned long *irqstack_end = (unsigned long *)cpu_pda(cpu)->irqstackptr;
+ unsigned used = 0;
+
+ printk("\nCall Trace:\n");
+
+ if (!tsk)
+ tsk = current;
+
+ if (call_trace >= 0) {
+ int unw_ret = 0;
+ struct unwind_frame_info info;
+
+ if (regs) {
+ if (unwind_init_frame_info(&info, tsk, regs) == 0)
+ unw_ret = show_trace_unwind(&info, NULL);
+ } else if (tsk == current)
+ unw_ret = unwind_init_running(&info, show_trace_unwind, NULL);
+ else {
+ if (unwind_init_blocked(&info, tsk) == 0)
+ unw_ret = show_trace_unwind(&info, NULL);
+ }
+ if (unw_ret > 0) {
+ if (call_trace == 1 && !arch_unw_user_mode(&info)) {
+ print_symbol("DWARF2 unwinder stuck at %s\n",
+ UNW_PC(&info));
+ if ((long)UNW_SP(&info) < 0) {
+ printk("Leftover inexact backtrace:\n");
+ stack = (unsigned long *)UNW_SP(&info);
+ } else
+ printk("Full inexact backtrace again:\n");
+ } else if (call_trace >= 1)
+ return;
+ else
+ printk("Full inexact backtrace again:\n");
+ } else
+ printk("Inexact backtrace:\n");
+ }
+
+ /*
+ * Print function call entries within a stack. 'cond' is the
+ * "end of stackframe" condition, that the 'stack++'
+ * iteration will eventually trigger.
+ */
+#define HANDLE_STACK(cond) \
+ do while (cond) { \
+ unsigned long addr = *stack++; \
+ if (kernel_text_address(addr)) { \
+ /* \
+ * If the address is either in the text segment of the \
+ * kernel, or in the region which contains vmalloc'ed \
+ * memory, it *may* be the address of a calling \
+ * routine; if so, print it so that someone tracing \
+ * down the cause of the crash will be able to figure \
+ * out the call path that was taken. \
+ */ \
+ printk_address(addr); \
+ } \
+ } while (0)
+
+ /*
+ * Print function call entries in all stacks, starting at the
+ * current stack address. If the stacks consist of nested
+ * exceptions
+ */
+ for ( ; ; ) {
+ const char *id;
+ unsigned long *estack_end;
+ estack_end = in_exception_stack(cpu, (unsigned long)stack,
+ &used, &id);
+
+ if (estack_end) {
+ printk(" <%s>", id);
+ HANDLE_STACK (stack < estack_end);
+ printk(" <EOE>");
+ /*
+ * We link to the next stack via the
+ * second-to-last pointer (index -2 to end) in the
+ * exception stack:
+ */
+ stack = (unsigned long *) estack_end[-2];
+ continue;
+ }
+ if (irqstack_end) {
+ unsigned long *irqstack;
+ irqstack = irqstack_end -
+ (IRQSTACKSIZE - 64) / sizeof(*irqstack);
+
+ if (stack >= irqstack && stack < irqstack_end) {
+ printk(" <IRQ>");
+ HANDLE_STACK (stack < irqstack_end);
+ /*
+ * We link to the next stack (which would be
+ * the process stack normally) the last
+ * pointer (index -1 to end) in the IRQ stack:
+ */
+ stack = (unsigned long *) (irqstack_end[-1]);
+ irqstack_end = NULL;
+ printk(" <EOI>");
+ continue;
+ }
+ }
+ break;
+ }
+
+ /*
+ * This prints the process stack:
+ */
+ HANDLE_STACK (((long) stack & (THREAD_SIZE-1)) != 0);
+#undef HANDLE_STACK
+
+ printk("\n");
+}
+
+static void _show_stack(struct task_struct *tsk, struct pt_regs *regs, unsigned long * rsp)
+{
+ unsigned long *stack;
+ int i;
+ const int cpu = safe_smp_processor_id();
+ unsigned long *irqstack_end = (unsigned long *) (cpu_pda(cpu)->irqstackptr);
+ unsigned long *irqstack = (unsigned long *) (cpu_pda(cpu)->irqstackptr - IRQSTACKSIZE);
+
+ // debugging aid: "show_stack(NULL, NULL);" prints the
+ // back trace for this cpu.
+
+ if (rsp == NULL) {
+ if (tsk)
+ rsp = (unsigned long *)tsk->thread.rsp;
+ else
+ rsp = (unsigned long *)&rsp;
+ }
+
+ stack = rsp;
+ for(i=0; i < kstack_depth_to_print; i++) {
+ if (stack >= irqstack && stack <= irqstack_end) {
+ if (stack == irqstack_end) {
+ stack = (unsigned long *) (irqstack_end[-1]);
+ printk(" <EOI> ");
+ }
+ } else {
+ if (((long) stack & (THREAD_SIZE-1)) == 0)
+ break;
+ }
+ if (i && ((i % 4) == 0))
+ printk("\n");
+ printk(" %016lx", *stack++);
+ touch_nmi_watchdog();
+ }
+ show_trace(tsk, regs, rsp);
+}
+
+void show_stack(struct task_struct *tsk, unsigned long * rsp)
+{
+ _show_stack(tsk, NULL, rsp);
+}
+
+/*
+ * The architecture-independent dump_stack generator
+ */
+void dump_stack(void)
+{
+ unsigned long dummy;
+ show_trace(NULL, NULL, &dummy);
+}
+
+EXPORT_SYMBOL(dump_stack);
+
+void show_registers(struct pt_regs *regs)
+{
+ int i;
+ int in_kernel = !user_mode(regs);
+ unsigned long rsp;
+ const int cpu = safe_smp_processor_id();
+ struct task_struct *cur = cpu_pda(cpu)->pcurrent;
+
+ rsp = regs->rsp;
+
+ printk("CPU %d ", cpu);
+ __show_regs(regs);
+ printk("Process %s (pid: %d, threadinfo %p, task %p)\n",
+ cur->comm, cur->pid, task_thread_info(cur), cur);
+
+ /*
+ * When in-kernel, we also print out the stack and code at the
+ * time of the fault..
+ */
+ if (in_kernel) {
+
+ printk("Stack: ");
+ _show_stack(NULL, regs, (unsigned long*)rsp);
+
+ printk("\nCode: ");
+ if (regs->rip < PAGE_OFFSET)
+ goto bad;
+
+ for (i=0; i<20; i++) {
+ unsigned char c;
+ if (__get_user(c, &((unsigned char*)regs->rip)[i])) {
+bad:
+ printk(" Bad RIP value.");
+ break;
+ }
+ printk("%02x ", c);
+ }
+ }
+ printk("\n");
+}
+
+void handle_BUG(struct pt_regs *regs)
+{
+ struct bug_frame f;
+ long len;
+ const char *prefix = "";
+
+ if (user_mode(regs))
+ return;
+ if (__copy_from_user(&f, (const void __user *) regs->rip,
+ sizeof(struct bug_frame)))
+ return;
+ if (f.filename >= 0 ||
+ f.ud2[0] != 0x0f || f.ud2[1] != 0x0b)
+ return;
+ len = __strnlen_user((char *)(long)f.filename, PATH_MAX) - 1;
+ if (len < 0 || len >= PATH_MAX)
+ f.filename = (int)(long)"unmapped filename";
+ else if (len > 50) {
+ f.filename += len - 50;
+ prefix = "...";
+ }
+ printk("----------- [cut here ] --------- [please bite here ] ---------\n");
+ printk(KERN_ALERT "Kernel BUG at %s%.50s:%d\n", prefix, (char *)(long)f.filename, f.line);
+}
+
+#ifdef CONFIG_BUG
+void out_of_line_bug(void)
+{
+ BUG();
+}
+EXPORT_SYMBOL(out_of_line_bug);
+#endif
+
+static DEFINE_SPINLOCK(die_lock);
+static int die_owner = -1;
+static unsigned int die_nest_count;
+
+unsigned __kprobes long oops_begin(void)
+{
+ int cpu = safe_smp_processor_id();
+ unsigned long flags;
+
+ /* racy, but better than risking deadlock. */
+ local_irq_save(flags);
+ if (!spin_trylock(&die_lock)) {
+ if (cpu == die_owner)
+ /* nested oops. should stop eventually */;
+ else
+ spin_lock(&die_lock);
+ }
+ die_nest_count++;
+ die_owner = cpu;
+ console_verbose();
+ bust_spinlocks(1);
+ return flags;
+}
+
+void __kprobes oops_end(unsigned long flags)
+{
+ die_owner = -1;
+ bust_spinlocks(0);
+ die_nest_count--;
+ if (die_nest_count)
+ /* We still own the lock */
+ local_irq_restore(flags);
+ else
+ /* Nest count reaches zero, release the lock. */
+ spin_unlock_irqrestore(&die_lock, flags);
+ if (panic_on_oops)
+ panic("Fatal exception");
+}
+
+void __kprobes __die(const char * str, struct pt_regs * regs, long err)
+{
+ static int die_counter;
+ printk(KERN_EMERG "%s: %04lx [%u] ", str, err & 0xffff,++die_counter);
+#ifdef CONFIG_PREEMPT
+ printk("PREEMPT ");
+#endif
+#ifdef CONFIG_SMP
+ printk("SMP ");
+#endif
+#ifdef CONFIG_DEBUG_PAGEALLOC
+ printk("DEBUG_PAGEALLOC");
+#endif
+ printk("\n");
+ notify_die(DIE_OOPS, str, regs, err, current->thread.trap_no, SIGSEGV);
+ show_registers(regs);
+ /* Executive summary in case the oops scrolled away */
+ printk(KERN_ALERT "RIP ");
+ printk_address(regs->rip);
+ printk(" RSP <%016lx>\n", regs->rsp);
+ if (kexec_should_crash(current))
+ crash_kexec(regs);
+}
+
+void die(const char * str, struct pt_regs * regs, long err)
+{
+ unsigned long flags = oops_begin();
+
+ handle_BUG(regs);
+ __die(str, regs, err);
+ oops_end(flags);
+ do_exit(SIGSEGV);
+}
+
+#ifdef CONFIG_X86_LOCAL_APIC
+void __kprobes die_nmi(char *str, struct pt_regs *regs)
+{
+ unsigned long flags = oops_begin();
+
+ /*
+ * We are in trouble anyway, lets at least try
+ * to get a message out.
+ */
+ printk(str, safe_smp_processor_id());
+ show_registers(regs);
+ if (kexec_should_crash(current))
+ crash_kexec(regs);
+ if (panic_on_timeout || panic_on_oops)
+ panic("nmi watchdog");
+ printk("console shuts up ...\n");
+ oops_end(flags);
+ nmi_exit();
+ local_irq_enable();
+ do_exit(SIGSEGV);
+}
+#endif
+
+static void __kprobes do_trap(int trapnr, int signr, char *str,
+ struct pt_regs * regs, long error_code,
+ siginfo_t *info)
+{
+ struct task_struct *tsk = current;
+
+ tsk->thread.error_code = error_code;
+ tsk->thread.trap_no = trapnr;
+
+ if (user_mode(regs)) {
+ if (exception_trace && unhandled_signal(tsk, signr))
+ printk(KERN_INFO
+ "%s[%d] trap %s rip:%lx rsp:%lx error:%lx\n",
+ tsk->comm, tsk->pid, str,
+ regs->rip, regs->rsp, error_code);
+
+ if (info)
+ force_sig_info(signr, info, tsk);
+ else
+ force_sig(signr, tsk);
+ return;
+ }
+
+
+ /* kernel trap */
+ {
+ const struct exception_table_entry *fixup;
+ fixup = search_exception_tables(regs->rip);
+ if (fixup)
+ regs->rip = fixup->fixup;
+ else
+ die(str, regs, error_code);
+ return;
+ }
+}
+
+#define DO_ERROR(trapnr, signr, str, name) \
+asmlinkage void do_##name(struct pt_regs * regs, long error_code) \
+{ \
+ if (notify_die(DIE_TRAP, str, regs, error_code, trapnr, signr) \
+ == NOTIFY_STOP) \
+ return; \
+ conditional_sti(regs); \
+ do_trap(trapnr, signr, str, regs, error_code, NULL); \
+}
+
+#define DO_ERROR_INFO(trapnr, signr, str, name, sicode, siaddr) \
+asmlinkage void do_##name(struct pt_regs * regs, long error_code) \
+{ \
+ siginfo_t info; \
+ info.si_signo = signr; \
+ info.si_errno = 0; \
+ info.si_code = sicode; \
+ info.si_addr = (void __user *)siaddr; \
+ if (notify_die(DIE_TRAP, str, regs, error_code, trapnr, signr) \
+ == NOTIFY_STOP) \
+ return; \
+ conditional_sti(regs); \
+ do_trap(trapnr, signr, str, regs, error_code, &info); \
+}
+
+DO_ERROR_INFO( 0, SIGFPE, "divide error", divide_error, FPE_INTDIV, regs->rip)
+DO_ERROR( 4, SIGSEGV, "overflow", overflow)
+DO_ERROR( 5, SIGSEGV, "bounds", bounds)
+DO_ERROR_INFO( 6, SIGILL, "invalid opcode", invalid_op, ILL_ILLOPN, regs->rip)
+DO_ERROR( 7, SIGSEGV, "device not available", device_not_available)
+DO_ERROR( 9, SIGFPE, "coprocessor segment overrun", coprocessor_segment_overrun)
+DO_ERROR(10, SIGSEGV, "invalid TSS", invalid_TSS)
+DO_ERROR(11, SIGBUS, "segment not present", segment_not_present)
+DO_ERROR_INFO(17, SIGBUS, "alignment check", alignment_check, BUS_ADRALN, 0)
+DO_ERROR(18, SIGSEGV, "reserved", reserved)
+
+/* Runs on IST stack */
+asmlinkage void do_stack_segment(struct pt_regs *regs, long error_code)
+{
+ if (notify_die(DIE_TRAP, "stack segment", regs, error_code,
+ 12, SIGBUS) == NOTIFY_STOP)
+ return;
+ preempt_conditional_sti(regs);
+ do_trap(12, SIGBUS, "stack segment", regs, error_code, NULL);
+ preempt_conditional_cli(regs);
+}
+
+asmlinkage void do_double_fault(struct pt_regs * regs, long error_code)
+{
+ static const char str[] = "double fault";
+ struct task_struct *tsk = current;
+
+ /* Return not checked because double check cannot be ignored */
+ notify_die(DIE_TRAP, str, regs, error_code, 8, SIGSEGV);
+
+ tsk->thread.error_code = error_code;
+ tsk->thread.trap_no = 8;
+
+ /* This is always a kernel trap and never fixable (and thus must
+ never return). */
+ for (;;)
+ die(str, regs, error_code);
+}
+
+asmlinkage void __kprobes do_general_protection(struct pt_regs * regs,
+ long error_code)
+{
+ struct task_struct *tsk = current;
+
+ conditional_sti(regs);
+
+ tsk->thread.error_code = error_code;
+ tsk->thread.trap_no = 13;
+
+ if (user_mode(regs)) {
+ if (exception_trace && unhandled_signal(tsk, SIGSEGV))
+ printk(KERN_INFO
+ "%s[%d] general protection rip:%lx rsp:%lx error:%lx\n",
+ tsk->comm, tsk->pid,
+ regs->rip, regs->rsp, error_code);
+
+ force_sig(SIGSEGV, tsk);
+ return;
+ }
+
+ /* kernel gp */
+ {
+ const struct exception_table_entry *fixup;
+ fixup = search_exception_tables(regs->rip);
+ if (fixup) {
+ regs->rip = fixup->fixup;
+ return;
+ }
+ if (notify_die(DIE_GPF, "general protection fault", regs,
+ error_code, 13, SIGSEGV) == NOTIFY_STOP)
+ return;
+ die("general protection fault", regs, error_code);
+ }
+}
+
+static __kprobes void
+mem_parity_error(unsigned char reason, struct pt_regs * regs)
+{
+ printk("Uhhuh. NMI received. Dazed and confused, but trying to continue\n");
+ printk("You probably have a hardware problem with your RAM chips\n");
+
+#if 0 /* XEN */
+ /* Clear and disable the memory parity error line. */
+ reason = (reason & 0xf) | 4;
+ outb(reason, 0x61);
+#endif /* XEN */
+}
+
+static __kprobes void
+io_check_error(unsigned char reason, struct pt_regs * regs)
+{
+ printk("NMI: IOCK error (debug interrupt?)\n");
+ show_registers(regs);
+
+#if 0 /* XEN */
+ /* Re-enable the IOCK line, wait for a few seconds */
+ reason = (reason & 0xf) | 8;
+ outb(reason, 0x61);
+ mdelay(2000);
+ reason &= ~8;
+ outb(reason, 0x61);
+#endif /* XEN */
+}
+
+static __kprobes void
+unknown_nmi_error(unsigned char reason, struct pt_regs * regs)
+{ printk("Uhhuh. NMI received for unknown reason %02x.\n", reason);
+ printk("Dazed and confused, but trying to continue\n");
+ printk("Do you have a strange power saving mode enabled?\n");
+}
+
+/* Runs on IST stack. This code must keep interrupts off all the time.
+ Nested NMIs are prevented by the CPU. */
+asmlinkage __kprobes void default_do_nmi(struct pt_regs *regs)
+{
+ unsigned char reason = 0;
+ int cpu;
+
+ cpu = smp_processor_id();
+
+ /* Only the BSP gets external NMIs from the system. */
+ if (!cpu)
+ reason = get_nmi_reason();
+
+ if (!(reason & 0xc0)) {
+ if (notify_die(DIE_NMI_IPI, "nmi_ipi", regs, reason, 2, SIGINT)
+ == NOTIFY_STOP)
+ return;
+#ifdef CONFIG_X86_LOCAL_APIC
+ /*
+ * Ok, so this is none of the documented NMI sources,
+ * so it must be the NMI watchdog.
+ */
+ if (nmi_watchdog > 0) {
+ nmi_watchdog_tick(regs,reason);
+ return;
+ }
+#endif
+ unknown_nmi_error(reason, regs);
+ return;
+ }
+ if (notify_die(DIE_NMI, "nmi", regs, reason, 2, SIGINT) == NOTIFY_STOP)
+ return;
+
+ /* AK: following checks seem to be broken on modern chipsets. FIXME */
+
+ if (reason & 0x80)
+ mem_parity_error(reason, regs);
+ if (reason & 0x40)
+ io_check_error(reason, regs);
+}
+
+/* runs on IST stack. */
+asmlinkage void __kprobes do_int3(struct pt_regs * regs, long error_code)
+{
+ if (notify_die(DIE_INT3, "int3", regs, error_code, 3, SIGTRAP) == NOTIFY_STOP) {
+ return;
+ }
+ preempt_conditional_sti(regs);
+ do_trap(3, SIGTRAP, "int3", regs, error_code, NULL);
+ preempt_conditional_cli(regs);
+}
+
+/* Help handler running on IST stack to switch back to user stack
+ for scheduling or signal handling. The actual stack switch is done in
+ entry.S */
+asmlinkage __kprobes struct pt_regs *sync_regs(struct pt_regs *eregs)
+{
+ struct pt_regs *regs = eregs;
+ /* Did already sync */
+ if (eregs == (struct pt_regs *)eregs->rsp)
+ ;
+ /* Exception from user space */
+ else if (user_mode(eregs))
+ regs = task_pt_regs(current);
+ /* Exception from kernel and interrupts are enabled. Move to
+ kernel process stack. */
+ else if (eregs->eflags & X86_EFLAGS_IF)
+ regs = (struct pt_regs *)(eregs->rsp -= sizeof(struct pt_regs));
+ if (eregs != regs)
+ *regs = *eregs;
+ return regs;
+}
+
+/* runs on IST stack. */
+asmlinkage void __kprobes do_debug(struct pt_regs * regs,
+ unsigned long error_code)
+{
+ unsigned long condition;
+ struct task_struct *tsk = current;
+ siginfo_t info;
+
+ get_debugreg(condition, 6);
+
+ if (notify_die(DIE_DEBUG, "debug", regs, condition, error_code,
+ SIGTRAP) == NOTIFY_STOP)
+ return;
+
+ preempt_conditional_sti(regs);
+
+ /* Mask out spurious debug traps due to lazy DR7 setting */
+ if (condition & (DR_TRAP0|DR_TRAP1|DR_TRAP2|DR_TRAP3)) {
+ if (!tsk->thread.debugreg7) {
+ goto clear_dr7;
+ }
+ }
+
+ tsk->thread.debugreg6 = condition;
+
+ /* Mask out spurious TF errors due to lazy TF clearing */
+ if (condition & DR_STEP) {
+ /*
+ * The TF error should be masked out only if the current
+ * process is not traced and if the TRAP flag has been set
+ * previously by a tracing process (condition detected by
+ * the PT_DTRACE flag); remember that the i386 TRAP flag
+ * can be modified by the process itself in user mode,
+ * allowing programs to debug themselves without the ptrace()
+ * interface.
+ */
+ if (!user_mode(regs))
+ goto clear_TF_reenable;
+ /*
+ * Was the TF flag set by a debugger? If so, clear it now,
+ * so that register information is correct.
+ */
+ if (tsk->ptrace & PT_DTRACE) {
+ regs->eflags &= ~TF_MASK;
+ tsk->ptrace &= ~PT_DTRACE;
+ }
+ }
+
+ /* Ok, finally something we can handle */
+ tsk->thread.trap_no = 1;
+ tsk->thread.error_code = error_code;
+ info.si_signo = SIGTRAP;
+ info.si_errno = 0;
+ info.si_code = TRAP_BRKPT;
+ info.si_addr = user_mode(regs) ? (void __user *)regs->rip : NULL;
+ force_sig_info(SIGTRAP, &info, tsk);
+
+clear_dr7:
+ set_debugreg(0UL, 7);
+ preempt_conditional_cli(regs);
+ return;
+
+clear_TF_reenable:
+ set_tsk_thread_flag(tsk, TIF_SINGLESTEP);
+ regs->eflags &= ~TF_MASK;
+ preempt_conditional_cli(regs);
+}
+
+static int kernel_math_error(struct pt_regs *regs, const char *str, int trapnr)
+{
+ const struct exception_table_entry *fixup;
+ fixup = search_exception_tables(regs->rip);
+ if (fixup) {
+ regs->rip = fixup->fixup;
+ return 1;
+ }
+ notify_die(DIE_GPF, str, regs, 0, trapnr, SIGFPE);
+ /* Illegal floating point operation in the kernel */
+ current->thread.trap_no = trapnr;
+ die(str, regs, 0);
+ return 0;
+}
+
+/*
+ * Note that we play around with the 'TS' bit in an attempt to get
+ * the correct behaviour even in the presence of the asynchronous
+ * IRQ13 behaviour
+ */
+asmlinkage void do_coprocessor_error(struct pt_regs *regs)
+{
+ void __user *rip = (void __user *)(regs->rip);
+ struct task_struct * task;
+ siginfo_t info;
+ unsigned short cwd, swd;
+
+ conditional_sti(regs);
+ if (!user_mode(regs) &&
+ kernel_math_error(regs, "kernel x87 math error", 16))
+ return;
+
+ /*
+ * Save the info for the exception handler and clear the error.
+ */
+ task = current;
+ save_init_fpu(task);
+ task->thread.trap_no = 16;
+ task->thread.error_code = 0;
+ info.si_signo = SIGFPE;
+ info.si_errno = 0;
+ info.si_code = __SI_FAULT;
+ info.si_addr = rip;
+ /*
+ * (~cwd & swd) will mask out exceptions that are not set to unmasked
+ * status. 0x3f is the exception bits in these regs, 0x200 is the
+ * C1 reg you need in case of a stack fault, 0x040 is the stack
+ * fault bit. We should only be taking one exception at a time,
+ * so if this combination doesn't produce any single exception,
+ * then we have a bad program that isn't synchronizing its FPU usage
+ * and it will suffer the consequences since we won't be able to
+ * fully reproduce the context of the exception
+ */
+ cwd = get_fpu_cwd(task);
+ swd = get_fpu_swd(task);
+ switch (swd & ~cwd & 0x3f) {
+ case 0x000:
+ default:
+ break;
+ case 0x001: /* Invalid Op */
+ /*
+ * swd & 0x240 == 0x040: Stack Underflow
+ * swd & 0x240 == 0x240: Stack Overflow
+ * User must clear the SF bit (0x40) if set
+ */
+ info.si_code = FPE_FLTINV;
+ break;
+ case 0x002: /* Denormalize */
+ case 0x010: /* Underflow */
+ info.si_code = FPE_FLTUND;
+ break;
+ case 0x004: /* Zero Divide */
+ info.si_code = FPE_FLTDIV;
+ break;
+ case 0x008: /* Overflow */
+ info.si_code = FPE_FLTOVF;
+ break;
+ case 0x020: /* Precision */
+ info.si_code = FPE_FLTRES;
+ break;
+ }
+ force_sig_info(SIGFPE, &info, task);
+}
+
+asmlinkage void bad_intr(void)
+{
+ printk("bad interrupt");
+}
+
+asmlinkage void do_simd_coprocessor_error(struct pt_regs *regs)
+{
+ void __user *rip = (void __user *)(regs->rip);
+ struct task_struct * task;
+ siginfo_t info;
+ unsigned short mxcsr;
+
+ conditional_sti(regs);
+ if (!user_mode(regs) &&
+ kernel_math_error(regs, "kernel simd math error", 19))
+ return;
+
+ /*
+ * Save the info for the exception handler and clear the error.
+ */
+ task = current;
+ save_init_fpu(task);
+ task->thread.trap_no = 19;
+ task->thread.error_code = 0;
+ info.si_signo = SIGFPE;
+ info.si_errno = 0;
+ info.si_code = __SI_FAULT;
+ info.si_addr = rip;
+ /*
+ * The SIMD FPU exceptions are handled a little differently, as there
+ * is only a single status/control register. Thus, to determine which
+ * unmasked exception was caught we must mask the exception mask bits
+ * at 0x1f80, and then use these to mask the exception bits at 0x3f.
+ */
+ mxcsr = get_fpu_mxcsr(task);
+ switch (~((mxcsr & 0x1f80) >> 7) & (mxcsr & 0x3f)) {
+ case 0x000:
+ default:
+ break;
+ case 0x001: /* Invalid Op */
+ info.si_code = FPE_FLTINV;
+ break;
+ case 0x002: /* Denormalize */
+ case 0x010: /* Underflow */
+ info.si_code = FPE_FLTUND;
+ break;
+ case 0x004: /* Zero Divide */
+ info.si_code = FPE_FLTDIV;
+ break;
+ case 0x008: /* Overflow */
+ info.si_code = FPE_FLTOVF;
+ break;
+ case 0x020: /* Precision */
+ info.si_code = FPE_FLTRES;
+ break;
+ }
+ force_sig_info(SIGFPE, &info, task);
+}
+
+asmlinkage void do_spurious_interrupt_bug(struct pt_regs * regs)
+{
+}
+
+#if 0
+asmlinkage void __attribute__((weak)) smp_thermal_interrupt(void)
+{
+}
+#endif
+
+asmlinkage void __attribute__((weak)) mce_threshold_interrupt(void)
+{
+}
+
+/*
+ * 'math_state_restore()' saves the current math information in the
+ * old math state array, and gets the new ones from the current task
+ *
+ * Careful.. There are problems with IBM-designed IRQ13 behaviour.
+ * Don't touch unless you *really* know how it works.
+ */
+asmlinkage void math_state_restore(void)
+{
+ struct task_struct *me = current;
+ /* clts(); */ /* 'clts' is done for us by Xen during virtual trap. */
+
+ if (!used_math())
+ init_fpu(me);
+ restore_fpu_checking(&me->thread.i387.fxsave);
+ task_thread_info(me)->status |= TS_USEDFPU;
+}
+
+
+/*
+ * NB. All these are "interrupt gates" (i.e. events_mask is set) because we
+ * specify <dpl>|4 in the second field.
+ */
+static trap_info_t __cpuinitdata trap_table[] = {
+ { 0, 0|4, __KERNEL_CS, (unsigned long)divide_error },
+ { 1, 0|4, __KERNEL_CS, (unsigned long)debug },
+ { 3, 3|4, __KERNEL_CS, (unsigned long)int3 },
+ { 4, 3|4, __KERNEL_CS, (unsigned long)overflow },
+ { 5, 0|4, __KERNEL_CS, (unsigned long)bounds },
+ { 6, 0|4, __KERNEL_CS, (unsigned long)invalid_op },
+ { 7, 0|4, __KERNEL_CS, (unsigned long)device_not_available },
+ { 9, 0|4, __KERNEL_CS, (unsigned long)coprocessor_segment_overrun},
+ { 10, 0|4, __KERNEL_CS, (unsigned long)invalid_TSS },
+ { 11, 0|4, __KERNEL_CS, (unsigned long)segment_not_present },
+ { 12, 0|4, __KERNEL_CS, (unsigned long)stack_segment },
+ { 13, 0|4, __KERNEL_CS, (unsigned long)general_protection },
+ { 14, 0|4, __KERNEL_CS, (unsigned long)page_fault },
+ { 15, 0|4, __KERNEL_CS, (unsigned long)spurious_interrupt_bug },
+ { 16, 0|4, __KERNEL_CS, (unsigned long)coprocessor_error },
+ { 17, 0|4, __KERNEL_CS, (unsigned long)alignment_check },
+#ifdef CONFIG_X86_MCE
+ { 18, 0|4, __KERNEL_CS, (unsigned long)machine_check },
+#endif
+ { 19, 0|4, __KERNEL_CS, (unsigned long)simd_coprocessor_error },
+#ifdef CONFIG_IA32_EMULATION
+ { IA32_SYSCALL_VECTOR, 3, __KERNEL_CS, (unsigned long)ia32_syscall},
+#endif
+ { 0, 0, 0, 0 }
+};
+
+void __init trap_init(void)
+{
+ int ret;
+
+ ret = HYPERVISOR_set_trap_table(trap_table);
+ if (ret)
+ printk("HYPERVISOR_set_trap_table failed: error %d\n", ret);
+
+ /*
+ * Should be a barrier for any external CPU state.
+ */
+ cpu_init();
+}
+
+void __cpuinit smp_trap_init(trap_info_t *trap_ctxt)
+{
+ const trap_info_t *t = trap_table;
+
+ for (t = trap_table; t->address; t++) {
+ trap_ctxt[t->vector].flags = t->flags;
+ trap_ctxt[t->vector].cs = t->cs;
+ trap_ctxt[t->vector].address = t->address;
+ }
+}
+
+
+/* Actual parsing is done early in setup.c. */
+static int __init oops_dummy(char *s)
+{
+ panic_on_oops = 1;
+ return 1;
+}
+__setup("oops=", oops_dummy);
+
+static int __init kstack_setup(char *s)
+{
+ kstack_depth_to_print = simple_strtoul(s,NULL,0);
+ return 1;
+}
+__setup("kstack=", kstack_setup);
+
+#ifdef CONFIG_STACK_UNWIND
+static int __init call_trace_setup(char *s)
+{
+ if (strcmp(s, "old") == 0)
+ call_trace = -1;
+ else if (strcmp(s, "both") == 0)
+ call_trace = 0;
+ else if (strcmp(s, "newfallback") == 0)
+ call_trace = 1;
+ else if (strcmp(s, "new") == 0)
+ call_trace = 2;
+ return 1;
+}
+__setup("call_trace=", call_trace_setup);
+#endif
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/kernel/vsyscall_64-xen.c 2007-06-18 08:38:13.000000000 +0200
@@ -0,0 +1,227 @@
+/*
+ * linux/arch/x86_64/kernel/vsyscall.c
+ *
+ * Copyright (C) 2001 Andrea Arcangeli <andrea@suse.de> SuSE
+ * Copyright 2003 Andi Kleen, SuSE Labs.
+ *
+ * Thanks to hpa@transmeta.com for some useful hint.
+ * Special thanks to Ingo Molnar for his early experience with
+ * a different vsyscall implementation for Linux/IA32 and for the name.
+ *
+ * vsyscall 1 is located at -10Mbyte, vsyscall 2 is located
+ * at virtual address -10Mbyte+1024bytes etc... There are at max 4
+ * vsyscalls. One vsyscall can reserve more than 1 slot to avoid
+ * jumping out of line if necessary. We cannot add more with this
+ * mechanism because older kernels won't return -ENOSYS.
+ * If we want more than four we need a vDSO.
+ *
+ * Note: the concept clashes with user mode linux. If you use UML and
+ * want per guest time just set the kernel.vsyscall64 sysctl to 0.
+ */
+
+#include <linux/time.h>
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/timer.h>
+#include <linux/seqlock.h>
+#include <linux/jiffies.h>
+#include <linux/sysctl.h>
+
+#include <asm/vsyscall.h>
+#include <asm/pgtable.h>
+#include <asm/page.h>
+#include <asm/fixmap.h>
+#include <asm/errno.h>
+#include <asm/io.h>
+
+#define __vsyscall(nr) __attribute__ ((unused,__section__(".vsyscall_" #nr)))
+
+int __sysctl_vsyscall __section_sysctl_vsyscall = 1;
+seqlock_t __xtime_lock __section_xtime_lock = SEQLOCK_UNLOCKED;
+
+#include <asm/unistd.h>
+
+static __always_inline void timeval_normalize(struct timeval * tv)
+{
+ time_t __sec;
+
+ __sec = tv->tv_usec / 1000000;
+ if (__sec) {
+ tv->tv_usec %= 1000000;
+ tv->tv_sec += __sec;
+ }
+}
+
+static __always_inline void do_vgettimeofday(struct timeval * tv)
+{
+ long sequence, t;
+ unsigned long sec, usec;
+
+ do {
+ sequence = read_seqbegin(&__xtime_lock);
+
+ sec = __xtime.tv_sec;
+ usec = (__xtime.tv_nsec / 1000) +
+ (__jiffies - __wall_jiffies) * (1000000 / HZ);
+
+ if (__vxtime.mode != VXTIME_HPET) {
+ t = get_cycles_sync();
+ if (t < __vxtime.last_tsc)
+ t = __vxtime.last_tsc;
+ usec += ((t - __vxtime.last_tsc) *
+ __vxtime.tsc_quot) >> 32;
+ /* See comment in x86_64 do_gettimeofday. */
+ } else {
+ usec += ((readl((void *)fix_to_virt(VSYSCALL_HPET) + 0xf0) -
+ __vxtime.last) * __vxtime.quot) >> 32;
+ }
+ } while (read_seqretry(&__xtime_lock, sequence));
+
+ tv->tv_sec = sec + usec / 1000000;
+ tv->tv_usec = usec % 1000000;
+}
+
+/* RED-PEN may want to readd seq locking, but then the variable should be write-once. */
+static __always_inline void do_get_tz(struct timezone * tz)
+{
+ *tz = __sys_tz;
+}
+
+static __always_inline int gettimeofday(struct timeval *tv, struct timezone *tz)
+{
+ int ret;
+ asm volatile("vsysc2: syscall"
+ : "=a" (ret)
+ : "0" (__NR_gettimeofday),"D" (tv),"S" (tz) : __syscall_clobber );
+ return ret;
+}
+
+static __always_inline long time_syscall(long *t)
+{
+ long secs;
+ asm volatile("vsysc1: syscall"
+ : "=a" (secs)
+ : "0" (__NR_time),"D" (t) : __syscall_clobber);
+ return secs;
+}
+
+int __vsyscall(0) vgettimeofday(struct timeval * tv, struct timezone * tz)
+{
+ if (!__sysctl_vsyscall)
+ return gettimeofday(tv,tz);
+ if (tv)
+ do_vgettimeofday(tv);
+ if (tz)
+ do_get_tz(tz);
+ return 0;
+}
+
+/* This will break when the xtime seconds get inaccurate, but that is
+ * unlikely */
+time_t __vsyscall(1) vtime(time_t *t)
+{
+ if (!__sysctl_vsyscall)
+ return time_syscall(t);
+ else if (t)
+ *t = __xtime.tv_sec;
+ return __xtime.tv_sec;
+}
+
+long __vsyscall(2) venosys_0(void)
+{
+ return -ENOSYS;
+}
+
+long __vsyscall(3) venosys_1(void)
+{
+ return -ENOSYS;
+}
+
+#ifdef CONFIG_SYSCTL
+
+#define SYSCALL 0x050f
+#define NOP2 0x9090
+
+/*
+ * NOP out syscall in vsyscall page when not needed.
+ */
+static int vsyscall_sysctl_change(ctl_table *ctl, int write, struct file * filp,
+ void __user *buffer, size_t *lenp, loff_t *ppos)
+{
+ extern u16 vsysc1, vsysc2;
+ u16 *map1, *map2;
+ int ret = proc_dointvec(ctl, write, filp, buffer, lenp, ppos);
+ if (!write)
+ return ret;
+ /* gcc has some trouble with __va(__pa()), so just do it this
+ way. */
+ map1 = ioremap(__pa_symbol(&vsysc1), 2);
+ if (!map1)
+ return -ENOMEM;
+ map2 = ioremap(__pa_symbol(&vsysc2), 2);
+ if (!map2) {
+ ret = -ENOMEM;
+ goto out;
+ }
+ if (!sysctl_vsyscall) {
+ *map1 = SYSCALL;
+ *map2 = SYSCALL;
+ } else {
+ *map1 = NOP2;
+ *map2 = NOP2;
+ }
+ iounmap(map2);
+out:
+ iounmap(map1);
+ return ret;
+}
+
+static int vsyscall_sysctl_nostrat(ctl_table *t, int __user *name, int nlen,
+ void __user *oldval, size_t __user *oldlenp,
+ void __user *newval, size_t newlen,
+ void **context)
+{
+ return -ENOSYS;
+}
+
+static ctl_table kernel_table2[] = {
+ { .ctl_name = 99, .procname = "vsyscall64",
+ .data = &sysctl_vsyscall, .maxlen = sizeof(int), .mode = 0644,
+ .strategy = vsyscall_sysctl_nostrat,
+ .proc_handler = vsyscall_sysctl_change },
+ { 0, }
+};
+
+static ctl_table kernel_root_table2[] = {
+ { .ctl_name = CTL_KERN, .procname = "kernel", .mode = 0555,
+ .child = kernel_table2 },
+ { 0 },
+};
+
+#endif
+
+static void __init map_vsyscall(void)
+{
+ extern char __vsyscall_0;
+ unsigned long physaddr_page0 = __pa_symbol(&__vsyscall_0);
+
+ __set_fixmap(VSYSCALL_FIRST_PAGE, physaddr_page0, PAGE_KERNEL_VSYSCALL);
+}
+
+static int __init vsyscall_init(void)
+{
+ BUG_ON(((unsigned long) &vgettimeofday !=
+ VSYSCALL_ADDR(__NR_vgettimeofday)));
+ BUG_ON((unsigned long) &vtime != VSYSCALL_ADDR(__NR_vtime));
+ BUG_ON((VSYSCALL_ADDR(0) != __fix_to_virt(VSYSCALL_FIRST_PAGE)));
+ map_vsyscall();
+#ifdef CONFIG_XEN
+ sysctl_vsyscall = 0; /* disable vgettimeofay() */
+#endif
+#ifdef CONFIG_SYSCTL
+ register_sysctl_table(kernel_root_table2, 0);
+#endif
+ return 0;
+}
+
+__initcall(vsyscall_init);
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/kernel/xen_entry_64.S 2008-04-02 12:34:02.000000000 +0200
@@ -0,0 +1,36 @@
+/*
+ * Copied from arch/xen/i386/kernel/entry.S
+ */
+/* Offsets into shared_info_t. */
+#define evtchn_upcall_pending /* 0 */
+#define evtchn_upcall_mask 1
+
+#define sizeof_vcpu_shift 6
+
+#ifdef CONFIG_SMP
+//#define preempt_disable(reg) incl threadinfo_preempt_count(reg)
+//#define preempt_enable(reg) decl threadinfo_preempt_count(reg)
+#define preempt_disable(reg)
+#define preempt_enable(reg)
+#define XEN_GET_VCPU_INFO(reg) preempt_disable(%rbp) ; \
+ movq %gs:pda_cpunumber,reg ; \
+ shl $32, reg ; \
+ shr $32-sizeof_vcpu_shift,reg ; \
+ addq HYPERVISOR_shared_info,reg
+#define XEN_PUT_VCPU_INFO(reg) preempt_enable(%rbp) ; \
+#define XEN_PUT_VCPU_INFO_fixup .byte 0xff,0xff,0xff
+#else
+#define XEN_GET_VCPU_INFO(reg) movq HYPERVISOR_shared_info,reg
+#define XEN_PUT_VCPU_INFO(reg)
+#define XEN_PUT_VCPU_INFO_fixup
+#endif
+
+#define XEN_LOCKED_BLOCK_EVENTS(reg) movb $1,evtchn_upcall_mask(reg)
+#define XEN_LOCKED_UNBLOCK_EVENTS(reg) movb $0,evtchn_upcall_mask(reg)
+#define XEN_BLOCK_EVENTS(reg) XEN_GET_VCPU_INFO(reg) ; \
+ XEN_LOCKED_BLOCK_EVENTS(reg) ; \
+ XEN_PUT_VCPU_INFO(reg)
+#define XEN_UNBLOCK_EVENTS(reg) XEN_GET_VCPU_INFO(reg) ; \
+ XEN_LOCKED_UNBLOCK_EVENTS(reg) ; \
+ XEN_PUT_VCPU_INFO(reg)
+#define XEN_TEST_PENDING(reg) testb $0xFF,evtchn_upcall_pending(reg)
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/mm/fault_64-xen.c 2010-09-23 15:39:04.000000000 +0200
@@ -0,0 +1,730 @@
+/*
+ * linux/arch/x86-64/mm/fault.c
+ *
+ * Copyright (C) 1995 Linus Torvalds
+ * Copyright (C) 2001,2002 Andi Kleen, SuSE Labs.
+ */
+
+#include <linux/signal.h>
+#include <linux/sched.h>
+#include <linux/kernel.h>
+#include <linux/errno.h>
+#include <linux/string.h>
+#include <linux/types.h>
+#include <linux/ptrace.h>
+#include <linux/mman.h>
+#include <linux/mm.h>
+#include <linux/smp.h>
+#include <linux/smp_lock.h>
+#include <linux/interrupt.h>
+#include <linux/init.h>
+#include <linux/tty.h>
+#include <linux/vt_kern.h> /* For unblank_screen() */
+#include <linux/compiler.h>
+#include <linux/module.h>
+#include <linux/kprobes.h>
+
+#include <asm/system.h>
+#include <asm/uaccess.h>
+#include <asm/pgalloc.h>
+#include <asm/smp.h>
+#include <asm/tlbflush.h>
+#include <asm/proto.h>
+#include <asm/kdebug.h>
+#include <asm-generic/sections.h>
+
+/* Page fault error code bits */
+#define PF_PROT (1<<0) /* or no page found */
+#define PF_WRITE (1<<1)
+#define PF_USER (1<<2)
+#define PF_RSVD (1<<3)
+#define PF_INSTR (1<<4)
+
+#ifdef CONFIG_KPROBES
+ATOMIC_NOTIFIER_HEAD(notify_page_fault_chain);
+
+/* Hook to register for page fault notifications */
+int register_page_fault_notifier(struct notifier_block *nb)
+{
+ vmalloc_sync_all();
+ return atomic_notifier_chain_register(&notify_page_fault_chain, nb);
+}
+
+int unregister_page_fault_notifier(struct notifier_block *nb)
+{
+ return atomic_notifier_chain_unregister(&notify_page_fault_chain, nb);
+}
+
+static inline int notify_page_fault(enum die_val val, const char *str,
+ struct pt_regs *regs, long err, int trap, int sig)
+{
+ struct die_args args = {
+ .regs = regs,
+ .str = str,
+ .err = err,
+ .trapnr = trap,
+ .signr = sig
+ };
+ return atomic_notifier_call_chain(&notify_page_fault_chain, val, &args);
+}
+#else
+static inline int notify_page_fault(enum die_val val, const char *str,
+ struct pt_regs *regs, long err, int trap, int sig)
+{
+ return NOTIFY_DONE;
+}
+#endif
+
+void bust_spinlocks(int yes)
+{
+ int loglevel_save = console_loglevel;
+ if (yes) {
+ oops_in_progress = 1;
+ } else {
+#ifdef CONFIG_VT
+ unblank_screen();
+#endif
+ oops_in_progress = 0;
+ /*
+ * OK, the message is on the console. Now we call printk()
+ * without oops_in_progress set so that printk will give klogd
+ * a poke. Hold onto your hats...
+ */
+ console_loglevel = 15; /* NMI oopser may have shut the console up */
+ printk(" ");
+ console_loglevel = loglevel_save;
+ }
+}
+
+/* Sometimes the CPU reports invalid exceptions on prefetch.
+ Check that here and ignore.
+ Opcode checker based on code by Richard Brunner */
+static noinline int is_prefetch(struct pt_regs *regs, unsigned long addr,
+ unsigned long error_code)
+{
+ unsigned char *instr;
+ int scan_more = 1;
+ int prefetch = 0;
+ unsigned char *max_instr;
+
+ /* If it was a exec fault ignore */
+ if (error_code & PF_INSTR)
+ return 0;
+
+ instr = (unsigned char *)convert_rip_to_linear(current, regs);
+ max_instr = instr + 15;
+
+ if (user_mode(regs) && instr >= (unsigned char *)TASK_SIZE)
+ return 0;
+
+ while (scan_more && instr < max_instr) {
+ unsigned char opcode;
+ unsigned char instr_hi;
+ unsigned char instr_lo;
+
+ if (__get_user(opcode, instr))
+ break;
+
+ instr_hi = opcode & 0xf0;
+ instr_lo = opcode & 0x0f;
+ instr++;
+
+ switch (instr_hi) {
+ case 0x20:
+ case 0x30:
+ /* Values 0x26,0x2E,0x36,0x3E are valid x86
+ prefixes. In long mode, the CPU will signal
+ invalid opcode if some of these prefixes are
+ present so we will never get here anyway */
+ scan_more = ((instr_lo & 7) == 0x6);
+ break;
+
+ case 0x40:
+ /* In AMD64 long mode, 0x40 to 0x4F are valid REX prefixes
+ Need to figure out under what instruction mode the
+ instruction was issued ... */
+ /* Could check the LDT for lm, but for now it's good
+ enough to assume that long mode only uses well known
+ segments or kernel. */
+ scan_more = (!user_mode(regs)) || (regs->cs == __USER_CS);
+ break;
+
+ case 0x60:
+ /* 0x64 thru 0x67 are valid prefixes in all modes. */
+ scan_more = (instr_lo & 0xC) == 0x4;
+ break;
+ case 0xF0:
+ /* 0xF0, 0xF2, and 0xF3 are valid prefixes in all modes. */
+ scan_more = !instr_lo || (instr_lo>>1) == 1;
+ break;
+ case 0x00:
+ /* Prefetch instruction is 0x0F0D or 0x0F18 */
+ scan_more = 0;
+ if (__get_user(opcode, instr))
+ break;
+ prefetch = (instr_lo == 0xF) &&
+ (opcode == 0x0D || opcode == 0x18);
+ break;
+ default:
+ scan_more = 0;
+ break;
+ }
+ }
+ return prefetch;
+}
+
+static int bad_address(void *p)
+{
+ unsigned long dummy;
+ return __get_user(dummy, (unsigned long *)p);
+}
+
+void dump_pagetable(unsigned long address)
+{
+ pgd_t *pgd;
+ pud_t *pud;
+ pmd_t *pmd;
+ pte_t *pte;
+
+ pgd = __va(read_cr3() & PHYSICAL_PAGE_MASK);
+ pgd += pgd_index(address);
+ if (bad_address(pgd)) goto bad;
+ printk("PGD %lx ", pgd_val(*pgd));
+ if (!pgd_present(*pgd)) goto ret;
+
+ pud = pud_offset(pgd, address);
+ if (bad_address(pud)) goto bad;
+ printk("PUD %lx ", pud_val(*pud));
+ if (!pud_present(*pud)) goto ret;
+
+ pmd = pmd_offset(pud, address);
+ if (bad_address(pmd)) goto bad;
+ printk("PMD %lx ", pmd_val(*pmd));
+ if (!pmd_present(*pmd)) goto ret;
+
+ pte = pte_offset_kernel(pmd, address);
+ if (bad_address(pte)) goto bad;
+ printk("PTE %lx", pte_val(*pte));
+ret:
+ printk("\n");
+ return;
+bad:
+ printk("BAD\n");
+}
+
+static const char errata93_warning[] =
+KERN_ERR "******* Your BIOS seems to not contain a fix for K8 errata #93\n"
+KERN_ERR "******* Working around it, but it may cause SEGVs or burn power.\n"
+KERN_ERR "******* Please consider a BIOS update.\n"
+KERN_ERR "******* Disabling USB legacy in the BIOS may also help.\n";
+
+/* Workaround for K8 erratum #93 & buggy BIOS.
+ BIOS SMM functions are required to use a specific workaround
+ to avoid corruption of the 64bit RIP register on C stepping K8.
+ A lot of BIOS that didn't get tested properly miss this.
+ The OS sees this as a page fault with the upper 32bits of RIP cleared.
+ Try to work around it here.
+ Note we only handle faults in kernel here. */
+
+static int is_errata93(struct pt_regs *regs, unsigned long address)
+{
+ static int warned;
+ if (address != regs->rip)
+ return 0;
+ if ((address >> 32) != 0)
+ return 0;
+ address |= 0xffffffffUL << 32;
+ if ((address >= (u64)_stext && address <= (u64)_etext) ||
+ (address >= MODULES_VADDR && address <= MODULES_END)) {
+ if (!warned) {
+ printk(errata93_warning);
+ warned = 1;
+ }
+ regs->rip = address;
+ return 1;
+ }
+ return 0;
+}
+
+int unhandled_signal(struct task_struct *tsk, int sig)
+{
+ if (tsk->pid == 1)
+ return 1;
+ if (tsk->ptrace & PT_PTRACED)
+ return 0;
+ return (tsk->sighand->action[sig-1].sa.sa_handler == SIG_IGN) ||
+ (tsk->sighand->action[sig-1].sa.sa_handler == SIG_DFL);
+}
+
+static noinline void pgtable_bad(unsigned long address, struct pt_regs *regs,
+ unsigned long error_code)
+{
+ unsigned long flags = oops_begin();
+ struct task_struct *tsk;
+
+ printk(KERN_ALERT "%s: Corrupted page table at address %lx\n",
+ current->comm, address);
+ dump_pagetable(address);
+ tsk = current;
+ tsk->thread.cr2 = address;
+ tsk->thread.trap_no = 14;
+ tsk->thread.error_code = error_code;
+ __die("Bad pagetable", regs, error_code);
+ oops_end(flags);
+ do_exit(SIGKILL);
+}
+
+/*
+ * Handle a fault on the vmalloc area
+ *
+ * This assumes no large pages in there.
+ */
+static int vmalloc_fault(unsigned long address)
+{
+ pgd_t *pgd, *pgd_ref;
+ pud_t *pud, *pud_ref;
+ pmd_t *pmd, *pmd_ref;
+ pte_t *pte, *pte_ref;
+
+ /* Copy kernel mappings over when needed. This can also
+ happen within a race in page table update. In the later
+ case just flush. */
+
+ /* On Xen the line below does not always work. Needs investigating! */
+ /*pgd = pgd_offset(current->mm ?: &init_mm, address);*/
+ pgd = __va(read_cr3() & PHYSICAL_PAGE_MASK);
+ pgd += pgd_index(address);
+ pgd_ref = pgd_offset_k(address);
+ if (pgd_none(*pgd_ref))
+ return -1;
+ if (pgd_none(*pgd))
+ set_pgd(pgd, *pgd_ref);
+ else
+ BUG_ON(pgd_page(*pgd) != pgd_page(*pgd_ref));
+
+ /* Below here mismatches are bugs because these lower tables
+ are shared */
+
+ pud = pud_offset(pgd, address);
+ pud_ref = pud_offset(pgd_ref, address);
+ if (pud_none(*pud_ref))
+ return -1;
+ if (pud_none(*pud) || pud_page(*pud) != pud_page(*pud_ref))
+ BUG();
+ pmd = pmd_offset(pud, address);
+ pmd_ref = pmd_offset(pud_ref, address);
+ if (pmd_none(*pmd_ref))
+ return -1;
+ if (pmd_none(*pmd) || pmd_page(*pmd) != pmd_page(*pmd_ref))
+ BUG();
+ pte_ref = pte_offset_kernel(pmd_ref, address);
+ if (!pte_present(*pte_ref))
+ return -1;
+ pte = pte_offset_kernel(pmd, address);
+ /* Don't use pte_page here, because the mappings can point
+ outside mem_map, and the NUMA hash lookup cannot handle
+ that. */
+ if (!pte_present(*pte) || pte_pfn(*pte) != pte_pfn(*pte_ref))
+ BUG();
+ return 0;
+}
+
+int page_fault_trace = 0;
+int exception_trace = 1;
+
+
+#define MEM_VERBOSE 1
+
+#ifdef MEM_VERBOSE
+#define MEM_LOG(_f, _a...) \
+ printk("fault.c:[%d]-> " _f "\n", \
+ __LINE__ , ## _a )
+#else
+#define MEM_LOG(_f, _a...) ((void)0)
+#endif
+
+static int spurious_fault(struct pt_regs *regs,
+ unsigned long address,
+ unsigned long error_code)
+{
+ pgd_t *pgd;
+ pud_t *pud;
+ pmd_t *pmd;
+ pte_t *pte;
+
+#ifdef CONFIG_XEN
+ /* Faults in hypervisor area are never spurious. */
+ if ((address >= HYPERVISOR_VIRT_START) &&
+ (address < HYPERVISOR_VIRT_END))
+ return 0;
+#endif
+
+ /* Reserved-bit violation or user access to kernel space? */
+ if (error_code & (PF_RSVD|PF_USER))
+ return 0;
+
+ pgd = init_mm.pgd + pgd_index(address);
+ if (!pgd_present(*pgd))
+ return 0;
+
+ pud = pud_offset(pgd, address);
+ if (!pud_present(*pud))
+ return 0;
+
+ pmd = pmd_offset(pud, address);
+ if (!pmd_present(*pmd))
+ return 0;
+
+ pte = pte_offset_kernel(pmd, address);
+ if (!pte_present(*pte))
+ return 0;
+ if ((error_code & PF_WRITE) && !pte_write(*pte))
+ return 0;
+ if ((error_code & PF_INSTR) && (__pte_val(*pte) & _PAGE_NX))
+ return 0;
+
+ return 1;
+}
+
+/*
+ * This routine handles page faults. It determines the address,
+ * and the problem, and then passes it off to one of the appropriate
+ * routines.
+ */
+asmlinkage void __kprobes do_page_fault(struct pt_regs *regs,
+ unsigned long error_code)
+{
+ struct task_struct *tsk;
+ struct mm_struct *mm;
+ struct vm_area_struct * vma;
+ unsigned long address;
+ const struct exception_table_entry *fixup;
+ int write;
+ unsigned long flags;
+ siginfo_t info;
+
+ if (!user_mode(regs))
+ error_code &= ~PF_USER; /* means kernel */
+
+ tsk = current;
+ mm = tsk->mm;
+ prefetchw(&mm->mmap_sem);
+
+ /* get the address */
+ address = current_vcpu_info()->arch.cr2;
+
+ info.si_code = SEGV_MAPERR;
+
+
+ /*
+ * We fault-in kernel-space virtual memory on-demand. The
+ * 'reference' page table is init_mm.pgd.
+ *
+ * NOTE! We MUST NOT take any locks for this case. We may
+ * be in an interrupt or a critical region, and should
+ * only copy the information from the master page table,
+ * nothing more.
+ *
+ * This verifies that the fault happens in kernel space
+ * (error_code & 4) == 0, and that the fault was not a
+ * protection error (error_code & 9) == 0.
+ */
+ if (unlikely(address >= TASK_SIZE64)) {
+ /*
+ * Don't check for the module range here: its PML4
+ * is always initialized because it's shared with the main
+ * kernel text. Only vmalloc may need PML4 syncups.
+ */
+ if (!(error_code & (PF_RSVD|PF_USER|PF_PROT)) &&
+ ((address >= VMALLOC_START && address < VMALLOC_END))) {
+ if (vmalloc_fault(address) >= 0)
+ return;
+ }
+ /* Can take a spurious fault if mapping changes R/O -> R/W. */
+ if (spurious_fault(regs, address, error_code))
+ return;
+ if (notify_page_fault(DIE_PAGE_FAULT, "page fault", regs, error_code, 14,
+ SIGSEGV) == NOTIFY_STOP)
+ return;
+ /*
+ * Don't take the mm semaphore here. If we fixup a prefetch
+ * fault we could otherwise deadlock.
+ */
+ goto bad_area_nosemaphore;
+ }
+
+ if (notify_page_fault(DIE_PAGE_FAULT, "page fault", regs, error_code, 14,
+ SIGSEGV) == NOTIFY_STOP)
+ return;
+
+ if (likely(regs->eflags & X86_EFLAGS_IF))
+ local_irq_enable();
+
+ if (unlikely(page_fault_trace))
+ printk("pagefault rip:%lx rsp:%lx cs:%lu ss:%lu address %lx error %lx\n",
+ regs->rip,regs->rsp,regs->cs,regs->ss,address,error_code);
+
+ if (unlikely(error_code & PF_RSVD))
+ pgtable_bad(address, regs, error_code);
+
+ /*
+ * If we're in an interrupt or have no user
+ * context, we must not take the fault..
+ */
+ if (unlikely(in_atomic() || !mm))
+ goto bad_area_nosemaphore;
+
+ again:
+ /* When running in the kernel we expect faults to occur only to
+ * addresses in user space. All other faults represent errors in the
+ * kernel and should generate an OOPS. Unfortunatly, in the case of an
+ * erroneous fault occurring in a code path which already holds mmap_sem
+ * we will deadlock attempting to validate the fault against the
+ * address space. Luckily the kernel only validly references user
+ * space from well defined areas of code, which are listed in the
+ * exceptions table.
+ *
+ * As the vast majority of faults will be valid we will only perform
+ * the source reference check when there is a possibilty of a deadlock.
+ * Attempt to lock the address space, if we cannot we then validate the
+ * source. If this is invalid we can skip the address space check,
+ * thus avoiding the deadlock.
+ */
+ if (!down_read_trylock(&mm->mmap_sem)) {
+ if ((error_code & PF_USER) == 0 &&
+ !search_exception_tables(regs->rip))
+ goto bad_area_nosemaphore;
+ down_read(&mm->mmap_sem);
+ }
+
+ vma = find_vma(mm, address);
+ if (!vma)
+ goto bad_area;
+ if (likely(vma->vm_start <= address))
+ goto good_area;
+ if (!(vma->vm_flags & VM_GROWSDOWN))
+ goto bad_area;
+ if (error_code & 4) {
+ /* Allow userspace just enough access below the stack pointer
+ * to let the 'enter' instruction work.
+ */
+ if (address + 65536 + 32 * sizeof(unsigned long) < regs->rsp)
+ goto bad_area;
+ }
+ if (expand_stack(vma, address))
+ goto bad_area;
+/*
+ * Ok, we have a good vm_area for this memory access, so
+ * we can handle it..
+ */
+good_area:
+ info.si_code = SEGV_ACCERR;
+ write = 0;
+ switch (error_code & (PF_PROT|PF_WRITE)) {
+ default: /* 3: write, present */
+ /* fall through */
+ case PF_WRITE: /* write, not present */
+ if (!(vma->vm_flags & VM_WRITE))
+ goto bad_area;
+ write++;
+ break;
+ case PF_PROT: /* read, present */
+ goto bad_area;
+ case 0: /* read, not present */
+ if (!(vma->vm_flags & (VM_READ | VM_EXEC)))
+ goto bad_area;
+ }
+
+ /*
+ * If for any reason at all we couldn't handle the fault,
+ * make sure we exit gracefully rather than endlessly redo
+ * the fault.
+ */
+ switch (handle_mm_fault(mm, vma, address, write)) {
+ case VM_FAULT_MINOR:
+ tsk->min_flt++;
+ break;
+ case VM_FAULT_MAJOR:
+ tsk->maj_flt++;
+ break;
+ case VM_FAULT_SIGBUS:
+ goto do_sigbus;
+ default:
+ goto out_of_memory;
+ }
+
+ up_read(&mm->mmap_sem);
+ return;
+
+/*
+ * Something tried to access memory that isn't in our memory map..
+ * Fix it, but check if it's kernel or user first..
+ */
+bad_area:
+ up_read(&mm->mmap_sem);
+
+bad_area_nosemaphore:
+ /* User mode accesses just cause a SIGSEGV */
+ if (error_code & PF_USER) {
+ if (is_prefetch(regs, address, error_code))
+ return;
+
+ /* Work around K8 erratum #100 K8 in compat mode
+ occasionally jumps to illegal addresses >4GB. We
+ catch this here in the page fault handler because
+ these addresses are not reachable. Just detect this
+ case and return. Any code segment in LDT is
+ compatibility mode. */
+ if ((regs->cs == __USER32_CS || (regs->cs & (1<<2))) &&
+ (address >> 32))
+ return;
+
+ if (exception_trace && unhandled_signal(tsk, SIGSEGV)) {
+ printk(
+ "%s%s[%d]: segfault at %016lx rip %016lx rsp %016lx error %lx\n",
+ tsk->pid > 1 ? KERN_INFO : KERN_EMERG,
+ tsk->comm, tsk->pid, address, regs->rip,
+ regs->rsp, error_code);
+ }
+
+ tsk->thread.cr2 = address;
+ /* Kernel addresses are always protection faults */
+ tsk->thread.error_code = error_code | (address >= TASK_SIZE);
+ tsk->thread.trap_no = 14;
+ info.si_signo = SIGSEGV;
+ info.si_errno = 0;
+ /* info.si_code has been set above */
+ info.si_addr = (void __user *)address;
+ force_sig_info(SIGSEGV, &info, tsk);
+ return;
+ }
+
+no_context:
+
+ /* Are we prepared to handle this kernel fault? */
+ fixup = search_exception_tables(regs->rip);
+ if (fixup) {
+ regs->rip = fixup->fixup;
+ return;
+ }
+
+ /*
+ * Hall of shame of CPU/BIOS bugs.
+ */
+
+ if (is_prefetch(regs, address, error_code))
+ return;
+
+ if (is_errata93(regs, address))
+ return;
+
+/*
+ * Oops. The kernel tried to access some bad page. We'll have to
+ * terminate things with extreme prejudice.
+ */
+
+ flags = oops_begin();
+
+ if (address < PAGE_SIZE)
+ printk(KERN_ALERT "Unable to handle kernel NULL pointer dereference");
+ else
+ printk(KERN_ALERT "Unable to handle kernel paging request");
+ printk(" at %016lx RIP: \n" KERN_ALERT,address);
+ printk_address(regs->rip);
+ dump_pagetable(address);
+ tsk->thread.cr2 = address;
+ tsk->thread.trap_no = 14;
+ tsk->thread.error_code = error_code;
+ __die("Oops", regs, error_code);
+ /* Executive summary in case the body of the oops scrolled away */
+ printk(KERN_EMERG "CR2: %016lx\n", address);
+ oops_end(flags);
+ do_exit(SIGKILL);
+
+/*
+ * We ran out of memory, or some other thing happened to us that made
+ * us unable to handle the page fault gracefully.
+ */
+out_of_memory:
+ up_read(&mm->mmap_sem);
+ if (current->pid == 1) {
+ yield();
+ goto again;
+ }
+ printk("VM: killing process %s\n", tsk->comm);
+ if (error_code & 4)
+ do_exit(SIGKILL);
+ goto no_context;
+
+do_sigbus:
+ up_read(&mm->mmap_sem);
+
+ /* Kernel mode? Handle exceptions or die */
+ if (!(error_code & PF_USER))
+ goto no_context;
+
+ tsk->thread.cr2 = address;
+ tsk->thread.error_code = error_code;
+ tsk->thread.trap_no = 14;
+ info.si_signo = SIGBUS;
+ info.si_errno = 0;
+ info.si_code = BUS_ADRERR;
+ info.si_addr = (void __user *)address;
+ force_sig_info(SIGBUS, &info, tsk);
+ return;
+}
+
+DEFINE_SPINLOCK(pgd_lock);
+struct page *pgd_list;
+
+#define pgd_page_table(what, pg) \
+ spin_##what(&((struct mm_struct *)(pg)->mapping)->page_table_lock)
+
+void vmalloc_sync_all(void)
+{
+ /* Note that races in the updates of insync and start aren't
+ problematic:
+ insync can only get set bits added, and updates to start are only
+ improving performance (without affecting correctness if undone). */
+ static DECLARE_BITMAP(insync, PTRS_PER_PGD);
+ static unsigned long start = VMALLOC_START & PGDIR_MASK;
+ unsigned long address;
+
+ for (address = start; address <= VMALLOC_END; address += PGDIR_SIZE) {
+ if (!test_bit(pgd_index(address), insync)) {
+ const pgd_t *pgd_ref = pgd_offset_k(address);
+ struct page *page;
+
+ if (pgd_none(*pgd_ref))
+ continue;
+ spin_lock(&pgd_lock);
+ for (page = pgd_list; page;
+ page = (struct page *)page->index) {
+ pgd_t *pgd;
+ pgd = (pgd_t *)page_address(page) + pgd_index(address);
+
+ pgd_page_table(lock, page);
+ if (pgd_none(*pgd))
+ set_pgd(pgd, *pgd_ref);
+ else
+ BUG_ON(pgd_page(*pgd) != pgd_page(*pgd_ref));
+ pgd_page_table(unlock, page);
+ }
+ spin_unlock(&pgd_lock);
+ set_bit(pgd_index(address), insync);
+ }
+ if (address == start)
+ start = address + PGDIR_SIZE;
+ }
+ /* Check that there is no need to do the same for the modules area. */
+ BUILD_BUG_ON(!(MODULES_VADDR > __START_KERNEL));
+ BUILD_BUG_ON(!(((MODULES_END - 1) & PGDIR_MASK) ==
+ (__START_KERNEL & PGDIR_MASK)));
+}
+
+static int __init enable_pagefaulttrace(char *str)
+{
+ page_fault_trace = 1;
+ return 1;
+}
+__setup("pagefaulttrace", enable_pagefaulttrace);
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/mm/init_64-xen.c 2010-04-29 09:34:47.000000000 +0200
@@ -0,0 +1,1244 @@
+/*
+ * linux/arch/x86_64/mm/init.c
+ *
+ * Copyright (C) 1995 Linus Torvalds
+ * Copyright (C) 2000 Pavel Machek <pavel@suse.cz>
+ * Copyright (C) 2002,2003 Andi Kleen <ak@suse.de>
+ *
+ * Jun Nakajima <jun.nakajima@intel.com>
+ * Modified for Xen.
+ */
+
+#include <linux/signal.h>
+#include <linux/sched.h>
+#include <linux/kernel.h>
+#include <linux/errno.h>
+#include <linux/string.h>
+#include <linux/types.h>
+#include <linux/ptrace.h>
+#include <linux/mman.h>
+#include <linux/mm.h>
+#include <linux/swap.h>
+#include <linux/smp.h>
+#include <linux/init.h>
+#include <linux/pagemap.h>
+#include <linux/bootmem.h>
+#include <linux/proc_fs.h>
+#include <linux/pci.h>
+#include <linux/poison.h>
+#include <linux/dma-mapping.h>
+#include <linux/module.h>
+#include <linux/memory_hotplug.h>
+
+#include <asm/processor.h>
+#include <asm/system.h>
+#include <asm/uaccess.h>
+#include <asm/pgtable.h>
+#include <asm/pgalloc.h>
+#include <asm/dma.h>
+#include <asm/fixmap.h>
+#include <asm/e820.h>
+#include <asm/apic.h>
+#include <asm/tlb.h>
+#include <asm/mmu_context.h>
+#include <asm/proto.h>
+#include <asm/smp.h>
+#include <asm/sections.h>
+
+#include <xen/features.h>
+
+#ifndef Dprintk
+#define Dprintk(x...)
+#endif
+
+struct dma_mapping_ops* dma_ops;
+EXPORT_SYMBOL(dma_ops);
+
+#if CONFIG_XEN_COMPAT <= 0x030002
+unsigned int __kernel_page_user;
+EXPORT_SYMBOL(__kernel_page_user);
+#endif
+
+int after_bootmem;
+
+static unsigned long dma_reserve __initdata;
+
+DEFINE_PER_CPU(struct mmu_gather, mmu_gathers);
+extern unsigned long start_pfn;
+
+/*
+ * Use this until direct mapping is established, i.e. before __va() is
+ * available in init_memory_mapping().
+ */
+
+#define addr_to_page(addr, page) \
+ (addr) &= PHYSICAL_PAGE_MASK; \
+ (page) = ((unsigned long *) ((unsigned long) \
+ (((mfn_to_pfn((addr) >> PAGE_SHIFT)) << PAGE_SHIFT) + \
+ __START_KERNEL_map)))
+
+static void __meminit early_make_page_readonly(void *va, unsigned int feature)
+{
+ unsigned long addr, _va = (unsigned long)va;
+ pte_t pte, *ptep;
+ unsigned long *page = (unsigned long *) init_level4_pgt;
+
+ BUG_ON(after_bootmem);
+
+ if (xen_feature(feature))
+ return;
+
+ addr = (unsigned long) page[pgd_index(_va)];
+ addr_to_page(addr, page);
+
+ addr = page[pud_index(_va)];
+ addr_to_page(addr, page);
+
+ addr = page[pmd_index(_va)];
+ addr_to_page(addr, page);
+
+ ptep = (pte_t *) &page[pte_index(_va)];
+
+ pte.pte = ptep->pte & ~_PAGE_RW;
+ if (HYPERVISOR_update_va_mapping(_va, pte, 0))
+ BUG();
+}
+
+static void __make_page_readonly(void *va)
+{
+ pgd_t *pgd; pud_t *pud; pmd_t *pmd; pte_t pte, *ptep;
+ unsigned long addr = (unsigned long) va;
+
+ pgd = pgd_offset_k(addr);
+ pud = pud_offset(pgd, addr);
+ pmd = pmd_offset(pud, addr);
+ ptep = pte_offset_kernel(pmd, addr);
+
+ pte.pte = ptep->pte & ~_PAGE_RW;
+ if (HYPERVISOR_update_va_mapping(addr, pte, 0))
+ xen_l1_entry_update(ptep, pte); /* fallback */
+
+ if ((addr >= VMALLOC_START) && (addr < VMALLOC_END))
+ __make_page_readonly(__va(pte_pfn(pte) << PAGE_SHIFT));
+}
+
+static void __make_page_writable(void *va)
+{
+ pgd_t *pgd; pud_t *pud; pmd_t *pmd; pte_t pte, *ptep;
+ unsigned long addr = (unsigned long) va;
+
+ pgd = pgd_offset_k(addr);
+ pud = pud_offset(pgd, addr);
+ pmd = pmd_offset(pud, addr);
+ ptep = pte_offset_kernel(pmd, addr);
+
+ pte.pte = ptep->pte | _PAGE_RW;
+ if (HYPERVISOR_update_va_mapping(addr, pte, 0))
+ xen_l1_entry_update(ptep, pte); /* fallback */
+
+ if ((addr >= VMALLOC_START) && (addr < VMALLOC_END))
+ __make_page_writable(__va(pte_pfn(pte) << PAGE_SHIFT));
+}
+
+void make_page_readonly(void *va, unsigned int feature)
+{
+ if (!xen_feature(feature))
+ __make_page_readonly(va);
+}
+
+void make_page_writable(void *va, unsigned int feature)
+{
+ if (!xen_feature(feature))
+ __make_page_writable(va);
+}
+
+void make_pages_readonly(void *va, unsigned nr, unsigned int feature)
+{
+ if (xen_feature(feature))
+ return;
+
+ while (nr-- != 0) {
+ __make_page_readonly(va);
+ va = (void*)((unsigned long)va + PAGE_SIZE);
+ }
+}
+
+void make_pages_writable(void *va, unsigned nr, unsigned int feature)
+{
+ if (xen_feature(feature))
+ return;
+
+ while (nr-- != 0) {
+ __make_page_writable(va);
+ va = (void*)((unsigned long)va + PAGE_SIZE);
+ }
+}
+
+/*
+ * NOTE: pagetable_init alloc all the fixmap pagetables contiguous on the
+ * physical space so we can cache the place of the first one and move
+ * around without checking the pgd every time.
+ */
+
+void show_mem(void)
+{
+ long i, total = 0, reserved = 0;
+ long shared = 0, cached = 0;
+ pg_data_t *pgdat;
+ struct page *page;
+
+ printk(KERN_INFO "Mem-info:\n");
+ show_free_areas();
+ printk(KERN_INFO "Free swap: %6ldkB\n", nr_swap_pages<<(PAGE_SHIFT-10));
+
+ for_each_online_pgdat(pgdat) {
+ for (i = 0; i < pgdat->node_spanned_pages; ++i) {
+ page = pfn_to_page(pgdat->node_start_pfn + i);
+ total++;
+ if (PageReserved(page))
+ reserved++;
+ else if (PageSwapCache(page))
+ cached++;
+ else if (page_count(page))
+ shared += page_count(page) - 1;
+ }
+ }
+ printk(KERN_INFO "%lu pages of RAM\n", total);
+ printk(KERN_INFO "%lu reserved pages\n",reserved);
+ printk(KERN_INFO "%lu pages shared\n",shared);
+ printk(KERN_INFO "%lu pages swap cached\n",cached);
+}
+
+
+static __init void *spp_getpage(void)
+{
+ void *ptr;
+ if (after_bootmem)
+ ptr = (void *) get_zeroed_page(GFP_ATOMIC);
+ else if (start_pfn < table_end) {
+ ptr = __va(start_pfn << PAGE_SHIFT);
+ start_pfn++;
+ memset(ptr, 0, PAGE_SIZE);
+ } else
+ ptr = alloc_bootmem_pages(PAGE_SIZE);
+ if (!ptr || ((unsigned long)ptr & ~PAGE_MASK))
+ panic("set_pte_phys: cannot allocate page data %s\n", after_bootmem?"after bootmem":"");
+
+ Dprintk("spp_getpage %p\n", ptr);
+ return ptr;
+}
+
+#define pgd_offset_u(address) (__user_pgd(init_level4_pgt) + pgd_index(address))
+#define pud_offset_u(address) (level3_user_pgt + pud_index(address))
+
+static __init void set_pte_phys(unsigned long vaddr,
+ unsigned long phys, pgprot_t prot, int user_mode)
+{
+ pgd_t *pgd;
+ pud_t *pud;
+ pmd_t *pmd;
+ pte_t *pte, new_pte;
+
+ Dprintk("set_pte_phys %lx to %lx\n", vaddr, phys);
+
+ pgd = (user_mode ? pgd_offset_u(vaddr) : pgd_offset_k(vaddr));
+ if (pgd_none(*pgd)) {
+ printk("PGD FIXMAP MISSING, it should be setup in head.S!\n");
+ return;
+ }
+ pud = (user_mode ? pud_offset_u(vaddr) : pud_offset(pgd, vaddr));
+ if (pud_none(*pud)) {
+ pmd = (pmd_t *) spp_getpage();
+ make_page_readonly(pmd, XENFEAT_writable_page_tables);
+ set_pud(pud, __pud(__pa(pmd) | _KERNPG_TABLE | _PAGE_USER));
+ if (pmd != pmd_offset(pud, 0)) {
+ printk("PAGETABLE BUG #01! %p <-> %p\n", pmd, pmd_offset(pud,0));
+ return;
+ }
+ }
+ pmd = pmd_offset(pud, vaddr);
+ if (pmd_none(*pmd)) {
+ pte = (pte_t *) spp_getpage();
+ make_page_readonly(pte, XENFEAT_writable_page_tables);
+ set_pmd(pmd, __pmd(__pa(pte) | _KERNPG_TABLE | _PAGE_USER));
+ if (pte != pte_offset_kernel(pmd, 0)) {
+ printk("PAGETABLE BUG #02!\n");
+ return;
+ }
+ }
+ if (pgprot_val(prot))
+ new_pte = pfn_pte(phys >> PAGE_SHIFT, prot);
+ else
+ new_pte = __pte(0);
+
+ pte = pte_offset_kernel(pmd, vaddr);
+ if (!pte_none(*pte) && __pte_val(new_pte) &&
+ __pte_val(*pte) != (__pte_val(new_pte) & __supported_pte_mask))
+ pte_ERROR(*pte);
+ set_pte(pte, new_pte);
+
+ /*
+ * It's enough to flush this one mapping.
+ * (PGE mappings get flushed as well)
+ */
+ __flush_tlb_one(vaddr);
+}
+
+static __init void set_pte_phys_ma(unsigned long vaddr,
+ unsigned long phys, pgprot_t prot)
+{
+ pgd_t *pgd;
+ pud_t *pud;
+ pmd_t *pmd;
+ pte_t *pte, new_pte;
+
+ Dprintk("set_pte_phys %lx to %lx\n", vaddr, phys);
+
+ pgd = pgd_offset_k(vaddr);
+ if (pgd_none(*pgd)) {
+ printk("PGD FIXMAP MISSING, it should be setup in head.S!\n");
+ return;
+ }
+ pud = pud_offset(pgd, vaddr);
+ if (pud_none(*pud)) {
+
+ pmd = (pmd_t *) spp_getpage();
+ make_page_readonly(pmd, XENFEAT_writable_page_tables);
+ set_pud(pud, __pud(__pa(pmd) | _KERNPG_TABLE | _PAGE_USER));
+ if (pmd != pmd_offset(pud, 0)) {
+ printk("PAGETABLE BUG #01! %p <-> %p\n", pmd, pmd_offset(pud,0));
+ return;
+ }
+ }
+ pmd = pmd_offset(pud, vaddr);
+ if (pmd_none(*pmd)) {
+ pte = (pte_t *) spp_getpage();
+ make_page_readonly(pte, XENFEAT_writable_page_tables);
+ set_pmd(pmd, __pmd(__pa(pte) | _KERNPG_TABLE | _PAGE_USER));
+ if (pte != pte_offset_kernel(pmd, 0)) {
+ printk("PAGETABLE BUG #02!\n");
+ return;
+ }
+ }
+ new_pte = pfn_pte_ma(phys >> PAGE_SHIFT, prot);
+
+ pte = pte_offset_kernel(pmd, vaddr);
+ if (!pte_none(*pte) && __pte_val(new_pte) &&
+#ifdef CONFIG_ACPI
+ /* __acpi_map_table() fails to properly call clear_fixmap() */
+ (vaddr < __fix_to_virt(FIX_ACPI_END) ||
+ vaddr > __fix_to_virt(FIX_ACPI_BEGIN)) &&
+#endif
+ __pte_val(*pte) != (__pte_val(new_pte) & __supported_pte_mask))
+ pte_ERROR(*pte);
+ set_pte(pte, new_pte);
+
+ /*
+ * It's enough to flush this one mapping.
+ * (PGE mappings get flushed as well)
+ */
+ __flush_tlb_one(vaddr);
+}
+
+/* NOTE: this is meant to be run only at boot */
+void __init
+__set_fixmap (enum fixed_addresses idx, unsigned long phys, pgprot_t prot)
+{
+ unsigned long address = __fix_to_virt(idx);
+
+ if (idx >= __end_of_fixed_addresses) {
+ printk("Invalid __set_fixmap\n");
+ return;
+ }
+ switch (idx) {
+ case VSYSCALL_LAST_PAGE ... VSYSCALL_FIRST_PAGE:
+ set_pte_phys(address, phys, prot, 0);
+ set_pte_phys(address, phys, prot, 1);
+ break;
+ default:
+ set_pte_phys_ma(address, phys, prot);
+ break;
+ }
+}
+
+unsigned long __initdata table_start, table_end;
+
+static __meminit void *alloc_static_page(unsigned long *phys)
+{
+ unsigned long va = (start_pfn << PAGE_SHIFT) + __START_KERNEL_map;
+
+ if (after_bootmem) {
+ void *adr = (void *)get_zeroed_page(GFP_ATOMIC);
+
+ *phys = __pa(adr);
+ return adr;
+ }
+
+ *phys = start_pfn << PAGE_SHIFT;
+ start_pfn++;
+ memset((void *)va, 0, PAGE_SIZE);
+ return (void *)va;
+}
+
+#define PTE_SIZE PAGE_SIZE
+
+static inline int make_readonly(unsigned long paddr)
+{
+ extern char __vsyscall_0;
+ int readonly = 0;
+
+ /* Make new page tables read-only. */
+ if (!xen_feature(XENFEAT_writable_page_tables)
+ && (paddr >= (table_start << PAGE_SHIFT))
+ && (paddr < (table_end << PAGE_SHIFT)))
+ readonly = 1;
+ /* Make old page tables read-only. */
+ if (!xen_feature(XENFEAT_writable_page_tables)
+ && (paddr >= (xen_start_info->pt_base - __START_KERNEL_map))
+ && (paddr < (start_pfn << PAGE_SHIFT)))
+ readonly = 1;
+
+ /*
+ * No need for writable mapping of kernel image. This also ensures that
+ * page and descriptor tables embedded inside don't have writable
+ * mappings. Exclude the vsyscall area here, allowing alternative
+ * instruction patching to work.
+ */
+ if ((paddr >= __pa_symbol(&_text)) && (paddr < __pa_symbol(&_end))
+ && !(paddr >= __pa_symbol(&__vsyscall_0)
+ && paddr < __pa_symbol(&__vsyscall_0) + PAGE_SIZE))
+ readonly = 1;
+
+ return readonly;
+}
+
+#ifndef CONFIG_XEN
+/* Must run before zap_low_mappings */
+__init void *early_ioremap(unsigned long addr, unsigned long size)
+{
+ unsigned long map = round_down(addr, LARGE_PAGE_SIZE);
+
+ /* actually usually some more */
+ if (size >= LARGE_PAGE_SIZE) {
+ printk("SMBIOS area too long %lu\n", size);
+ return NULL;
+ }
+ set_pmd(temp_mappings[0].pmd, __pmd(map | _KERNPG_TABLE | _PAGE_PSE));
+ map += LARGE_PAGE_SIZE;
+ set_pmd(temp_mappings[1].pmd, __pmd(map | _KERNPG_TABLE | _PAGE_PSE));
+ __flush_tlb();
+ return temp_mappings[0].address + (addr & (LARGE_PAGE_SIZE-1));
+}
+
+/* To avoid virtual aliases later */
+__init void early_iounmap(void *addr, unsigned long size)
+{
+ if ((void *)round_down((unsigned long)addr, LARGE_PAGE_SIZE) != temp_mappings[0].address)
+ printk("early_iounmap: bad address %p\n", addr);
+ set_pmd(temp_mappings[0].pmd, __pmd(0));
+ set_pmd(temp_mappings[1].pmd, __pmd(0));
+ __flush_tlb();
+}
+#endif
+
+static void __meminit
+phys_pmd_init(pmd_t *pmd, unsigned long address, unsigned long end)
+{
+ int i, k;
+
+ for (i = 0; i < PTRS_PER_PMD; pmd++, i++) {
+ unsigned long pte_phys;
+ pte_t *pte, *pte_save;
+
+ if (address >= end)
+ break;
+ pte = alloc_static_page(&pte_phys);
+ pte_save = pte;
+ for (k = 0; k < PTRS_PER_PTE; pte++, k++, address += PTE_SIZE) {
+ unsigned long pteval = address | _PAGE_NX | _KERNPG_TABLE;
+
+ if (address >= end ||
+ (!after_bootmem &&
+ (address >> PAGE_SHIFT) >= xen_start_info->nr_pages))
+ pteval = 0;
+ else if (make_readonly(address))
+ pteval &= ~_PAGE_RW;
+ set_pte(pte, __pte(pteval & __supported_pte_mask));
+ }
+ if (!after_bootmem) {
+ early_make_page_readonly(pte_save, XENFEAT_writable_page_tables);
+ *pmd = __pmd(pte_phys | _KERNPG_TABLE);
+ } else {
+ make_page_readonly(pte_save, XENFEAT_writable_page_tables);
+ set_pmd(pmd, __pmd(pte_phys | _KERNPG_TABLE));
+ }
+ }
+}
+
+static void __meminit
+phys_pmd_update(pud_t *pud, unsigned long address, unsigned long end)
+{
+ pmd_t *pmd = pmd_offset(pud, (unsigned long)__va(address));
+
+ if (pmd_none(*pmd)) {
+ spin_lock(&init_mm.page_table_lock);
+ phys_pmd_init(pmd, address, end);
+ spin_unlock(&init_mm.page_table_lock);
+ __flush_tlb_all();
+ }
+}
+
+static void __meminit phys_pud_init(pud_t *pud, unsigned long address, unsigned long end)
+{
+ long i = pud_index(address);
+
+ pud = pud + i;
+
+ if (after_bootmem && pud_val(*pud)) {
+ phys_pmd_update(pud, address, end);
+ return;
+ }
+
+ for (; i < PTRS_PER_PUD; pud++, i++) {
+ unsigned long paddr, pmd_phys;
+ pmd_t *pmd;
+
+ paddr = (address & PGDIR_MASK) + i*PUD_SIZE;
+ if (paddr >= end)
+ break;
+
+ pmd = alloc_static_page(&pmd_phys);
+
+ spin_lock(&init_mm.page_table_lock);
+ *pud = __pud(pmd_phys | _KERNPG_TABLE);
+ phys_pmd_init(pmd, paddr, end);
+ spin_unlock(&init_mm.page_table_lock);
+
+ early_make_page_readonly(pmd, XENFEAT_writable_page_tables);
+ }
+ __flush_tlb();
+}
+
+void __init xen_init_pt(void)
+{
+ unsigned long addr, *page;
+
+ /* Find the initial pte page that was built for us. */
+ page = (unsigned long *)xen_start_info->pt_base;
+ addr = page[pgd_index(__START_KERNEL_map)];
+ addr_to_page(addr, page);
+
+#if CONFIG_XEN_COMPAT <= 0x030002
+ /* On Xen 3.0.2 and older we may need to explicitly specify _PAGE_USER
+ in kernel PTEs. We check that here. */
+ if (HYPERVISOR_xen_version(XENVER_version, NULL) <= 0x30000) {
+ unsigned long *pg;
+ pte_t pte;
+
+ /* Mess with the initial mapping of page 0. It's not needed. */
+ BUILD_BUG_ON(__START_KERNEL <= __START_KERNEL_map);
+ addr = page[pud_index(__START_KERNEL_map)];
+ addr_to_page(addr, pg);
+ addr = pg[pmd_index(__START_KERNEL_map)];
+ addr_to_page(addr, pg);
+ pte.pte = pg[pte_index(__START_KERNEL_map)];
+ BUG_ON(!(pte.pte & _PAGE_PRESENT));
+
+ /* If _PAGE_USER isn't set, we obviously do not need it. */
+ if (pte.pte & _PAGE_USER) {
+ /* _PAGE_USER is needed, but is it set implicitly? */
+ pte.pte &= ~_PAGE_USER;
+ if ((HYPERVISOR_update_va_mapping(__START_KERNEL_map,
+ pte, 0) != 0) ||
+ !(pg[pte_index(__START_KERNEL_map)] & _PAGE_USER))
+ /* We need to explicitly specify _PAGE_USER. */
+ __kernel_page_user = _PAGE_USER;
+ }
+ }
+#endif
+
+ /* Construct mapping of initial pte page in our own directories. */
+ init_level4_pgt[pgd_index(__START_KERNEL_map)] =
+ __pgd(__pa_symbol(level3_kernel_pgt) | _PAGE_TABLE);
+ memcpy(level3_kernel_pgt + pud_index(__START_KERNEL_map),
+ page + pud_index(__START_KERNEL_map),
+ (PTRS_PER_PUD - pud_index(__START_KERNEL_map))
+ * sizeof(*level3_kernel_pgt));
+
+ __user_pgd(init_level4_pgt)[pgd_index(VSYSCALL_START)] =
+ __pgd(__pa_symbol(level3_user_pgt) | _PAGE_TABLE);
+
+ early_make_page_readonly(init_level4_pgt,
+ XENFEAT_writable_page_tables);
+ early_make_page_readonly(__user_pgd(init_level4_pgt),
+ XENFEAT_writable_page_tables);
+ early_make_page_readonly(level3_kernel_pgt,
+ XENFEAT_writable_page_tables);
+ early_make_page_readonly(level3_user_pgt,
+ XENFEAT_writable_page_tables);
+
+ if (!xen_feature(XENFEAT_writable_page_tables)) {
+ xen_pgd_pin(__pa_symbol(init_level4_pgt));
+ xen_pgd_pin(__pa_symbol(__user_pgd(init_level4_pgt)));
+ }
+}
+
+static void __init extend_init_mapping(unsigned long tables_space)
+{
+ unsigned long va = __START_KERNEL_map;
+ unsigned long phys, addr, *pte_page;
+ pmd_t *pmd;
+ pte_t *pte, new_pte;
+ unsigned long *page = (unsigned long *)init_level4_pgt;
+
+ addr = page[pgd_index(va)];
+ addr_to_page(addr, page);
+ addr = page[pud_index(va)];
+ addr_to_page(addr, page);
+
+ /* Kill mapping of low 1MB. */
+ while (va < (unsigned long)&_text) {
+ if (HYPERVISOR_update_va_mapping(va, __pte_ma(0), 0))
+ BUG();
+ va += PAGE_SIZE;
+ }
+
+ /* Ensure init mappings cover kernel text/data and initial tables. */
+ while (va < (__START_KERNEL_map
+ + (start_pfn << PAGE_SHIFT)
+ + tables_space)) {
+ if (!(pmd_index(va) | pte_index(va))) {
+ pud_t *pud;
+
+ page = (unsigned long *)init_level4_pgt;
+ addr = page[pgd_index(va)];
+ addr_to_page(addr, page);
+ pud = (pud_t *)&page[pud_index(va)];
+ if (pud_none(*pud)) {
+ page = alloc_static_page(&phys);
+ early_make_page_readonly(
+ page, XENFEAT_writable_page_tables);
+ set_pud(pud, __pud(phys | _KERNPG_TABLE));
+ } else {
+ addr = page[pud_index(va)];
+ addr_to_page(addr, page);
+ }
+ }
+ pmd = (pmd_t *)&page[pmd_index(va)];
+ if (pmd_none(*pmd)) {
+ pte_page = alloc_static_page(&phys);
+ early_make_page_readonly(
+ pte_page, XENFEAT_writable_page_tables);
+ set_pmd(pmd, __pmd(phys | _KERNPG_TABLE));
+ } else {
+ addr = page[pmd_index(va)];
+ addr_to_page(addr, pte_page);
+ }
+ pte = (pte_t *)&pte_page[pte_index(va)];
+ if (pte_none(*pte)) {
+ new_pte = pfn_pte(
+ (va - __START_KERNEL_map) >> PAGE_SHIFT,
+ __pgprot(_KERNPG_TABLE));
+ xen_l1_entry_update(pte, new_pte);
+ }
+ va += PAGE_SIZE;
+ }
+
+ /* Finally, blow away any spurious initial mappings. */
+ while (1) {
+ if (!(pmd_index(va) | pte_index(va))) {
+ page = (unsigned long *)init_level4_pgt;
+ addr = page[pgd_index(va)];
+ addr_to_page(addr, page);
+ if (pud_none(((pud_t *)page)[pud_index(va)]))
+ break;
+ addr = page[pud_index(va)];
+ addr_to_page(addr, page);
+ }
+ pmd = (pmd_t *)&page[pmd_index(va)];
+ if (pmd_none(*pmd))
+ break;
+ if (HYPERVISOR_update_va_mapping(va, __pte_ma(0), 0))
+ BUG();
+ va += PAGE_SIZE;
+ }
+}
+
+static void __init find_early_table_space(unsigned long end)
+{
+ unsigned long puds, pmds, ptes, tables;
+
+ puds = (end + PUD_SIZE - 1) >> PUD_SHIFT;
+ pmds = (end + PMD_SIZE - 1) >> PMD_SHIFT;
+ ptes = (end + PTE_SIZE - 1) >> PAGE_SHIFT;
+
+ tables = round_up(puds * 8, PAGE_SIZE) +
+ round_up(pmds * 8, PAGE_SIZE) +
+ round_up(ptes * 8, PAGE_SIZE);
+
+ extend_init_mapping(tables);
+
+ table_start = start_pfn;
+ table_end = table_start + (tables>>PAGE_SHIFT);
+
+ early_printk("kernel direct mapping tables up to %lx @ %lx-%lx\n",
+ end, table_start << PAGE_SHIFT,
+ (table_start << PAGE_SHIFT) + tables);
+}
+
+static void xen_finish_init_mapping(void)
+{
+ unsigned long i, start, end;
+
+ /* Re-vector virtual addresses pointing into the initial
+ mapping to the just-established permanent ones. */
+ xen_start_info = __va(__pa(xen_start_info));
+ xen_start_info->pt_base = (unsigned long)
+ __va(__pa(xen_start_info->pt_base));
+ if (!xen_feature(XENFEAT_auto_translated_physmap)) {
+ phys_to_machine_mapping =
+ __va(__pa(xen_start_info->mfn_list));
+ xen_start_info->mfn_list = (unsigned long)
+ phys_to_machine_mapping;
+ }
+ if (xen_start_info->mod_start)
+ xen_start_info->mod_start = (unsigned long)
+ __va(__pa(xen_start_info->mod_start));
+
+ /* Destroy the Xen-created mappings beyond the kernel image as
+ * well as the temporary mappings created above. Prevents
+ * overlap with modules area (if init mapping is very big).
+ */
+ start = PAGE_ALIGN((unsigned long)_end);
+ end = __START_KERNEL_map + (table_end << PAGE_SHIFT);
+ for (; start < end; start += PAGE_SIZE)
+ if (HYPERVISOR_update_va_mapping(start, __pte_ma(0), 0))
+ BUG();
+
+ /* Allocate pte's for initial fixmaps from 'start_pfn' allocator. */
+ table_end = ~0UL;
+
+ /*
+ * Prefetch pte's for the bt_ioremap() area. It gets used before the
+ * boot-time allocator is online, so allocate-on-demand would fail.
+ */
+ for (i = FIX_BTMAP_END; i <= FIX_BTMAP_BEGIN; i++)
+ __set_fixmap(i, 0, __pgprot(0));
+
+ /* Switch to the real shared_info page, and clear the dummy page. */
+ set_fixmap(FIX_SHARED_INFO, xen_start_info->shared_info);
+ HYPERVISOR_shared_info = (shared_info_t *)fix_to_virt(FIX_SHARED_INFO);
+ memset(empty_zero_page, 0, sizeof(empty_zero_page));
+
+ /* Set up mapping of lowest 1MB of physical memory. */
+ for (i = 0; i < NR_FIX_ISAMAPS; i++)
+ if (is_initial_xendomain())
+ set_fixmap(FIX_ISAMAP_BEGIN - i, i * PAGE_SIZE);
+ else
+ __set_fixmap(FIX_ISAMAP_BEGIN - i,
+ virt_to_mfn(empty_zero_page)
+ << PAGE_SHIFT,
+ PAGE_KERNEL_RO);
+
+ /* Disable the 'start_pfn' allocator. */
+ table_end = start_pfn;
+}
+
+/* Setup the direct mapping of the physical memory at PAGE_OFFSET.
+ This runs before bootmem is initialized and gets pages directly from the
+ physical memory. To access them they are temporarily mapped. */
+void __meminit init_memory_mapping(unsigned long start, unsigned long end)
+{
+ unsigned long next;
+
+ Dprintk("init_memory_mapping\n");
+
+ /*
+ * Find space for the kernel direct mapping tables.
+ * Later we should allocate these tables in the local node of the memory
+ * mapped. Unfortunately this is done currently before the nodes are
+ * discovered.
+ */
+ if (!after_bootmem)
+ find_early_table_space(end);
+
+ start = (unsigned long)__va(start);
+ end = (unsigned long)__va(end);
+
+ for (; start < end; start = next) {
+ unsigned long pud_phys;
+ pgd_t *pgd = pgd_offset_k(start);
+ pud_t *pud;
+
+ if (after_bootmem)
+ pud = pud_offset(pgd, start & PGDIR_MASK);
+ else
+ pud = alloc_static_page(&pud_phys);
+ next = start + PGDIR_SIZE;
+ if (next > end)
+ next = end;
+ phys_pud_init(pud, __pa(start), __pa(next));
+ if (!after_bootmem) {
+ early_make_page_readonly(pud, XENFEAT_writable_page_tables);
+ set_pgd(pgd_offset_k(start), mk_kernel_pgd(pud_phys));
+ }
+ }
+
+ if (!after_bootmem) {
+ BUG_ON(start_pfn != table_end);
+ xen_finish_init_mapping();
+ }
+
+ __flush_tlb_all();
+}
+
+void __cpuinit zap_low_mappings(int cpu)
+{
+ /* this is not required for Xen */
+#if 0
+ swap_low_mappings();
+#endif
+}
+
+/* Compute zone sizes for the DMA and DMA32 zones in a node. */
+__init void
+size_zones(unsigned long *z, unsigned long *h,
+ unsigned long start_pfn, unsigned long end_pfn)
+{
+ int i;
+ unsigned long w;
+
+ for (i = 0; i < MAX_NR_ZONES; i++)
+ z[i] = 0;
+
+ if (start_pfn < MAX_DMA_PFN)
+ z[ZONE_DMA] = MAX_DMA_PFN - start_pfn;
+ if (start_pfn < MAX_DMA32_PFN) {
+ unsigned long dma32_pfn = MAX_DMA32_PFN;
+ if (dma32_pfn > end_pfn)
+ dma32_pfn = end_pfn;
+ z[ZONE_DMA32] = dma32_pfn - start_pfn;
+ }
+ z[ZONE_NORMAL] = end_pfn - start_pfn;
+
+ /* Remove lower zones from higher ones. */
+ w = 0;
+ for (i = 0; i < MAX_NR_ZONES; i++) {
+ if (z[i])
+ z[i] -= w;
+ w += z[i];
+ }
+
+ /* Compute holes */
+ w = start_pfn;
+ for (i = 0; i < MAX_NR_ZONES; i++) {
+ unsigned long s = w;
+ w += z[i];
+ h[i] = e820_hole_size(s, w);
+ }
+
+ /* Add the space pace needed for mem_map to the holes too. */
+ for (i = 0; i < MAX_NR_ZONES; i++)
+ h[i] += (z[i] * sizeof(struct page)) / PAGE_SIZE;
+
+ /* The 16MB DMA zone has the kernel and other misc mappings.
+ Account them too */
+ if (h[ZONE_DMA]) {
+ h[ZONE_DMA] += dma_reserve;
+ if (h[ZONE_DMA] >= z[ZONE_DMA]) {
+ printk(KERN_WARNING
+ "Kernel too large and filling up ZONE_DMA?\n");
+ h[ZONE_DMA] = z[ZONE_DMA];
+ }
+ }
+}
+
+#ifndef CONFIG_NUMA
+void __init paging_init(void)
+{
+ unsigned long zones[MAX_NR_ZONES], holes[MAX_NR_ZONES];
+
+ memory_present(0, 0, end_pfn);
+ sparse_init();
+ size_zones(zones, holes, 0, end_pfn);
+ free_area_init_node(0, NODE_DATA(0), zones,
+ __pa(PAGE_OFFSET) >> PAGE_SHIFT, holes);
+
+ init_mm.context.pinned = 1;
+}
+#endif
+
+/* Unmap a kernel mapping if it exists. This is useful to avoid prefetches
+ from the CPU leading to inconsistent cache lines. address and size
+ must be aligned to 2MB boundaries.
+ Does nothing when the mapping doesn't exist. */
+void __init clear_kernel_mapping(unsigned long address, unsigned long size)
+{
+ unsigned long end = address + size;
+
+ BUG_ON(address & ~LARGE_PAGE_MASK);
+ BUG_ON(size & ~LARGE_PAGE_MASK);
+
+ for (; address < end; address += LARGE_PAGE_SIZE) {
+ pgd_t *pgd = pgd_offset_k(address);
+ pud_t *pud;
+ pmd_t *pmd;
+ if (pgd_none(*pgd))
+ continue;
+ pud = pud_offset(pgd, address);
+ if (pud_none(*pud))
+ continue;
+ pmd = pmd_offset(pud, address);
+ if (!pmd || pmd_none(*pmd))
+ continue;
+ if (0 == (__pmd_val(*pmd) & _PAGE_PSE)) {
+ /* Could handle this, but it should not happen currently. */
+ printk(KERN_ERR
+ "clear_kernel_mapping: mapping has been split. will leak memory\n");
+ pmd_ERROR(*pmd);
+ }
+ set_pmd(pmd, __pmd(0));
+ }
+ __flush_tlb_all();
+}
+
+/*
+ * Memory hotplug specific functions
+ */
+void online_page(struct page *page)
+{
+ ClearPageReserved(page);
+ init_page_count(page);
+ __free_page(page);
+ totalram_pages++;
+ num_physpages++;
+}
+
+#ifdef CONFIG_MEMORY_HOTPLUG
+/*
+ * XXX: memory_add_physaddr_to_nid() is to find node id from physical address
+ * via probe interface of sysfs. If acpi notifies hot-add event, then it
+ * can tell node id by searching dsdt. But, probe interface doesn't have
+ * node id. So, return 0 as node id at this time.
+ */
+#ifdef CONFIG_NUMA
+int memory_add_physaddr_to_nid(u64 start)
+{
+ return 0;
+}
+#endif
+
+/*
+ * Memory is added always to NORMAL zone. This means you will never get
+ * additional DMA/DMA32 memory.
+ */
+int arch_add_memory(int nid, u64 start, u64 size)
+{
+ struct pglist_data *pgdat = NODE_DATA(nid);
+ struct zone *zone = pgdat->node_zones + MAX_NR_ZONES-2;
+ unsigned long start_pfn = start >> PAGE_SHIFT;
+ unsigned long nr_pages = size >> PAGE_SHIFT;
+ int ret;
+
+ ret = __add_pages(zone, start_pfn, nr_pages);
+ if (ret)
+ goto error;
+
+ init_memory_mapping(start, (start + size -1));
+
+ return ret;
+error:
+ printk("%s: Problem encountered in __add_pages!\n", __func__);
+ return ret;
+}
+EXPORT_SYMBOL_GPL(arch_add_memory);
+
+int remove_memory(u64 start, u64 size)
+{
+ return -EINVAL;
+}
+EXPORT_SYMBOL_GPL(remove_memory);
+
+#else /* CONFIG_MEMORY_HOTPLUG */
+/*
+ * Memory Hotadd without sparsemem. The mem_maps have been allocated in advance,
+ * just online the pages.
+ */
+int __add_pages(struct zone *z, unsigned long start_pfn, unsigned long nr_pages)
+{
+ int err = -EIO;
+ unsigned long pfn;
+ unsigned long total = 0, mem = 0;
+ for (pfn = start_pfn; pfn < start_pfn + nr_pages; pfn++) {
+ if (pfn_valid(pfn)) {
+ online_page(pfn_to_page(pfn));
+ err = 0;
+ mem++;
+ }
+ total++;
+ }
+ if (!err) {
+ z->spanned_pages += total;
+ z->present_pages += mem;
+ z->zone_pgdat->node_spanned_pages += total;
+ z->zone_pgdat->node_present_pages += mem;
+ }
+ return err;
+}
+#endif /* CONFIG_MEMORY_HOTPLUG */
+
+static struct kcore_list kcore_mem, kcore_vmalloc, kcore_kernel, kcore_modules,
+ kcore_vsyscall;
+
+void __init mem_init(void)
+{
+ long codesize, reservedpages, datasize, initsize;
+ unsigned long pfn;
+
+ pci_iommu_alloc();
+
+ /* How many end-of-memory variables you have, grandma! */
+ max_low_pfn = end_pfn;
+ max_pfn = end_pfn;
+ num_physpages = end_pfn;
+ high_memory = (void *) __va(end_pfn * PAGE_SIZE);
+
+ /* clear the zero-page */
+ memset(empty_zero_page, 0, PAGE_SIZE);
+
+ reservedpages = 0;
+
+ /* this will put all low memory onto the freelists */
+#ifdef CONFIG_NUMA
+ totalram_pages = numa_free_all_bootmem();
+#else
+ totalram_pages = free_all_bootmem();
+#endif
+ /* XEN: init and count pages outside initial allocation. */
+ for (pfn = xen_start_info->nr_pages; pfn < max_pfn; pfn++) {
+ ClearPageReserved(pfn_to_page(pfn));
+ init_page_count(pfn_to_page(pfn));
+ totalram_pages++;
+ }
+ reservedpages = end_pfn - totalram_pages - e820_hole_size(0, end_pfn);
+
+ after_bootmem = 1;
+
+ codesize = (unsigned long) &_etext - (unsigned long) &_text;
+ datasize = (unsigned long) &_edata - (unsigned long) &_etext;
+ initsize = (unsigned long) &__init_end - (unsigned long) &__init_begin;
+
+ /* Register memory areas for /proc/kcore */
+ kclist_add(&kcore_mem, __va(0), max_low_pfn << PAGE_SHIFT);
+ kclist_add(&kcore_vmalloc, (void *)VMALLOC_START,
+ VMALLOC_END-VMALLOC_START);
+ kclist_add(&kcore_kernel, &_stext, _end - _stext);
+ kclist_add(&kcore_modules, (void *)MODULES_VADDR, MODULES_LEN);
+ kclist_add(&kcore_vsyscall, (void *)VSYSCALL_START,
+ VSYSCALL_END - VSYSCALL_START);
+
+ printk("Memory: %luk/%luk available (%ldk kernel code, %ldk reserved, %ldk data, %ldk init)\n",
+ (unsigned long) nr_free_pages() << (PAGE_SHIFT-10),
+ end_pfn << (PAGE_SHIFT-10),
+ codesize >> 10,
+ reservedpages << (PAGE_SHIFT-10),
+ datasize >> 10,
+ initsize >> 10);
+
+#ifndef CONFIG_XEN
+#ifdef CONFIG_SMP
+ /*
+ * Sync boot_level4_pgt mappings with the init_level4_pgt
+ * except for the low identity mappings which are already zapped
+ * in init_level4_pgt. This sync-up is essential for AP's bringup
+ */
+ memcpy(boot_level4_pgt+1, init_level4_pgt+1, (PTRS_PER_PGD-1)*sizeof(pgd_t));
+#endif
+#endif
+}
+
+void free_init_pages(char *what, unsigned long begin, unsigned long end)
+{
+ unsigned long addr;
+
+ if (begin >= end)
+ return;
+
+ printk(KERN_INFO "Freeing %s: %ldk freed\n", what, (end - begin) >> 10);
+ for (addr = begin; addr < end; addr += PAGE_SIZE) {
+ ClearPageReserved(virt_to_page(addr));
+ init_page_count(virt_to_page(addr));
+ memset((void *)(addr & ~(PAGE_SIZE-1)),
+ POISON_FREE_INITMEM, PAGE_SIZE);
+ if (addr >= __START_KERNEL_map) {
+ /* make_readonly() reports all kernel addresses. */
+ __make_page_writable(__va(__pa(addr)));
+ if (HYPERVISOR_update_va_mapping(addr, __pte(0), 0)) {
+ pgd_t *pgd = pgd_offset_k(addr);
+ pud_t *pud = pud_offset(pgd, addr);
+ pmd_t *pmd = pmd_offset(pud, addr);
+ pte_t *pte = pte_offset_kernel(pmd, addr);
+
+ xen_l1_entry_update(pte, __pte(0)); /* fallback */
+ }
+ }
+ free_page(addr);
+ totalram_pages++;
+ }
+}
+
+void free_initmem(void)
+{
+ memset(__initdata_begin, POISON_FREE_INITDATA,
+ __initdata_end - __initdata_begin);
+ free_init_pages("unused kernel memory",
+ (unsigned long)(&__init_begin),
+ (unsigned long)(&__init_end));
+}
+
+#ifdef CONFIG_DEBUG_RODATA
+
+void mark_rodata_ro(void)
+{
+ unsigned long addr = (unsigned long)__start_rodata;
+
+ for (; addr < (unsigned long)__end_rodata; addr += PAGE_SIZE)
+ change_page_attr_addr(addr, 1, PAGE_KERNEL_RO);
+
+ printk ("Write protecting the kernel read-only data: %luk\n",
+ (__end_rodata - __start_rodata) >> 10);
+
+ /*
+ * change_page_attr_addr() requires a global_flush_tlb() call after it.
+ * We do this after the printk so that if something went wrong in the
+ * change, the printk gets out at least to give a better debug hint
+ * of who is the culprit.
+ */
+ global_flush_tlb();
+}
+#endif
+
+#ifdef CONFIG_BLK_DEV_INITRD
+void free_initrd_mem(unsigned long start, unsigned long end)
+{
+ free_init_pages("initrd memory", start, end);
+}
+#endif
+
+void __init reserve_bootmem_generic(unsigned long phys, unsigned len)
+{
+ /* Should check here against the e820 map to avoid double free */
+#ifdef CONFIG_NUMA
+ int nid = phys_to_nid(phys);
+ reserve_bootmem_node(NODE_DATA(nid), phys, len);
+#else
+ reserve_bootmem(phys, len);
+#endif
+ if (phys+len <= MAX_DMA_PFN*PAGE_SIZE)
+ dma_reserve += len / PAGE_SIZE;
+}
+
+int kern_addr_valid(unsigned long addr)
+{
+ unsigned long above = ((long)addr) >> __VIRTUAL_MASK_SHIFT;
+ pgd_t *pgd;
+ pud_t *pud;
+ pmd_t *pmd;
+ pte_t *pte;
+
+ if (above != 0 && above != -1UL)
+ return 0;
+
+#ifdef CONFIG_XEN
+ /*
+ * Don't walk page tables for hypervisor addresses, but allow
+ * the M2P table to be accessed through e.g. /proc/kcore.
+ */
+ if (addr >= (unsigned long)machine_to_phys_mapping &&
+ addr < (unsigned long)(machine_to_phys_mapping +
+ (1UL << machine_to_phys_order)))
+ return 1;
+ if (addr >= HYPERVISOR_VIRT_START && addr < HYPERVISOR_VIRT_END)
+ return 0;
+#endif
+
+ pgd = pgd_offset_k(addr);
+ if (pgd_none(*pgd))
+ return 0;
+
+ pud = pud_offset(pgd, addr);
+ if (pud_none(*pud))
+ return 0;
+
+ pmd = pmd_offset(pud, addr);
+ if (pmd_none(*pmd))
+ return 0;
+ if (pmd_large(*pmd))
+ return pfn_valid(pmd_pfn(*pmd));
+
+ pte = pte_offset_kernel(pmd, addr);
+ if (pte_none(*pte))
+ return 0;
+ return pfn_valid(pte_pfn(*pte));
+}
+
+#ifdef CONFIG_SYSCTL
+#include <linux/sysctl.h>
+
+extern int exception_trace, page_fault_trace;
+
+static ctl_table debug_table2[] = {
+ { 99, "exception-trace", &exception_trace, sizeof(int), 0644, NULL,
+ proc_dointvec },
+ { 0, }
+};
+
+static ctl_table debug_root_table2[] = {
+ { .ctl_name = CTL_DEBUG, .procname = "debug", .mode = 0555,
+ .child = debug_table2 },
+ { 0 },
+};
+
+static __init int x8664_sysctl_init(void)
+{
+ register_sysctl_table(debug_root_table2, 1);
+ return 0;
+}
+__initcall(x8664_sysctl_init);
+#endif
+
+/* A pseudo VMAs to allow ptrace access for the vsyscall page. This only
+ covers the 64bit vsyscall page now. 32bit has a real VMA now and does
+ not need special handling anymore. */
+
+static struct vm_area_struct gate_vma = {
+ .vm_start = VSYSCALL_START,
+ .vm_end = VSYSCALL_END,
+ .vm_page_prot = PAGE_READONLY
+};
+
+struct vm_area_struct *get_gate_vma(struct task_struct *tsk)
+{
+#ifdef CONFIG_IA32_EMULATION
+ if (test_tsk_thread_flag(tsk, TIF_IA32))
+ return NULL;
+#endif
+ return &gate_vma;
+}
+
+int in_gate_area(struct task_struct *task, unsigned long addr)
+{
+ struct vm_area_struct *vma = get_gate_vma(task);
+ if (!vma)
+ return 0;
+ return (addr >= vma->vm_start) && (addr < vma->vm_end);
+}
+
+/* Use this when you have no reliable task/vma, typically from interrupt
+ * context. It is less reliable than using the task's vma and may give
+ * false positives.
+ */
+int in_gate_area_no_task(unsigned long addr)
+{
+ return (addr >= VSYSCALL_START) && (addr < VSYSCALL_END);
+}
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/mm/pageattr_64-xen.c 2009-03-18 10:39:31.000000000 +0100
@@ -0,0 +1,508 @@
+/*
+ * Copyright 2002 Andi Kleen, SuSE Labs.
+ * Thanks to Ben LaHaise for precious feedback.
+ */
+
+#include <linux/mm.h>
+#include <linux/sched.h>
+#include <linux/highmem.h>
+#include <linux/module.h>
+#include <linux/slab.h>
+#include <asm/uaccess.h>
+#include <asm/processor.h>
+#include <asm/tlbflush.h>
+#include <asm/io.h>
+
+#ifdef CONFIG_XEN
+#include <asm/pgalloc.h>
+#include <asm/mmu_context.h>
+
+LIST_HEAD(mm_unpinned);
+DEFINE_SPINLOCK(mm_unpinned_lock);
+
+static void _pin_lock(struct mm_struct *mm, int lock) {
+ if (lock)
+ spin_lock(&mm->page_table_lock);
+#if NR_CPUS >= CONFIG_SPLIT_PTLOCK_CPUS
+ /* While mm->page_table_lock protects us against insertions and
+ * removals of higher level page table pages, it doesn't protect
+ * against updates of pte-s. Such updates, however, require the
+ * pte pages to be in consistent state (unpinned+writable or
+ * pinned+readonly). The pinning and attribute changes, however
+ * cannot be done atomically, which is why such updates must be
+ * prevented from happening concurrently.
+ * Note that no pte lock can ever elsewhere be acquired nesting
+ * with an already acquired one in the same mm, or with the mm's
+ * page_table_lock already acquired, as that would break in the
+ * non-split case (where all these are actually resolving to the
+ * one page_table_lock). Thus acquiring all of them here is not
+ * going to result in dead locks, and the order of acquires
+ * doesn't matter.
+ */
+ {
+ pgd_t *pgd = mm->pgd;
+ unsigned g;
+
+ for (g = 0; g <= ((TASK_SIZE64-1) / PGDIR_SIZE); g++, pgd++) {
+ pud_t *pud;
+ unsigned u;
+
+ if (pgd_none(*pgd))
+ continue;
+ pud = pud_offset(pgd, 0);
+ for (u = 0; u < PTRS_PER_PUD; u++, pud++) {
+ pmd_t *pmd;
+ unsigned m;
+
+ if (pud_none(*pud))
+ continue;
+ pmd = pmd_offset(pud, 0);
+ for (m = 0; m < PTRS_PER_PMD; m++, pmd++) {
+ spinlock_t *ptl;
+
+ if (pmd_none(*pmd))
+ continue;
+ ptl = pte_lockptr(0, pmd);
+ if (lock)
+ spin_lock(ptl);
+ else
+ spin_unlock(ptl);
+ }
+ }
+ }
+ }
+#endif
+ if (!lock)
+ spin_unlock(&mm->page_table_lock);
+}
+#define pin_lock(mm) _pin_lock(mm, 1)
+#define pin_unlock(mm) _pin_lock(mm, 0)
+
+#define PIN_BATCH 8
+static DEFINE_PER_CPU(multicall_entry_t[PIN_BATCH], pb_mcl);
+
+static inline unsigned int mm_walk_set_prot(void *pt, pgprot_t flags,
+ unsigned int cpu, unsigned int seq)
+{
+ struct page *page = virt_to_page(pt);
+ unsigned long pfn = page_to_pfn(page);
+
+ MULTI_update_va_mapping(per_cpu(pb_mcl, cpu) + seq,
+ (unsigned long)__va(pfn << PAGE_SHIFT),
+ pfn_pte(pfn, flags), 0);
+ if (unlikely(++seq == PIN_BATCH)) {
+ if (unlikely(HYPERVISOR_multicall_check(per_cpu(pb_mcl, cpu),
+ PIN_BATCH, NULL)))
+ BUG();
+ seq = 0;
+ }
+
+ return seq;
+}
+
+static void mm_walk(struct mm_struct *mm, pgprot_t flags)
+{
+ pgd_t *pgd;
+ pud_t *pud;
+ pmd_t *pmd;
+ pte_t *pte;
+ int g,u,m;
+ unsigned int cpu, seq;
+ multicall_entry_t *mcl;
+
+ pgd = mm->pgd;
+ cpu = get_cpu();
+
+ /*
+ * Cannot iterate up to USER_PTRS_PER_PGD as these pagetables may not
+ * be the 'current' task's pagetables (e.g., current may be 32-bit,
+ * but the pagetables may be for a 64-bit task).
+ * Subtracting 1 from TASK_SIZE64 means the loop limit is correct
+ * regardless of whether TASK_SIZE64 is a multiple of PGDIR_SIZE.
+ */
+ for (g = 0, seq = 0; g <= ((TASK_SIZE64-1) / PGDIR_SIZE); g++, pgd++) {
+ if (pgd_none(*pgd))
+ continue;
+ pud = pud_offset(pgd, 0);
+ if (PTRS_PER_PUD > 1) /* not folded */
+ seq = mm_walk_set_prot(pud,flags,cpu,seq);
+ for (u = 0; u < PTRS_PER_PUD; u++, pud++) {
+ if (pud_none(*pud))
+ continue;
+ pmd = pmd_offset(pud, 0);
+ if (PTRS_PER_PMD > 1) /* not folded */
+ seq = mm_walk_set_prot(pmd,flags,cpu,seq);
+ for (m = 0; m < PTRS_PER_PMD; m++, pmd++) {
+ if (pmd_none(*pmd))
+ continue;
+ pte = pte_offset_kernel(pmd,0);
+ seq = mm_walk_set_prot(pte,flags,cpu,seq);
+ }
+ }
+ }
+
+ mcl = per_cpu(pb_mcl, cpu);
+ if (unlikely(seq > PIN_BATCH - 2)) {
+ if (unlikely(HYPERVISOR_multicall_check(mcl, seq, NULL)))
+ BUG();
+ seq = 0;
+ }
+ MULTI_update_va_mapping(mcl + seq,
+ (unsigned long)__user_pgd(mm->pgd),
+ pfn_pte(virt_to_phys(__user_pgd(mm->pgd))>>PAGE_SHIFT, flags),
+ 0);
+ MULTI_update_va_mapping(mcl + seq + 1,
+ (unsigned long)mm->pgd,
+ pfn_pte(virt_to_phys(mm->pgd)>>PAGE_SHIFT, flags),
+ UVMF_TLB_FLUSH);
+ if (unlikely(HYPERVISOR_multicall_check(mcl, seq + 2, NULL)))
+ BUG();
+
+ put_cpu();
+}
+
+void mm_pin(struct mm_struct *mm)
+{
+ if (xen_feature(XENFEAT_writable_page_tables))
+ return;
+
+ pin_lock(mm);
+
+ mm_walk(mm, PAGE_KERNEL_RO);
+ xen_pgd_pin(__pa(mm->pgd)); /* kernel */
+ xen_pgd_pin(__pa(__user_pgd(mm->pgd))); /* user */
+ mm->context.pinned = 1;
+ spin_lock(&mm_unpinned_lock);
+ list_del(&mm->context.unpinned);
+ spin_unlock(&mm_unpinned_lock);
+
+ pin_unlock(mm);
+}
+
+void mm_unpin(struct mm_struct *mm)
+{
+ if (xen_feature(XENFEAT_writable_page_tables))
+ return;
+
+ pin_lock(mm);
+
+ xen_pgd_unpin(__pa(mm->pgd));
+ xen_pgd_unpin(__pa(__user_pgd(mm->pgd)));
+ mm_walk(mm, PAGE_KERNEL);
+ mm->context.pinned = 0;
+ spin_lock(&mm_unpinned_lock);
+ list_add(&mm->context.unpinned, &mm_unpinned);
+ spin_unlock(&mm_unpinned_lock);
+
+ pin_unlock(mm);
+}
+
+void mm_pin_all(void)
+{
+ if (xen_feature(XENFEAT_writable_page_tables))
+ return;
+
+ /*
+ * Allow uninterrupted access to the mm_unpinned list. We don't
+ * actually take the mm_unpinned_lock as it is taken inside mm_pin().
+ * All other CPUs must be at a safe point (e.g., in stop_machine
+ * or offlined entirely).
+ */
+ preempt_disable();
+ while (!list_empty(&mm_unpinned))
+ mm_pin(list_entry(mm_unpinned.next, struct mm_struct,
+ context.unpinned));
+ preempt_enable();
+}
+
+void _arch_dup_mmap(struct mm_struct *mm)
+{
+ if (!mm->context.pinned)
+ mm_pin(mm);
+}
+
+void _arch_exit_mmap(struct mm_struct *mm)
+{
+ struct task_struct *tsk = current;
+
+ task_lock(tsk);
+
+ /*
+ * We aggressively remove defunct pgd from cr3. We execute unmap_vmas()
+ * *much* faster this way, as no tlb flushes means bigger wrpt batches.
+ */
+ if (tsk->active_mm == mm) {
+ tsk->active_mm = &init_mm;
+ atomic_inc(&init_mm.mm_count);
+
+ switch_mm(mm, &init_mm, tsk);
+
+ atomic_dec(&mm->mm_count);
+ BUG_ON(atomic_read(&mm->mm_count) == 0);
+ }
+
+ task_unlock(tsk);
+
+ if ( mm->context.pinned && (atomic_read(&mm->mm_count) == 1) &&
+ !mm->context.has_foreign_mappings )
+ mm_unpin(mm);
+}
+
+static void _pte_free(struct page *page, unsigned int order)
+{
+ BUG_ON(order);
+ pte_free(page);
+}
+
+struct page *pte_alloc_one(struct mm_struct *mm, unsigned long address)
+{
+ struct page *pte;
+
+ pte = alloc_pages(GFP_KERNEL|__GFP_REPEAT|__GFP_ZERO, 0);
+ if (pte) {
+ SetPageForeign(pte, _pte_free);
+ init_page_count(pte);
+ }
+ return pte;
+}
+
+void pte_free(struct page *pte)
+{
+ unsigned long va = (unsigned long)__va(page_to_pfn(pte)<<PAGE_SHIFT);
+
+ if (!pte_write(*virt_to_ptep(va)))
+ if (HYPERVISOR_update_va_mapping(
+ va, pfn_pte(page_to_pfn(pte), PAGE_KERNEL), 0))
+ BUG();
+
+ ClearPageForeign(pte);
+ init_page_count(pte);
+
+ __free_page(pte);
+}
+#endif /* CONFIG_XEN */
+
+pte_t *lookup_address(unsigned long address)
+{
+ pgd_t *pgd = pgd_offset_k(address);
+ pud_t *pud;
+ pmd_t *pmd;
+ pte_t *pte;
+ if (pgd_none(*pgd))
+ return NULL;
+ pud = pud_offset(pgd, address);
+ if (!pud_present(*pud))
+ return NULL;
+ pmd = pmd_offset(pud, address);
+ if (!pmd_present(*pmd))
+ return NULL;
+ if (pmd_large(*pmd))
+ return (pte_t *)pmd;
+ pte = pte_offset_kernel(pmd, address);
+ if (pte && !pte_present(*pte))
+ pte = NULL;
+ return pte;
+}
+
+static struct page *split_large_page(unsigned long address, pgprot_t prot,
+ pgprot_t ref_prot)
+{
+ int i;
+ unsigned long addr;
+ struct page *base = alloc_pages(GFP_KERNEL, 0);
+ pte_t *pbase;
+ if (!base)
+ return NULL;
+ /*
+ * page_private is used to track the number of entries in
+ * the page table page have non standard attributes.
+ */
+ SetPagePrivate(base);
+ page_private(base) = 0;
+
+ address = __pa(address);
+ addr = address & LARGE_PAGE_MASK;
+ pbase = (pte_t *)page_address(base);
+ for (i = 0; i < PTRS_PER_PTE; i++, addr += PAGE_SIZE) {
+ pbase[i] = pfn_pte(addr >> PAGE_SHIFT,
+ addr == address ? prot : ref_prot);
+ }
+ return base;
+}
+
+
+static void flush_kernel_map(void *address)
+{
+ if (0 && address && cpu_has_clflush) {
+ /* is this worth it? */
+ int i;
+ for (i = 0; i < PAGE_SIZE; i += boot_cpu_data.x86_clflush_size)
+ asm volatile("clflush (%0)" :: "r" (address + i));
+ } else
+ asm volatile("wbinvd":::"memory");
+ if (address)
+ __flush_tlb_one(address);
+ else
+ __flush_tlb_all();
+}
+
+
+static inline void flush_map(unsigned long address)
+{
+ on_each_cpu(flush_kernel_map, (void *)address, 1, 1);
+}
+
+static struct page *deferred_pages; /* protected by init_mm.mmap_sem */
+
+static inline void save_page(struct page *fpage)
+{
+ fpage->lru.next = (struct list_head *)deferred_pages;
+ deferred_pages = fpage;
+}
+
+/*
+ * No more special protections in this 2/4MB area - revert to a
+ * large page again.
+ */
+static void revert_page(unsigned long address, pgprot_t ref_prot)
+{
+ pgd_t *pgd;
+ pud_t *pud;
+ pmd_t *pmd;
+ pte_t large_pte;
+
+ pgd = pgd_offset_k(address);
+ BUG_ON(pgd_none(*pgd));
+ pud = pud_offset(pgd,address);
+ BUG_ON(pud_none(*pud));
+ pmd = pmd_offset(pud, address);
+ BUG_ON(__pmd_val(*pmd) & _PAGE_PSE);
+ pgprot_val(ref_prot) |= _PAGE_PSE;
+ large_pte = mk_pte_phys(__pa(address) & LARGE_PAGE_MASK, ref_prot);
+ set_pte((pte_t *)pmd, large_pte);
+}
+
+static int
+__change_page_attr(unsigned long address, unsigned long pfn, pgprot_t prot,
+ pgprot_t ref_prot)
+{
+ pte_t *kpte;
+ struct page *kpte_page;
+ unsigned kpte_flags;
+ pgprot_t ref_prot2;
+ kpte = lookup_address(address);
+ if (!kpte) return 0;
+ kpte_page = virt_to_page(((unsigned long)kpte) & PAGE_MASK);
+ kpte_flags = pte_val(*kpte);
+ if (pgprot_val(prot) != pgprot_val(ref_prot)) {
+ if ((kpte_flags & _PAGE_PSE) == 0) {
+ set_pte(kpte, pfn_pte(pfn, prot));
+ } else {
+ /*
+ * split_large_page will take the reference for this
+ * change_page_attr on the split page.
+ */
+
+ struct page *split;
+ ref_prot2 = __pgprot(pgprot_val(pte_pgprot(*lookup_address(address))) & ~(1<<_PAGE_BIT_PSE));
+
+ split = split_large_page(address, prot, ref_prot2);
+ if (!split)
+ return -ENOMEM;
+ set_pte(kpte,mk_pte(split, ref_prot2));
+ kpte_page = split;
+ }
+ page_private(kpte_page)++;
+ } else if ((kpte_flags & _PAGE_PSE) == 0) {
+ set_pte(kpte, pfn_pte(pfn, ref_prot));
+ BUG_ON(page_private(kpte_page) == 0);
+ page_private(kpte_page)--;
+ } else
+ BUG();
+
+ /* on x86-64 the direct mapping set at boot is not using 4k pages */
+ /*
+ * ..., but the XEN guest kernels (currently) do:
+ * If the pte was reserved, it means it was created at boot
+ * time (not via split_large_page) and in turn we must not
+ * replace it with a large page.
+ */
+#ifndef CONFIG_XEN
+ BUG_ON(PageReserved(kpte_page));
+#else
+ if (PageReserved(kpte_page))
+ return 0;
+#endif
+
+ if (page_private(kpte_page) == 0) {
+ save_page(kpte_page);
+ revert_page(address, ref_prot);
+ }
+ return 0;
+}
+
+/*
+ * Change the page attributes of an page in the linear mapping.
+ *
+ * This should be used when a page is mapped with a different caching policy
+ * than write-back somewhere - some CPUs do not like it when mappings with
+ * different caching policies exist. This changes the page attributes of the
+ * in kernel linear mapping too.
+ *
+ * The caller needs to ensure that there are no conflicting mappings elsewhere.
+ * This function only deals with the kernel linear map.
+ *
+ * Caller must call global_flush_tlb() after this.
+ */
+int change_page_attr_addr(unsigned long address, int numpages, pgprot_t prot)
+{
+ int err = 0;
+ int i;
+
+ down_write(&init_mm.mmap_sem);
+ for (i = 0; i < numpages; i++, address += PAGE_SIZE) {
+ unsigned long pfn = __pa(address) >> PAGE_SHIFT;
+
+ err = __change_page_attr(address, pfn, prot, PAGE_KERNEL);
+ if (err)
+ break;
+ /* Handle kernel mapping too which aliases part of the
+ * lowmem */
+ if (__pa(address) < KERNEL_TEXT_SIZE) {
+ unsigned long addr2;
+ pgprot_t prot2 = prot;
+ addr2 = __START_KERNEL_map + __pa(address);
+ pgprot_val(prot2) &= ~_PAGE_NX;
+ err = __change_page_attr(addr2, pfn, prot2, PAGE_KERNEL_EXEC);
+ }
+ }
+ up_write(&init_mm.mmap_sem);
+ return err;
+}
+
+/* Don't call this for MMIO areas that may not have a mem_map entry */
+int change_page_attr(struct page *page, int numpages, pgprot_t prot)
+{
+ unsigned long addr = (unsigned long)page_address(page);
+ return change_page_attr_addr(addr, numpages, prot);
+}
+
+void global_flush_tlb(void)
+{
+ struct page *dpage;
+
+ down_read(&init_mm.mmap_sem);
+ dpage = xchg(&deferred_pages, NULL);
+ up_read(&init_mm.mmap_sem);
+
+ flush_map((dpage && !dpage->lru.next) ? (unsigned long)page_address(dpage) : 0);
+ while (dpage) {
+ struct page *tmp = dpage;
+ dpage = (struct page *)dpage->lru.next;
+ ClearPagePrivate(tmp);
+ __free_page(tmp);
+ }
+}
+
+EXPORT_SYMBOL(change_page_attr);
+EXPORT_SYMBOL(global_flush_tlb);
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/drivers/pci/msi-xen.c 2009-12-04 08:45:56.000000000 +0100
@@ -0,0 +1,910 @@
+/*
+ * File: msi.c
+ * Purpose: PCI Message Signaled Interrupt (MSI)
+ *
+ * Copyright (C) 2003-2004 Intel
+ * Copyright (C) Tom Long Nguyen (tom.l.nguyen@intel.com)
+ */
+
+#include <linux/mm.h>
+#include <linux/irq.h>
+#include <linux/interrupt.h>
+#include <linux/init.h>
+#include <linux/ioport.h>
+#include <linux/smp_lock.h>
+#include <linux/pci.h>
+#include <linux/proc_fs.h>
+
+#include <xen/evtchn.h>
+
+#include <asm/errno.h>
+#include <asm/io.h>
+#include <asm/smp.h>
+
+#include "pci.h"
+#include "msi.h"
+
+static int pci_msi_enable = 1;
+
+static struct msi_ops *msi_ops;
+
+int msi_register(struct msi_ops *ops)
+{
+ msi_ops = ops;
+ return 0;
+}
+
+static LIST_HEAD(msi_dev_head);
+DEFINE_SPINLOCK(msi_dev_lock);
+
+struct msi_dev_list {
+ struct pci_dev *dev;
+ struct list_head list;
+ spinlock_t pirq_list_lock;
+ struct list_head pirq_list_head;
+ /* Used for saving/restoring MSI-X tables */
+ void __iomem *mask_base;
+ /* Store default pre-assigned irq */
+ unsigned int default_irq;
+};
+
+struct msi_pirq_entry {
+ struct list_head list;
+ int pirq;
+ int entry_nr;
+#ifdef CONFIG_PM
+ /* PM save area for MSIX address/data */
+ u32 address_hi_save;
+ u32 address_lo_save;
+ u32 data_save;
+#endif
+};
+
+static struct msi_dev_list *get_msi_dev_pirq_list(struct pci_dev *dev)
+{
+ struct msi_dev_list *msi_dev_list, *ret = NULL;
+ unsigned long flags;
+
+ spin_lock_irqsave(&msi_dev_lock, flags);
+
+ list_for_each_entry(msi_dev_list, &msi_dev_head, list)
+ if ( msi_dev_list->dev == dev )
+ ret = msi_dev_list;
+
+ if ( ret ) {
+ spin_unlock_irqrestore(&msi_dev_lock, flags);
+ return ret;
+ }
+
+ /* Has not allocate msi_dev until now. */
+ ret = kzalloc(sizeof(struct msi_dev_list), GFP_ATOMIC);
+
+ /* Failed to allocate msi_dev structure */
+ if ( !ret ) {
+ spin_unlock_irqrestore(&msi_dev_lock, flags);
+ return NULL;
+ }
+
+ ret->dev = dev;
+ spin_lock_init(&ret->pirq_list_lock);
+ INIT_LIST_HEAD(&ret->pirq_list_head);
+ list_add_tail(&ret->list, &msi_dev_head);
+ spin_unlock_irqrestore(&msi_dev_lock, flags);
+ return ret;
+}
+
+static int attach_pirq_entry(int pirq, int entry_nr,
+ struct msi_dev_list *msi_dev_entry)
+{
+ struct msi_pirq_entry *entry = kmalloc(sizeof(*entry), GFP_ATOMIC);
+ unsigned long flags;
+
+ if (!entry)
+ return -ENOMEM;
+ entry->pirq = pirq;
+ entry->entry_nr = entry_nr;
+ spin_lock_irqsave(&msi_dev_entry->pirq_list_lock, flags);
+ list_add_tail(&entry->list, &msi_dev_entry->pirq_list_head);
+ spin_unlock_irqrestore(&msi_dev_entry->pirq_list_lock, flags);
+ return 0;
+}
+
+static void detach_pirq_entry(int entry_nr,
+ struct msi_dev_list *msi_dev_entry)
+{
+ unsigned long flags;
+ struct msi_pirq_entry *pirq_entry;
+
+ list_for_each_entry(pirq_entry, &msi_dev_entry->pirq_list_head, list) {
+ if (pirq_entry->entry_nr == entry_nr) {
+ spin_lock_irqsave(&msi_dev_entry->pirq_list_lock, flags);
+ list_del(&pirq_entry->list);
+ spin_unlock_irqrestore(&msi_dev_entry->pirq_list_lock, flags);
+ kfree(pirq_entry);
+ return;
+ }
+ }
+}
+
+/*
+ * pciback will provide device's owner
+ */
+static int (*get_owner)(struct pci_dev *dev);
+
+int register_msi_get_owner(int (*func)(struct pci_dev *dev))
+{
+ if (get_owner) {
+ printk(KERN_WARNING "register msi_get_owner again\n");
+ return -EEXIST;
+ }
+ get_owner = func;
+ return 0;
+}
+
+int unregister_msi_get_owner(int (*func)(struct pci_dev *dev))
+{
+ if (get_owner != func)
+ return -EINVAL;
+ get_owner = NULL;
+ return 0;
+}
+
+static int msi_get_dev_owner(struct pci_dev *dev)
+{
+ int owner;
+
+ BUG_ON(!is_initial_xendomain());
+ if (get_owner && (owner = get_owner(dev)) >= 0) {
+ printk(KERN_INFO "get owner for dev %x get %x \n",
+ dev->devfn, owner);
+ return owner;
+ }
+
+ return DOMID_SELF;
+}
+
+static int msi_unmap_pirq(struct pci_dev *dev, int pirq)
+{
+ struct physdev_unmap_pirq unmap;
+ int rc;
+
+ unmap.domid = msi_get_dev_owner(dev);
+ /* See comments in msi_map_pirq_to_vector, input parameter pirq
+ * mean irq number only if the device belongs to dom0 itself.
+ */
+ unmap.pirq = (unmap.domid != DOMID_SELF)
+ ? pirq : evtchn_get_xen_pirq(pirq);
+
+ if ((rc = HYPERVISOR_physdev_op(PHYSDEVOP_unmap_pirq, &unmap)))
+ printk(KERN_WARNING "unmap irq %x failed\n", pirq);
+
+ if (rc < 0)
+ return rc;
+
+ if (unmap.domid == DOMID_SELF)
+ evtchn_map_pirq(pirq, 0);
+
+ return 0;
+}
+
+static u64 find_table_base(struct pci_dev *dev, int pos)
+{
+ u8 bar;
+ u32 reg;
+ unsigned long flags;
+
+ pci_read_config_dword(dev, msix_table_offset_reg(pos), &reg);
+ bar = reg & PCI_MSIX_FLAGS_BIRMASK;
+
+ flags = pci_resource_flags(dev, bar);
+ if (flags & (IORESOURCE_DISABLED | IORESOURCE_UNSET | IORESOURCE_BUSY))
+ return 0;
+
+ return pci_resource_start(dev, bar);
+}
+
+/*
+ * Protected by msi_lock
+ */
+static int msi_map_pirq_to_vector(struct pci_dev *dev, int pirq,
+ int entry_nr, u64 table_base)
+{
+ struct physdev_map_pirq map_irq;
+ int rc;
+ domid_t domid = DOMID_SELF;
+
+ domid = msi_get_dev_owner(dev);
+
+ map_irq.domid = domid;
+ map_irq.type = MAP_PIRQ_TYPE_MSI;
+ map_irq.index = -1;
+ map_irq.pirq = pirq < 0 ? -1 : evtchn_get_xen_pirq(pirq);
+ map_irq.bus = dev->bus->number;
+ map_irq.devfn = dev->devfn;
+ map_irq.entry_nr = entry_nr;
+ map_irq.table_base = table_base;
+
+ if ((rc = HYPERVISOR_physdev_op(PHYSDEVOP_map_pirq, &map_irq)))
+ printk(KERN_WARNING "map irq failed\n");
+
+ if (rc < 0)
+ return rc;
+ /* This happens when MSI support is not enabled in Xen. */
+ if (rc == 0 && map_irq.pirq < 0)
+ return -ENOSYS;
+
+ BUG_ON(map_irq.pirq <= 0);
+
+ /* If mapping of this particular MSI is on behalf of another domain,
+ * we do not need to get an irq in dom0. This also implies:
+ * dev->irq in dom0 will be 'Xen pirq' if this device belongs to
+ * to another domain, and will be 'Linux irq' if it belongs to dom0.
+ */
+ return ((domid != DOMID_SELF) ?
+ map_irq.pirq : evtchn_map_pirq(pirq, map_irq.pirq));
+}
+
+static int msi_map_vector(struct pci_dev *dev, int entry_nr, u64 table_base)
+{
+ return msi_map_pirq_to_vector(dev, -1, entry_nr, table_base);
+}
+
+static int msi_init(void)
+{
+ static int status = 0;
+
+ if (pci_msi_quirk) {
+ pci_msi_enable = 0;
+ printk(KERN_WARNING "PCI: MSI quirk detected. MSI disabled.\n");
+ status = -EINVAL;
+ }
+
+ return status;
+}
+
+void pci_scan_msi_device(struct pci_dev *dev) { }
+
+void disable_msi_mode(struct pci_dev *dev, int pos, int type)
+{
+ u16 control;
+
+ pci_read_config_word(dev, msi_control_reg(pos), &control);
+ if (type == PCI_CAP_ID_MSI) {
+ /* Set enabled bits to single MSI & enable MSI_enable bit */
+ msi_disable(control);
+ pci_write_config_word(dev, msi_control_reg(pos), control);
+ dev->msi_enabled = 0;
+ } else {
+ msix_disable(control);
+ pci_write_config_word(dev, msi_control_reg(pos), control);
+ dev->msix_enabled = 0;
+ }
+ if (pci_find_capability(dev, PCI_CAP_ID_EXP)) {
+ /* PCI Express Endpoint device detected */
+ pci_intx(dev, 1); /* enable intx */
+ }
+}
+
+static void enable_msi_mode(struct pci_dev *dev, int pos, int type)
+{
+ u16 control;
+
+ pci_read_config_word(dev, msi_control_reg(pos), &control);
+ if (type == PCI_CAP_ID_MSI) {
+ /* Set enabled bits to single MSI & enable MSI_enable bit */
+ msi_enable(control, 1);
+ pci_write_config_word(dev, msi_control_reg(pos), control);
+ dev->msi_enabled = 1;
+ } else {
+ msix_enable(control);
+ pci_write_config_word(dev, msi_control_reg(pos), control);
+ dev->msix_enabled = 1;
+ }
+ if (pci_find_capability(dev, PCI_CAP_ID_EXP)) {
+ /* PCI Express Endpoint device detected */
+ pci_intx(dev, 0); /* disable intx */
+ }
+}
+
+#ifdef CONFIG_PM
+int pci_save_msi_state(struct pci_dev *dev)
+{
+ int pos, i = 0;
+ u16 control;
+ struct pci_cap_saved_state *save_state;
+ u32 *cap;
+
+ pos = pci_find_capability(dev, PCI_CAP_ID_MSI);
+ if (pos <= 0 || dev->no_msi)
+ return 0;
+
+ pci_read_config_word(dev, msi_control_reg(pos), &control);
+ if (!(control & PCI_MSI_FLAGS_ENABLE))
+ return 0;
+
+ save_state = kzalloc(sizeof(struct pci_cap_saved_state) + sizeof(u32) * 5,
+ GFP_KERNEL);
+ if (!save_state) {
+ printk(KERN_ERR "Out of memory in pci_save_msi_state\n");
+ return -ENOMEM;
+ }
+ cap = &save_state->data[0];
+
+ pci_read_config_dword(dev, pos, &cap[i++]);
+ control = cap[0] >> 16;
+ pci_read_config_dword(dev, pos + PCI_MSI_ADDRESS_LO, &cap[i++]);
+ if (control & PCI_MSI_FLAGS_64BIT) {
+ pci_read_config_dword(dev, pos + PCI_MSI_ADDRESS_HI, &cap[i++]);
+ pci_read_config_dword(dev, pos + PCI_MSI_DATA_64, &cap[i++]);
+ } else
+ pci_read_config_dword(dev, pos + PCI_MSI_DATA_32, &cap[i++]);
+ if (control & PCI_MSI_FLAGS_MASKBIT)
+ pci_read_config_dword(dev, pos + PCI_MSI_MASK_BIT, &cap[i++]);
+ save_state->cap_nr = PCI_CAP_ID_MSI;
+ pci_add_saved_cap(dev, save_state);
+ return 0;
+}
+
+void pci_restore_msi_state(struct pci_dev *dev)
+{
+ int i = 0, pos;
+ u16 control;
+ struct pci_cap_saved_state *save_state;
+ u32 *cap;
+
+ save_state = pci_find_saved_cap(dev, PCI_CAP_ID_MSI);
+ pos = pci_find_capability(dev, PCI_CAP_ID_MSI);
+ if (!save_state || pos <= 0)
+ return;
+ cap = &save_state->data[0];
+
+ control = cap[i++] >> 16;
+ pci_write_config_dword(dev, pos + PCI_MSI_ADDRESS_LO, cap[i++]);
+ if (control & PCI_MSI_FLAGS_64BIT) {
+ pci_write_config_dword(dev, pos + PCI_MSI_ADDRESS_HI, cap[i++]);
+ pci_write_config_dword(dev, pos + PCI_MSI_DATA_64, cap[i++]);
+ } else
+ pci_write_config_dword(dev, pos + PCI_MSI_DATA_32, cap[i++]);
+ if (control & PCI_MSI_FLAGS_MASKBIT)
+ pci_write_config_dword(dev, pos + PCI_MSI_MASK_BIT, cap[i++]);
+ pci_write_config_word(dev, pos + PCI_MSI_FLAGS, control);
+ enable_msi_mode(dev, pos, PCI_CAP_ID_MSI);
+ pci_remove_saved_cap(save_state);
+ kfree(save_state);
+}
+
+int pci_save_msix_state(struct pci_dev *dev)
+{
+ int pos;
+ u16 control;
+ struct pci_cap_saved_state *save_state;
+ unsigned long flags;
+ struct msi_dev_list *msi_dev_entry;
+ struct msi_pirq_entry *pirq_entry;
+ void __iomem *base;
+
+ pos = pci_find_capability(dev, PCI_CAP_ID_MSIX);
+ if (pos <= 0 || dev->no_msi)
+ return 0;
+
+ /* save the capability */
+ pci_read_config_word(dev, msi_control_reg(pos), &control);
+ if (!(control & PCI_MSIX_FLAGS_ENABLE))
+ return 0;
+
+ msi_dev_entry = get_msi_dev_pirq_list(dev);
+ /* If we failed to map the MSI-X table at pci_enable_msix,
+ * We could not support saving them here.
+ */
+ if (!(base = msi_dev_entry->mask_base))
+ return -ENOMEM;
+
+ save_state = kzalloc(sizeof(struct pci_cap_saved_state) + sizeof(u16),
+ GFP_KERNEL);
+ if (!save_state) {
+ printk(KERN_ERR "Out of memory in pci_save_msix_state\n");
+ return -ENOMEM;
+ }
+ *((u16 *)&save_state->data[0]) = control;
+
+ spin_lock_irqsave(&msi_dev_entry->pirq_list_lock, flags);
+ list_for_each_entry(pirq_entry, &msi_dev_entry->pirq_list_head, list) {
+ int j;
+
+ /* save the table */
+ j = pirq_entry->entry_nr;
+ pirq_entry->address_lo_save =
+ readl(base + j * PCI_MSIX_ENTRY_SIZE +
+ PCI_MSIX_ENTRY_LOWER_ADDR_OFFSET);
+ pirq_entry->address_hi_save =
+ readl(base + j * PCI_MSIX_ENTRY_SIZE +
+ PCI_MSIX_ENTRY_UPPER_ADDR_OFFSET);
+ pirq_entry->data_save =
+ readl(base + j * PCI_MSIX_ENTRY_SIZE +
+ PCI_MSIX_ENTRY_DATA_OFFSET);
+ }
+ spin_unlock_irqrestore(&msi_dev_entry->pirq_list_lock, flags);
+
+ save_state->cap_nr = PCI_CAP_ID_MSIX;
+ pci_add_saved_cap(dev, save_state);
+ return 0;
+}
+
+void pci_restore_msix_state(struct pci_dev *dev)
+{
+ u16 save;
+ int pos, j;
+ void __iomem *base;
+ struct pci_cap_saved_state *save_state;
+ unsigned long flags;
+ struct msi_dev_list *msi_dev_entry;
+ struct msi_pirq_entry *pirq_entry;
+
+ save_state = pci_find_saved_cap(dev, PCI_CAP_ID_MSIX);
+ if (!save_state)
+ return;
+
+ save = *((u16 *)&save_state->data[0]);
+ pci_remove_saved_cap(save_state);
+ kfree(save_state);
+
+ pos = pci_find_capability(dev, PCI_CAP_ID_MSIX);
+ if (pos <= 0)
+ return;
+
+ msi_dev_entry = get_msi_dev_pirq_list(dev);
+ base = msi_dev_entry->mask_base;
+
+ spin_lock_irqsave(&msi_dev_entry->pirq_list_lock, flags);
+ list_for_each_entry(pirq_entry, &msi_dev_entry->pirq_list_head, list) {
+ /* route the table */
+ j = pirq_entry->entry_nr;
+ writel(pirq_entry->address_lo_save,
+ base + j * PCI_MSIX_ENTRY_SIZE +
+ PCI_MSIX_ENTRY_LOWER_ADDR_OFFSET);
+ writel(pirq_entry->address_hi_save,
+ base + j * PCI_MSIX_ENTRY_SIZE +
+ PCI_MSIX_ENTRY_UPPER_ADDR_OFFSET);
+ writel(pirq_entry->data_save,
+ base + j * PCI_MSIX_ENTRY_SIZE +
+ PCI_MSIX_ENTRY_DATA_OFFSET);
+ }
+ spin_unlock_irqrestore(&msi_dev_entry->pirq_list_lock, flags);
+
+ pci_write_config_word(dev, msi_control_reg(pos), save);
+ enable_msi_mode(dev, pos, PCI_CAP_ID_MSIX);
+}
+#endif
+
+/**
+ * msi_capability_init - configure device's MSI capability structure
+ * @dev: pointer to the pci_dev data structure of MSI device function
+ *
+ * Setup the MSI capability structure of device function with a single
+ * MSI vector, regardless of device function is capable of handling
+ * multiple messages. A return of zero indicates the successful setup
+ * of an entry zero with the new MSI vector or non-zero for otherwise.
+ **/
+static int msi_capability_init(struct pci_dev *dev)
+{
+ int pos, pirq;
+ u16 control;
+
+ pos = pci_find_capability(dev, PCI_CAP_ID_MSI);
+ pci_read_config_word(dev, msi_control_reg(pos), &control);
+
+ pirq = msi_map_vector(dev, 0, 0);
+ if (pirq < 0)
+ return -EBUSY;
+
+ dev->irq = pirq;
+ /* Set MSI enabled bits */
+ enable_msi_mode(dev, pos, PCI_CAP_ID_MSI);
+ dev->msi_enabled = 1;
+
+ return 0;
+}
+
+/**
+ * msix_capability_init - configure device's MSI-X capability
+ * @dev: pointer to the pci_dev data structure of MSI-X device function
+ * @entries: pointer to an array of struct msix_entry entries
+ * @nvec: number of @entries
+ *
+ * Setup the MSI-X capability structure of device function with a
+ * single MSI-X vector. A return of zero indicates the successful setup of
+ * requested MSI-X entries with allocated vectors or non-zero for otherwise.
+ **/
+static int msix_capability_init(struct pci_dev *dev,
+ struct msix_entry *entries, int nvec)
+{
+ u64 table_base;
+ u16 control;
+ int pirq, i, j, mapped, pos, nr_entries;
+ struct msi_dev_list *msi_dev_entry = get_msi_dev_pirq_list(dev);
+ struct msi_pirq_entry *pirq_entry;
+
+ if (!msi_dev_entry)
+ return -ENOMEM;
+
+ pos = pci_find_capability(dev, PCI_CAP_ID_MSIX);
+ table_base = find_table_base(dev, pos);
+ if (!table_base)
+ return -ENODEV;
+
+ pci_read_config_word(dev, msi_control_reg(pos), &control);
+ nr_entries = multi_msix_capable(control);
+ if (!msi_dev_entry->mask_base)
+ msi_dev_entry->mask_base =
+ ioremap_nocache(table_base, nr_entries * PCI_MSIX_ENTRY_SIZE);
+
+ /* MSI-X Table Initialization */
+ for (i = 0; i < nvec; i++) {
+ mapped = 0;
+ list_for_each_entry(pirq_entry, &msi_dev_entry->pirq_list_head, list) {
+ if (pirq_entry->entry_nr == entries[i].entry) {
+ printk(KERN_WARNING "msix entry %d for dev %02x:%02x:%01x are \
+ not freed before acquire again.\n", entries[i].entry,
+ dev->bus->number, PCI_SLOT(dev->devfn),
+ PCI_FUNC(dev->devfn));
+ (entries + i)->vector = pirq_entry->pirq;
+ mapped = 1;
+ break;
+ }
+ }
+ if (mapped)
+ continue;
+ pirq = msi_map_vector(dev, entries[i].entry, table_base);
+ if (pirq < 0)
+ break;
+ attach_pirq_entry(pirq, entries[i].entry, msi_dev_entry);
+ (entries + i)->vector = pirq;
+ }
+
+ if (i != nvec) {
+ for (j = --i; j >= 0; j--) {
+ msi_unmap_pirq(dev, entries[j].vector);
+ detach_pirq_entry(entries[j].entry, msi_dev_entry);
+ entries[j].vector = 0;
+ }
+ return -EBUSY;
+ }
+
+ enable_msi_mode(dev, pos, PCI_CAP_ID_MSIX);
+ dev->msix_enabled = 1;
+
+ return 0;
+}
+
+/**
+ * pci_enable_msi - configure device's MSI capability structure
+ * @dev: pointer to the pci_dev data structure of MSI device function
+ *
+ * Setup the MSI capability structure of device function with
+ * a single MSI vector upon its software driver call to request for
+ * MSI mode enabled on its hardware device function. A return of zero
+ * indicates the successful setup of an entry zero with the new MSI
+ * vector or non-zero for otherwise.
+ **/
+extern int pci_frontend_enable_msi(struct pci_dev *dev);
+int pci_enable_msi(struct pci_dev* dev)
+{
+ struct pci_bus *bus;
+ int pos, temp, status = -EINVAL;
+ struct msi_dev_list *msi_dev_entry = get_msi_dev_pirq_list(dev);
+
+ if (!pci_msi_enable || !dev)
+ return status;
+
+ if (dev->no_msi)
+ return status;
+
+ for (bus = dev->bus; bus; bus = bus->parent)
+ if (bus->bus_flags & PCI_BUS_FLAGS_NO_MSI)
+ return -EINVAL;
+
+ status = msi_init();
+ if (status < 0)
+ return status;
+
+#ifdef CONFIG_XEN_PCIDEV_FRONTEND
+ if (!is_initial_xendomain())
+ {
+ int ret;
+
+ temp = dev->irq;
+ ret = pci_frontend_enable_msi(dev);
+ if (ret)
+ return ret;
+
+ dev->irq = evtchn_map_pirq(-1, dev->irq);
+ dev->msi_enabled = 1;
+ msi_dev_entry->default_irq = temp;
+
+ return ret;
+ }
+#endif
+
+ temp = dev->irq;
+
+ pos = pci_find_capability(dev, PCI_CAP_ID_MSI);
+ if (!pos)
+ return -EINVAL;
+
+ /* Check whether driver already requested for MSI-X vectors */
+ if (dev->msix_enabled) {
+ printk(KERN_INFO "PCI: %s: Can't enable MSI. "
+ "Device already has MSI-X vectors assigned\n",
+ pci_name(dev));
+ dev->irq = temp;
+ return -EINVAL;
+ }
+
+ status = msi_capability_init(dev);
+ if ( !status )
+ msi_dev_entry->default_irq = temp;
+ else
+ dev->irq = temp;
+
+ return status;
+}
+
+extern void pci_frontend_disable_msi(struct pci_dev* dev);
+void pci_disable_msi(struct pci_dev* dev)
+{
+ int pos;
+ int pirq;
+ struct msi_dev_list *msi_dev_entry = get_msi_dev_pirq_list(dev);
+
+ if (!pci_msi_enable)
+ return;
+ if (!dev)
+ return;
+
+#ifdef CONFIG_XEN_PCIDEV_FRONTEND
+ if (!is_initial_xendomain()) {
+ if (!(dev->msi_enabled)) {
+ printk(KERN_INFO "PCI: %s: Device did not enabled MSI.\n",
+ pci_name(dev));
+ return;
+ }
+ evtchn_map_pirq(dev->irq, 0);
+ pci_frontend_disable_msi(dev);
+ dev->irq = msi_dev_entry->default_irq;
+ dev->msi_enabled = 0;
+ return;
+ }
+#endif
+
+ pos = pci_find_capability(dev, PCI_CAP_ID_MSI);
+ if (!pos)
+ return;
+
+ if (!(dev->msi_enabled)) {
+ printk(KERN_INFO "PCI: %s: Device did not enabled MSI.\n",
+ pci_name(dev));
+ return;
+ }
+
+ pirq = dev->irq;
+ /* Restore dev->irq to its default pin-assertion vector */
+ dev->irq = msi_dev_entry->default_irq;
+ msi_unmap_pirq(dev, pirq);
+
+ /* Disable MSI mode */
+ disable_msi_mode(dev, pos, PCI_CAP_ID_MSI);
+}
+
+/**
+ * pci_enable_msix - configure device's MSI-X capability structure
+ * @dev: pointer to the pci_dev data structure of MSI-X device function
+ * @entries: pointer to an array of MSI-X entries
+ * @nvec: number of MSI-X vectors requested for allocation by device driver
+ *
+ * Setup the MSI-X capability structure of device function with the number
+ * of requested vectors upon its software driver call to request for
+ * MSI-X mode enabled on its hardware device function. A return of zero
+ * indicates the successful configuration of MSI-X capability structure
+ * with new allocated MSI-X vectors. A return of < 0 indicates a failure.
+ * Or a return of > 0 indicates that driver request is exceeding the number
+ * of vectors available. Driver should use the returned value to re-send
+ * its request.
+ **/
+extern int pci_frontend_enable_msix(struct pci_dev *dev,
+ struct msix_entry *entries, int nvec);
+int pci_enable_msix(struct pci_dev* dev, struct msix_entry *entries, int nvec)
+{
+ struct pci_bus *bus;
+ int status, pos, nr_entries;
+ int i, j, temp;
+ u16 control;
+ struct msi_dev_list *msi_dev_entry = get_msi_dev_pirq_list(dev);
+
+ if (!pci_msi_enable || !dev || !entries)
+ return -EINVAL;
+
+ if (dev->no_msi)
+ return -EINVAL;
+
+ for (bus = dev->bus; bus; bus = bus->parent)
+ if (bus->bus_flags & PCI_BUS_FLAGS_NO_MSI)
+ return -EINVAL;
+
+#ifdef CONFIG_XEN_PCIDEV_FRONTEND
+ if (!is_initial_xendomain()) {
+ struct msi_pirq_entry *pirq_entry;
+ int ret, irq;
+
+ temp = dev->irq;
+ ret = pci_frontend_enable_msix(dev, entries, nvec);
+ if (ret) {
+ printk("get %x from pci_frontend_enable_msix\n", ret);
+ return ret;
+ }
+ dev->msix_enabled = 1;
+ msi_dev_entry->default_irq = temp;
+
+ for (i = 0; i < nvec; i++) {
+ int mapped = 0;
+
+ list_for_each_entry(pirq_entry, &msi_dev_entry->pirq_list_head, list) {
+ if (pirq_entry->entry_nr == entries[i].entry) {
+ irq = pirq_entry->pirq;
+ BUG_ON(entries[i].vector != evtchn_get_xen_pirq(irq));
+ entries[i].vector = irq;
+ mapped = 1;
+ break;
+ }
+ }
+ if (mapped)
+ continue;
+ irq = evtchn_map_pirq(-1, entries[i].vector);
+ attach_pirq_entry(irq, entries[i].entry, msi_dev_entry);
+ entries[i].vector = irq;
+ }
+ return 0;
+ }
+#endif
+
+ status = msi_init();
+ if (status < 0)
+ return status;
+
+ pos = pci_find_capability(dev, PCI_CAP_ID_MSIX);
+ if (!pos)
+ return -EINVAL;
+
+ pci_read_config_word(dev, msi_control_reg(pos), &control);
+ nr_entries = multi_msix_capable(control);
+ if (nvec > nr_entries)
+ return -EINVAL;
+
+ /* Check for any invalid entries */
+ for (i = 0; i < nvec; i++) {
+ if (entries[i].entry >= nr_entries)
+ return -EINVAL; /* invalid entry */
+ for (j = i + 1; j < nvec; j++) {
+ if (entries[i].entry == entries[j].entry)
+ return -EINVAL; /* duplicate entry */
+ }
+ }
+
+ temp = dev->irq;
+ /* Check whether driver already requested for MSI vector */
+ if (dev->msi_enabled) {
+ printk(KERN_INFO "PCI: %s: Can't enable MSI-X. "
+ "Device already has an MSI vector assigned\n",
+ pci_name(dev));
+ dev->irq = temp;
+ return -EINVAL;
+ }
+
+ status = msix_capability_init(dev, entries, nvec);
+
+ if ( !status )
+ msi_dev_entry->default_irq = temp;
+ else
+ dev->irq = temp;
+
+ return status;
+}
+
+extern void pci_frontend_disable_msix(struct pci_dev* dev);
+void pci_disable_msix(struct pci_dev* dev)
+{
+ int pos;
+ u16 control;
+
+ if (!pci_msi_enable)
+ return;
+ if (!dev)
+ return;
+ if (!dev->msix_enabled) {
+ printk(KERN_INFO "PCI: %s: Device did not enabled MSI-X.\n",
+ pci_name(dev));
+ return;
+ }
+
+#ifdef CONFIG_XEN_PCIDEV_FRONTEND
+ if (!is_initial_xendomain()) {
+ struct msi_dev_list *msi_dev_entry;
+ struct msi_pirq_entry *pirq_entry, *tmp;
+
+ pci_frontend_disable_msix(dev);
+
+ msi_dev_entry = get_msi_dev_pirq_list(dev);
+ list_for_each_entry_safe(pirq_entry, tmp,
+ &msi_dev_entry->pirq_list_head, list) {
+ evtchn_map_pirq(pirq_entry->pirq, 0);
+ list_del(&pirq_entry->list);
+ kfree(pirq_entry);
+ }
+
+ dev->irq = msi_dev_entry->default_irq;
+ dev->msix_enabled = 0;
+ return;
+ }
+#endif
+
+ pos = pci_find_capability(dev, PCI_CAP_ID_MSIX);
+ if (!pos)
+ return;
+
+ pci_read_config_word(dev, msi_control_reg(pos), &control);
+ if (!(control & PCI_MSIX_FLAGS_ENABLE))
+ return;
+
+ msi_remove_pci_irq_vectors(dev);
+
+ /* Disable MSI mode */
+ disable_msi_mode(dev, pos, PCI_CAP_ID_MSIX);
+}
+
+/**
+ * msi_remove_pci_irq_vectors - reclaim MSI(X) vectors to unused state
+ * @dev: pointer to the pci_dev data structure of MSI(X) device function
+ *
+ * Being called during hotplug remove, from which the device function
+ * is hot-removed. All previous assigned MSI/MSI-X vectors, if
+ * allocated for this device function, are reclaimed to unused state,
+ * which may be used later on.
+ **/
+void msi_remove_pci_irq_vectors(struct pci_dev* dev)
+{
+ unsigned long flags;
+ struct msi_dev_list *msi_dev_entry;
+ struct msi_pirq_entry *pirq_entry, *tmp;
+
+ if (!pci_msi_enable || !dev)
+ return;
+
+ msi_dev_entry = get_msi_dev_pirq_list(dev);
+
+ spin_lock_irqsave(&msi_dev_entry->pirq_list_lock, flags);
+ if (!list_empty(&msi_dev_entry->pirq_list_head))
+ list_for_each_entry_safe(pirq_entry, tmp,
+ &msi_dev_entry->pirq_list_head, list) {
+ msi_unmap_pirq(dev, pirq_entry->pirq);
+ list_del(&pirq_entry->list);
+ kfree(pirq_entry);
+ }
+ spin_unlock_irqrestore(&msi_dev_entry->pirq_list_lock, flags);
+ iounmap(msi_dev_entry->mask_base);
+ msi_dev_entry->mask_base = NULL;
+ dev->irq = msi_dev_entry->default_irq;
+}
+
+void pci_no_msi(void)
+{
+ pci_msi_enable = 0;
+}
+
+EXPORT_SYMBOL(pci_enable_msi);
+EXPORT_SYMBOL(pci_disable_msi);
+EXPORT_SYMBOL(pci_enable_msix);
+EXPORT_SYMBOL(pci_disable_msix);
+#ifdef CONFIG_XEN
+EXPORT_SYMBOL(register_msi_get_owner);
+EXPORT_SYMBOL(unregister_msi_get_owner);
+#endif
+
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/include/mach-xen/asm/agp.h 2007-06-22 09:08:06.000000000 +0200
@@ -0,0 +1,44 @@
+#ifndef AGP_H
+#define AGP_H 1
+
+#include <asm/pgtable.h>
+#include <asm/cacheflush.h>
+#include <asm/system.h>
+
+/*
+ * Functions to keep the agpgart mappings coherent with the MMU.
+ * The GART gives the CPU a physical alias of pages in memory. The alias region is
+ * mapped uncacheable. Make sure there are no conflicting mappings
+ * with different cachability attributes for the same page. This avoids
+ * data corruption on some CPUs.
+ */
+
+/* Caller's responsibility to call global_flush_tlb() for
+ * performance reasons */
+#define map_page_into_agp(page) ( \
+ xen_create_contiguous_region((unsigned long)page_address(page), 0, 32) \
+ ?: change_page_attr(page, 1, PAGE_KERNEL_NOCACHE))
+#define unmap_page_from_agp(page) ( \
+ xen_destroy_contiguous_region((unsigned long)page_address(page), 0), \
+ /* only a fallback: xen_destroy_contiguous_region uses PAGE_KERNEL */ \
+ change_page_attr(page, 1, PAGE_KERNEL))
+#define flush_agp_mappings() global_flush_tlb()
+
+/* Could use CLFLUSH here if the cpu supports it. But then it would
+ need to be called for each cacheline of the whole page so it may not be
+ worth it. Would need a page for it. */
+#define flush_agp_cache() wbinvd()
+
+/* Convert a physical address to an address suitable for the GART. */
+#define phys_to_gart(x) phys_to_machine(x)
+#define gart_to_phys(x) machine_to_phys(x)
+
+/* GATT allocation. Returns/accepts GATT kernel virtual address. */
+#define alloc_gatt_pages(order) ({ \
+ char *_t; dma_addr_t _d; \
+ _t = dma_alloc_coherent(NULL,PAGE_SIZE<<(order),&_d,GFP_KERNEL); \
+ _t; })
+#define free_gatt_pages(table, order) \
+ dma_free_coherent(NULL,PAGE_SIZE<<(order),(table),virt_to_bus(table))
+
+#endif
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/include/mach-xen/asm/desc_32.h 2008-01-28 12:24:19.000000000 +0100
@@ -0,0 +1,166 @@
+#ifndef __ARCH_DESC_H
+#define __ARCH_DESC_H
+
+#include <asm/ldt.h>
+#include <asm/segment.h>
+
+#define CPU_16BIT_STACK_SIZE 1024
+
+#ifndef __ASSEMBLY__
+
+#include <linux/preempt.h>
+#include <linux/smp.h>
+
+#include <asm/mmu.h>
+
+extern struct desc_struct cpu_gdt_table[GDT_ENTRIES];
+
+DECLARE_PER_CPU(unsigned char, cpu_16bit_stack[CPU_16BIT_STACK_SIZE]);
+
+struct Xgt_desc_struct {
+ unsigned short size;
+ unsigned long address __attribute__((packed));
+ unsigned short pad;
+} __attribute__ ((packed));
+
+extern struct Xgt_desc_struct idt_descr;
+DECLARE_PER_CPU(struct Xgt_desc_struct, cpu_gdt_descr);
+
+
+static inline struct desc_struct *get_cpu_gdt_table(unsigned int cpu)
+{
+ return (struct desc_struct *)per_cpu(cpu_gdt_descr, cpu).address;
+}
+
+#define load_TR_desc() __asm__ __volatile__("ltr %w0"::"q" (GDT_ENTRY_TSS*8))
+#define load_LDT_desc() __asm__ __volatile__("lldt %w0"::"q" (GDT_ENTRY_LDT*8))
+
+#define load_gdt(dtr) __asm__ __volatile("lgdt %0"::"m" (*dtr))
+#define load_idt(dtr) __asm__ __volatile("lidt %0"::"m" (*dtr))
+#define load_tr(tr) __asm__ __volatile("ltr %0"::"mr" (tr))
+#define load_ldt(ldt) __asm__ __volatile("lldt %0"::"mr" (ldt))
+
+#define store_gdt(dtr) __asm__ ("sgdt %0":"=m" (*dtr))
+#define store_idt(dtr) __asm__ ("sidt %0":"=m" (*dtr))
+#define store_tr(tr) __asm__ ("str %0":"=mr" (tr))
+#define store_ldt(ldt) __asm__ ("sldt %0":"=mr" (ldt))
+
+/*
+ * This is the ldt that every process will get unless we need
+ * something other than this.
+ */
+extern struct desc_struct default_ldt[];
+extern void set_intr_gate(unsigned int irq, void * addr);
+
+#define _set_tssldt_desc(n,addr,limit,type) \
+__asm__ __volatile__ ("movw %w3,0(%2)\n\t" \
+ "movw %w1,2(%2)\n\t" \
+ "rorl $16,%1\n\t" \
+ "movb %b1,4(%2)\n\t" \
+ "movb %4,5(%2)\n\t" \
+ "movb $0,6(%2)\n\t" \
+ "movb %h1,7(%2)\n\t" \
+ "rorl $16,%1" \
+ : "=m"(*(n)) : "q" (addr), "r"(n), "ir"(limit), "i"(type))
+
+#ifndef CONFIG_X86_NO_TSS
+static inline void __set_tss_desc(unsigned int cpu, unsigned int entry, void *addr)
+{
+ _set_tssldt_desc(&get_cpu_gdt_table(cpu)[entry], (int)addr,
+ offsetof(struct tss_struct, __cacheline_filler) - 1, 0x89);
+}
+
+#define set_tss_desc(cpu,addr) __set_tss_desc(cpu, GDT_ENTRY_TSS, addr)
+#endif
+
+static inline void set_ldt_desc(unsigned int cpu, void *addr, unsigned int size)
+{
+ _set_tssldt_desc(&get_cpu_gdt_table(cpu)[GDT_ENTRY_LDT], (int)addr, ((size << 3)-1), 0x82);
+}
+
+#define LDT_entry_a(info) \
+ ((((info)->base_addr & 0x0000ffff) << 16) | ((info)->limit & 0x0ffff))
+
+#define LDT_entry_b(info) \
+ (((info)->base_addr & 0xff000000) | \
+ (((info)->base_addr & 0x00ff0000) >> 16) | \
+ ((info)->limit & 0xf0000) | \
+ (((info)->read_exec_only ^ 1) << 9) | \
+ ((info)->contents << 10) | \
+ (((info)->seg_not_present ^ 1) << 15) | \
+ ((info)->seg_32bit << 22) | \
+ ((info)->limit_in_pages << 23) | \
+ ((info)->useable << 20) | \
+ 0x7000)
+
+#define LDT_empty(info) (\
+ (info)->base_addr == 0 && \
+ (info)->limit == 0 && \
+ (info)->contents == 0 && \
+ (info)->read_exec_only == 1 && \
+ (info)->seg_32bit == 0 && \
+ (info)->limit_in_pages == 0 && \
+ (info)->seg_not_present == 1 && \
+ (info)->useable == 0 )
+
+extern int write_ldt_entry(void *ldt, int entry, __u32 entry_a, __u32 entry_b);
+
+#if TLS_SIZE != 24
+# error update this code.
+#endif
+
+static inline void load_TLS(struct thread_struct *t, unsigned int cpu)
+{
+#define C(i) if (HYPERVISOR_update_descriptor(virt_to_machine(&get_cpu_gdt_table(cpu)[GDT_ENTRY_TLS_MIN + i]), \
+ *(u64 *)&t->tls_array[i])) \
+ BUG();
+ C(0); C(1); C(2);
+#undef C
+}
+
+static inline void clear_LDT(void)
+{
+ int cpu = get_cpu();
+
+ /*
+ * NB. We load the default_ldt for lcall7/27 handling on demand, as
+ * it slows down context switching. Noone uses it anyway.
+ */
+ cpu = cpu; /* XXX avoid compiler warning */
+ xen_set_ldt(NULL, 0);
+ put_cpu();
+}
+
+/*
+ * load one particular LDT into the current CPU
+ */
+static inline void load_LDT_nolock(mm_context_t *pc, int cpu)
+{
+ void *segments = pc->ldt;
+ int count = pc->size;
+
+ if (likely(!count))
+ segments = NULL;
+
+ xen_set_ldt(segments, count);
+}
+
+static inline void load_LDT(mm_context_t *pc)
+{
+ int cpu = get_cpu();
+ load_LDT_nolock(pc, cpu);
+ put_cpu();
+}
+
+static inline unsigned long get_desc_base(unsigned long *desc)
+{
+ unsigned long base;
+ base = ((desc[0] >> 16) & 0x0000ffff) |
+ ((desc[1] << 16) & 0x00ff0000) |
+ (desc[1] & 0xff000000);
+ return base;
+}
+
+#endif /* !__ASSEMBLY__ */
+
+#endif
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/include/mach-xen/asm/fixmap_32.h 2007-06-12 13:14:02.000000000 +0200
@@ -0,0 +1,155 @@
+/*
+ * fixmap.h: compile-time virtual memory allocation
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License. See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 1998 Ingo Molnar
+ *
+ * Support of BIGMEM added by Gerhard Wichert, Siemens AG, July 1999
+ */
+
+#ifndef _ASM_FIXMAP_H
+#define _ASM_FIXMAP_H
+
+
+/* used by vmalloc.c, vsyscall.lds.S.
+ *
+ * Leave one empty page between vmalloc'ed areas and
+ * the start of the fixmap.
+ */
+extern unsigned long __FIXADDR_TOP;
+
+#ifndef __ASSEMBLY__
+#include <linux/kernel.h>
+#include <asm/acpi.h>
+#include <asm/apicdef.h>
+#include <asm/page.h>
+#ifdef CONFIG_HIGHMEM
+#include <linux/threads.h>
+#include <asm/kmap_types.h>
+#endif
+
+/*
+ * Here we define all the compile-time 'special' virtual
+ * addresses. The point is to have a constant address at
+ * compile time, but to set the physical address only
+ * in the boot process. We allocate these special addresses
+ * from the end of virtual memory (0xfffff000) backwards.
+ * Also this lets us do fail-safe vmalloc(), we
+ * can guarantee that these special addresses and
+ * vmalloc()-ed addresses never overlap.
+ *
+ * these 'compile-time allocated' memory buffers are
+ * fixed-size 4k pages. (or larger if used with an increment
+ * highger than 1) use fixmap_set(idx,phys) to associate
+ * physical memory with fixmap indices.
+ *
+ * TLB entries of such buffers will not be flushed across
+ * task switches.
+ */
+enum fixed_addresses {
+ FIX_HOLE,
+ FIX_VDSO,
+#ifdef CONFIG_X86_LOCAL_APIC
+ FIX_APIC_BASE, /* local (CPU) APIC) -- required for SMP or not */
+#endif
+#ifdef CONFIG_X86_IO_APIC
+ FIX_IO_APIC_BASE_0,
+ FIX_IO_APIC_BASE_END = FIX_IO_APIC_BASE_0 + MAX_IO_APICS-1,
+#endif
+#ifdef CONFIG_X86_VISWS_APIC
+ FIX_CO_CPU, /* Cobalt timer */
+ FIX_CO_APIC, /* Cobalt APIC Redirection Table */
+ FIX_LI_PCIA, /* Lithium PCI Bridge A */
+ FIX_LI_PCIB, /* Lithium PCI Bridge B */
+#endif
+#ifdef CONFIG_X86_F00F_BUG
+ FIX_F00F_IDT, /* Virtual mapping for IDT */
+#endif
+#ifdef CONFIG_X86_CYCLONE_TIMER
+ FIX_CYCLONE_TIMER, /*cyclone timer register*/
+#endif
+#ifdef CONFIG_HIGHMEM
+ FIX_KMAP_BEGIN, /* reserved pte's for temporary kernel mappings */
+ FIX_KMAP_END = FIX_KMAP_BEGIN+(KM_TYPE_NR*NR_CPUS)-1,
+#endif
+#ifdef CONFIG_ACPI
+ FIX_ACPI_BEGIN,
+ FIX_ACPI_END = FIX_ACPI_BEGIN + FIX_ACPI_PAGES - 1,
+#endif
+#ifdef CONFIG_PCI_MMCONFIG
+ FIX_PCIE_MCFG,
+#endif
+ FIX_SHARED_INFO,
+#define NR_FIX_ISAMAPS 256
+ FIX_ISAMAP_END,
+ FIX_ISAMAP_BEGIN = FIX_ISAMAP_END + NR_FIX_ISAMAPS - 1,
+ __end_of_permanent_fixed_addresses,
+ /* temporary boot-time mappings, used before ioremap() is functional */
+#define NR_FIX_BTMAPS 16
+ FIX_BTMAP_END = __end_of_permanent_fixed_addresses,
+ FIX_BTMAP_BEGIN = FIX_BTMAP_END + NR_FIX_BTMAPS - 1,
+ FIX_WP_TEST,
+ __end_of_fixed_addresses
+};
+
+extern void set_fixaddr_top(unsigned long top);
+
+extern void __set_fixmap(enum fixed_addresses idx,
+ maddr_t phys, pgprot_t flags);
+
+#define set_fixmap(idx, phys) \
+ __set_fixmap(idx, phys, PAGE_KERNEL)
+/*
+ * Some hardware wants to get fixmapped without caching.
+ */
+#define set_fixmap_nocache(idx, phys) \
+ __set_fixmap(idx, phys, PAGE_KERNEL_NOCACHE)
+
+#define clear_fixmap(idx) \
+ __set_fixmap(idx, 0, __pgprot(0))
+
+#define FIXADDR_TOP ((unsigned long)__FIXADDR_TOP)
+
+#define __FIXADDR_SIZE (__end_of_permanent_fixed_addresses << PAGE_SHIFT)
+#define __FIXADDR_BOOT_SIZE (__end_of_fixed_addresses << PAGE_SHIFT)
+#define FIXADDR_START (FIXADDR_TOP - __FIXADDR_SIZE)
+#define FIXADDR_BOOT_START (FIXADDR_TOP - __FIXADDR_BOOT_SIZE)
+
+#define __fix_to_virt(x) (FIXADDR_TOP - ((x) << PAGE_SHIFT))
+#define __virt_to_fix(x) ((FIXADDR_TOP - ((x)&PAGE_MASK)) >> PAGE_SHIFT)
+
+extern void __this_fixmap_does_not_exist(void);
+
+/*
+ * 'index to address' translation. If anyone tries to use the idx
+ * directly without tranlation, we catch the bug with a NULL-deference
+ * kernel oops. Illegal ranges of incoming indices are caught too.
+ */
+static __always_inline unsigned long fix_to_virt(const unsigned int idx)
+{
+ /*
+ * this branch gets completely eliminated after inlining,
+ * except when someone tries to use fixaddr indices in an
+ * illegal way. (such as mixing up address types or using
+ * out-of-range indices).
+ *
+ * If it doesn't get removed, the linker will complain
+ * loudly with a reasonably clear error message..
+ */
+ if (idx >= __end_of_fixed_addresses)
+ __this_fixmap_does_not_exist();
+
+ return __fix_to_virt(idx);
+}
+
+static inline unsigned long virt_to_fix(const unsigned long vaddr)
+{
+ BUG_ON(vaddr >= FIXADDR_TOP || vaddr < FIXADDR_START);
+ return __virt_to_fix(vaddr);
+}
+
+#endif /* !__ASSEMBLY__ */
+#endif
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/include/mach-xen/asm/gnttab_dma.h 2007-08-06 15:10:49.000000000 +0200
@@ -0,0 +1,41 @@
+/*
+ * Copyright (c) 2007 Herbert Xu <herbert@gondor.apana.org.au>
+ * Copyright (c) 2007 Isaku Yamahata <yamahata at valinux co jp>
+ * VA Linux Systems Japan K.K.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ */
+
+#ifndef _ASM_I386_GNTTAB_DMA_H
+#define _ASM_I386_GNTTAB_DMA_H
+
+static inline int gnttab_dma_local_pfn(struct page *page)
+{
+ /* Has it become a local MFN? */
+ return pfn_valid(mfn_to_local_pfn(pfn_to_mfn(page_to_pfn(page))));
+}
+
+static inline maddr_t gnttab_dma_map_page(struct page *page)
+{
+ __gnttab_dma_map_page(page);
+ return ((maddr_t)pfn_to_mfn(page_to_pfn(page)) << PAGE_SHIFT);
+}
+
+static inline void gnttab_dma_unmap_page(maddr_t maddr)
+{
+ __gnttab_dma_unmap_page(virt_to_page(bus_to_virt(maddr)));
+}
+
+#endif /* _ASM_I386_GNTTAB_DMA_H */
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/include/mach-xen/asm/highmem.h 2008-10-29 09:55:56.000000000 +0100
@@ -0,0 +1,97 @@
+/*
+ * highmem.h: virtual kernel memory mappings for high memory
+ *
+ * Used in CONFIG_HIGHMEM systems for memory pages which
+ * are not addressable by direct kernel virtual addresses.
+ *
+ * Copyright (C) 1999 Gerhard Wichert, Siemens AG
+ * Gerhard.Wichert@pdb.siemens.de
+ *
+ *
+ * Redesigned the x86 32-bit VM architecture to deal with
+ * up to 16 Terabyte physical memory. With current x86 CPUs
+ * we now support up to 64 Gigabytes physical RAM.
+ *
+ * Copyright (C) 1999 Ingo Molnar <mingo@redhat.com>
+ */
+
+#ifndef _ASM_HIGHMEM_H
+#define _ASM_HIGHMEM_H
+
+#ifdef __KERNEL__
+
+#include <linux/interrupt.h>
+#include <linux/threads.h>
+#include <asm/kmap_types.h>
+#include <asm/tlbflush.h>
+
+/* declarations for highmem.c */
+extern unsigned long highstart_pfn, highend_pfn;
+
+extern pte_t *kmap_pte;
+extern pgprot_t kmap_prot;
+extern pte_t *pkmap_page_table;
+
+/*
+ * Right now we initialize only a single pte table. It can be extended
+ * easily, subsequent pte tables have to be allocated in one physical
+ * chunk of RAM.
+ */
+#ifdef CONFIG_X86_PAE
+#define LAST_PKMAP 512
+#else
+#define LAST_PKMAP 1024
+#endif
+/*
+ * Ordering is:
+ *
+ * FIXADDR_TOP
+ * fixed_addresses
+ * FIXADDR_START
+ * temp fixed addresses
+ * FIXADDR_BOOT_START
+ * Persistent kmap area
+ * PKMAP_BASE
+ * VMALLOC_END
+ * Vmalloc area
+ * VMALLOC_START
+ * high_memory
+ */
+#define PKMAP_BASE ( (FIXADDR_BOOT_START - PAGE_SIZE*(LAST_PKMAP + 1)) & PMD_MASK )
+#define LAST_PKMAP_MASK (LAST_PKMAP-1)
+#define PKMAP_NR(virt) ((virt-PKMAP_BASE) >> PAGE_SHIFT)
+#define PKMAP_ADDR(nr) (PKMAP_BASE + ((nr) << PAGE_SHIFT))
+
+extern void * FASTCALL(kmap_high(struct page *page));
+extern void FASTCALL(kunmap_high(struct page *page));
+
+void *kmap(struct page *page);
+void kunmap(struct page *page);
+void *kmap_atomic(struct page *page, enum km_type type);
+void *kmap_atomic_pte(struct page *page, enum km_type type);
+void kunmap_atomic(void *kvaddr, enum km_type type);
+void *kmap_atomic_pfn(unsigned long pfn, enum km_type type);
+struct page *kmap_atomic_to_page(void *ptr);
+
+#define flush_cache_kmaps() do { } while (0)
+
+void clear_highpage(struct page *);
+static inline void clear_user_highpage(struct page *page, unsigned long vaddr)
+{
+ clear_highpage(page);
+}
+#define __HAVE_ARCH_CLEAR_HIGHPAGE
+#define __HAVE_ARCH_CLEAR_USER_HIGHPAGE
+
+void copy_highpage(struct page *to, struct page *from);
+static inline void copy_user_highpage(struct page *to, struct page *from,
+ unsigned long vaddr)
+{
+ copy_highpage(to, from);
+}
+#define __HAVE_ARCH_COPY_HIGHPAGE
+#define __HAVE_ARCH_COPY_USER_HIGHPAGE
+
+#endif /* __KERNEL__ */
+
+#endif /* _ASM_HIGHMEM_H */
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/include/mach-xen/asm/hypercall_32.h 2009-06-23 09:28:21.000000000 +0200
@@ -0,0 +1,415 @@
+/******************************************************************************
+ * hypercall.h
+ *
+ * Linux-specific hypervisor handling.
+ *
+ * Copyright (c) 2002-2004, K A Fraser
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License version 2
+ * as published by the Free Software Foundation; or, when distributed
+ * separately from the Linux kernel or incorporated into other
+ * software packages, subject to the following license:
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this source file (the "Software"), to deal in the Software without
+ * restriction, including without limitation the rights to use, copy, modify,
+ * merge, publish, distribute, sublicense, and/or sell copies of the Software,
+ * and to permit persons to whom the Software is furnished to do so, subject to
+ * the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
+ * IN THE SOFTWARE.
+ */
+
+#ifndef __HYPERCALL_H__
+#define __HYPERCALL_H__
+
+#include <linux/string.h> /* memcpy() */
+#include <linux/stringify.h>
+
+#ifndef __HYPERVISOR_H__
+# error "please don't include this file directly"
+#endif
+
+#ifdef CONFIG_XEN
+#define HYPERCALL_STR(name) \
+ "call hypercall_page + ("__stringify(__HYPERVISOR_##name)" * 32)"
+#else
+#define HYPERCALL_STR(name) \
+ "mov hypercall_stubs,%%eax; " \
+ "add $("__stringify(__HYPERVISOR_##name)" * 32),%%eax; "\
+ "call *%%eax"
+#endif
+
+#define _hypercall0(type, name) \
+({ \
+ type __res; \
+ asm volatile ( \
+ HYPERCALL_STR(name) \
+ : "=a" (__res) \
+ : \
+ : "memory" ); \
+ __res; \
+})
+
+#define _hypercall1(type, name, a1) \
+({ \
+ type __res; \
+ long __ign1; \
+ asm volatile ( \
+ HYPERCALL_STR(name) \
+ : "=a" (__res), "=b" (__ign1) \
+ : "1" ((long)(a1)) \
+ : "memory" ); \
+ __res; \
+})
+
+#define _hypercall2(type, name, a1, a2) \
+({ \
+ type __res; \
+ long __ign1, __ign2; \
+ asm volatile ( \
+ HYPERCALL_STR(name) \
+ : "=a" (__res), "=b" (__ign1), "=c" (__ign2) \
+ : "1" ((long)(a1)), "2" ((long)(a2)) \
+ : "memory" ); \
+ __res; \
+})
+
+#define _hypercall3(type, name, a1, a2, a3) \
+({ \
+ type __res; \
+ long __ign1, __ign2, __ign3; \
+ asm volatile ( \
+ HYPERCALL_STR(name) \
+ : "=a" (__res), "=b" (__ign1), "=c" (__ign2), \
+ "=d" (__ign3) \
+ : "1" ((long)(a1)), "2" ((long)(a2)), \
+ "3" ((long)(a3)) \
+ : "memory" ); \
+ __res; \
+})
+
+#define _hypercall4(type, name, a1, a2, a3, a4) \
+({ \
+ type __res; \
+ long __ign1, __ign2, __ign3, __ign4; \
+ asm volatile ( \
+ HYPERCALL_STR(name) \
+ : "=a" (__res), "=b" (__ign1), "=c" (__ign2), \
+ "=d" (__ign3), "=S" (__ign4) \
+ : "1" ((long)(a1)), "2" ((long)(a2)), \
+ "3" ((long)(a3)), "4" ((long)(a4)) \
+ : "memory" ); \
+ __res; \
+})
+
+#define _hypercall5(type, name, a1, a2, a3, a4, a5) \
+({ \
+ type __res; \
+ long __ign1, __ign2, __ign3, __ign4, __ign5; \
+ asm volatile ( \
+ HYPERCALL_STR(name) \
+ : "=a" (__res), "=b" (__ign1), "=c" (__ign2), \
+ "=d" (__ign3), "=S" (__ign4), "=D" (__ign5) \
+ : "1" ((long)(a1)), "2" ((long)(a2)), \
+ "3" ((long)(a3)), "4" ((long)(a4)), \
+ "5" ((long)(a5)) \
+ : "memory" ); \
+ __res; \
+})
+
+static inline int __must_check
+HYPERVISOR_set_trap_table(
+ const trap_info_t *table)
+{
+ return _hypercall1(int, set_trap_table, table);
+}
+
+static inline int __must_check
+HYPERVISOR_mmu_update(
+ mmu_update_t *req, unsigned int count, unsigned int *success_count,
+ domid_t domid)
+{
+ return _hypercall4(int, mmu_update, req, count, success_count, domid);
+}
+
+static inline int __must_check
+HYPERVISOR_mmuext_op(
+ struct mmuext_op *op, unsigned int count, unsigned int *success_count,
+ domid_t domid)
+{
+ return _hypercall4(int, mmuext_op, op, count, success_count, domid);
+}
+
+static inline int __must_check
+HYPERVISOR_set_gdt(
+ unsigned long *frame_list, unsigned int entries)
+{
+ return _hypercall2(int, set_gdt, frame_list, entries);
+}
+
+static inline int __must_check
+HYPERVISOR_stack_switch(
+ unsigned long ss, unsigned long esp)
+{
+ return _hypercall2(int, stack_switch, ss, esp);
+}
+
+static inline int __must_check
+HYPERVISOR_set_callbacks(
+ unsigned long event_selector, unsigned long event_address,
+ unsigned long failsafe_selector, unsigned long failsafe_address)
+{
+ return _hypercall4(int, set_callbacks,
+ event_selector, event_address,
+ failsafe_selector, failsafe_address);
+}
+
+static inline int
+HYPERVISOR_fpu_taskswitch(
+ int set)
+{
+ return _hypercall1(int, fpu_taskswitch, set);
+}
+
+static inline int __must_check
+HYPERVISOR_sched_op_compat(
+ int cmd, unsigned long arg)
+{
+ return _hypercall2(int, sched_op_compat, cmd, arg);
+}
+
+static inline int __must_check
+HYPERVISOR_sched_op(
+ int cmd, void *arg)
+{
+ return _hypercall2(int, sched_op, cmd, arg);
+}
+
+static inline long __must_check
+HYPERVISOR_set_timer_op(
+ u64 timeout)
+{
+ unsigned long timeout_hi = (unsigned long)(timeout>>32);
+ unsigned long timeout_lo = (unsigned long)timeout;
+ return _hypercall2(long, set_timer_op, timeout_lo, timeout_hi);
+}
+
+static inline int __must_check
+HYPERVISOR_platform_op(
+ struct xen_platform_op *platform_op)
+{
+ platform_op->interface_version = XENPF_INTERFACE_VERSION;
+ return _hypercall1(int, platform_op, platform_op);
+}
+
+static inline int __must_check
+HYPERVISOR_set_debugreg(
+ unsigned int reg, unsigned long value)
+{
+ return _hypercall2(int, set_debugreg, reg, value);
+}
+
+static inline unsigned long __must_check
+HYPERVISOR_get_debugreg(
+ unsigned int reg)
+{
+ return _hypercall1(unsigned long, get_debugreg, reg);
+}
+
+static inline int __must_check
+HYPERVISOR_update_descriptor(
+ u64 ma, u64 desc)
+{
+ return _hypercall4(int, update_descriptor, ma, ma>>32, desc, desc>>32);
+}
+
+static inline int __must_check
+HYPERVISOR_memory_op(
+ unsigned int cmd, void *arg)
+{
+ return _hypercall2(int, memory_op, cmd, arg);
+}
+
+static inline int __must_check
+HYPERVISOR_multicall(
+ multicall_entry_t *call_list, unsigned int nr_calls)
+{
+ return _hypercall2(int, multicall, call_list, nr_calls);
+}
+
+static inline int __must_check
+HYPERVISOR_update_va_mapping(
+ unsigned long va, pte_t new_val, unsigned long flags)
+{
+ unsigned long pte_hi = 0;
+#ifdef CONFIG_X86_PAE
+ pte_hi = new_val.pte_high;
+#endif
+ return _hypercall4(int, update_va_mapping, va,
+ new_val.pte_low, pte_hi, flags);
+}
+
+static inline int __must_check
+HYPERVISOR_event_channel_op(
+ int cmd, void *arg)
+{
+ int rc = _hypercall2(int, event_channel_op, cmd, arg);
+
+#if CONFIG_XEN_COMPAT <= 0x030002
+ if (unlikely(rc == -ENOSYS)) {
+ struct evtchn_op op;
+ op.cmd = cmd;
+ memcpy(&op.u, arg, sizeof(op.u));
+ rc = _hypercall1(int, event_channel_op_compat, &op);
+ memcpy(arg, &op.u, sizeof(op.u));
+ }
+#endif
+
+ return rc;
+}
+
+static inline int __must_check
+HYPERVISOR_xen_version(
+ int cmd, void *arg)
+{
+ return _hypercall2(int, xen_version, cmd, arg);
+}
+
+static inline int __must_check
+HYPERVISOR_console_io(
+ int cmd, unsigned int count, char *str)
+{
+ return _hypercall3(int, console_io, cmd, count, str);
+}
+
+static inline int __must_check
+HYPERVISOR_physdev_op(
+ int cmd, void *arg)
+{
+ int rc = _hypercall2(int, physdev_op, cmd, arg);
+
+#if CONFIG_XEN_COMPAT <= 0x030002
+ if (unlikely(rc == -ENOSYS)) {
+ struct physdev_op op;
+ op.cmd = cmd;
+ memcpy(&op.u, arg, sizeof(op.u));
+ rc = _hypercall1(int, physdev_op_compat, &op);
+ memcpy(arg, &op.u, sizeof(op.u));
+ }
+#endif
+
+ return rc;
+}
+
+static inline int __must_check
+HYPERVISOR_grant_table_op(
+ unsigned int cmd, void *uop, unsigned int count)
+{
+ return _hypercall3(int, grant_table_op, cmd, uop, count);
+}
+
+static inline int __must_check
+HYPERVISOR_update_va_mapping_otherdomain(
+ unsigned long va, pte_t new_val, unsigned long flags, domid_t domid)
+{
+ unsigned long pte_hi = 0;
+#ifdef CONFIG_X86_PAE
+ pte_hi = new_val.pte_high;
+#endif
+ return _hypercall5(int, update_va_mapping_otherdomain, va,
+ new_val.pte_low, pte_hi, flags, domid);
+}
+
+static inline int __must_check
+HYPERVISOR_vm_assist(
+ unsigned int cmd, unsigned int type)
+{
+ return _hypercall2(int, vm_assist, cmd, type);
+}
+
+static inline int __must_check
+HYPERVISOR_vcpu_op(
+ int cmd, unsigned int vcpuid, void *extra_args)
+{
+ return _hypercall3(int, vcpu_op, cmd, vcpuid, extra_args);
+}
+
+static inline int __must_check
+HYPERVISOR_suspend(
+ unsigned long srec)
+{
+ struct sched_shutdown sched_shutdown = {
+ .reason = SHUTDOWN_suspend
+ };
+
+ int rc = _hypercall3(int, sched_op, SCHEDOP_shutdown,
+ &sched_shutdown, srec);
+
+#if CONFIG_XEN_COMPAT <= 0x030002
+ if (rc == -ENOSYS)
+ rc = _hypercall3(int, sched_op_compat, SCHEDOP_shutdown,
+ SHUTDOWN_suspend, srec);
+#endif
+
+ return rc;
+}
+
+#if CONFIG_XEN_COMPAT <= 0x030002
+static inline int
+HYPERVISOR_nmi_op(
+ unsigned long op, void *arg)
+{
+ return _hypercall2(int, nmi_op, op, arg);
+}
+#endif
+
+#ifndef CONFIG_XEN
+static inline unsigned long __must_check
+HYPERVISOR_hvm_op(
+ int op, void *arg)
+{
+ return _hypercall2(unsigned long, hvm_op, op, arg);
+}
+#endif
+
+static inline int __must_check
+HYPERVISOR_callback_op(
+ int cmd, const void *arg)
+{
+ return _hypercall2(int, callback_op, cmd, arg);
+}
+
+static inline int __must_check
+HYPERVISOR_xenoprof_op(
+ int op, void *arg)
+{
+ return _hypercall2(int, xenoprof_op, op, arg);
+}
+
+static inline int __must_check
+HYPERVISOR_kexec_op(
+ unsigned long op, void *args)
+{
+ return _hypercall2(int, kexec_op, op, args);
+}
+
+static inline int __must_check
+HYPERVISOR_tmem_op(
+ struct tmem_op *op)
+{
+ return _hypercall1(int, tmem_op, op);
+}
+
+
+#endif /* __HYPERCALL_H__ */
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/include/mach-xen/asm/hypervisor.h 2009-07-13 14:25:35.000000000 +0200
@@ -0,0 +1,263 @@
+/******************************************************************************
+ * hypervisor.h
+ *
+ * Linux-specific hypervisor handling.
+ *
+ * Copyright (c) 2002-2004, K A Fraser
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License version 2
+ * as published by the Free Software Foundation; or, when distributed
+ * separately from the Linux kernel or incorporated into other
+ * software packages, subject to the following license:
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this source file (the "Software"), to deal in the Software without
+ * restriction, including without limitation the rights to use, copy, modify,
+ * merge, publish, distribute, sublicense, and/or sell copies of the Software,
+ * and to permit persons to whom the Software is furnished to do so, subject to
+ * the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
+ * IN THE SOFTWARE.
+ */
+
+#ifndef __HYPERVISOR_H__
+#define __HYPERVISOR_H__
+
+#include <linux/types.h>
+#include <linux/kernel.h>
+#include <linux/version.h>
+#include <linux/errno.h>
+#include <xen/interface/xen.h>
+#include <xen/interface/platform.h>
+#include <xen/interface/event_channel.h>
+#include <xen/interface/physdev.h>
+#include <xen/interface/sched.h>
+#include <xen/interface/nmi.h>
+#include <xen/interface/tmem.h>
+#include <asm/ptrace.h>
+#include <asm/page.h>
+#if defined(__i386__)
+# ifdef CONFIG_X86_PAE
+# include <asm-generic/pgtable-nopud.h>
+# else
+# include <asm-generic/pgtable-nopmd.h>
+# endif
+#elif defined(__x86_64__) && LINUX_VERSION_CODE < KERNEL_VERSION(2,6,11)
+# include <asm-generic/pgtable-nopud.h>
+#endif
+
+extern shared_info_t *HYPERVISOR_shared_info;
+
+#define vcpu_info(cpu) (HYPERVISOR_shared_info->vcpu_info + (cpu))
+#ifdef CONFIG_SMP
+#define current_vcpu_info() vcpu_info(smp_processor_id())
+#else
+#define current_vcpu_info() vcpu_info(0)
+#endif
+
+#ifdef CONFIG_X86_32
+extern unsigned long hypervisor_virt_start;
+#endif
+
+/* arch/xen/i386/kernel/setup.c */
+extern start_info_t *xen_start_info;
+#ifdef CONFIG_XEN_PRIVILEGED_GUEST
+#define is_initial_xendomain() (xen_start_info->flags & SIF_INITDOMAIN)
+#else
+#define is_initial_xendomain() 0
+#endif
+
+/* arch/xen/kernel/evtchn.c */
+/* Force a proper event-channel callback from Xen. */
+void force_evtchn_callback(void);
+
+/* arch/xen/kernel/process.c */
+void xen_cpu_idle (void);
+
+/* arch/xen/i386/kernel/hypervisor.c */
+void do_hypervisor_callback(struct pt_regs *regs);
+
+/* arch/xen/i386/mm/hypervisor.c */
+/*
+ * NB. ptr values should be PHYSICAL, not MACHINE. 'vals' should be already
+ * be MACHINE addresses.
+ */
+
+void xen_pt_switch(unsigned long ptr);
+void xen_new_user_pt(unsigned long ptr); /* x86_64 only */
+void xen_load_gs(unsigned int selector); /* x86_64 only */
+void xen_tlb_flush(void);
+void xen_invlpg(unsigned long ptr);
+
+void xen_l1_entry_update(pte_t *ptr, pte_t val);
+void xen_l2_entry_update(pmd_t *ptr, pmd_t val);
+void xen_l3_entry_update(pud_t *ptr, pud_t val); /* x86_64/PAE */
+void xen_l4_entry_update(pgd_t *ptr, pgd_t val); /* x86_64 only */
+void xen_pgd_pin(unsigned long ptr);
+void xen_pgd_unpin(unsigned long ptr);
+
+void xen_set_ldt(const void *ptr, unsigned int ents);
+
+#ifdef CONFIG_SMP
+#include <linux/cpumask.h>
+void xen_tlb_flush_all(void);
+void xen_invlpg_all(unsigned long ptr);
+void xen_tlb_flush_mask(cpumask_t *mask);
+void xen_invlpg_mask(cpumask_t *mask, unsigned long ptr);
+#else
+#define xen_tlb_flush_all xen_tlb_flush
+#define xen_invlpg_all xen_invlpg
+#endif
+
+/* Returns zero on success else negative errno. */
+int xen_create_contiguous_region(
+ unsigned long vstart, unsigned int order, unsigned int address_bits);
+void xen_destroy_contiguous_region(
+ unsigned long vstart, unsigned int order);
+
+struct page;
+
+int xen_limit_pages_to_max_mfn(
+ struct page *pages, unsigned int order, unsigned int address_bits);
+
+/* Turn jiffies into Xen system time. */
+u64 jiffies_to_st(unsigned long jiffies);
+
+#ifdef CONFIG_XEN_SCRUB_PAGES
+void scrub_pages(void *, unsigned int);
+#else
+#define scrub_pages(_p,_n) ((void)0)
+#endif
+
+#include <xen/hypercall.h>
+
+#if defined(CONFIG_X86_64)
+#define MULTI_UVMFLAGS_INDEX 2
+#define MULTI_UVMDOMID_INDEX 3
+#else
+#define MULTI_UVMFLAGS_INDEX 3
+#define MULTI_UVMDOMID_INDEX 4
+#endif
+
+#ifdef CONFIG_XEN
+#define is_running_on_xen() 1
+#else
+extern char *hypercall_stubs;
+#define is_running_on_xen() (!!hypercall_stubs)
+#endif
+
+static inline int
+HYPERVISOR_yield(
+ void)
+{
+ int rc = HYPERVISOR_sched_op(SCHEDOP_yield, NULL);
+
+#if CONFIG_XEN_COMPAT <= 0x030002
+ if (rc == -ENOSYS)
+ rc = HYPERVISOR_sched_op_compat(SCHEDOP_yield, 0);
+#endif
+
+ return rc;
+}
+
+static inline int
+HYPERVISOR_block(
+ void)
+{
+ int rc = HYPERVISOR_sched_op(SCHEDOP_block, NULL);
+
+#if CONFIG_XEN_COMPAT <= 0x030002
+ if (rc == -ENOSYS)
+ rc = HYPERVISOR_sched_op_compat(SCHEDOP_block, 0);
+#endif
+
+ return rc;
+}
+
+static inline void /*__noreturn*/
+HYPERVISOR_shutdown(
+ unsigned int reason)
+{
+ struct sched_shutdown sched_shutdown = {
+ .reason = reason
+ };
+
+ VOID(HYPERVISOR_sched_op(SCHEDOP_shutdown, &sched_shutdown));
+#if CONFIG_XEN_COMPAT <= 0x030002
+ VOID(HYPERVISOR_sched_op_compat(SCHEDOP_shutdown, reason));
+#endif
+ /* Don't recurse needlessly. */
+ BUG_ON(reason != SHUTDOWN_crash);
+ for(;;);
+}
+
+static inline int __must_check
+HYPERVISOR_poll(
+ evtchn_port_t *ports, unsigned int nr_ports, u64 timeout)
+{
+ int rc;
+ struct sched_poll sched_poll = {
+ .nr_ports = nr_ports,
+ .timeout = jiffies_to_st(timeout)
+ };
+ set_xen_guest_handle(sched_poll.ports, ports);
+
+ rc = HYPERVISOR_sched_op(SCHEDOP_poll, &sched_poll);
+#if CONFIG_XEN_COMPAT <= 0x030002
+ if (rc == -ENOSYS)
+ rc = HYPERVISOR_sched_op_compat(SCHEDOP_yield, 0);
+#endif
+
+ return rc;
+}
+
+#ifdef CONFIG_XEN
+
+static inline void
+MULTI_update_va_mapping(
+ multicall_entry_t *mcl, unsigned long va,
+ pte_t new_val, unsigned long flags)
+{
+ mcl->op = __HYPERVISOR_update_va_mapping;
+ mcl->args[0] = va;
+#if defined(CONFIG_X86_64)
+ mcl->args[1] = new_val.pte;
+#elif defined(CONFIG_X86_PAE)
+ mcl->args[1] = new_val.pte_low;
+ mcl->args[2] = new_val.pte_high;
+#else
+ mcl->args[1] = new_val.pte_low;
+ mcl->args[2] = 0;
+#endif
+ mcl->args[MULTI_UVMFLAGS_INDEX] = flags;
+}
+
+static inline void
+MULTI_grant_table_op(multicall_entry_t *mcl, unsigned int cmd,
+ void *uop, unsigned int count)
+{
+ mcl->op = __HYPERVISOR_grant_table_op;
+ mcl->args[0] = cmd;
+ mcl->args[1] = (unsigned long)uop;
+ mcl->args[2] = count;
+}
+
+#else /* !defined(CONFIG_XEN) */
+
+/* Multicalls not supported for HVM guests. */
+#define MULTI_update_va_mapping(a,b,c,d) ((void)0)
+#define MULTI_grant_table_op(a,b,c,d) ((void)0)
+
+#endif
+
+#endif /* __HYPERVISOR_H__ */
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/include/mach-xen/asm/irqflags_32.h 2007-06-12 13:14:02.000000000 +0200
@@ -0,0 +1,127 @@
+/*
+ * include/asm-i386/irqflags.h
+ *
+ * IRQ flags handling
+ *
+ * This file gets included from lowlevel asm headers too, to provide
+ * wrapped versions of the local_irq_*() APIs, based on the
+ * raw_local_irq_*() functions from the lowlevel headers.
+ */
+#ifndef _ASM_IRQFLAGS_H
+#define _ASM_IRQFLAGS_H
+
+#ifndef __ASSEMBLY__
+
+/*
+ * The use of 'barrier' in the following reflects their use as local-lock
+ * operations. Reentrancy must be prevented (e.g., __cli()) /before/ following
+ * critical operations are executed. All critical operations must complete
+ * /before/ reentrancy is permitted (e.g., __sti()). Alpha architecture also
+ * includes these barriers, for example.
+ */
+
+#define __raw_local_save_flags() (current_vcpu_info()->evtchn_upcall_mask)
+
+#define raw_local_save_flags(flags) \
+ do { (flags) = __raw_local_save_flags(); } while (0)
+
+#define raw_local_irq_restore(x) \
+do { \
+ vcpu_info_t *_vcpu; \
+ barrier(); \
+ _vcpu = current_vcpu_info(); \
+ if ((_vcpu->evtchn_upcall_mask = (x)) == 0) { \
+ barrier(); /* unmask then check (avoid races) */ \
+ if (unlikely(_vcpu->evtchn_upcall_pending)) \
+ force_evtchn_callback(); \
+ } \
+} while (0)
+
+#define raw_local_irq_disable() \
+do { \
+ current_vcpu_info()->evtchn_upcall_mask = 1; \
+ barrier(); \
+} while (0)
+
+#define raw_local_irq_enable() \
+do { \
+ vcpu_info_t *_vcpu; \
+ barrier(); \
+ _vcpu = current_vcpu_info(); \
+ _vcpu->evtchn_upcall_mask = 0; \
+ barrier(); /* unmask then check (avoid races) */ \
+ if (unlikely(_vcpu->evtchn_upcall_pending)) \
+ force_evtchn_callback(); \
+} while (0)
+
+/*
+ * Used in the idle loop; sti takes one instruction cycle
+ * to complete:
+ */
+void raw_safe_halt(void);
+
+/*
+ * Used when interrupts are already enabled or to
+ * shutdown the processor:
+ */
+void halt(void);
+
+static inline int raw_irqs_disabled_flags(unsigned long flags)
+{
+ return (flags != 0);
+}
+
+#define raw_irqs_disabled() \
+({ \
+ unsigned long flags = __raw_local_save_flags(); \
+ \
+ raw_irqs_disabled_flags(flags); \
+})
+
+/*
+ * For spinlocks, etc:
+ */
+#define __raw_local_irq_save() \
+({ \
+ unsigned long flags = __raw_local_save_flags(); \
+ \
+ raw_local_irq_disable(); \
+ \
+ flags; \
+})
+
+#define raw_local_irq_save(flags) \
+ do { (flags) = __raw_local_irq_save(); } while (0)
+
+#endif /* __ASSEMBLY__ */
+
+/*
+ * Do the CPU's IRQ-state tracing from assembly code. We call a
+ * C function, so save all the C-clobbered registers:
+ */
+#ifdef CONFIG_TRACE_IRQFLAGS
+
+# define TRACE_IRQS_ON \
+ pushl %eax; \
+ pushl %ecx; \
+ pushl %edx; \
+ call trace_hardirqs_on; \
+ popl %edx; \
+ popl %ecx; \
+ popl %eax;
+
+# define TRACE_IRQS_OFF \
+ pushl %eax; \
+ pushl %ecx; \
+ pushl %edx; \
+ call trace_hardirqs_off; \
+ popl %edx; \
+ popl %ecx; \
+ popl %eax;
+
+#else
+# define TRACE_IRQS_ON
+# define TRACE_IRQS_OFF
+#endif
+
+#endif
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/include/mach-xen/asm/maddr_32.h 2008-04-02 12:34:02.000000000 +0200
@@ -0,0 +1,193 @@
+#ifndef _I386_MADDR_H
+#define _I386_MADDR_H
+
+#include <xen/features.h>
+#include <xen/interface/xen.h>
+
+/**** MACHINE <-> PHYSICAL CONVERSION MACROS ****/
+#define INVALID_P2M_ENTRY (~0UL)
+#define FOREIGN_FRAME_BIT (1UL<<31)
+#define FOREIGN_FRAME(m) ((m) | FOREIGN_FRAME_BIT)
+
+/* Definitions for machine and pseudophysical addresses. */
+#ifdef CONFIG_X86_PAE
+typedef unsigned long long paddr_t;
+typedef unsigned long long maddr_t;
+#else
+typedef unsigned long paddr_t;
+typedef unsigned long maddr_t;
+#endif
+
+#ifdef CONFIG_XEN
+
+extern unsigned long *phys_to_machine_mapping;
+extern unsigned long max_mapnr;
+
+#undef machine_to_phys_mapping
+extern unsigned long *machine_to_phys_mapping;
+extern unsigned int machine_to_phys_order;
+
+static inline unsigned long pfn_to_mfn(unsigned long pfn)
+{
+ if (xen_feature(XENFEAT_auto_translated_physmap))
+ return pfn;
+ BUG_ON(max_mapnr && pfn >= max_mapnr);
+ return phys_to_machine_mapping[pfn] & ~FOREIGN_FRAME_BIT;
+}
+
+static inline int phys_to_machine_mapping_valid(unsigned long pfn)
+{
+ if (xen_feature(XENFEAT_auto_translated_physmap))
+ return 1;
+ BUG_ON(max_mapnr && pfn >= max_mapnr);
+ return (phys_to_machine_mapping[pfn] != INVALID_P2M_ENTRY);
+}
+
+static inline unsigned long mfn_to_pfn(unsigned long mfn)
+{
+ unsigned long pfn;
+
+ if (xen_feature(XENFEAT_auto_translated_physmap))
+ return mfn;
+
+ if (unlikely((mfn >> machine_to_phys_order) != 0))
+ return max_mapnr;
+
+ /* The array access can fail (e.g., device space beyond end of RAM). */
+ asm (
+ "1: movl %1,%0\n"
+ "2:\n"
+ ".section .fixup,\"ax\"\n"
+ "3: movl %2,%0\n"
+ " jmp 2b\n"
+ ".previous\n"
+ ".section __ex_table,\"a\"\n"
+ " .align 4\n"
+ " .long 1b,3b\n"
+ ".previous"
+ : "=r" (pfn)
+ : "m" (machine_to_phys_mapping[mfn]), "m" (max_mapnr) );
+
+ return pfn;
+}
+
+/*
+ * We detect special mappings in one of two ways:
+ * 1. If the MFN is an I/O page then Xen will set the m2p entry
+ * to be outside our maximum possible pseudophys range.
+ * 2. If the MFN belongs to a different domain then we will certainly
+ * not have MFN in our p2m table. Conversely, if the page is ours,
+ * then we'll have p2m(m2p(MFN))==MFN.
+ * If we detect a special mapping then it doesn't have a 'struct page'.
+ * We force !pfn_valid() by returning an out-of-range pointer.
+ *
+ * NB. These checks require that, for any MFN that is not in our reservation,
+ * there is no PFN such that p2m(PFN) == MFN. Otherwise we can get confused if
+ * we are foreign-mapping the MFN, and the other domain as m2p(MFN) == PFN.
+ * Yikes! Various places must poke in INVALID_P2M_ENTRY for safety.
+ *
+ * NB2. When deliberately mapping foreign pages into the p2m table, you *must*
+ * use FOREIGN_FRAME(). This will cause pte_pfn() to choke on it, as we
+ * require. In all the cases we care about, the FOREIGN_FRAME bit is
+ * masked (e.g., pfn_to_mfn()) so behaviour there is correct.
+ */
+static inline unsigned long mfn_to_local_pfn(unsigned long mfn)
+{
+ unsigned long pfn = mfn_to_pfn(mfn);
+ if ((pfn < max_mapnr)
+ && !xen_feature(XENFEAT_auto_translated_physmap)
+ && (phys_to_machine_mapping[pfn] != mfn))
+ return max_mapnr; /* force !pfn_valid() */
+ return pfn;
+}
+
+static inline void set_phys_to_machine(unsigned long pfn, unsigned long mfn)
+{
+ BUG_ON(max_mapnr && pfn >= max_mapnr);
+ if (xen_feature(XENFEAT_auto_translated_physmap)) {
+ BUG_ON(pfn != mfn && mfn != INVALID_P2M_ENTRY);
+ return;
+ }
+ phys_to_machine_mapping[pfn] = mfn;
+}
+
+static inline maddr_t phys_to_machine(paddr_t phys)
+{
+ maddr_t machine = pfn_to_mfn(phys >> PAGE_SHIFT);
+ machine = (machine << PAGE_SHIFT) | (phys & ~PAGE_MASK);
+ return machine;
+}
+
+static inline paddr_t machine_to_phys(maddr_t machine)
+{
+ paddr_t phys = mfn_to_pfn(machine >> PAGE_SHIFT);
+ phys = (phys << PAGE_SHIFT) | (machine & ~PAGE_MASK);
+ return phys;
+}
+
+#ifdef CONFIG_X86_PAE
+static inline paddr_t pte_phys_to_machine(paddr_t phys)
+{
+ /*
+ * In PAE mode, the NX bit needs to be dealt with in the value
+ * passed to pfn_to_mfn(). On x86_64, we need to mask it off,
+ * but for i386 the conversion to ulong for the argument will
+ * clip it off.
+ */
+ maddr_t machine = pfn_to_mfn(phys >> PAGE_SHIFT);
+ machine = (machine << PAGE_SHIFT) | (phys & ~PHYSICAL_PAGE_MASK);
+ return machine;
+}
+
+static inline paddr_t pte_machine_to_phys(maddr_t machine)
+{
+ /*
+ * In PAE mode, the NX bit needs to be dealt with in the value
+ * passed to mfn_to_pfn(). On x86_64, we need to mask it off,
+ * but for i386 the conversion to ulong for the argument will
+ * clip it off.
+ */
+ paddr_t phys = mfn_to_pfn(machine >> PAGE_SHIFT);
+ phys = (phys << PAGE_SHIFT) | (machine & ~PHYSICAL_PAGE_MASK);
+ return phys;
+}
+#endif
+
+#ifdef CONFIG_X86_PAE
+#define __pte_ma(x) ((pte_t) { (x), (maddr_t)(x) >> 32 } )
+static inline pte_t pfn_pte_ma(unsigned long page_nr, pgprot_t pgprot)
+{
+ pte_t pte;
+
+ pte.pte_high = (page_nr >> (32 - PAGE_SHIFT)) | \
+ (pgprot_val(pgprot) >> 32);
+ pte.pte_high &= (__supported_pte_mask >> 32);
+ pte.pte_low = ((page_nr << PAGE_SHIFT) | pgprot_val(pgprot)) & \
+ __supported_pte_mask;
+ return pte;
+}
+#else
+#define __pte_ma(x) ((pte_t) { (x) } )
+#define pfn_pte_ma(pfn, prot) __pte_ma(((pfn) << PAGE_SHIFT) | pgprot_val(prot))
+#endif
+
+#else /* !CONFIG_XEN */
+
+#define pfn_to_mfn(pfn) (pfn)
+#define mfn_to_pfn(mfn) (mfn)
+#define mfn_to_local_pfn(mfn) (mfn)
+#define set_phys_to_machine(pfn, mfn) ((void)0)
+#define phys_to_machine_mapping_valid(pfn) (1)
+#define phys_to_machine(phys) ((maddr_t)(phys))
+#define machine_to_phys(mach) ((paddr_t)(mach))
+#define pfn_pte_ma(pfn, prot) pfn_pte(pfn, prot)
+#define __pte_ma(x) __pte(x)
+
+#endif /* !CONFIG_XEN */
+
+/* VIRT <-> MACHINE conversion */
+#define virt_to_machine(v) (phys_to_machine(__pa(v)))
+#define virt_to_mfn(v) (pfn_to_mfn(__pa(v) >> PAGE_SHIFT))
+#define mfn_to_virt(m) (__va(mfn_to_pfn(m) << PAGE_SHIFT))
+
+#endif /* _I386_MADDR_H */
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/include/mach-xen/asm/mmu_context_32.h 2007-06-12 13:14:02.000000000 +0200
@@ -0,0 +1,108 @@
+#ifndef __I386_SCHED_H
+#define __I386_SCHED_H
+
+#include <asm/desc.h>
+#include <asm/atomic.h>
+#include <asm/pgalloc.h>
+#include <asm/tlbflush.h>
+
+/*
+ * Used for LDT copy/destruction.
+ */
+int init_new_context(struct task_struct *tsk, struct mm_struct *mm);
+void destroy_context(struct mm_struct *mm);
+
+
+static inline void enter_lazy_tlb(struct mm_struct *mm, struct task_struct *tsk)
+{
+#if 0 /* XEN: no lazy tlb */
+ unsigned cpu = smp_processor_id();
+ if (per_cpu(cpu_tlbstate, cpu).state == TLBSTATE_OK)
+ per_cpu(cpu_tlbstate, cpu).state = TLBSTATE_LAZY;
+#endif
+}
+
+#define prepare_arch_switch(next) __prepare_arch_switch()
+
+static inline void __prepare_arch_switch(void)
+{
+ /*
+ * Save away %fs and %gs. No need to save %es and %ds, as those
+ * are always kernel segments while inside the kernel. Must
+ * happen before reload of cr3/ldt (i.e., not in __switch_to).
+ */
+ asm volatile ( "mov %%fs,%0 ; mov %%gs,%1"
+ : "=m" (current->thread.fs),
+ "=m" (current->thread.gs));
+ asm volatile ( "movl %0,%%fs ; movl %0,%%gs"
+ : : "r" (0) );
+}
+
+extern void mm_pin(struct mm_struct *mm);
+extern void mm_unpin(struct mm_struct *mm);
+void mm_pin_all(void);
+
+static inline void switch_mm(struct mm_struct *prev,
+ struct mm_struct *next,
+ struct task_struct *tsk)
+{
+ int cpu = smp_processor_id();
+ struct mmuext_op _op[2], *op = _op;
+
+ if (likely(prev != next)) {
+ BUG_ON(!xen_feature(XENFEAT_writable_page_tables) &&
+ !test_bit(PG_pinned, &virt_to_page(next->pgd)->flags));
+
+ /* stop flush ipis for the previous mm */
+ cpu_clear(cpu, prev->cpu_vm_mask);
+#if 0 /* XEN: no lazy tlb */
+ per_cpu(cpu_tlbstate, cpu).state = TLBSTATE_OK;
+ per_cpu(cpu_tlbstate, cpu).active_mm = next;
+#endif
+ cpu_set(cpu, next->cpu_vm_mask);
+
+ /* Re-load page tables: load_cr3(next->pgd) */
+ op->cmd = MMUEXT_NEW_BASEPTR;
+ op->arg1.mfn = pfn_to_mfn(__pa(next->pgd) >> PAGE_SHIFT);
+ op++;
+
+ /*
+ * load the LDT, if the LDT is different:
+ */
+ if (unlikely(prev->context.ldt != next->context.ldt)) {
+ /* load_LDT_nolock(&next->context, cpu) */
+ op->cmd = MMUEXT_SET_LDT;
+ op->arg1.linear_addr = (unsigned long)next->context.ldt;
+ op->arg2.nr_ents = next->context.size;
+ op++;
+ }
+
+ BUG_ON(HYPERVISOR_mmuext_op(_op, op-_op, NULL, DOMID_SELF));
+ }
+#if 0 /* XEN: no lazy tlb */
+ else {
+ per_cpu(cpu_tlbstate, cpu).state = TLBSTATE_OK;
+ BUG_ON(per_cpu(cpu_tlbstate, cpu).active_mm != next);
+
+ if (!cpu_test_and_set(cpu, next->cpu_vm_mask)) {
+ /* We were in lazy tlb mode and leave_mm disabled
+ * tlb flush IPI delivery. We must reload %cr3.
+ */
+ load_cr3(next->pgd);
+ load_LDT_nolock(&next->context, cpu);
+ }
+ }
+#endif
+}
+
+#define deactivate_mm(tsk, mm) \
+ asm("movl %0,%%fs ; movl %0,%%gs": :"r" (0))
+
+static inline void activate_mm(struct mm_struct *prev, struct mm_struct *next)
+{
+ if (!test_bit(PG_pinned, &virt_to_page(next->pgd)->flags))
+ mm_pin(next);
+ switch_mm(prev, next, NULL);
+}
+
+#endif
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/include/mach-xen/asm/pgalloc_32.h 2008-07-21 11:00:33.000000000 +0200
@@ -0,0 +1,59 @@
+#ifndef _I386_PGALLOC_H
+#define _I386_PGALLOC_H
+
+#include <asm/fixmap.h>
+#include <linux/threads.h>
+#include <linux/mm.h> /* for struct page */
+#include <asm/io.h> /* for phys_to_virt and page_to_pseudophys */
+
+#define pmd_populate_kernel(mm, pmd, pte) \
+ set_pmd(pmd, __pmd(_PAGE_TABLE + __pa(pte)))
+
+#define pmd_populate(mm, pmd, pte) \
+do { \
+ unsigned long pfn = page_to_pfn(pte); \
+ if (test_bit(PG_pinned, &virt_to_page((mm)->pgd)->flags)) { \
+ if (!PageHighMem(pte)) \
+ BUG_ON(HYPERVISOR_update_va_mapping( \
+ (unsigned long)__va(pfn << PAGE_SHIFT), \
+ pfn_pte(pfn, PAGE_KERNEL_RO), 0)); \
+ else if (!test_and_set_bit(PG_pinned, &pte->flags)) \
+ kmap_flush_unused(); \
+ set_pmd(pmd, \
+ __pmd(_PAGE_TABLE + ((paddr_t)pfn << PAGE_SHIFT))); \
+ } else \
+ *(pmd) = __pmd(_PAGE_TABLE + ((paddr_t)pfn << PAGE_SHIFT)); \
+} while (0)
+
+/*
+ * Allocate and free page tables.
+ */
+extern pgd_t *pgd_alloc(struct mm_struct *);
+extern void pgd_free(pgd_t *pgd);
+
+extern pte_t *pte_alloc_one_kernel(struct mm_struct *, unsigned long);
+extern struct page *pte_alloc_one(struct mm_struct *, unsigned long);
+
+static inline void pte_free_kernel(pte_t *pte)
+{
+ make_lowmem_page_writable(pte, XENFEAT_writable_page_tables);
+ free_page((unsigned long)pte);
+}
+
+extern void pte_free(struct page *pte);
+
+#define __pte_free_tlb(tlb,pte) tlb_remove_page((tlb),(pte))
+
+#ifdef CONFIG_X86_PAE
+/*
+ * In the PAE case we free the pmds as part of the pgd.
+ */
+#define pmd_alloc_one(mm, addr) ({ BUG(); ((pmd_t *)2); })
+#define pmd_free(x) do { } while (0)
+#define __pmd_free_tlb(tlb,x) do { } while (0)
+#define pud_populate(mm, pmd, pte) BUG()
+#endif
+
+#define check_pgt_cache() do { } while (0)
+
+#endif /* _I386_PGALLOC_H */
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/include/mach-xen/asm/pgtable-3level-defs.h 2007-06-12 13:14:02.000000000 +0200
@@ -0,0 +1,24 @@
+#ifndef _I386_PGTABLE_3LEVEL_DEFS_H
+#define _I386_PGTABLE_3LEVEL_DEFS_H
+
+#define HAVE_SHARED_KERNEL_PMD 0
+
+/*
+ * PGDIR_SHIFT determines what a top-level page table entry can map
+ */
+#define PGDIR_SHIFT 30
+#define PTRS_PER_PGD 4
+
+/*
+ * PMD_SHIFT determines the size of the area a middle-level
+ * page table can map
+ */
+#define PMD_SHIFT 21
+#define PTRS_PER_PMD 512
+
+/*
+ * entries per page directory level
+ */
+#define PTRS_PER_PTE 512
+
+#endif /* _I386_PGTABLE_3LEVEL_DEFS_H */
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/include/mach-xen/asm/pgtable-3level.h 2008-04-02 12:34:02.000000000 +0200
@@ -0,0 +1,211 @@
+#ifndef _I386_PGTABLE_3LEVEL_H
+#define _I386_PGTABLE_3LEVEL_H
+
+#include <asm-generic/pgtable-nopud.h>
+
+/*
+ * Intel Physical Address Extension (PAE) Mode - three-level page
+ * tables on PPro+ CPUs.
+ *
+ * Copyright (C) 1999 Ingo Molnar <mingo@redhat.com>
+ */
+
+#define pte_ERROR(e) \
+ printk("%s:%d: bad pte %p(%016Lx pfn %08lx).\n", __FILE__, __LINE__, \
+ &(e), __pte_val(e), pte_pfn(e))
+#define pmd_ERROR(e) \
+ printk("%s:%d: bad pmd %p(%016Lx pfn %08Lx).\n", __FILE__, __LINE__, \
+ &(e), __pmd_val(e), (pmd_val(e) & PTE_MASK) >> PAGE_SHIFT)
+#define pgd_ERROR(e) \
+ printk("%s:%d: bad pgd %p(%016Lx pfn %08Lx).\n", __FILE__, __LINE__, \
+ &(e), __pgd_val(e), (pgd_val(e) & PTE_MASK) >> PAGE_SHIFT)
+
+#define pud_none(pud) 0
+#define pud_bad(pud) 0
+#define pud_present(pud) 1
+
+/*
+ * Is the pte executable?
+ */
+static inline int pte_x(pte_t pte)
+{
+ return !(__pte_val(pte) & _PAGE_NX);
+}
+
+/*
+ * All present user-pages with !NX bit are user-executable:
+ */
+static inline int pte_exec(pte_t pte)
+{
+ return pte_user(pte) && pte_x(pte);
+}
+/*
+ * All present pages with !NX bit are kernel-executable:
+ */
+static inline int pte_exec_kernel(pte_t pte)
+{
+ return pte_x(pte);
+}
+
+/* Rules for using set_pte: the pte being assigned *must* be
+ * either not present or in a state where the hardware will
+ * not attempt to update the pte. In places where this is
+ * not possible, use pte_get_and_clear to obtain the old pte
+ * value and then use set_pte to update it. -ben
+ */
+#define __HAVE_ARCH_SET_PTE_ATOMIC
+
+static inline void set_pte(pte_t *ptep, pte_t pte)
+{
+ ptep->pte_high = pte.pte_high;
+ smp_wmb();
+ ptep->pte_low = pte.pte_low;
+}
+#define set_pte_atomic(pteptr,pteval) \
+ set_64bit((unsigned long long *)(pteptr),__pte_val(pteval))
+
+#define set_pte_at(_mm,addr,ptep,pteval) do { \
+ if (((_mm) != current->mm && (_mm) != &init_mm) || \
+ HYPERVISOR_update_va_mapping((addr), (pteval), 0)) \
+ set_pte((ptep), (pteval)); \
+} while (0)
+
+#define set_pte_at_sync(_mm,addr,ptep,pteval) do { \
+ if (((_mm) != current->mm && (_mm) != &init_mm) || \
+ HYPERVISOR_update_va_mapping((addr), (pteval), UVMF_INVLPG)) { \
+ set_pte((ptep), (pteval)); \
+ xen_invlpg((addr)); \
+ } \
+} while (0)
+
+#define set_pmd(pmdptr,pmdval) \
+ xen_l2_entry_update((pmdptr), (pmdval))
+#define set_pud(pudptr,pudval) \
+ xen_l3_entry_update((pudptr), (pudval))
+
+/*
+ * Pentium-II erratum A13: in PAE mode we explicitly have to flush
+ * the TLB via cr3 if the top-level pgd is changed...
+ * We do not let the generic code free and clear pgd entries due to
+ * this erratum.
+ */
+static inline void pud_clear (pud_t * pud) { }
+
+#define pud_page(pud) \
+((struct page *) __va(pud_val(pud) & PAGE_MASK))
+
+#define pud_page_kernel(pud) \
+((unsigned long) __va(pud_val(pud) & PAGE_MASK))
+
+
+/* Find an entry in the second-level page table.. */
+#define pmd_offset(pud, address) ((pmd_t *) pud_page(*(pud)) + \
+ pmd_index(address))
+
+static inline int pte_none(pte_t pte)
+{
+ return !(pte.pte_low | pte.pte_high);
+}
+
+/*
+ * For PTEs and PDEs, we must clear the P-bit first when clearing a page table
+ * entry, so clear the bottom half first and enforce ordering with a compiler
+ * barrier.
+ */
+static inline void pte_clear(struct mm_struct *mm, unsigned long addr, pte_t *ptep)
+{
+ if ((mm != current->mm && mm != &init_mm)
+ || HYPERVISOR_update_va_mapping(addr, __pte(0), 0)) {
+ ptep->pte_low = 0;
+ smp_wmb();
+ ptep->pte_high = 0;
+ }
+}
+
+#define pmd_clear(xp) do { set_pmd(xp, __pmd(0)); } while (0)
+
+static inline pte_t ptep_get_and_clear(struct mm_struct *mm, unsigned long addr, pte_t *ptep)
+{
+ pte_t pte = *ptep;
+ if (!pte_none(pte)) {
+ if ((mm != &init_mm) ||
+ HYPERVISOR_update_va_mapping(addr, __pte(0), 0)) {
+ uint64_t val = __pte_val(pte);
+ if (__cmpxchg64(ptep, val, 0) != val) {
+ /* xchg acts as a barrier before the setting of the high bits */
+ pte.pte_low = xchg(&ptep->pte_low, 0);
+ pte.pte_high = ptep->pte_high;
+ ptep->pte_high = 0;
+ }
+ }
+ }
+ return pte;
+}
+
+#define ptep_clear_flush(vma, addr, ptep) \
+({ \
+ pte_t *__ptep = (ptep); \
+ pte_t __res = *__ptep; \
+ if (!pte_none(__res) && \
+ ((vma)->vm_mm != current->mm || \
+ HYPERVISOR_update_va_mapping(addr, __pte(0), \
+ (unsigned long)(vma)->vm_mm->cpu_vm_mask.bits| \
+ UVMF_INVLPG|UVMF_MULTI))) { \
+ __ptep->pte_low = 0; \
+ smp_wmb(); \
+ __ptep->pte_high = 0; \
+ flush_tlb_page(vma, addr); \
+ } \
+ __res; \
+})
+
+static inline int pte_same(pte_t a, pte_t b)
+{
+ return a.pte_low == b.pte_low && a.pte_high == b.pte_high;
+}
+
+#define pte_page(x) pfn_to_page(pte_pfn(x))
+
+#define __pte_mfn(_pte) (((_pte).pte_low >> PAGE_SHIFT) | \
+ ((_pte).pte_high << (32-PAGE_SHIFT)))
+#define pte_mfn(_pte) ((_pte).pte_low & _PAGE_PRESENT ? \
+ __pte_mfn(_pte) : pfn_to_mfn(__pte_mfn(_pte)))
+#define pte_pfn(_pte) ((_pte).pte_low & _PAGE_IO ? max_mapnr : \
+ (_pte).pte_low & _PAGE_PRESENT ? \
+ mfn_to_local_pfn(__pte_mfn(_pte)) : \
+ __pte_mfn(_pte))
+
+extern unsigned long long __supported_pte_mask;
+
+static inline pte_t pfn_pte(unsigned long page_nr, pgprot_t pgprot)
+{
+ return __pte((((unsigned long long)page_nr << PAGE_SHIFT) |
+ pgprot_val(pgprot)) & __supported_pte_mask);
+}
+
+static inline pmd_t pfn_pmd(unsigned long page_nr, pgprot_t pgprot)
+{
+ return __pmd((((unsigned long long)page_nr << PAGE_SHIFT) |
+ pgprot_val(pgprot)) & __supported_pte_mask);
+}
+
+/*
+ * Bits 0, 6 and 7 are taken in the low part of the pte,
+ * put the 32 bits of offset into the high part.
+ */
+#define pte_to_pgoff(pte) ((pte).pte_high)
+#define pgoff_to_pte(off) ((pte_t) { _PAGE_FILE, (off) })
+#define PTE_FILE_MAX_BITS 32
+
+/* Encode and de-code a swap entry */
+#define __swp_type(x) (((x).val) & 0x1f)
+#define __swp_offset(x) ((x).val >> 5)
+#define __swp_entry(type, offset) ((swp_entry_t){(type) | (offset) << 5})
+#define __pte_to_swp_entry(pte) ((swp_entry_t){ (pte).pte_high })
+#define __swp_entry_to_pte(x) ((pte_t){ 0, (x).val })
+
+#define __pmd_free_tlb(tlb, x) do { } while (0)
+
+void vmalloc_sync_all(void);
+
+#endif /* _I386_PGTABLE_3LEVEL_H */
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/include/mach-xen/asm/pgtable_32.h 2009-03-18 10:39:32.000000000 +0100
@@ -0,0 +1,550 @@
+#ifndef _I386_PGTABLE_H
+#define _I386_PGTABLE_H
+
+#include <asm/hypervisor.h>
+
+/*
+ * The Linux memory management assumes a three-level page table setup. On
+ * the i386, we use that, but "fold" the mid level into the top-level page
+ * table, so that we physically have the same two-level page table as the
+ * i386 mmu expects.
+ *
+ * This file contains the functions and defines necessary to modify and use
+ * the i386 page table tree.
+ */
+#ifndef __ASSEMBLY__
+#include <asm/processor.h>
+#include <asm/fixmap.h>
+#include <linux/threads.h>
+
+#ifndef _I386_BITOPS_H
+#include <asm/bitops.h>
+#endif
+
+#include <linux/slab.h>
+#include <linux/list.h>
+#include <linux/spinlock.h>
+
+/* Is this pagetable pinned? */
+#define PG_pinned PG_arch_1
+
+struct mm_struct;
+struct vm_area_struct;
+
+/*
+ * ZERO_PAGE is a global shared page that is always zero: used
+ * for zero-mapped memory areas etc..
+ */
+#define ZERO_PAGE(vaddr) (virt_to_page(empty_zero_page))
+extern unsigned long empty_zero_page[1024];
+extern pgd_t *swapper_pg_dir;
+extern kmem_cache_t *pgd_cache;
+extern kmem_cache_t *pmd_cache;
+extern spinlock_t pgd_lock;
+extern struct page *pgd_list;
+
+void pmd_ctor(void *, kmem_cache_t *, unsigned long);
+void pgd_ctor(void *, kmem_cache_t *, unsigned long);
+void pgd_dtor(void *, kmem_cache_t *, unsigned long);
+void pgtable_cache_init(void);
+void paging_init(void);
+
+/*
+ * The Linux x86 paging architecture is 'compile-time dual-mode', it
+ * implements both the traditional 2-level x86 page tables and the
+ * newer 3-level PAE-mode page tables.
+ */
+#ifdef CONFIG_X86_PAE
+# include <asm/pgtable-3level-defs.h>
+# define PMD_SIZE (1UL << PMD_SHIFT)
+# define PMD_MASK (~(PMD_SIZE-1))
+#else
+# include <asm/pgtable-2level-defs.h>
+#endif
+
+#define PGDIR_SIZE (1UL << PGDIR_SHIFT)
+#define PGDIR_MASK (~(PGDIR_SIZE-1))
+
+#define USER_PTRS_PER_PGD (TASK_SIZE/PGDIR_SIZE)
+#define FIRST_USER_ADDRESS 0
+
+#define USER_PGD_PTRS (PAGE_OFFSET >> PGDIR_SHIFT)
+#define KERNEL_PGD_PTRS (PTRS_PER_PGD-USER_PGD_PTRS)
+
+#define TWOLEVEL_PGDIR_SHIFT 22
+#define BOOT_USER_PGD_PTRS (__PAGE_OFFSET >> TWOLEVEL_PGDIR_SHIFT)
+#define BOOT_KERNEL_PGD_PTRS (1024-BOOT_USER_PGD_PTRS)
+
+/* Just any arbitrary offset to the start of the vmalloc VM area: the
+ * current 8MB value just means that there will be a 8MB "hole" after the
+ * physical memory until the kernel virtual memory starts. That means that
+ * any out-of-bounds memory accesses will hopefully be caught.
+ * The vmalloc() routines leaves a hole of 4kB between each vmalloced
+ * area for the same reason. ;)
+ */
+#define VMALLOC_OFFSET (8*1024*1024)
+#define VMALLOC_START (((unsigned long) high_memory + vmalloc_earlyreserve + \
+ 2*VMALLOC_OFFSET-1) & ~(VMALLOC_OFFSET-1))
+#ifdef CONFIG_HIGHMEM
+# define VMALLOC_END (PKMAP_BASE-2*PAGE_SIZE)
+#else
+# define VMALLOC_END (FIXADDR_START-2*PAGE_SIZE)
+#endif
+
+/*
+ * _PAGE_PSE set in the page directory entry just means that
+ * the page directory entry points directly to a 4MB-aligned block of
+ * memory.
+ */
+#define _PAGE_BIT_PRESENT 0
+#define _PAGE_BIT_RW 1
+#define _PAGE_BIT_USER 2
+#define _PAGE_BIT_PWT 3
+#define _PAGE_BIT_PCD 4
+#define _PAGE_BIT_ACCESSED 5
+#define _PAGE_BIT_DIRTY 6
+#define _PAGE_BIT_PSE 7 /* 4 MB (or 2MB) page, Pentium+, if present.. */
+#define _PAGE_BIT_GLOBAL 8 /* Global TLB entry PPro+ */
+/*#define _PAGE_BIT_UNUSED1 9*/ /* available for programmer */
+#define _PAGE_BIT_UNUSED2 10
+#define _PAGE_BIT_UNUSED3 11
+#define _PAGE_BIT_NX 63
+
+#define _PAGE_PRESENT 0x001
+#define _PAGE_RW 0x002
+#define _PAGE_USER 0x004
+#define _PAGE_PWT 0x008
+#define _PAGE_PCD 0x010
+#define _PAGE_ACCESSED 0x020
+#define _PAGE_DIRTY 0x040
+#define _PAGE_PSE 0x080 /* 4 MB (or 2MB) page, Pentium+, if present.. */
+#define _PAGE_GLOBAL 0x100 /* Global TLB entry PPro+ */
+/*#define _PAGE_UNUSED1 0x200*/ /* available for programmer */
+#define _PAGE_UNUSED2 0x400
+#define _PAGE_UNUSED3 0x800
+
+/* If _PAGE_PRESENT is clear, we use these: */
+#define _PAGE_FILE 0x040 /* nonlinear file mapping, saved PTE; unset:swap */
+#define _PAGE_PROTNONE 0x080 /* if the user mapped it with PROT_NONE;
+ pte_present gives true */
+#ifdef CONFIG_X86_PAE
+#define _PAGE_NX (1ULL<<_PAGE_BIT_NX)
+#else
+#define _PAGE_NX 0
+#endif
+
+/* Mapped page is I/O or foreign and has no associated page struct. */
+#define _PAGE_IO 0x200
+
+#define _PAGE_TABLE (_PAGE_PRESENT | _PAGE_RW | _PAGE_USER | _PAGE_ACCESSED | _PAGE_DIRTY)
+#define _KERNPG_TABLE (_PAGE_PRESENT | _PAGE_RW | _PAGE_ACCESSED | _PAGE_DIRTY)
+#define _PAGE_CHG_MASK (PTE_MASK | _PAGE_ACCESSED | _PAGE_DIRTY | _PAGE_IO)
+
+#define PAGE_NONE \
+ __pgprot(_PAGE_PROTNONE | _PAGE_ACCESSED)
+#define PAGE_SHARED \
+ __pgprot(_PAGE_PRESENT | _PAGE_RW | _PAGE_USER | _PAGE_ACCESSED)
+
+#define PAGE_SHARED_EXEC \
+ __pgprot(_PAGE_PRESENT | _PAGE_RW | _PAGE_USER | _PAGE_ACCESSED)
+#define PAGE_COPY_NOEXEC \
+ __pgprot(_PAGE_PRESENT | _PAGE_USER | _PAGE_ACCESSED | _PAGE_NX)
+#define PAGE_COPY_EXEC \
+ __pgprot(_PAGE_PRESENT | _PAGE_USER | _PAGE_ACCESSED)
+#define PAGE_COPY \
+ PAGE_COPY_NOEXEC
+#define PAGE_READONLY \
+ __pgprot(_PAGE_PRESENT | _PAGE_USER | _PAGE_ACCESSED | _PAGE_NX)
+#define PAGE_READONLY_EXEC \
+ __pgprot(_PAGE_PRESENT | _PAGE_USER | _PAGE_ACCESSED)
+
+#define _PAGE_KERNEL \
+ (_PAGE_PRESENT | _PAGE_RW | _PAGE_DIRTY | _PAGE_ACCESSED | _PAGE_NX)
+#define _PAGE_KERNEL_EXEC \
+ (_PAGE_PRESENT | _PAGE_RW | _PAGE_DIRTY | _PAGE_ACCESSED)
+
+extern unsigned long long __PAGE_KERNEL, __PAGE_KERNEL_EXEC;
+#define __PAGE_KERNEL_RO (__PAGE_KERNEL & ~_PAGE_RW)
+#define __PAGE_KERNEL_NOCACHE (__PAGE_KERNEL | _PAGE_PCD)
+#define __PAGE_KERNEL_LARGE (__PAGE_KERNEL | _PAGE_PSE)
+#define __PAGE_KERNEL_LARGE_EXEC (__PAGE_KERNEL_EXEC | _PAGE_PSE)
+
+#define PAGE_KERNEL __pgprot(__PAGE_KERNEL)
+#define PAGE_KERNEL_RO __pgprot(__PAGE_KERNEL_RO)
+#define PAGE_KERNEL_EXEC __pgprot(__PAGE_KERNEL_EXEC)
+#define PAGE_KERNEL_NOCACHE __pgprot(__PAGE_KERNEL_NOCACHE)
+#define PAGE_KERNEL_LARGE __pgprot(__PAGE_KERNEL_LARGE)
+#define PAGE_KERNEL_LARGE_EXEC __pgprot(__PAGE_KERNEL_LARGE_EXEC)
+
+/*
+ * The i386 can't do page protection for execute, and considers that
+ * the same are read. Also, write permissions imply read permissions.
+ * This is the closest we can get..
+ */
+#define __P000 PAGE_NONE
+#define __P001 PAGE_READONLY
+#define __P010 PAGE_COPY
+#define __P011 PAGE_COPY
+#define __P100 PAGE_READONLY_EXEC
+#define __P101 PAGE_READONLY_EXEC
+#define __P110 PAGE_COPY_EXEC
+#define __P111 PAGE_COPY_EXEC
+
+#define __S000 PAGE_NONE
+#define __S001 PAGE_READONLY
+#define __S010 PAGE_SHARED
+#define __S011 PAGE_SHARED
+#define __S100 PAGE_READONLY_EXEC
+#define __S101 PAGE_READONLY_EXEC
+#define __S110 PAGE_SHARED_EXEC
+#define __S111 PAGE_SHARED_EXEC
+
+/*
+ * Define this if things work differently on an i386 and an i486:
+ * it will (on an i486) warn about kernel memory accesses that are
+ * done without a 'access_ok(VERIFY_WRITE,..)'
+ */
+#undef TEST_ACCESS_OK
+
+/* The boot page tables (all created as a single array) */
+extern unsigned long pg0[];
+
+#define pte_present(x) ((x).pte_low & (_PAGE_PRESENT | _PAGE_PROTNONE))
+
+/* To avoid harmful races, pmd_none(x) should check only the lower when PAE */
+#define pmd_none(x) (!(unsigned long)__pmd_val(x))
+#if CONFIG_XEN_COMPAT <= 0x030002
+/* pmd_present doesn't just test the _PAGE_PRESENT bit since wr.p.t.
+ can temporarily clear it. */
+#define pmd_present(x) (__pmd_val(x))
+#define pmd_bad(x) ((__pmd_val(x) & (~PAGE_MASK & ~_PAGE_USER & ~_PAGE_PRESENT)) != (_KERNPG_TABLE & ~_PAGE_PRESENT))
+#else
+#define pmd_present(x) (__pmd_val(x) & _PAGE_PRESENT)
+#define pmd_bad(x) ((__pmd_val(x) & (~PAGE_MASK & ~_PAGE_USER)) != _KERNPG_TABLE)
+#endif
+
+
+#define pages_to_mb(x) ((x) >> (20-PAGE_SHIFT))
+
+/*
+ * The following only work if pte_present() is true.
+ * Undefined behaviour if not..
+ */
+static inline int pte_user(pte_t pte) { return (pte).pte_low & _PAGE_USER; }
+static inline int pte_read(pte_t pte) { return (pte).pte_low & _PAGE_USER; }
+static inline int pte_dirty(pte_t pte) { return (pte).pte_low & _PAGE_DIRTY; }
+static inline int pte_young(pte_t pte) { return (pte).pte_low & _PAGE_ACCESSED; }
+static inline int pte_write(pte_t pte) { return (pte).pte_low & _PAGE_RW; }
+static inline int pte_huge(pte_t pte) { return (pte).pte_low & _PAGE_PSE; }
+
+/*
+ * The following only works if pte_present() is not true.
+ */
+static inline int pte_file(pte_t pte) { return (pte).pte_low & _PAGE_FILE; }
+
+static inline pte_t pte_rdprotect(pte_t pte) { (pte).pte_low &= ~_PAGE_USER; return pte; }
+static inline pte_t pte_exprotect(pte_t pte) { (pte).pte_low &= ~_PAGE_USER; return pte; }
+static inline pte_t pte_mkclean(pte_t pte) { (pte).pte_low &= ~_PAGE_DIRTY; return pte; }
+static inline pte_t pte_mkold(pte_t pte) { (pte).pte_low &= ~_PAGE_ACCESSED; return pte; }
+static inline pte_t pte_wrprotect(pte_t pte) { (pte).pte_low &= ~_PAGE_RW; return pte; }
+static inline pte_t pte_mkread(pte_t pte) { (pte).pte_low |= _PAGE_USER; return pte; }
+static inline pte_t pte_mkexec(pte_t pte) { (pte).pte_low |= _PAGE_USER; return pte; }
+static inline pte_t pte_mkdirty(pte_t pte) { (pte).pte_low |= _PAGE_DIRTY; return pte; }
+static inline pte_t pte_mkyoung(pte_t pte) { (pte).pte_low |= _PAGE_ACCESSED; return pte; }
+static inline pte_t pte_mkwrite(pte_t pte) { (pte).pte_low |= _PAGE_RW; return pte; }
+static inline pte_t pte_mkhuge(pte_t pte) { (pte).pte_low |= _PAGE_PSE; return pte; }
+
+#ifdef CONFIG_X86_PAE
+# include <asm/pgtable-3level.h>
+#else
+# include <asm/pgtable-2level.h>
+#endif
+
+#define ptep_test_and_clear_dirty(vma, addr, ptep) \
+({ \
+ pte_t __pte = *(ptep); \
+ int __ret = pte_dirty(__pte); \
+ if (__ret) { \
+ __pte = pte_mkclean(__pte); \
+ if ((vma)->vm_mm != current->mm || \
+ HYPERVISOR_update_va_mapping(addr, __pte, 0)) \
+ (ptep)->pte_low = __pte.pte_low; \
+ } \
+ __ret; \
+})
+
+#define ptep_test_and_clear_young(vma, addr, ptep) \
+({ \
+ pte_t __pte = *(ptep); \
+ int __ret = pte_young(__pte); \
+ if (__ret) \
+ __pte = pte_mkold(__pte); \
+ if ((vma)->vm_mm != current->mm || \
+ HYPERVISOR_update_va_mapping(addr, __pte, 0)) \
+ (ptep)->pte_low = __pte.pte_low; \
+ __ret; \
+})
+
+#define ptep_get_and_clear_full(mm, addr, ptep, full) \
+ ((full) ? ({ \
+ pte_t __res = *(ptep); \
+ if (test_bit(PG_pinned, &virt_to_page((mm)->pgd)->flags)) \
+ xen_l1_entry_update(ptep, __pte(0)); \
+ else \
+ *(ptep) = __pte(0); \
+ __res; \
+ }) : \
+ ptep_get_and_clear(mm, addr, ptep))
+
+static inline void ptep_set_wrprotect(struct mm_struct *mm, unsigned long addr, pte_t *ptep)
+{
+ pte_t pte = *ptep;
+ if (pte_write(pte))
+ set_pte_at(mm, addr, ptep, pte_wrprotect(pte));
+}
+
+/*
+ * clone_pgd_range(pgd_t *dst, pgd_t *src, int count);
+ *
+ * dst - pointer to pgd range anwhere on a pgd page
+ * src - ""
+ * count - the number of pgds to copy.
+ *
+ * dst and src can be on the same page, but the range must not overlap,
+ * and must not cross a page boundary.
+ */
+static inline void clone_pgd_range(pgd_t *dst, pgd_t *src, int count)
+{
+ memcpy(dst, src, count * sizeof(pgd_t));
+}
+
+/*
+ * Macro to mark a page protection value as "uncacheable". On processors which do not support
+ * it, this is a no-op.
+ */
+#define pgprot_noncached(prot) ((boot_cpu_data.x86 > 3) \
+ ? (__pgprot(pgprot_val(prot) | _PAGE_PCD | _PAGE_PWT)) : (prot))
+
+/*
+ * Conversion functions: convert a page and protection to a page entry,
+ * and a page entry and page directory to the page they refer to.
+ */
+
+#define mk_pte(page, pgprot) pfn_pte(page_to_pfn(page), (pgprot))
+
+static inline pte_t pte_modify(pte_t pte, pgprot_t newprot)
+{
+ /*
+ * Since this might change the present bit (which controls whether
+ * a pte_t object has undergone p2m translation), we must use
+ * pte_val() on the input pte and __pte() for the return value.
+ */
+ paddr_t pteval = pte_val(pte);
+
+ pteval &= _PAGE_CHG_MASK;
+ pteval |= pgprot_val(newprot);
+#ifdef CONFIG_X86_PAE
+ pteval &= __supported_pte_mask;
+#endif
+ return __pte(pteval);
+}
+
+#define pmd_large(pmd) \
+((__pmd_val(pmd) & (_PAGE_PSE|_PAGE_PRESENT)) == (_PAGE_PSE|_PAGE_PRESENT))
+
+/*
+ * the pgd page can be thought of an array like this: pgd_t[PTRS_PER_PGD]
+ *
+ * this macro returns the index of the entry in the pgd page which would
+ * control the given virtual address
+ */
+#define pgd_index(address) (((address) >> PGDIR_SHIFT) & (PTRS_PER_PGD-1))
+#define pgd_index_k(addr) pgd_index(addr)
+
+/*
+ * pgd_offset() returns a (pgd_t *)
+ * pgd_index() is used get the offset into the pgd page's array of pgd_t's;
+ */
+#define pgd_offset(mm, address) ((mm)->pgd+pgd_index(address))
+
+/*
+ * a shortcut which implies the use of the kernel's pgd, instead
+ * of a process's
+ */
+#define pgd_offset_k(address) pgd_offset(&init_mm, address)
+
+/*
+ * the pmd page can be thought of an array like this: pmd_t[PTRS_PER_PMD]
+ *
+ * this macro returns the index of the entry in the pmd page which would
+ * control the given virtual address
+ */
+#define pmd_index(address) \
+ (((address) >> PMD_SHIFT) & (PTRS_PER_PMD-1))
+
+/*
+ * the pte page can be thought of an array like this: pte_t[PTRS_PER_PTE]
+ *
+ * this macro returns the index of the entry in the pte page which would
+ * control the given virtual address
+ */
+#define pte_index(address) \
+ (((address) >> PAGE_SHIFT) & (PTRS_PER_PTE - 1))
+#define pte_offset_kernel(dir, address) \
+ ((pte_t *) pmd_page_kernel(*(dir)) + pte_index(address))
+
+#define pmd_page(pmd) (pfn_to_page(pmd_val(pmd) >> PAGE_SHIFT))
+
+#define pmd_page_kernel(pmd) \
+ ((unsigned long) __va(pmd_val(pmd) & PAGE_MASK))
+
+/*
+ * Helper function that returns the kernel pagetable entry controlling
+ * the virtual address 'address'. NULL means no pagetable entry present.
+ * NOTE: the return type is pte_t but if the pmd is PSE then we return it
+ * as a pte too.
+ */
+extern pte_t *lookup_address(unsigned long address);
+
+/*
+ * Make a given kernel text page executable/non-executable.
+ * Returns the previous executability setting of that page (which
+ * is used to restore the previous state). Used by the SMP bootup code.
+ * NOTE: this is an __init function for security reasons.
+ */
+#ifdef CONFIG_X86_PAE
+ extern int set_kernel_exec(unsigned long vaddr, int enable);
+#else
+ static inline int set_kernel_exec(unsigned long vaddr, int enable) { return 0;}
+#endif
+
+extern void noexec_setup(const char *str);
+
+#if defined(CONFIG_HIGHPTE)
+#define pte_offset_map(dir, address) \
+ ((pte_t *)kmap_atomic_pte(pmd_page(*(dir)),KM_PTE0) + \
+ pte_index(address))
+#define pte_offset_map_nested(dir, address) \
+ ((pte_t *)kmap_atomic_pte(pmd_page(*(dir)),KM_PTE1) + \
+ pte_index(address))
+#define pte_unmap(pte) kunmap_atomic(pte, KM_PTE0)
+#define pte_unmap_nested(pte) kunmap_atomic(pte, KM_PTE1)
+#else
+#define pte_offset_map(dir, address) \
+ ((pte_t *)page_address(pmd_page(*(dir))) + pte_index(address))
+#define pte_offset_map_nested(dir, address) pte_offset_map(dir, address)
+#define pte_unmap(pte) do { } while (0)
+#define pte_unmap_nested(pte) do { } while (0)
+#endif
+
+#define __HAVE_ARCH_PTEP_ESTABLISH
+#define ptep_establish(vma, address, ptep, pteval) \
+ do { \
+ if ( likely((vma)->vm_mm == current->mm) ) { \
+ BUG_ON(HYPERVISOR_update_va_mapping(address, \
+ pteval, \
+ (unsigned long)(vma)->vm_mm->cpu_vm_mask.bits| \
+ UVMF_INVLPG|UVMF_MULTI)); \
+ } else { \
+ xen_l1_entry_update(ptep, pteval); \
+ flush_tlb_page(vma, address); \
+ } \
+ } while (0)
+
+/*
+ * The i386 doesn't have any external MMU info: the kernel page
+ * tables contain all the necessary information.
+ *
+ * Also, we only update the dirty/accessed state if we set
+ * the dirty bit by hand in the kernel, since the hardware
+ * will do the accessed bit for us, and we don't want to
+ * race with other CPU's that might be updating the dirty
+ * bit at the same time.
+ */
+#define update_mmu_cache(vma,address,pte) do { } while (0)
+#define __HAVE_ARCH_PTEP_SET_ACCESS_FLAGS
+#define ptep_set_access_flags(vma, address, ptep, entry, dirty) \
+ do { \
+ if (dirty) \
+ ptep_establish(vma, address, ptep, entry); \
+ } while (0)
+
+#include <xen/features.h>
+void make_lowmem_page_readonly(void *va, unsigned int feature);
+void make_lowmem_page_writable(void *va, unsigned int feature);
+void make_page_readonly(void *va, unsigned int feature);
+void make_page_writable(void *va, unsigned int feature);
+void make_pages_readonly(void *va, unsigned int nr, unsigned int feature);
+void make_pages_writable(void *va, unsigned int nr, unsigned int feature);
+
+#define virt_to_ptep(va) \
+({ \
+ pte_t *__ptep = lookup_address((unsigned long)(va)); \
+ BUG_ON(!__ptep || !pte_present(*__ptep)); \
+ __ptep; \
+})
+
+#define arbitrary_virt_to_machine(va) \
+ (((maddr_t)pte_mfn(*virt_to_ptep(va)) << PAGE_SHIFT) \
+ | ((unsigned long)(va) & (PAGE_SIZE - 1)))
+
+#ifdef CONFIG_HIGHPTE
+#include <asm/io.h>
+struct page *kmap_atomic_to_page(void *);
+#define ptep_to_machine(ptep) \
+({ \
+ pte_t *__ptep = (ptep); \
+ page_to_phys(kmap_atomic_to_page(__ptep)) \
+ | ((unsigned long)__ptep & (PAGE_SIZE - 1)); \
+})
+#else
+#define ptep_to_machine(ptep) virt_to_machine(ptep)
+#endif
+
+#endif /* !__ASSEMBLY__ */
+
+#ifdef CONFIG_FLATMEM
+#define kern_addr_valid(addr) (1)
+#endif /* CONFIG_FLATMEM */
+
+int direct_remap_pfn_range(struct vm_area_struct *vma,
+ unsigned long address,
+ unsigned long mfn,
+ unsigned long size,
+ pgprot_t prot,
+ domid_t domid);
+int direct_kernel_remap_pfn_range(unsigned long address,
+ unsigned long mfn,
+ unsigned long size,
+ pgprot_t prot,
+ domid_t domid);
+int create_lookup_pte_addr(struct mm_struct *mm,
+ unsigned long address,
+ uint64_t *ptep);
+int touch_pte_range(struct mm_struct *mm,
+ unsigned long address,
+ unsigned long size);
+
+int xen_change_pte_range(struct mm_struct *mm, pmd_t *pmd,
+ unsigned long addr, unsigned long end, pgprot_t newprot);
+
+#define arch_change_pte_range(mm, pmd, addr, end, newprot) \
+ xen_change_pte_range(mm, pmd, addr, end, newprot)
+
+#define io_remap_pfn_range(vma,from,pfn,size,prot) \
+direct_remap_pfn_range(vma,from,pfn,size,prot,DOMID_IO)
+
+#define MK_IOSPACE_PFN(space, pfn) (pfn)
+#define GET_IOSPACE(pfn) 0
+#define GET_PFN(pfn) (pfn)
+
+#define __HAVE_ARCH_PTEP_TEST_AND_CLEAR_YOUNG
+#define __HAVE_ARCH_PTEP_TEST_AND_CLEAR_DIRTY
+#define __HAVE_ARCH_PTEP_GET_AND_CLEAR
+#define __HAVE_ARCH_PTEP_GET_AND_CLEAR_FULL
+#define __HAVE_ARCH_PTEP_CLEAR_FLUSH
+#define __HAVE_ARCH_PTEP_SET_WRPROTECT
+#define __HAVE_ARCH_PTE_SAME
+#include <asm-generic/pgtable.h>
+
+#endif /* _I386_PGTABLE_H */
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/include/mach-xen/asm/processor_32.h 2008-01-28 12:24:19.000000000 +0100
@@ -0,0 +1,743 @@
+/*
+ * include/asm-i386/processor.h
+ *
+ * Copyright (C) 1994 Linus Torvalds
+ */
+
+#ifndef __ASM_I386_PROCESSOR_H
+#define __ASM_I386_PROCESSOR_H
+
+#include <asm/vm86.h>
+#include <asm/math_emu.h>
+#include <asm/segment.h>
+#include <asm/page.h>
+#include <asm/types.h>
+#include <asm/sigcontext.h>
+#include <asm/cpufeature.h>
+#include <asm/msr.h>
+#include <asm/system.h>
+#include <linux/cache.h>
+#include <linux/threads.h>
+#include <asm/percpu.h>
+#include <linux/cpumask.h>
+#include <xen/interface/physdev.h>
+
+/* flag for disabling the tsc */
+extern int tsc_disable;
+
+struct desc_struct {
+ unsigned long a,b;
+};
+
+#define desc_empty(desc) \
+ (!((desc)->a | (desc)->b))
+
+#define desc_equal(desc1, desc2) \
+ (((desc1)->a == (desc2)->a) && ((desc1)->b == (desc2)->b))
+/*
+ * Default implementation of macro that returns current
+ * instruction pointer ("program counter").
+ */
+#define current_text_addr() ({ void *pc; __asm__("movl $1f,%0\n1:":"=g" (pc)); pc; })
+
+/*
+ * CPU type and hardware bug flags. Kept separately for each CPU.
+ * Members of this structure are referenced in head.S, so think twice
+ * before touching them. [mj]
+ */
+
+struct cpuinfo_x86 {
+ __u8 x86; /* CPU family */
+ __u8 x86_vendor; /* CPU vendor */
+ __u8 x86_model;
+ __u8 x86_mask;
+ char wp_works_ok; /* It doesn't on 386's */
+ char hlt_works_ok; /* Problems on some 486Dx4's and old 386's */
+ char hard_math;
+ char rfu;
+ int cpuid_level; /* Maximum supported CPUID level, -1=no CPUID */
+ unsigned long x86_capability[NCAPINTS];
+ char x86_vendor_id[16];
+ char x86_model_id[64];
+ int x86_cache_size; /* in KB - valid for CPUS which support this
+ call */
+ int x86_cache_alignment; /* In bytes */
+ char fdiv_bug;
+ char f00f_bug;
+ char coma_bug;
+ char pad0;
+ int x86_power;
+ unsigned long loops_per_jiffy;
+#ifdef CONFIG_SMP
+ cpumask_t llc_shared_map; /* cpus sharing the last level cache */
+#endif
+ unsigned char x86_max_cores; /* cpuid returned max cores value */
+ unsigned char apicid;
+#ifdef CONFIG_SMP
+ unsigned char booted_cores; /* number of cores as seen by OS */
+ __u8 phys_proc_id; /* Physical processor id. */
+ __u8 cpu_core_id; /* Core id */
+#endif
+} __attribute__((__aligned__(SMP_CACHE_BYTES)));
+
+#define X86_VENDOR_INTEL 0
+#define X86_VENDOR_CYRIX 1
+#define X86_VENDOR_AMD 2
+#define X86_VENDOR_UMC 3
+#define X86_VENDOR_NEXGEN 4
+#define X86_VENDOR_CENTAUR 5
+#define X86_VENDOR_RISE 6
+#define X86_VENDOR_TRANSMETA 7
+#define X86_VENDOR_NSC 8
+#define X86_VENDOR_NUM 9
+#define X86_VENDOR_UNKNOWN 0xff
+
+/*
+ * capabilities of CPUs
+ */
+
+extern struct cpuinfo_x86 boot_cpu_data;
+extern struct cpuinfo_x86 new_cpu_data;
+#ifndef CONFIG_X86_NO_TSS
+extern struct tss_struct doublefault_tss;
+DECLARE_PER_CPU(struct tss_struct, init_tss);
+#endif
+
+#ifdef CONFIG_SMP
+extern struct cpuinfo_x86 cpu_data[];
+#define current_cpu_data cpu_data[smp_processor_id()]
+#else
+#define cpu_data (&boot_cpu_data)
+#define current_cpu_data boot_cpu_data
+#endif
+
+extern int cpu_llc_id[NR_CPUS];
+extern char ignore_fpu_irq;
+
+extern void identify_cpu(struct cpuinfo_x86 *);
+extern void print_cpu_info(struct cpuinfo_x86 *);
+extern unsigned int init_intel_cacheinfo(struct cpuinfo_x86 *c);
+extern unsigned short num_cache_leaves;
+
+#ifdef CONFIG_X86_HT
+extern void detect_ht(struct cpuinfo_x86 *c);
+#else
+static inline void detect_ht(struct cpuinfo_x86 *c) {}
+#endif
+
+/*
+ * EFLAGS bits
+ */
+#define X86_EFLAGS_CF 0x00000001 /* Carry Flag */
+#define X86_EFLAGS_PF 0x00000004 /* Parity Flag */
+#define X86_EFLAGS_AF 0x00000010 /* Auxillary carry Flag */
+#define X86_EFLAGS_ZF 0x00000040 /* Zero Flag */
+#define X86_EFLAGS_SF 0x00000080 /* Sign Flag */
+#define X86_EFLAGS_TF 0x00000100 /* Trap Flag */
+#define X86_EFLAGS_IF 0x00000200 /* Interrupt Flag */
+#define X86_EFLAGS_DF 0x00000400 /* Direction Flag */
+#define X86_EFLAGS_OF 0x00000800 /* Overflow Flag */
+#define X86_EFLAGS_IOPL 0x00003000 /* IOPL mask */
+#define X86_EFLAGS_NT 0x00004000 /* Nested Task */
+#define X86_EFLAGS_RF 0x00010000 /* Resume Flag */
+#define X86_EFLAGS_VM 0x00020000 /* Virtual Mode */
+#define X86_EFLAGS_AC 0x00040000 /* Alignment Check */
+#define X86_EFLAGS_VIF 0x00080000 /* Virtual Interrupt Flag */
+#define X86_EFLAGS_VIP 0x00100000 /* Virtual Interrupt Pending */
+#define X86_EFLAGS_ID 0x00200000 /* CPUID detection flag */
+
+/*
+ * Generic CPUID function
+ * clear %ecx since some cpus (Cyrix MII) do not set or clear %ecx
+ * resulting in stale register contents being returned.
+ */
+static inline void cpuid(unsigned int op, unsigned int *eax, unsigned int *ebx, unsigned int *ecx, unsigned int *edx)
+{
+ __asm__(XEN_CPUID
+ : "=a" (*eax),
+ "=b" (*ebx),
+ "=c" (*ecx),
+ "=d" (*edx)
+ : "0" (op), "c"(0));
+}
+
+/* Some CPUID calls want 'count' to be placed in ecx */
+static inline void cpuid_count(int op, int count, int *eax, int *ebx, int *ecx,
+ int *edx)
+{
+ __asm__(XEN_CPUID
+ : "=a" (*eax),
+ "=b" (*ebx),
+ "=c" (*ecx),
+ "=d" (*edx)
+ : "0" (op), "c" (count));
+}
+
+/*
+ * CPUID functions returning a single datum
+ */
+static inline unsigned int cpuid_eax(unsigned int op)
+{
+ unsigned int eax;
+
+ __asm__(XEN_CPUID
+ : "=a" (eax)
+ : "0" (op)
+ : "bx", "cx", "dx");
+ return eax;
+}
+static inline unsigned int cpuid_ebx(unsigned int op)
+{
+ unsigned int eax, ebx;
+
+ __asm__(XEN_CPUID
+ : "=a" (eax), "=b" (ebx)
+ : "0" (op)
+ : "cx", "dx" );
+ return ebx;
+}
+static inline unsigned int cpuid_ecx(unsigned int op)
+{
+ unsigned int eax, ecx;
+
+ __asm__(XEN_CPUID
+ : "=a" (eax), "=c" (ecx)
+ : "0" (op)
+ : "bx", "dx" );
+ return ecx;
+}
+static inline unsigned int cpuid_edx(unsigned int op)
+{
+ unsigned int eax, edx;
+
+ __asm__(XEN_CPUID
+ : "=a" (eax), "=d" (edx)
+ : "0" (op)
+ : "bx", "cx");
+ return edx;
+}
+
+#define load_cr3(pgdir) write_cr3(__pa(pgdir))
+
+/*
+ * Intel CPU features in CR4
+ */
+#define X86_CR4_VME 0x0001 /* enable vm86 extensions */
+#define X86_CR4_PVI 0x0002 /* virtual interrupts flag enable */
+#define X86_CR4_TSD 0x0004 /* disable time stamp at ipl 3 */
+#define X86_CR4_DE 0x0008 /* enable debugging extensions */
+#define X86_CR4_PSE 0x0010 /* enable page size extensions */
+#define X86_CR4_PAE 0x0020 /* enable physical address extensions */
+#define X86_CR4_MCE 0x0040 /* Machine check enable */
+#define X86_CR4_PGE 0x0080 /* enable global pages */
+#define X86_CR4_PCE 0x0100 /* enable performance counters at ipl 3 */
+#define X86_CR4_OSFXSR 0x0200 /* enable fast FPU save and restore */
+#define X86_CR4_OSXMMEXCPT 0x0400 /* enable unmasked SSE exceptions */
+
+/*
+ * Save the cr4 feature set we're using (ie
+ * Pentium 4MB enable and PPro Global page
+ * enable), so that any CPU's that boot up
+ * after us can get the correct flags.
+ */
+extern unsigned long mmu_cr4_features;
+
+static inline void set_in_cr4 (unsigned long mask)
+{
+ unsigned cr4;
+ mmu_cr4_features |= mask;
+ cr4 = read_cr4();
+ cr4 |= mask;
+ write_cr4(cr4);
+}
+
+static inline void clear_in_cr4 (unsigned long mask)
+{
+ unsigned cr4;
+ mmu_cr4_features &= ~mask;
+ cr4 = read_cr4();
+ cr4 &= ~mask;
+ write_cr4(cr4);
+}
+
+/*
+ * NSC/Cyrix CPU configuration register indexes
+ */
+
+#define CX86_PCR0 0x20
+#define CX86_GCR 0xb8
+#define CX86_CCR0 0xc0
+#define CX86_CCR1 0xc1
+#define CX86_CCR2 0xc2
+#define CX86_CCR3 0xc3
+#define CX86_CCR4 0xe8
+#define CX86_CCR5 0xe9
+#define CX86_CCR6 0xea
+#define CX86_CCR7 0xeb
+#define CX86_PCR1 0xf0
+#define CX86_DIR0 0xfe
+#define CX86_DIR1 0xff
+#define CX86_ARR_BASE 0xc4
+#define CX86_RCR_BASE 0xdc
+
+/*
+ * NSC/Cyrix CPU indexed register access macros
+ */
+
+#define getCx86(reg) ({ outb((reg), 0x22); inb(0x23); })
+
+#define setCx86(reg, data) do { \
+ outb((reg), 0x22); \
+ outb((data), 0x23); \
+} while (0)
+
+/* Stop speculative execution */
+static inline void sync_core(void)
+{
+ int tmp;
+ asm volatile("cpuid" : "=a" (tmp) : "0" (1) : "ebx","ecx","edx","memory");
+}
+
+static inline void __monitor(const void *eax, unsigned long ecx,
+ unsigned long edx)
+{
+ /* "monitor %eax,%ecx,%edx;" */
+ asm volatile(
+ ".byte 0x0f,0x01,0xc8;"
+ : :"a" (eax), "c" (ecx), "d"(edx));
+}
+
+static inline void __mwait(unsigned long eax, unsigned long ecx)
+{
+ /* "mwait %eax,%ecx;" */
+ asm volatile(
+ ".byte 0x0f,0x01,0xc9;"
+ : :"a" (eax), "c" (ecx));
+}
+
+/* from system description table in BIOS. Mostly for MCA use, but
+others may find it useful. */
+extern unsigned int machine_id;
+extern unsigned int machine_submodel_id;
+extern unsigned int BIOS_revision;
+extern unsigned int mca_pentium_flag;
+
+/* Boot loader type from the setup header */
+extern int bootloader_type;
+
+/*
+ * User space process size: 3GB (default).
+ */
+#define TASK_SIZE (PAGE_OFFSET)
+
+/* This decides where the kernel will search for a free chunk of vm
+ * space during mmap's.
+ */
+#define TASK_UNMAPPED_BASE (PAGE_ALIGN(TASK_SIZE / 3))
+
+#define HAVE_ARCH_PICK_MMAP_LAYOUT
+
+/*
+ * Size of io_bitmap.
+ */
+#define IO_BITMAP_BITS 65536
+#define IO_BITMAP_BYTES (IO_BITMAP_BITS/8)
+#define IO_BITMAP_LONGS (IO_BITMAP_BYTES/sizeof(long))
+#ifndef CONFIG_X86_NO_TSS
+#define IO_BITMAP_OFFSET offsetof(struct tss_struct,io_bitmap)
+#endif
+#define INVALID_IO_BITMAP_OFFSET 0x8000
+#define INVALID_IO_BITMAP_OFFSET_LAZY 0x9000
+
+struct i387_fsave_struct {
+ long cwd;
+ long swd;
+ long twd;
+ long fip;
+ long fcs;
+ long foo;
+ long fos;
+ long st_space[20]; /* 8*10 bytes for each FP-reg = 80 bytes */
+ long status; /* software status information */
+};
+
+struct i387_fxsave_struct {
+ unsigned short cwd;
+ unsigned short swd;
+ unsigned short twd;
+ unsigned short fop;
+ long fip;
+ long fcs;
+ long foo;
+ long fos;
+ long mxcsr;
+ long mxcsr_mask;
+ long st_space[32]; /* 8*16 bytes for each FP-reg = 128 bytes */
+ long xmm_space[32]; /* 8*16 bytes for each XMM-reg = 128 bytes */
+ long padding[56];
+} __attribute__ ((aligned (16)));
+
+struct i387_soft_struct {
+ long cwd;
+ long swd;
+ long twd;
+ long fip;
+ long fcs;
+ long foo;
+ long fos;
+ long st_space[20]; /* 8*10 bytes for each FP-reg = 80 bytes */
+ unsigned char ftop, changed, lookahead, no_update, rm, alimit;
+ struct info *info;
+ unsigned long entry_eip;
+};
+
+union i387_union {
+ struct i387_fsave_struct fsave;
+ struct i387_fxsave_struct fxsave;
+ struct i387_soft_struct soft;
+};
+
+typedef struct {
+ unsigned long seg;
+} mm_segment_t;
+
+struct thread_struct;
+
+#ifndef CONFIG_X86_NO_TSS
+struct tss_struct {
+ unsigned short back_link,__blh;
+ unsigned long esp0;
+ unsigned short ss0,__ss0h;
+ unsigned long esp1;
+ unsigned short ss1,__ss1h; /* ss1 is used to cache MSR_IA32_SYSENTER_CS */
+ unsigned long esp2;
+ unsigned short ss2,__ss2h;
+ unsigned long __cr3;
+ unsigned long eip;
+ unsigned long eflags;
+ unsigned long eax,ecx,edx,ebx;
+ unsigned long esp;
+ unsigned long ebp;
+ unsigned long esi;
+ unsigned long edi;
+ unsigned short es, __esh;
+ unsigned short cs, __csh;
+ unsigned short ss, __ssh;
+ unsigned short ds, __dsh;
+ unsigned short fs, __fsh;
+ unsigned short gs, __gsh;
+ unsigned short ldt, __ldth;
+ unsigned short trace, io_bitmap_base;
+ /*
+ * The extra 1 is there because the CPU will access an
+ * additional byte beyond the end of the IO permission
+ * bitmap. The extra byte must be all 1 bits, and must
+ * be within the limit.
+ */
+ unsigned long io_bitmap[IO_BITMAP_LONGS + 1];
+ /*
+ * Cache the current maximum and the last task that used the bitmap:
+ */
+ unsigned long io_bitmap_max;
+ struct thread_struct *io_bitmap_owner;
+ /*
+ * pads the TSS to be cacheline-aligned (size is 0x100)
+ */
+ unsigned long __cacheline_filler[35];
+ /*
+ * .. and then another 0x100 bytes for emergency kernel stack
+ */
+ unsigned long stack[64];
+} __attribute__((packed));
+#endif
+
+#define ARCH_MIN_TASKALIGN 16
+
+struct thread_struct {
+/* cached TLS descriptors. */
+ struct desc_struct tls_array[GDT_ENTRY_TLS_ENTRIES];
+ unsigned long esp0;
+ unsigned long sysenter_cs;
+ unsigned long eip;
+ unsigned long esp;
+ unsigned long fs;
+ unsigned long gs;
+/* Hardware debugging registers */
+ unsigned long debugreg[8]; /* %%db0-7 debug registers */
+/* fault info */
+ unsigned long cr2, trap_no, error_code;
+/* floating point info */
+ union i387_union i387;
+/* virtual 86 mode info */
+ struct vm86_struct __user * vm86_info;
+ unsigned long screen_bitmap;
+ unsigned long v86flags, v86mask, saved_esp0;
+ unsigned int saved_fs, saved_gs;
+/* IO permissions */
+ unsigned long *io_bitmap_ptr;
+ unsigned long iopl;
+/* max allowed port in the bitmap, in bytes: */
+ unsigned long io_bitmap_max;
+};
+
+#define INIT_THREAD { \
+ .vm86_info = NULL, \
+ .sysenter_cs = __KERNEL_CS, \
+ .io_bitmap_ptr = NULL, \
+}
+
+#ifndef CONFIG_X86_NO_TSS
+/*
+ * Note that the .io_bitmap member must be extra-big. This is because
+ * the CPU will access an additional byte beyond the end of the IO
+ * permission bitmap. The extra byte must be all 1 bits, and must
+ * be within the limit.
+ */
+#define INIT_TSS { \
+ .esp0 = sizeof(init_stack) + (long)&init_stack, \
+ .ss0 = __KERNEL_DS, \
+ .ss1 = __KERNEL_CS, \
+ .io_bitmap_base = INVALID_IO_BITMAP_OFFSET, \
+ .io_bitmap = { [ 0 ... IO_BITMAP_LONGS] = ~0 }, \
+}
+
+static inline void __load_esp0(struct tss_struct *tss, struct thread_struct *thread)
+{
+ tss->esp0 = thread->esp0;
+ /* This can only happen when SEP is enabled, no need to test "SEP"arately */
+ if (unlikely(tss->ss1 != thread->sysenter_cs)) {
+ tss->ss1 = thread->sysenter_cs;
+ wrmsr(MSR_IA32_SYSENTER_CS, thread->sysenter_cs, 0);
+ }
+}
+#define load_esp0(tss, thread) \
+ __load_esp0(tss, thread)
+#else
+#define load_esp0(tss, thread) do { \
+ if (HYPERVISOR_stack_switch(__KERNEL_DS, (thread)->esp0)) \
+ BUG(); \
+} while (0)
+#endif
+
+#define start_thread(regs, new_eip, new_esp) do { \
+ __asm__("movl %0,%%fs ; movl %0,%%gs": :"r" (0)); \
+ set_fs(USER_DS); \
+ regs->xds = __USER_DS; \
+ regs->xes = __USER_DS; \
+ regs->xss = __USER_DS; \
+ regs->xcs = __USER_CS; \
+ regs->eip = new_eip; \
+ regs->esp = new_esp; \
+} while (0)
+
+/*
+ * These special macros can be used to get or set a debugging register
+ */
+#define get_debugreg(var, register) \
+ (var) = HYPERVISOR_get_debugreg((register))
+#define set_debugreg(value, register) \
+ WARN_ON(HYPERVISOR_set_debugreg((register), (value)))
+
+/*
+ * Set IOPL bits in EFLAGS from given mask
+ */
+static inline void set_iopl_mask(unsigned mask)
+{
+ struct physdev_set_iopl set_iopl;
+
+ /* Force the change at ring 0. */
+ set_iopl.iopl = (mask == 0) ? 1 : (mask >> 12) & 3;
+ WARN_ON(HYPERVISOR_physdev_op(PHYSDEVOP_set_iopl, &set_iopl));
+}
+
+/* Forward declaration, a strange C thing */
+struct task_struct;
+struct mm_struct;
+
+/* Free all resources held by a thread. */
+extern void release_thread(struct task_struct *);
+
+/* Prepare to copy thread state - unlazy all lazy status */
+extern void prepare_to_copy(struct task_struct *tsk);
+
+/*
+ * create a kernel thread without removing it from tasklists
+ */
+extern int kernel_thread(int (*fn)(void *), void * arg, unsigned long flags);
+
+extern unsigned long thread_saved_pc(struct task_struct *tsk);
+void show_trace(struct task_struct *task, struct pt_regs *regs, unsigned long *stack);
+
+unsigned long get_wchan(struct task_struct *p);
+
+#define THREAD_SIZE_LONGS (THREAD_SIZE/sizeof(unsigned long))
+#define KSTK_TOP(info) \
+({ \
+ unsigned long *__ptr = (unsigned long *)(info); \
+ (unsigned long)(&__ptr[THREAD_SIZE_LONGS]); \
+})
+
+/*
+ * The below -8 is to reserve 8 bytes on top of the ring0 stack.
+ * This is necessary to guarantee that the entire "struct pt_regs"
+ * is accessable even if the CPU haven't stored the SS/ESP registers
+ * on the stack (interrupt gate does not save these registers
+ * when switching to the same priv ring).
+ * Therefore beware: accessing the xss/esp fields of the
+ * "struct pt_regs" is possible, but they may contain the
+ * completely wrong values.
+ */
+#define task_pt_regs(task) \
+({ \
+ struct pt_regs *__regs__; \
+ __regs__ = (struct pt_regs *)(KSTK_TOP(task_stack_page(task))-8); \
+ __regs__ - 1; \
+})
+
+#define KSTK_EIP(task) (task_pt_regs(task)->eip)
+#define KSTK_ESP(task) (task_pt_regs(task)->esp)
+
+
+struct microcode_header {
+ unsigned int hdrver;
+ unsigned int rev;
+ unsigned int date;
+ unsigned int sig;
+ unsigned int cksum;
+ unsigned int ldrver;
+ unsigned int pf;
+ unsigned int datasize;
+ unsigned int totalsize;
+ unsigned int reserved[3];
+};
+
+struct microcode {
+ struct microcode_header hdr;
+ unsigned int bits[0];
+};
+
+typedef struct microcode microcode_t;
+typedef struct microcode_header microcode_header_t;
+
+/* microcode format is extended from prescott processors */
+struct extended_signature {
+ unsigned int sig;
+ unsigned int pf;
+ unsigned int cksum;
+};
+
+struct extended_sigtable {
+ unsigned int count;
+ unsigned int cksum;
+ unsigned int reserved[3];
+ struct extended_signature sigs[0];
+};
+
+/* REP NOP (PAUSE) is a good thing to insert into busy-wait loops. */
+static inline void rep_nop(void)
+{
+ __asm__ __volatile__("rep;nop": : :"memory");
+}
+
+#define cpu_relax() rep_nop()
+
+/* generic versions from gas */
+#define GENERIC_NOP1 ".byte 0x90\n"
+#define GENERIC_NOP2 ".byte 0x89,0xf6\n"
+#define GENERIC_NOP3 ".byte 0x8d,0x76,0x00\n"
+#define GENERIC_NOP4 ".byte 0x8d,0x74,0x26,0x00\n"
+#define GENERIC_NOP5 GENERIC_NOP1 GENERIC_NOP4
+#define GENERIC_NOP6 ".byte 0x8d,0xb6,0x00,0x00,0x00,0x00\n"
+#define GENERIC_NOP7 ".byte 0x8d,0xb4,0x26,0x00,0x00,0x00,0x00\n"
+#define GENERIC_NOP8 GENERIC_NOP1 GENERIC_NOP7
+
+/* Opteron nops */
+#define K8_NOP1 GENERIC_NOP1
+#define K8_NOP2 ".byte 0x66,0x90\n"
+#define K8_NOP3 ".byte 0x66,0x66,0x90\n"
+#define K8_NOP4 ".byte 0x66,0x66,0x66,0x90\n"
+#define K8_NOP5 K8_NOP3 K8_NOP2
+#define K8_NOP6 K8_NOP3 K8_NOP3
+#define K8_NOP7 K8_NOP4 K8_NOP3
+#define K8_NOP8 K8_NOP4 K8_NOP4
+
+/* K7 nops */
+/* uses eax dependencies (arbitary choice) */
+#define K7_NOP1 GENERIC_NOP1
+#define K7_NOP2 ".byte 0x8b,0xc0\n"
+#define K7_NOP3 ".byte 0x8d,0x04,0x20\n"
+#define K7_NOP4 ".byte 0x8d,0x44,0x20,0x00\n"
+#define K7_NOP5 K7_NOP4 ASM_NOP1
+#define K7_NOP6 ".byte 0x8d,0x80,0,0,0,0\n"
+#define K7_NOP7 ".byte 0x8D,0x04,0x05,0,0,0,0\n"
+#define K7_NOP8 K7_NOP7 ASM_NOP1
+
+#ifdef CONFIG_MK8
+#define ASM_NOP1 K8_NOP1
+#define ASM_NOP2 K8_NOP2
+#define ASM_NOP3 K8_NOP3
+#define ASM_NOP4 K8_NOP4
+#define ASM_NOP5 K8_NOP5
+#define ASM_NOP6 K8_NOP6
+#define ASM_NOP7 K8_NOP7
+#define ASM_NOP8 K8_NOP8
+#elif defined(CONFIG_MK7)
+#define ASM_NOP1 K7_NOP1
+#define ASM_NOP2 K7_NOP2
+#define ASM_NOP3 K7_NOP3
+#define ASM_NOP4 K7_NOP4
+#define ASM_NOP5 K7_NOP5
+#define ASM_NOP6 K7_NOP6
+#define ASM_NOP7 K7_NOP7
+#define ASM_NOP8 K7_NOP8
+#else
+#define ASM_NOP1 GENERIC_NOP1
+#define ASM_NOP2 GENERIC_NOP2
+#define ASM_NOP3 GENERIC_NOP3
+#define ASM_NOP4 GENERIC_NOP4
+#define ASM_NOP5 GENERIC_NOP5
+#define ASM_NOP6 GENERIC_NOP6
+#define ASM_NOP7 GENERIC_NOP7
+#define ASM_NOP8 GENERIC_NOP8
+#endif
+
+#define ASM_NOP_MAX 8
+
+/* Prefetch instructions for Pentium III and AMD Athlon */
+/* It's not worth to care about 3dnow! prefetches for the K6
+ because they are microcoded there and very slow.
+ However we don't do prefetches for pre XP Athlons currently
+ That should be fixed. */
+#define ARCH_HAS_PREFETCH
+static inline void prefetch(const void *x)
+{
+ alternative_input(ASM_NOP4,
+ "prefetchnta (%1)",
+ X86_FEATURE_XMM,
+ "r" (x));
+}
+
+#define ARCH_HAS_PREFETCH
+#define ARCH_HAS_PREFETCHW
+#define ARCH_HAS_SPINLOCK_PREFETCH
+
+/* 3dnow! prefetch to get an exclusive cache line. Useful for
+ spinlocks to avoid one state transition in the cache coherency protocol. */
+static inline void prefetchw(const void *x)
+{
+ alternative_input(ASM_NOP4,
+ "prefetchw (%1)",
+ X86_FEATURE_3DNOW,
+ "r" (x));
+}
+#define spin_lock_prefetch(x) prefetchw(x)
+
+extern void select_idle_routine(const struct cpuinfo_x86 *c);
+
+#define cache_line_size() (boot_cpu_data.x86_cache_alignment)
+
+extern unsigned long boot_option_idle_override;
+extern void enable_sep_cpu(void);
+extern int sysenter_setup(void);
+
+#endif /* __ASM_I386_PROCESSOR_H */
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/include/mach-xen/asm/smp_32.h 2007-06-12 13:14:02.000000000 +0200
@@ -0,0 +1,103 @@
+#ifndef __ASM_SMP_H
+#define __ASM_SMP_H
+
+/*
+ * We need the APIC definitions automatically as part of 'smp.h'
+ */
+#ifndef __ASSEMBLY__
+#include <linux/kernel.h>
+#include <linux/threads.h>
+#include <linux/cpumask.h>
+#endif
+
+#ifdef CONFIG_X86_LOCAL_APIC
+#ifndef __ASSEMBLY__
+#include <asm/fixmap.h>
+#include <asm/bitops.h>
+#include <asm/mpspec.h>
+#ifdef CONFIG_X86_IO_APIC
+#include <asm/io_apic.h>
+#endif
+#include <asm/apic.h>
+#endif
+#endif
+
+#define BAD_APICID 0xFFu
+#ifdef CONFIG_SMP
+#ifndef __ASSEMBLY__
+
+/*
+ * Private routines/data
+ */
+
+extern void smp_alloc_memory(void);
+extern int pic_mode;
+extern int smp_num_siblings;
+extern cpumask_t cpu_sibling_map[];
+extern cpumask_t cpu_core_map[];
+
+extern void (*mtrr_hook) (void);
+extern void zap_low_mappings (void);
+extern void lock_ipi_call_lock(void);
+extern void unlock_ipi_call_lock(void);
+
+#define MAX_APICID 256
+extern u8 x86_cpu_to_apicid[];
+
+#define cpu_physical_id(cpu) x86_cpu_to_apicid[cpu]
+
+#ifdef CONFIG_HOTPLUG_CPU
+extern void cpu_exit_clear(void);
+extern void cpu_uninit(void);
+#endif
+
+/*
+ * This function is needed by all SMP systems. It must _always_ be valid
+ * from the initial startup. We map APIC_BASE very early in page_setup(),
+ * so this is correct in the x86 case.
+ */
+#define raw_smp_processor_id() (current_thread_info()->cpu)
+
+extern cpumask_t cpu_possible_map;
+#define cpu_callin_map cpu_possible_map
+
+/* We don't mark CPUs online until __cpu_up(), so we need another measure */
+static inline int num_booting_cpus(void)
+{
+ return cpus_weight(cpu_possible_map);
+}
+
+#ifdef CONFIG_X86_LOCAL_APIC
+
+#ifdef APIC_DEFINITION
+extern int hard_smp_processor_id(void);
+#else
+#include <mach_apicdef.h>
+static inline int hard_smp_processor_id(void)
+{
+ /* we don't want to mark this access volatile - bad code generation */
+ return GET_APIC_ID(*(unsigned long *)(APIC_BASE+APIC_ID));
+}
+#endif
+
+static __inline int logical_smp_processor_id(void)
+{
+ /* we don't want to mark this access volatile - bad code generation */
+ return GET_APIC_LOGICAL_ID(*(unsigned long *)(APIC_BASE+APIC_LDR));
+}
+
+#endif
+
+extern int __cpu_disable(void);
+extern void __cpu_die(unsigned int cpu);
+extern void prefill_possible_map(void);
+#endif /* !__ASSEMBLY__ */
+
+#else /* CONFIG_SMP */
+
+#define cpu_physical_id(cpu) boot_cpu_physical_apicid
+
+#define NO_PROC_ID 0xFF /* No processor magic marker */
+
+#endif
+#endif
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/include/mach-xen/asm/synch_bitops.h 2008-04-02 12:34:02.000000000 +0200
@@ -0,0 +1,126 @@
+#ifndef __XEN_SYNCH_BITOPS_H__
+#define __XEN_SYNCH_BITOPS_H__
+
+/*
+ * Copyright 1992, Linus Torvalds.
+ * Heavily modified to provide guaranteed strong synchronisation
+ * when communicating with Xen or other guest OSes running on other CPUs.
+ */
+
+#ifdef HAVE_XEN_PLATFORM_COMPAT_H
+#include <xen/platform-compat.h>
+#endif
+
+#define ADDR (*(volatile long *) addr)
+
+static __inline__ void synch_set_bit(int nr, volatile void * addr)
+{
+ __asm__ __volatile__ (
+ "lock btsl %1,%0"
+ : "+m" (ADDR) : "Ir" (nr) : "memory" );
+}
+
+static __inline__ void synch_clear_bit(int nr, volatile void * addr)
+{
+ __asm__ __volatile__ (
+ "lock btrl %1,%0"
+ : "+m" (ADDR) : "Ir" (nr) : "memory" );
+}
+
+static __inline__ void synch_change_bit(int nr, volatile void * addr)
+{
+ __asm__ __volatile__ (
+ "lock btcl %1,%0"
+ : "+m" (ADDR) : "Ir" (nr) : "memory" );
+}
+
+static __inline__ int synch_test_and_set_bit(int nr, volatile void * addr)
+{
+ int oldbit;
+ __asm__ __volatile__ (
+ "lock btsl %2,%1\n\tsbbl %0,%0"
+ : "=r" (oldbit), "+m" (ADDR) : "Ir" (nr) : "memory");
+ return oldbit;
+}
+
+static __inline__ int synch_test_and_clear_bit(int nr, volatile void * addr)
+{
+ int oldbit;
+ __asm__ __volatile__ (
+ "lock btrl %2,%1\n\tsbbl %0,%0"
+ : "=r" (oldbit), "+m" (ADDR) : "Ir" (nr) : "memory");
+ return oldbit;
+}
+
+static __inline__ int synch_test_and_change_bit(int nr, volatile void * addr)
+{
+ int oldbit;
+
+ __asm__ __volatile__ (
+ "lock btcl %2,%1\n\tsbbl %0,%0"
+ : "=r" (oldbit), "+m" (ADDR) : "Ir" (nr) : "memory");
+ return oldbit;
+}
+
+struct __synch_xchg_dummy { unsigned long a[100]; };
+#define __synch_xg(x) ((struct __synch_xchg_dummy *)(x))
+
+#define synch_cmpxchg(ptr, old, new) \
+((__typeof__(*(ptr)))__synch_cmpxchg((ptr),\
+ (unsigned long)(old), \
+ (unsigned long)(new), \
+ sizeof(*(ptr))))
+
+static inline unsigned long __synch_cmpxchg(volatile void *ptr,
+ unsigned long old,
+ unsigned long new, int size)
+{
+ unsigned long prev;
+ switch (size) {
+ case 1:
+ __asm__ __volatile__("lock; cmpxchgb %b1,%2"
+ : "=a"(prev)
+ : "q"(new), "m"(*__synch_xg(ptr)),
+ "0"(old)
+ : "memory");
+ return prev;
+ case 2:
+ __asm__ __volatile__("lock; cmpxchgw %w1,%2"
+ : "=a"(prev)
+ : "r"(new), "m"(*__synch_xg(ptr)),
+ "0"(old)
+ : "memory");
+ return prev;
+#ifdef CONFIG_X86_64
+ case 4:
+ __asm__ __volatile__("lock; cmpxchgl %k1,%2"
+ : "=a"(prev)
+ : "r"(new), "m"(*__synch_xg(ptr)),
+ "0"(old)
+ : "memory");
+ return prev;
+ case 8:
+ __asm__ __volatile__("lock; cmpxchgq %1,%2"
+ : "=a"(prev)
+ : "r"(new), "m"(*__synch_xg(ptr)),
+ "0"(old)
+ : "memory");
+ return prev;
+#else
+ case 4:
+ __asm__ __volatile__("lock; cmpxchgl %1,%2"
+ : "=a"(prev)
+ : "r"(new), "m"(*__synch_xg(ptr)),
+ "0"(old)
+ : "memory");
+ return prev;
+#endif
+ }
+ return old;
+}
+
+#define synch_test_bit test_bit
+
+#define synch_cmpxchg_subword synch_cmpxchg
+
+#endif /* __XEN_SYNCH_BITOPS_H__ */
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/include/mach-xen/asm/system_32.h 2007-06-12 13:14:02.000000000 +0200
@@ -0,0 +1,488 @@
+#ifndef __ASM_SYSTEM_H
+#define __ASM_SYSTEM_H
+
+#include <linux/kernel.h>
+#include <asm/segment.h>
+#include <asm/cpufeature.h>
+#include <linux/bitops.h> /* for LOCK_PREFIX */
+#include <asm/synch_bitops.h>
+#include <asm/hypervisor.h>
+
+#ifdef __KERNEL__
+
+struct task_struct; /* one of the stranger aspects of C forward declarations.. */
+extern struct task_struct * FASTCALL(__switch_to(struct task_struct *prev, struct task_struct *next));
+
+/*
+ * Saving eflags is important. It switches not only IOPL between tasks,
+ * it also protects other tasks from NT leaking through sysenter etc.
+ */
+#define switch_to(prev,next,last) do { \
+ unsigned long esi,edi; \
+ asm volatile("pushfl\n\t" /* Save flags */ \
+ "pushl %%ebp\n\t" \
+ "movl %%esp,%0\n\t" /* save ESP */ \
+ "movl %5,%%esp\n\t" /* restore ESP */ \
+ "movl $1f,%1\n\t" /* save EIP */ \
+ "pushl %6\n\t" /* restore EIP */ \
+ "jmp __switch_to\n" \
+ "1:\t" \
+ "popl %%ebp\n\t" \
+ "popfl" \
+ :"=m" (prev->thread.esp),"=m" (prev->thread.eip), \
+ "=a" (last),"=S" (esi),"=D" (edi) \
+ :"m" (next->thread.esp),"m" (next->thread.eip), \
+ "2" (prev), "d" (next)); \
+} while (0)
+
+#define _set_base(addr,base) do { unsigned long __pr; \
+__asm__ __volatile__ ("movw %%dx,%1\n\t" \
+ "rorl $16,%%edx\n\t" \
+ "movb %%dl,%2\n\t" \
+ "movb %%dh,%3" \
+ :"=&d" (__pr) \
+ :"m" (*((addr)+2)), \
+ "m" (*((addr)+4)), \
+ "m" (*((addr)+7)), \
+ "0" (base) \
+ ); } while(0)
+
+#define _set_limit(addr,limit) do { unsigned long __lr; \
+__asm__ __volatile__ ("movw %%dx,%1\n\t" \
+ "rorl $16,%%edx\n\t" \
+ "movb %2,%%dh\n\t" \
+ "andb $0xf0,%%dh\n\t" \
+ "orb %%dh,%%dl\n\t" \
+ "movb %%dl,%2" \
+ :"=&d" (__lr) \
+ :"m" (*(addr)), \
+ "m" (*((addr)+6)), \
+ "0" (limit) \
+ ); } while(0)
+
+#define set_base(ldt,base) _set_base( ((char *)&(ldt)) , (base) )
+#define set_limit(ldt,limit) _set_limit( ((char *)&(ldt)) , ((limit)-1) )
+
+/*
+ * Load a segment. Fall back on loading the zero
+ * segment if something goes wrong..
+ */
+#define loadsegment(seg,value) \
+ asm volatile("\n" \
+ "1:\t" \
+ "mov %0,%%" #seg "\n" \
+ "2:\n" \
+ ".section .fixup,\"ax\"\n" \
+ "3:\t" \
+ "pushl $0\n\t" \
+ "popl %%" #seg "\n\t" \
+ "jmp 2b\n" \
+ ".previous\n" \
+ ".section __ex_table,\"a\"\n\t" \
+ ".align 4\n\t" \
+ ".long 1b,3b\n" \
+ ".previous" \
+ : :"rm" (value))
+
+/*
+ * Save a segment register away
+ */
+#define savesegment(seg, value) \
+ asm volatile("mov %%" #seg ",%0":"=rm" (value))
+
+#define read_cr0() ({ \
+ unsigned int __dummy; \
+ __asm__ __volatile__( \
+ "movl %%cr0,%0\n\t" \
+ :"=r" (__dummy)); \
+ __dummy; \
+})
+#define write_cr0(x) \
+ __asm__ __volatile__("movl %0,%%cr0": :"r" (x))
+
+#define read_cr2() (current_vcpu_info()->arch.cr2)
+#define write_cr2(x) \
+ __asm__ __volatile__("movl %0,%%cr2": :"r" (x))
+
+#define read_cr3() ({ \
+ unsigned int __dummy; \
+ __asm__ ( \
+ "movl %%cr3,%0\n\t" \
+ :"=r" (__dummy)); \
+ __dummy = xen_cr3_to_pfn(__dummy); \
+ mfn_to_pfn(__dummy) << PAGE_SHIFT; \
+})
+#define write_cr3(x) ({ \
+ unsigned int __dummy = pfn_to_mfn((x) >> PAGE_SHIFT); \
+ __dummy = xen_pfn_to_cr3(__dummy); \
+ __asm__ __volatile__("movl %0,%%cr3": :"r" (__dummy)); \
+})
+#define read_cr4() ({ \
+ unsigned int __dummy; \
+ __asm__( \
+ "movl %%cr4,%0\n\t" \
+ :"=r" (__dummy)); \
+ __dummy; \
+})
+#define read_cr4_safe() ({ \
+ unsigned int __dummy; \
+ /* This could fault if %cr4 does not exist */ \
+ __asm__("1: movl %%cr4, %0 \n" \
+ "2: \n" \
+ ".section __ex_table,\"a\" \n" \
+ ".long 1b,2b \n" \
+ ".previous \n" \
+ : "=r" (__dummy): "0" (0)); \
+ __dummy; \
+})
+
+#define write_cr4(x) \
+ __asm__ __volatile__("movl %0,%%cr4": :"r" (x))
+
+/*
+ * Clear and set 'TS' bit respectively
+ */
+#define clts() (HYPERVISOR_fpu_taskswitch(0))
+#define stts() (HYPERVISOR_fpu_taskswitch(1))
+
+#endif /* __KERNEL__ */
+
+#define wbinvd() \
+ __asm__ __volatile__ ("wbinvd": : :"memory")
+
+static inline unsigned long get_limit(unsigned long segment)
+{
+ unsigned long __limit;
+ __asm__("lsll %1,%0"
+ :"=r" (__limit):"r" (segment));
+ return __limit+1;
+}
+
+#define nop() __asm__ __volatile__ ("nop")
+
+#define xchg(ptr,v) ((__typeof__(*(ptr)))__xchg((unsigned long)(v),(ptr),sizeof(*(ptr))))
+
+#define tas(ptr) (xchg((ptr),1))
+
+struct __xchg_dummy { unsigned long a[100]; };
+#define __xg(x) ((struct __xchg_dummy *)(x))
+
+
+#ifdef CONFIG_X86_CMPXCHG64
+
+/*
+ * The semantics of XCHGCMP8B are a bit strange, this is why
+ * there is a loop and the loading of %%eax and %%edx has to
+ * be inside. This inlines well in most cases, the cached
+ * cost is around ~38 cycles. (in the future we might want
+ * to do an SIMD/3DNOW!/MMX/FPU 64-bit store here, but that
+ * might have an implicit FPU-save as a cost, so it's not
+ * clear which path to go.)
+ *
+ * cmpxchg8b must be used with the lock prefix here to allow
+ * the instruction to be executed atomically, see page 3-102
+ * of the instruction set reference 24319102.pdf. We need
+ * the reader side to see the coherent 64bit value.
+ */
+static inline void __set_64bit (unsigned long long * ptr,
+ unsigned int low, unsigned int high)
+{
+ __asm__ __volatile__ (
+ "\n1:\t"
+ "movl (%0), %%eax\n\t"
+ "movl 4(%0), %%edx\n\t"
+ "lock cmpxchg8b (%0)\n\t"
+ "jnz 1b"
+ : /* no outputs */
+ : "D"(ptr),
+ "b"(low),
+ "c"(high)
+ : "ax","dx","memory");
+}
+
+static inline void __set_64bit_constant (unsigned long long *ptr,
+ unsigned long long value)
+{
+ __set_64bit(ptr,(unsigned int)(value), (unsigned int)((value)>>32ULL));
+}
+#define ll_low(x) *(((unsigned int*)&(x))+0)
+#define ll_high(x) *(((unsigned int*)&(x))+1)
+
+static inline void __set_64bit_var (unsigned long long *ptr,
+ unsigned long long value)
+{
+ __set_64bit(ptr,ll_low(value), ll_high(value));
+}
+
+#define set_64bit(ptr,value) \
+(__builtin_constant_p(value) ? \
+ __set_64bit_constant(ptr, value) : \
+ __set_64bit_var(ptr, value) )
+
+#define _set_64bit(ptr,value) \
+(__builtin_constant_p(value) ? \
+ __set_64bit(ptr, (unsigned int)(value), (unsigned int)((value)>>32ULL) ) : \
+ __set_64bit(ptr, ll_low(value), ll_high(value)) )
+
+#endif
+
+/*
+ * Note: no "lock" prefix even on SMP: xchg always implies lock anyway
+ * Note 2: xchg has side effect, so that attribute volatile is necessary,
+ * but generally the primitive is invalid, *ptr is output argument. --ANK
+ */
+static inline unsigned long __xchg(unsigned long x, volatile void * ptr, int size)
+{
+ switch (size) {
+ case 1:
+ __asm__ __volatile__("xchgb %b0,%1"
+ :"=q" (x)
+ :"m" (*__xg(ptr)), "0" (x)
+ :"memory");
+ break;
+ case 2:
+ __asm__ __volatile__("xchgw %w0,%1"
+ :"=r" (x)
+ :"m" (*__xg(ptr)), "0" (x)
+ :"memory");
+ break;
+ case 4:
+ __asm__ __volatile__("xchgl %0,%1"
+ :"=r" (x)
+ :"m" (*__xg(ptr)), "0" (x)
+ :"memory");
+ break;
+ }
+ return x;
+}
+
+/*
+ * Atomic compare and exchange. Compare OLD with MEM, if identical,
+ * store NEW in MEM. Return the initial value in MEM. Success is
+ * indicated by comparing RETURN with OLD.
+ */
+
+#ifdef CONFIG_X86_CMPXCHG
+#define __HAVE_ARCH_CMPXCHG 1
+#define cmpxchg(ptr,o,n)\
+ ((__typeof__(*(ptr)))__cmpxchg((ptr),(unsigned long)(o),\
+ (unsigned long)(n),sizeof(*(ptr))))
+#endif
+
+static inline unsigned long __cmpxchg(volatile void *ptr, unsigned long old,
+ unsigned long new, int size)
+{
+ unsigned long prev;
+ switch (size) {
+ case 1:
+ __asm__ __volatile__(LOCK_PREFIX "cmpxchgb %b1,%2"
+ : "=a"(prev)
+ : "q"(new), "m"(*__xg(ptr)), "0"(old)
+ : "memory");
+ return prev;
+ case 2:
+ __asm__ __volatile__(LOCK_PREFIX "cmpxchgw %w1,%2"
+ : "=a"(prev)
+ : "r"(new), "m"(*__xg(ptr)), "0"(old)
+ : "memory");
+ return prev;
+ case 4:
+ __asm__ __volatile__(LOCK_PREFIX "cmpxchgl %1,%2"
+ : "=a"(prev)
+ : "r"(new), "m"(*__xg(ptr)), "0"(old)
+ : "memory");
+ return prev;
+ }
+ return old;
+}
+
+#ifndef CONFIG_X86_CMPXCHG
+/*
+ * Building a kernel capable running on 80386. It may be necessary to
+ * simulate the cmpxchg on the 80386 CPU. For that purpose we define
+ * a function for each of the sizes we support.
+ */
+
+extern unsigned long cmpxchg_386_u8(volatile void *, u8, u8);
+extern unsigned long cmpxchg_386_u16(volatile void *, u16, u16);
+extern unsigned long cmpxchg_386_u32(volatile void *, u32, u32);
+
+static inline unsigned long cmpxchg_386(volatile void *ptr, unsigned long old,
+ unsigned long new, int size)
+{
+ switch (size) {
+ case 1:
+ return cmpxchg_386_u8(ptr, old, new);
+ case 2:
+ return cmpxchg_386_u16(ptr, old, new);
+ case 4:
+ return cmpxchg_386_u32(ptr, old, new);
+ }
+ return old;
+}
+
+#define cmpxchg(ptr,o,n) \
+({ \
+ __typeof__(*(ptr)) __ret; \
+ if (likely(boot_cpu_data.x86 > 3)) \
+ __ret = __cmpxchg((ptr), (unsigned long)(o), \
+ (unsigned long)(n), sizeof(*(ptr))); \
+ else \
+ __ret = cmpxchg_386((ptr), (unsigned long)(o), \
+ (unsigned long)(n), sizeof(*(ptr))); \
+ __ret; \
+})
+#endif
+
+#ifdef CONFIG_X86_CMPXCHG64
+
+static inline unsigned long long __cmpxchg64(volatile void *ptr, unsigned long long old,
+ unsigned long long new)
+{
+ unsigned long long prev;
+ __asm__ __volatile__(LOCK_PREFIX "cmpxchg8b %3"
+ : "=A"(prev)
+ : "b"((unsigned long)new),
+ "c"((unsigned long)(new >> 32)),
+ "m"(*__xg(ptr)),
+ "0"(old)
+ : "memory");
+ return prev;
+}
+
+#define cmpxchg64(ptr,o,n)\
+ ((__typeof__(*(ptr)))__cmpxchg64((ptr),(unsigned long long)(o),\
+ (unsigned long long)(n)))
+
+#endif
+
+/*
+ * Force strict CPU ordering.
+ * And yes, this is required on UP too when we're talking
+ * to devices.
+ *
+ * For now, "wmb()" doesn't actually do anything, as all
+ * Intel CPU's follow what Intel calls a *Processor Order*,
+ * in which all writes are seen in the program order even
+ * outside the CPU.
+ *
+ * I expect future Intel CPU's to have a weaker ordering,
+ * but I'd also expect them to finally get their act together
+ * and add some real memory barriers if so.
+ *
+ * Some non intel clones support out of order store. wmb() ceases to be a
+ * nop for these.
+ */
+
+
+/*
+ * Actually only lfence would be needed for mb() because all stores done
+ * by the kernel should be already ordered. But keep a full barrier for now.
+ */
+
+#define mb() alternative("lock; addl $0,0(%%esp)", "mfence", X86_FEATURE_XMM2)
+#define rmb() alternative("lock; addl $0,0(%%esp)", "lfence", X86_FEATURE_XMM2)
+
+/**
+ * read_barrier_depends - Flush all pending reads that subsequents reads
+ * depend on.
+ *
+ * No data-dependent reads from memory-like regions are ever reordered
+ * over this barrier. All reads preceding this primitive are guaranteed
+ * to access memory (but not necessarily other CPUs' caches) before any
+ * reads following this primitive that depend on the data return by
+ * any of the preceding reads. This primitive is much lighter weight than
+ * rmb() on most CPUs, and is never heavier weight than is
+ * rmb().
+ *
+ * These ordering constraints are respected by both the local CPU
+ * and the compiler.
+ *
+ * Ordering is not guaranteed by anything other than these primitives,
+ * not even by data dependencies. See the documentation for
+ * memory_barrier() for examples and URLs to more information.
+ *
+ * For example, the following code would force ordering (the initial
+ * value of "a" is zero, "b" is one, and "p" is "&a"):
+ *
+ * <programlisting>
+ * CPU 0 CPU 1
+ *
+ * b = 2;
+ * memory_barrier();
+ * p = &b; q = p;
+ * read_barrier_depends();
+ * d = *q;
+ * </programlisting>
+ *
+ * because the read of "*q" depends on the read of "p" and these
+ * two reads are separated by a read_barrier_depends(). However,
+ * the following code, with the same initial values for "a" and "b":
+ *
+ * <programlisting>
+ * CPU 0 CPU 1
+ *
+ * a = 2;
+ * memory_barrier();
+ * b = 3; y = b;
+ * read_barrier_depends();
+ * x = a;
+ * </programlisting>
+ *
+ * does not enforce ordering, since there is no data dependency between
+ * the read of "a" and the read of "b". Therefore, on some CPUs, such
+ * as Alpha, "y" could be set to 3 and "x" to 0. Use rmb()
+ * in cases like this where there are no data dependencies.
+ **/
+
+#define read_barrier_depends() do { } while(0)
+
+#ifdef CONFIG_X86_OOSTORE
+/* Actually there are no OOO store capable CPUs for now that do SSE,
+ but make it already an possibility. */
+#define wmb() alternative("lock; addl $0,0(%%esp)", "sfence", X86_FEATURE_XMM)
+#else
+#define wmb() __asm__ __volatile__ ("": : :"memory")
+#endif
+
+#ifdef CONFIG_SMP
+#define smp_mb() mb()
+#define smp_rmb() rmb()
+#define smp_wmb() wmb()
+#define smp_read_barrier_depends() read_barrier_depends()
+#define set_mb(var, value) do { (void) xchg(&var, value); } while (0)
+#else
+#define smp_mb() barrier()
+#define smp_rmb() barrier()
+#define smp_wmb() barrier()
+#define smp_read_barrier_depends() do { } while(0)
+#define set_mb(var, value) do { var = value; barrier(); } while (0)
+#endif
+
+#include <linux/irqflags.h>
+
+/*
+ * disable hlt during certain critical i/o operations
+ */
+#define HAVE_DISABLE_HLT
+void disable_hlt(void);
+void enable_hlt(void);
+
+extern int es7000_plat;
+void cpu_idle_wait(void);
+
+/*
+ * On SMP systems, when the scheduler does migration-cost autodetection,
+ * it needs a way to flush as much of the CPU's caches as possible:
+ */
+static inline void sched_cacheflush(void)
+{
+ wbinvd();
+}
+
+extern unsigned long arch_align_stack(unsigned long sp);
+extern void free_init_pages(char *what, unsigned long begin, unsigned long end);
+
+void default_idle(void);
+
+#endif
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/include/mach-xen/asm/tlbflush_32.h 2007-11-26 16:59:25.000000000 +0100
@@ -0,0 +1,101 @@
+#ifndef _I386_TLBFLUSH_H
+#define _I386_TLBFLUSH_H
+
+#include <linux/mm.h>
+#include <asm/processor.h>
+
+#define __flush_tlb() xen_tlb_flush()
+#define __flush_tlb_global() xen_tlb_flush()
+#define __flush_tlb_all() xen_tlb_flush()
+
+extern unsigned long pgkern_mask;
+
+#define cpu_has_invlpg (boot_cpu_data.x86 > 3)
+
+#define __flush_tlb_single(addr) xen_invlpg(addr)
+
+#define __flush_tlb_one(addr) __flush_tlb_single(addr)
+
+/*
+ * TLB flushing:
+ *
+ * - flush_tlb() flushes the current mm struct TLBs
+ * - flush_tlb_all() flushes all processes TLBs
+ * - flush_tlb_mm(mm) flushes the specified mm context TLB's
+ * - flush_tlb_page(vma, vmaddr) flushes one page
+ * - flush_tlb_range(vma, start, end) flushes a range of pages
+ * - flush_tlb_kernel_range(start, end) flushes a range of kernel pages
+ * - flush_tlb_pgtables(mm, start, end) flushes a range of page tables
+ *
+ * ..but the i386 has somewhat limited tlb flushing capabilities,
+ * and page-granular flushes are available only on i486 and up.
+ */
+
+#ifndef CONFIG_SMP
+
+#define flush_tlb() __flush_tlb()
+#define flush_tlb_all() __flush_tlb_all()
+#define local_flush_tlb() __flush_tlb()
+
+static inline void flush_tlb_mm(struct mm_struct *mm)
+{
+ if (mm == current->active_mm)
+ __flush_tlb();
+}
+
+static inline void flush_tlb_page(struct vm_area_struct *vma,
+ unsigned long addr)
+{
+ if (vma->vm_mm == current->active_mm)
+ __flush_tlb_one(addr);
+}
+
+static inline void flush_tlb_range(struct vm_area_struct *vma,
+ unsigned long start, unsigned long end)
+{
+ if (vma->vm_mm == current->active_mm)
+ __flush_tlb();
+}
+
+#else
+
+#include <asm/smp.h>
+
+#define local_flush_tlb() \
+ __flush_tlb()
+
+#define flush_tlb_all xen_tlb_flush_all
+#define flush_tlb_current_task() xen_tlb_flush_mask(&current->mm->cpu_vm_mask)
+#define flush_tlb_mm(mm) xen_tlb_flush_mask(&(mm)->cpu_vm_mask)
+#define flush_tlb_page(vma, va) xen_invlpg_mask(&(vma)->vm_mm->cpu_vm_mask, va)
+
+#define flush_tlb() flush_tlb_current_task()
+
+static inline void flush_tlb_range(struct vm_area_struct * vma, unsigned long start, unsigned long end)
+{
+ flush_tlb_mm(vma->vm_mm);
+}
+
+#define TLBSTATE_OK 1
+#define TLBSTATE_LAZY 2
+
+struct tlb_state
+{
+ struct mm_struct *active_mm;
+ int state;
+ char __cacheline_padding[L1_CACHE_BYTES-8];
+};
+DECLARE_PER_CPU(struct tlb_state, cpu_tlbstate);
+
+
+#endif
+
+#define flush_tlb_kernel_range(start, end) flush_tlb_all()
+
+static inline void flush_tlb_pgtables(struct mm_struct *mm,
+ unsigned long start, unsigned long end)
+{
+ /* i386 does not keep any page table caches in TLB */
+}
+
+#endif /* _I386_TLBFLUSH_H */
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/include/mach-xen/asm/vga.h 2007-06-12 13:14:02.000000000 +0200
@@ -0,0 +1,20 @@
+/*
+ * Access to VGA videoram
+ *
+ * (c) 1998 Martin Mares <mj@ucw.cz>
+ */
+
+#ifndef _LINUX_ASM_VGA_H_
+#define _LINUX_ASM_VGA_H_
+
+/*
+ * On the PC, we can just recalculate addresses and then
+ * access the videoram directly without any black magic.
+ */
+
+#define VGA_MAP_MEM(x,s) (unsigned long)isa_bus_to_virt(x)
+
+#define vga_readb(x) (*(x))
+#define vga_writeb(x,y) (*(y) = (x))
+
+#endif
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/include/mach-xen/asm/xenoprof.h 2007-06-12 13:14:02.000000000 +0200
@@ -0,0 +1,48 @@
+/******************************************************************************
+ * asm-i386/mach-xen/asm/xenoprof.h
+ *
+ * Copyright (c) 2006 Isaku Yamahata <yamahata at valinux co jp>
+ * VA Linux Systems Japan K.K.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ *
+ */
+#ifndef __ASM_XENOPROF_H__
+#define __ASM_XENOPROF_H__
+#ifdef CONFIG_XEN
+
+struct super_block;
+struct dentry;
+int xenoprof_create_files(struct super_block * sb, struct dentry * root);
+#define HAVE_XENOPROF_CREATE_FILES
+
+struct xenoprof_init;
+void xenoprof_arch_init_counter(struct xenoprof_init *init);
+void xenoprof_arch_counter(void);
+void xenoprof_arch_start(void);
+void xenoprof_arch_stop(void);
+
+struct xenoprof_arch_shared_buffer {
+ /* nothing */
+};
+struct xenoprof_shared_buffer;
+void xenoprof_arch_unmap_shared_buffer(struct xenoprof_shared_buffer* sbuf);
+struct xenoprof_get_buffer;
+int xenoprof_arch_map_shared_buffer(struct xenoprof_get_buffer* get_buffer, struct xenoprof_shared_buffer* sbuf);
+struct xenoprof_passive;
+int xenoprof_arch_set_passive(struct xenoprof_passive* pdomain, struct xenoprof_shared_buffer* sbuf);
+
+#endif /* CONFIG_XEN */
+#endif /* __ASM_XENOPROF_H__ */
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/include/mach-xen/irq_vectors.h 2008-09-25 13:55:32.000000000 +0200
@@ -0,0 +1,125 @@
+/*
+ * This file should contain #defines for all of the interrupt vector
+ * numbers used by this architecture.
+ *
+ * In addition, there are some standard defines:
+ *
+ * FIRST_EXTERNAL_VECTOR:
+ * The first free place for external interrupts
+ *
+ * SYSCALL_VECTOR:
+ * The IRQ vector a syscall makes the user to kernel transition
+ * under.
+ *
+ * TIMER_IRQ:
+ * The IRQ number the timer interrupt comes in at.
+ *
+ * NR_IRQS:
+ * The total number of interrupt vectors (including all the
+ * architecture specific interrupts) needed.
+ *
+ */
+#ifndef _ASM_IRQ_VECTORS_H
+#define _ASM_IRQ_VECTORS_H
+
+/*
+ * IDT vectors usable for external interrupt sources start
+ * at 0x20:
+ */
+#define FIRST_EXTERNAL_VECTOR 0x20
+
+#define SYSCALL_VECTOR 0x80
+
+/*
+ * Vectors 0x20-0x2f are used for ISA interrupts.
+ */
+
+#if 0
+/*
+ * Special IRQ vectors used by the SMP architecture, 0xf0-0xff
+ *
+ * some of the following vectors are 'rare', they are merged
+ * into a single vector (CALL_FUNCTION_VECTOR) to save vector space.
+ * TLB, reschedule and local APIC vectors are performance-critical.
+ *
+ * Vectors 0xf0-0xfa are free (reserved for future Linux use).
+ */
+#define SPURIOUS_APIC_VECTOR 0xff
+#define ERROR_APIC_VECTOR 0xfe
+#define INVALIDATE_TLB_VECTOR 0xfd
+#define RESCHEDULE_VECTOR 0xfc
+#define CALL_FUNCTION_VECTOR 0xfb
+
+#define THERMAL_APIC_VECTOR 0xf0
+/*
+ * Local APIC timer IRQ vector is on a different priority level,
+ * to work around the 'lost local interrupt if more than 2 IRQ
+ * sources per level' errata.
+ */
+#define LOCAL_TIMER_VECTOR 0xef
+#endif
+
+#define SPURIOUS_APIC_VECTOR 0xff
+#define ERROR_APIC_VECTOR 0xfe
+
+/*
+ * First APIC vector available to drivers: (vectors 0x30-0xee)
+ * we start at 0x31 to spread out vectors evenly between priority
+ * levels. (0x80 is the syscall vector)
+ */
+#define FIRST_DEVICE_VECTOR 0x31
+#define FIRST_SYSTEM_VECTOR 0xef
+
+/*
+ * 16 8259A IRQ's, 208 potential APIC interrupt sources.
+ * Right now the APIC is mostly only used for SMP.
+ * 256 vectors is an architectural limit. (we can have
+ * more than 256 devices theoretically, but they will
+ * have to use shared interrupts)
+ * Since vectors 0x00-0x1f are used/reserved for the CPU,
+ * the usable vector space is 0x20-0xff (224 vectors)
+ */
+
+#define RESCHEDULE_VECTOR 0
+#define CALL_FUNCTION_VECTOR 1
+#define NR_IPIS 2
+
+/*
+ * The maximum number of vectors supported by i386 processors
+ * is limited to 256. For processors other than i386, NR_VECTORS
+ * should be changed accordingly.
+ */
+#define NR_VECTORS 256
+
+#define FPU_IRQ 13
+
+#define FIRST_VM86_IRQ 3
+#define LAST_VM86_IRQ 15
+#define invalid_vm86_irq(irq) ((irq) < 3 || (irq) > 15)
+
+/*
+ * The flat IRQ space is divided into two regions:
+ * 1. A one-to-one mapping of real physical IRQs. This space is only used
+ * if we have physical device-access privilege. This region is at the
+ * start of the IRQ space so that existing device drivers do not need
+ * to be modified to translate physical IRQ numbers into our IRQ space.
+ * 3. A dynamic mapping of inter-domain and Xen-sourced virtual IRQs. These
+ * are bound using the provided bind/unbind functions.
+ */
+
+#define PIRQ_BASE 0
+#if !defined(MAX_IO_APICS)
+# define NR_PIRQS (NR_VECTORS + 32 * NR_CPUS)
+#elif NR_CPUS < MAX_IO_APICS
+# define NR_PIRQS (NR_VECTORS + 32 * NR_CPUS)
+#else
+# define NR_PIRQS (NR_VECTORS + 32 * MAX_IO_APICS)
+#endif
+
+#define DYNIRQ_BASE (PIRQ_BASE + NR_PIRQS)
+#define NR_DYNIRQS 256
+
+#define NR_IRQS (NR_PIRQS + NR_DYNIRQS)
+#define NR_IRQ_VECTORS NR_IRQS
+
+#endif /* _ASM_IRQ_VECTORS_H */
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/include/mach-xen/asm/mach_traps.h 2007-06-12 13:14:02.000000000 +0200
@@ -0,0 +1,33 @@
+/*
+ * include/asm-xen/asm-i386/mach-xen/mach_traps.h
+ *
+ * Machine specific NMI handling for Xen
+ */
+#ifndef _MACH_TRAPS_H
+#define _MACH_TRAPS_H
+
+#include <linux/bitops.h>
+#include <xen/interface/nmi.h>
+
+static inline void clear_mem_error(unsigned char reason) {}
+static inline void clear_io_check_error(unsigned char reason) {}
+
+static inline unsigned char get_nmi_reason(void)
+{
+ shared_info_t *s = HYPERVISOR_shared_info;
+ unsigned char reason = 0;
+
+ /* construct a value which looks like it came from
+ * port 0x61.
+ */
+ if (test_bit(_XEN_NMIREASON_io_error, &s->arch.nmi_reason))
+ reason |= 0x40;
+ if (test_bit(_XEN_NMIREASON_parity_error, &s->arch.nmi_reason))
+ reason |= 0x80;
+
+ return reason;
+}
+
+static inline void reassert_nmi(void) {}
+
+#endif /* !_MACH_TRAPS_H */
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/include/mach-xen/asm/desc_64.h 2008-01-28 12:24:19.000000000 +0100
@@ -0,0 +1,265 @@
+/* Written 2000 by Andi Kleen */
+#ifndef __ARCH_DESC_H
+#define __ARCH_DESC_H
+
+#include <linux/threads.h>
+#include <asm/ldt.h>
+
+#ifndef __ASSEMBLY__
+
+#include <linux/string.h>
+#include <linux/smp.h>
+
+#include <asm/segment.h>
+#include <asm/mmu.h>
+
+// 8 byte segment descriptor
+struct desc_struct {
+ u16 limit0;
+ u16 base0;
+ unsigned base1 : 8, type : 4, s : 1, dpl : 2, p : 1;
+ unsigned limit : 4, avl : 1, l : 1, d : 1, g : 1, base2 : 8;
+} __attribute__((packed));
+
+struct n_desc_struct {
+ unsigned int a,b;
+};
+
+enum {
+ GATE_INTERRUPT = 0xE,
+ GATE_TRAP = 0xF,
+ GATE_CALL = 0xC,
+};
+
+// 16byte gate
+struct gate_struct {
+ u16 offset_low;
+ u16 segment;
+ unsigned ist : 3, zero0 : 5, type : 5, dpl : 2, p : 1;
+ u16 offset_middle;
+ u32 offset_high;
+ u32 zero1;
+} __attribute__((packed));
+
+#define PTR_LOW(x) ((unsigned long)(x) & 0xFFFF)
+#define PTR_MIDDLE(x) (((unsigned long)(x) >> 16) & 0xFFFF)
+#define PTR_HIGH(x) ((unsigned long)(x) >> 32)
+
+enum {
+ DESC_TSS = 0x9,
+ DESC_LDT = 0x2,
+};
+
+// LDT or TSS descriptor in the GDT. 16 bytes.
+struct ldttss_desc {
+ u16 limit0;
+ u16 base0;
+ unsigned base1 : 8, type : 5, dpl : 2, p : 1;
+ unsigned limit1 : 4, zero0 : 3, g : 1, base2 : 8;
+ u32 base3;
+ u32 zero1;
+} __attribute__((packed));
+
+struct desc_ptr {
+ unsigned short size;
+ unsigned long address;
+} __attribute__((packed)) ;
+
+extern struct desc_ptr idt_descr, cpu_gdt_descr[NR_CPUS];
+
+extern struct desc_struct cpu_gdt_table[GDT_ENTRIES];
+
+#define load_TR_desc() asm volatile("ltr %w0"::"r" (GDT_ENTRY_TSS*8))
+#define load_LDT_desc() asm volatile("lldt %w0"::"r" (GDT_ENTRY_LDT*8))
+
+static inline void clear_LDT(void)
+{
+ int cpu = get_cpu();
+
+ /*
+ * NB. We load the default_ldt for lcall7/27 handling on demand, as
+ * it slows down context switching. Noone uses it anyway.
+ */
+ cpu = cpu; /* XXX avoid compiler warning */
+ xen_set_ldt(NULL, 0);
+ put_cpu();
+}
+
+/*
+ * This is the ldt that every process will get unless we need
+ * something other than this.
+ */
+extern struct desc_struct default_ldt[];
+#ifndef CONFIG_X86_NO_IDT
+extern struct gate_struct idt_table[];
+#endif
+extern struct desc_ptr cpu_gdt_descr[];
+
+/* the cpu gdt accessor */
+#define cpu_gdt(_cpu) ((struct desc_struct *)cpu_gdt_descr[_cpu].address)
+
+static inline void _set_gate(void *adr, unsigned type, unsigned long func, unsigned dpl, unsigned ist)
+{
+ struct gate_struct s;
+ s.offset_low = PTR_LOW(func);
+ s.segment = __KERNEL_CS;
+ s.ist = ist;
+ s.p = 1;
+ s.dpl = dpl;
+ s.zero0 = 0;
+ s.zero1 = 0;
+ s.type = type;
+ s.offset_middle = PTR_MIDDLE(func);
+ s.offset_high = PTR_HIGH(func);
+ /* does not need to be atomic because it is only done once at setup time */
+ memcpy(adr, &s, 16);
+}
+
+#ifndef CONFIG_X86_NO_IDT
+static inline void set_intr_gate(int nr, void *func)
+{
+ BUG_ON((unsigned)nr > 0xFF);
+ _set_gate(&idt_table[nr], GATE_INTERRUPT, (unsigned long) func, 0, 0);
+}
+
+static inline void set_intr_gate_ist(int nr, void *func, unsigned ist)
+{
+ BUG_ON((unsigned)nr > 0xFF);
+ _set_gate(&idt_table[nr], GATE_INTERRUPT, (unsigned long) func, 0, ist);
+}
+
+static inline void set_system_gate(int nr, void *func)
+{
+ BUG_ON((unsigned)nr > 0xFF);
+ _set_gate(&idt_table[nr], GATE_INTERRUPT, (unsigned long) func, 3, 0);
+}
+
+static inline void set_system_gate_ist(int nr, void *func, unsigned ist)
+{
+ _set_gate(&idt_table[nr], GATE_INTERRUPT, (unsigned long) func, 3, ist);
+}
+#endif
+
+static inline void set_tssldt_descriptor(void *ptr, unsigned long tss, unsigned type,
+ unsigned size)
+{
+ struct ldttss_desc d;
+ memset(&d,0,sizeof(d));
+ d.limit0 = size & 0xFFFF;
+ d.base0 = PTR_LOW(tss);
+ d.base1 = PTR_MIDDLE(tss) & 0xFF;
+ d.type = type;
+ d.p = 1;
+ d.limit1 = (size >> 16) & 0xF;
+ d.base2 = (PTR_MIDDLE(tss) >> 8) & 0xFF;
+ d.base3 = PTR_HIGH(tss);
+ memcpy(ptr, &d, 16);
+}
+
+#ifndef CONFIG_X86_NO_TSS
+static inline void set_tss_desc(unsigned cpu, void *addr)
+{
+ /*
+ * sizeof(unsigned long) coming from an extra "long" at the end
+ * of the iobitmap. See tss_struct definition in processor.h
+ *
+ * -1? seg base+limit should be pointing to the address of the
+ * last valid byte
+ */
+ set_tssldt_descriptor(&cpu_gdt(cpu)[GDT_ENTRY_TSS],
+ (unsigned long)addr, DESC_TSS,
+ IO_BITMAP_OFFSET + IO_BITMAP_BYTES + sizeof(unsigned long) - 1);
+}
+#endif
+
+static inline void set_ldt_desc(unsigned cpu, void *addr, int size)
+{
+ set_tssldt_descriptor(&cpu_gdt(cpu)[GDT_ENTRY_LDT], (unsigned long)addr,
+ DESC_LDT, size * 8 - 1);
+}
+
+static inline void set_seg_base(unsigned cpu, int entry, void *base)
+{
+ struct desc_struct *d = &cpu_gdt(cpu)[entry];
+ u32 addr = (u32)(u64)base;
+ BUG_ON((u64)base >> 32);
+ d->base0 = addr & 0xffff;
+ d->base1 = (addr >> 16) & 0xff;
+ d->base2 = (addr >> 24) & 0xff;
+}
+
+#define LDT_entry_a(info) \
+ ((((info)->base_addr & 0x0000ffff) << 16) | ((info)->limit & 0x0ffff))
+/* Don't allow setting of the lm bit. It is useless anyways because
+ 64bit system calls require __USER_CS. */
+#define LDT_entry_b(info) \
+ (((info)->base_addr & 0xff000000) | \
+ (((info)->base_addr & 0x00ff0000) >> 16) | \
+ ((info)->limit & 0xf0000) | \
+ (((info)->read_exec_only ^ 1) << 9) | \
+ ((info)->contents << 10) | \
+ (((info)->seg_not_present ^ 1) << 15) | \
+ ((info)->seg_32bit << 22) | \
+ ((info)->limit_in_pages << 23) | \
+ ((info)->useable << 20) | \
+ /* ((info)->lm << 21) | */ \
+ 0x7000)
+
+#define LDT_empty(info) (\
+ (info)->base_addr == 0 && \
+ (info)->limit == 0 && \
+ (info)->contents == 0 && \
+ (info)->read_exec_only == 1 && \
+ (info)->seg_32bit == 0 && \
+ (info)->limit_in_pages == 0 && \
+ (info)->seg_not_present == 1 && \
+ (info)->useable == 0 && \
+ (info)->lm == 0)
+
+#if TLS_SIZE != 24
+# error update this code.
+#endif
+
+static inline void load_TLS(struct thread_struct *t, unsigned int cpu)
+{
+#if 0
+ u64 *gdt = (u64 *)(cpu_gdt(cpu) + GDT_ENTRY_TLS_MIN);
+ gdt[0] = t->tls_array[0];
+ gdt[1] = t->tls_array[1];
+ gdt[2] = t->tls_array[2];
+#endif
+#define C(i) \
+ if (HYPERVISOR_update_descriptor(virt_to_machine(&cpu_gdt(cpu)[GDT_ENTRY_TLS_MIN + i]), \
+ t->tls_array[i])) \
+ BUG();
+
+ C(0); C(1); C(2);
+#undef C
+}
+
+/*
+ * load one particular LDT into the current CPU
+ */
+static inline void load_LDT_nolock (mm_context_t *pc, int cpu)
+{
+ void *segments = pc->ldt;
+ int count = pc->size;
+
+ if (likely(!count))
+ segments = NULL;
+
+ xen_set_ldt(segments, count);
+}
+
+static inline void load_LDT(mm_context_t *pc)
+{
+ int cpu = get_cpu();
+ load_LDT_nolock(pc, cpu);
+ put_cpu();
+}
+
+extern struct desc_ptr idt_descr;
+
+#endif /* !__ASSEMBLY__ */
+
+#endif
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/include/mach-xen/asm/fixmap_64.h 2007-06-12 13:14:13.000000000 +0200
@@ -0,0 +1,112 @@
+/*
+ * fixmap.h: compile-time virtual memory allocation
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License. See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 1998 Ingo Molnar
+ */
+
+#ifndef _ASM_FIXMAP_H
+#define _ASM_FIXMAP_H
+
+#include <linux/kernel.h>
+#include <asm/apicdef.h>
+#include <asm/page.h>
+#include <asm/vsyscall.h>
+#include <asm/vsyscall32.h>
+#include <asm/acpi.h>
+
+/*
+ * Here we define all the compile-time 'special' virtual
+ * addresses. The point is to have a constant address at
+ * compile time, but to set the physical address only
+ * in the boot process.
+ *
+ * these 'compile-time allocated' memory buffers are
+ * fixed-size 4k pages. (or larger if used with an increment
+ * highger than 1) use fixmap_set(idx,phys) to associate
+ * physical memory with fixmap indices.
+ *
+ * TLB entries of such buffers will not be flushed across
+ * task switches.
+ */
+
+enum fixed_addresses {
+ VSYSCALL_LAST_PAGE,
+ VSYSCALL_FIRST_PAGE = VSYSCALL_LAST_PAGE + ((VSYSCALL_END-VSYSCALL_START) >> PAGE_SHIFT) - 1,
+ VSYSCALL_HPET,
+ FIX_HPET_BASE,
+#ifdef CONFIG_X86_LOCAL_APIC
+ FIX_APIC_BASE, /* local (CPU) APIC) -- required for SMP or not */
+#endif
+#ifdef CONFIG_X86_IO_APIC
+ FIX_IO_APIC_BASE_0,
+ FIX_IO_APIC_BASE_END = FIX_IO_APIC_BASE_0 + MAX_IO_APICS-1,
+#endif
+#ifdef CONFIG_ACPI
+ FIX_ACPI_BEGIN,
+ FIX_ACPI_END = FIX_ACPI_BEGIN + FIX_ACPI_PAGES - 1,
+#endif
+ FIX_SHARED_INFO,
+#define NR_FIX_ISAMAPS 256
+ FIX_ISAMAP_END,
+ FIX_ISAMAP_BEGIN = FIX_ISAMAP_END + NR_FIX_ISAMAPS - 1,
+ __end_of_permanent_fixed_addresses,
+ /* temporary boot-time mappings, used before ioremap() is functional */
+#define NR_FIX_BTMAPS 16
+ FIX_BTMAP_END = __end_of_permanent_fixed_addresses,
+ FIX_BTMAP_BEGIN = FIX_BTMAP_END + NR_FIX_BTMAPS - 1,
+ __end_of_fixed_addresses
+};
+
+extern void __set_fixmap (enum fixed_addresses idx,
+ unsigned long phys, pgprot_t flags);
+
+#define set_fixmap(idx, phys) \
+ __set_fixmap(idx, phys, PAGE_KERNEL)
+/*
+ * Some hardware wants to get fixmapped without caching.
+ */
+#define set_fixmap_nocache(idx, phys) \
+ __set_fixmap(idx, phys, PAGE_KERNEL_NOCACHE)
+
+#define clear_fixmap(idx) \
+ __set_fixmap(idx, 0, __pgprot(0))
+
+#define FIXADDR_TOP (VSYSCALL_END-PAGE_SIZE)
+#define FIXADDR_SIZE (__end_of_fixed_addresses << PAGE_SHIFT)
+#define FIXADDR_START (FIXADDR_TOP - FIXADDR_SIZE)
+
+/* Only covers 32bit vsyscalls currently. Need another set for 64bit. */
+#define FIXADDR_USER_START ((unsigned long)VSYSCALL32_VSYSCALL)
+#define FIXADDR_USER_END (FIXADDR_USER_START + PAGE_SIZE)
+
+#define __fix_to_virt(x) (FIXADDR_TOP - ((x) << PAGE_SHIFT))
+
+extern void __this_fixmap_does_not_exist(void);
+
+/*
+ * 'index to address' translation. If anyone tries to use the idx
+ * directly without translation, we catch the bug with a NULL-deference
+ * kernel oops. Illegal ranges of incoming indices are caught too.
+ */
+static __always_inline unsigned long fix_to_virt(const unsigned int idx)
+{
+ /*
+ * this branch gets completely eliminated after inlining,
+ * except when someone tries to use fixaddr indices in an
+ * illegal way. (such as mixing up address types or using
+ * out-of-range indices).
+ *
+ * If it doesn't get removed, the linker will complain
+ * loudly with a reasonably clear error message..
+ */
+ if (idx >= __end_of_fixed_addresses)
+ __this_fixmap_does_not_exist();
+
+ return __fix_to_virt(idx);
+}
+
+#endif
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/include/mach-xen/asm/hypercall_64.h 2009-06-23 09:28:21.000000000 +0200
@@ -0,0 +1,422 @@
+/******************************************************************************
+ * hypercall.h
+ *
+ * Linux-specific hypervisor handling.
+ *
+ * Copyright (c) 2002-2004, K A Fraser
+ *
+ * 64-bit updates:
+ * Benjamin Liu <benjamin.liu@intel.com>
+ * Jun Nakajima <jun.nakajima@intel.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License version 2
+ * as published by the Free Software Foundation; or, when distributed
+ * separately from the Linux kernel or incorporated into other
+ * software packages, subject to the following license:
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this source file (the "Software"), to deal in the Software without
+ * restriction, including without limitation the rights to use, copy, modify,
+ * merge, publish, distribute, sublicense, and/or sell copies of the Software,
+ * and to permit persons to whom the Software is furnished to do so, subject to
+ * the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
+ * IN THE SOFTWARE.
+ */
+
+#ifndef __HYPERCALL_H__
+#define __HYPERCALL_H__
+
+#include <linux/string.h> /* memcpy() */
+#include <linux/stringify.h>
+#include <xen/interface/arch-x86/xen-mca.h>
+
+#ifndef __HYPERVISOR_H__
+# error "please don't include this file directly"
+#endif
+
+#ifdef CONFIG_XEN
+#define HYPERCALL_STR(name) \
+ "call hypercall_page + ("__stringify(__HYPERVISOR_##name)" * 32)"
+#else
+#define HYPERCALL_STR(name) \
+ "mov $("__stringify(__HYPERVISOR_##name)" * 32),%%eax; "\
+ "add hypercall_stubs(%%rip),%%rax; " \
+ "call *%%rax"
+#endif
+
+#define _hypercall0(type, name) \
+({ \
+ type __res; \
+ asm volatile ( \
+ HYPERCALL_STR(name) \
+ : "=a" (__res) \
+ : \
+ : "memory" ); \
+ __res; \
+})
+
+#define _hypercall1(type, name, a1) \
+({ \
+ type __res; \
+ long __ign1; \
+ asm volatile ( \
+ HYPERCALL_STR(name) \
+ : "=a" (__res), "=D" (__ign1) \
+ : "1" ((long)(a1)) \
+ : "memory" ); \
+ __res; \
+})
+
+#define _hypercall2(type, name, a1, a2) \
+({ \
+ type __res; \
+ long __ign1, __ign2; \
+ asm volatile ( \
+ HYPERCALL_STR(name) \
+ : "=a" (__res), "=D" (__ign1), "=S" (__ign2) \
+ : "1" ((long)(a1)), "2" ((long)(a2)) \
+ : "memory" ); \
+ __res; \
+})
+
+#define _hypercall3(type, name, a1, a2, a3) \
+({ \
+ type __res; \
+ long __ign1, __ign2, __ign3; \
+ asm volatile ( \
+ HYPERCALL_STR(name) \
+ : "=a" (__res), "=D" (__ign1), "=S" (__ign2), \
+ "=d" (__ign3) \
+ : "1" ((long)(a1)), "2" ((long)(a2)), \
+ "3" ((long)(a3)) \
+ : "memory" ); \
+ __res; \
+})
+
+#define _hypercall4(type, name, a1, a2, a3, a4) \
+({ \
+ type __res; \
+ long __ign1, __ign2, __ign3; \
+ register long __arg4 asm("r10") = (long)(a4); \
+ asm volatile ( \
+ HYPERCALL_STR(name) \
+ : "=a" (__res), "=D" (__ign1), "=S" (__ign2), \
+ "=d" (__ign3), "+r" (__arg4) \
+ : "1" ((long)(a1)), "2" ((long)(a2)), \
+ "3" ((long)(a3)) \
+ : "memory" ); \
+ __res; \
+})
+
+#define _hypercall5(type, name, a1, a2, a3, a4, a5) \
+({ \
+ type __res; \
+ long __ign1, __ign2, __ign3; \
+ register long __arg4 asm("r10") = (long)(a4); \
+ register long __arg5 asm("r8") = (long)(a5); \
+ asm volatile ( \
+ HYPERCALL_STR(name) \
+ : "=a" (__res), "=D" (__ign1), "=S" (__ign2), \
+ "=d" (__ign3), "+r" (__arg4), "+r" (__arg5) \
+ : "1" ((long)(a1)), "2" ((long)(a2)), \
+ "3" ((long)(a3)) \
+ : "memory" ); \
+ __res; \
+})
+
+static inline int __must_check
+HYPERVISOR_set_trap_table(
+ const trap_info_t *table)
+{
+ return _hypercall1(int, set_trap_table, table);
+}
+
+static inline int __must_check
+HYPERVISOR_mmu_update(
+ mmu_update_t *req, unsigned int count, unsigned int *success_count,
+ domid_t domid)
+{
+ return _hypercall4(int, mmu_update, req, count, success_count, domid);
+}
+
+static inline int __must_check
+HYPERVISOR_mmuext_op(
+ struct mmuext_op *op, unsigned int count, unsigned int *success_count,
+ domid_t domid)
+{
+ return _hypercall4(int, mmuext_op, op, count, success_count, domid);
+}
+
+static inline int __must_check
+HYPERVISOR_set_gdt(
+ unsigned long *frame_list, unsigned int entries)
+{
+ return _hypercall2(int, set_gdt, frame_list, entries);
+}
+
+static inline int __must_check
+HYPERVISOR_stack_switch(
+ unsigned long ss, unsigned long esp)
+{
+ return _hypercall2(int, stack_switch, ss, esp);
+}
+
+static inline int __must_check
+HYPERVISOR_set_callbacks(
+ unsigned long event_address, unsigned long failsafe_address,
+ unsigned long syscall_address)
+{
+ return _hypercall3(int, set_callbacks,
+ event_address, failsafe_address, syscall_address);
+}
+
+static inline int
+HYPERVISOR_fpu_taskswitch(
+ int set)
+{
+ return _hypercall1(int, fpu_taskswitch, set);
+}
+
+static inline int __must_check
+HYPERVISOR_sched_op_compat(
+ int cmd, unsigned long arg)
+{
+ return _hypercall2(int, sched_op_compat, cmd, arg);
+}
+
+static inline int __must_check
+HYPERVISOR_sched_op(
+ int cmd, void *arg)
+{
+ return _hypercall2(int, sched_op, cmd, arg);
+}
+
+static inline long __must_check
+HYPERVISOR_set_timer_op(
+ u64 timeout)
+{
+ return _hypercall1(long, set_timer_op, timeout);
+}
+
+static inline int __must_check
+HYPERVISOR_platform_op(
+ struct xen_platform_op *platform_op)
+{
+ platform_op->interface_version = XENPF_INTERFACE_VERSION;
+ return _hypercall1(int, platform_op, platform_op);
+}
+static inline int __must_check
+HYPERVISOR_mca(
+ struct xen_mc *mc_op)
+{
+ mc_op->interface_version = XEN_MCA_INTERFACE_VERSION;
+ return _hypercall1(int, mca, mc_op);
+}
+static inline int __must_check
+HYPERVISOR_set_debugreg(
+ unsigned int reg, unsigned long value)
+{
+ return _hypercall2(int, set_debugreg, reg, value);
+}
+
+static inline unsigned long __must_check
+HYPERVISOR_get_debugreg(
+ unsigned int reg)
+{
+ return _hypercall1(unsigned long, get_debugreg, reg);
+}
+
+static inline int __must_check
+HYPERVISOR_update_descriptor(
+ unsigned long ma, unsigned long word)
+{
+ return _hypercall2(int, update_descriptor, ma, word);
+}
+
+static inline int __must_check
+HYPERVISOR_memory_op(
+ unsigned int cmd, void *arg)
+{
+ return _hypercall2(int, memory_op, cmd, arg);
+}
+
+static inline int __must_check
+HYPERVISOR_multicall(
+ multicall_entry_t *call_list, unsigned int nr_calls)
+{
+ return _hypercall2(int, multicall, call_list, nr_calls);
+}
+
+static inline int __must_check
+HYPERVISOR_update_va_mapping(
+ unsigned long va, pte_t new_val, unsigned long flags)
+{
+ return _hypercall3(int, update_va_mapping, va, new_val.pte, flags);
+}
+
+static inline int __must_check
+HYPERVISOR_event_channel_op(
+ int cmd, void *arg)
+{
+ int rc = _hypercall2(int, event_channel_op, cmd, arg);
+
+#if CONFIG_XEN_COMPAT <= 0x030002
+ if (unlikely(rc == -ENOSYS)) {
+ struct evtchn_op op;
+ op.cmd = cmd;
+ memcpy(&op.u, arg, sizeof(op.u));
+ rc = _hypercall1(int, event_channel_op_compat, &op);
+ memcpy(arg, &op.u, sizeof(op.u));
+ }
+#endif
+
+ return rc;
+}
+
+static inline int __must_check
+HYPERVISOR_xen_version(
+ int cmd, void *arg)
+{
+ return _hypercall2(int, xen_version, cmd, arg);
+}
+
+static inline int __must_check
+HYPERVISOR_console_io(
+ int cmd, unsigned int count, char *str)
+{
+ return _hypercall3(int, console_io, cmd, count, str);
+}
+
+static inline int __must_check
+HYPERVISOR_physdev_op(
+ int cmd, void *arg)
+{
+ int rc = _hypercall2(int, physdev_op, cmd, arg);
+
+#if CONFIG_XEN_COMPAT <= 0x030002
+ if (unlikely(rc == -ENOSYS)) {
+ struct physdev_op op;
+ op.cmd = cmd;
+ memcpy(&op.u, arg, sizeof(op.u));
+ rc = _hypercall1(int, physdev_op_compat, &op);
+ memcpy(arg, &op.u, sizeof(op.u));
+ }
+#endif
+
+ return rc;
+}
+
+static inline int __must_check
+HYPERVISOR_grant_table_op(
+ unsigned int cmd, void *uop, unsigned int count)
+{
+ return _hypercall3(int, grant_table_op, cmd, uop, count);
+}
+
+static inline int __must_check
+HYPERVISOR_update_va_mapping_otherdomain(
+ unsigned long va, pte_t new_val, unsigned long flags, domid_t domid)
+{
+ return _hypercall4(int, update_va_mapping_otherdomain, va,
+ new_val.pte, flags, domid);
+}
+
+static inline int __must_check
+HYPERVISOR_vm_assist(
+ unsigned int cmd, unsigned int type)
+{
+ return _hypercall2(int, vm_assist, cmd, type);
+}
+
+static inline int __must_check
+HYPERVISOR_vcpu_op(
+ int cmd, unsigned int vcpuid, void *extra_args)
+{
+ return _hypercall3(int, vcpu_op, cmd, vcpuid, extra_args);
+}
+
+static inline int __must_check
+HYPERVISOR_set_segment_base(
+ int reg, unsigned long value)
+{
+ return _hypercall2(int, set_segment_base, reg, value);
+}
+
+static inline int __must_check
+HYPERVISOR_suspend(
+ unsigned long srec)
+{
+ struct sched_shutdown sched_shutdown = {
+ .reason = SHUTDOWN_suspend
+ };
+
+ int rc = _hypercall3(int, sched_op, SCHEDOP_shutdown,
+ &sched_shutdown, srec);
+
+#if CONFIG_XEN_COMPAT <= 0x030002
+ if (rc == -ENOSYS)
+ rc = _hypercall3(int, sched_op_compat, SCHEDOP_shutdown,
+ SHUTDOWN_suspend, srec);
+#endif
+
+ return rc;
+}
+
+#if CONFIG_XEN_COMPAT <= 0x030002
+static inline int
+HYPERVISOR_nmi_op(
+ unsigned long op, void *arg)
+{
+ return _hypercall2(int, nmi_op, op, arg);
+}
+#endif
+
+#ifndef CONFIG_XEN
+static inline unsigned long __must_check
+HYPERVISOR_hvm_op(
+ int op, void *arg)
+{
+ return _hypercall2(unsigned long, hvm_op, op, arg);
+}
+#endif
+
+static inline int __must_check
+HYPERVISOR_callback_op(
+ int cmd, const void *arg)
+{
+ return _hypercall2(int, callback_op, cmd, arg);
+}
+
+static inline int __must_check
+HYPERVISOR_xenoprof_op(
+ int op, void *arg)
+{
+ return _hypercall2(int, xenoprof_op, op, arg);
+}
+
+static inline int __must_check
+HYPERVISOR_kexec_op(
+ unsigned long op, void *args)
+{
+ return _hypercall2(int, kexec_op, op, args);
+}
+
+static inline int __must_check
+HYPERVISOR_tmem_op(
+ struct tmem_op *op)
+{
+ return _hypercall1(int, tmem_op, op);
+}
+
+#endif /* __HYPERCALL_H__ */
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/include/mach-xen/asm/irqflags_64.h 2007-06-12 13:14:13.000000000 +0200
@@ -0,0 +1,139 @@
+/*
+ * include/asm-x86_64/irqflags.h
+ *
+ * IRQ flags handling
+ *
+ * This file gets included from lowlevel asm headers too, to provide
+ * wrapped versions of the local_irq_*() APIs, based on the
+ * raw_local_irq_*() functions from the lowlevel headers.
+ */
+#ifndef _ASM_IRQFLAGS_H
+#define _ASM_IRQFLAGS_H
+
+#ifndef __ASSEMBLY__
+/*
+ * Interrupt control:
+ */
+
+/*
+ * The use of 'barrier' in the following reflects their use as local-lock
+ * operations. Reentrancy must be prevented (e.g., __cli()) /before/ following
+ * critical operations are executed. All critical operations must complete
+ * /before/ reentrancy is permitted (e.g., __sti()). Alpha architecture also
+ * includes these barriers, for example.
+ */
+
+#define __raw_local_save_flags() (current_vcpu_info()->evtchn_upcall_mask)
+
+#define raw_local_save_flags(flags) \
+ do { (flags) = __raw_local_save_flags(); } while (0)
+
+#define raw_local_irq_restore(x) \
+do { \
+ vcpu_info_t *_vcpu; \
+ barrier(); \
+ _vcpu = current_vcpu_info(); \
+ if ((_vcpu->evtchn_upcall_mask = (x)) == 0) { \
+ barrier(); /* unmask then check (avoid races) */ \
+ if ( unlikely(_vcpu->evtchn_upcall_pending) ) \
+ force_evtchn_callback(); \
+ } \
+} while (0)
+
+#ifdef CONFIG_X86_VSMP
+
+/*
+ * Interrupt control for the VSMP architecture:
+ */
+
+static inline void raw_local_irq_disable(void)
+{
+ unsigned long flags = __raw_local_save_flags();
+
+ raw_local_irq_restore((flags & ~(1 << 9)) | (1 << 18));
+}
+
+static inline void raw_local_irq_enable(void)
+{
+ unsigned long flags = __raw_local_save_flags();
+
+ raw_local_irq_restore((flags | (1 << 9)) & ~(1 << 18));
+}
+
+static inline int raw_irqs_disabled_flags(unsigned long flags)
+{
+ return !(flags & (1<<9)) || (flags & (1 << 18));
+}
+
+#else /* CONFIG_X86_VSMP */
+
+#define raw_local_irq_disable() \
+do { \
+ current_vcpu_info()->evtchn_upcall_mask = 1; \
+ barrier(); \
+} while (0)
+
+#define raw_local_irq_enable() \
+do { \
+ vcpu_info_t *_vcpu; \
+ barrier(); \
+ _vcpu = current_vcpu_info(); \
+ _vcpu->evtchn_upcall_mask = 0; \
+ barrier(); /* unmask then check (avoid races) */ \
+ if ( unlikely(_vcpu->evtchn_upcall_pending) ) \
+ force_evtchn_callback(); \
+} while (0)
+
+static inline int raw_irqs_disabled_flags(unsigned long flags)
+{
+ return (flags != 0);
+}
+
+#endif
+
+/*
+ * For spinlocks, etc.:
+ */
+
+#define __raw_local_irq_save() \
+({ \
+ unsigned long flags = __raw_local_save_flags(); \
+ \
+ raw_local_irq_disable(); \
+ \
+ flags; \
+})
+
+#define raw_local_irq_save(flags) \
+ do { (flags) = __raw_local_irq_save(); } while (0)
+
+#define raw_irqs_disabled() \
+({ \
+ unsigned long flags = __raw_local_save_flags(); \
+ \
+ raw_irqs_disabled_flags(flags); \
+})
+
+/*
+ * Used in the idle loop; sti takes one instruction cycle
+ * to complete:
+ */
+void raw_safe_halt(void);
+
+/*
+ * Used when interrupts are already enabled or to
+ * shutdown the processor:
+ */
+void halt(void);
+
+#else /* __ASSEMBLY__: */
+# ifdef CONFIG_TRACE_IRQFLAGS
+# define TRACE_IRQS_ON call trace_hardirqs_on_thunk
+# define TRACE_IRQS_OFF call trace_hardirqs_off_thunk
+# else
+# define TRACE_IRQS_ON
+# define TRACE_IRQS_OFF
+# endif
+#endif
+
+#endif
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/include/mach-xen/asm/maddr_64.h 2007-06-12 13:14:13.000000000 +0200
@@ -0,0 +1,161 @@
+#ifndef _X86_64_MADDR_H
+#define _X86_64_MADDR_H
+
+#include <xen/features.h>
+#include <xen/interface/xen.h>
+
+/**** MACHINE <-> PHYSICAL CONVERSION MACROS ****/
+#define INVALID_P2M_ENTRY (~0UL)
+#define FOREIGN_FRAME_BIT (1UL<<63)
+#define FOREIGN_FRAME(m) ((m) | FOREIGN_FRAME_BIT)
+
+/* Definitions for machine and pseudophysical addresses. */
+typedef unsigned long paddr_t;
+typedef unsigned long maddr_t;
+
+#ifdef CONFIG_XEN
+
+extern unsigned long *phys_to_machine_mapping;
+
+#undef machine_to_phys_mapping
+extern unsigned long *machine_to_phys_mapping;
+extern unsigned int machine_to_phys_order;
+
+static inline unsigned long pfn_to_mfn(unsigned long pfn)
+{
+ if (xen_feature(XENFEAT_auto_translated_physmap))
+ return pfn;
+ BUG_ON(end_pfn && pfn >= end_pfn);
+ return phys_to_machine_mapping[pfn] & ~FOREIGN_FRAME_BIT;
+}
+
+static inline int phys_to_machine_mapping_valid(unsigned long pfn)
+{
+ if (xen_feature(XENFEAT_auto_translated_physmap))
+ return 1;
+ BUG_ON(end_pfn && pfn >= end_pfn);
+ return (phys_to_machine_mapping[pfn] != INVALID_P2M_ENTRY);
+}
+
+static inline unsigned long mfn_to_pfn(unsigned long mfn)
+{
+ unsigned long pfn;
+
+ if (xen_feature(XENFEAT_auto_translated_physmap))
+ return mfn;
+
+ if (unlikely((mfn >> machine_to_phys_order) != 0))
+ return end_pfn;
+
+ /* The array access can fail (e.g., device space beyond end of RAM). */
+ asm (
+ "1: movq %1,%0\n"
+ "2:\n"
+ ".section .fixup,\"ax\"\n"
+ "3: movq %2,%0\n"
+ " jmp 2b\n"
+ ".previous\n"
+ ".section __ex_table,\"a\"\n"
+ " .align 8\n"
+ " .quad 1b,3b\n"
+ ".previous"
+ : "=r" (pfn)
+ : "m" (machine_to_phys_mapping[mfn]), "m" (end_pfn) );
+
+ return pfn;
+}
+
+/*
+ * We detect special mappings in one of two ways:
+ * 1. If the MFN is an I/O page then Xen will set the m2p entry
+ * to be outside our maximum possible pseudophys range.
+ * 2. If the MFN belongs to a different domain then we will certainly
+ * not have MFN in our p2m table. Conversely, if the page is ours,
+ * then we'll have p2m(m2p(MFN))==MFN.
+ * If we detect a special mapping then it doesn't have a 'struct page'.
+ * We force !pfn_valid() by returning an out-of-range pointer.
+ *
+ * NB. These checks require that, for any MFN that is not in our reservation,
+ * there is no PFN such that p2m(PFN) == MFN. Otherwise we can get confused if
+ * we are foreign-mapping the MFN, and the other domain as m2p(MFN) == PFN.
+ * Yikes! Various places must poke in INVALID_P2M_ENTRY for safety.
+ *
+ * NB2. When deliberately mapping foreign pages into the p2m table, you *must*
+ * use FOREIGN_FRAME(). This will cause pte_pfn() to choke on it, as we
+ * require. In all the cases we care about, the FOREIGN_FRAME bit is
+ * masked (e.g., pfn_to_mfn()) so behaviour there is correct.
+ */
+static inline unsigned long mfn_to_local_pfn(unsigned long mfn)
+{
+ unsigned long pfn = mfn_to_pfn(mfn);
+ if ((pfn < end_pfn)
+ && !xen_feature(XENFEAT_auto_translated_physmap)
+ && (phys_to_machine_mapping[pfn] != mfn))
+ return end_pfn; /* force !pfn_valid() */
+ return pfn;
+}
+
+static inline void set_phys_to_machine(unsigned long pfn, unsigned long mfn)
+{
+ BUG_ON(end_pfn && pfn >= end_pfn);
+ if (xen_feature(XENFEAT_auto_translated_physmap)) {
+ BUG_ON(pfn != mfn && mfn != INVALID_P2M_ENTRY);
+ return;
+ }
+ phys_to_machine_mapping[pfn] = mfn;
+}
+
+static inline maddr_t phys_to_machine(paddr_t phys)
+{
+ maddr_t machine = pfn_to_mfn(phys >> PAGE_SHIFT);
+ machine = (machine << PAGE_SHIFT) | (phys & ~PAGE_MASK);
+ return machine;
+}
+
+static inline paddr_t machine_to_phys(maddr_t machine)
+{
+ paddr_t phys = mfn_to_pfn(machine >> PAGE_SHIFT);
+ phys = (phys << PAGE_SHIFT) | (machine & ~PAGE_MASK);
+ return phys;
+}
+
+static inline paddr_t pte_phys_to_machine(paddr_t phys)
+{
+ maddr_t machine;
+ machine = pfn_to_mfn((phys & PHYSICAL_PAGE_MASK) >> PAGE_SHIFT);
+ machine = (machine << PAGE_SHIFT) | (phys & ~PHYSICAL_PAGE_MASK);
+ return machine;
+}
+
+static inline paddr_t pte_machine_to_phys(maddr_t machine)
+{
+ paddr_t phys;
+ phys = mfn_to_pfn((machine & PHYSICAL_PAGE_MASK) >> PAGE_SHIFT);
+ phys = (phys << PAGE_SHIFT) | (machine & ~PHYSICAL_PAGE_MASK);
+ return phys;
+}
+
+#define __pte_ma(x) ((pte_t) { (x) } )
+#define pfn_pte_ma(pfn, prot) __pte_ma((((pfn) << PAGE_SHIFT) | pgprot_val(prot)) & __supported_pte_mask)
+
+#else /* !CONFIG_XEN */
+
+#define pfn_to_mfn(pfn) (pfn)
+#define mfn_to_pfn(mfn) (mfn)
+#define mfn_to_local_pfn(mfn) (mfn)
+#define set_phys_to_machine(pfn, mfn) ((void)0)
+#define phys_to_machine_mapping_valid(pfn) (1)
+#define phys_to_machine(phys) ((maddr_t)(phys))
+#define machine_to_phys(mach) ((paddr_t)(mach))
+#define pfn_pte_ma(pfn, prot) pfn_pte(pfn, prot)
+#define __pte_ma(x) __pte(x)
+
+#endif /* !CONFIG_XEN */
+
+/* VIRT <-> MACHINE conversion */
+#define virt_to_machine(v) (phys_to_machine(__pa(v)))
+#define virt_to_mfn(v) (pfn_to_mfn(__pa(v) >> PAGE_SHIFT))
+#define mfn_to_virt(m) (__va(mfn_to_pfn(m) << PAGE_SHIFT))
+
+#endif /* _X86_64_MADDR_H */
+
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/include/mach-xen/asm/mmu_context_64.h 2007-06-12 13:14:13.000000000 +0200
@@ -0,0 +1,136 @@
+#ifndef __X86_64_MMU_CONTEXT_H
+#define __X86_64_MMU_CONTEXT_H
+
+#include <asm/desc.h>
+#include <asm/atomic.h>
+#include <asm/pgalloc.h>
+#include <asm/page.h>
+#include <asm/pda.h>
+#include <asm/pgtable.h>
+#include <asm/tlbflush.h>
+
+/*
+ * possibly do the LDT unload here?
+ */
+int init_new_context(struct task_struct *tsk, struct mm_struct *mm);
+void destroy_context(struct mm_struct *mm);
+
+static inline void enter_lazy_tlb(struct mm_struct *mm, struct task_struct *tsk)
+{
+#if defined(CONFIG_SMP) && !defined(CONFIG_XEN)
+ if (read_pda(mmu_state) == TLBSTATE_OK)
+ write_pda(mmu_state, TLBSTATE_LAZY);
+#endif
+}
+
+#define prepare_arch_switch(next) __prepare_arch_switch()
+
+static inline void __prepare_arch_switch(void)
+{
+ /*
+ * Save away %es, %ds, %fs and %gs. Must happen before reload
+ * of cr3/ldt (i.e., not in __switch_to).
+ */
+ __asm__ __volatile__ (
+ "mov %%es,%0 ; mov %%ds,%1 ; mov %%fs,%2 ; mov %%gs,%3"
+ : "=m" (current->thread.es),
+ "=m" (current->thread.ds),
+ "=m" (current->thread.fsindex),
+ "=m" (current->thread.gsindex) );
+
+ if (current->thread.ds)
+ __asm__ __volatile__ ( "movl %0,%%ds" : : "r" (0) );
+
+ if (current->thread.es)
+ __asm__ __volatile__ ( "movl %0,%%es" : : "r" (0) );
+
+ if (current->thread.fsindex) {
+ __asm__ __volatile__ ( "movl %0,%%fs" : : "r" (0) );
+ current->thread.fs = 0;
+ }
+
+ if (current->thread.gsindex) {
+ load_gs_index(0);
+ current->thread.gs = 0;
+ }
+}
+
+extern void mm_pin(struct mm_struct *mm);
+extern void mm_unpin(struct mm_struct *mm);
+void mm_pin_all(void);
+
+static inline void load_cr3(pgd_t *pgd)
+{
+ asm volatile("movq %0,%%cr3" :: "r" (phys_to_machine(__pa(pgd))) :
+ "memory");
+}
+
+static inline void switch_mm(struct mm_struct *prev, struct mm_struct *next,
+ struct task_struct *tsk)
+{
+ unsigned cpu = smp_processor_id();
+ struct mmuext_op _op[3], *op = _op;
+
+ if (likely(prev != next)) {
+ BUG_ON(!xen_feature(XENFEAT_writable_page_tables) &&
+ !next->context.pinned);
+
+ /* stop flush ipis for the previous mm */
+ cpu_clear(cpu, prev->cpu_vm_mask);
+#if defined(CONFIG_SMP) && !defined(CONFIG_XEN)
+ write_pda(mmu_state, TLBSTATE_OK);
+ write_pda(active_mm, next);
+#endif
+ cpu_set(cpu, next->cpu_vm_mask);
+
+ /* load_cr3(next->pgd) */
+ op->cmd = MMUEXT_NEW_BASEPTR;
+ op->arg1.mfn = pfn_to_mfn(__pa(next->pgd) >> PAGE_SHIFT);
+ op++;
+
+ /* xen_new_user_pt(__pa(__user_pgd(next->pgd))) */
+ op->cmd = MMUEXT_NEW_USER_BASEPTR;
+ op->arg1.mfn = pfn_to_mfn(__pa(__user_pgd(next->pgd)) >> PAGE_SHIFT);
+ op++;
+
+ if (unlikely(next->context.ldt != prev->context.ldt)) {
+ /* load_LDT_nolock(&next->context, cpu) */
+ op->cmd = MMUEXT_SET_LDT;
+ op->arg1.linear_addr = (unsigned long)next->context.ldt;
+ op->arg2.nr_ents = next->context.size;
+ op++;
+ }
+
+ BUG_ON(HYPERVISOR_mmuext_op(_op, op-_op, NULL, DOMID_SELF));
+ }
+#if defined(CONFIG_SMP) && !defined(CONFIG_XEN)
+ else {
+ write_pda(mmu_state, TLBSTATE_OK);
+ if (read_pda(active_mm) != next)
+ out_of_line_bug();
+ if (!cpu_test_and_set(cpu, next->cpu_vm_mask)) {
+ /* We were in lazy tlb mode and leave_mm disabled
+ * tlb flush IPI delivery. We must reload CR3
+ * to make sure to use no freed page tables.
+ */
+ load_cr3(next->pgd);
+ xen_new_user_pt(__pa(__user_pgd(next->pgd)));
+ load_LDT_nolock(&next->context, cpu);
+ }
+ }
+#endif
+}
+
+#define deactivate_mm(tsk,mm) do { \
+ load_gs_index(0); \
+ asm volatile("movl %0,%%fs"::"r"(0)); \
+} while(0)
+
+static inline void activate_mm(struct mm_struct *prev, struct mm_struct *next)
+{
+ if (!next->context.pinned)
+ mm_pin(next);
+ switch_mm(prev, next, NULL);
+}
+
+#endif
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/include/mach-xen/asm/pgalloc_64.h 2010-09-23 15:39:04.000000000 +0200
@@ -0,0 +1,210 @@
+#ifndef _X86_64_PGALLOC_H
+#define _X86_64_PGALLOC_H
+
+#include <asm/fixmap.h>
+#include <asm/pda.h>
+#include <linux/threads.h>
+#include <linux/mm.h>
+#include <asm/io.h> /* for phys_to_virt and page_to_pseudophys */
+
+#include <xen/features.h>
+void make_page_readonly(void *va, unsigned int feature);
+void make_page_writable(void *va, unsigned int feature);
+void make_pages_readonly(void *va, unsigned int nr, unsigned int feature);
+void make_pages_writable(void *va, unsigned int nr, unsigned int feature);
+
+#define __user_pgd(pgd) ((pgd) + PTRS_PER_PGD)
+
+static inline void pmd_populate_kernel(struct mm_struct *mm, pmd_t *pmd, pte_t *pte)
+{
+ set_pmd(pmd, __pmd(_PAGE_TABLE | __pa(pte)));
+}
+
+static inline void pmd_populate(struct mm_struct *mm, pmd_t *pmd, struct page *pte)
+{
+ if (unlikely((mm)->context.pinned)) {
+ BUG_ON(HYPERVISOR_update_va_mapping(
+ (unsigned long)__va(page_to_pfn(pte) << PAGE_SHIFT),
+ pfn_pte(page_to_pfn(pte), PAGE_KERNEL_RO), 0));
+ set_pmd(pmd, __pmd(_PAGE_TABLE | (page_to_pfn(pte) << PAGE_SHIFT)));
+ } else {
+ *(pmd) = __pmd(_PAGE_TABLE | (page_to_pfn(pte) << PAGE_SHIFT));
+ }
+}
+
+static inline void pud_populate(struct mm_struct *mm, pud_t *pud, pmd_t *pmd)
+{
+ if (unlikely((mm)->context.pinned)) {
+ BUG_ON(HYPERVISOR_update_va_mapping(
+ (unsigned long)pmd,
+ pfn_pte(virt_to_phys(pmd)>>PAGE_SHIFT,
+ PAGE_KERNEL_RO), 0));
+ set_pud(pud, __pud(_PAGE_TABLE | __pa(pmd)));
+ } else {
+ *(pud) = __pud(_PAGE_TABLE | __pa(pmd));
+ }
+}
+
+/*
+ * We need to use the batch mode here, but pgd_pupulate() won't be
+ * be called frequently.
+ */
+static inline void pgd_populate(struct mm_struct *mm, pgd_t *pgd, pud_t *pud)
+{
+ if (unlikely((mm)->context.pinned)) {
+ BUG_ON(HYPERVISOR_update_va_mapping(
+ (unsigned long)pud,
+ pfn_pte(virt_to_phys(pud)>>PAGE_SHIFT,
+ PAGE_KERNEL_RO), 0));
+ set_pgd(pgd, __pgd(_PAGE_TABLE | __pa(pud)));
+ set_pgd(__user_pgd(pgd), __pgd(_PAGE_TABLE | __pa(pud)));
+ } else {
+ *(pgd) = __pgd(_PAGE_TABLE | __pa(pud));
+ *(__user_pgd(pgd)) = *(pgd);
+ }
+}
+
+extern struct page *pte_alloc_one(struct mm_struct *mm, unsigned long addr);
+extern void pte_free(struct page *pte);
+
+static inline pmd_t *pmd_alloc_one(struct mm_struct *mm, unsigned long addr)
+{
+ struct page *pg;
+
+ pg = pte_alloc_one(mm, addr);
+ return pg ? page_address(pg) : NULL;
+}
+
+static inline void pmd_free(pmd_t *pmd)
+{
+ BUG_ON((unsigned long)pmd & (PAGE_SIZE-1));
+ pte_free(virt_to_page(pmd));
+}
+
+static inline pud_t *pud_alloc_one(struct mm_struct *mm, unsigned long addr)
+{
+ struct page *pg;
+
+ pg = pte_alloc_one(mm, addr);
+ return pg ? page_address(pg) : NULL;
+}
+
+static inline void pud_free(pud_t *pud)
+{
+ BUG_ON((unsigned long)pud & (PAGE_SIZE-1));
+ pte_free(virt_to_page(pud));
+}
+
+static inline void pgd_list_add(pgd_t *pgd, void *mm)
+{
+ struct page *page = virt_to_page(pgd);
+
+ /* Store a back link for vmalloc_sync_all(). */
+ page->mapping = mm;
+
+ spin_lock(&pgd_lock);
+ page->index = (pgoff_t)pgd_list;
+ if (pgd_list)
+ pgd_list->private = (unsigned long)&page->index;
+ pgd_list = page;
+ page->private = (unsigned long)&pgd_list;
+ spin_unlock(&pgd_lock);
+}
+
+static inline void pgd_list_del(pgd_t *pgd)
+{
+ struct page *next, **pprev, *page = virt_to_page(pgd);
+
+ spin_lock(&pgd_lock);
+ next = (struct page *)page->index;
+ pprev = (struct page **)page->private;
+ *pprev = next;
+ if (next)
+ next->private = (unsigned long)pprev;
+ spin_unlock(&pgd_lock);
+
+ page->mapping = NULL;
+}
+
+static inline pgd_t *pgd_alloc(struct mm_struct *mm)
+{
+ /*
+ * We allocate two contiguous pages for kernel and user.
+ */
+ unsigned boundary;
+ pgd_t *pgd;
+
+ pgd = (pgd_t *)__get_free_pages(GFP_KERNEL|__GFP_REPEAT|__GFP_ZERO, 1);
+ if (!pgd)
+ return NULL;
+ pgd_list_add(pgd, mm);
+ /*
+ * Copy kernel pointers in from init.
+ * Could keep a freelist or slab cache of those because the kernel
+ * part never changes.
+ */
+ boundary = pgd_index(__PAGE_OFFSET);
+ memcpy(pgd + boundary,
+ init_level4_pgt + boundary,
+ (PTRS_PER_PGD - boundary) * sizeof(pgd_t));
+
+ /*
+ * Set level3_user_pgt for vsyscall area
+ */
+ __user_pgd(pgd)[pgd_index(VSYSCALL_START)] =
+ __pgd(__pa_symbol(level3_user_pgt) | _PAGE_TABLE);
+ return pgd;
+}
+
+static inline void pgd_free(pgd_t *pgd)
+{
+ pte_t *ptep = virt_to_ptep(pgd);
+
+ pgd_list_del(pgd);
+
+ if (!pte_write(*ptep)) {
+ xen_pgd_unpin(__pa(pgd));
+ BUG_ON(HYPERVISOR_update_va_mapping(
+ (unsigned long)pgd,
+ pfn_pte(virt_to_phys(pgd)>>PAGE_SHIFT, PAGE_KERNEL),
+ 0));
+ }
+
+ ptep = virt_to_ptep(__user_pgd(pgd));
+
+ if (!pte_write(*ptep)) {
+ xen_pgd_unpin(__pa(__user_pgd(pgd)));
+ BUG_ON(HYPERVISOR_update_va_mapping(
+ (unsigned long)__user_pgd(pgd),
+ pfn_pte(virt_to_phys(__user_pgd(pgd))>>PAGE_SHIFT,
+ PAGE_KERNEL),
+ 0));
+ }
+
+ free_pages((unsigned long)pgd, 1);
+}
+
+static inline pte_t *pte_alloc_one_kernel(struct mm_struct *mm, unsigned long address)
+{
+ pte_t *pte = (pte_t *)get_zeroed_page(GFP_KERNEL|__GFP_REPEAT);
+ if (pte)
+ make_page_readonly(pte, XENFEAT_writable_page_tables);
+
+ return pte;
+}
+
+/* Should really implement gc for free page table pages. This could be
+ done with a reference count in struct page. */
+
+static inline void pte_free_kernel(pte_t *pte)
+{
+ BUG_ON((unsigned long)pte & (PAGE_SIZE-1));
+ make_page_writable(pte, XENFEAT_writable_page_tables);
+ free_page((unsigned long)pte);
+}
+
+#define __pte_free_tlb(tlb,pte) tlb_remove_page((tlb),(pte))
+#define __pmd_free_tlb(tlb,x) tlb_remove_page((tlb),virt_to_page(x))
+#define __pud_free_tlb(tlb,x) tlb_remove_page((tlb),virt_to_page(x))
+
+#endif /* _X86_64_PGALLOC_H */
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/include/mach-xen/asm/pgtable_64.h 2009-06-23 09:28:21.000000000 +0200
@@ -0,0 +1,585 @@
+#ifndef _X86_64_PGTABLE_H
+#define _X86_64_PGTABLE_H
+
+/*
+ * This file contains the functions and defines necessary to modify and use
+ * the x86-64 page table tree.
+ */
+#include <asm/processor.h>
+#include <asm/fixmap.h>
+#include <asm/bitops.h>
+#include <linux/threads.h>
+#include <linux/sched.h>
+#include <asm/pda.h>
+#ifdef CONFIG_XEN
+#include <asm/hypervisor.h>
+
+extern pud_t level3_user_pgt[512];
+
+extern void xen_init_pt(void);
+
+extern pte_t *lookup_address(unsigned long address);
+
+#define virt_to_ptep(va) \
+({ \
+ pte_t *__ptep = lookup_address((unsigned long)(va)); \
+ BUG_ON(!__ptep || !pte_present(*__ptep)); \
+ __ptep; \
+})
+
+#define arbitrary_virt_to_machine(va) \
+ (((maddr_t)pte_mfn(*virt_to_ptep(va)) << PAGE_SHIFT) \
+ | ((unsigned long)(va) & (PAGE_SIZE - 1)))
+
+#define ptep_to_machine(ptep) virt_to_machine(ptep)
+#endif
+
+extern pud_t level3_kernel_pgt[512];
+extern pud_t level3_physmem_pgt[512];
+extern pud_t level3_ident_pgt[512];
+extern pmd_t level2_kernel_pgt[512];
+extern pgd_t init_level4_pgt[];
+extern pgd_t boot_level4_pgt[];
+extern unsigned long __supported_pte_mask;
+
+#define swapper_pg_dir init_level4_pgt
+
+extern int nonx_setup(char *str);
+extern void paging_init(void);
+extern void clear_kernel_mapping(unsigned long addr, unsigned long size);
+
+extern unsigned long pgkern_mask;
+
+/*
+ * ZERO_PAGE is a global shared page that is always zero: used
+ * for zero-mapped memory areas etc..
+ */
+extern unsigned long empty_zero_page[PAGE_SIZE/sizeof(unsigned long)];
+#define ZERO_PAGE(vaddr) (virt_to_page(empty_zero_page))
+
+/*
+ * PGDIR_SHIFT determines what a top-level page table entry can map
+ */
+#define PGDIR_SHIFT 39
+#define PTRS_PER_PGD 512
+
+/*
+ * 3rd level page
+ */
+#define PUD_SHIFT 30
+#define PTRS_PER_PUD 512
+
+/*
+ * PMD_SHIFT determines the size of the area a middle-level
+ * page table can map
+ */
+#define PMD_SHIFT 21
+#define PTRS_PER_PMD 512
+
+/*
+ * entries per page directory level
+ */
+#define PTRS_PER_PTE 512
+
+#define pte_ERROR(e) \
+ printk("%s:%d: bad pte %p(%016lx pfn %010lx).\n", __FILE__, __LINE__, \
+ &(e), __pte_val(e), pte_pfn(e))
+#define pmd_ERROR(e) \
+ printk("%s:%d: bad pmd %p(%016lx pfn %010lx).\n", __FILE__, __LINE__, \
+ &(e), __pmd_val(e), pmd_pfn(e))
+#define pud_ERROR(e) \
+ printk("%s:%d: bad pud %p(%016lx pfn %010lx).\n", __FILE__, __LINE__, \
+ &(e), __pud_val(e), (pud_val(e) & __PHYSICAL_MASK) >> PAGE_SHIFT)
+#define pgd_ERROR(e) \
+ printk("%s:%d: bad pgd %p(%016lx pfn %010lx).\n", __FILE__, __LINE__, \
+ &(e), __pgd_val(e), (pgd_val(e) & __PHYSICAL_MASK) >> PAGE_SHIFT)
+
+#define pgd_none(x) (!__pgd_val(x))
+#define pud_none(x) (!__pud_val(x))
+
+static inline void set_pte(pte_t *dst, pte_t val)
+{
+ *dst = val;
+}
+
+#define set_pmd(pmdptr, pmdval) xen_l2_entry_update(pmdptr, (pmdval))
+#define set_pud(pudptr, pudval) xen_l3_entry_update(pudptr, (pudval))
+#define set_pgd(pgdptr, pgdval) xen_l4_entry_update(pgdptr, (pgdval))
+
+static inline void pud_clear (pud_t * pud)
+{
+ set_pud(pud, __pud(0));
+}
+
+#define __user_pgd(pgd) ((pgd) + PTRS_PER_PGD)
+
+static inline void pgd_clear (pgd_t * pgd)
+{
+ set_pgd(pgd, __pgd(0));
+ set_pgd(__user_pgd(pgd), __pgd(0));
+}
+
+#define pud_page(pud) \
+ ((unsigned long) __va(pud_val(pud) & PHYSICAL_PAGE_MASK))
+
+#define pte_same(a, b) ((a).pte == (b).pte)
+
+#define pte_pgprot(a) (__pgprot((a).pte & ~PHYSICAL_PAGE_MASK))
+
+#define PMD_SIZE (1UL << PMD_SHIFT)
+#define PMD_MASK (~(PMD_SIZE-1))
+#define PUD_SIZE (1UL << PUD_SHIFT)
+#define PUD_MASK (~(PUD_SIZE-1))
+#define PGDIR_SIZE (1UL << PGDIR_SHIFT)
+#define PGDIR_MASK (~(PGDIR_SIZE-1))
+
+#define USER_PTRS_PER_PGD ((TASK_SIZE-1)/PGDIR_SIZE+1)
+#define FIRST_USER_ADDRESS 0
+
+#ifndef __ASSEMBLY__
+#define MAXMEM 0x6fffffffffUL
+#define VMALLOC_START 0xffffc20000000000UL
+#define VMALLOC_END 0xffffe1ffffffffffUL
+#define MODULES_VADDR 0xffffffff88000000UL
+#define MODULES_END 0xffffffffff000000UL
+#define MODULES_LEN (MODULES_END - MODULES_VADDR)
+
+#define _PAGE_BIT_PRESENT 0
+#define _PAGE_BIT_RW 1
+#define _PAGE_BIT_USER 2
+#define _PAGE_BIT_PWT 3
+#define _PAGE_BIT_PCD 4
+#define _PAGE_BIT_ACCESSED 5
+#define _PAGE_BIT_DIRTY 6
+#define _PAGE_BIT_PSE 7 /* 4 MB (or 2MB) page */
+#define _PAGE_BIT_GLOBAL 8 /* Global TLB entry PPro+ */
+#define _PAGE_BIT_NX 63 /* No execute: only valid after cpuid check */
+
+#define _PAGE_PRESENT 0x001
+#define _PAGE_RW 0x002
+#define _PAGE_USER 0x004
+#define _PAGE_PWT 0x008
+#define _PAGE_PCD 0x010
+#define _PAGE_ACCESSED 0x020
+#define _PAGE_DIRTY 0x040
+#define _PAGE_PSE 0x080 /* 2MB page */
+#define _PAGE_FILE 0x040 /* nonlinear file mapping, saved PTE; unset:swap */
+#define _PAGE_GLOBAL 0x100 /* Global TLB entry */
+
+#define _PAGE_PROTNONE 0x080 /* If not present */
+#define _PAGE_NX (1UL<<_PAGE_BIT_NX)
+
+/* Mapped page is I/O or foreign and has no associated page struct. */
+#define _PAGE_IO 0x200
+
+#if CONFIG_XEN_COMPAT <= 0x030002
+extern unsigned int __kernel_page_user;
+#else
+#define __kernel_page_user 0
+#endif
+
+#define _PAGE_TABLE (_PAGE_PRESENT | _PAGE_RW | _PAGE_USER | _PAGE_ACCESSED | _PAGE_DIRTY)
+#define _KERNPG_TABLE (_PAGE_PRESENT | _PAGE_RW | _PAGE_ACCESSED | _PAGE_DIRTY | __kernel_page_user)
+
+#define _PAGE_CHG_MASK (PTE_MASK | _PAGE_ACCESSED | _PAGE_DIRTY | _PAGE_IO)
+
+#define PAGE_NONE __pgprot(_PAGE_PROTNONE | _PAGE_ACCESSED)
+#define PAGE_SHARED __pgprot(_PAGE_PRESENT | _PAGE_RW | _PAGE_USER | _PAGE_ACCESSED | _PAGE_NX)
+#define PAGE_SHARED_EXEC __pgprot(_PAGE_PRESENT | _PAGE_RW | _PAGE_USER | _PAGE_ACCESSED)
+#define PAGE_COPY_NOEXEC __pgprot(_PAGE_PRESENT | _PAGE_USER | _PAGE_ACCESSED | _PAGE_NX)
+#define PAGE_COPY PAGE_COPY_NOEXEC
+#define PAGE_COPY_EXEC __pgprot(_PAGE_PRESENT | _PAGE_USER | _PAGE_ACCESSED)
+#define PAGE_READONLY __pgprot(_PAGE_PRESENT | _PAGE_USER | _PAGE_ACCESSED | _PAGE_NX)
+#define PAGE_READONLY_EXEC __pgprot(_PAGE_PRESENT | _PAGE_USER | _PAGE_ACCESSED)
+#define __PAGE_KERNEL \
+ (_PAGE_PRESENT | _PAGE_RW | _PAGE_DIRTY | _PAGE_ACCESSED | _PAGE_NX | __kernel_page_user)
+#define __PAGE_KERNEL_EXEC \
+ (_PAGE_PRESENT | _PAGE_RW | _PAGE_DIRTY | _PAGE_ACCESSED | __kernel_page_user)
+#define __PAGE_KERNEL_NOCACHE \
+ (_PAGE_PRESENT | _PAGE_RW | _PAGE_DIRTY | _PAGE_PCD | _PAGE_ACCESSED | _PAGE_NX | __kernel_page_user)
+#define __PAGE_KERNEL_RO \
+ (_PAGE_PRESENT | _PAGE_DIRTY | _PAGE_ACCESSED | _PAGE_NX | __kernel_page_user)
+#define __PAGE_KERNEL_VSYSCALL \
+ (_PAGE_PRESENT | _PAGE_USER | _PAGE_ACCESSED)
+#define __PAGE_KERNEL_VSYSCALL_NOCACHE \
+ (_PAGE_PRESENT | _PAGE_USER | _PAGE_ACCESSED | _PAGE_PCD)
+#define __PAGE_KERNEL_LARGE \
+ (__PAGE_KERNEL | _PAGE_PSE)
+#define __PAGE_KERNEL_LARGE_EXEC \
+ (__PAGE_KERNEL_EXEC | _PAGE_PSE)
+
+/*
+ * We don't support GLOBAL page in xenolinux64
+ */
+#define MAKE_GLOBAL(x) __pgprot((x))
+
+#define PAGE_KERNEL MAKE_GLOBAL(__PAGE_KERNEL)
+#define PAGE_KERNEL_EXEC MAKE_GLOBAL(__PAGE_KERNEL_EXEC)
+#define PAGE_KERNEL_RO MAKE_GLOBAL(__PAGE_KERNEL_RO)
+#define PAGE_KERNEL_NOCACHE MAKE_GLOBAL(__PAGE_KERNEL_NOCACHE)
+#define PAGE_KERNEL_VSYSCALL32 __pgprot(__PAGE_KERNEL_VSYSCALL)
+#define PAGE_KERNEL_VSYSCALL MAKE_GLOBAL(__PAGE_KERNEL_VSYSCALL)
+#define PAGE_KERNEL_LARGE MAKE_GLOBAL(__PAGE_KERNEL_LARGE)
+#define PAGE_KERNEL_VSYSCALL_NOCACHE MAKE_GLOBAL(__PAGE_KERNEL_VSYSCALL_NOCACHE)
+
+/* xwr */
+#define __P000 PAGE_NONE
+#define __P001 PAGE_READONLY
+#define __P010 PAGE_COPY
+#define __P011 PAGE_COPY
+#define __P100 PAGE_READONLY_EXEC
+#define __P101 PAGE_READONLY_EXEC
+#define __P110 PAGE_COPY_EXEC
+#define __P111 PAGE_COPY_EXEC
+
+#define __S000 PAGE_NONE
+#define __S001 PAGE_READONLY
+#define __S010 PAGE_SHARED
+#define __S011 PAGE_SHARED
+#define __S100 PAGE_READONLY_EXEC
+#define __S101 PAGE_READONLY_EXEC
+#define __S110 PAGE_SHARED_EXEC
+#define __S111 PAGE_SHARED_EXEC
+
+static inline unsigned long pgd_bad(pgd_t pgd)
+{
+ unsigned long val = __pgd_val(pgd);
+ val &= ~PTE_MASK;
+ val &= ~(_PAGE_USER | _PAGE_DIRTY);
+ return val & ~(_PAGE_PRESENT | _PAGE_RW | _PAGE_ACCESSED);
+}
+
+static inline unsigned long pud_bad(pud_t pud)
+{
+ unsigned long val = __pud_val(pud);
+ val &= ~PTE_MASK;
+ val &= ~(_PAGE_USER | _PAGE_DIRTY);
+ return val & ~(_PAGE_PRESENT | _PAGE_RW | _PAGE_ACCESSED);
+}
+
+#define set_pte_at(_mm,addr,ptep,pteval) do { \
+ if (((_mm) != current->mm && (_mm) != &init_mm) || \
+ HYPERVISOR_update_va_mapping((addr), (pteval), 0)) \
+ set_pte((ptep), (pteval)); \
+} while (0)
+
+#define pte_none(x) (!(x).pte)
+#define pte_present(x) ((x).pte & (_PAGE_PRESENT | _PAGE_PROTNONE))
+#define pte_clear(mm,addr,xp) do { set_pte_at(mm, addr, xp, __pte(0)); } while (0)
+
+#define pages_to_mb(x) ((x) >> (20-PAGE_SHIFT))
+
+#define __pte_mfn(_pte) (((_pte).pte & PTE_MASK) >> PAGE_SHIFT)
+#define pte_mfn(_pte) ((_pte).pte & _PAGE_PRESENT ? \
+ __pte_mfn(_pte) : pfn_to_mfn(__pte_mfn(_pte)))
+#define pte_pfn(_pte) ((_pte).pte & _PAGE_IO ? end_pfn : \
+ (_pte).pte & _PAGE_PRESENT ? \
+ mfn_to_local_pfn(__pte_mfn(_pte)) : \
+ __pte_mfn(_pte))
+
+#define pte_page(x) pfn_to_page(pte_pfn(x))
+
+static inline pte_t pfn_pte(unsigned long page_nr, pgprot_t pgprot)
+{
+ unsigned long pte = page_nr << PAGE_SHIFT;
+ pte |= pgprot_val(pgprot);
+ pte &= __supported_pte_mask;
+ return __pte(pte);
+}
+
+static inline pte_t ptep_get_and_clear(struct mm_struct *mm, unsigned long addr, pte_t *ptep)
+{
+ pte_t pte = *ptep;
+ if (!pte_none(pte)) {
+ if ((mm != &init_mm) ||
+ HYPERVISOR_update_va_mapping(addr, __pte(0), 0))
+ pte = __pte_ma(xchg(&ptep->pte, 0));
+ }
+ return pte;
+}
+
+static inline pte_t ptep_get_and_clear_full(struct mm_struct *mm, unsigned long addr, pte_t *ptep, int full)
+{
+ if (full) {
+ pte_t pte = *ptep;
+ if (mm->context.pinned)
+ xen_l1_entry_update(ptep, __pte(0));
+ else
+ *ptep = __pte(0);
+ return pte;
+ }
+ return ptep_get_and_clear(mm, addr, ptep);
+}
+
+#define ptep_clear_flush(vma, addr, ptep) \
+({ \
+ pte_t *__ptep = (ptep); \
+ pte_t __res = *__ptep; \
+ if (!pte_none(__res) && \
+ ((vma)->vm_mm != current->mm || \
+ HYPERVISOR_update_va_mapping(addr, __pte(0), \
+ (unsigned long)(vma)->vm_mm->cpu_vm_mask.bits| \
+ UVMF_INVLPG|UVMF_MULTI))) { \
+ __ptep->pte = 0; \
+ flush_tlb_page(vma, addr); \
+ } \
+ __res; \
+})
+
+/*
+ * The following only work if pte_present() is true.
+ * Undefined behaviour if not..
+ */
+#define __LARGE_PTE (_PAGE_PSE|_PAGE_PRESENT)
+static inline int pte_user(pte_t pte) { return __pte_val(pte) & _PAGE_USER; }
+static inline int pte_read(pte_t pte) { return __pte_val(pte) & _PAGE_USER; }
+static inline int pte_exec(pte_t pte) { return __pte_val(pte) & _PAGE_USER; }
+static inline int pte_dirty(pte_t pte) { return __pte_val(pte) & _PAGE_DIRTY; }
+static inline int pte_young(pte_t pte) { return __pte_val(pte) & _PAGE_ACCESSED; }
+static inline int pte_write(pte_t pte) { return __pte_val(pte) & _PAGE_RW; }
+static inline int pte_file(pte_t pte) { return __pte_val(pte) & _PAGE_FILE; }
+static inline int pte_huge(pte_t pte) { return __pte_val(pte) & _PAGE_PSE; }
+
+static inline pte_t pte_rdprotect(pte_t pte) { __pte_val(pte) &= ~_PAGE_USER; return pte; }
+static inline pte_t pte_exprotect(pte_t pte) { __pte_val(pte) &= ~_PAGE_USER; return pte; }
+static inline pte_t pte_mkclean(pte_t pte) { __pte_val(pte) &= ~_PAGE_DIRTY; return pte; }
+static inline pte_t pte_mkold(pte_t pte) { __pte_val(pte) &= ~_PAGE_ACCESSED; return pte; }
+static inline pte_t pte_wrprotect(pte_t pte) { __pte_val(pte) &= ~_PAGE_RW; return pte; }
+static inline pte_t pte_mkread(pte_t pte) { __pte_val(pte) |= _PAGE_USER; return pte; }
+static inline pte_t pte_mkexec(pte_t pte) { __pte_val(pte) |= _PAGE_USER; return pte; }
+static inline pte_t pte_mkdirty(pte_t pte) { __pte_val(pte) |= _PAGE_DIRTY; return pte; }
+static inline pte_t pte_mkyoung(pte_t pte) { __pte_val(pte) |= _PAGE_ACCESSED; return pte; }
+static inline pte_t pte_mkwrite(pte_t pte) { __pte_val(pte) |= _PAGE_RW; return pte; }
+static inline pte_t pte_mkhuge(pte_t pte) { __pte_val(pte) |= _PAGE_PSE; return pte; }
+
+#define ptep_test_and_clear_dirty(vma, addr, ptep) \
+({ \
+ pte_t __pte = *(ptep); \
+ int __ret = pte_dirty(__pte); \
+ if (__ret) \
+ set_pte_at((vma)->vm_mm, addr, ptep, pte_mkclean(__pte)); \
+ __ret; \
+})
+
+#define ptep_test_and_clear_young(vma, addr, ptep) \
+({ \
+ pte_t __pte = *(ptep); \
+ int __ret = pte_young(__pte); \
+ if (__ret) \
+ set_pte_at((vma)->vm_mm, addr, ptep, pte_mkold(__pte)); \
+ __ret; \
+})
+
+static inline void ptep_set_wrprotect(struct mm_struct *mm, unsigned long addr, pte_t *ptep)
+{
+ pte_t pte = *ptep;
+ if (pte_write(pte))
+ set_pte_at(mm, addr, ptep, pte_wrprotect(pte));
+}
+
+/*
+ * Macro to mark a page protection value as "uncacheable".
+ */
+#define pgprot_noncached(prot) (__pgprot(pgprot_val(prot) | _PAGE_PCD | _PAGE_PWT))
+
+static inline int pmd_large(pmd_t pte) {
+ return (__pmd_val(pte) & __LARGE_PTE) == __LARGE_PTE;
+}
+
+
+/*
+ * Conversion functions: convert a page and protection to a page entry,
+ * and a page entry and page directory to the page they refer to.
+ */
+
+/*
+ * Level 4 access.
+ * Never use these in the common code.
+ */
+#define pgd_page(pgd) ((unsigned long) __va(pgd_val(pgd) & PTE_MASK))
+#define pgd_index(address) (((address) >> PGDIR_SHIFT) & (PTRS_PER_PGD-1))
+#define pgd_offset(mm, addr) ((mm)->pgd + pgd_index(addr))
+#define pgd_offset_k(address) (init_level4_pgt + pgd_index(address))
+#define pgd_present(pgd) (__pgd_val(pgd) & _PAGE_PRESENT)
+#define mk_kernel_pgd(address) __pgd((address) | _KERNPG_TABLE)
+
+/* PUD - Level3 access */
+/* to find an entry in a page-table-directory. */
+#define pud_index(address) (((address) >> PUD_SHIFT) & (PTRS_PER_PUD-1))
+#define pud_offset(pgd, address) ((pud_t *) pgd_page(*(pgd)) + pud_index(address))
+#define pud_present(pud) (__pud_val(pud) & _PAGE_PRESENT)
+
+/* PMD - Level 2 access */
+#define pmd_page_kernel(pmd) ((unsigned long) __va(pmd_val(pmd) & PTE_MASK))
+#define pmd_page(pmd) (pfn_to_page(pmd_val(pmd) >> PAGE_SHIFT))
+
+#define pmd_index(address) (((address) >> PMD_SHIFT) & (PTRS_PER_PMD-1))
+#define pmd_offset(dir, address) ((pmd_t *) pud_page(*(dir)) + \
+ pmd_index(address))
+#define pmd_none(x) (!__pmd_val(x))
+#if CONFIG_XEN_COMPAT <= 0x030002
+/* pmd_present doesn't just test the _PAGE_PRESENT bit since wr.p.t.
+ can temporarily clear it. */
+#define pmd_present(x) (__pmd_val(x))
+#else
+#define pmd_present(x) (__pmd_val(x) & _PAGE_PRESENT)
+#endif
+#define pmd_clear(xp) do { set_pmd(xp, __pmd(0)); } while (0)
+#define pmd_bad(x) ((__pmd_val(x) & ~(PTE_MASK | _PAGE_USER | _PAGE_PRESENT)) \
+ != (_KERNPG_TABLE & ~(_PAGE_USER | _PAGE_PRESENT)))
+#define pfn_pmd(nr,prot) (__pmd(((nr) << PAGE_SHIFT) | pgprot_val(prot)))
+#define pmd_pfn(x) ((pmd_val(x) & __PHYSICAL_MASK) >> PAGE_SHIFT)
+
+#define pte_to_pgoff(pte) ((__pte_val(pte) & PHYSICAL_PAGE_MASK) >> PAGE_SHIFT)
+#define pgoff_to_pte(off) ((pte_t) { ((off) << PAGE_SHIFT) | _PAGE_FILE })
+#define PTE_FILE_MAX_BITS __PHYSICAL_MASK_SHIFT
+
+/* PTE - Level 1 access. */
+
+/* page, protection -> pte */
+#define mk_pte(page, pgprot) pfn_pte(page_to_pfn(page), (pgprot))
+#define mk_pte_huge(entry) (__pte_val(entry) |= _PAGE_PRESENT | _PAGE_PSE)
+
+/* physical address -> PTE */
+static inline pte_t mk_pte_phys(unsigned long physpage, pgprot_t pgprot)
+{
+ unsigned long pteval;
+ pteval = physpage | pgprot_val(pgprot);
+ return __pte(pteval);
+}
+
+/* Change flags of a PTE */
+static inline pte_t pte_modify(pte_t pte, pgprot_t newprot)
+{
+ /*
+ * Since this might change the present bit (which controls whether
+ * a pte_t object has undergone p2m translation), we must use
+ * pte_val() on the input pte and __pte() for the return value.
+ */
+ unsigned long pteval = pte_val(pte);
+
+ pteval &= _PAGE_CHG_MASK;
+ pteval |= pgprot_val(newprot);
+ pteval &= __supported_pte_mask;
+ return __pte(pteval);
+}
+
+#define pte_index(address) \
+ (((address) >> PAGE_SHIFT) & (PTRS_PER_PTE - 1))
+#define pte_offset_kernel(dir, address) ((pte_t *) pmd_page_kernel(*(dir)) + \
+ pte_index(address))
+
+/* x86-64 always has all page tables mapped. */
+#define pte_offset_map(dir,address) pte_offset_kernel(dir,address)
+#define pte_offset_map_nested(dir,address) pte_offset_kernel(dir,address)
+#define pte_unmap(pte) /* NOP */
+#define pte_unmap_nested(pte) /* NOP */
+
+#define update_mmu_cache(vma,address,pte) do { } while (0)
+
+/*
+ * Rules for using ptep_establish: the pte MUST be a user pte, and
+ * must be a present->present transition.
+ */
+#define __HAVE_ARCH_PTEP_ESTABLISH
+#define ptep_establish(vma, address, ptep, pteval) \
+ do { \
+ if ( likely((vma)->vm_mm == current->mm) ) { \
+ BUG_ON(HYPERVISOR_update_va_mapping(address, \
+ pteval, \
+ (unsigned long)(vma)->vm_mm->cpu_vm_mask.bits| \
+ UVMF_INVLPG|UVMF_MULTI)); \
+ } else { \
+ xen_l1_entry_update(ptep, pteval); \
+ flush_tlb_page(vma, address); \
+ } \
+ } while (0)
+
+/* We only update the dirty/accessed state if we set
+ * the dirty bit by hand in the kernel, since the hardware
+ * will do the accessed bit for us, and we don't want to
+ * race with other CPU's that might be updating the dirty
+ * bit at the same time. */
+#define __HAVE_ARCH_PTEP_SET_ACCESS_FLAGS
+#define ptep_set_access_flags(vma, address, ptep, entry, dirty) \
+ do { \
+ if (dirty) \
+ ptep_establish(vma, address, ptep, entry); \
+ } while (0)
+
+/* Encode and de-code a swap entry */
+#define __swp_type(x) (((x).val >> 1) & 0x3f)
+#define __swp_offset(x) ((x).val >> 8)
+#define __swp_entry(type, offset) ((swp_entry_t) { ((type) << 1) | ((offset) << 8) })
+#define __pte_to_swp_entry(pte) ((swp_entry_t) { __pte_val(pte) })
+#define __swp_entry_to_pte(x) ((pte_t) { (x).val })
+
+extern spinlock_t pgd_lock;
+extern struct page *pgd_list;
+void vmalloc_sync_all(void);
+
+#endif /* !__ASSEMBLY__ */
+
+extern int kern_addr_valid(unsigned long addr);
+
+#define DOMID_LOCAL (0xFFFFU)
+
+struct vm_area_struct;
+
+int direct_remap_pfn_range(struct vm_area_struct *vma,
+ unsigned long address,
+ unsigned long mfn,
+ unsigned long size,
+ pgprot_t prot,
+ domid_t domid);
+
+int direct_kernel_remap_pfn_range(unsigned long address,
+ unsigned long mfn,
+ unsigned long size,
+ pgprot_t prot,
+ domid_t domid);
+
+int create_lookup_pte_addr(struct mm_struct *mm,
+ unsigned long address,
+ uint64_t *ptep);
+
+int touch_pte_range(struct mm_struct *mm,
+ unsigned long address,
+ unsigned long size);
+
+int xen_change_pte_range(struct mm_struct *mm, pmd_t *pmd,
+ unsigned long addr, unsigned long end, pgprot_t newprot);
+
+#define arch_change_pte_range(mm, pmd, addr, end, newprot) \
+ xen_change_pte_range(mm, pmd, addr, end, newprot)
+
+#define io_remap_pfn_range(vma, vaddr, pfn, size, prot) \
+ direct_remap_pfn_range(vma,vaddr,pfn,size,prot,DOMID_IO)
+
+#define MK_IOSPACE_PFN(space, pfn) (pfn)
+#define GET_IOSPACE(pfn) 0
+#define GET_PFN(pfn) (pfn)
+
+#define HAVE_ARCH_UNMAPPED_AREA
+
+#define pgtable_cache_init() do { } while (0)
+#define check_pgt_cache() do { } while (0)
+
+#define PAGE_AGP PAGE_KERNEL_NOCACHE
+#define HAVE_PAGE_AGP 1
+
+/* fs/proc/kcore.c */
+#define kc_vaddr_to_offset(v) ((v) & __VIRTUAL_MASK)
+#define kc_offset_to_vaddr(o) \
+ (((o) & (1UL << (__VIRTUAL_MASK_SHIFT-1))) ? ((o) | (~__VIRTUAL_MASK)) : (o))
+
+#define __HAVE_ARCH_PTEP_TEST_AND_CLEAR_YOUNG
+#define __HAVE_ARCH_PTEP_TEST_AND_CLEAR_DIRTY
+#define __HAVE_ARCH_PTEP_GET_AND_CLEAR
+#define __HAVE_ARCH_PTEP_GET_AND_CLEAR_FULL
+#define __HAVE_ARCH_PTEP_CLEAR_FLUSH
+#define __HAVE_ARCH_PTEP_SET_WRPROTECT
+#define __HAVE_ARCH_PTE_SAME
+#include <asm-generic/pgtable.h>
+
+#endif /* _X86_64_PGTABLE_H */
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/include/mach-xen/asm/processor_64.h 2008-03-06 08:54:32.000000000 +0100
@@ -0,0 +1,502 @@
+/*
+ * include/asm-x86_64/processor.h
+ *
+ * Copyright (C) 1994 Linus Torvalds
+ */
+
+#ifndef __ASM_X86_64_PROCESSOR_H
+#define __ASM_X86_64_PROCESSOR_H
+
+#include <asm/segment.h>
+#include <asm/page.h>
+#include <asm/types.h>
+#include <asm/sigcontext.h>
+#include <asm/cpufeature.h>
+#include <linux/threads.h>
+#include <asm/msr.h>
+#include <asm/current.h>
+#include <asm/system.h>
+#include <asm/mmsegment.h>
+#include <asm/percpu.h>
+#include <linux/personality.h>
+#include <linux/cpumask.h>
+
+#define TF_MASK 0x00000100
+#define IF_MASK 0x00000200
+#define IOPL_MASK 0x00003000
+#define NT_MASK 0x00004000
+#define VM_MASK 0x00020000
+#define AC_MASK 0x00040000
+#define VIF_MASK 0x00080000 /* virtual interrupt flag */
+#define VIP_MASK 0x00100000 /* virtual interrupt pending */
+#define ID_MASK 0x00200000
+
+#define desc_empty(desc) \
+ (!((desc)->a | (desc)->b))
+
+#define desc_equal(desc1, desc2) \
+ (((desc1)->a == (desc2)->a) && ((desc1)->b == (desc2)->b))
+
+/*
+ * Default implementation of macro that returns current
+ * instruction pointer ("program counter").
+ */
+#define current_text_addr() ({ void *pc; asm volatile("leaq 1f(%%rip),%0\n1:":"=r"(pc)); pc; })
+
+/*
+ * CPU type and hardware bug flags. Kept separately for each CPU.
+ */
+
+struct cpuinfo_x86 {
+ __u8 x86; /* CPU family */
+ __u8 x86_vendor; /* CPU vendor */
+ __u8 x86_model;
+ __u8 x86_mask;
+ int cpuid_level; /* Maximum supported CPUID level, -1=no CPUID */
+ __u32 x86_capability[NCAPINTS];
+ char x86_vendor_id[16];
+ char x86_model_id[64];
+ int x86_cache_size; /* in KB */
+ int x86_clflush_size;
+ int x86_cache_alignment;
+ int x86_tlbsize; /* number of 4K pages in DTLB/ITLB combined(in pages)*/
+ __u8 x86_virt_bits, x86_phys_bits;
+ __u8 x86_max_cores; /* cpuid returned max cores value */
+ __u32 x86_power;
+ __u32 extended_cpuid_level; /* Max extended CPUID function supported */
+ unsigned long loops_per_jiffy;
+#ifdef CONFIG_SMP
+ cpumask_t llc_shared_map; /* cpus sharing the last level cache */
+#endif
+ __u8 apicid;
+#ifdef CONFIG_SMP
+ __u8 booted_cores; /* number of cores as seen by OS */
+ __u8 phys_proc_id; /* Physical Processor id. */
+ __u8 cpu_core_id; /* Core id. */
+#endif
+} ____cacheline_aligned;
+
+#define X86_VENDOR_INTEL 0
+#define X86_VENDOR_CYRIX 1
+#define X86_VENDOR_AMD 2
+#define X86_VENDOR_UMC 3
+#define X86_VENDOR_NEXGEN 4
+#define X86_VENDOR_CENTAUR 5
+#define X86_VENDOR_RISE 6
+#define X86_VENDOR_TRANSMETA 7
+#define X86_VENDOR_NUM 8
+#define X86_VENDOR_UNKNOWN 0xff
+
+#ifdef CONFIG_SMP
+extern struct cpuinfo_x86 cpu_data[];
+#define current_cpu_data cpu_data[smp_processor_id()]
+#else
+#define cpu_data (&boot_cpu_data)
+#define current_cpu_data boot_cpu_data
+#endif
+
+extern char ignore_irq13;
+
+extern void identify_cpu(struct cpuinfo_x86 *);
+extern void print_cpu_info(struct cpuinfo_x86 *);
+extern unsigned int init_intel_cacheinfo(struct cpuinfo_x86 *c);
+extern unsigned short num_cache_leaves;
+
+/*
+ * EFLAGS bits
+ */
+#define X86_EFLAGS_CF 0x00000001 /* Carry Flag */
+#define X86_EFLAGS_PF 0x00000004 /* Parity Flag */
+#define X86_EFLAGS_AF 0x00000010 /* Auxillary carry Flag */
+#define X86_EFLAGS_ZF 0x00000040 /* Zero Flag */
+#define X86_EFLAGS_SF 0x00000080 /* Sign Flag */
+#define X86_EFLAGS_TF 0x00000100 /* Trap Flag */
+#define X86_EFLAGS_IF 0x00000200 /* Interrupt Flag */
+#define X86_EFLAGS_DF 0x00000400 /* Direction Flag */
+#define X86_EFLAGS_OF 0x00000800 /* Overflow Flag */
+#define X86_EFLAGS_IOPL 0x00003000 /* IOPL mask */
+#define X86_EFLAGS_NT 0x00004000 /* Nested Task */
+#define X86_EFLAGS_RF 0x00010000 /* Resume Flag */
+#define X86_EFLAGS_VM 0x00020000 /* Virtual Mode */
+#define X86_EFLAGS_AC 0x00040000 /* Alignment Check */
+#define X86_EFLAGS_VIF 0x00080000 /* Virtual Interrupt Flag */
+#define X86_EFLAGS_VIP 0x00100000 /* Virtual Interrupt Pending */
+#define X86_EFLAGS_ID 0x00200000 /* CPUID detection flag */
+
+/*
+ * Intel CPU features in CR4
+ */
+#define X86_CR4_VME 0x0001 /* enable vm86 extensions */
+#define X86_CR4_PVI 0x0002 /* virtual interrupts flag enable */
+#define X86_CR4_TSD 0x0004 /* disable time stamp at ipl 3 */
+#define X86_CR4_DE 0x0008 /* enable debugging extensions */
+#define X86_CR4_PSE 0x0010 /* enable page size extensions */
+#define X86_CR4_PAE 0x0020 /* enable physical address extensions */
+#define X86_CR4_MCE 0x0040 /* Machine check enable */
+#define X86_CR4_PGE 0x0080 /* enable global pages */
+#define X86_CR4_PCE 0x0100 /* enable performance counters at ipl 3 */
+#define X86_CR4_OSFXSR 0x0200 /* enable fast FPU save and restore */
+#define X86_CR4_OSXMMEXCPT 0x0400 /* enable unmasked SSE exceptions */
+
+/*
+ * Save the cr4 feature set we're using (ie
+ * Pentium 4MB enable and PPro Global page
+ * enable), so that any CPU's that boot up
+ * after us can get the correct flags.
+ */
+extern unsigned long mmu_cr4_features;
+
+static inline void set_in_cr4 (unsigned long mask)
+{
+ mmu_cr4_features |= mask;
+ __asm__("movq %%cr4,%%rax\n\t"
+ "orq %0,%%rax\n\t"
+ "movq %%rax,%%cr4\n"
+ : : "irg" (mask)
+ :"ax");
+}
+
+static inline void clear_in_cr4 (unsigned long mask)
+{
+ mmu_cr4_features &= ~mask;
+ __asm__("movq %%cr4,%%rax\n\t"
+ "andq %0,%%rax\n\t"
+ "movq %%rax,%%cr4\n"
+ : : "irg" (~mask)
+ :"ax");
+}
+
+
+/*
+ * User space process size. 47bits minus one guard page.
+ */
+#define TASK_SIZE64 (0x800000000000UL - 4096)
+
+/* This decides where the kernel will search for a free chunk of vm
+ * space during mmap's.
+ */
+#define IA32_PAGE_OFFSET ((current->personality & ADDR_LIMIT_3GB) ? 0xc0000000 : 0xFFFFe000)
+
+#define TASK_SIZE (test_thread_flag(TIF_IA32) ? IA32_PAGE_OFFSET : TASK_SIZE64)
+#define TASK_SIZE_OF(child) ((test_tsk_thread_flag(child, TIF_IA32)) ? IA32_PAGE_OFFSET : TASK_SIZE64)
+
+#define TASK_UNMAPPED_BASE PAGE_ALIGN(TASK_SIZE/3)
+
+/*
+ * Size of io_bitmap.
+ */
+#define IO_BITMAP_BITS 65536
+#define IO_BITMAP_BYTES (IO_BITMAP_BITS/8)
+#define IO_BITMAP_LONGS (IO_BITMAP_BYTES/sizeof(long))
+#ifndef CONFIG_X86_NO_TSS
+#define IO_BITMAP_OFFSET offsetof(struct tss_struct,io_bitmap)
+#endif
+#define INVALID_IO_BITMAP_OFFSET 0x8000
+
+struct i387_fxsave_struct {
+ u16 cwd;
+ u16 swd;
+ u16 twd;
+ u16 fop;
+ u64 rip;
+ u64 rdp;
+ u32 mxcsr;
+ u32 mxcsr_mask;
+ u32 st_space[32]; /* 8*16 bytes for each FP-reg = 128 bytes */
+ u32 xmm_space[64]; /* 16*16 bytes for each XMM-reg = 128 bytes */
+ u32 padding[24];
+} __attribute__ ((aligned (16)));
+
+union i387_union {
+ struct i387_fxsave_struct fxsave;
+};
+
+#ifndef CONFIG_X86_NO_TSS
+struct tss_struct {
+ u32 reserved1;
+ u64 rsp0;
+ u64 rsp1;
+ u64 rsp2;
+ u64 reserved2;
+ u64 ist[7];
+ u32 reserved3;
+ u32 reserved4;
+ u16 reserved5;
+ u16 io_bitmap_base;
+ /*
+ * The extra 1 is there because the CPU will access an
+ * additional byte beyond the end of the IO permission
+ * bitmap. The extra byte must be all 1 bits, and must
+ * be within the limit. Thus we have:
+ *
+ * 128 bytes, the bitmap itself, for ports 0..0x3ff
+ * 8 bytes, for an extra "long" of ~0UL
+ */
+ unsigned long io_bitmap[IO_BITMAP_LONGS + 1];
+} __attribute__((packed)) ____cacheline_aligned;
+
+DECLARE_PER_CPU(struct tss_struct,init_tss);
+#endif
+
+
+extern struct cpuinfo_x86 boot_cpu_data;
+#ifndef CONFIG_X86_NO_TSS
+/* Save the original ist values for checking stack pointers during debugging */
+struct orig_ist {
+ unsigned long ist[7];
+};
+DECLARE_PER_CPU(struct orig_ist, orig_ist);
+#endif
+
+#ifdef CONFIG_X86_VSMP
+#define ARCH_MIN_TASKALIGN (1 << INTERNODE_CACHE_SHIFT)
+#define ARCH_MIN_MMSTRUCT_ALIGN (1 << INTERNODE_CACHE_SHIFT)
+#else
+#define ARCH_MIN_TASKALIGN 16
+#define ARCH_MIN_MMSTRUCT_ALIGN 0
+#endif
+
+struct thread_struct {
+ unsigned long rsp0;
+ unsigned long rsp;
+ unsigned long userrsp; /* Copy from PDA */
+ unsigned long fs;
+ unsigned long gs;
+ unsigned short es, ds, fsindex, gsindex;
+/* Hardware debugging registers */
+ unsigned long debugreg0;
+ unsigned long debugreg1;
+ unsigned long debugreg2;
+ unsigned long debugreg3;
+ unsigned long debugreg6;
+ unsigned long debugreg7;
+/* fault info */
+ unsigned long cr2, trap_no, error_code;
+/* floating point info */
+ union i387_union i387 __attribute__((aligned(16)));
+/* IO permissions. the bitmap could be moved into the GDT, that would make
+ switch faster for a limited number of ioperm using tasks. -AK */
+ int ioperm;
+ unsigned long *io_bitmap_ptr;
+ unsigned io_bitmap_max;
+/* cached TLS descriptors. */
+ u64 tls_array[GDT_ENTRY_TLS_ENTRIES];
+ unsigned int iopl;
+} __attribute__((aligned(16)));
+
+#define INIT_THREAD { \
+ .rsp0 = (unsigned long)&init_stack + sizeof(init_stack) \
+}
+
+#ifndef CONFIG_X86_NO_TSS
+#define INIT_TSS { \
+ .rsp0 = (unsigned long)&init_stack + sizeof(init_stack) \
+}
+#endif
+
+#define INIT_MMAP \
+{ &init_mm, 0, 0, NULL, PAGE_SHARED, VM_READ | VM_WRITE | VM_EXEC, 1, NULL, NULL }
+
+#define start_thread(regs,new_rip,new_rsp) do { \
+ asm volatile("movl %0,%%fs; movl %0,%%es; movl %0,%%ds": :"r" (0)); \
+ load_gs_index(0); \
+ (regs)->rip = (new_rip); \
+ (regs)->rsp = (new_rsp); \
+ write_pda(oldrsp, (new_rsp)); \
+ (regs)->cs = __USER_CS; \
+ (regs)->ss = __USER_DS; \
+ (regs)->eflags = 0x200; \
+ set_fs(USER_DS); \
+} while(0)
+
+#define get_debugreg(var, register) \
+ var = HYPERVISOR_get_debugreg(register)
+#define set_debugreg(value, register) do { \
+ if (HYPERVISOR_set_debugreg(register, value)) \
+ BUG(); \
+} while (0)
+
+struct task_struct;
+struct mm_struct;
+
+/* Free all resources held by a thread. */
+extern void release_thread(struct task_struct *);
+
+/* Prepare to copy thread state - unlazy all lazy status */
+extern void prepare_to_copy(struct task_struct *tsk);
+
+/*
+ * create a kernel thread without removing it from tasklists
+ */
+extern long kernel_thread(int (*fn)(void *), void * arg, unsigned long flags);
+
+/*
+ * Return saved PC of a blocked thread.
+ * What is this good for? it will be always the scheduler or ret_from_fork.
+ */
+#define thread_saved_pc(t) (*(unsigned long *)((t)->thread.rsp - 8))
+
+extern unsigned long get_wchan(struct task_struct *p);
+#define task_pt_regs(tsk) ((struct pt_regs *)(tsk)->thread.rsp0 - 1)
+#define KSTK_EIP(tsk) (task_pt_regs(tsk)->rip)
+#define KSTK_ESP(tsk) -1 /* sorry. doesn't work for syscall. */
+
+
+struct microcode_header {
+ unsigned int hdrver;
+ unsigned int rev;
+ unsigned int date;
+ unsigned int sig;
+ unsigned int cksum;
+ unsigned int ldrver;
+ unsigned int pf;
+ unsigned int datasize;
+ unsigned int totalsize;
+ unsigned int reserved[3];
+};
+
+struct microcode {
+ struct microcode_header hdr;
+ unsigned int bits[0];
+};
+
+typedef struct microcode microcode_t;
+typedef struct microcode_header microcode_header_t;
+
+/* microcode format is extended from prescott processors */
+struct extended_signature {
+ unsigned int sig;
+ unsigned int pf;
+ unsigned int cksum;
+};
+
+struct extended_sigtable {
+ unsigned int count;
+ unsigned int cksum;
+ unsigned int reserved[3];
+ struct extended_signature sigs[0];
+};
+
+
+#define ASM_NOP1 K8_NOP1
+#define ASM_NOP2 K8_NOP2
+#define ASM_NOP3 K8_NOP3
+#define ASM_NOP4 K8_NOP4
+#define ASM_NOP5 K8_NOP5
+#define ASM_NOP6 K8_NOP6
+#define ASM_NOP7 K8_NOP7
+#define ASM_NOP8 K8_NOP8
+
+/* Opteron nops */
+#define K8_NOP1 ".byte 0x90\n"
+#define K8_NOP2 ".byte 0x66,0x90\n"
+#define K8_NOP3 ".byte 0x66,0x66,0x90\n"
+#define K8_NOP4 ".byte 0x66,0x66,0x66,0x90\n"
+#define K8_NOP5 K8_NOP3 K8_NOP2
+#define K8_NOP6 K8_NOP3 K8_NOP3
+#define K8_NOP7 K8_NOP4 K8_NOP3
+#define K8_NOP8 K8_NOP4 K8_NOP4
+
+#define ASM_NOP_MAX 8
+
+/* REP NOP (PAUSE) is a good thing to insert into busy-wait loops. */
+static inline void rep_nop(void)
+{
+ __asm__ __volatile__("rep;nop": : :"memory");
+}
+
+/* Stop speculative execution */
+static inline void sync_core(void)
+{
+ int tmp;
+ asm volatile("cpuid" : "=a" (tmp) : "0" (1) : "ebx","ecx","edx","memory");
+}
+
+#define cpu_has_fpu 1
+
+#define ARCH_HAS_PREFETCH
+static inline void prefetch(void *x)
+{
+ asm volatile("prefetcht0 %0" :: "m" (*(unsigned long *)x));
+}
+
+#define ARCH_HAS_PREFETCHW 1
+static inline void prefetchw(void *x)
+{
+ alternative_input("prefetcht0 (%1)",
+ "prefetchw (%1)",
+ X86_FEATURE_3DNOW,
+ "r" (x));
+}
+
+#define ARCH_HAS_SPINLOCK_PREFETCH 1
+
+#define spin_lock_prefetch(x) prefetchw(x)
+
+#define cpu_relax() rep_nop()
+
+/*
+ * NSC/Cyrix CPU configuration register indexes
+ */
+#define CX86_CCR0 0xc0
+#define CX86_CCR1 0xc1
+#define CX86_CCR2 0xc2
+#define CX86_CCR3 0xc3
+#define CX86_CCR4 0xe8
+#define CX86_CCR5 0xe9
+#define CX86_CCR6 0xea
+#define CX86_CCR7 0xeb
+#define CX86_DIR0 0xfe
+#define CX86_DIR1 0xff
+#define CX86_ARR_BASE 0xc4
+#define CX86_RCR_BASE 0xdc
+
+/*
+ * NSC/Cyrix CPU indexed register access macros
+ */
+
+#define getCx86(reg) ({ outb((reg), 0x22); inb(0x23); })
+
+#define setCx86(reg, data) do { \
+ outb((reg), 0x22); \
+ outb((data), 0x23); \
+} while (0)
+
+static inline void serialize_cpu(void)
+{
+ __asm__ __volatile__ ("cpuid" : : : "ax", "bx", "cx", "dx");
+}
+
+static inline void __monitor(const void *eax, unsigned long ecx,
+ unsigned long edx)
+{
+ /* "monitor %eax,%ecx,%edx;" */
+ asm volatile(
+ ".byte 0x0f,0x01,0xc8;"
+ : :"a" (eax), "c" (ecx), "d"(edx));
+}
+
+static inline void __mwait(unsigned long eax, unsigned long ecx)
+{
+ /* "mwait %eax,%ecx;" */
+ asm volatile(
+ ".byte 0x0f,0x01,0xc9;"
+ : :"a" (eax), "c" (ecx));
+}
+
+#define stack_current() \
+({ \
+ struct thread_info *ti; \
+ asm("andq %%rsp,%0; ":"=r" (ti) : "0" (CURRENT_MASK)); \
+ ti->task; \
+})
+
+#define cache_line_size() (boot_cpu_data.x86_cache_alignment)
+
+extern unsigned long boot_option_idle_override;
+/* Boot loader type from the setup header */
+extern int bootloader_type;
+
+#define HAVE_ARCH_PICK_MMAP_LAYOUT 1
+
+#endif /* __ASM_X86_64_PROCESSOR_H */
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/include/mach-xen/asm/smp_64.h 2007-06-12 13:14:13.000000000 +0200
@@ -0,0 +1,150 @@
+#ifndef __ASM_SMP_H
+#define __ASM_SMP_H
+
+/*
+ * We need the APIC definitions automatically as part of 'smp.h'
+ */
+#ifndef __ASSEMBLY__
+#include <linux/threads.h>
+#include <linux/cpumask.h>
+#include <linux/bitops.h>
+extern int disable_apic;
+#endif
+
+#ifdef CONFIG_X86_LOCAL_APIC
+#ifndef __ASSEMBLY__
+#include <asm/fixmap.h>
+#include <asm/mpspec.h>
+#ifdef CONFIG_X86_IO_APIC
+#include <asm/io_apic.h>
+#endif
+#include <asm/apic.h>
+#include <asm/thread_info.h>
+#endif
+#endif
+
+#ifdef CONFIG_SMP
+#ifndef ASSEMBLY
+
+#include <asm/pda.h>
+
+struct pt_regs;
+
+extern cpumask_t cpu_present_mask;
+extern cpumask_t cpu_possible_map;
+extern cpumask_t cpu_online_map;
+extern cpumask_t cpu_initialized;
+
+/*
+ * Private routines/data
+ */
+
+extern void smp_alloc_memory(void);
+extern volatile unsigned long smp_invalidate_needed;
+extern int pic_mode;
+extern void lock_ipi_call_lock(void);
+extern void unlock_ipi_call_lock(void);
+extern int smp_num_siblings;
+extern void smp_send_reschedule(int cpu);
+void smp_stop_cpu(void);
+extern int smp_call_function_single(int cpuid, void (*func) (void *info),
+ void *info, int retry, int wait);
+
+extern cpumask_t cpu_sibling_map[NR_CPUS];
+extern cpumask_t cpu_core_map[NR_CPUS];
+extern u8 cpu_llc_id[NR_CPUS];
+
+#define SMP_TRAMPOLINE_BASE 0x6000
+
+/*
+ * On x86 all CPUs are mapped 1:1 to the APIC space.
+ * This simplifies scheduling and IPI sending and
+ * compresses data structures.
+ */
+
+static inline int num_booting_cpus(void)
+{
+ return cpus_weight(cpu_possible_map);
+}
+
+#define raw_smp_processor_id() read_pda(cpunumber)
+
+#ifdef CONFIG_X86_LOCAL_APIC
+static inline int hard_smp_processor_id(void)
+{
+ /* we don't want to mark this access volatile - bad code generation */
+ return GET_APIC_ID(*(unsigned int *)(APIC_BASE+APIC_ID));
+}
+#endif
+
+extern int safe_smp_processor_id(void);
+extern int __cpu_disable(void);
+extern void __cpu_die(unsigned int cpu);
+extern void prefill_possible_map(void);
+extern unsigned num_processors;
+extern unsigned disabled_cpus;
+
+#endif /* !ASSEMBLY */
+
+#define NO_PROC_ID 0xFF /* No processor magic marker */
+
+#endif
+
+#ifndef ASSEMBLY
+/*
+ * Some lowlevel functions might want to know about
+ * the real APIC ID <-> CPU # mapping.
+ */
+extern u8 x86_cpu_to_apicid[NR_CPUS]; /* physical ID */
+extern u8 x86_cpu_to_log_apicid[NR_CPUS];
+extern u8 bios_cpu_apicid[];
+
+#ifdef CONFIG_X86_LOCAL_APIC
+static inline unsigned int cpu_mask_to_apicid(cpumask_t cpumask)
+{
+ return cpus_addr(cpumask)[0];
+}
+
+static inline int cpu_present_to_apicid(int mps_cpu)
+{
+ if (mps_cpu < NR_CPUS)
+ return (int)bios_cpu_apicid[mps_cpu];
+ else
+ return BAD_APICID;
+}
+#endif
+
+#endif /* !ASSEMBLY */
+
+#ifndef CONFIG_SMP
+#define stack_smp_processor_id() 0
+#define safe_smp_processor_id() 0
+#define cpu_logical_map(x) (x)
+#else
+#include <asm/thread_info.h>
+#define stack_smp_processor_id() \
+({ \
+ struct thread_info *ti; \
+ __asm__("andq %%rsp,%0; ":"=r" (ti) : "0" (CURRENT_MASK)); \
+ ti->cpu; \
+})
+#endif
+
+#ifndef __ASSEMBLY__
+#ifdef CONFIG_X86_LOCAL_APIC
+static __inline int logical_smp_processor_id(void)
+{
+ /* we don't want to mark this access volatile - bad code generation */
+ return GET_APIC_LOGICAL_ID(*(unsigned long *)(APIC_BASE+APIC_LDR));
+}
+#endif
+#endif
+
+#ifdef CONFIG_SMP
+#define cpu_physical_id(cpu) x86_cpu_to_apicid[cpu]
+#else
+#define cpu_physical_id(cpu) boot_cpu_id
+#endif
+
+#endif
+
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/include/mach-xen/asm/system_64.h 2007-11-26 16:59:25.000000000 +0100
@@ -0,0 +1,256 @@
+#ifndef __ASM_SYSTEM_H
+#define __ASM_SYSTEM_H
+
+#include <linux/kernel.h>
+#include <asm/segment.h>
+#include <asm/alternative.h>
+
+#include <asm/synch_bitops.h>
+#include <asm/hypervisor.h>
+#include <xen/interface/arch-x86_64.h>
+
+#ifdef __KERNEL__
+
+#define __STR(x) #x
+#define STR(x) __STR(x)
+
+#define __SAVE(reg,offset) "movq %%" #reg ",(14-" #offset ")*8(%%rsp)\n\t"
+#define __RESTORE(reg,offset) "movq (14-" #offset ")*8(%%rsp),%%" #reg "\n\t"
+
+/* frame pointer must be last for get_wchan */
+#define SAVE_CONTEXT "pushf ; pushq %%rbp ; movq %%rsi,%%rbp\n\t"
+#define RESTORE_CONTEXT "movq %%rbp,%%rsi ; popq %%rbp ; popf\n\t"
+
+#define __EXTRA_CLOBBER \
+ ,"rcx","rbx","rdx","r8","r9","r10","r11","r12","r13","r14","r15"
+
+#define switch_to(prev,next,last) \
+ asm volatile(SAVE_CONTEXT \
+ "movq %%rsp,%P[threadrsp](%[prev])\n\t" /* save RSP */ \
+ "movq %P[threadrsp](%[next]),%%rsp\n\t" /* restore RSP */ \
+ "call __switch_to\n\t" \
+ ".globl thread_return\n" \
+ "thread_return:\n\t" \
+ "movq %%gs:%P[pda_pcurrent],%%rsi\n\t" \
+ "movq %P[thread_info](%%rsi),%%r8\n\t" \
+ LOCK_PREFIX "btr %[tif_fork],%P[ti_flags](%%r8)\n\t" \
+ "movq %%rax,%%rdi\n\t" \
+ "jc ret_from_fork\n\t" \
+ RESTORE_CONTEXT \
+ : "=a" (last) \
+ : [next] "S" (next), [prev] "D" (prev), \
+ [threadrsp] "i" (offsetof(struct task_struct, thread.rsp)), \
+ [ti_flags] "i" (offsetof(struct thread_info, flags)),\
+ [tif_fork] "i" (TIF_FORK), \
+ [thread_info] "i" (offsetof(struct task_struct, thread_info)), \
+ [pda_pcurrent] "i" (offsetof(struct x8664_pda, pcurrent)) \
+ : "memory", "cc" __EXTRA_CLOBBER)
+
+extern void load_gs_index(unsigned);
+
+/*
+ * Load a segment. Fall back on loading the zero
+ * segment if something goes wrong..
+ */
+#define loadsegment(seg,value) \
+ asm volatile("\n" \
+ "1:\t" \
+ "movl %k0,%%" #seg "\n" \
+ "2:\n" \
+ ".section .fixup,\"ax\"\n" \
+ "3:\t" \
+ "movl %1,%%" #seg "\n\t" \
+ "jmp 2b\n" \
+ ".previous\n" \
+ ".section __ex_table,\"a\"\n\t" \
+ ".align 8\n\t" \
+ ".quad 1b,3b\n" \
+ ".previous" \
+ : :"r" (value), "r" (0))
+
+/*
+ * Clear and set 'TS' bit respectively
+ */
+#define clts() (HYPERVISOR_fpu_taskswitch(0))
+
+static inline unsigned long read_cr0(void)
+{
+ unsigned long cr0;
+ asm volatile("movq %%cr0,%0" : "=r" (cr0));
+ return cr0;
+}
+
+static inline void write_cr0(unsigned long val)
+{
+ asm volatile("movq %0,%%cr0" :: "r" (val));
+}
+
+#define read_cr3() ({ \
+ unsigned long __dummy; \
+ asm("movq %%cr3,%0" : "=r" (__dummy)); \
+ machine_to_phys(__dummy); \
+})
+
+static inline unsigned long read_cr4(void)
+{
+ unsigned long cr4;
+ asm("movq %%cr4,%0" : "=r" (cr4));
+ return cr4;
+}
+
+static inline void write_cr4(unsigned long val)
+{
+ asm volatile("movq %0,%%cr4" :: "r" (val));
+}
+
+#define stts() (HYPERVISOR_fpu_taskswitch(1))
+
+#define wbinvd() \
+ __asm__ __volatile__ ("wbinvd": : :"memory");
+
+/*
+ * On SMP systems, when the scheduler does migration-cost autodetection,
+ * it needs a way to flush as much of the CPU's caches as possible.
+ */
+static inline void sched_cacheflush(void)
+{
+ wbinvd();
+}
+
+#endif /* __KERNEL__ */
+
+#define nop() __asm__ __volatile__ ("nop")
+
+#define xchg(ptr,v) ((__typeof__(*(ptr)))__xchg((unsigned long)(v),(ptr),sizeof(*(ptr))))
+
+#define tas(ptr) (xchg((ptr),1))
+
+#define __xg(x) ((volatile long *)(x))
+
+static inline void set_64bit(volatile unsigned long *ptr, unsigned long val)
+{
+ *ptr = val;
+}
+
+#define _set_64bit set_64bit
+
+/*
+ * Note: no "lock" prefix even on SMP: xchg always implies lock anyway
+ * Note 2: xchg has side effect, so that attribute volatile is necessary,
+ * but generally the primitive is invalid, *ptr is output argument. --ANK
+ */
+static inline unsigned long __xchg(unsigned long x, volatile void * ptr, int size)
+{
+ switch (size) {
+ case 1:
+ __asm__ __volatile__("xchgb %b0,%1"
+ :"=q" (x)
+ :"m" (*__xg(ptr)), "0" (x)
+ :"memory");
+ break;
+ case 2:
+ __asm__ __volatile__("xchgw %w0,%1"
+ :"=r" (x)
+ :"m" (*__xg(ptr)), "0" (x)
+ :"memory");
+ break;
+ case 4:
+ __asm__ __volatile__("xchgl %k0,%1"
+ :"=r" (x)
+ :"m" (*__xg(ptr)), "0" (x)
+ :"memory");
+ break;
+ case 8:
+ __asm__ __volatile__("xchgq %0,%1"
+ :"=r" (x)
+ :"m" (*__xg(ptr)), "0" (x)
+ :"memory");
+ break;
+ }
+ return x;
+}
+
+/*
+ * Atomic compare and exchange. Compare OLD with MEM, if identical,
+ * store NEW in MEM. Return the initial value in MEM. Success is
+ * indicated by comparing RETURN with OLD.
+ */
+
+#define __HAVE_ARCH_CMPXCHG 1
+
+static inline unsigned long __cmpxchg(volatile void *ptr, unsigned long old,
+ unsigned long new, int size)
+{
+ unsigned long prev;
+ switch (size) {
+ case 1:
+ __asm__ __volatile__(LOCK_PREFIX "cmpxchgb %b1,%2"
+ : "=a"(prev)
+ : "q"(new), "m"(*__xg(ptr)), "0"(old)
+ : "memory");
+ return prev;
+ case 2:
+ __asm__ __volatile__(LOCK_PREFIX "cmpxchgw %w1,%2"
+ : "=a"(prev)
+ : "r"(new), "m"(*__xg(ptr)), "0"(old)
+ : "memory");
+ return prev;
+ case 4:
+ __asm__ __volatile__(LOCK_PREFIX "cmpxchgl %k1,%2"
+ : "=a"(prev)
+ : "r"(new), "m"(*__xg(ptr)), "0"(old)
+ : "memory");
+ return prev;
+ case 8:
+ __asm__ __volatile__(LOCK_PREFIX "cmpxchgq %1,%2"
+ : "=a"(prev)
+ : "r"(new), "m"(*__xg(ptr)), "0"(old)
+ : "memory");
+ return prev;
+ }
+ return old;
+}
+
+#define cmpxchg(ptr,o,n)\
+ ((__typeof__(*(ptr)))__cmpxchg((ptr),(unsigned long)(o),\
+ (unsigned long)(n),sizeof(*(ptr))))
+
+#ifdef CONFIG_SMP
+#define smp_mb() mb()
+#define smp_rmb() rmb()
+#define smp_wmb() wmb()
+#define smp_read_barrier_depends() do {} while(0)
+#else
+#define smp_mb() barrier()
+#define smp_rmb() barrier()
+#define smp_wmb() barrier()
+#define smp_read_barrier_depends() do {} while(0)
+#endif
+
+
+/*
+ * Force strict CPU ordering.
+ * And yes, this is required on UP too when we're talking
+ * to devices.
+ */
+#define mb() asm volatile("mfence":::"memory")
+#define rmb() asm volatile("lfence":::"memory")
+
+#ifdef CONFIG_UNORDERED_IO
+#define wmb() asm volatile("sfence" ::: "memory")
+#else
+#define wmb() asm volatile("" ::: "memory")
+#endif
+#define read_barrier_depends() do {} while(0)
+#define set_mb(var, value) do { (void) xchg(&var, value); } while (0)
+
+#define warn_if_not_ulong(x) do { unsigned long foo; (void) (&(x) == &foo); } while (0)
+
+#include <linux/irqflags.h>
+
+void cpu_idle_wait(void);
+
+extern unsigned long arch_align_stack(unsigned long sp);
+extern void free_init_pages(char *what, unsigned long begin, unsigned long end);
+
+#endif
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/include/mach-xen/asm/tlbflush_64.h 2007-11-26 16:59:25.000000000 +0100
@@ -0,0 +1,103 @@
+#ifndef _X8664_TLBFLUSH_H
+#define _X8664_TLBFLUSH_H
+
+#include <linux/mm.h>
+#include <asm/processor.h>
+
+#define __flush_tlb() xen_tlb_flush()
+
+/*
+ * Global pages have to be flushed a bit differently. Not a real
+ * performance problem because this does not happen often.
+ */
+#define __flush_tlb_global() xen_tlb_flush()
+
+
+extern unsigned long pgkern_mask;
+
+#define __flush_tlb_all() __flush_tlb_global()
+
+#define __flush_tlb_one(addr) xen_invlpg((unsigned long)addr)
+
+
+/*
+ * TLB flushing:
+ *
+ * - flush_tlb() flushes the current mm struct TLBs
+ * - flush_tlb_all() flushes all processes TLBs
+ * - flush_tlb_mm(mm) flushes the specified mm context TLB's
+ * - flush_tlb_page(vma, vmaddr) flushes one page
+ * - flush_tlb_range(vma, start, end) flushes a range of pages
+ * - flush_tlb_kernel_range(start, end) flushes a range of kernel pages
+ * - flush_tlb_pgtables(mm, start, end) flushes a range of page tables
+ *
+ * x86-64 can only flush individual pages or full VMs. For a range flush
+ * we always do the full VM. Might be worth trying if for a small
+ * range a few INVLPGs in a row are a win.
+ */
+
+#ifndef CONFIG_SMP
+
+#define flush_tlb() __flush_tlb()
+#define flush_tlb_all() __flush_tlb_all()
+#define local_flush_tlb() __flush_tlb()
+
+static inline void flush_tlb_mm(struct mm_struct *mm)
+{
+ if (mm == current->active_mm)
+ __flush_tlb();
+}
+
+static inline void flush_tlb_page(struct vm_area_struct *vma,
+ unsigned long addr)
+{
+ if (vma->vm_mm == current->active_mm)
+ __flush_tlb_one(addr);
+}
+
+static inline void flush_tlb_range(struct vm_area_struct *vma,
+ unsigned long start, unsigned long end)
+{
+ if (vma->vm_mm == current->active_mm)
+ __flush_tlb();
+}
+
+#else
+
+#include <asm/smp.h>
+
+#define local_flush_tlb() \
+ __flush_tlb()
+
+#define flush_tlb_all xen_tlb_flush_all
+#define flush_tlb_current_task() xen_tlb_flush_mask(&current->mm->cpu_vm_mask)
+#define flush_tlb_mm(mm) xen_tlb_flush_mask(&(mm)->cpu_vm_mask)
+#define flush_tlb_page(vma, va) xen_invlpg_mask(&(vma)->vm_mm->cpu_vm_mask, va)
+
+#define flush_tlb() flush_tlb_current_task()
+
+static inline void flush_tlb_range(struct vm_area_struct * vma, unsigned long start, unsigned long end)
+{
+ flush_tlb_mm(vma->vm_mm);
+}
+
+#define TLBSTATE_OK 1
+#define TLBSTATE_LAZY 2
+
+/* Roughly an IPI every 20MB with 4k pages for freeing page table
+ ranges. Cost is about 42k of memory for each CPU. */
+#define ARCH_FREE_PTE_NR 5350
+
+#endif
+
+#define flush_tlb_kernel_range(start, end) flush_tlb_all()
+
+static inline void flush_tlb_pgtables(struct mm_struct *mm,
+ unsigned long start, unsigned long end)
+{
+ /* x86_64 does not keep any page table caches in a software TLB.
+ The CPUs do in their hardware TLBs, but they are handled
+ by the normal TLB flushing algorithms. */
+}
+
+#endif /* _X8664_TLBFLUSH_H */
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/include/mach-xen/asm/xor_64.h 2007-06-12 13:14:13.000000000 +0200
@@ -0,0 +1,328 @@
+/*
+ * x86-64 changes / gcc fixes from Andi Kleen.
+ * Copyright 2002 Andi Kleen, SuSE Labs.
+ *
+ * This hasn't been optimized for the hammer yet, but there are likely
+ * no advantages to be gotten from x86-64 here anyways.
+ */
+
+typedef struct { unsigned long a,b; } __attribute__((aligned(16))) xmm_store_t;
+
+/* Doesn't use gcc to save the XMM registers, because there is no easy way to
+ tell it to do a clts before the register saving. */
+#define XMMS_SAVE do { \
+ preempt_disable(); \
+ if (!(current_thread_info()->status & TS_USEDFPU)) \
+ clts(); \
+ __asm__ __volatile__ ( \
+ "movups %%xmm0,(%1) ;\n\t" \
+ "movups %%xmm1,0x10(%1) ;\n\t" \
+ "movups %%xmm2,0x20(%1) ;\n\t" \
+ "movups %%xmm3,0x30(%1) ;\n\t" \
+ : "=&r" (cr0) \
+ : "r" (xmm_save) \
+ : "memory"); \
+} while(0)
+
+#define XMMS_RESTORE do { \
+ asm volatile ( \
+ "sfence ;\n\t" \
+ "movups (%1),%%xmm0 ;\n\t" \
+ "movups 0x10(%1),%%xmm1 ;\n\t" \
+ "movups 0x20(%1),%%xmm2 ;\n\t" \
+ "movups 0x30(%1),%%xmm3 ;\n\t" \
+ : \
+ : "r" (cr0), "r" (xmm_save) \
+ : "memory"); \
+ if (!(current_thread_info()->status & TS_USEDFPU)) \
+ stts(); \
+ preempt_enable(); \
+} while(0)
+
+#define OFFS(x) "16*("#x")"
+#define PF_OFFS(x) "256+16*("#x")"
+#define PF0(x) " prefetchnta "PF_OFFS(x)"(%[p1]) ;\n"
+#define LD(x,y) " movaps "OFFS(x)"(%[p1]), %%xmm"#y" ;\n"
+#define ST(x,y) " movaps %%xmm"#y", "OFFS(x)"(%[p1]) ;\n"
+#define PF1(x) " prefetchnta "PF_OFFS(x)"(%[p2]) ;\n"
+#define PF2(x) " prefetchnta "PF_OFFS(x)"(%[p3]) ;\n"
+#define PF3(x) " prefetchnta "PF_OFFS(x)"(%[p4]) ;\n"
+#define PF4(x) " prefetchnta "PF_OFFS(x)"(%[p5]) ;\n"
+#define PF5(x) " prefetchnta "PF_OFFS(x)"(%[p6]) ;\n"
+#define XO1(x,y) " xorps "OFFS(x)"(%[p2]), %%xmm"#y" ;\n"
+#define XO2(x,y) " xorps "OFFS(x)"(%[p3]), %%xmm"#y" ;\n"
+#define XO3(x,y) " xorps "OFFS(x)"(%[p4]), %%xmm"#y" ;\n"
+#define XO4(x,y) " xorps "OFFS(x)"(%[p5]), %%xmm"#y" ;\n"
+#define XO5(x,y) " xorps "OFFS(x)"(%[p6]), %%xmm"#y" ;\n"
+
+
+static void
+xor_sse_2(unsigned long bytes, unsigned long *p1, unsigned long *p2)
+{
+ unsigned int lines = bytes >> 8;
+ unsigned long cr0;
+ xmm_store_t xmm_save[4];
+
+ XMMS_SAVE;
+
+ asm volatile (
+#undef BLOCK
+#define BLOCK(i) \
+ LD(i,0) \
+ LD(i+1,1) \
+ PF1(i) \
+ PF1(i+2) \
+ LD(i+2,2) \
+ LD(i+3,3) \
+ PF0(i+4) \
+ PF0(i+6) \
+ XO1(i,0) \
+ XO1(i+1,1) \
+ XO1(i+2,2) \
+ XO1(i+3,3) \
+ ST(i,0) \
+ ST(i+1,1) \
+ ST(i+2,2) \
+ ST(i+3,3) \
+
+
+ PF0(0)
+ PF0(2)
+
+ " .align 32 ;\n"
+ " 1: ;\n"
+
+ BLOCK(0)
+ BLOCK(4)
+ BLOCK(8)
+ BLOCK(12)
+
+ " addq %[inc], %[p1] ;\n"
+ " addq %[inc], %[p2] ;\n"
+ " decl %[cnt] ; jnz 1b"
+ : [p1] "+r" (p1), [p2] "+r" (p2), [cnt] "+r" (lines)
+ : [inc] "r" (256UL)
+ : "memory");
+
+ XMMS_RESTORE;
+}
+
+static void
+xor_sse_3(unsigned long bytes, unsigned long *p1, unsigned long *p2,
+ unsigned long *p3)
+{
+ unsigned int lines = bytes >> 8;
+ xmm_store_t xmm_save[4];
+ unsigned long cr0;
+
+ XMMS_SAVE;
+
+ __asm__ __volatile__ (
+#undef BLOCK
+#define BLOCK(i) \
+ PF1(i) \
+ PF1(i+2) \
+ LD(i,0) \
+ LD(i+1,1) \
+ LD(i+2,2) \
+ LD(i+3,3) \
+ PF2(i) \
+ PF2(i+2) \
+ PF0(i+4) \
+ PF0(i+6) \
+ XO1(i,0) \
+ XO1(i+1,1) \
+ XO1(i+2,2) \
+ XO1(i+3,3) \
+ XO2(i,0) \
+ XO2(i+1,1) \
+ XO2(i+2,2) \
+ XO2(i+3,3) \
+ ST(i,0) \
+ ST(i+1,1) \
+ ST(i+2,2) \
+ ST(i+3,3) \
+
+
+ PF0(0)
+ PF0(2)
+
+ " .align 32 ;\n"
+ " 1: ;\n"
+
+ BLOCK(0)
+ BLOCK(4)
+ BLOCK(8)
+ BLOCK(12)
+
+ " addq %[inc], %[p1] ;\n"
+ " addq %[inc], %[p2] ;\n"
+ " addq %[inc], %[p3] ;\n"
+ " decl %[cnt] ; jnz 1b"
+ : [cnt] "+r" (lines),
+ [p1] "+r" (p1), [p2] "+r" (p2), [p3] "+r" (p3)
+ : [inc] "r" (256UL)
+ : "memory");
+ XMMS_RESTORE;
+}
+
+static void
+xor_sse_4(unsigned long bytes, unsigned long *p1, unsigned long *p2,
+ unsigned long *p3, unsigned long *p4)
+{
+ unsigned int lines = bytes >> 8;
+ xmm_store_t xmm_save[4];
+ unsigned long cr0;
+
+ XMMS_SAVE;
+
+ __asm__ __volatile__ (
+#undef BLOCK
+#define BLOCK(i) \
+ PF1(i) \
+ PF1(i+2) \
+ LD(i,0) \
+ LD(i+1,1) \
+ LD(i+2,2) \
+ LD(i+3,3) \
+ PF2(i) \
+ PF2(i+2) \
+ XO1(i,0) \
+ XO1(i+1,1) \
+ XO1(i+2,2) \
+ XO1(i+3,3) \
+ PF3(i) \
+ PF3(i+2) \
+ PF0(i+4) \
+ PF0(i+6) \
+ XO2(i,0) \
+ XO2(i+1,1) \
+ XO2(i+2,2) \
+ XO2(i+3,3) \
+ XO3(i,0) \
+ XO3(i+1,1) \
+ XO3(i+2,2) \
+ XO3(i+3,3) \
+ ST(i,0) \
+ ST(i+1,1) \
+ ST(i+2,2) \
+ ST(i+3,3) \
+
+
+ PF0(0)
+ PF0(2)
+
+ " .align 32 ;\n"
+ " 1: ;\n"
+
+ BLOCK(0)
+ BLOCK(4)
+ BLOCK(8)
+ BLOCK(12)
+
+ " addq %[inc], %[p1] ;\n"
+ " addq %[inc], %[p2] ;\n"
+ " addq %[inc], %[p3] ;\n"
+ " addq %[inc], %[p4] ;\n"
+ " decl %[cnt] ; jnz 1b"
+ : [cnt] "+c" (lines),
+ [p1] "+r" (p1), [p2] "+r" (p2), [p3] "+r" (p3), [p4] "+r" (p4)
+ : [inc] "r" (256UL)
+ : "memory" );
+
+ XMMS_RESTORE;
+}
+
+static void
+xor_sse_5(unsigned long bytes, unsigned long *p1, unsigned long *p2,
+ unsigned long *p3, unsigned long *p4, unsigned long *p5)
+{
+ unsigned int lines = bytes >> 8;
+ xmm_store_t xmm_save[4];
+ unsigned long cr0;
+
+ XMMS_SAVE;
+
+ __asm__ __volatile__ (
+#undef BLOCK
+#define BLOCK(i) \
+ PF1(i) \
+ PF1(i+2) \
+ LD(i,0) \
+ LD(i+1,1) \
+ LD(i+2,2) \
+ LD(i+3,3) \
+ PF2(i) \
+ PF2(i+2) \
+ XO1(i,0) \
+ XO1(i+1,1) \
+ XO1(i+2,2) \
+ XO1(i+3,3) \
+ PF3(i) \
+ PF3(i+2) \
+ XO2(i,0) \
+ XO2(i+1,1) \
+ XO2(i+2,2) \
+ XO2(i+3,3) \
+ PF4(i) \
+ PF4(i+2) \
+ PF0(i+4) \
+ PF0(i+6) \
+ XO3(i,0) \
+ XO3(i+1,1) \
+ XO3(i+2,2) \
+ XO3(i+3,3) \
+ XO4(i,0) \
+ XO4(i+1,1) \
+ XO4(i+2,2) \
+ XO4(i+3,3) \
+ ST(i,0) \
+ ST(i+1,1) \
+ ST(i+2,2) \
+ ST(i+3,3) \
+
+
+ PF0(0)
+ PF0(2)
+
+ " .align 32 ;\n"
+ " 1: ;\n"
+
+ BLOCK(0)
+ BLOCK(4)
+ BLOCK(8)
+ BLOCK(12)
+
+ " addq %[inc], %[p1] ;\n"
+ " addq %[inc], %[p2] ;\n"
+ " addq %[inc], %[p3] ;\n"
+ " addq %[inc], %[p4] ;\n"
+ " addq %[inc], %[p5] ;\n"
+ " decl %[cnt] ; jnz 1b"
+ : [cnt] "+c" (lines),
+ [p1] "+r" (p1), [p2] "+r" (p2), [p3] "+r" (p3), [p4] "+r" (p4),
+ [p5] "+r" (p5)
+ : [inc] "r" (256UL)
+ : "memory");
+
+ XMMS_RESTORE;
+}
+
+static struct xor_block_template xor_block_sse = {
+ .name = "generic_sse",
+ .do_2 = xor_sse_2,
+ .do_3 = xor_sse_3,
+ .do_4 = xor_sse_4,
+ .do_5 = xor_sse_5,
+};
+
+#undef XOR_TRY_TEMPLATES
+#define XOR_TRY_TEMPLATES \
+ do { \
+ xor_speed(&xor_block_sse); \
+ } while (0)
+
+/* We force the use of the SSE xor block because it can write around L2.
+ We may also be able to load into the L1 only depending on how the cpu
+ deals with a load to a line that is being prefetched. */
+#define XOR_SELECT_TEMPLATE(FASTEST) (&xor_block_sse)
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/include/mach-xen/setup_arch_post.h 2007-06-12 13:14:13.000000000 +0200
@@ -0,0 +1,63 @@
+/**
+ * machine_specific_* - Hooks for machine specific setup.
+ *
+ * Description:
+ * This is included late in kernel/setup.c so that it can make
+ * use of all of the static functions.
+ **/
+
+#include <xen/interface/callback.h>
+
+extern void hypervisor_callback(void);
+extern void failsafe_callback(void);
+extern void nmi(void);
+
+static void __init machine_specific_arch_setup(void)
+{
+ int ret;
+ static struct callback_register __initdata event = {
+ .type = CALLBACKTYPE_event,
+ .address = (unsigned long) hypervisor_callback,
+ };
+ static struct callback_register __initdata failsafe = {
+ .type = CALLBACKTYPE_failsafe,
+ .address = (unsigned long)failsafe_callback,
+ };
+ static struct callback_register __initdata syscall = {
+ .type = CALLBACKTYPE_syscall,
+ .address = (unsigned long)system_call,
+ };
+#ifdef CONFIG_X86_LOCAL_APIC
+ static struct callback_register __initdata nmi_cb = {
+ .type = CALLBACKTYPE_nmi,
+ .address = (unsigned long)nmi,
+ };
+#endif
+
+ ret = HYPERVISOR_callback_op(CALLBACKOP_register, &event);
+ if (ret == 0)
+ ret = HYPERVISOR_callback_op(CALLBACKOP_register, &failsafe);
+ if (ret == 0)
+ ret = HYPERVISOR_callback_op(CALLBACKOP_register, &syscall);
+#if CONFIG_XEN_COMPAT <= 0x030002
+ if (ret == -ENOSYS)
+ ret = HYPERVISOR_set_callbacks(
+ event.address,
+ failsafe.address,
+ syscall.address);
+#endif
+ BUG_ON(ret);
+
+#ifdef CONFIG_X86_LOCAL_APIC
+ ret = HYPERVISOR_callback_op(CALLBACKOP_register, &nmi_cb);
+#if CONFIG_XEN_COMPAT <= 0x030002
+ if (ret == -ENOSYS) {
+ static struct xennmi_callback __initdata cb = {
+ .handler_address = (unsigned long)nmi
+ };
+
+ HYPERVISOR_nmi_op(XENNMI_register_callback, &cb);
+ }
+#endif
+#endif
+}
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/arch/x86/include/mach-xen/setup_arch_pre.h 2007-06-12 13:14:13.000000000 +0200
@@ -0,0 +1,5 @@
+/* Hook to call BIOS initialisation function */
+
+#define ARCH_SETUP machine_specific_arch_setup();
+
+static void __init machine_specific_arch_setup(void);
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/include/xen/balloon.h 2007-06-12 13:14:19.000000000 +0200
@@ -0,0 +1,57 @@
+/******************************************************************************
+ * balloon.h
+ *
+ * Xen balloon driver - enables returning/claiming memory to/from Xen.
+ *
+ * Copyright (c) 2003, B Dragovic
+ * Copyright (c) 2003-2004, M Williamson, K Fraser
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License version 2
+ * as published by the Free Software Foundation; or, when distributed
+ * separately from the Linux kernel or incorporated into other
+ * software packages, subject to the following license:
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this source file (the "Software"), to deal in the Software without
+ * restriction, including without limitation the rights to use, copy, modify,
+ * merge, publish, distribute, sublicense, and/or sell copies of the Software,
+ * and to permit persons to whom the Software is furnished to do so, subject to
+ * the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
+ * IN THE SOFTWARE.
+ */
+
+#ifndef __ASM_BALLOON_H__
+#define __ASM_BALLOON_H__
+
+/*
+ * Inform the balloon driver that it should allow some slop for device-driver
+ * memory activities.
+ */
+void balloon_update_driver_allowance(long delta);
+
+/* Allocate/free a set of empty pages in low memory (i.e., no RAM mapped). */
+struct page **alloc_empty_pages_and_pagevec(int nr_pages);
+void free_empty_pages_and_pagevec(struct page **pagevec, int nr_pages);
+
+void balloon_release_driver_page(struct page *page);
+
+/*
+ * Prevent the balloon driver from changing the memory reservation during
+ * a driver critical region.
+ */
+extern spinlock_t balloon_lock;
+#define balloon_lock(__flags) spin_lock_irqsave(&balloon_lock, __flags)
+#define balloon_unlock(__flags) spin_unlock_irqrestore(&balloon_lock, __flags)
+
+#endif /* __ASM_BALLOON_H__ */
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/include/xen/blkif.h 2008-07-21 11:00:33.000000000 +0200
@@ -0,0 +1,123 @@
+/*
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this software and associated documentation files (the "Software"), to
+ * deal in the Software without restriction, including without limitation the
+ * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or
+ * sell copies of the Software, and to permit persons to whom the Software is
+ * furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
+ * DEALINGS IN THE SOFTWARE.
+ */
+
+#ifndef __XEN_BLKIF_H__
+#define __XEN_BLKIF_H__
+
+#include <xen/interface/io/ring.h>
+#include <xen/interface/io/blkif.h>
+#include <xen/interface/io/protocols.h>
+
+/* Not a real protocol. Used to generate ring structs which contain
+ * the elements common to all protocols only. This way we get a
+ * compiler-checkable way to use common struct elements, so we can
+ * avoid using switch(protocol) in a number of places. */
+struct blkif_common_request {
+ char dummy;
+};
+struct blkif_common_response {
+ char dummy;
+};
+
+/* i386 protocol version */
+#pragma pack(push, 4)
+struct blkif_x86_32_request {
+ uint8_t operation; /* BLKIF_OP_??? */
+ uint8_t nr_segments; /* number of segments */
+ blkif_vdev_t handle; /* only for read/write requests */
+ uint64_t id; /* private guest value, echoed in resp */
+ blkif_sector_t sector_number;/* start sector idx on disk (r/w only) */
+ struct blkif_request_segment seg[BLKIF_MAX_SEGMENTS_PER_REQUEST];
+};
+struct blkif_x86_32_response {
+ uint64_t id; /* copied from request */
+ uint8_t operation; /* copied from request */
+ int16_t status; /* BLKIF_RSP_??? */
+};
+typedef struct blkif_x86_32_request blkif_x86_32_request_t;
+typedef struct blkif_x86_32_response blkif_x86_32_response_t;
+#pragma pack(pop)
+
+/* x86_64 protocol version */
+struct blkif_x86_64_request {
+ uint8_t operation; /* BLKIF_OP_??? */
+ uint8_t nr_segments; /* number of segments */
+ blkif_vdev_t handle; /* only for read/write requests */
+ uint64_t __attribute__((__aligned__(8))) id;
+ blkif_sector_t sector_number;/* start sector idx on disk (r/w only) */
+ struct blkif_request_segment seg[BLKIF_MAX_SEGMENTS_PER_REQUEST];
+};
+struct blkif_x86_64_response {
+ uint64_t __attribute__((__aligned__(8))) id;
+ uint8_t operation; /* copied from request */
+ int16_t status; /* BLKIF_RSP_??? */
+};
+typedef struct blkif_x86_64_request blkif_x86_64_request_t;
+typedef struct blkif_x86_64_response blkif_x86_64_response_t;
+
+DEFINE_RING_TYPES(blkif_common, struct blkif_common_request, struct blkif_common_response);
+DEFINE_RING_TYPES(blkif_x86_32, struct blkif_x86_32_request, struct blkif_x86_32_response);
+DEFINE_RING_TYPES(blkif_x86_64, struct blkif_x86_64_request, struct blkif_x86_64_response);
+
+union blkif_back_rings {
+ blkif_back_ring_t native;
+ blkif_common_back_ring_t common;
+ blkif_x86_32_back_ring_t x86_32;
+ blkif_x86_64_back_ring_t x86_64;
+};
+typedef union blkif_back_rings blkif_back_rings_t;
+
+enum blkif_protocol {
+ BLKIF_PROTOCOL_NATIVE = 1,
+ BLKIF_PROTOCOL_X86_32 = 2,
+ BLKIF_PROTOCOL_X86_64 = 3,
+};
+
+static void inline blkif_get_x86_32_req(blkif_request_t *dst, blkif_x86_32_request_t *src)
+{
+ int i, n = BLKIF_MAX_SEGMENTS_PER_REQUEST;
+ dst->operation = src->operation;
+ dst->nr_segments = src->nr_segments;
+ dst->handle = src->handle;
+ dst->id = src->id;
+ dst->sector_number = src->sector_number;
+ barrier();
+ if (n > dst->nr_segments)
+ n = dst->nr_segments;
+ for (i = 0; i < n; i++)
+ dst->seg[i] = src->seg[i];
+}
+
+static void inline blkif_get_x86_64_req(blkif_request_t *dst, blkif_x86_64_request_t *src)
+{
+ int i, n = BLKIF_MAX_SEGMENTS_PER_REQUEST;
+ dst->operation = src->operation;
+ dst->nr_segments = src->nr_segments;
+ dst->handle = src->handle;
+ dst->id = src->id;
+ dst->sector_number = src->sector_number;
+ barrier();
+ if (n > dst->nr_segments)
+ n = dst->nr_segments;
+ for (i = 0; i < n; i++)
+ dst->seg[i] = src->seg[i];
+}
+
+#endif /* __XEN_BLKIF_H__ */
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/include/xen/compat_ioctl.h 2010-01-18 15:23:12.000000000 +0100
@@ -0,0 +1,75 @@
+/*
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation; either version 2 of the
+ * License, or (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
+ *
+ * Copyright IBM Corp. 2007
+ *
+ * Authors: Jimi Xenidis <jimix@watson.ibm.com>
+ * Hollis Blanchard <hollisb@us.ibm.com>
+ */
+
+#ifndef __LINUX_XEN_COMPAT_H__
+#define __LINUX_XEN_COMPAT_H__
+
+#include <linux/compat.h>
+#include <linux/compiler.h>
+
+#if defined(CONFIG_X86) || defined(CONFIG_IA64)
+#define xen_pfn32_t __u32
+#endif
+
+extern int privcmd_ioctl_32(int fd, unsigned int cmd, unsigned long arg);
+struct privcmd_mmap_32 {
+ int num;
+ domid_t dom;
+ compat_uptr_t entry;
+};
+
+struct privcmd_mmapbatch_32 {
+ int num; /* number of pages to populate */
+ domid_t dom; /* target domain */
+#if defined(CONFIG_X86) || defined(CONFIG_IA64)
+ union { /* virtual address */
+ __u64 addr __attribute__((packed));
+ __u32 va; /* ensures union is 4-byte aligned */
+ };
+#else
+ __u64 addr; /* virtual address */
+#endif
+ compat_uptr_t arr; /* array of mfns - top nibble set on err */
+};
+
+struct privcmd_mmapbatch_v2_32 {
+ unsigned int num; /* number of pages to populate */
+ domid_t dom; /* target domain */
+#if defined(CONFIG_X86) || defined(CONFIG_IA64)
+ union { /* virtual address */
+ __u64 addr __attribute__((packed));
+ __u32 va; /* ensures union is 4-byte aligned */
+ };
+#else
+ __u64 addr; /* virtual address */
+#endif
+ compat_uptr_t arr; /* array of mfns */
+ compat_uptr_t err; /* array of error codes */
+};
+
+#define IOCTL_PRIVCMD_MMAP_32 \
+ _IOC(_IOC_NONE, 'P', 2, sizeof(struct privcmd_mmap_32))
+#define IOCTL_PRIVCMD_MMAPBATCH_32 \
+ _IOC(_IOC_NONE, 'P', 3, sizeof(struct privcmd_mmapbatch_32))
+#define IOCTL_PRIVCMD_MMAPBATCH_V2_32 \
+ _IOC(_IOC_NONE, 'P', 4, sizeof(struct privcmd_mmapbatch_v2_32))
+
+#endif /* __LINUX_XEN_COMPAT_H__ */
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/include/xen/cpu_hotplug.h 2007-08-16 18:07:01.000000000 +0200
@@ -0,0 +1,41 @@
+#ifndef __XEN_CPU_HOTPLUG_H__
+#define __XEN_CPU_HOTPLUG_H__
+
+#include <linux/kernel.h>
+#include <linux/cpumask.h>
+
+#if defined(CONFIG_X86) && defined(CONFIG_SMP)
+extern cpumask_t cpu_initialized_map;
+#endif
+
+#if defined(CONFIG_HOTPLUG_CPU)
+
+int cpu_up_check(unsigned int cpu);
+void init_xenbus_allowed_cpumask(void);
+int smp_suspend(void);
+void smp_resume(void);
+
+void cpu_bringup(void);
+
+#else /* !defined(CONFIG_HOTPLUG_CPU) */
+
+#define cpu_up_check(cpu) (0)
+#define init_xenbus_allowed_cpumask() ((void)0)
+
+static inline int smp_suspend(void)
+{
+ if (num_online_cpus() > 1) {
+ printk(KERN_WARNING "Can't suspend SMP guests "
+ "without CONFIG_HOTPLUG_CPU\n");
+ return -EOPNOTSUPP;
+ }
+ return 0;
+}
+
+static inline void smp_resume(void)
+{
+}
+
+#endif /* !defined(CONFIG_HOTPLUG_CPU) */
+
+#endif /* __XEN_CPU_HOTPLUG_H__ */
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/include/xen/driver_util.h 2007-06-12 13:14:19.000000000 +0200
@@ -0,0 +1,14 @@
+
+#ifndef __ASM_XEN_DRIVER_UTIL_H__
+#define __ASM_XEN_DRIVER_UTIL_H__
+
+#include <linux/vmalloc.h>
+#include <linux/device.h>
+
+/* Allocate/destroy a 'vmalloc' VM area. */
+extern struct vm_struct *alloc_vm_area(unsigned long size);
+extern void free_vm_area(struct vm_struct *area);
+
+extern struct class *get_xen_class(void);
+
+#endif /* __ASM_XEN_DRIVER_UTIL_H__ */
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/include/xen/firmware.h 2007-07-02 08:16:19.000000000 +0200
@@ -0,0 +1,10 @@
+#ifndef __XEN_FIRMWARE_H__
+#define __XEN_FIRMWARE_H__
+
+#if defined(CONFIG_EDD) || defined(CONFIG_EDD_MODULE)
+void copy_edd(void);
+#endif
+
+void copy_edid(void);
+
+#endif /* __XEN_FIRMWARE_H__ */
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/include/xen/gnttab.h 2010-09-23 15:39:04.000000000 +0200
@@ -0,0 +1,202 @@
+/******************************************************************************
+ * gnttab.h
+ *
+ * Two sets of functionality:
+ * 1. Granting foreign access to our memory reservation.
+ * 2. Accessing others' memory reservations via grant references.
+ * (i.e., mechanisms for both sender and recipient of grant references)
+ *
+ * Copyright (c) 2004-2005, K A Fraser
+ * Copyright (c) 2005, Christopher Clark
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License version 2
+ * as published by the Free Software Foundation; or, when distributed
+ * separately from the Linux kernel or incorporated into other
+ * software packages, subject to the following license:
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this source file (the "Software"), to deal in the Software without
+ * restriction, including without limitation the rights to use, copy, modify,
+ * merge, publish, distribute, sublicense, and/or sell copies of the Software,
+ * and to permit persons to whom the Software is furnished to do so, subject to
+ * the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
+ * IN THE SOFTWARE.
+ */
+
+#ifndef __ASM_GNTTAB_H__
+#define __ASM_GNTTAB_H__
+
+#include <asm/hypervisor.h>
+#include <asm/maddr.h> /* maddr_t */
+#include <linux/mm.h>
+#include <linux/delay.h>
+#include <xen/interface/grant_table.h>
+#include <xen/features.h>
+
+struct gnttab_free_callback {
+ struct gnttab_free_callback *next;
+ void (*fn)(void *);
+ void *arg;
+ u16 count;
+ u8 queued;
+};
+
+int gnttab_grant_foreign_access(domid_t domid, unsigned long frame,
+ int flags);
+
+/*
+ * End access through the given grant reference, iff the grant entry is no
+ * longer in use. Return 1 if the grant entry was freed, 0 if it is still in
+ * use.
+ */
+int gnttab_end_foreign_access_ref(grant_ref_t ref);
+
+/*
+ * Eventually end access through the given grant reference, and once that
+ * access has been ended, free the given page too. Access will be ended
+ * immediately iff the grant entry is not in use, otherwise it will happen
+ * some time later. page may be 0, in which case no freeing will occur.
+ */
+void gnttab_end_foreign_access(grant_ref_t ref, unsigned long page);
+
+int gnttab_grant_foreign_transfer(domid_t domid, unsigned long pfn);
+
+unsigned long gnttab_end_foreign_transfer_ref(grant_ref_t ref);
+unsigned long gnttab_end_foreign_transfer(grant_ref_t ref);
+
+int gnttab_query_foreign_access(grant_ref_t ref);
+
+/*
+ * operations on reserved batches of grant references
+ */
+int gnttab_alloc_grant_references(u16 count, grant_ref_t *pprivate_head);
+
+void gnttab_free_grant_reference(grant_ref_t ref);
+
+void gnttab_free_grant_references(grant_ref_t head);
+
+int gnttab_empty_grant_references(const grant_ref_t *pprivate_head);
+
+int gnttab_claim_grant_reference(grant_ref_t *pprivate_head);
+
+void gnttab_release_grant_reference(grant_ref_t *private_head,
+ grant_ref_t release);
+
+void gnttab_request_free_callback(struct gnttab_free_callback *callback,
+ void (*fn)(void *), void *arg, u16 count);
+void gnttab_cancel_free_callback(struct gnttab_free_callback *callback);
+
+void gnttab_grant_foreign_access_ref(grant_ref_t ref, domid_t domid,
+ unsigned long frame, int flags);
+
+void gnttab_grant_foreign_transfer_ref(grant_ref_t, domid_t domid,
+ unsigned long pfn);
+
+int gnttab_copy_grant_page(grant_ref_t ref, struct page **pagep);
+void __gnttab_dma_map_page(struct page *page);
+static inline void __gnttab_dma_unmap_page(struct page *page)
+{
+}
+
+void gnttab_reset_grant_page(struct page *page);
+
+int gnttab_suspend(void);
+int gnttab_resume(void);
+
+void *arch_gnttab_alloc_shared(unsigned long *frames);
+
+static inline void
+gnttab_set_map_op(struct gnttab_map_grant_ref *map, maddr_t addr,
+ uint32_t flags, grant_ref_t ref, domid_t domid)
+{
+ if (flags & GNTMAP_contains_pte)
+ map->host_addr = addr;
+ else if (xen_feature(XENFEAT_auto_translated_physmap))
+ map->host_addr = __pa(addr);
+ else
+ map->host_addr = addr;
+
+ map->flags = flags;
+ map->ref = ref;
+ map->dom = domid;
+}
+
+static inline void
+gnttab_set_unmap_op(struct gnttab_unmap_grant_ref *unmap, maddr_t addr,
+ uint32_t flags, grant_handle_t handle)
+{
+ if (flags & GNTMAP_contains_pte)
+ unmap->host_addr = addr;
+ else if (xen_feature(XENFEAT_auto_translated_physmap))
+ unmap->host_addr = __pa(addr);
+ else
+ unmap->host_addr = addr;
+
+ unmap->handle = handle;
+ unmap->dev_bus_addr = 0;
+}
+
+static inline void
+gnttab_set_replace_op(struct gnttab_unmap_and_replace *unmap, maddr_t addr,
+ maddr_t new_addr, grant_handle_t handle)
+{
+ if (xen_feature(XENFEAT_auto_translated_physmap)) {
+ unmap->host_addr = __pa(addr);
+ unmap->new_addr = __pa(new_addr);
+ } else {
+ unmap->host_addr = addr;
+ unmap->new_addr = new_addr;
+ }
+
+ unmap->handle = handle;
+}
+
+#define gnttab_check_GNTST_eagain_while(__HCop, __HCarg_p) \
+{ \
+ u8 __hc_delay = 1; \
+ int __ret; \
+ while (unlikely((__HCarg_p)->status == GNTST_eagain && __hc_delay)) { \
+ msleep(__hc_delay++); \
+ __ret = HYPERVISOR_grant_table_op(__HCop, (__HCarg_p), 1); \
+ BUG_ON(__ret); \
+ } \
+ if (__hc_delay == 0) { \
+ printk(KERN_ERR "%s: %s gnt busy\n", __func__, current->comm); \
+ (__HCarg_p)->status = GNTST_bad_page; \
+ } \
+ if ((__HCarg_p)->status != GNTST_okay) \
+ printk(KERN_ERR "%s: %s gnt status %x\n", \
+ __func__, current->comm, (__HCarg_p)->status); \
+}
+
+#define gnttab_check_GNTST_eagain_do_while(__HCop, __HCarg_p) \
+{ \
+ u8 __hc_delay = 1; \
+ int __ret; \
+ do { \
+ __ret = HYPERVISOR_grant_table_op(__HCop, (__HCarg_p), 1); \
+ BUG_ON(__ret); \
+ if ((__HCarg_p)->status == GNTST_eagain) \
+ msleep(__hc_delay++); \
+ } while ((__HCarg_p)->status == GNTST_eagain && __hc_delay); \
+ if (__hc_delay == 0) { \
+ printk(KERN_ERR "%s: %s gnt busy\n", __func__, current->comm); \
+ (__HCarg_p)->status = GNTST_bad_page; \
+ } \
+ if ((__HCarg_p)->status != GNTST_okay) \
+ printk(KERN_ERR "%s: %s gnt status %x\n", \
+ __func__, current->comm, (__HCarg_p)->status); \
+}
+
+#endif /* __ASM_GNTTAB_H__ */
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/include/xen/hypercall.h 2008-01-28 12:24:19.000000000 +0100
@@ -0,0 +1,30 @@
+#ifndef __XEN_HYPERCALL_H__
+#define __XEN_HYPERCALL_H__
+
+#include <asm/hypercall.h>
+
+static inline int __must_check
+HYPERVISOR_multicall_check(
+ multicall_entry_t *call_list, unsigned int nr_calls,
+ const unsigned long *rc_list)
+{
+ int rc = HYPERVISOR_multicall(call_list, nr_calls);
+
+ if (unlikely(rc < 0))
+ return rc;
+ BUG_ON(rc);
+ BUG_ON((int)nr_calls < 0);
+
+ for ( ; nr_calls > 0; --nr_calls, ++call_list)
+ if (unlikely(call_list->result != (rc_list ? *rc_list++ : 0)))
+ return nr_calls;
+
+ return 0;
+}
+
+/* A construct to ignore the return value of hypercall wrappers in a few
+ * exceptional cases (simply casting the function result to void doesn't
+ * avoid the compiler warning): */
+#define VOID(expr) ((void)((expr)?:0))
+
+#endif /* __XEN_HYPERCALL_H__ */
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/include/xen/hypervisor_sysfs.h 2007-06-22 09:08:06.000000000 +0200
@@ -0,0 +1,30 @@
+/*
+ * copyright (c) 2006 IBM Corporation
+ * Authored by: Mike D. Day <ncmike@us.ibm.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#ifndef _HYP_SYSFS_H_
+#define _HYP_SYSFS_H_
+
+#include <linux/kobject.h>
+#include <linux/sysfs.h>
+
+#define HYPERVISOR_ATTR_RO(_name) \
+static struct hyp_sysfs_attr _name##_attr = __ATTR_RO(_name)
+
+#define HYPERVISOR_ATTR_RW(_name) \
+static struct hyp_sysfs_attr _name##_attr = \
+ __ATTR(_name, 0644, _name##_show, _name##_store)
+
+struct hyp_sysfs_attr {
+ struct attribute attr;
+ ssize_t (*show)(struct hyp_sysfs_attr *, char *);
+ ssize_t (*store)(struct hyp_sysfs_attr *, const char *, size_t);
+ void *hyp_attr_data;
+};
+
+#endif /* _HYP_SYSFS_H_ */
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/include/xen/pcifront.h 2007-06-18 08:38:13.000000000 +0200
@@ -0,0 +1,83 @@
+/*
+ * PCI Frontend - arch-dependendent declarations
+ *
+ * Author: Ryan Wilson <hap9@epoch.ncsc.mil>
+ */
+#ifndef __XEN_ASM_PCIFRONT_H__
+#define __XEN_ASM_PCIFRONT_H__
+
+#include <linux/spinlock.h>
+
+#ifdef __KERNEL__
+
+#ifndef __ia64__
+
+struct pcifront_device;
+struct pci_bus;
+
+struct pcifront_sd {
+ int domain;
+ struct pcifront_device *pdev;
+};
+
+static inline struct pcifront_device *
+pcifront_get_pdev(struct pcifront_sd *sd)
+{
+ return sd->pdev;
+}
+
+static inline void pcifront_init_sd(struct pcifront_sd *sd,
+ unsigned int domain, unsigned int bus,
+ struct pcifront_device *pdev)
+{
+ sd->domain = domain;
+ sd->pdev = pdev;
+}
+
+#if defined(CONFIG_PCI_DOMAINS)
+static inline int pci_domain_nr(struct pci_bus *bus)
+{
+ struct pcifront_sd *sd = bus->sysdata;
+ return sd->domain;
+}
+static inline int pci_proc_domain(struct pci_bus *bus)
+{
+ return pci_domain_nr(bus);
+}
+#endif /* CONFIG_PCI_DOMAINS */
+
+static inline void pcifront_setup_root_resources(struct pci_bus *bus,
+ struct pcifront_sd *sd)
+{
+}
+
+#else /* __ia64__ */
+
+#include <linux/acpi.h>
+#include <asm/pci.h>
+#define pcifront_sd pci_controller
+
+extern void xen_add_resource(struct pci_controller *, unsigned int,
+ unsigned int, struct acpi_resource *);
+extern void xen_pcibios_setup_root_windows(struct pci_bus *,
+ struct pci_controller *);
+
+static inline struct pcifront_device *
+pcifront_get_pdev(struct pcifront_sd *sd)
+{
+ return (struct pcifront_device *)sd->platform_data;
+}
+
+static inline void pcifront_setup_root_resources(struct pci_bus *bus,
+ struct pcifront_sd *sd)
+{
+ xen_pcibios_setup_root_windows(bus, sd);
+}
+
+#endif /* __ia64__ */
+
+extern struct rw_semaphore pci_bus_sem;
+
+#endif /* __KERNEL__ */
+
+#endif /* __XEN_ASM_PCIFRONT_H__ */
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/include/xen/public/evtchn.h 2007-06-12 13:14:19.000000000 +0200
@@ -0,0 +1,88 @@
+/******************************************************************************
+ * evtchn.h
+ *
+ * Interface to /dev/xen/evtchn.
+ *
+ * Copyright (c) 2003-2005, K A Fraser
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License version 2
+ * as published by the Free Software Foundation; or, when distributed
+ * separately from the Linux kernel or incorporated into other
+ * software packages, subject to the following license:
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this source file (the "Software"), to deal in the Software without
+ * restriction, including without limitation the rights to use, copy, modify,
+ * merge, publish, distribute, sublicense, and/or sell copies of the Software,
+ * and to permit persons to whom the Software is furnished to do so, subject to
+ * the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
+ * IN THE SOFTWARE.
+ */
+
+#ifndef __LINUX_PUBLIC_EVTCHN_H__
+#define __LINUX_PUBLIC_EVTCHN_H__
+
+/*
+ * Bind a fresh port to VIRQ @virq.
+ * Return allocated port.
+ */
+#define IOCTL_EVTCHN_BIND_VIRQ \
+ _IOC(_IOC_NONE, 'E', 0, sizeof(struct ioctl_evtchn_bind_virq))
+struct ioctl_evtchn_bind_virq {
+ unsigned int virq;
+};
+
+/*
+ * Bind a fresh port to remote <@remote_domain, @remote_port>.
+ * Return allocated port.
+ */
+#define IOCTL_EVTCHN_BIND_INTERDOMAIN \
+ _IOC(_IOC_NONE, 'E', 1, sizeof(struct ioctl_evtchn_bind_interdomain))
+struct ioctl_evtchn_bind_interdomain {
+ unsigned int remote_domain, remote_port;
+};
+
+/*
+ * Allocate a fresh port for binding to @remote_domain.
+ * Return allocated port.
+ */
+#define IOCTL_EVTCHN_BIND_UNBOUND_PORT \
+ _IOC(_IOC_NONE, 'E', 2, sizeof(struct ioctl_evtchn_bind_unbound_port))
+struct ioctl_evtchn_bind_unbound_port {
+ unsigned int remote_domain;
+};
+
+/*
+ * Unbind previously allocated @port.
+ */
+#define IOCTL_EVTCHN_UNBIND \
+ _IOC(_IOC_NONE, 'E', 3, sizeof(struct ioctl_evtchn_unbind))
+struct ioctl_evtchn_unbind {
+ unsigned int port;
+};
+
+/*
+ * Unbind previously allocated @port.
+ */
+#define IOCTL_EVTCHN_NOTIFY \
+ _IOC(_IOC_NONE, 'E', 4, sizeof(struct ioctl_evtchn_notify))
+struct ioctl_evtchn_notify {
+ unsigned int port;
+};
+
+/* Clear and reinitialise the event buffer. Clear error condition. */
+#define IOCTL_EVTCHN_RESET \
+ _IOC(_IOC_NONE, 'E', 5, 0)
+
+#endif /* __LINUX_PUBLIC_EVTCHN_H__ */
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/include/xen/public/gntdev.h 2008-04-02 12:34:02.000000000 +0200
@@ -0,0 +1,119 @@
+/******************************************************************************
+ * gntdev.h
+ *
+ * Interface to /dev/xen/gntdev.
+ *
+ * Copyright (c) 2007, D G Murray
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License version 2
+ * as published by the Free Software Foundation; or, when distributed
+ * separately from the Linux kernel or incorporated into other
+ * software packages, subject to the following license:
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this source file (the "Software"), to deal in the Software without
+ * restriction, including without limitation the rights to use, copy, modify,
+ * merge, publish, distribute, sublicense, and/or sell copies of the Software,
+ * and to permit persons to whom the Software is furnished to do so, subject to
+ * the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
+ * IN THE SOFTWARE.
+ */
+
+#ifndef __LINUX_PUBLIC_GNTDEV_H__
+#define __LINUX_PUBLIC_GNTDEV_H__
+
+struct ioctl_gntdev_grant_ref {
+ /* The domain ID of the grant to be mapped. */
+ uint32_t domid;
+ /* The grant reference of the grant to be mapped. */
+ uint32_t ref;
+};
+
+/*
+ * Inserts the grant references into the mapping table of an instance
+ * of gntdev. N.B. This does not perform the mapping, which is deferred
+ * until mmap() is called with @index as the offset.
+ */
+#define IOCTL_GNTDEV_MAP_GRANT_REF \
+_IOC(_IOC_NONE, 'G', 0, sizeof(struct ioctl_gntdev_map_grant_ref))
+struct ioctl_gntdev_map_grant_ref {
+ /* IN parameters */
+ /* The number of grants to be mapped. */
+ uint32_t count;
+ uint32_t pad;
+ /* OUT parameters */
+ /* The offset to be used on a subsequent call to mmap(). */
+ uint64_t index;
+ /* Variable IN parameter. */
+ /* Array of grant references, of size @count. */
+ struct ioctl_gntdev_grant_ref refs[1];
+};
+
+/*
+ * Removes the grant references from the mapping table of an instance of
+ * of gntdev. N.B. munmap() must be called on the relevant virtual address(es)
+ * before this ioctl is called, or an error will result.
+ */
+#define IOCTL_GNTDEV_UNMAP_GRANT_REF \
+_IOC(_IOC_NONE, 'G', 1, sizeof(struct ioctl_gntdev_unmap_grant_ref))
+struct ioctl_gntdev_unmap_grant_ref {
+ /* IN parameters */
+ /* The offset was returned by the corresponding map operation. */
+ uint64_t index;
+ /* The number of pages to be unmapped. */
+ uint32_t count;
+ uint32_t pad;
+};
+
+/*
+ * Returns the offset in the driver's address space that corresponds
+ * to @vaddr. This can be used to perform a munmap(), followed by an
+ * UNMAP_GRANT_REF ioctl, where no state about the offset is retained by
+ * the caller. The number of pages that were allocated at the same time as
+ * @vaddr is returned in @count.
+ *
+ * N.B. Where more than one page has been mapped into a contiguous range, the
+ * supplied @vaddr must correspond to the start of the range; otherwise
+ * an error will result. It is only possible to munmap() the entire
+ * contiguously-allocated range at once, and not any subrange thereof.
+ */
+#define IOCTL_GNTDEV_GET_OFFSET_FOR_VADDR \
+_IOC(_IOC_NONE, 'G', 2, sizeof(struct ioctl_gntdev_get_offset_for_vaddr))
+struct ioctl_gntdev_get_offset_for_vaddr {
+ /* IN parameters */
+ /* The virtual address of the first mapped page in a range. */
+ uint64_t vaddr;
+ /* OUT parameters */
+ /* The offset that was used in the initial mmap() operation. */
+ uint64_t offset;
+ /* The number of pages mapped in the VM area that begins at @vaddr. */
+ uint32_t count;
+ uint32_t pad;
+};
+
+/*
+ * Sets the maximum number of grants that may mapped at once by this gntdev
+ * instance.
+ *
+ * N.B. This must be called before any other ioctl is performed on the device.
+ */
+#define IOCTL_GNTDEV_SET_MAX_GRANTS \
+_IOC(_IOC_NONE, 'G', 3, sizeof(struct ioctl_gntdev_set_max_grants))
+struct ioctl_gntdev_set_max_grants {
+ /* IN parameter */
+ /* The maximum number of grants that may be mapped at once. */
+ uint32_t count;
+};
+
+#endif /* __LINUX_PUBLIC_GNTDEV_H__ */
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/include/xen/public/privcmd.h 2010-01-18 15:23:12.000000000 +0100
@@ -0,0 +1,89 @@
+/******************************************************************************
+ * privcmd.h
+ *
+ * Interface to /proc/xen/privcmd.
+ *
+ * Copyright (c) 2003-2005, K A Fraser
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License version 2
+ * as published by the Free Software Foundation; or, when distributed
+ * separately from the Linux kernel or incorporated into other
+ * software packages, subject to the following license:
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this source file (the "Software"), to deal in the Software without
+ * restriction, including without limitation the rights to use, copy, modify,
+ * merge, publish, distribute, sublicense, and/or sell copies of the Software,
+ * and to permit persons to whom the Software is furnished to do so, subject to
+ * the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
+ * IN THE SOFTWARE.
+ */
+
+#ifndef __LINUX_PUBLIC_PRIVCMD_H__
+#define __LINUX_PUBLIC_PRIVCMD_H__
+
+#include <linux/types.h>
+
+#ifndef __user
+#define __user
+#endif
+
+typedef struct privcmd_hypercall
+{
+ __u64 op;
+ __u64 arg[5];
+} privcmd_hypercall_t;
+
+typedef struct privcmd_mmap_entry {
+ __u64 va;
+ __u64 mfn;
+ __u64 npages;
+} privcmd_mmap_entry_t;
+
+typedef struct privcmd_mmap {
+ int num;
+ domid_t dom; /* target domain */
+ privcmd_mmap_entry_t __user *entry;
+} privcmd_mmap_t;
+
+typedef struct privcmd_mmapbatch {
+ int num; /* number of pages to populate */
+ domid_t dom; /* target domain */
+ __u64 addr; /* virtual address */
+ xen_pfn_t __user *arr; /* array of mfns - top nibble set on err */
+} privcmd_mmapbatch_t;
+
+typedef struct privcmd_mmapbatch_v2 {
+ unsigned int num; /* number of pages to populate */
+ domid_t dom; /* target domain */
+ __u64 addr; /* virtual address */
+ const xen_pfn_t __user *arr; /* array of mfns */
+ int __user *err; /* array of error codes */
+} privcmd_mmapbatch_v2_t;
+
+/*
+ * @cmd: IOCTL_PRIVCMD_HYPERCALL
+ * @arg: &privcmd_hypercall_t
+ * Return: Value returned from execution of the specified hypercall.
+ */
+#define IOCTL_PRIVCMD_HYPERCALL \
+ _IOC(_IOC_NONE, 'P', 0, sizeof(privcmd_hypercall_t))
+#define IOCTL_PRIVCMD_MMAP \
+ _IOC(_IOC_NONE, 'P', 2, sizeof(privcmd_mmap_t))
+#define IOCTL_PRIVCMD_MMAPBATCH \
+ _IOC(_IOC_NONE, 'P', 3, sizeof(privcmd_mmapbatch_t))
+#define IOCTL_PRIVCMD_MMAPBATCH_V2 \
+ _IOC(_IOC_NONE, 'P', 4, sizeof(privcmd_mmapbatch_v2_t))
+
+#endif /* __LINUX_PUBLIC_PRIVCMD_H__ */
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/include/xen/public/xenbus.h 2009-05-29 10:25:53.000000000 +0200
@@ -0,0 +1,56 @@
+/******************************************************************************
+ * xenbus.h
+ *
+ * Interface to /proc/xen/xenbus.
+ *
+ * Copyright (c) 2008, Diego Ongaro <diego.ongaro@citrix.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License version 2
+ * as published by the Free Software Foundation; or, when distributed
+ * separately from the Linux kernel or incorporated into other
+ * software packages, subject to the following license:
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this source file (the "Software"), to deal in the Software without
+ * restriction, including without limitation the rights to use, copy, modify,
+ * merge, publish, distribute, sublicense, and/or sell copies of the Software,
+ * and to permit persons to whom the Software is furnished to do so, subject to
+ * the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
+ * IN THE SOFTWARE.
+ */
+
+#ifndef __LINUX_PUBLIC_XENBUS_H__
+#define __LINUX_PUBLIC_XENBUS_H__
+
+#include <linux/types.h>
+
+#ifndef __user
+#define __user
+#endif
+
+typedef struct xenbus_alloc {
+ domid_t dom;
+ __u32 port;
+ __u32 grant_ref;
+} xenbus_alloc_t;
+
+/*
+ * @cmd: IOCTL_XENBUS_ALLOC
+ * @arg: &xenbus_alloc_t
+ * Return: 0, or -1 for error
+ */
+#define IOCTL_XENBUS_ALLOC \
+ _IOC(_IOC_NONE, 'X', 0, sizeof(xenbus_alloc_t))
+
+#endif /* __LINUX_PUBLIC_XENBUS_H__ */
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/include/xen/xen_proc.h 2007-06-12 13:14:19.000000000 +0200
@@ -0,0 +1,12 @@
+
+#ifndef __ASM_XEN_PROC_H__
+#define __ASM_XEN_PROC_H__
+
+#include <linux/proc_fs.h>
+
+extern struct proc_dir_entry *create_xen_proc_entry(
+ const char *name, mode_t mode);
+extern void remove_xen_proc_entry(
+ const char *name);
+
+#endif /* __ASM_XEN_PROC_H__ */
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/include/xen/xencons.h 2007-10-15 09:39:38.000000000 +0200
@@ -0,0 +1,17 @@
+#ifndef __ASM_XENCONS_H__
+#define __ASM_XENCONS_H__
+
+struct dom0_vga_console_info;
+void dom0_init_screen_info(const struct dom0_vga_console_info *, size_t);
+
+void xencons_force_flush(void);
+void xencons_resume(void);
+
+/* Interrupt work hooks. Receive data, or kick data out. */
+void xencons_rx(char *buf, unsigned len, struct pt_regs *regs);
+void xencons_tx(void);
+
+int xencons_ring_init(void);
+int xencons_ring_send(const char *data, unsigned len);
+
+#endif /* __ASM_XENCONS_H__ */
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/include/xen/xenoprof.h 2007-06-12 13:14:19.000000000 +0200
@@ -0,0 +1,42 @@
+/******************************************************************************
+ * xen/xenoprof.h
+ *
+ * Copyright (c) 2006 Isaku Yamahata <yamahata at valinux co jp>
+ * VA Linux Systems Japan K.K.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ *
+ */
+
+#ifndef __XEN_XENOPROF_H__
+#define __XEN_XENOPROF_H__
+#ifdef CONFIG_XEN
+
+#include <asm/xenoprof.h>
+
+struct oprofile_operations;
+int xenoprofile_init(struct oprofile_operations * ops);
+void xenoprofile_exit(void);
+
+struct xenoprof_shared_buffer {
+ char *buffer;
+ struct xenoprof_arch_shared_buffer arch;
+};
+#else
+#define xenoprofile_init(ops) (-ENOSYS)
+#define xenoprofile_exit() do { } while (0)
+
+#endif /* CONFIG_XEN */
+#endif /* __XEN_XENOPROF_H__ */
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/lib/swiotlb-xen.c 2010-09-16 13:31:46.000000000 +0200
@@ -0,0 +1,765 @@
+/*
+ * Dynamic DMA mapping support.
+ *
+ * This implementation is a fallback for platforms that do not support
+ * I/O TLBs (aka DMA address translation hardware).
+ * Copyright (C) 2000 Asit Mallick <Asit.K.Mallick@intel.com>
+ * Copyright (C) 2000 Goutham Rao <goutham.rao@intel.com>
+ * Copyright (C) 2000, 2003 Hewlett-Packard Co
+ * David Mosberger-Tang <davidm@hpl.hp.com>
+ * Copyright (C) 2005 Keir Fraser <keir@xensource.com>
+ */
+
+#include <linux/cache.h>
+#include <linux/mm.h>
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/spinlock.h>
+#include <linux/string.h>
+#include <linux/types.h>
+#include <linux/ctype.h>
+#include <linux/init.h>
+#include <linux/bootmem.h>
+#include <linux/highmem.h>
+#include <asm/io.h>
+#include <asm/pci.h>
+#include <asm/dma.h>
+#include <asm/uaccess.h>
+#include <xen/gnttab.h>
+#include <xen/interface/memory.h>
+#include <asm-i386/mach-xen/asm/gnttab_dma.h>
+
+int swiotlb;
+EXPORT_SYMBOL(swiotlb);
+
+#define OFFSET(val,align) ((unsigned long)((val) & ( (align) - 1)))
+
+/*
+ * Maximum allowable number of contiguous slabs to map,
+ * must be a power of 2. What is the appropriate value ?
+ * The complexity of {map,unmap}_single is linearly dependent on this value.
+ */
+#define IO_TLB_SEGSIZE 128
+
+/*
+ * log of the size of each IO TLB slab. The number of slabs is command line
+ * controllable.
+ */
+#define IO_TLB_SHIFT 11
+
+/*
+ * Enumeration for sync targets
+ */
+enum dma_sync_target {
+ SYNC_FOR_CPU = 0,
+ SYNC_FOR_DEVICE = 1,
+};
+
+int swiotlb_force;
+
+static char *iotlb_virt_start;
+static unsigned long iotlb_nslabs;
+
+/*
+ * Used to do a quick range check in swiotlb_unmap_single and
+ * swiotlb_sync_single_*, to see if the memory was in fact allocated by this
+ * API.
+ */
+static unsigned long iotlb_pfn_start, iotlb_pfn_end;
+
+/* Does the given dma address reside within the swiotlb aperture? */
+static inline int in_swiotlb_aperture(dma_addr_t dev_addr)
+{
+ unsigned long pfn = mfn_to_local_pfn(dev_addr >> PAGE_SHIFT);
+ return (pfn_valid(pfn)
+ && (pfn >= iotlb_pfn_start)
+ && (pfn < iotlb_pfn_end));
+}
+
+/*
+ * When the IOMMU overflows we return a fallback buffer. This sets the size.
+ */
+static unsigned long io_tlb_overflow = 32*1024;
+
+void *io_tlb_overflow_buffer;
+
+/*
+ * This is a free list describing the number of free entries available from
+ * each index
+ */
+static unsigned int *io_tlb_list;
+static unsigned int io_tlb_index;
+
+/*
+ * We need to save away the original address corresponding to a mapped entry
+ * for the sync operations.
+ */
+static struct phys_addr {
+ struct page *page;
+ unsigned int offset;
+} *io_tlb_orig_addr;
+
+/*
+ * Protect the above data structures in the map and unmap calls
+ */
+static DEFINE_SPINLOCK(io_tlb_lock);
+
+static unsigned int dma_bits;
+static unsigned int __initdata max_dma_bits = 32;
+static int __init
+setup_dma_bits(char *str)
+{
+ max_dma_bits = simple_strtoul(str, NULL, 0);
+ return 0;
+}
+__setup("dma_bits=", setup_dma_bits);
+
+static int __init
+setup_io_tlb_npages(char *str)
+{
+ /* Unlike ia64, the size is aperture in megabytes, not 'slabs'! */
+ if (isdigit(*str)) {
+ iotlb_nslabs = simple_strtoul(str, &str, 0) <<
+ (20 - IO_TLB_SHIFT);
+ iotlb_nslabs = ALIGN(iotlb_nslabs, IO_TLB_SEGSIZE);
+ }
+ if (*str == ',')
+ ++str;
+ /*
+ * NB. 'force' enables the swiotlb, but doesn't force its use for
+ * every DMA like it does on native Linux. 'off' forcibly disables
+ * use of the swiotlb.
+ */
+ if (!strcmp(str, "force"))
+ swiotlb_force = 1;
+ else if (!strcmp(str, "off"))
+ swiotlb_force = -1;
+ return 1;
+}
+__setup("swiotlb=", setup_io_tlb_npages);
+/* make io_tlb_overflow tunable too? */
+
+/*
+ * Statically reserve bounce buffer space and initialize bounce buffer data
+ * structures for the software IO TLB used to implement the PCI DMA API.
+ */
+void
+swiotlb_init_with_default_size (size_t default_size)
+{
+ unsigned long i, bytes;
+ int rc;
+
+ if (!iotlb_nslabs) {
+ iotlb_nslabs = (default_size >> IO_TLB_SHIFT);
+ iotlb_nslabs = ALIGN(iotlb_nslabs, IO_TLB_SEGSIZE);
+ }
+
+ bytes = iotlb_nslabs * (1UL << IO_TLB_SHIFT);
+
+ /*
+ * Get IO TLB memory from the low pages
+ */
+ iotlb_virt_start = alloc_bootmem_pages(bytes);
+ if (!iotlb_virt_start)
+ panic("Cannot allocate SWIOTLB buffer!\n");
+
+ dma_bits = get_order(IO_TLB_SEGSIZE << IO_TLB_SHIFT) + PAGE_SHIFT;
+ for (i = 0; i < iotlb_nslabs; i += IO_TLB_SEGSIZE) {
+ do {
+ rc = xen_create_contiguous_region(
+ (unsigned long)iotlb_virt_start + (i << IO_TLB_SHIFT),
+ get_order(IO_TLB_SEGSIZE << IO_TLB_SHIFT),
+ dma_bits);
+ } while (rc && dma_bits++ < max_dma_bits);
+ if (rc) {
+ if (i == 0)
+ panic("No suitable physical memory available for SWIOTLB buffer!\n"
+ "Use dom0_mem Xen boot parameter to reserve\n"
+ "some DMA memory (e.g., dom0_mem=-128M).\n");
+ iotlb_nslabs = i;
+ i <<= IO_TLB_SHIFT;
+ free_bootmem(__pa(iotlb_virt_start + i), bytes - i);
+ bytes = i;
+ for (dma_bits = 0; i > 0; i -= IO_TLB_SEGSIZE << IO_TLB_SHIFT) {
+ unsigned int bits = fls64(virt_to_bus(iotlb_virt_start + i - 1));
+
+ if (bits > dma_bits)
+ dma_bits = bits;
+ }
+ break;
+ }
+ }
+
+ /*
+ * Allocate and initialize the free list array. This array is used
+ * to find contiguous free memory regions of size up to IO_TLB_SEGSIZE.
+ */
+ io_tlb_list = alloc_bootmem(iotlb_nslabs * sizeof(int));
+ for (i = 0; i < iotlb_nslabs; i++)
+ io_tlb_list[i] = IO_TLB_SEGSIZE - OFFSET(i, IO_TLB_SEGSIZE);
+ io_tlb_index = 0;
+ io_tlb_orig_addr = alloc_bootmem(
+ iotlb_nslabs * sizeof(*io_tlb_orig_addr));
+
+ /*
+ * Get the overflow emergency buffer
+ */
+ io_tlb_overflow_buffer = alloc_bootmem(io_tlb_overflow);
+ if (!io_tlb_overflow_buffer)
+ panic("Cannot allocate SWIOTLB overflow buffer!\n");
+
+ do {
+ rc = xen_create_contiguous_region(
+ (unsigned long)io_tlb_overflow_buffer,
+ get_order(io_tlb_overflow),
+ dma_bits);
+ } while (rc && dma_bits++ < max_dma_bits);
+ if (rc)
+ panic("No suitable physical memory available for SWIOTLB overflow buffer!\n");
+
+ iotlb_pfn_start = __pa(iotlb_virt_start) >> PAGE_SHIFT;
+ iotlb_pfn_end = iotlb_pfn_start + (bytes >> PAGE_SHIFT);
+
+ printk(KERN_INFO "Software IO TLB enabled: \n"
+ " Aperture: %lu megabytes\n"
+ " Kernel range: %p - %p\n"
+ " Address size: %u bits\n",
+ bytes >> 20,
+ iotlb_virt_start, iotlb_virt_start + bytes,
+ dma_bits);
+}
+
+void
+swiotlb_init(void)
+{
+ long ram_end;
+ size_t defsz = 64 * (1 << 20); /* 64MB default size */
+
+ if (swiotlb_force == 1) {
+ swiotlb = 1;
+ } else if ((swiotlb_force != -1) &&
+ is_running_on_xen() &&
+ is_initial_xendomain()) {
+ /* Domain 0 always has a swiotlb. */
+ ram_end = HYPERVISOR_memory_op(XENMEM_maximum_ram_page, NULL);
+ if (ram_end <= 0x7ffff)
+ defsz = 2 * (1 << 20); /* 2MB on <2GB on systems. */
+ swiotlb = 1;
+ }
+
+ if (swiotlb)
+ swiotlb_init_with_default_size(defsz);
+ else
+ printk(KERN_INFO "Software IO TLB disabled\n");
+}
+
+/*
+ * We use __copy_to_user_inatomic to transfer to the host buffer because the
+ * buffer may be mapped read-only (e.g, in blkback driver) but lower-level
+ * drivers map the buffer for DMA_BIDIRECTIONAL access. This causes an
+ * unnecessary copy from the aperture to the host buffer, and a page fault.
+ */
+static void
+__sync_single(struct phys_addr buffer, char *dma_addr, size_t size, int dir)
+{
+ if (PageHighMem(buffer.page)) {
+ size_t len, bytes;
+ char *dev, *host, *kmp;
+ len = size;
+ while (len != 0) {
+ unsigned long flags;
+
+ if (((bytes = len) + buffer.offset) > PAGE_SIZE)
+ bytes = PAGE_SIZE - buffer.offset;
+ local_irq_save(flags); /* protects KM_BOUNCE_READ */
+ kmp = kmap_atomic(buffer.page, KM_BOUNCE_READ);
+ dev = dma_addr + size - len;
+ host = kmp + buffer.offset;
+ if (dir == DMA_FROM_DEVICE) {
+ if (__copy_to_user_inatomic(host, dev, bytes))
+ /* inaccessible */;
+ } else
+ memcpy(dev, host, bytes);
+ kunmap_atomic(kmp, KM_BOUNCE_READ);
+ local_irq_restore(flags);
+ len -= bytes;
+ buffer.page++;
+ buffer.offset = 0;
+ }
+ } else {
+ char *host = (char *)phys_to_virt(
+ page_to_pseudophys(buffer.page)) + buffer.offset;
+ if (dir == DMA_FROM_DEVICE) {
+ if (__copy_to_user_inatomic(host, dma_addr, size))
+ /* inaccessible */;
+ } else if (dir == DMA_TO_DEVICE)
+ memcpy(dma_addr, host, size);
+ }
+}
+
+/*
+ * Allocates bounce buffer and returns its kernel virtual address.
+ */
+static void *
+map_single(struct device *hwdev, struct phys_addr buffer, size_t size, int dir)
+{
+ unsigned long flags;
+ char *dma_addr;
+ unsigned int nslots, stride, index, wrap;
+ struct phys_addr slot_buf;
+ int i;
+
+ /*
+ * For mappings greater than a page, we limit the stride (and
+ * hence alignment) to a page size.
+ */
+ nslots = ALIGN(size, 1 << IO_TLB_SHIFT) >> IO_TLB_SHIFT;
+ if (size > PAGE_SIZE)
+ stride = (1 << (PAGE_SHIFT - IO_TLB_SHIFT));
+ else
+ stride = 1;
+
+ BUG_ON(!nslots);
+
+ /*
+ * Find suitable number of IO TLB entries size that will fit this
+ * request and allocate a buffer from that IO TLB pool.
+ */
+ spin_lock_irqsave(&io_tlb_lock, flags);
+ {
+ wrap = index = ALIGN(io_tlb_index, stride);
+
+ if (index >= iotlb_nslabs)
+ wrap = index = 0;
+
+ do {
+ /*
+ * If we find a slot that indicates we have 'nslots'
+ * number of contiguous buffers, we allocate the
+ * buffers from that slot and mark the entries as '0'
+ * indicating unavailable.
+ */
+ if (io_tlb_list[index] >= nslots) {
+ int count = 0;
+
+ for (i = index; i < (int)(index + nslots); i++)
+ io_tlb_list[i] = 0;
+ for (i = index - 1;
+ (OFFSET(i, IO_TLB_SEGSIZE) !=
+ IO_TLB_SEGSIZE -1) && io_tlb_list[i];
+ i--)
+ io_tlb_list[i] = ++count;
+ dma_addr = iotlb_virt_start +
+ (index << IO_TLB_SHIFT);
+
+ /*
+ * Update the indices to avoid searching in
+ * the next round.
+ */
+ io_tlb_index =
+ ((index + nslots) < iotlb_nslabs
+ ? (index + nslots) : 0);
+
+ goto found;
+ }
+ index += stride;
+ if (index >= iotlb_nslabs)
+ index = 0;
+ } while (index != wrap);
+
+ spin_unlock_irqrestore(&io_tlb_lock, flags);
+ return NULL;
+ }
+ found:
+ spin_unlock_irqrestore(&io_tlb_lock, flags);
+
+ /*
+ * Save away the mapping from the original address to the DMA address.
+ * This is needed when we sync the memory. Then we sync the buffer if
+ * needed.
+ */
+ slot_buf = buffer;
+ for (i = 0; i < nslots; i++) {
+ slot_buf.page += slot_buf.offset >> PAGE_SHIFT;
+ slot_buf.offset &= PAGE_SIZE - 1;
+ io_tlb_orig_addr[index+i] = slot_buf;
+ slot_buf.offset += 1 << IO_TLB_SHIFT;
+ }
+ if ((dir == DMA_TO_DEVICE) || (dir == DMA_BIDIRECTIONAL))
+ __sync_single(buffer, dma_addr, size, DMA_TO_DEVICE);
+
+ return dma_addr;
+}
+
+static struct phys_addr dma_addr_to_phys_addr(char *dma_addr)
+{
+ int index = (dma_addr - iotlb_virt_start) >> IO_TLB_SHIFT;
+ struct phys_addr buffer = io_tlb_orig_addr[index];
+ buffer.offset += (long)dma_addr & ((1 << IO_TLB_SHIFT) - 1);
+ buffer.page += buffer.offset >> PAGE_SHIFT;
+ buffer.offset &= PAGE_SIZE - 1;
+ return buffer;
+}
+
+/*
+ * dma_addr is the kernel virtual address of the bounce buffer to unmap.
+ */
+static void
+unmap_single(struct device *hwdev, char *dma_addr, size_t size, int dir)
+{
+ unsigned long flags;
+ int i, count, nslots = ALIGN(size, 1 << IO_TLB_SHIFT) >> IO_TLB_SHIFT;
+ int index = (dma_addr - iotlb_virt_start) >> IO_TLB_SHIFT;
+ struct phys_addr buffer = dma_addr_to_phys_addr(dma_addr);
+
+ /*
+ * First, sync the memory before unmapping the entry
+ */
+ if ((dir == DMA_FROM_DEVICE) || (dir == DMA_BIDIRECTIONAL))
+ __sync_single(buffer, dma_addr, size, DMA_FROM_DEVICE);
+
+ /*
+ * Return the buffer to the free list by setting the corresponding
+ * entries to indicate the number of contigous entries available.
+ * While returning the entries to the free list, we merge the entries
+ * with slots below and above the pool being returned.
+ */
+ spin_lock_irqsave(&io_tlb_lock, flags);
+ {
+ count = ((index + nslots) < ALIGN(index + 1, IO_TLB_SEGSIZE) ?
+ io_tlb_list[index + nslots] : 0);
+ /*
+ * Step 1: return the slots to the free list, merging the
+ * slots with superceeding slots
+ */
+ for (i = index + nslots - 1; i >= index; i--)
+ io_tlb_list[i] = ++count;
+ /*
+ * Step 2: merge the returned slots with the preceding slots,
+ * if available (non zero)
+ */
+ for (i = index - 1;
+ (OFFSET(i, IO_TLB_SEGSIZE) !=
+ IO_TLB_SEGSIZE -1) && io_tlb_list[i];
+ i--)
+ io_tlb_list[i] = ++count;
+ }
+ spin_unlock_irqrestore(&io_tlb_lock, flags);
+}
+
+static void
+sync_single(struct device *hwdev, char *dma_addr, size_t size,
+ int dir, int target)
+{
+ struct phys_addr buffer = dma_addr_to_phys_addr(dma_addr);
+ switch (target) {
+ case SYNC_FOR_CPU:
+ if (likely(dir == DMA_FROM_DEVICE || dir == DMA_BIDIRECTIONAL))
+ __sync_single(buffer, dma_addr, size, DMA_FROM_DEVICE);
+ else
+ BUG_ON(dir != DMA_TO_DEVICE);
+ break;
+ case SYNC_FOR_DEVICE:
+ if (likely(dir == DMA_TO_DEVICE || dir == DMA_BIDIRECTIONAL))
+ __sync_single(buffer, dma_addr, size, DMA_TO_DEVICE);
+ else
+ BUG_ON(dir != DMA_FROM_DEVICE);
+ break;
+ default:
+ BUG();
+ }
+}
+
+static void
+swiotlb_full(struct device *dev, size_t size, int dir, int do_panic)
+{
+ /*
+ * Ran out of IOMMU space for this operation. This is very bad.
+ * Unfortunately the drivers cannot handle this operation properly.
+ * unless they check for pci_dma_mapping_error (most don't)
+ * When the mapping is small enough return a static buffer to limit
+ * the damage, or panic when the transfer is too big.
+ */
+ printk(KERN_ERR "PCI-DMA: Out of SW-IOMMU space for %lu bytes at "
+ "device %s\n", (unsigned long)size, dev ? dev->bus_id : "?");
+
+ if (size > io_tlb_overflow && do_panic) {
+ if (dir == PCI_DMA_FROMDEVICE || dir == PCI_DMA_BIDIRECTIONAL)
+ panic("PCI-DMA: Memory would be corrupted\n");
+ if (dir == PCI_DMA_TODEVICE || dir == PCI_DMA_BIDIRECTIONAL)
+ panic("PCI-DMA: Random memory would be DMAed\n");
+ }
+}
+
+/*
+ * Map a single buffer of the indicated size for DMA in streaming mode. The
+ * PCI address to use is returned.
+ *
+ * Once the device is given the dma address, the device owns this memory until
+ * either swiotlb_unmap_single or swiotlb_dma_sync_single is performed.
+ */
+dma_addr_t
+swiotlb_map_single(struct device *hwdev, void *ptr, size_t size, int dir)
+{
+ dma_addr_t dev_addr = gnttab_dma_map_page(virt_to_page(ptr)) +
+ offset_in_page(ptr);
+ void *map;
+ struct phys_addr buffer;
+
+ BUG_ON(dir == DMA_NONE);
+
+ /*
+ * If the pointer passed in happens to be in the device's DMA window,
+ * we can safely return the device addr and not worry about bounce
+ * buffering it.
+ */
+ if (!range_straddles_page_boundary(__pa(ptr), size) &&
+ !address_needs_mapping(hwdev, dev_addr))
+ return dev_addr;
+
+ /*
+ * Oh well, have to allocate and map a bounce buffer.
+ */
+ gnttab_dma_unmap_page(dev_addr);
+ buffer.page = virt_to_page(ptr);
+ buffer.offset = (unsigned long)ptr & ~PAGE_MASK;
+ map = map_single(hwdev, buffer, size, dir);
+ if (!map) {
+ swiotlb_full(hwdev, size, dir, 1);
+ map = io_tlb_overflow_buffer;
+ }
+
+ dev_addr = virt_to_bus(map);
+ return dev_addr;
+}
+
+/*
+ * Unmap a single streaming mode DMA translation. The dma_addr and size must
+ * match what was provided for in a previous swiotlb_map_single call. All
+ * other usages are undefined.
+ *
+ * After this call, reads by the cpu to the buffer are guaranteed to see
+ * whatever the device wrote there.
+ */
+void
+swiotlb_unmap_single(struct device *hwdev, dma_addr_t dev_addr, size_t size,
+ int dir)
+{
+ BUG_ON(dir == DMA_NONE);
+ if (in_swiotlb_aperture(dev_addr))
+ unmap_single(hwdev, bus_to_virt(dev_addr), size, dir);
+ else
+ gnttab_dma_unmap_page(dev_addr);
+}
+
+/*
+ * Make physical memory consistent for a single streaming mode DMA translation
+ * after a transfer.
+ *
+ * If you perform a swiotlb_map_single() but wish to interrogate the buffer
+ * using the cpu, yet do not wish to teardown the PCI dma mapping, you must
+ * call this function before doing so. At the next point you give the PCI dma
+ * address back to the card, you must first perform a
+ * swiotlb_dma_sync_for_device, and then the device again owns the buffer
+ */
+static inline void
+swiotlb_sync_single(struct device *hwdev, dma_addr_t dev_addr,
+ size_t size, int dir, int target)
+{
+ BUG_ON(dir == DMA_NONE);
+ if (in_swiotlb_aperture(dev_addr))
+ sync_single(hwdev, bus_to_virt(dev_addr), size, dir, target);
+}
+
+void
+swiotlb_sync_single_for_cpu(struct device *hwdev, dma_addr_t dev_addr,
+ size_t size, int dir)
+{
+ swiotlb_sync_single(hwdev, dev_addr, size, dir, SYNC_FOR_CPU);
+}
+
+void
+swiotlb_sync_single_for_device(struct device *hwdev, dma_addr_t dev_addr,
+ size_t size, int dir)
+{
+ swiotlb_sync_single(hwdev, dev_addr, size, dir, SYNC_FOR_DEVICE);
+}
+
+/*
+ * Map a set of buffers described by scatterlist in streaming mode for DMA.
+ * This is the scatter-gather version of the above swiotlb_map_single
+ * interface. Here the scatter gather list elements are each tagged with the
+ * appropriate dma address and length. They are obtained via
+ * sg_dma_{address,length}(SG).
+ *
+ * NOTE: An implementation may be able to use a smaller number of
+ * DMA address/length pairs than there are SG table elements.
+ * (for example via virtual mapping capabilities)
+ * The routine returns the number of addr/length pairs actually
+ * used, at most nents.
+ *
+ * Device ownership issues as mentioned above for swiotlb_map_single are the
+ * same here.
+ */
+int
+swiotlb_map_sg(struct device *hwdev, struct scatterlist *sg, int nelems,
+ int dir)
+{
+ struct phys_addr buffer;
+ dma_addr_t dev_addr;
+ char *map;
+ int i;
+
+ BUG_ON(dir == DMA_NONE);
+
+ for (i = 0; i < nelems; i++, sg++) {
+ dev_addr = gnttab_dma_map_page(sg->page) + sg->offset;
+
+ if (range_straddles_page_boundary(page_to_pseudophys(sg->page)
+ + sg->offset, sg->length)
+ || address_needs_mapping(hwdev, dev_addr)) {
+ gnttab_dma_unmap_page(dev_addr);
+ buffer.page = sg->page;
+ buffer.offset = sg->offset;
+ map = map_single(hwdev, buffer, sg->length, dir);
+ if (!map) {
+ /* Don't panic here, we expect map_sg users
+ to do proper error handling. */
+ swiotlb_full(hwdev, sg->length, dir, 0);
+ swiotlb_unmap_sg(hwdev, sg - i, i, dir);
+ sg[0].dma_length = 0;
+ return 0;
+ }
+ sg->dma_address = (dma_addr_t)virt_to_bus(map);
+ } else
+ sg->dma_address = dev_addr;
+ sg->dma_length = sg->length;
+ }
+ return nelems;
+}
+
+/*
+ * Unmap a set of streaming mode DMA translations. Again, cpu read rules
+ * concerning calls here are the same as for swiotlb_unmap_single() above.
+ */
+void
+swiotlb_unmap_sg(struct device *hwdev, struct scatterlist *sg, int nelems,
+ int dir)
+{
+ int i;
+
+ BUG_ON(dir == DMA_NONE);
+
+ for (i = 0; i < nelems; i++, sg++)
+ if (in_swiotlb_aperture(sg->dma_address))
+ unmap_single(hwdev,
+ (void *)bus_to_virt(sg->dma_address),
+ sg->dma_length, dir);
+ else
+ gnttab_dma_unmap_page(sg->dma_address);
+}
+
+/*
+ * Make physical memory consistent for a set of streaming mode DMA translations
+ * after a transfer.
+ *
+ * The same as swiotlb_sync_single_* but for a scatter-gather list, same rules
+ * and usage.
+ */
+static inline void
+swiotlb_sync_sg(struct device *hwdev, struct scatterlist *sg,
+ int nelems, int dir, int target)
+{
+ int i;
+
+ BUG_ON(dir == DMA_NONE);
+
+ for (i = 0; i < nelems; i++, sg++)
+ if (in_swiotlb_aperture(sg->dma_address))
+ sync_single(hwdev, bus_to_virt(sg->dma_address),
+ sg->dma_length, dir, target);
+}
+
+void
+swiotlb_sync_sg_for_cpu(struct device *hwdev, struct scatterlist *sg,
+ int nelems, int dir)
+{
+ swiotlb_sync_sg(hwdev, sg, nelems, dir, SYNC_FOR_CPU);
+}
+
+void
+swiotlb_sync_sg_for_device(struct device *hwdev, struct scatterlist *sg,
+ int nelems, int dir)
+{
+ swiotlb_sync_sg(hwdev, sg, nelems, dir, SYNC_FOR_DEVICE);
+}
+
+#ifdef CONFIG_HIGHMEM
+
+dma_addr_t
+swiotlb_map_page(struct device *hwdev, struct page *page,
+ unsigned long offset, size_t size,
+ enum dma_data_direction direction)
+{
+ struct phys_addr buffer;
+ dma_addr_t dev_addr;
+ char *map;
+
+ dev_addr = gnttab_dma_map_page(page) + offset;
+ if (address_needs_mapping(hwdev, dev_addr)) {
+ gnttab_dma_unmap_page(dev_addr);
+ buffer.page = page;
+ buffer.offset = offset;
+ map = map_single(hwdev, buffer, size, direction);
+ if (!map) {
+ swiotlb_full(hwdev, size, direction, 1);
+ map = io_tlb_overflow_buffer;
+ }
+ dev_addr = (dma_addr_t)virt_to_bus(map);
+ }
+
+ return dev_addr;
+}
+
+void
+swiotlb_unmap_page(struct device *hwdev, dma_addr_t dma_address,
+ size_t size, enum dma_data_direction direction)
+{
+ BUG_ON(direction == DMA_NONE);
+ if (in_swiotlb_aperture(dma_address))
+ unmap_single(hwdev, bus_to_virt(dma_address), size, direction);
+ else
+ gnttab_dma_unmap_page(dma_address);
+}
+
+#endif
+
+int
+swiotlb_dma_mapping_error(dma_addr_t dma_addr)
+{
+ return (dma_addr == virt_to_bus(io_tlb_overflow_buffer));
+}
+
+/*
+ * Return whether the given PCI device DMA address mask can be supported
+ * properly. For example, if your device can only drive the low 24-bits
+ * during PCI bus mastering, then you would pass 0x00ffffff as the mask to
+ * this function.
+ */
+int
+swiotlb_dma_supported (struct device *hwdev, u64 mask)
+{
+ return (mask >= ((1UL << dma_bits) - 1));
+}
+
+EXPORT_SYMBOL(swiotlb_init);
+EXPORT_SYMBOL(swiotlb_map_single);
+EXPORT_SYMBOL(swiotlb_unmap_single);
+EXPORT_SYMBOL(swiotlb_map_sg);
+EXPORT_SYMBOL(swiotlb_unmap_sg);
+EXPORT_SYMBOL(swiotlb_sync_single_for_cpu);
+EXPORT_SYMBOL(swiotlb_sync_single_for_device);
+EXPORT_SYMBOL(swiotlb_sync_sg_for_cpu);
+EXPORT_SYMBOL(swiotlb_sync_sg_for_device);
+EXPORT_SYMBOL(swiotlb_dma_mapping_error);
+EXPORT_SYMBOL(swiotlb_dma_supported);
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/mm/tmem-xen.c 2009-06-23 09:28:21.000000000 +0200
@@ -0,0 +1,41 @@
+/*
+ * Xen implementation for transcendent memory (tmem)
+ *
+ * Dan Magenheimer <dan.magenheimer@oracle.com> 2009
+ */
+
+#include <linux/types.h>
+#include <xen/interface/xen.h>
+#include <asm/hypervisor.h>
+
+int xen_tmem_op(u32 tmem_cmd, u32 tmem_pool, u64 object, u32 index,
+ unsigned long gmfn, u32 tmem_offset, u32 pfn_offset, u32 len)
+{
+ struct tmem_op op;
+ int rc = 0;
+
+ op.cmd = tmem_cmd;
+ op.pool_id = tmem_pool;
+ op.u.gen.object = object;
+ op.u.gen.index = index;
+ op.u.gen.tmem_offset = tmem_offset;
+ op.u.gen.pfn_offset = pfn_offset;
+ op.u.gen.len = len;
+ op.u.gen.cmfn = gmfn;
+ rc = HYPERVISOR_tmem_op(&op);
+ return rc;
+}
+
+int xen_tmem_new_pool(uint32_t tmem_cmd, uint64_t uuid_lo,
+ uint64_t uuid_hi, uint32_t flags)
+{
+ struct tmem_op op;
+ int rc = 0;
+
+ op.cmd = tmem_cmd;
+ op.u.new.uuid[0] = uuid_lo;
+ op.u.new.uuid[1] = uuid_hi;
+ op.u.new.flags = flags;
+ rc = HYPERVISOR_tmem_op(&op);
+ return rc;
+}
--- /dev/null 1970-01-01 00:00:00.000000000 +0000
+++ b/scripts/Makefile.xen.awk 2007-08-06 15:10:49.000000000 +0200
@@ -0,0 +1,34 @@
+BEGIN {
+ is_rule = 0
+}
+
+/^[[:space:]]*#/ {
+ next
+}
+
+/^[[:space:]]*$/ {
+ if (is_rule)
+ print("")
+ is_rule = 0
+ next
+}
+
+/:[[:space:]]*%\.[cS][[:space:]]/ {
+ line = gensub(/%.([cS])/, "%-xen.\\1", "g", $0)
+ line = gensub(/(single-used-m)/, "xen-\\1", "g", line)
+ print line
+ is_rule = 1
+ next
+}
+
+/^[^\t]$/ {
+ if (is_rule)
+ print("")
+ is_rule = 0
+ next
+}
+
+is_rule {
+ print $0
+ next
+}
Reference in new issue View Git Blame Copy Permalink