From: Linux Kernel Mailing List Subject: [PATCH] Linux: Update to 2.6.27 Patch-mainline: 2.6.27 This patch contains the differences between Linux 2.6.26 and 2.6.27. Acked-by: Jeff Mahoney Automatically created from "patches.kernel.org/patch-2.6.27" by xen-port-patches.py Removed adjustments NO_HZ -> NO_HZ || NO_IDLE_HZ from kernel/{hr,}timer.c, as they would get removed again by xen-clockevents (and really shouldn't have been needed - see SLE11 SPn). --- head-2011-03-11.orig/arch/x86/Kconfig 2011-01-31 18:07:35.000000000 +0100 +++ head-2011-03-11/arch/x86/Kconfig 2011-02-01 14:38:38.000000000 +0100 @@ -723,7 +723,7 @@ config AMD_IOMMU bool "AMD IOMMU support" select SWIOTLB select PCI_MSI - depends on X86_64 && PCI && ACPI + depends on X86_64 && PCI && ACPI && !XEN ---help--- With this option you can enable support for AMD IOMMU hardware in your system. An IOMMU is a hardware component which provides @@ -1465,7 +1465,7 @@ config MTRR config MTRR_SANITIZER def_bool y prompt "MTRR cleanup support" - depends on MTRR + depends on MTRR && !XEN ---help--- Convert MTRR layout from continuous to discrete, so X drivers can add writeback entries. --- head-2011-03-11.orig/arch/x86/Kconfig.debug 2011-01-31 18:01:51.000000000 +0100 +++ head-2011-03-11/arch/x86/Kconfig.debug 2011-02-01 14:38:38.000000000 +0100 @@ -25,6 +25,7 @@ config STRICT_DEVMEM config X86_VERBOSE_BOOTUP bool "Enable verbose x86 bootup info messages" default y + depends on !XEN ---help--- Enables the informational output from the decompression stage (e.g. bzImage) of the boot. If you disable this you will still @@ -179,6 +180,7 @@ config IOMMU_LEAK config HAVE_MMIOTRACE_SUPPORT def_bool y + depends on !XEN config X86_DECODER_SELFTEST bool "x86 instruction decoder selftest" --- head-2011-03-11.orig/arch/x86/Makefile 2011-02-01 14:11:04.000000000 +0100 +++ head-2011-03-11/arch/x86/Makefile 2011-02-01 14:38:38.000000000 +0100 @@ -117,8 +117,8 @@ endif KBUILD_CFLAGS += $(call cc-option,-mno-sse -mno-mmx -mno-sse2 -mno-3dnow,) # Xen subarch support -mflags-$(CONFIG_X86_XEN) := -Iinclude/asm-x86/mach-xen -mcore-$(CONFIG_X86_XEN) := arch/x86/mach-xen/ +mflags-$(CONFIG_XEN) := -Iinclude/asm-x86/mach-xen +mcore-$(CONFIG_XEN) := arch/x86/mach-xen/ KBUILD_CFLAGS += $(mflags-y) KBUILD_AFLAGS += $(mflags-y) --- head-2011-03-11.orig/arch/x86/ia32/ia32entry-xen.S 2011-01-31 18:07:35.000000000 +0100 +++ head-2011-03-11/arch/x86/ia32/ia32entry-xen.S 2011-02-01 14:38:38.000000000 +0100 @@ -15,6 +15,16 @@ #include #include +/* Avoid __ASSEMBLER__'ifying just for this. */ +#include +#define AUDIT_ARCH_I386 (EM_386|__AUDIT_ARCH_LE) +#define __AUDIT_ARCH_LE 0x40000000 + +#ifndef CONFIG_AUDITSYSCALL +#define sysexit_audit int_ret_from_sys_call +#define sysretl_audit int_ret_from_sys_call +#endif + #define IA32_NR_syscalls ((ia32_syscall_end - ia32_sys_call_table)/8) .macro IA32_ARG_FIXUP noebp=0 @@ -37,6 +47,11 @@ movq %rax,R8(%rsp) .endm + /* + * Reload arg registers from stack in case ptrace changed them. + * We don't reload %eax because syscall_trace_enter() returned + * the value it wants us to use in the table lookup. + */ .macro LOAD_ARGS32 offset movl \offset(%rsp),%r11d movl \offset+8(%rsp),%r10d @@ -46,7 +61,6 @@ 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 @@ -61,6 +75,19 @@ CFI_UNDEFINED r15 .endm +#ifdef CONFIG_PARAVIRT +ENTRY(native_usergs_sysret32) + swapgs + sysretl +ENDPROC(native_usergs_sysret32) + +ENTRY(native_irq_enable_sysexit) + swapgs + sti + sysexit +ENDPROC(native_irq_enable_sysexit) +#endif + /* * 32bit SYSENTER instruction entry. * @@ -98,7 +125,7 @@ ENTRY(ia32_sysenter_target) CFI_RESTORE rcx movl %ebp,%ebp /* zero extension */ movl %eax,%eax - movl 48-THREAD_SIZE+threadinfo_sysenter_return(%rsp),%r10d + movl 48-THREAD_SIZE+TI_sysenter_return(%rsp),%r10d movl $__USER32_DS,40(%rsp) movq %rbp,32(%rsp) movl $__USER32_CS,16(%rsp) @@ -113,19 +140,75 @@ ENTRY(ia32_sysenter_target) .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) + orl $TS_COMPAT,TI_status(%r10) + testl $_TIF_WORK_SYSCALL_ENTRY,TI_flags(%r10) jnz sysenter_tracesys -sysenter_do_call: cmpl $(IA32_NR_syscalls-1),%eax ja ia32_badsys +sysenter_do_call: IA32_ARG_FIXUP 1 +sysenter_dispatch: call *ia32_sys_call_table(,%rax,8) movq %rax,RAX-ARGOFFSET(%rsp) + GET_THREAD_INFO(%r10) + DISABLE_INTERRUPTS(CLBR_NONE) + TRACE_IRQS_OFF + testl $_TIF_ALLWORK_MASK,TI_flags(%r10) + jnz sysexit_audit jmp int_ret_from_sys_call +#ifdef CONFIG_AUDITSYSCALL + .macro auditsys_entry_common + movl %esi,%r9d /* 6th arg: 4th syscall arg */ + movl %edx,%r8d /* 5th arg: 3rd syscall arg */ + /* (already in %ecx) 4th arg: 2nd syscall arg */ + movl %ebx,%edx /* 3rd arg: 1st syscall arg */ + movl %eax,%esi /* 2nd arg: syscall number */ + movl $AUDIT_ARCH_I386,%edi /* 1st arg: audit arch */ + call audit_syscall_entry + movl RAX-ARGOFFSET(%rsp),%eax /* reload syscall number */ + cmpl $(IA32_NR_syscalls-1),%eax + ja ia32_badsys + movl %ebx,%edi /* reload 1st syscall arg */ + movl RCX-ARGOFFSET(%rsp),%esi /* reload 2nd syscall arg */ + movl RDX-ARGOFFSET(%rsp),%edx /* reload 3rd syscall arg */ + movl RSI-ARGOFFSET(%rsp),%ecx /* reload 4th syscall arg */ + movl RDI-ARGOFFSET(%rsp),%r8d /* reload 5th syscall arg */ + .endm + + .macro auditsys_exit exit,ebpsave=RBP + testl $(_TIF_ALLWORK_MASK & ~_TIF_SYSCALL_AUDIT),TI_flags(%r10) + jnz int_ret_from_sys_call + TRACE_IRQS_ON + ENABLE_INTERRUPTS(CLBR_NONE) + movl %eax,%esi /* second arg, syscall return value */ + cmpl $0,%eax /* is it < 0? */ + setl %al /* 1 if so, 0 if not */ + movzbl %al,%edi /* zero-extend that into %edi */ + inc %edi /* first arg, 0->1(AUDITSC_SUCCESS), 1->2(AUDITSC_FAILURE) */ + call audit_syscall_exit + movl \ebpsave-ARGOFFSET(%rsp),%ebp /* reload user register value */ + movl $(_TIF_ALLWORK_MASK & ~_TIF_SYSCALL_AUDIT),%edi + DISABLE_INTERRUPTS(CLBR_NONE) + TRACE_IRQS_OFF + jmp int_with_check + .endm + +sysenter_auditsys: + auditsys_entry_common + movl %ebp,%r9d /* reload 6th syscall arg */ + jmp sysenter_dispatch + +sysexit_audit: + auditsys_exit sysexit_from_sys_call +#endif + sysenter_tracesys: xchgl %r9d,%ebp +#ifdef CONFIG_AUDITSYSCALL + testl $(_TIF_WORK_SYSCALL_ENTRY & ~_TIF_SYSCALL_AUDIT),TI_flags(%r10) + jz sysenter_auditsys +#endif SAVE_REST CLEAR_RREGS movq %r9,R9(%rsp) @@ -186,18 +269,38 @@ ENTRY(ia32_cstar_target) .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) + orl $TS_COMPAT,TI_status(%r10) + testl $_TIF_WORK_SYSCALL_ENTRY,TI_flags(%r10) jnz cstar_tracesys cstar_do_call: cmpl $IA32_NR_syscalls-1,%eax ja ia32_badsys IA32_ARG_FIXUP 1 +cstar_dispatch: call *ia32_sys_call_table(,%rax,8) movq %rax,RAX-ARGOFFSET(%rsp) + GET_THREAD_INFO(%r10) + DISABLE_INTERRUPTS(CLBR_NONE) + testl $_TIF_ALLWORK_MASK,TI_flags(%r10) + jnz sysretl_audit jmp int_ret_from_sys_call -cstar_tracesys: +#ifdef CONFIG_AUDITSYSCALL +cstar_auditsys: + movl %r9d,R9-ARGOFFSET(%rsp) /* register to be clobbered by call */ + auditsys_entry_common + movl R9-ARGOFFSET(%rsp),%r9d /* reload 6th syscall arg */ + jmp cstar_dispatch + +sysretl_audit: + auditsys_exit sysretl_from_sys_call, RCX /* user %ebp in RCX slot */ +#endif + +cstar_tracesys: +#ifdef CONFIG_AUDITSYSCALL + testl $(_TIF_WORK_SYSCALL_ENTRY & ~_TIF_SYSCALL_AUDIT),TI_flags(%r10) + jz cstar_auditsys +#endif xchgl %r9d,%ebp SAVE_REST CLEAR_RREGS @@ -263,8 +366,8 @@ ENTRY(ia32_syscall) 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) + orl $TS_COMPAT,TI_status(%r10) + testl $_TIF_WORK_SYSCALL_ENTRY,TI_flags(%r10) jnz ia32_tracesys ia32_do_syscall: cmpl $(IA32_NR_syscalls-1),%eax @@ -309,13 +412,11 @@ quiet_ni_syscall: PTREGSCALL stub32_rt_sigreturn, sys32_rt_sigreturn, %rdi PTREGSCALL stub32_sigreturn, sys32_sigreturn, %rdi PTREGSCALL stub32_sigaltstack, sys32_sigaltstack, %rdx - PTREGSCALL stub32_sigsuspend, sys32_sigsuspend, %rcx PTREGSCALL stub32_execve, sys32_execve, %rcx PTREGSCALL stub32_fork, sys_fork, %rdi PTREGSCALL stub32_clone, sys32_clone, %rdx PTREGSCALL stub32_vfork, sys_vfork, %rdi PTREGSCALL stub32_iopl, sys_iopl, %rsi - PTREGSCALL stub32_rt_sigsuspend, sys_rt_sigsuspend, %rdx ENTRY(ia32_ptregs_common) popq %r11 @@ -415,7 +516,7 @@ ia32_sys_call_table: .quad sys_ssetmask .quad sys_setreuid16 /* 70 */ .quad sys_setregid16 - .quad stub32_sigsuspend + .quad sys32_sigsuspend .quad compat_sys_sigpending .quad sys_sethostname .quad compat_sys_setrlimit /* 75 */ @@ -522,7 +623,7 @@ ia32_sys_call_table: .quad sys32_rt_sigpending .quad compat_sys_rt_sigtimedwait .quad sys32_rt_sigqueueinfo - .quad stub32_rt_sigsuspend + .quad sys_rt_sigsuspend .quad sys32_pread /* 180 */ .quad sys32_pwrite .quad sys_chown16 @@ -670,4 +771,10 @@ ia32_sys_call_table: .quad sys32_fallocate .quad compat_sys_timerfd_settime /* 325 */ .quad compat_sys_timerfd_gettime + .quad compat_sys_signalfd4 + .quad sys_eventfd2 + .quad sys_epoll_create1 + .quad sys_dup3 /* 330 */ + .quad sys_pipe2 + .quad sys_inotify_init1 ia32_syscall_end: --- head-2011-03-11.orig/arch/x86/kernel/Makefile 2011-01-31 18:07:35.000000000 +0100 +++ head-2011-03-11/arch/x86/kernel/Makefile 2011-02-01 14:38:38.000000000 +0100 @@ -125,9 +125,11 @@ ifeq ($(CONFIG_X86_64),y) obj-$(CONFIG_PCI_MMCONFIG) += mmconf-fam10h_64.o obj-y += vsmp_64.o - obj-$(CONFIG_XEN) += nmi_64.o + obj-$(CONFIG_XEN) += nmi.o time_64-$(CONFIG_XEN) += time_32.o endif -disabled-obj-$(CONFIG_XEN) := crash.o early-quirks.o hpet.o i8253.o i8259_$(BITS).o \ - pci-swiotlb_64.o reboot.o smpboot.o tlb_$(BITS).o tsc_$(BITS).o tsc_sync.o vsmp_64.o +disabled-obj-$(CONFIG_XEN) := %_uv.o crash.o early-quirks.o hpet.o i8253.o \ + i8259.o irqinit_$(BITS).o pci-swiotlb_64.o reboot.o smpboot.o \ + tlb_$(BITS).o tsc.o tsc_sync.o vsmp_64.o +disabled-obj-$(CONFIG_XEN_UNPRIVILEGED_GUEST) += probe_roms_32.o --- head-2011-03-11.orig/arch/x86/kernel/acpi/boot.c 2011-03-11 10:59:02.000000000 +0100 +++ head-2011-03-11/arch/x86/kernel/acpi/boot.c 2011-03-11 10:59:30.000000000 +0100 @@ -1349,6 +1349,7 @@ static int __init dmi_disable_acpi(const return 0; } +#ifndef CONFIG_XEN /* * Force ignoring BIOS IRQ0 pin2 override */ @@ -1366,6 +1367,7 @@ static int __init dmi_ignore_irq0_timer_ } return 0; } +#endif static int __init force_acpi_rsdt(const struct dmi_system_id *d) { @@ -1486,6 +1488,7 @@ static struct dmi_system_id __initdata a {} }; +#ifndef CONFIG_XEN /* second table for DMI checks that should run after early-quirks */ static struct dmi_system_id __initdata acpi_dmi_table_late[] = { /* @@ -1532,6 +1535,7 @@ static struct dmi_system_id __initdata a }, {} }; +#endif /* * acpi_boot_table_init() and acpi_boot_init() @@ -1604,8 +1608,10 @@ int __init early_acpi_boot_init(void) int __init acpi_boot_init(void) { +#ifndef CONFIG_XEN /* those are executed after early-quirks are executed */ dmi_check_system(acpi_dmi_table_late); +#endif /* * If acpi_disabled, bail out --- head-2011-03-11.orig/arch/x86/kernel/acpi/sleep-xen.c 2011-01-31 18:07:35.000000000 +0100 +++ head-2011-03-11/arch/x86/kernel/acpi/sleep-xen.c 2011-02-01 14:38:38.000000000 +0100 @@ -9,6 +9,7 @@ #include #include #include +#include #include "realmode/wakeup.h" #include "sleep.h" @@ -20,7 +21,7 @@ unsigned long acpi_realmode_flags; /* address in low memory of the wakeup routine. */ static unsigned long acpi_realmode; -#ifdef CONFIG_64BIT +#if defined(CONFIG_SMP) && defined(CONFIG_64BIT) static char temp_stack[10240]; #endif #endif @@ -54,18 +55,27 @@ int acpi_save_state_mem(void) header->video_mode = saved_video_mode; header->wakeup_jmp_seg = acpi_wakeup_address >> 4; + + /* + * Set up the wakeup GDT. We set these up as Big Real Mode, + * that is, with limits set to 4 GB. At least the Lenovo + * Thinkpad X61 is known to need this for the video BIOS + * initialization quirk to work; this is likely to also + * be the case for other laptops or integrated video devices. + */ + /* GDT[0]: GDT self-pointer */ header->wakeup_gdt[0] = (u64)(sizeof(header->wakeup_gdt) - 1) + ((u64)(acpi_wakeup_address + ((char *)&header->wakeup_gdt - (char *)acpi_realmode)) << 16); - /* GDT[1]: real-mode-like code segment */ - header->wakeup_gdt[1] = (0x009bULL << 40) + - ((u64)acpi_wakeup_address << 16) + 0xffff; - /* GDT[2]: real-mode-like data segment */ - header->wakeup_gdt[2] = (0x0093ULL << 40) + - ((u64)acpi_wakeup_address << 16) + 0xffff; + /* GDT[1]: big real mode-like code segment */ + header->wakeup_gdt[1] = + GDT_ENTRY(0x809b, acpi_wakeup_address, 0xfffff); + /* GDT[2]: big real mode-like data segment */ + header->wakeup_gdt[2] = + GDT_ENTRY(0x8093, acpi_wakeup_address, 0xfffff); #ifndef CONFIG_64BIT store_gdt((struct desc_ptr *)&header->pmode_gdt); @@ -79,7 +89,7 @@ int acpi_save_state_mem(void) #endif /* !CONFIG_64BIT */ header->pmode_cr0 = read_cr0(); - header->pmode_cr4 = read_cr4(); + header->pmode_cr4 = read_cr4_safe(); header->realmode_flags = acpi_realmode_flags; header->real_magic = 0x12345678; @@ -89,7 +99,9 @@ int acpi_save_state_mem(void) saved_magic = 0x12345678; #else /* CONFIG_64BIT */ header->trampoline_segment = setup_trampoline() >> 4; - init_rsp = (unsigned long)temp_stack + 4096; +#ifdef CONFIG_SMP + stack_start.sp = temp_stack + 4096; +#endif initial_code = (unsigned long)wakeup_long64; saved_magic = 0x123456789abcdef0; #endif /* CONFIG_64BIT */ @@ -145,6 +157,12 @@ static int __init acpi_sleep_setup(char acpi_realmode_flags |= 2; if (strncmp(str, "s3_beep", 7) == 0) acpi_realmode_flags |= 4; +#ifdef CONFIG_HIBERNATION + if (strncmp(str, "s4_nohwsig", 10) == 0) + acpi_no_s4_hw_signature(); +#endif + if (strncmp(str, "old_ordering", 12) == 0) + acpi_old_suspend_ordering(); str = strchr(str, ','); if (str != NULL) str += strspn(str, ", \t"); --- head-2011-03-11.orig/arch/x86/kernel/amd_nb.c 2011-03-15 16:45:55.000000000 +0100 +++ head-2011-03-11/arch/x86/kernel/amd_nb.c 2011-02-01 14:38:38.000000000 +0100 @@ -15,6 +15,10 @@ static u32 *flush_words; struct pci_device_id amd_nb_misc_ids[] = { { PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_K8_NB_MISC) }, { PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_10H_NB_MISC) }, +#ifdef CONFIG_XEN + { PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_11H_NB_MISC) }, + { PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_CNB17H_F3) }, /* Fam12, Fam14 */ +#endif { PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_15H_NB_MISC) }, {} }; --- head-2011-03-11.orig/arch/x86/kernel/apic/apic-xen.c 2011-02-24 15:45:13.000000000 +0100 +++ head-2011-03-11/arch/x86/kernel/apic/apic-xen.c 2011-02-01 14:38:38.000000000 +0100 @@ -59,7 +59,10 @@ static cpumask_t timer_bcast_ipi; /* * Debug level, exported for io_apic.c */ -int apic_verbosity; +unsigned int apic_verbosity; + +/* Have we found an MP table */ +int smp_found_config; #ifndef CONFIG_XEN static int modern_apic(void) --- head-2011-03-11.orig/arch/x86/kernel/asm-offsets_64.c 2010-01-19 16:00:48.000000000 +0100 +++ head-2011-03-11/arch/x86/kernel/asm-offsets_64.c 2011-02-01 14:38:38.000000000 +0100 @@ -132,7 +132,7 @@ int main(void) BLANK(); DEFINE(PAGE_SIZE_asm, PAGE_SIZE); -#ifdef CONFIG_XEN +#ifdef CONFIG_PARAVIRT_XEN BLANK(); OFFSET(XEN_vcpu_info_mask, vcpu_info, evtchn_upcall_mask); OFFSET(XEN_vcpu_info_pending, vcpu_info, evtchn_upcall_pending); --- head-2011-03-11.orig/arch/x86/kernel/cpu/amd.c 2011-03-15 16:45:55.000000000 +0100 +++ head-2011-03-11/arch/x86/kernel/cpu/amd.c 2011-02-01 14:38:38.000000000 +0100 @@ -575,6 +575,7 @@ static void __cpuinit init_amd(struct cp fam10h_check_enable_mmcfg(); } +#ifndef CONFIG_XEN if (c == &boot_cpu_data && c->x86 >= 0xf) { unsigned long long tseg; @@ -594,6 +595,7 @@ static void __cpuinit init_amd(struct cp } } #endif +#endif } #ifdef CONFIG_X86_32 --- head-2011-03-11.orig/arch/x86/kernel/cpu/bugs_64.c 2011-03-15 16:45:55.000000000 +0100 +++ head-2011-03-11/arch/x86/kernel/cpu/bugs_64.c 2011-02-01 14:38:38.000000000 +0100 @@ -20,6 +20,7 @@ void __init check_bugs(void) #endif alternative_instructions(); +#ifndef CONFIG_XEN /* * Make sure the first 2MB area is not mapped by huge pages * There are typically fixed size MTRRs in there and overlapping @@ -30,4 +31,5 @@ void __init check_bugs(void) */ if (!direct_gbpages) set_memory_4k((unsigned long)__va(0), 1); +#endif } --- head-2011-03-11.orig/arch/x86/kernel/cpu/common-xen.c 2011-01-31 18:07:35.000000000 +0100 +++ head-2011-03-11/arch/x86/kernel/cpu/common-xen.c 2011-02-01 14:38:38.000000000 +0100 @@ -13,6 +13,7 @@ #include #include #include +#include #ifdef CONFIG_X86_LOCAL_APIC #include #include @@ -341,11 +342,24 @@ static void __init early_cpu_detect(void get_cpu_vendor(c, 1); + early_get_cap(c); + if (c->x86_vendor != X86_VENDOR_UNKNOWN && cpu_devs[c->x86_vendor]->c_early_init) cpu_devs[c->x86_vendor]->c_early_init(c); +} - early_get_cap(c); +/* + * The NOPL instruction is supposed to exist on all CPUs with + * family >= 6; unfortunately, that's not true in practice because + * of early VIA chips and (more importantly) broken virtualizers that + * are not easy to detect. In the latter case it doesn't even *fail* + * reliably, so probing for it doesn't even work. Disable it completely + * unless we can find a reliable way to detect all the broken cases. + */ +static void __cpuinit detect_nopl(struct cpuinfo_x86 *c) +{ + clear_cpu_cap(c, X86_FEATURE_NOPL); } static void __cpuinit generic_identify(struct cpuinfo_x86 *c) @@ -404,8 +418,8 @@ static void __cpuinit generic_identify(s } init_scattered_cpuid_features(c); + detect_nopl(c); } - } static void __cpuinit squash_the_stupid_serial_number(struct cpuinfo_x86 *c) @@ -436,7 +450,7 @@ __setup("serialnumber", x86_serial_nr_se /* * This does the hard work of actually picking apart the CPU stuff... */ -void __cpuinit identify_cpu(struct cpuinfo_x86 *c) +static void __cpuinit identify_cpu(struct cpuinfo_x86 *c) { int i; @@ -452,6 +466,8 @@ void __cpuinit identify_cpu(struct cpuin #endif c->x86_clflush_size = 32; memset(&c->x86_capability, 0, sizeof c->x86_capability); + if (boot_cpu_has(X86_FEATURE_SYSCALL32)) + set_cpu_cap(c, X86_FEATURE_SYSCALL32); if (!have_cpuid_p()) { /* --- /dev/null 1970-01-01 00:00:00.000000000 +0000 +++ head-2011-03-11/arch/x86/kernel/cpu/common_64-xen.c 2011-02-01 14:38:38.000000000 +0100 @@ -0,0 +1,777 @@ +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#ifdef CONFIG_X86_LOCAL_APIC +#include +#include +#include +#elif defined(CONFIG_XEN) +#include +#endif +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include "cpu.h" + +/* 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 + */ +/* The TLS descriptors are currently at a different place compared to i386. + Hopefully nobody expects them at a fixed place (Wine?) */ +DEFINE_PER_CPU(struct gdt_page, gdt_page) = { .gdt = { + [GDT_ENTRY_KERNEL32_CS] = { { { 0x0000ffff, 0x00cf9b00 } } }, + [GDT_ENTRY_KERNEL_CS] = { { { 0x0000ffff, 0x00af9b00 } } }, + [GDT_ENTRY_KERNEL_DS] = { { { 0x0000ffff, 0x00cf9300 } } }, + [GDT_ENTRY_DEFAULT_USER32_CS] = { { { 0x0000ffff, 0x00cffb00 } } }, + [GDT_ENTRY_DEFAULT_USER_DS] = { { { 0x0000ffff, 0x00cff300 } } }, + [GDT_ENTRY_DEFAULT_USER_CS] = { { { 0x0000ffff, 0x00affb00 } } }, +} }; +EXPORT_PER_CPU_SYMBOL_GPL(gdt_page); + +__u32 cleared_cpu_caps[NCAPINTS] __cpuinitdata; + +/* Current gdt points %fs at the "master" per-cpu area: after this, + * it's on the real one. */ +void switch_to_new_gdt(void) +{ +#ifndef CONFIG_XEN + struct desc_ptr gdt_descr; + + gdt_descr.address = (long)get_cpu_gdt_table(smp_processor_id()); + gdt_descr.size = GDT_SIZE - 1; + load_gdt(&gdt_descr); +#else + void *va, *gdt_addr = get_cpu_gdt_table(smp_processor_id()); + unsigned long frames[16]; + unsigned int f = 0; + + for (va = gdt_addr; va < gdt_addr + GDT_SIZE; va += PAGE_SIZE) { + frames[f++] = virt_to_mfn(va); + make_page_readonly(va, XENFEAT_writable_descriptor_tables); + } + if (HYPERVISOR_set_gdt(frames, GDT_SIZE / sizeof(struct desc_struct))) + BUG(); +#endif +} + +struct cpu_dev *cpu_devs[X86_VENDOR_NUM] = {}; + +static void __cpuinit default_init(struct cpuinfo_x86 *c) +{ + display_cacheinfo(c); +} + +static struct cpu_dev __cpuinitdata default_cpu = { + .c_init = default_init, + .c_vendor = "Unknown", +}; +static struct cpu_dev *this_cpu __cpuinitdata = &default_cpu; + +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; +} + + +void __cpuinit display_cacheinfo(struct cpuinfo_x86 *c) +{ + unsigned int n, dummy, 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); + } +} + +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 " + "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 +} + +static void __cpuinit get_cpu_vendor(struct cpuinfo_x86 *c) +{ + 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; + 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; +} + +static void __init early_cpu_support_print(void) +{ + int i,j; + struct cpu_dev *cpu_devx; + + printk("KERNEL supported cpus:\n"); + for (i = 0; i < X86_VENDOR_NUM; i++) { + cpu_devx = cpu_devs[i]; + if (!cpu_devx) + continue; + for (j = 0; j < 2; j++) { + if (!cpu_devx->c_ident[j]) + continue; + printk(" %s %s\n", cpu_devx->c_vendor, + cpu_devx->c_ident[j]); + } + } +} + +/* + * The NOPL instruction is supposed to exist on all CPUs with + * family >= 6, unfortunately, that's not true in practice because + * of early VIA chips and (more importantly) broken virtualizers that + * are not easy to detect. Hence, probe for it based on first + * principles. + * + * Note: no 64-bit chip is known to lack these, but put the code here + * for consistency with 32 bits, and to make it utterly trivial to + * diagnose the problem should it ever surface. + */ +static void __cpuinit detect_nopl(struct cpuinfo_x86 *c) +{ + const u32 nopl_signature = 0x888c53b1; /* Random number */ + u32 has_nopl = nopl_signature; + + clear_cpu_cap(c, X86_FEATURE_NOPL); + if (c->x86 >= 6) { + asm volatile("\n" + "1: .byte 0x0f,0x1f,0xc0\n" /* nopl %eax */ + "2:\n" + " .section .fixup,\"ax\"\n" + "3: xor %0,%0\n" + " jmp 2b\n" + " .previous\n" + _ASM_EXTABLE(1b,3b) + : "+a" (has_nopl)); + + if (has_nopl == nopl_signature) + set_cpu_cap(c, X86_FEATURE_NOPL); + } +} + +static void __cpuinit early_identify_cpu(struct cpuinfo_x86 *c); + +void __init early_cpu_init(void) +{ + struct cpu_vendor_dev *cvdev; + + for (cvdev = __x86cpuvendor_start ; + cvdev < __x86cpuvendor_end ; + cvdev++) + cpu_devs[cvdev->vendor] = cvdev->cpu_dev; + early_cpu_support_print(); + early_identify_cpu(&boot_cpu_data); +} + +/* 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. */ +static void __cpuinit early_identify_cpu(struct cpuinfo_x86 *c) +{ + u32 tfms, xlvl; + + 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; +#ifndef CONFIG_XEN + c->x86_max_cores = 1; + c->x86_coreid_bits = 0; +#endif + 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 (test_cpu_cap(c, X86_FEATURE_CLFLSH)) + c->x86_clflush_size = ((misc >> 8) & 0xff) * 8; + } else { + /* Have CPUID level 0 only - unheard of */ + c->x86 = 4; + } + +#ifndef CONFIG_XEN + c->initial_apicid = (cpuid_ebx(1) >> 24) & 0xff; +#ifdef CONFIG_SMP + c->phys_proc_id = c->initial_apicid; +#endif +#endif + /* 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); + } + + if (c->extended_cpuid_level >= 0x80000007) + c->x86_power = cpuid_edx(0x80000007); + + if (c->extended_cpuid_level >= 0x80000008) { + u32 eax = cpuid_eax(0x80000008); + + c->x86_virt_bits = (eax >> 8) & 0xff; + c->x86_phys_bits = eax & 0xff; + } + + detect_nopl(c); + + if (c->x86_vendor != X86_VENDOR_UNKNOWN && + cpu_devs[c->x86_vendor]->c_early_init) + cpu_devs[c->x86_vendor]->c_early_init(c); + + validate_pat_support(c); +} + +/* + * This does the hard work of actually picking apart the CPU stuff... + */ +static void __cpuinit identify_cpu(struct cpuinfo_x86 *c) +{ + int i; + + early_identify_cpu(c); + + init_scattered_cpuid_features(c); + +#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! + */ + if (this_cpu->c_init) + this_cpu->c_init(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]; + } + + /* Clear all flags overriden by options */ + for (i = 0; i < NCAPINTS; i++) + c->x86_capability[i] &= ~cleared_cpu_caps[i]; + +#ifdef CONFIG_X86_MCE + mcheck_init(c); +#endif + select_idle_routine(c); + +#ifdef CONFIG_NUMA + numa_add_cpu(smp_processor_id()); +#endif + +} + +void __cpuinit identify_boot_cpu(void) +{ + identify_cpu(&boot_cpu_data); +} + +void __cpuinit identify_secondary_cpu(struct cpuinfo_x86 *c) +{ + BUG_ON(c == &boot_cpu_data); + identify_cpu(c); + mtrr_ap_init(); +} + +static __init int setup_noclflush(char *arg) +{ + setup_clear_cpu_cap(X86_FEATURE_CLFLSH); + return 1; +} +__setup("noclflush", setup_noclflush); + +void __cpuinit print_cpu_info(struct cpuinfo_x86 *c) +{ + if (c->x86_model_id[0]) + printk(KERN_CONT "%s", c->x86_model_id); + + if (c->x86_mask || c->cpuid_level >= 0) + printk(KERN_CONT " stepping %02x\n", c->x86_mask); + else + printk(KERN_CONT "\n"); +} + +static __init int setup_disablecpuid(char *arg) +{ + int bit; + if (get_option(&arg, &bit) && bit < NCAPINTS*32) + setup_clear_cpu_cap(bit); + else + return 0; + return 1; +} +__setup("clearcpuid=", setup_disablecpuid); + +cpumask_t cpu_initialized __cpuinitdata = CPU_MASK_NONE; + +struct x8664_pda **_cpu_pda __read_mostly; +EXPORT_SYMBOL(_cpu_pda); + +#ifndef CONFIG_X86_NO_IDT +struct desc_ptr idt_descr = { 256 * 16 - 1, (unsigned long) idt_table }; +#endif + +char boot_cpu_stack[IRQSTACKSIZE] __page_aligned_bss; + +unsigned long __supported_pte_mask __read_mostly = ~0UL; +EXPORT_SYMBOL_GPL(__supported_pte_mask); + +static int do_not_nx __cpuinitdata; + +/* noexec=on|off +Control non executable mappings for 64bit processes. + +on Enable(default) +off Disable +*/ +static int __init nonx_setup(char *str) +{ + if (!str) + return -EINVAL; + 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 0; +} +early_param("noexec", nonx_setup); + +int force_personality32; + +/* 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 (default) +off PROT_READ implies PROT_EXEC +*/ +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); + +static void __init_refok switch_pt(int cpu) +{ +#ifdef CONFIG_XEN + if (cpu == 0) + xen_init_pt(); + xen_pt_switch(__pa_symbol(init_level4_pgt)); + xen_new_user_pt(__pa_symbol(__user_pgd(init_level4_pgt))); +#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 */ + loadsegment(fs, 0); + loadsegment(gs, 0); +#ifndef CONFIG_XEN + /* Memory clobbers used to order PDA accessed */ + mb(); + wrmsrl(MSR_GS_BASE, pda); + mb(); +#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) { + /* others are initialized in smpboot.c */ + pda->pcurrent = &init_task; + pda->irqstackptr = boot_cpu_stack; + pda->irqstackptr += IRQSTACKSIZE - 64; + } else { + if (!pda->irqstackptr) { + pda->irqstackptr = (char *) + __get_free_pages(GFP_ATOMIC, IRQSTACK_ORDER); + if (!pda->irqstackptr) + panic("cannot allocate irqstack for cpu %d", + cpu); + pda->irqstackptr += IRQSTACKSIZE - 64; + } + + if (pda->nodenumber == 0 && cpu_to_node(cpu) != NUMA_NO_NODE) + pda->nodenumber = cpu_to_node(cpu); + } + + switch_pt(cpu); +} + +#ifndef CONFIG_X86_NO_TSS +char boot_exception_stacks[(N_EXCEPTION_STACKS - 1) * EXCEPTION_STKSZ + + DEBUG_STKSZ] __page_aligned_bss; +#endif + +extern asmlinkage void ignore_sysret(void); + +void __cpuinit 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); + wrmsrl(MSR_CSTAR, ignore_sysret); + + /* Flags to clear on syscall */ + wrmsrl(MSR_SYSCALL_MASK, + X86_EFLAGS_TF|X86_EFLAGS_DF|X86_EFLAGS_IF|X86_EFLAGS_IOPL); +#endif +#ifdef CONFIG_IA32_EMULATION + syscall32_cpu_init(); +#else + static const struct callback_register __cpuinitconst cstar = { + .type = CALLBACKTYPE_syscall32, + .address = (unsigned long)ignore_sysret + }; + + if (HYPERVISOR_callback_op(CALLBACKOP_register, &cstar)) + printk(KERN_WARNING "Unable to register CSTAR callback\n"); +#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; + +#ifndef CONFIG_X86_NO_TSS +/* + * Copies of the original ist values from the tss are only accessed during + * debugging, no special alignment required. + */ +DEFINE_PER_CPU(struct orig_ist, orig_ist); +#endif + +/* + * 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; + int i; +#endif + struct task_struct *me; + + /* CPU 0 is initialised in head64.c */ + if (cpu != 0) + pda_init(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(KERN_INFO "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: + */ + + switch_to_new_gdt(); +#ifndef CONFIG_X86_NO_IDT + load_idt((const struct desc_ptr *)&idt_descr); +#endif + + 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 + */ + if (!orig_ist->ist[0]) { + static const unsigned int order[N_EXCEPTION_STACKS] = { + [0 ... N_EXCEPTION_STACKS - 1] = EXCEPTION_STACK_ORDER, + [DEBUG_STACK - 1] = DEBUG_STACK_ORDER + }; + for (v = 0; v < N_EXCEPTION_STACKS; v++) { + if (cpu) { + estacks = (char *)__get_free_pages(GFP_ATOMIC, order[v]); + if (!estacks) + panic("Cannot allocate exception " + "stack %ld %d\n", v, cpu); + } + estacks += PAGE_SIZE << order[v]; + orig_ist->ist[v] = t->x86_tss.ist[v] = + (unsigned long)estacks; + } + } + + t->x86_tss.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); + + load_sp0(t, ¤t->thread); +#ifndef CONFIG_X86_NO_TSS + set_tss_desc(cpu, t); + load_TR_desc(); +#endif + load_LDT(&init_mm.context); + +#ifdef CONFIG_KGDB + /* + * If the kgdb is connected no debug regs should be altered. This + * is only applicable when KGDB and a KGDB I/O module are built + * into the kernel and you are using early debugging with + * kgdbwait. KGDB will control the kernel HW breakpoint registers. + */ + if (kgdb_connected && arch_kgdb_ops.correct_hw_break) + arch_kgdb_ops.correct_hw_break(); + else { +#endif + /* + * 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); +#ifdef CONFIG_KGDB + /* If the kgdb is connected no debug regs should be altered. */ + } +#endif + + fpu_init(); + + asm ("pushfq; popq %0" : "=rm" (kernel_eflags)); + if (raw_irqs_disabled()) + kernel_eflags &= ~X86_EFLAGS_IF; + + if (is_uv_system()) + uv_cpu_init(); +} --- /dev/null 1970-01-01 00:00:00.000000000 +0000 +++ head-2011-03-11/arch/x86/kernel/e820-xen.c 2011-02-01 14:38:38.000000000 +0100 @@ -0,0 +1,1553 @@ +/* + * 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 , December 2002. + * Venkatesh Pallipadi + * + */ +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include +#include +#include + +/* + * The e820 map is the map that gets modified e.g. with command line parameters + * and that is also registered with modifications in the kernel resource tree + * with the iomem_resource as parent. + * + * The e820_saved is directly saved after the BIOS-provided memory map is + * copied. It doesn't get modified afterwards. It's registered for the + * /sys/firmware/memmap interface. + * + * That memory map is not modified and is used as base for kexec. The kexec'd + * kernel should get the same memory map as the firmware provides. Then the + * user can e.g. boot the original kernel with mem=1G while still booting the + * next kernel with full memory. + */ +struct e820map e820; +#ifndef CONFIG_XEN +struct e820map e820_saved; +#else +static struct e820map machine_e820; +#define e820_saved machine_e820 +#endif + +/* For PCI or other memory-mapped resources */ +unsigned long pci_mem_start = 0xaeedbabe; +#ifdef CONFIG_PCI +EXPORT_SYMBOL(pci_mem_start); +#endif + +/* + * This function checks if any part of the range 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++) { + 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 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(u64 start, u64 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 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; + } + return 0; +} + +/* + * Add a memory region to the kernel e820 map. + */ +void __init e820_add_region(u64 start, u64 size, int type) +{ + int x = e820.nr_map; + + if (x == ARRAY_SIZE(e820.map)) { + 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++; +} + +static void __init _e820_print_map(const struct e820map *e820, const char *who) +{ + int i; + + for (i = 0; i < e820->nr_map; i++) { + printk(KERN_INFO " %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: + case E820_RESERVED_KERN: + printk(KERN_CONT "(usable)\n"); + break; + case E820_RESERVED: + printk(KERN_CONT "(reserved)\n"); + break; + case E820_ACPI: + printk(KERN_CONT "(ACPI data)\n"); + break; + case E820_NVS: + printk(KERN_CONT "(ACPI NVS)\n"); + break; + default: + printk(KERN_CONT "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, + * and resolving conflicting memory types in favor of highest + * numbered type. + * + * The input parameter biosmap points to an array of 'struct + * e820entry' which on entry has elements in the range [0, *pnr_map) + * valid, and which has space for up to max_nr_map entries. + * On return, the resulting sanitized e820 map entries will be in + * overwritten in the same location, starting at biosmap. + * + * The integer pointed to by pnr_map must be valid on entry (the + * current number of valid entries located at biosmap) and will + * be updated on return, with the new number of valid entries + * (something no more than max_nr_map.) + * + * The return value from sanitize_e820_map() is zero if it + * successfully 'sanitized' the map entries passed in, and is -1 + * if it did nothing, which can happen if either of (1) it was + * only passed one map entry, or (2) any of the input map entries + * were invalid (start + size < start, meaning that the size was + * so big the described memory range wrapped around through zero.) + * + * 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_ + */ + +int __init sanitize_e820_map(struct e820entry *biosmap, int max_nr_map, + int *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*E820_X_MAX] __initdata; + static struct change_member *change_point[2*E820_X_MAX] __initdata; + static struct e820entry *overlap_list[E820_X_MAX] __initdata; + static struct e820entry new_bios[E820_X_MAX] __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; + + /* if there's only one memory region, don't bother */ +#ifdef CONFIG_XEN + if (*pnr_map == 1) + return 0; +#endif + if (*pnr_map < 2) + return -1; + + old_nr = *pnr_map; + BUG_ON(old_nr > max_nr_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++) { + unsigned long long curaddr, lastaddr; + unsigned long long curpbaddr, lastpbaddr; + + curaddr = change_point[i]->addr; + lastaddr = change_point[i - 1]->addr; + curpbaddr = change_point[i]->pbios->addr; + lastpbaddr = change_point[i - 1]->pbios->addr; + + /* + * swap entries, when: + * + * curaddr > lastaddr or + * curaddr == lastaddr and curaddr == curpbaddr and + * lastaddr != lastpbaddr + */ + if (curaddr < lastaddr || + (curaddr == lastaddr && curaddr == curpbaddr && + lastaddr != lastpbaddr)) { + 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) + /* + * no more space left for new + * bios entries ? + */ + if (++new_bios_entry >= max_nr_map) + break; + } + 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; + } + } + /* retain count for new bios entries */ + new_nr = new_bios_entry; + + /* copy new bios mapping into original location */ + memcpy(biosmap, new_bios, new_nr * sizeof(struct e820entry)); + *pnr_map = new_nr; + + return 0; +} + +static int __init __append_e820_map(struct e820entry *biosmap, int nr_map) +{ + while (nr_map) { + u64 start = biosmap->addr; + u64 size = biosmap->size; + u64 end = start + size; + u32 type = biosmap->type; + + /* Overflow in 64 bits? Ignore the memory map. */ + if (start > end) + return -1; + + e820_add_region(start, size, type); + + biosmap++; + nr_map--; + } + 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. + */ +static int __init append_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 + + return __append_e820_map(biosmap, nr_map); +} + +static u64 __init e820_update_range_map(struct e820map *e820x, u64 start, + u64 size, unsigned old_type, + unsigned new_type) +{ + unsigned int i, x; + u64 real_updated_size = 0; + + BUG_ON(old_type == new_type); + + if (size > (ULLONG_MAX - start)) + size = ULLONG_MAX - start; + + for (i = 0; i < e820x->nr_map; i++) { + struct e820entry *ei = &e820x->map[i]; + u64 final_start, final_end; + if (ei->type != old_type) + continue; + /* totally covered? */ + if (ei->addr >= start && + (ei->addr + ei->size) <= (start + size)) { + ei->type = new_type; + real_updated_size += ei->size; + continue; + } + /* partially covered */ + final_start = max(start, ei->addr); + final_end = min(start + size, ei->addr + ei->size); + if (final_start >= final_end) + continue; + + x = e820x->nr_map; + if (x == ARRAY_SIZE(e820x->map)) { + printk(KERN_ERR "Too many memory map entries!\n"); + break; + } + e820x->map[x].addr = final_start; + e820x->map[x].size = final_end - final_start; + e820x->map[x].type = new_type; + e820x->nr_map++; + + real_updated_size += final_end - final_start; + + if (ei->addr < final_start) + continue; + ei->addr = final_end; + ei->size -= final_end - final_start; + } + return real_updated_size; +} + +u64 __init e820_update_range(u64 start, u64 size, unsigned old_type, + unsigned new_type) +{ + return e820_update_range_map(&e820, start, size, old_type, new_type); +} + +static u64 __init e820_update_range_saved(u64 start, u64 size, + unsigned old_type, unsigned new_type) +{ +#ifdef CONFIG_XEN + if (is_initial_xendomain()) + return e820_update_range_map(&machine_e820, + phys_to_machine(start), size, + old_type, new_type); +#endif + return e820_update_range_map(&e820_saved, start, size, old_type, + new_type); +} + +/* make e820 not cover the range */ +u64 __init e820_remove_range(u64 start, u64 size, unsigned old_type, + int checktype) +{ + int i; + u64 real_removed_size = 0; + + if (size > (ULLONG_MAX - start)) + size = ULLONG_MAX - start; + + for (i = 0; i < e820.nr_map; i++) { + struct e820entry *ei = &e820.map[i]; + u64 final_start, final_end; + + if (checktype && ei->type != old_type) + continue; + /* totally covered? */ + if (ei->addr >= start && + (ei->addr + ei->size) <= (start + size)) { + real_removed_size += ei->size; + memset(ei, 0, sizeof(struct e820entry)); + continue; + } + /* partially covered */ + final_start = max(start, ei->addr); + final_end = min(start + size, ei->addr + ei->size); + if (final_start >= final_end) + continue; + real_removed_size += final_end - final_start; + + ei->size -= final_end - final_start; + if (ei->addr < final_start) + continue; + ei->addr = final_end; + } + return real_removed_size; +} + +void __init update_e820(void) +{ + int nr_map; + + nr_map = e820.nr_map; + if (sanitize_e820_map(e820.map, ARRAY_SIZE(e820.map), &nr_map)) + return; + e820.nr_map = nr_map; + printk(KERN_INFO "modified physical RAM map:\n"); + _e820_print_map(&e820, "modified"); +} +static void __init update_e820_saved(void) +{ + int nr_map; + + nr_map = e820_saved.nr_map; + if (sanitize_e820_map(e820_saved.map, ARRAY_SIZE(e820_saved.map), &nr_map)) + return; + e820_saved.nr_map = nr_map; +} + +#ifdef CONFIG_XEN +#define e820 machine_e820 +#endif + +#define MAX_GAP_END 0x100000000ull +/* + * Search for a gap in the e820 memory space from start_addr to end_addr. + */ +__init int e820_search_gap(unsigned long *gapstart, unsigned long *gapsize, + unsigned long start_addr, unsigned long long end_addr) +{ + unsigned long long last; + int i = e820.nr_map; + int found = 0; + + last = (end_addr && end_addr < MAX_GAP_END) ? end_addr : MAX_GAP_END; +#ifdef CONFIG_X86_64 + if (start_addr >= MAX_GAP_END) + last = end_addr ?: (1UL << boot_cpu_data.x86_phys_bits); +#endif + + while (--i >= 0) { + unsigned long long start = e820.map[i].addr; + unsigned long long end = start + e820.map[i].size; + + if (end < start_addr) + continue; + + /* + * 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; + } + return found; +} + +/* + * 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(void) +{ + unsigned long gapstart, gapsize, round; + int found; + + gapstart = 0x10000000; + gapsize = 0x400000; + found = e820_search_gap(&gapstart, &gapsize, 0, MAX_GAP_END); + +#ifdef CONFIG_X86_64 + if (!found) { + 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"); + found = e820_search_gap(&gapstart, &gapsize, MAX_GAP_END, 0); + WARN_ON(!found); + } +#endif + + /* + * 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); +} + +#undef e820 + +#ifndef CONFIG_XEN +/** + * Because of the size limitation of struct boot_params, only first + * 128 E820 memory entries are passed to kernel via + * boot_params.e820_map, others are passed via SETUP_E820_EXT node of + * linked list of struct setup_data, which is parsed here. + */ +void __init parse_e820_ext(struct setup_data *sdata, unsigned long pa_data) +{ + u32 map_len; + int entries; + struct e820entry *extmap; + + entries = sdata->len / sizeof(struct e820entry); + map_len = sdata->len + sizeof(struct setup_data); + if (map_len > PAGE_SIZE) + sdata = early_ioremap(pa_data, map_len); + extmap = (struct e820entry *)(sdata->data); + __append_e820_map(extmap, entries); + sanitize_e820_map(e820.map, ARRAY_SIZE(e820.map), &e820.nr_map); + if (map_len > PAGE_SIZE) + early_iounmap(sdata, map_len); + printk(KERN_INFO "extended physical RAM map:\n"); + _e820_print_map(&e820, "extended"); +} + +#if defined(CONFIG_X86_64) || \ + (defined(CONFIG_X86_32) && defined(CONFIG_HIBERNATION)) +/** + * Find the ranges of physical addresses that do not correspond to + * e820 RAM areas and mark the corresponding pages as nosave for + * hibernation (32 bit) or software suspend and suspend to RAM (64 bit). + * + * This function requires the e820 map to be sorted and without any + * overlapping entries and assumes the first e820 area to be RAM. + */ +void __init e820_mark_nosave_regions(unsigned long limit_pfn) +{ + int i; + unsigned long pfn; + + pfn = PFN_DOWN(e820.map[0].addr + e820.map[0].size); + for (i = 1; i < e820.nr_map; i++) { + struct e820entry *ei = &e820.map[i]; + + if (pfn < PFN_UP(ei->addr)) + register_nosave_region(pfn, PFN_UP(ei->addr)); + + pfn = PFN_DOWN(ei->addr + ei->size); + if (ei->type != E820_RAM && ei->type != E820_RESERVED_KERN) + register_nosave_region(PFN_UP(ei->addr), pfn); + + if (pfn >= limit_pfn) + break; + } +} +#endif +#endif + +/* + * Early reserved memory areas. + */ +#define MAX_EARLY_RES 20 + +struct early_res { + u64 start, end; + char name[16]; + char overlap_ok; +}; +static struct early_res early_res[MAX_EARLY_RES] __initdata = { +#ifndef CONFIG_XEN + { 0, PAGE_SIZE, "BIOS data page" }, /* BIOS data page */ +#if defined(CONFIG_X86_64) && defined(CONFIG_X86_TRAMPOLINE) + { TRAMPOLINE_BASE, TRAMPOLINE_BASE + 2 * PAGE_SIZE, "TRAMPOLINE" }, +#endif +#if defined(CONFIG_X86_32) && defined(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) + */ + { PAGE_SIZE, PAGE_SIZE + PAGE_SIZE, "EX TRAMPOLINE" }, + /* + * Has to be in very low memory so we can execute + * real-mode AP code. + */ + { TRAMPOLINE_BASE, TRAMPOLINE_BASE + PAGE_SIZE, "TRAMPOLINE" }, +#endif +#endif + {} +}; + +static int __init find_overlapped_early(u64 start, u64 end) +{ + int i; + struct early_res *r; + + for (i = 0; i < MAX_EARLY_RES && early_res[i].end; i++) { + r = &early_res[i]; + if (end > r->start && start < r->end) + break; + } + + return i; +} + +/* + * Drop the i-th range from the early reservation map, + * by copying any higher ranges down one over it, and + * clearing what had been the last slot. + */ +static void __init drop_range(int i) +{ + int j; + + for (j = i + 1; j < MAX_EARLY_RES && early_res[j].end; j++) + ; + + memmove(&early_res[i], &early_res[i + 1], + (j - 1 - i) * sizeof(struct early_res)); + + early_res[j - 1].end = 0; +} + +/* + * Split any existing ranges that: + * 1) are marked 'overlap_ok', and + * 2) overlap with the stated range [start, end) + * into whatever portion (if any) of the existing range is entirely + * below or entirely above the stated range. Drop the portion + * of the existing range that overlaps with the stated range, + * which will allow the caller of this routine to then add that + * stated range without conflicting with any existing range. + */ +static void __init drop_overlaps_that_are_ok(u64 start, u64 end) +{ + int i; + struct early_res *r; + u64 lower_start, lower_end; + u64 upper_start, upper_end; + char name[16]; + + for (i = 0; i < MAX_EARLY_RES && early_res[i].end; i++) { + r = &early_res[i]; + + /* Continue past non-overlapping ranges */ + if (end <= r->start || start >= r->end) + continue; + + /* + * Leave non-ok overlaps as is; let caller + * panic "Overlapping early reservations" + * when it hits this overlap. + */ + if (!r->overlap_ok) + return; + + /* + * We have an ok overlap. We will drop it from the early + * reservation map, and add back in any non-overlapping + * portions (lower or upper) as separate, overlap_ok, + * non-overlapping ranges. + */ + + /* 1. Note any non-overlapping (lower or upper) ranges. */ + strncpy(name, r->name, sizeof(name) - 1); + + lower_start = lower_end = 0; + upper_start = upper_end = 0; + if (r->start < start) { + lower_start = r->start; + lower_end = start; + } + if (r->end > end) { + upper_start = end; + upper_end = r->end; + } + + /* 2. Drop the original ok overlapping range */ + drop_range(i); + + i--; /* resume for-loop on copied down entry */ + + /* 3. Add back in any non-overlapping ranges. */ + if (lower_end) + reserve_early_overlap_ok(lower_start, lower_end, name); + if (upper_end) + reserve_early_overlap_ok(upper_start, upper_end, name); + } +} + +static void __init __reserve_early(u64 start, u64 end, char *name, + int overlap_ok) +{ + int i; + struct early_res *r; + + i = find_overlapped_early(start, end); + if (i >= MAX_EARLY_RES) + panic("Too many early reservations"); + r = &early_res[i]; + if (r->end) + panic("Overlapping early reservations " + "%llx-%llx %s to %llx-%llx %s\n", + start, end - 1, name?name:"", r->start, + r->end - 1, r->name); + r->start = start; + r->end = end; + r->overlap_ok = overlap_ok; + if (name) + strncpy(r->name, name, sizeof(r->name) - 1); +} + +/* + * A few early reservtations come here. + * + * The 'overlap_ok' in the name of this routine does -not- mean it + * is ok for these reservations to overlap an earlier reservation. + * Rather it means that it is ok for subsequent reservations to + * overlap this one. + * + * Use this entry point to reserve early ranges when you are doing + * so out of "Paranoia", reserving perhaps more memory than you need, + * just in case, and don't mind a subsequent overlapping reservation + * that is known to be needed. + * + * The drop_overlaps_that_are_ok() call here isn't really needed. + * It would be needed if we had two colliding 'overlap_ok' + * reservations, so that the second such would not panic on the + * overlap with the first. We don't have any such as of this + * writing, but might as well tolerate such if it happens in + * the future. + */ +void __init reserve_early_overlap_ok(u64 start, u64 end, char *name) +{ + drop_overlaps_that_are_ok(start, end); + __reserve_early(start, end, name, 1); +} + +/* + * Most early reservations come here. + * + * We first have drop_overlaps_that_are_ok() drop any pre-existing + * 'overlap_ok' ranges, so that we can then reserve this memory + * range without risk of panic'ing on an overlapping overlap_ok + * early reservation. + */ +void __init reserve_early(u64 start, u64 end, char *name) +{ + drop_overlaps_that_are_ok(start, end); + __reserve_early(start, end, name, 0); +} + +void __init free_early(u64 start, u64 end) +{ + struct early_res *r; + int i; + + i = find_overlapped_early(start, end); + r = &early_res[i]; + if (i >= MAX_EARLY_RES || r->end != end || r->start != start) + panic("free_early on not reserved area: %llx-%llx!", + start, end - 1); + + drop_range(i); +} + +void __init early_res_to_bootmem(u64 start, u64 end) +{ + int i, count; + u64 final_start, final_end; + + count = 0; + for (i = 0; i < MAX_EARLY_RES && early_res[i].end; i++) + count++; + + printk(KERN_INFO "(%d early reservations) ==> bootmem [%010llx - %010llx]\n", + count, start, end); + for (i = 0; i < count; i++) { + struct early_res *r = &early_res[i]; + printk(KERN_INFO " #%d [%010llx - %010llx] %16s", i, + r->start, r->end, r->name); + final_start = max(start, r->start); + final_end = min(end, r->end); + if (final_start >= final_end) { + printk(KERN_CONT "\n"); + continue; + } + printk(KERN_CONT " ==> [%010llx - %010llx]\n", + final_start, final_end); + reserve_bootmem_generic(final_start, final_end - final_start, + BOOTMEM_DEFAULT); + } +} + +/* Check for already reserved areas */ +static inline int __init bad_addr(u64 *addrp, u64 size, u64 align) +{ + int i; + u64 addr = *addrp; + int changed = 0; + struct early_res *r; +again: + i = find_overlapped_early(addr, addr + size); + r = &early_res[i]; + if (i < MAX_EARLY_RES && r->end) { + *addrp = addr = round_up(r->end, align); + changed = 1; + goto again; + } + return changed; +} + +/* Check for already reserved areas */ +static inline int __init bad_addr_size(u64 *addrp, u64 *sizep, u64 align) +{ + int i; + u64 addr = *addrp, last; + u64 size = *sizep; + int changed = 0; +again: + last = addr + size; + for (i = 0; i < MAX_EARLY_RES && early_res[i].end; i++) { + struct early_res *r = &early_res[i]; + if (last > r->start && addr < r->start) { + size = r->start - addr; + changed = 1; + goto again; + } + if (last > r->end && addr < r->end) { + addr = round_up(r->end, align); + size = last - addr; + changed = 1; + goto again; + } + if (last <= r->end && addr >= r->start) { + (*sizep)++; + return 0; + } + } + if (changed) { + *addrp = addr; + *sizep = size; + } + return changed; +} + +/* + * Find a free area with specified alignment in a specific range. + */ +u64 __init find_e820_area(u64 start, u64 end, u64 size, u64 align) +{ + int i; + + for (i = 0; i < e820.nr_map; i++) { + struct e820entry *ei = &e820.map[i]; + u64 addr, last; + u64 ei_last; + + if (ei->type != E820_RAM) + continue; + addr = round_up(ei->addr, align); + ei_last = ei->addr + ei->size; + if (addr < start) + addr = round_up(start, align); + if (addr >= ei_last) + continue; + while (bad_addr(&addr, size, align) && addr+size <= ei_last) + ; + last = addr + size; + if (last > ei_last) + continue; + if (last > end) + continue; + return addr; + } + return -1ULL; +} + +/* + * Find next free range after *start + */ +u64 __init find_e820_area_size(u64 start, u64 *sizep, u64 align) +{ + int i; + + for (i = 0; i < e820.nr_map; i++) { + struct e820entry *ei = &e820.map[i]; + u64 addr, last; + u64 ei_last; + + if (ei->type != E820_RAM) + continue; + addr = round_up(ei->addr, align); + ei_last = ei->addr + ei->size; + if (addr < start) + addr = round_up(start, align); + if (addr >= ei_last) + continue; + *sizep = ei_last - addr; + while (bad_addr_size(&addr, sizep, align) && + addr + *sizep <= ei_last) + ; + last = addr + *sizep; + if (last > ei_last) + continue; + return addr; + } + + return -1ULL; +} + +/* + * pre allocated 4k and reserved it in e820 + */ +u64 __init early_reserve_e820(u64 startt, u64 sizet, u64 align) +{ + u64 size = 0; + u64 addr; + u64 start; +#ifdef CONFIG_XEN + unsigned int order = get_order(sizet); + + if (is_initial_xendomain()) { + sizet = PAGE_SIZE << order; + if (align < PAGE_SIZE) + align = PAGE_SIZE; + } +#endif + for (start = startt; ; start += size) { + start = find_e820_area_size(start, &size, align); + if (!(start + 1)) + return 0; + if (size >= sizet) + break; + } + +#ifdef CONFIG_X86_32 + if (start >= MAXMEM) + return 0; + if (start + size > MAXMEM) + size = MAXMEM - start; +#endif +#ifdef CONFIG_XEN + if ((start >> PAGE_SHIFT) >= xen_start_info->nr_pages) + return 0; + if (PFN_UP(start + size) > xen_start_info->nr_pages) + size = ((u64)xen_start_info->nr_pages << PAGE_SHIFT) - start; +#endif + + addr = round_down(start + size - sizet, align); + if (addr < start) + return 0; +#ifdef CONFIG_XEN + if (is_initial_xendomain()) { + int rc; + unsigned long max_initmap_pfn; + + max_initmap_pfn = ALIGN(PFN_UP(__pa(xen_start_info->pt_base)) + + xen_start_info->nr_pt_frames + + 1 + (1 << (19 - PAGE_SHIFT)), + 1UL << (22 - PAGE_SHIFT)); +#ifdef CONFIG_X86_32 + if ((addr >> PAGE_SHIFT) + < max(max_initmap_pfn, max_pfn_mapped)) + rc = xen_create_contiguous_region((unsigned long) + __va(addr), + order, 32); +#else + if ((addr >> PAGE_SHIFT) < max_pfn_mapped) + rc = xen_create_contiguous_region((unsigned long) + __va(addr), + order, 32); + else if ((addr >> PAGE_SHIFT) < max_initmap_pfn) + rc = xen_create_contiguous_region(__START_KERNEL_map + + addr, + order, 32); +#endif + else + rc = early_create_contiguous_region(addr >> PAGE_SHIFT, + order, 32); + if (rc) + return 0; + } +#endif + e820_update_range(addr, sizet, E820_RAM, E820_RESERVED); + e820_update_range_saved(addr, sizet, E820_RAM, E820_RESERVED); + printk(KERN_INFO "update e820 for early_reserve_e820\n"); + update_e820(); + update_e820_saved(); + + return addr; +} + +#ifdef CONFIG_X86_32 +# ifdef CONFIG_X86_PAE +# define MAX_ARCH_PFN (1ULL<<(40-PAGE_SHIFT)) +# else +# define MAX_ARCH_PFN (1ULL<<(32-PAGE_SHIFT)) +# endif +#else /* CONFIG_X86_32 */ +# define MAX_ARCH_PFN MAXMEM>>PAGE_SHIFT +#endif + +/* + * Find the highest page frame number we have available + */ +static unsigned long __init e820_end_pfn(unsigned long limit_pfn, unsigned type) +{ + int i; + unsigned long last_pfn = 0; + unsigned long max_arch_pfn = MAX_ARCH_PFN; + + for (i = 0; i < e820.nr_map; i++) { + struct e820entry *ei = &e820.map[i]; + unsigned long start_pfn; + unsigned long end_pfn; + + if (ei->type != type) + continue; + + start_pfn = ei->addr >> PAGE_SHIFT; + end_pfn = (ei->addr + ei->size) >> PAGE_SHIFT; + + if (start_pfn >= limit_pfn) + continue; + if (end_pfn > limit_pfn) { + last_pfn = limit_pfn; + break; + } + if (end_pfn > last_pfn) + last_pfn = end_pfn; + } + + if (last_pfn > max_arch_pfn) + last_pfn = max_arch_pfn; + + printk(KERN_INFO "last_pfn = %#lx max_arch_pfn = %#lx\n", + last_pfn, max_arch_pfn); + return last_pfn; +} +unsigned long __init e820_end_of_ram_pfn(void) +{ + return e820_end_pfn(MAX_ARCH_PFN, E820_RAM); +} + +unsigned long __init e820_end_of_low_ram_pfn(void) +{ + return e820_end_pfn(1UL<<(32 - PAGE_SHIFT), E820_RAM); +} +/* + * Finds an active region in the address range from start_pfn to last_pfn and + * returns its range in ei_startpfn and ei_endpfn for the e820 entry. + */ +int __init e820_find_active_region(const struct e820entry *ei, + unsigned long start_pfn, + unsigned long last_pfn, + unsigned long *ei_startpfn, + unsigned long *ei_endpfn) +{ + u64 align = PAGE_SIZE; + +#ifdef CONFIG_XEN + if (last_pfn > xen_start_info->nr_pages) + last_pfn = xen_start_info->nr_pages; +#endif + + *ei_startpfn = round_up(ei->addr, align) >> PAGE_SHIFT; + *ei_endpfn = round_down(ei->addr + ei->size, align) >> PAGE_SHIFT; + + /* Skip map entries smaller than a page */ + if (*ei_startpfn >= *ei_endpfn) + return 0; + + /* Skip if map is outside the node */ + if (ei->type != E820_RAM || *ei_endpfn <= start_pfn || + *ei_startpfn >= last_pfn) + return 0; + + /* Check for overlaps */ + if (*ei_startpfn < start_pfn) + *ei_startpfn = start_pfn; + if (*ei_endpfn > last_pfn) + *ei_endpfn = last_pfn; + + return 1; +} + +/* Walk the e820 map and register active regions within a node */ +void __init e820_register_active_regions(int nid, unsigned long start_pfn, + unsigned long last_pfn) +{ + unsigned long ei_startpfn; + unsigned long ei_endpfn; + int i; + + for (i = 0; i < e820.nr_map; i++) + if (e820_find_active_region(&e820.map[i], + start_pfn, last_pfn, + &ei_startpfn, &ei_endpfn)) + add_active_range(nid, ei_startpfn, ei_endpfn); +#ifdef CONFIG_XEN + BUG_ON(nid); + add_active_range(nid, last_pfn, last_pfn); +#endif +} + +/* + * Find the hole size (in bytes) in the memory range. + * @start: starting address of the memory range to scan + * @end: ending address of the memory range to scan + */ +u64 __init e820_hole_size(u64 start, u64 end) +{ + unsigned long start_pfn = start >> PAGE_SHIFT; + unsigned long last_pfn = end >> PAGE_SHIFT; + unsigned long ei_startpfn, ei_endpfn, ram = 0; + int i; + + for (i = 0; i < e820.nr_map; i++) { + if (e820_find_active_region(&e820.map[i], + start_pfn, last_pfn, + &ei_startpfn, &ei_endpfn)) + ram += ei_endpfn - ei_startpfn; + } + return end - start - ((u64)ram << PAGE_SHIFT); +} + +static void early_panic(char *msg) +{ + early_printk(msg); + panic(msg); +} + +static int userdef __initdata; + +/* "mem=nopentium" disables the 4MB page tables. */ +static int __init parse_memopt(char *p) +{ + u64 mem_size, current_end; + unsigned int i; + + if (!p) + return -EINVAL; + +#ifdef CONFIG_X86_32 + if (!strcmp(p, "nopentium")) { + setup_clear_cpu_cap(X86_FEATURE_PSE); + return 0; + } +#endif + + userdef = 1; + mem_size = memparse(p, &p); + e820_remove_range(mem_size, ULLONG_MAX - mem_size, E820_RAM, 1); + + i = e820.nr_map - 1; + current_end = e820.map[i].addr + e820.map[i].size; + if (current_end < mem_size) { + /* + * 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 = mem_size - e820.map[i].addr; + else + e820_add_region(current_end, mem_size - current_end, E820_RAM); + } + + return 0; +} +early_param("mem", parse_memopt); + +#ifndef CONFIG_XEN +static int __init parse_memmap_opt(char *p) +{ + char *oldp; + u64 start_at, mem_size; + + if (!p) + return -EINVAL; + + if (!strncmp(p, "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_pfn(); +#endif + e820.nr_map = 0; + userdef = 1; + return 0; + } + + oldp = p; + mem_size = memparse(p, &p); + if (p == oldp) + return -EINVAL; + + userdef = 1; + if (*p == '@') { + start_at = memparse(p+1, &p); + e820_add_region(start_at, mem_size, E820_RAM); + } else if (*p == '#') { + start_at = memparse(p+1, &p); + e820_add_region(start_at, mem_size, E820_ACPI); + } else if (*p == '$') { + start_at = memparse(p+1, &p); + e820_add_region(start_at, mem_size, E820_RESERVED); + } else + e820_remove_range(mem_size, ULLONG_MAX - mem_size, E820_RAM, 1); + + return *p == '\0' ? 0 : -EINVAL; +} +early_param("memmap", parse_memmap_opt); +#endif + +void __init finish_e820_parsing(void) +{ + if (userdef) { + int nr = e820.nr_map; + + if (sanitize_e820_map(e820.map, ARRAY_SIZE(e820.map), &nr) < 0) + early_panic("Invalid user supplied memory map"); + e820.nr_map = nr; + + printk(KERN_INFO "user-defined physical RAM map:\n"); + _e820_print_map(&e820, "user"); + } +} + +static inline const char *e820_type_to_string(int e820_type) +{ + switch (e820_type) { + case E820_RESERVED_KERN: + case E820_RAM: return "System RAM"; + case E820_ACPI: return "ACPI Tables"; + case E820_NVS: return "ACPI Non-volatile Storage"; + default: return "reserved"; + } +} + +#ifdef CONFIG_XEN +#define e820 machine_e820 +#endif + +/* + * Mark e820 reserved areas as busy for the resource manager. + */ +void __init e820_reserve_resources(void) +{ + int i; + struct resource *res; + u64 end; + + res = alloc_bootmem_low(sizeof(struct resource) * e820.nr_map); + for (i = 0; i < e820.nr_map; i++) { + end = e820.map[i].addr + e820.map[i].size - 1; +#ifndef CONFIG_RESOURCES_64BIT + if (end > 0x100000000ULL) { + res++; + continue; + } +#endif + res->name = e820_type_to_string(e820.map[i].type); + res->start = e820.map[i].addr; + res->end = end; + + res->flags = IORESOURCE_MEM | IORESOURCE_BUSY; + insert_resource(&iomem_resource, res); + res++; + } + + for (i = 0; i < e820_saved.nr_map; i++) { + struct e820entry *entry = &e820_saved.map[i]; + firmware_map_add_early(entry->addr, + entry->addr + entry->size - 1, + e820_type_to_string(entry->type)); + } +} + +#undef e820 + +#ifndef CONFIG_XEN +char *__init default_machine_specific_memory_setup(void) +{ + char *who = "BIOS-e820"; + int new_nr; + /* + * 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 + */ + new_nr = boot_params.e820_entries; + sanitize_e820_map(boot_params.e820_map, + ARRAY_SIZE(boot_params.e820_map), + &new_nr); + boot_params.e820_entries = new_nr; + if (append_e820_map(boot_params.e820_map, boot_params.e820_entries) + < 0) { + u64 mem_size; + + /* compare results from other methods and take the greater */ + if (boot_params.alt_mem_k + < boot_params.screen_info.ext_mem_k) { + mem_size = boot_params.screen_info.ext_mem_k; + who = "BIOS-88"; + } else { + mem_size = boot_params.alt_mem_k; + who = "BIOS-e801"; + } + + e820.nr_map = 0; + e820_add_region(0, LOWMEMSIZE(), E820_RAM); + e820_add_region(HIGH_MEMORY, mem_size << 10, E820_RAM); + } + + /* In case someone cares... */ + return who; +} + +char *__init __attribute__((weak)) machine_specific_memory_setup(void) +{ + if (x86_quirks->arch_memory_setup) { + char *who = x86_quirks->arch_memory_setup(); + + if (who) + return who; + } + return default_machine_specific_memory_setup(); +} +#endif + +static char * __init _memory_setup(void) +{ + int rc, nr_map; + struct xen_memory_map memmap; + static struct e820entry __initdata 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); + + nr_map = memmap.nr_entries; + sanitize_e820_map(map, ARRAY_SIZE(map), &nr_map); + + if (append_e820_map(map, nr_map) < 0) + BUG(); + +#ifdef CONFIG_XEN + if (is_initial_xendomain()) { + 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; + } +#endif + + return "Xen"; +} + +void __init setup_memory_map(void) +{ + char *who; + + who = _memory_setup(); +#ifdef CONFIG_XEN + if (is_initial_xendomain()) { + printk(KERN_INFO "Xen-provided machine memory map:\n"); + _e820_print_map(&machine_e820, "BIOS"); + } else +#endif + memcpy(&e820_saved, &e820, sizeof(struct e820map)); + printk(KERN_INFO "Xen-provided physical RAM map:\n"); + _e820_print_map(&e820, who); +} --- head-2011-03-11.orig/arch/x86/kernel/e820_32-xen.c 2011-01-31 18:07:35.000000000 +0100 +++ /dev/null 1970-01-01 00:00:00.000000000 +0000 @@ -1,860 +0,0 @@ -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include - -#include -#include -#include -#include -#include - -struct e820map e820; -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; -/* For PCI or other memory-mapped resources */ -unsigned long pci_mem_start = 0x10000000; -#ifdef CONFIG_PCI -EXPORT_SYMBOL(pci_mem_start); -#endif -extern int user_defined_memmap; - -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 -} }; - -static struct resource video_rom_resource = { - .name = "Video ROM", - .start = 0xc0000, - .end = 0xc7fff, - .flags = IORESOURCE_BUSY | IORESOURCE_READONLY | IORESOURCE_MEM -}; - -#define ROMSIGNATURE 0xaa55 - -static int __init romsignature(const unsigned char *rom) -{ - const unsigned short * const ptr = (const unsigned short *)rom; - unsigned short sig; - - return probe_kernel_address(ptr, sig) == 0 && sig == ROMSIGNATURE; -} - -static int __init romchecksum(const unsigned char *rom, unsigned long length) -{ - unsigned char sum, c; - - for (sum = 0; length && probe_kernel_address(rom++, c) == 0; length--) - sum += c; - return !length && !sum; -} - -static void __init probe_roms(void) -{ - const unsigned char *rom; - unsigned long start, length, upper; - unsigned char c; - 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; - - if (probe_kernel_address(rom + 2, c) != 0) - continue; - - /* 0 < length <= 0x7f * 512, historically */ - length = c * 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((unsigned long)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 < ARRAY_SIZE(adapter_rom_resources) && start < upper; start += 2048) { - rom = isa_bus_to_virt(start); - if (!romsignature(rom)) - continue; - - if (probe_kernel_address(rom + 2, c) != 0) - continue; - - /* 0 < length <= 0x7f * 512, historically */ - length = c * 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; - } -} - -#ifdef CONFIG_XEN -static struct e820map machine_e820; -#define e820 machine_e820 -#endif - -/* - * Request address space for all standard RAM and ROM resources - * and also for regions reported as reserved by the e820. - */ -void __init init_iomem_resources(struct resource *code_resource, - struct resource *data_resource, - struct resource *bss_resource) -{ - int i; - - probe_roms(); - for (i = 0; i < e820.nr_map; i++) { - struct resource *res; -#ifndef CONFIG_RESOURCES_64BIT - if (e820.map[i].addr + e820.map[i].size > 0x100000000ULL) - continue; -#endif - res = kzalloc(sizeof(struct resource), GFP_ATOMIC); - switch (e820.map[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.map[i].addr; - res->end = res->start + e820.map[i].size - 1; - res->flags = IORESOURCE_MEM | IORESOURCE_BUSY; - if (request_resource(&iomem_resource, res)) { - kfree(res); - continue; - } - if (e820.map[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); - request_resource(res, bss_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 - } - } -} - -#undef e820 - -#if defined(CONFIG_PM) && defined(CONFIG_HIBERNATION) -/** - * e820_mark_nosave_regions - Find the ranges of physical addresses that do not - * correspond to e820 RAM areas and mark the corresponding pages as nosave for - * hibernation. - * - * This function requires the e820 map to be sorted and without any - * overlapping entries and assumes the first e820 area to be RAM. - */ -void __init e820_mark_nosave_regions(void) -{ - int i; - unsigned long pfn; - - pfn = PFN_DOWN(e820.map[0].addr + e820.map[0].size); - for (i = 1; i < e820.nr_map; i++) { - struct e820entry *ei = &e820.map[i]; - - if (pfn < PFN_UP(ei->addr)) - register_nosave_region(pfn, PFN_UP(ei->addr)); - - pfn = PFN_DOWN(ei->addr + ei->size); - if (ei->type != E820_RAM) - register_nosave_region(PFN_UP(ei->addr), pfn); - - if (pfn >= max_low_pfn) - break; - } -} -#endif - -void __init add_memory_region(unsigned long long start, - unsigned long long size, int type) -{ - int x; - - 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 */ - -/* - * Sanitize the BIOS e820 map. - * - * Some e820 responses include overlapping entries. The following - * replaces the original e820 map with a new one, removing overlaps. - * - */ -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; iaddr = 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 > , swap */ - /* or, if current= & last=, 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; ipbios) - 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; itype > 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 { - u64 start = biosmap->addr; - u64 size = biosmap->size; - u64 end = start + size; - u32 type = biosmap->type; - - /* Overflow in 64 bits? Ignore the memory map. */ - if (start > end) - return -1; - - 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; -} - -/* - * Find the highest page frame number we have available - */ -void __init propagate_e820_map(void) -{ - int i; - - max_pfn = 0; - - 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); - } -} - -/* - * Register fully available low RAM pages with the bootmem allocator. - */ -void __init register_bootmem_low_pages(unsigned long max_low_pfn) -{ - int i; - - 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)); - } -} - -void __init e820_register_memory(void) -{ - unsigned long gapstart, gapsize, round; - unsigned long long last; - int i; - -#ifdef CONFIG_XEN -#define e820 machine_e820 -#endif - /* - * Search for the biggest gap in the low 32 bits of the e820 - * memory space. - */ - last = 0x100000000ull; - gapstart = 0x10000000; - gapsize = 0x400000; - i = e820.nr_map; - while (--i >= 0) { - unsigned long long start = e820.map[i].addr; - unsigned long long end = start + e820.map[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; - } -#undef e820 - - /* - * 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); -} - -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 %u\n", e820.map[i].type); - break; - } - } -} - -void __init limit_regions(unsigned long long size) -{ - unsigned long long current_addr = 0; - int i; - - print_memory_map("limit_regions start"); - 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; - } - print_memory_map("limit_regions endfor"); - return; - } -#ifdef CONFIG_XEN - if (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 - print_memory_map("limit_regions endfunc"); -} - -/* - * This function checks if any part of the range 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 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 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; -} - -static int __init parse_memmap(char *arg) -{ - if (!arg) - return -EINVAL; - - if (strcmp(arg, "exactmap") == 0) { -#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. - */ - propagate_e820_map(); - saved_max_pfn = max_pfn; -#endif - e820.nr_map = 0; - user_defined_memmap = 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(arg, &arg); - if (*arg == '@') { - start_at = memparse(arg+1, &arg); - add_memory_region(start_at, mem_size, E820_RAM); - } else if (*arg == '#') { - start_at = memparse(arg+1, &arg); - add_memory_region(start_at, mem_size, E820_ACPI); - } else if (*arg == '$') { - start_at = memparse(arg+1, &arg); - add_memory_region(start_at, mem_size, E820_RESERVED); - } else { - limit_regions(mem_size); - user_defined_memmap = 1; - } - } - return 0; -} -early_param("memmap", parse_memmap); - -#ifndef CONFIG_XEN -void __init update_memory_range(u64 start, u64 size, unsigned old_type, - unsigned new_type) -{ - int i; - - BUG_ON(old_type == new_type); - - for (i = 0; i < e820.nr_map; i++) { - struct e820entry *ei = &e820.map[i]; - u64 final_start, final_end; - if (ei->type != old_type) - continue; - /* totally covered? */ - if (ei->addr >= start && ei->size <= size) { - ei->type = new_type; - continue; - } - /* partially covered */ - final_start = max(start, ei->addr); - final_end = min(start + size, ei->addr + ei->size); - if (final_start >= final_end) - continue; - add_memory_region(final_start, final_end - final_start, - new_type); - } -} - -void __init update_e820(void) -{ - u8 nr_map; - - nr_map = e820.nr_map; - if (sanitize_e820_map(e820.map, &nr_map)) - return; - e820.nr_map = nr_map; - printk(KERN_INFO "modified physical RAM map:\n"); - print_memory_map("modified"); -} -#endif --- head-2011-03-11.orig/arch/x86/kernel/e820_64-xen.c 2011-01-31 18:07:35.000000000 +0100 +++ /dev/null 1970-01-01 00:00:00.000000000 +0000 @@ -1,1052 +0,0 @@ -/* - * 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 , December 2002. - * Venkatesh Pallipadi - * - */ -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include - -#include -#include -#include -#include -#include -#include -#include -#include - -struct e820map e820 __initdata; -#ifdef CONFIG_XEN -struct e820map machine_e820; -#endif - -/* - * PFN of last memory page. - */ -unsigned long end_pfn; - -#ifndef CONFIG_XEN -/* - * end_pfn only includes RAM, while max_pfn_mapped includes all e820 entries. - * The direct mapping extends to max_pfn_mapped, so that we can directly access - * apertures, ACPI and other tables without having to play with fixmaps. - */ -unsigned long max_pfn_mapped; -#endif - -/* - * Last pfn which the user wants to use. - */ -static unsigned long __initdata end_user_pfn = MAXMEM>>PAGE_SHIFT; - -/* - * Early reserved memory areas. - */ -#define MAX_EARLY_RES 20 - -struct early_res { - unsigned long start, end; - char name[16]; -}; -static struct early_res early_res[MAX_EARLY_RES] __initdata = { -#ifndef CONFIG_XEN - { 0, PAGE_SIZE, "BIOS data page" }, /* BIOS data page */ -#ifdef CONFIG_X86_TRAMPOLINE - { TRAMPOLINE_BASE, TRAMPOLINE_BASE + 2 * PAGE_SIZE, "TRAMPOLINE" }, -#endif -#endif - {} -}; - -void __init reserve_early(unsigned long start, unsigned long end, char *name) -{ - int i; - struct early_res *r; - for (i = 0; i < MAX_EARLY_RES && early_res[i].end; i++) { - r = &early_res[i]; - if (end > r->start && start < r->end) - panic("Overlapping early reservations %lx-%lx %s to %lx-%lx %s\n", - start, end - 1, name?name:"", r->start, r->end - 1, r->name); - } - if (i >= MAX_EARLY_RES) - panic("Too many early reservations"); - r = &early_res[i]; - r->start = start; - r->end = end; - if (name) - strncpy(r->name, name, sizeof(r->name) - 1); -} - -void __init free_early(unsigned long start, unsigned long end) -{ - struct early_res *r; - int i, j; - - for (i = 0; i < MAX_EARLY_RES && early_res[i].end; i++) { - r = &early_res[i]; - if (start == r->start && end == r->end) - break; - } - if (i >= MAX_EARLY_RES || !early_res[i].end) - panic("free_early on not reserved area: %lx-%lx!", start, end); - - for (j = i + 1; j < MAX_EARLY_RES && early_res[j].end; j++) - ; - - memmove(&early_res[i], &early_res[i + 1], - (j - 1 - i) * sizeof(struct early_res)); - - early_res[j - 1].end = 0; -} - -void __init early_res_to_bootmem(unsigned long start, unsigned long end) -{ - int i; - unsigned long final_start, final_end; - for (i = 0; i < MAX_EARLY_RES && early_res[i].end; i++) { - struct early_res *r = &early_res[i]; - final_start = max(start, r->start); - final_end = min(end, r->end); - if (final_start >= final_end) - continue; - printk(KERN_INFO " early res: %d [%lx-%lx] %s\n", i, - final_start, final_end - 1, r->name); - reserve_bootmem_generic(final_start, final_end - final_start); - } -} - -/* Check for already reserved areas */ -static inline int __init -bad_addr(unsigned long *addrp, unsigned long size, unsigned long align) -{ - int i; - unsigned long addr = *addrp, last; - int changed = 0; -again: - last = addr + size; - for (i = 0; i < MAX_EARLY_RES && early_res[i].end; i++) { - struct early_res *r = &early_res[i]; - if (last >= r->start && addr < r->end) { - *addrp = addr = round_up(r->end, align); - changed = 1; - goto again; - } - } - return changed; -} - -/* Check for already reserved areas */ -static inline int __init -bad_addr_size(unsigned long *addrp, unsigned long *sizep, unsigned long align) -{ - int i; - unsigned long addr = *addrp, last; - unsigned long size = *sizep; - int changed = 0; -again: - last = addr + size; - for (i = 0; i < MAX_EARLY_RES && early_res[i].end; i++) { - struct early_res *r = &early_res[i]; - if (last > r->start && addr < r->start) { - size = r->start - addr; - changed = 1; - goto again; - } - if (last > r->end && addr < r->end) { - addr = round_up(r->end, align); - size = last - addr; - changed = 1; - goto again; - } - if (last <= r->end && addr >= r->start) { - (*sizep)++; - return 0; - } - } - if (changed) { - *addrp = addr; - *sizep = size; - } - return changed; -} -/* - * This function checks if any part of the range 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 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 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; - } - return 0; -} - -/* - * Find a free area with specified alignment in a specific range. - */ -unsigned long __init find_e820_area(unsigned long start, unsigned long end, - unsigned long size, unsigned long align) -{ - int i; - - for (i = 0; i < e820.nr_map; i++) { - struct e820entry *ei = &e820.map[i]; - unsigned long addr, last; - unsigned long ei_last; - - if (ei->type != E820_RAM) - continue; - addr = round_up(ei->addr, align); - ei_last = ei->addr + ei->size; - if (addr < start) - addr = round_up(start, align); - if (addr >= ei_last) - continue; - while (bad_addr(&addr, size, align) && addr+size <= ei_last) - ; - last = addr + size; - if (last > ei_last) - continue; - if (last > end) - continue; - return addr; - } - return -1UL; -} - -/* - * Find next free range after *start - */ -unsigned long __init find_e820_area_size(unsigned long start, - unsigned long *sizep, - unsigned long align) -{ - int i; - - for (i = 0; i < e820.nr_map; i++) { - struct e820entry *ei = &e820.map[i]; - unsigned long addr, last; - unsigned long ei_last; - - if (ei->type != E820_RAM) - continue; - addr = round_up(ei->addr, align); - ei_last = ei->addr + ei->size; - if (addr < start) - addr = round_up(start, align); - if (addr >= ei_last) - continue; - *sizep = ei_last - addr; - while (bad_addr_size(&addr, sizep, align) && - addr + *sizep <= ei_last) - ; - last = addr + *sizep; - if (last > ei_last) - continue; - return addr; - } - return -1UL; - -} -/* - * Find the highest page frame number we have available - */ -unsigned long __init e820_end_of_ram(void) -{ - unsigned long end_pfn; - - end_pfn = find_max_pfn_with_active_regions(); - - if (end_pfn > max_pfn_mapped) - max_pfn_mapped = end_pfn; - if (max_pfn_mapped > MAXMEM>>PAGE_SHIFT) - max_pfn_mapped = MAXMEM>>PAGE_SHIFT; - if (end_pfn > end_user_pfn) - end_pfn = end_user_pfn; - if (end_pfn > max_pfn_mapped) - end_pfn = max_pfn_mapped; - - printk(KERN_INFO "max_pfn_mapped = %lu\n", max_pfn_mapped); - return end_pfn; -} - -/* - * Mark e820 reserved areas as busy for the resource manager. - */ -void __init e820_reserve_resources(struct e820entry *e820, int nr_map) -{ - int i; - struct resource *res; - - res = alloc_bootmem_low(sizeof(struct resource) * nr_map); - for (i = 0; i < nr_map; i++) { - 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; - insert_resource(&iomem_resource, res); - res++; - } -} - -#ifndef CONFIG_XEN -/* - * Find the ranges of physical addresses that do not correspond to - * e820 RAM areas and mark the corresponding pages as nosave for software - * suspend and suspend to RAM. - * - * This function requires the e820 map to be sorted and without any - * overlapping entries and assumes the first e820 area to be RAM. - */ -void __init e820_mark_nosave_regions(void) -{ - int i; - unsigned long paddr; - - paddr = round_down(e820.map[0].addr + e820.map[0].size, PAGE_SIZE); - for (i = 1; i < e820.nr_map; i++) { - struct e820entry *ei = &e820.map[i]; - - if (paddr < ei->addr) - register_nosave_region(PFN_DOWN(paddr), - PFN_UP(ei->addr)); - - paddr = round_down(ei->addr + ei->size, PAGE_SIZE); - if (ei->type != E820_RAM) - register_nosave_region(PFN_UP(ei->addr), - PFN_DOWN(paddr)); - - if (paddr >= (end_pfn << PAGE_SHIFT)) - break; - } -} -#endif - -/* - * Finds an active region in the address range from start_pfn to end_pfn and - * returns its range in ei_startpfn and ei_endpfn for the e820 entry. - */ -static int __init e820_find_active_region(const struct e820entry *ei, - unsigned long start_pfn, - unsigned long end_pfn, - unsigned long *ei_startpfn, - unsigned long *ei_endpfn) -{ -#ifdef CONFIG_XEN - if (end_pfn > xen_start_info->nr_pages) - end_pfn = xen_start_info->nr_pages; -#endif - - *ei_startpfn = round_up(ei->addr, PAGE_SIZE) >> PAGE_SHIFT; - *ei_endpfn = round_down(ei->addr + ei->size, PAGE_SIZE) >> PAGE_SHIFT; - - /* Skip map entries smaller than a page */ - if (*ei_startpfn >= *ei_endpfn) - return 0; - - /* Check if max_pfn_mapped should be updated */ - if (ei->type != E820_RAM && *ei_endpfn > max_pfn_mapped) - max_pfn_mapped = *ei_endpfn; - - /* Skip if map is outside the node */ - if (ei->type != E820_RAM || *ei_endpfn <= start_pfn || - *ei_startpfn >= end_pfn) - return 0; - - /* Check for overlaps */ - if (*ei_startpfn < start_pfn) - *ei_startpfn = start_pfn; - if (*ei_endpfn > end_pfn) - *ei_endpfn = end_pfn; - - /* Obey end_user_pfn to save on memmap */ - if (*ei_startpfn >= end_user_pfn) - return 0; - if (*ei_endpfn > end_user_pfn) - *ei_endpfn = end_user_pfn; - - return 1; -} - -/* Walk the e820 map and register active regions within a node */ -void __init -e820_register_active_regions(int nid, unsigned long start_pfn, - unsigned long end_pfn) -{ - unsigned long ei_startpfn; - unsigned long ei_endpfn; - int i; - - for (i = 0; i < e820.nr_map; i++) - if (e820_find_active_region(&e820.map[i], - start_pfn, end_pfn, - &ei_startpfn, &ei_endpfn)) - add_active_range(nid, ei_startpfn, ei_endpfn); -#ifdef CONFIG_XEN - BUG_ON(nid); - add_active_range(nid, end_pfn, end_pfn); -#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++; -} - -/* - * Find the hole size (in bytes) in the memory range. - * @start: starting address of the memory range to scan - * @end: ending address of the memory range to scan - */ -unsigned long __init e820_hole_size(unsigned long start, unsigned long end) -{ - unsigned long start_pfn = start >> PAGE_SHIFT; - unsigned long end_pfn = end >> PAGE_SHIFT; - unsigned long ei_startpfn, ei_endpfn, ram = 0; - int i; - - for (i = 0; i < e820.nr_map; i++) { - if (e820_find_active_region(&e820.map[i], - start_pfn, end_pfn, - &ei_startpfn, &ei_endpfn)) - ram += ei_endpfn - ei_startpfn; - } - return end - start - (ram << PAGE_SHIFT); -} - -static void __init e820_print_map(char *who) -{ - int i; - - for (i = 0; i < e820.nr_map; i++) { - printk(KERN_INFO " %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(KERN_CONT "(usable)\n"); - break; - case E820_RESERVED: - printk(KERN_CONT "(reserved)\n"); - break; - case E820_ACPI: - printk(KERN_CONT "(ACPI data)\n"); - break; - case E820_NVS: - printk(KERN_CONT "(ACPI NVS)\n"); - break; - default: - printk(KERN_CONT "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++) { - unsigned long long curaddr, lastaddr; - unsigned long long curpbaddr, lastpbaddr; - - curaddr = change_point[i]->addr; - lastaddr = change_point[i - 1]->addr; - curpbaddr = change_point[i]->pbios->addr; - lastpbaddr = change_point[i - 1]->pbios->addr; - - /* - * swap entries, when: - * - * curaddr > lastaddr or - * curaddr == lastaddr and curaddr == curpbaddr and - * lastaddr != lastpbaddr - */ - if (curaddr < lastaddr || - (curaddr == lastaddr && curaddr == curpbaddr && - lastaddr != lastpbaddr)) { - 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) - /* - * no more space left for new - * bios entries ? - */ - if (++new_bios_entry >= E820MAX) - break; - } - 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; - } - } - /* retain count for new bios entries */ - new_nr = new_bios_entry; - - /* 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. - */ -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 { - u64 start = biosmap->addr; - u64 size = biosmap->size; - u64 end = start + size; - u32 type = biosmap->type; - - /* Overflow in 64 bits? Ignore the memory map. */ - if (start > end) - return -1; - - 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; -} - -static void early_panic(char *msg) -{ - early_printk(msg); - panic(msg); -} - -/* We're not void only for x86 32-bit compat */ -char * __init machine_specific_memory_setup(void) -{ -#ifndef CONFIG_XEN - 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(boot_params.e820_map, &boot_params.e820_entries); - if (copy_e820_map(boot_params.e820_map, boot_params.e820_entries) < 0) - early_panic("Cannot find a valid memory map"); -#else /* CONFIG_XEN */ - char *who = "Xen"; - int rc; - struct xen_memory_map memmap; - static struct e820entry __initdata 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); - - if (copy_e820_map(map, (char)memmap.nr_entries) < 0) - early_panic("Cannot find a valid memory map"); -#endif - printk(KERN_INFO "BIOS-provided physical RAM map:\n"); - e820_print_map(who); - - /* In case someone cares... */ - return who; -} - -static int __init parse_memopt(char *p) -{ - int i; - unsigned long current_end; - unsigned long end; - - if (!p) - return -EINVAL; - end_user_pfn = memparse(p, &p); - end_user_pfn >>= PAGE_SHIFT; - - end = end_user_pfn<> PAGE_SHIFT); - } - return *p == '\0' ? 0 : -EINVAL; -} -early_param("memmap", parse_memmap_opt); - -void __init finish_e820_parsing(void) -{ - if (userdef) { - char nr = e820.nr_map; - - if (sanitize_e820_map(e820.map, &nr) < 0) - early_panic("Invalid user supplied memory map"); - e820.nr_map = nr; - - printk(KERN_INFO "user-defined physical RAM map:\n"); - e820_print_map("user"); - } -} - -#ifndef CONFIG_XEN -void __init update_memory_range(u64 start, u64 size, unsigned old_type, - unsigned new_type) -{ - int i; - - BUG_ON(old_type == new_type); - - for (i = 0; i < e820.nr_map; i++) { - struct e820entry *ei = &e820.map[i]; - u64 final_start, final_end; - if (ei->type != old_type) - continue; - /* totally covered? */ - if (ei->addr >= start && ei->size <= size) { - ei->type = new_type; - continue; - } - /* partially covered */ - final_start = max(start, ei->addr); - final_end = min(start + size, ei->addr + ei->size); - if (final_start >= final_end) - continue; - add_memory_region(final_start, final_end - final_start, - new_type); - } -} - -void __init update_e820(void) -{ - u8 nr_map; - - nr_map = e820.nr_map; - if (sanitize_e820_map(e820.map, &nr_map)) - return; - e820.nr_map = nr_map; - printk(KERN_INFO "modified physical RAM map:\n"); - e820_print_map("modified"); -} -#endif - -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); -} - -int __init arch_get_ram_range(int slot, u64 *addr, u64 *size) -{ - int i; - - if (slot < 0 || slot >= e820.nr_map) - return -1; - for (i = slot; i < e820.nr_map; i++) { - if (e820.map[i].type != E820_RAM) - continue; - break; - } - if (i == e820.nr_map || e820.map[i].addr > (max_pfn << PAGE_SHIFT)) - return -1; - *addr = e820.map[i].addr; - *size = min_t(u64, e820.map[i].size + e820.map[i].addr, - max_pfn << PAGE_SHIFT) - *addr; - return i + 1; -} --- head-2011-03-11.orig/arch/x86/kernel/early_printk-xen.c 2011-01-31 18:07:35.000000000 +0100 +++ head-2011-03-11/arch/x86/kernel/early_printk-xen.c 2011-02-01 14:38:38.000000000 +0100 @@ -225,7 +225,7 @@ static struct console simnow_console = { static struct console *early_console = &early_vga_console; static int early_console_initialized; -void early_printk(const char *fmt, ...) +asmlinkage void early_printk(const char *fmt, ...) { char buf[512]; int n; --- head-2011-03-11.orig/arch/x86/kernel/entry_32-xen.S 2011-01-31 18:07:35.000000000 +0100 +++ head-2011-03-11/arch/x86/kernel/entry_32-xen.S 2011-02-01 14:38:38.000000000 +0100 @@ -51,15 +51,26 @@ #include #include #include -#include "irq_vectors.h" +#include +#include #include +/* Avoid __ASSEMBLER__'ifying just for this. */ +#include +#define AUDIT_ARCH_I386 (EM_386|__AUDIT_ARCH_LE) +#define __AUDIT_ARCH_LE 0x40000000 + +#ifndef CONFIG_AUDITSYSCALL +#define sysenter_audit syscall_trace_entry +#define sysexit_audit syscall_exit_work +#endif + /* * We use macros for low-level operations which need to be overridden * for paravirtualization. The following will never clobber any registers: * INTERRUPT_RETURN (aka. "iret") * GET_CR0_INTO_EAX (aka. "movl %cr0, %eax") - * ENABLE_INTERRUPTS_SYSCALL_RET (aka "sti; sysexit"). + * ENABLE_INTERRUPTS_SYSEXIT (aka "sti; sysexit"). * * For DISABLE_INTERRUPTS/ENABLE_INTERRUPTS (aka "cli"/"sti"), you must * specify what registers can be overwritten (CLBR_NONE, CLBR_EAX/EDX/ECX/ANY). @@ -277,11 +288,6 @@ END(resume_kernel) #endif CFI_ENDPROC - .macro test_tif ti_reg # 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(\ti_reg) - .endm - /* SYSENTER_RETURN points to after the "sysenter" instruction in the vsyscall page. See vsyscall-sysentry.S, which defines the symbol. */ @@ -338,8 +344,9 @@ sysenter_past_esp: .previous GET_THREAD_INFO(%ebp) - test_tif %ebp - jnz syscall_trace_entry + testw $_TIF_WORK_SYSCALL_ENTRY,TI_flags(%ebp) + jnz sysenter_audit +sysenter_do_call: cmpl $(nr_syscalls), %eax jae syscall_badsys call *sys_call_table(,%eax,4) @@ -349,14 +356,54 @@ sysenter_past_esp: TRACE_IRQS_OFF movl TI_flags(%ebp), %ecx testw $_TIF_ALLWORK_MASK, %cx - jne syscall_exit_work + jne sysexit_audit +sysenter_exit: /* if something modifies registers it must also disable sysexit */ movl PT_EIP(%esp), %edx movl PT_OLDESP(%esp), %ecx xorl %ebp,%ebp TRACE_IRQS_ON 1: mov PT_FS(%esp), %fs - ENABLE_INTERRUPTS_SYSCALL_RET + ENABLE_INTERRUPTS_SYSEXIT + +#ifdef CONFIG_AUDITSYSCALL +sysenter_audit: + testw $(_TIF_WORK_SYSCALL_ENTRY & ~_TIF_SYSCALL_AUDIT),TI_flags(%ebp) + jnz syscall_trace_entry + addl $4,%esp + CFI_ADJUST_CFA_OFFSET -4 + /* %esi already in 8(%esp) 6th arg: 4th syscall arg */ + /* %edx already in 4(%esp) 5th arg: 3rd syscall arg */ + /* %ecx already in 0(%esp) 4th arg: 2nd syscall arg */ + movl %ebx,%ecx /* 3rd arg: 1st syscall arg */ + movl %eax,%edx /* 2nd arg: syscall number */ + movl $AUDIT_ARCH_I386,%eax /* 1st arg: audit arch */ + call audit_syscall_entry + pushl %ebx + CFI_ADJUST_CFA_OFFSET 4 + movl PT_EAX(%esp),%eax /* reload syscall number */ + jmp sysenter_do_call + +sysexit_audit: + testw $(_TIF_ALLWORK_MASK & ~_TIF_SYSCALL_AUDIT), %cx + jne syscall_exit_work + TRACE_IRQS_ON + ENABLE_INTERRUPTS(CLBR_ANY) + movl %eax,%edx /* second arg, syscall return value */ + cmpl $0,%eax /* is it < 0? */ + setl %al /* 1 if so, 0 if not */ + movzbl %al,%eax /* zero-extend that */ + inc %eax /* first arg, 0->1(AUDITSC_SUCCESS), 1->2(AUDITSC_FAILURE) */ + call audit_syscall_exit + DISABLE_INTERRUPTS(CLBR_ANY) + TRACE_IRQS_OFF + movl TI_flags(%ebp), %ecx + testw $(_TIF_ALLWORK_MASK & ~_TIF_SYSCALL_AUDIT), %cx + jne syscall_exit_work + movl PT_EAX(%esp),%eax /* reload syscall return value */ + jmp sysenter_exit +#endif + CFI_ENDPROC .pushsection .fixup,"ax" 2: movl $0,PT_FS(%esp) @@ -400,7 +447,7 @@ ENTRY(system_call) CFI_ADJUST_CFA_OFFSET 4 SAVE_ALL GET_THREAD_INFO(%ebp) - test_tif %ebp + testw $_TIF_WORK_SYSCALL_ENTRY,TI_flags(%ebp) jnz syscall_trace_entry cmpl $(nr_syscalls), %eax jae syscall_badsys @@ -413,10 +460,6 @@ syscall_exit: # setting need_resched or sigpending # between sampling and the iret TRACE_IRQS_OFF - testl $X86_EFLAGS_TF,PT_EFLAGS(%esp) # If tracing set singlestep flag on exit - jz no_singlestep - orl $_TIF_SINGLESTEP,TI_flags(%ebp) -no_singlestep: movl TI_flags(%ebp), %ecx testw $_TIF_ALLWORK_MASK, %cx # current->work jne syscall_exit_work @@ -588,12 +631,8 @@ END(work_pending) syscall_trace_entry: movl $-ENOSYS,PT_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 PT_ORIG_EAX(%esp), %eax + call syscall_trace_enter + /* What it returned is what we'll actually use. */ cmpl $(nr_syscalls), %eax jnae syscall_call jmp syscall_exit @@ -602,14 +641,13 @@ END(syscall_trace_entry) # perform syscall exit tracing ALIGN syscall_exit_work: - testb $(_TIF_SYSCALL_TRACE|_TIF_SYSCALL_AUDIT|_TIF_SINGLESTEP), %cl + testb $_TIF_WORK_SYSCALL_EXIT, %cl jz work_pending TRACE_IRQS_ON - ENABLE_INTERRUPTS(CLBR_ANY) # could let do_syscall_trace() call + ENABLE_INTERRUPTS(CLBR_ANY) # could let syscall_trace_leave() call # schedule() instead movl %esp, %eax - movl $1, %edx - call do_syscall_trace + call syscall_trace_leave jmp resume_userspace END(syscall_exit_work) CFI_ENDPROC @@ -1113,10 +1151,10 @@ ENTRY(native_iret) .previous END(native_iret) -ENTRY(native_irq_enable_syscall_ret) +ENTRY(native_irq_enable_sysexit) sti sysexit -END(native_irq_enable_syscall_ret) +END(native_irq_enable_sysexit) #endif KPROBE_ENTRY(int3) @@ -1265,6 +1303,77 @@ ENTRY(kernel_thread_helper) CFI_ENDPROC ENDPROC(kernel_thread_helper) +#ifdef CONFIG_FTRACE +#ifdef CONFIG_DYNAMIC_FTRACE + +ENTRY(mcount) + pushl %eax + pushl %ecx + pushl %edx + movl 0xc(%esp), %eax + subl $MCOUNT_INSN_SIZE, %eax + +.globl mcount_call +mcount_call: + call ftrace_stub + + popl %edx + popl %ecx + popl %eax + + ret +END(mcount) + +ENTRY(ftrace_caller) + pushl %eax + pushl %ecx + pushl %edx + movl 0xc(%esp), %eax + movl 0x4(%ebp), %edx + subl $MCOUNT_INSN_SIZE, %eax + +.globl ftrace_call +ftrace_call: + call ftrace_stub + + popl %edx + popl %ecx + popl %eax + +.globl ftrace_stub +ftrace_stub: + ret +END(ftrace_caller) + +#else /* ! CONFIG_DYNAMIC_FTRACE */ + +ENTRY(mcount) + cmpl $ftrace_stub, ftrace_trace_function + jnz trace +.globl ftrace_stub +ftrace_stub: + ret + + /* taken from glibc */ +trace: + pushl %eax + pushl %ecx + pushl %edx + movl 0xc(%esp), %eax + movl 0x4(%ebp), %edx + subl $MCOUNT_INSN_SIZE, %eax + + call *ftrace_trace_function + + popl %edx + popl %ecx + popl %eax + + jmp ftrace_stub +END(mcount) +#endif /* CONFIG_DYNAMIC_FTRACE */ +#endif /* CONFIG_FTRACE */ + #include # pv syscall call handler stub @@ -1290,7 +1399,7 @@ ENTRY(ia32pv_cstar_target) .previous SAVE_ALL GET_THREAD_INFO(%ebp) - test_tif %ebp + testw $_TIF_WORK_SYSCALL_ENTRY,TI_flags(%ebp) jnz cstar_trace_entry cmpl $nr_syscalls,%eax jae cstar_badsys @@ -1324,29 +1433,21 @@ cstar_trace_entry: btl %eax,cstar_special jc .Lcstar_trace_special 1: movl %esp,%eax - xorl %edx,%edx LOCK_PREFIX orl $_TIF_CSTAR,TI_flags(%ebp) - call do_syscall_trace + call syscall_trace_enter LOCK_PREFIX andl $~_TIF_CSTAR,TI_flags(%ebp) - testl %eax,%eax - jne .Lcstar_resume # ret != 0 -> running under PTRACE_SYSEMU, - # so must skip actual syscall - movl PT_ORIG_EAX(%esp),%eax + /* What it returned is what we'll actually use. */ cmpl $nr_syscalls,%eax jb .Lcstar_call jmp .Lcstar_exit .Lcstar_trace_special: movl PT_ECX(%esp),%ecx movl %esp,%eax - xorl %edx,%edx movl %ecx,PT_EBP(%esp) # put user EBP back in place - call do_syscall_trace - testl %eax,%eax - jne resume_userspace # ret != 0 -> running under PTRACE_SYSEMU, - # so must skip actual syscall - movl PT_ORIG_EAX(%esp),%eax + call syscall_trace_enter + /* What it returned is what we'll actually use. */ cmpl $nr_syscalls,%eax jb syscall_call jmp syscall_exit --- head-2011-03-11.orig/arch/x86/kernel/entry_64.S 2011-03-15 16:45:55.000000000 +0100 +++ head-2011-03-11/arch/x86/kernel/entry_64.S 2011-02-16 16:02:30.000000000 +0100 @@ -1253,7 +1253,7 @@ ENTRY(arch_unwind_init_running) END(arch_unwind_init_running) #endif -#ifdef CONFIG_XEN +#ifdef CONFIG_PARAVIRT_XEN zeroentry xen_hypervisor_callback xen_do_hypervisor_callback /* @@ -1353,7 +1353,7 @@ END(xen_failsafe_callback) apicinterrupt XEN_HVM_EVTCHN_CALLBACK \ xen_hvm_callback_vector xen_evtchn_do_upcall -#endif /* CONFIG_XEN */ +#endif /* CONFIG_PARAVIRT_XEN */ /* * Some functions should be protected against kprobes --- head-2011-03-11.orig/arch/x86/kernel/entry_64-xen.S 2011-01-31 18:07:35.000000000 +0100 +++ head-2011-03-11/arch/x86/kernel/entry_64-xen.S 2011-02-01 14:38:38.000000000 +0100 @@ -53,12 +53,124 @@ #include #include #include +#include #include #include #include +/* Avoid __ASSEMBLER__'ifying just for this. */ +#include +#define AUDIT_ARCH_X86_64 (EM_X86_64|__AUDIT_ARCH_64BIT|__AUDIT_ARCH_LE) +#define __AUDIT_ARCH_64BIT 0x80000000 +#define __AUDIT_ARCH_LE 0x40000000 + .code64 +#ifdef CONFIG_FTRACE +#ifdef CONFIG_DYNAMIC_FTRACE +ENTRY(mcount) + + subq $0x38, %rsp + movq %rax, (%rsp) + movq %rcx, 8(%rsp) + movq %rdx, 16(%rsp) + movq %rsi, 24(%rsp) + movq %rdi, 32(%rsp) + movq %r8, 40(%rsp) + movq %r9, 48(%rsp) + + movq 0x38(%rsp), %rdi + subq $MCOUNT_INSN_SIZE, %rdi + +.globl mcount_call +mcount_call: + call ftrace_stub + + movq 48(%rsp), %r9 + movq 40(%rsp), %r8 + movq 32(%rsp), %rdi + movq 24(%rsp), %rsi + movq 16(%rsp), %rdx + movq 8(%rsp), %rcx + movq (%rsp), %rax + addq $0x38, %rsp + + retq +END(mcount) + +ENTRY(ftrace_caller) + + /* taken from glibc */ + subq $0x38, %rsp + movq %rax, (%rsp) + movq %rcx, 8(%rsp) + movq %rdx, 16(%rsp) + movq %rsi, 24(%rsp) + movq %rdi, 32(%rsp) + movq %r8, 40(%rsp) + movq %r9, 48(%rsp) + + movq 0x38(%rsp), %rdi + movq 8(%rbp), %rsi + subq $MCOUNT_INSN_SIZE, %rdi + +.globl ftrace_call +ftrace_call: + call ftrace_stub + + movq 48(%rsp), %r9 + movq 40(%rsp), %r8 + movq 32(%rsp), %rdi + movq 24(%rsp), %rsi + movq 16(%rsp), %rdx + movq 8(%rsp), %rcx + movq (%rsp), %rax + addq $0x38, %rsp + +.globl ftrace_stub +ftrace_stub: + retq +END(ftrace_caller) + +#else /* ! CONFIG_DYNAMIC_FTRACE */ +ENTRY(mcount) + cmpq $ftrace_stub, ftrace_trace_function + jnz trace +.globl ftrace_stub +ftrace_stub: + retq + +trace: + /* taken from glibc */ + subq $0x38, %rsp + movq %rax, (%rsp) + movq %rcx, 8(%rsp) + movq %rdx, 16(%rsp) + movq %rsi, 24(%rsp) + movq %rdi, 32(%rsp) + movq %r8, 40(%rsp) + movq %r9, 48(%rsp) + + movq 0x38(%rsp), %rdi + movq 8(%rbp), %rsi + subq $MCOUNT_INSN_SIZE, %rdi + + call *ftrace_trace_function + + movq 48(%rsp), %r9 + movq 40(%rsp), %r8 + movq 32(%rsp), %rdi + movq 24(%rsp), %rsi + movq 16(%rsp), %rdx + movq 8(%rsp), %rcx + movq (%rsp), %rax + addq $0x38, %rsp + + jmp ftrace_stub +END(mcount) +#endif /* CONFIG_DYNAMIC_FTRACE */ +#endif /* CONFIG_FTRACE */ + #ifndef CONFIG_PREEMPT #define retint_kernel retint_restore_args #endif @@ -95,7 +207,7 @@ NMI_MASK = 0x80000000 .macro FAKE_STACK_FRAME child_rip /* push in order ss, rsp, eflags, cs, rip */ xorl %eax, %eax - pushq %rax /* ss */ + pushq $__KERNEL_DS /* ss */ CFI_ADJUST_CFA_OFFSET 8 /*CFI_REL_OFFSET ss,0*/ pushq %rax /* rsp */ @@ -190,13 +302,13 @@ ENTRY(ret_from_fork) CFI_ADJUST_CFA_OFFSET -4 call schedule_tail GET_THREAD_INFO(%rcx) - testl $(_TIF_SYSCALL_TRACE|_TIF_SYSCALL_AUDIT),threadinfo_flags(%rcx) + testl $(_TIF_SYSCALL_TRACE|_TIF_SYSCALL_AUDIT),TI_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) + testl $_TIF_IA32,TI_flags(%rcx) jnz int_ret_from_sys_call RESTORE_TOP_OF_STACK %rdi,ARGOFFSET jmp ret_from_sys_call @@ -258,8 +370,9 @@ ENTRY(system_call) 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) + testl $_TIF_WORK_SYSCALL_ENTRY,TI_flags(%rcx) jnz tracesys +system_call_fastpath: cmpq $__NR_syscall_max,%rax ja badsys movq %r10,%rcx @@ -277,7 +390,7 @@ sysret_check: GET_THREAD_INFO(%rcx) DISABLE_INTERRUPTS(CLBR_NONE) TRACE_IRQS_OFF - movl threadinfo_flags(%rcx),%edx + movl TI_flags(%rcx),%edx andl %edi,%edx jnz sysret_careful CFI_REMEMBER_STATE @@ -308,16 +421,16 @@ sysret_careful: sysret_signal: TRACE_IRQS_ON ENABLE_INTERRUPTS(CLBR_NONE) - testl $_TIF_DO_NOTIFY_MASK,%edx - jz 1f - - /* Really a signal */ +#ifdef CONFIG_AUDITSYSCALL + bt $TIF_SYSCALL_AUDIT,%edx + jc sysret_audit +#endif /* 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 + movl $_TIF_WORK_MASK,%edi /* Use IRET because user could have changed frame. This works because ptregscall_common has called FIXUP_TOP_OF_STACK. */ DISABLE_INTERRUPTS(CLBR_NONE) @@ -328,14 +441,56 @@ badsys: movq $-ENOSYS,RAX-ARGOFFSET(%rsp) jmp ret_from_sys_call +#ifdef CONFIG_AUDITSYSCALL + /* + * Fast path for syscall audit without full syscall trace. + * We just call audit_syscall_entry() directly, and then + * jump back to the normal fast path. + */ +auditsys: + movq %r10,%r9 /* 6th arg: 4th syscall arg */ + movq %rdx,%r8 /* 5th arg: 3rd syscall arg */ + movq %rsi,%rcx /* 4th arg: 2nd syscall arg */ + movq %rdi,%rdx /* 3rd arg: 1st syscall arg */ + movq %rax,%rsi /* 2nd arg: syscall number */ + movl $AUDIT_ARCH_X86_64,%edi /* 1st arg: audit arch */ + call audit_syscall_entry + LOAD_ARGS 0 /* reload call-clobbered registers */ + jmp system_call_fastpath + + /* + * Return fast path for syscall audit. Call audit_syscall_exit() + * directly and then jump back to the fast path with TIF_SYSCALL_AUDIT + * masked off. + */ +sysret_audit: + movq %rax,%rsi /* second arg, syscall return value */ + cmpq $0,%rax /* is it < 0? */ + setl %al /* 1 if so, 0 if not */ + movzbl %al,%edi /* zero-extend that into %edi */ + inc %edi /* first arg, 0->1(AUDITSC_SUCCESS), 1->2(AUDITSC_FAILURE) */ + call audit_syscall_exit + movl $(_TIF_ALLWORK_MASK & ~_TIF_SYSCALL_AUDIT),%edi + jmp sysret_check +#endif /* CONFIG_AUDITSYSCALL */ + /* Do syscall tracing */ tracesys: +#ifdef CONFIG_AUDITSYSCALL + testl $(_TIF_WORK_SYSCALL_ENTRY & ~_TIF_SYSCALL_AUDIT),TI_flags(%rcx) + jz auditsys +#endif SAVE_REST movq $-ENOSYS,RAX(%rsp) /* ptrace can change this for a bad syscall */ FIXUP_TOP_OF_STACK %rdi movq %rsp,%rdi call syscall_trace_enter - LOAD_ARGS ARGOFFSET /* reload args from stack in case ptrace changed it */ + /* + * Reload arg registers from stack in case ptrace changed them. + * We don't reload %rax because syscall_trace_enter() returned + * the value it wants us to use in the table lookup. + */ + LOAD_ARGS ARGOFFSET, 1 RESTORE_REST cmpq $__NR_syscall_max,%rax ja int_ret_from_sys_call /* RAX(%rsp) set to -ENOSYS above */ @@ -349,6 +504,7 @@ tracesys: * Has correct top of stack, but partial stack frame. */ .globl int_ret_from_sys_call + .globl int_with_check int_ret_from_sys_call: DISABLE_INTERRUPTS(CLBR_NONE) TRACE_IRQS_OFF @@ -363,10 +519,10 @@ int_ret_from_sys_call: int_with_check: LOCKDEP_SYS_EXIT_IRQ GET_THREAD_INFO(%rcx) - movl threadinfo_flags(%rcx),%edx + movl TI_flags(%rcx),%edx andl %edi,%edx jnz int_careful - andl $~TS_COMPAT,threadinfo_status(%rcx) + andl $~TS_COMPAT,TI_status(%rcx) jmp retint_restore_args /* Either reschedule or signal or syscall exit tracking needed. */ @@ -392,7 +548,7 @@ int_very_careful: ENABLE_INTERRUPTS(CLBR_NONE) SAVE_REST /* Check for syscall exit trace */ - testl $(_TIF_SYSCALL_TRACE|_TIF_SYSCALL_AUDIT|_TIF_SINGLESTEP),%edx + testl $_TIF_WORK_SYSCALL_EXIT,%edx jz int_signal pushq %rdi CFI_ADJUST_CFA_OFFSET 8 @@ -400,7 +556,7 @@ int_very_careful: call syscall_trace_leave popq %rdi CFI_ADJUST_CFA_OFFSET -8 - andl $~(_TIF_SYSCALL_TRACE|_TIF_SYSCALL_AUDIT|_TIF_SINGLESTEP),%edi + andl $~(_TIF_WORK_SYSCALL_EXIT|_TIF_SYSCALL_EMU),%edi jmp int_restore_rest int_signal: @@ -409,7 +565,7 @@ int_signal: movq %rsp,%rdi # &ptregs -> arg1 xorl %esi,%esi # oldset -> arg2 call do_notify_resume -1: movl $_TIF_NEED_RESCHED,%edi +1: movl $_TIF_WORK_MASK,%edi int_restore_rest: RESTORE_REST DISABLE_INTERRUPTS(CLBR_NONE) @@ -436,7 +592,6 @@ END(\label) 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 @@ -510,10 +665,12 @@ END(stub_rt_sigreturn) * */ -retint_check: +retint_with_reschedule: CFI_DEFAULT_STACK adj=1 + movl $_TIF_WORK_MASK,%edi +retint_check: LOCKDEP_SYS_EXIT_IRQ - movl threadinfo_flags(%rcx),%edx + movl TI_flags(%rcx),%edx andl %edi,%edx CFI_REMEMBER_STATE jnz retint_careful @@ -558,17 +715,16 @@ retint_signal: RESTORE_REST DISABLE_INTERRUPTS(CLBR_NONE) TRACE_IRQS_OFF - movl $_TIF_NEED_RESCHED,%edi GET_THREAD_INFO(%rcx) - jmp retint_check + jmp retint_with_reschedule #ifdef CONFIG_PREEMPT /* Returning to kernel space. Check if we need preemption */ /* rcx: threadinfo. interrupts off. */ ENTRY(retint_kernel) - cmpl $0,threadinfo_preempt_count(%rcx) + cmpl $0,TI_preempt_count(%rcx) jnz retint_restore_args - bt $TIF_NEED_RESCHED,threadinfo_flags(%rcx) + bt $TIF_NEED_RESCHED,TI_flags(%rcx) jnc retint_restore_args bt $9,EFLAGS-ARGOFFSET(%rsp) /* interrupts off? */ jnc retint_restore_args @@ -623,6 +779,9 @@ END(invalidate_interrupt\num) ENTRY(call_function_interrupt) apicinterrupt CALL_FUNCTION_VECTOR,smp_call_function_interrupt END(call_function_interrupt) +ENTRY(call_function_single_interrupt) + apicinterrupt CALL_FUNCTION_SINGLE_VECTOR,smp_call_function_single_interrupt +END(call_function_single_interrupt) ENTRY(irq_move_cleanup_interrupt) apicinterrupt IRQ_MOVE_CLEANUP_VECTOR,smp_irq_move_cleanup_interrupt END(irq_move_cleanup_interrupt) @@ -632,6 +791,10 @@ ENTRY(apic_timer_interrupt) apicinterrupt LOCAL_TIMER_VECTOR,smp_apic_timer_interrupt END(apic_timer_interrupt) +ENTRY(uv_bau_message_intr1) + apicinterrupt 220,uv_bau_message_interrupt +END(uv_bau_message_intr1) + ENTRY(error_interrupt) apicinterrupt ERROR_APIC_VECTOR,smp_error_interrupt END(error_interrupt) @@ -745,7 +908,7 @@ paranoid_restore\trace: jmp irq_return paranoid_userspace\trace: GET_THREAD_INFO(%rcx) - movl threadinfo_flags(%rcx),%ebx + movl TI_flags(%rcx),%ebx andl $_TIF_WORK_MASK,%ebx jz paranoid_swapgs\trace movq %rsp,%rdi /* &pt_regs */ @@ -842,7 +1005,7 @@ error_exit: testb $3,CS-ARGOFFSET(%rsp) jz retint_kernel LOCKDEP_SYS_EXIT_IRQ - movl threadinfo_flags(%rcx),%edx + movl TI_flags(%rcx),%edx movl $_TIF_WORK_MASK,%edi andl %edi,%edx jnz retint_careful @@ -864,11 +1027,11 @@ error_kernelspace: 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 irq_return(%rip),%rbp - cmpq %rbp,RIP(%rsp) + leaq irq_return(%rip),%rcx + cmpq %rcx,RIP(%rsp) je error_swapgs - movl %ebp,%ebp /* zero extend */ - cmpq %rbp,RIP(%rsp) + movl %ecx,%ecx /* zero extend */ + cmpq %rcx,RIP(%rsp) je error_swapgs cmpq $gs_change,RIP(%rsp) je error_swapgs @@ -1114,6 +1277,7 @@ END(device_not_available) /* runs on exception stack */ KPROBE_ENTRY(debug) /* INTR_FRAME + PARAVIRT_ADJUST_EXCEPTION_FRAME pushq $0 CFI_ADJUST_CFA_OFFSET 8 */ zeroentry do_debug @@ -1141,6 +1305,7 @@ END(do_nmi_callback) KPROBE_ENTRY(int3) /* INTR_FRAME + PARAVIRT_ADJUST_EXCEPTION_FRAME pushq $0 CFI_ADJUST_CFA_OFFSET 8 */ zeroentry do_int3 @@ -1164,14 +1329,11 @@ 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 + PARAVIRT_ADJUST_EXCEPTION_FRAME paranoidentry do_double_fault jmp paranoid_exit1 CFI_ENDPROC @@ -1189,6 +1351,7 @@ END(segment_not_present) /* runs on exception stack */ ENTRY(stack_segment) /* XCPT_FRAME + PARAVIRT_ADJUST_EXCEPTION_FRAME paranoidentry do_stack_segment */ errorentry do_stack_segment /* jmp paranoid_exit1 --- head-2011-03-11.orig/arch/x86/kernel/fixup.c 2011-01-31 18:01:51.000000000 +0100 +++ head-2011-03-11/arch/x86/kernel/fixup.c 2011-02-01 14:38:38.000000000 +0100 @@ -33,6 +33,7 @@ #include #include #include +#include #define DP(_f, _args...) pr_alert(" " _f "\n" , ## _args ) --- /dev/null 1970-01-01 00:00:00.000000000 +0000 +++ head-2011-03-11/arch/x86/kernel/head-xen.c 2011-02-01 14:38:38.000000000 +0100 @@ -0,0 +1,57 @@ +#include +#include + +#include +#include + +#define BIOS_LOWMEM_KILOBYTES 0x413 + +/* + * The BIOS places the EBDA/XBDA at the top of conventional + * memory, and usually decreases the reported amount of + * conventional memory (int 0x12) too. This also contains a + * workaround for Dell systems that neglect to reserve EBDA. + * The same workaround also avoids a problem with the AMD768MPX + * 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. + */ +void __init reserve_ebda_region(void) +{ +#ifndef CONFIG_XEN + unsigned int lowmem, ebda_addr; + + /* To determine the position of the EBDA and the */ + /* end of conventional memory, we need to look at */ + /* the BIOS data area. In a paravirtual environment */ + /* that area is absent. We'll just have to assume */ + /* that the paravirt case can handle memory setup */ + /* correctly, without our help. */ + if (paravirt_enabled()) + return; + + /* end of low (conventional) memory */ + lowmem = *(unsigned short *)__va(BIOS_LOWMEM_KILOBYTES); + lowmem <<= 10; + + /* start of EBDA area */ + ebda_addr = get_bios_ebda(); + + /* Fixup: bios puts an EBDA in the top 64K segment */ + /* of conventional memory, but does not adjust lowmem. */ + if ((lowmem - ebda_addr) <= 0x10000) + lowmem = ebda_addr; + + /* Fixup: bios does not report an EBDA at all. */ + /* Some old Dells seem to need 4k anyhow (bugzilla 2990) */ + if ((ebda_addr == 0) && (lowmem >= 0x9f000)) + lowmem = 0x9f000; + + /* Paranoia: should never happen, but... */ + if ((lowmem == 0) || (lowmem >= 0x100000)) + lowmem = 0x9f000; + + /* reserve all memory between lowmem and the 1MB mark */ + reserve_early_overlap_ok(lowmem, 0x100000, "BIOS reserved"); +#endif +} --- /dev/null 1970-01-01 00:00:00.000000000 +0000 +++ head-2011-03-11/arch/x86/kernel/head32-xen.c 2011-02-01 14:38:38.000000000 +0100 @@ -0,0 +1,57 @@ +/* + * linux/arch/i386/kernel/head32.c -- prepare to run common code + * + * Copyright (C) 2000 Andrea Arcangeli SuSE + * Copyright (C) 2007 Eric Biederman + */ + +#include +#include + +#include +#include +#include +#include + +void __init i386_start_kernel(void) +{ + reserve_early(__pa_symbol(&_text), __pa_symbol(&_end), "TEXT DATA BSS"); + +#ifndef CONFIG_XEN +#ifdef CONFIG_BLK_DEV_INITRD + /* Reserve INITRD */ + if (boot_params.hdr.type_of_loader && boot_params.hdr.ramdisk_image) { + u64 ramdisk_image = boot_params.hdr.ramdisk_image; + u64 ramdisk_size = boot_params.hdr.ramdisk_size; + u64 ramdisk_end = ramdisk_image + ramdisk_size; + reserve_early(ramdisk_image, ramdisk_end, "RAMDISK"); + } +#endif + reserve_early(init_pg_tables_start, init_pg_tables_end, + "INIT_PG_TABLE"); +#else + reserve_early(ALIGN(__pa_symbol(&_end), PAGE_SIZE), + __pa(xen_start_info->pt_base) + + (xen_start_info->nr_pt_frames << PAGE_SHIFT), + "Xen provided"); + + { + int max_cmdline; + + if ((max_cmdline = MAX_GUEST_CMDLINE) > COMMAND_LINE_SIZE) + max_cmdline = COMMAND_LINE_SIZE; + memcpy(boot_command_line, xen_start_info->cmd_line, max_cmdline); + boot_command_line[max_cmdline-1] = '\0'; + } +#endif + + reserve_ebda_region(); + + /* + * At this point everything still needed from the boot loader + * or BIOS or kernel text should be early reserved or marked not + * RAM in e820. All other memory is free game. + */ + + start_kernel(); +} --- head-2011-03-11.orig/arch/x86/kernel/head64-xen.c 2011-01-31 18:07:35.000000000 +0100 +++ head-2011-03-11/arch/x86/kernel/head64-xen.c 2011-02-01 14:38:38.000000000 +0100 @@ -32,7 +32,26 @@ #include #include -unsigned long start_pfn; +/* boot cpu pda */ +static struct x8664_pda _boot_cpu_pda __read_mostly; + +#ifdef CONFIG_SMP +/* + * We install an empty cpu_pda pointer table to indicate to early users + * (numa_set_node) that the cpu_pda pointer table for cpus other than + * the boot cpu is not yet setup. + */ +static struct x8664_pda *__cpu_pda[NR_CPUS] __initdata; +#else +static struct x8664_pda *__cpu_pda[NR_CPUS] __read_mostly; +#endif + +void __init x86_64_init_pda(void) +{ + _cpu_pda = __cpu_pda; + cpu_pda(0) = &_boot_cpu_pda; + pda_init(0); +} #ifndef CONFIG_XEN static void __init zap_identity_mappings(void) @@ -77,83 +96,10 @@ EXPORT_SYMBOL(machine_to_phys_mapping); unsigned int machine_to_phys_order; EXPORT_SYMBOL(machine_to_phys_order); -#define BIOS_LOWMEM_KILOBYTES 0x413 - -/* - * The BIOS places the EBDA/XBDA at the top of conventional - * memory, and usually decreases the reported amount of - * conventional memory (int 0x12) too. This also contains a - * workaround for Dell systems that neglect to reserve EBDA. - * The same workaround also avoids a problem with the AMD768MPX - * 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. - */ -static void __init reserve_ebda_region(void) -{ -#ifndef CONFIG_XEN - unsigned int lowmem, ebda_addr; - - /* To determine the position of the EBDA and the */ - /* end of conventional memory, we need to look at */ - /* the BIOS data area. In a paravirtual environment */ - /* that area is absent. We'll just have to assume */ - /* that the paravirt case can handle memory setup */ - /* correctly, without our help. */ - if (paravirt_enabled()) - return; - - /* end of low (conventional) memory */ - lowmem = *(unsigned short *)__va(BIOS_LOWMEM_KILOBYTES); - lowmem <<= 10; - - /* start of EBDA area */ - ebda_addr = get_bios_ebda(); - - /* Fixup: bios puts an EBDA in the top 64K segment */ - /* of conventional memory, but does not adjust lowmem. */ - if ((lowmem - ebda_addr) <= 0x10000) - lowmem = ebda_addr; - - /* Fixup: bios does not report an EBDA at all. */ - /* Some old Dells seem to need 4k anyhow (bugzilla 2990) */ - if ((ebda_addr == 0) && (lowmem >= 0x9f000)) - lowmem = 0x9f000; - - /* Paranoia: should never happen, but... */ - if ((lowmem == 0) || (lowmem >= 0x100000)) - lowmem = 0x9f000; - - /* reserve all memory between lowmem and the 1MB mark */ - reserve_early(lowmem, 0x100000, "BIOS reserved"); -#endif -} - -static void __init reserve_setup_data(void) -{ -#ifndef CONFIG_XEN - struct setup_data *data; - unsigned long pa_data; - char buf[32]; - - if (boot_params.hdr.version < 0x0209) - return; - pa_data = boot_params.hdr.setup_data; - while (pa_data) { - data = early_ioremap(pa_data, sizeof(*data)); - sprintf(buf, "setup data %x", data->type); - reserve_early(pa_data, pa_data+sizeof(*data)+data->len, buf); - pa_data = data->next; - early_iounmap(data, sizeof(*data)); - } -#endif -} - void __init x86_64_start_kernel(char * real_mode_data) { struct xen_machphys_mapping mapping; unsigned long machine_to_phys_nr_ents; - int i; /* * Build-time sanity checks on the kernel image and module @@ -167,6 +113,7 @@ void __init x86_64_start_kernel(char * r BUILD_BUG_ON(!(MODULES_VADDR > __START_KERNEL)); BUILD_BUG_ON(!(((MODULES_END - 1) & PGDIR_MASK) == (__START_KERNEL & PGDIR_MASK))); + BUILD_BUG_ON(__fix_to_virt(__end_of_fixed_addresses) <= MODULES_END); xen_setup_features(); @@ -174,8 +121,6 @@ void __init x86_64_start_kernel(char * r 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; @@ -208,19 +153,23 @@ void __init x86_64_start_kernel(char * r early_printk("Kernel alive\n"); - for (i = 0; i < NR_CPUS; i++) - cpu_pda(i) = &boot_cpu_pda[i]; + x86_64_init_pda(); - pda_init(0); + early_printk("Kernel really alive\n"); + + x86_64_start_reservations(real_mode_data); +} + +void __init x86_64_start_reservations(char *real_mode_data) +{ copy_bootdata(__va(real_mode_data)); reserve_early(__pa_symbol(&_text), __pa_symbol(&_end), "TEXT DATA BSS"); reserve_early(round_up(__pa_symbol(&_end), PAGE_SIZE), - start_pfn << PAGE_SHIFT, "Xen provided"); - - reserve_ebda_region(); - reserve_setup_data(); + __pa(xen_start_info->pt_base) + + (xen_start_info->nr_pt_frames << PAGE_SHIFT), + "Xen provided"); /* * At this point everything still needed from the boot loader --- head-2011-03-11.orig/arch/x86/kernel/head_32-xen.S 2011-01-31 18:07:35.000000000 +0100 +++ head-2011-03-11/arch/x86/kernel/head_32-xen.S 2011-03-03 16:19:21.000000000 +0100 @@ -61,8 +61,6 @@ ENTRY(startup_32) movb %cl,X86_MASK movl %edx,X86_CAPABILITY - movb $1,X86_HARD_MATH - xorl %eax,%eax # Clear GS movl %eax,%gs --- head-2011-03-11.orig/arch/x86/kernel/head_64-xen.S 2011-01-31 17:49:31.000000000 +0100 +++ head-2011-03-11/arch/x86/kernel/head_64-xen.S 2011-02-01 14:38:38.000000000 +0100 @@ -92,53 +92,6 @@ NEXT_PAGE(hypercall_page) #undef NEXT_PAGE - .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 0x00cf9b000000ffff /* __KERNEL32_CS */ - .quad 0x00af9b000000ffff /* __KERNEL_CS */ - .quad 0x00cf93000000ffff /* __KERNEL_DS */ - .quad 0x00cffb000000ffff /* __USER32_CS */ - .quad 0x00cff3000000ffff /* __USER_DS, __USER32_DS */ - .quad 0x00affb000000ffff /* __USER_CS */ - .quad 0x0 /* unused */ - .quad 0,0 /* TSS */ - .quad 0,0 /* LDT */ - .quad 0,0,0 /* three TLS descriptors */ - .quad 0x0000f40000000000 /* node/CPU stored in limit */ -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) --- head-2011-03-11.orig/arch/x86/kernel/io_apic_32-xen.c 2011-01-31 18:07:35.000000000 +0100 +++ head-2011-03-11/arch/x86/kernel/io_apic_32-xen.c 2011-02-01 14:38:38.000000000 +0100 @@ -25,6 +25,7 @@ #include #include #include +#include #include #include #include @@ -75,7 +76,7 @@ static struct { int pin, apic; } ioapic_ static DEFINE_SPINLOCK(ioapic_lock); static DEFINE_SPINLOCK(vector_lock); -int timer_over_8254 __initdata = 1; +int timer_through_8259 __initdata; /* * Is the SiS APIC rmw bug present ? @@ -89,15 +90,21 @@ int sis_apic_bug = -1; int nr_ioapic_registers[MAX_IO_APICS]; /* I/O APIC entries */ -struct mpc_config_ioapic mp_ioapics[MAX_IO_APICS]; +struct mp_config_ioapic mp_ioapics[MAX_IO_APICS]; int nr_ioapics; /* MP IRQ source entries */ -struct mpc_config_intsrc mp_irqs[MAX_IRQ_SOURCES]; +struct mp_config_intsrc mp_irqs[MAX_IRQ_SOURCES]; /* # of MP IRQ source entries */ int mp_irq_entries; +#if defined (CONFIG_MCA) || defined (CONFIG_EISA) +int mp_bus_id_to_type[MAX_MP_BUSSES]; +#endif + +DECLARE_BITMAP(mp_bus_not_pci, MAX_MP_BUSSES); + static int disable_timer_pin_1 __initdata; /* @@ -128,7 +135,7 @@ struct io_apic { static __attribute_const__ struct io_apic __iomem *io_apic_base(int idx) { return (void __iomem *) __fix_to_virt(FIX_IO_APIC_BASE_0 + idx) - + (mp_ioapics[idx].mpc_apicaddr & ~PAGE_MASK); + + (mp_ioapics[idx].mp_apicaddr & ~PAGE_MASK); } #endif @@ -142,7 +149,7 @@ static inline unsigned int io_apic_read( struct physdev_apic apic_op; int ret; - apic_op.apic_physbase = mp_ioapics[apic].mpc_apicaddr; + apic_op.apic_physbase = mp_ioapics[apic].mp_apicaddr; apic_op.reg = reg; ret = HYPERVISOR_physdev_op(PHYSDEVOP_apic_read, &apic_op); if (ret) @@ -160,7 +167,7 @@ static inline void io_apic_write(unsigne #else struct physdev_apic apic_op; - apic_op.apic_physbase = mp_ioapics[apic].mpc_apicaddr; + apic_op.apic_physbase = mp_ioapics[apic].mp_apicaddr; apic_op.reg = reg; apic_op.value = value; WARN_ON(HYPERVISOR_physdev_op(PHYSDEVOP_apic_write, &apic_op)); @@ -288,7 +295,7 @@ static void __init replace_pin_at_irq(un } } -static void __modify_IO_APIC_irq (unsigned int irq, unsigned long enable, unsigned long disable) +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; @@ -308,30 +315,32 @@ static void __modify_IO_APIC_irq (unsign } /* mask = 1 */ -static void __mask_IO_APIC_irq (unsigned int irq) +static void __mask_IO_APIC_irq(unsigned int irq) { - __modify_IO_APIC_irq(irq, 0x00010000, 0); + __modify_IO_APIC_irq(irq, IO_APIC_REDIR_MASKED, 0); } /* mask = 0 */ -static void __unmask_IO_APIC_irq (unsigned int irq) +static void __unmask_IO_APIC_irq(unsigned int irq) { - __modify_IO_APIC_irq(irq, 0, 0x00010000); + __modify_IO_APIC_irq(irq, 0, IO_APIC_REDIR_MASKED); } /* mask = 1, trigger = 0 */ -static void __mask_and_edge_IO_APIC_irq (unsigned int irq) +static void __mask_and_edge_IO_APIC_irq(unsigned int irq) { - __modify_IO_APIC_irq(irq, 0x00010000, 0x00008000); + __modify_IO_APIC_irq(irq, IO_APIC_REDIR_MASKED, + IO_APIC_REDIR_LEVEL_TRIGGER); } /* mask = 0, trigger = 1 */ -static void __unmask_and_level_IO_APIC_irq (unsigned int irq) +static void __unmask_and_level_IO_APIC_irq(unsigned int irq) { - __modify_IO_APIC_irq(irq, 0x00008000, 0x00010000); + __modify_IO_APIC_irq(irq, IO_APIC_REDIR_LEVEL_TRIGGER, + IO_APIC_REDIR_MASKED); } -static void mask_IO_APIC_irq (unsigned int irq) +static void mask_IO_APIC_irq(unsigned int irq) { unsigned long flags; @@ -340,7 +349,7 @@ static void mask_IO_APIC_irq (unsigned i spin_unlock_irqrestore(&ioapic_lock, flags); } -static void unmask_IO_APIC_irq (unsigned int irq) +static void unmask_IO_APIC_irq(unsigned int irq) { unsigned long flags; @@ -352,7 +361,7 @@ static void unmask_IO_APIC_irq (unsigned static void clear_IO_APIC_pin(unsigned int apic, unsigned int pin) { struct IO_APIC_route_entry entry; - + /* Check delivery_mode to be sure we're not clearing an SMI pin */ entry = ioapic_read_entry(apic, pin); if (entry.delivery_mode == dest_SMI) @@ -364,7 +373,7 @@ static void clear_IO_APIC_pin(unsigned i ioapic_mask_entry(apic, pin); } -static void clear_IO_APIC (void) +static void clear_IO_APIC(void) { int apic, pin; @@ -381,7 +390,7 @@ static void set_ioapic_affinity_irq(unsi 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; @@ -410,7 +419,7 @@ static void set_ioapic_affinity_irq(unsi # include /* kstat */ # include /* kmalloc() */ # include - + #define IRQBALANCE_CHECK_ARCH -999 #define MAX_BALANCED_IRQ_INTERVAL (5*HZ) #define MIN_BALANCED_IRQ_INTERVAL (HZ/2) @@ -422,14 +431,14 @@ static int physical_balance __read_mostl 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 *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 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)) @@ -468,8 +477,8 @@ inside: if (cpu == -1) cpu = NR_CPUS-1; } - } while (!cpu_online(cpu) || !IRQ_ALLOWED(cpu,allowed_mask) || - (search_idle && !IDLE_ENOUGH(cpu,now))); + } while (!cpu_online(cpu) || !IRQ_ALLOWED(cpu, allowed_mask) || + (search_idle && !IDLE_ENOUGH(cpu, now))); return cpu; } @@ -479,15 +488,14 @@ static inline void balance_irq(int cpu, unsigned long now = jiffies; cpumask_t allowed_mask; unsigned int new_cpu; - + if (irqbalance_disabled) - return; + return; cpus_and(allowed_mask, cpu_online_map, balance_irq_affinity[irq]); new_cpu = move(cpu, allowed_mask, now, 1); - if (cpu != new_cpu) { + 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) @@ -499,14 +507,14 @@ static inline void rotate_irqs_among_cpu if (!irq_desc[j].action) continue; /* Is it a significant load ? */ - if (IRQ_DELTA(CPU_TO_PACKAGEINDEX(i),j) < + 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); + balanced_irq_interval - BALANCED_IRQ_LESS_DELTA); return; } @@ -535,22 +543,22 @@ static void do_irq_balance(void) /* Is this an active IRQ or balancing disabled ? */ if (!irq_desc[j].action || irq_balancing_disabled(j)) continue; - if ( package_index == i ) - IRQ_DELTA(package_index,j) = 0; + 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); + delta = value_now - LAST_CPU_IRQ(i, j); /* Update last_cpu_irq[][] for the next time */ - LAST_CPU_IRQ(i,j) = value_now; + 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; + IRQ_DELTA(package_index, j) += delta; /* Keep track of the higher numbered sibling as well */ if (i != package_index) @@ -576,7 +584,8 @@ static void do_irq_balance(void) max_cpu_irq = ULONG_MAX; tryanothercpu: - /* Look for heaviest loaded processor. + /* + * Look for heaviest loaded processor. * We may come back to get the next heaviest loaded processor. * Skip processors with trivial loads. */ @@ -585,7 +594,7 @@ tryanothercpu: for_each_online_cpu(i) { if (i != CPU_TO_PACKAGEINDEX(i)) continue; - if (max_cpu_irq <= CPU_IRQ(i)) + if (max_cpu_irq <= CPU_IRQ(i)) continue; if (tmp_cpu_irq < CPU_IRQ(i)) { tmp_cpu_irq = CPU_IRQ(i); @@ -594,8 +603,9 @@ tryanothercpu: } 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 + /* + * 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) { @@ -604,13 +614,14 @@ tryanothercpu: } 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; - - /* if imbalance is less than approx 10% of max load, then + + /* + * if imbalance is less than approx 10% of max load, then * observe diminishing returns action. - quit */ if (imbalance < (max_cpu_irq >> 3)) @@ -626,26 +637,25 @@ tryanotherirq: /* Is this an active IRQ? */ if (!irq_desc[j].action) continue; - if (imbalance <= IRQ_DELTA(max_loaded,j)) + 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); + if (move_this_load < IRQ_DELTA(max_loaded, j)) { + move_this_load = IRQ_DELTA(max_loaded, j); selected_irq = j; } } - if (selected_irq == -1) { + if (selected_irq == -1) goto tryanothercpu; - } imbalance = move_this_load; - + /* For physical_balance case, we accumulated 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. + * 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 @@ -674,11 +684,11 @@ tryanotherirq: /* mark for change destination */ set_pending_irq(selected_irq, cpumask_of_cpu(min_loaded)); - /* Since we made a change, come back sooner to + /* 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); + balanced_irq_interval - BALANCED_IRQ_LESS_DELTA); return; } goto tryanotherirq; @@ -689,7 +699,7 @@ not_worth_the_effort: * upward */ balanced_irq_interval = min((long)MAX_BALANCED_IRQ_INTERVAL, - balanced_irq_interval + BALANCED_IRQ_MORE_DELTA); + balanced_irq_interval + BALANCED_IRQ_MORE_DELTA); return; } @@ -728,13 +738,13 @@ static int __init balanced_irq_init(void cpumask_t tmp; cpus_shift_right(tmp, cpu_online_map, 2); - c = &boot_cpu_data; + 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; @@ -748,16 +758,14 @@ static int __init balanced_irq_init(void 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); + irq_cpu_data[i].irq_delta = kzalloc(sizeof(unsigned long) * NR_IRQS, GFP_KERNEL); + irq_cpu_data[i].last_irq = kzalloc(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 (!IS_ERR(kthread_run(balanced_irq, NULL, "kirqd"))) return 0; @@ -799,7 +807,7 @@ void send_IPI_self(int vector) /* * Send the IPI. The write to APIC_ICR fires this off. */ - apic_write_around(APIC_ICR, cfg); + apic_write(APIC_ICR, cfg); #endif } #endif /* !CONFIG_SMP */ @@ -853,10 +861,10 @@ static int find_irq_entry(int apic, int 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) + if (mp_irqs[i].mp_irqtype == type && + (mp_irqs[i].mp_dstapic == mp_ioapics[apic].mp_apicid || + mp_irqs[i].mp_dstapic == MP_APIC_ALL) && + mp_irqs[i].mp_dstirq == pin) return i; return -1; @@ -871,13 +879,13 @@ static int __init find_isa_irq_pin(int i int i; for (i = 0; i < mp_irq_entries; i++) { - int lbus = mp_irqs[i].mpc_srcbus; + int lbus = mp_irqs[i].mp_srcbus; if (test_bit(lbus, mp_bus_not_pci) && - (mp_irqs[i].mpc_irqtype == type) && - (mp_irqs[i].mpc_srcbusirq == irq)) + (mp_irqs[i].mp_irqtype == type) && + (mp_irqs[i].mp_srcbusirq == irq)) - return mp_irqs[i].mpc_dstirq; + return mp_irqs[i].mp_dstirq; } return -1; } @@ -887,17 +895,17 @@ static int __init find_isa_irq_apic(int int i; for (i = 0; i < mp_irq_entries; i++) { - int lbus = mp_irqs[i].mpc_srcbus; + int lbus = mp_irqs[i].mp_srcbus; if (test_bit(lbus, mp_bus_not_pci) && - (mp_irqs[i].mpc_irqtype == type) && - (mp_irqs[i].mpc_srcbusirq == irq)) + (mp_irqs[i].mp_irqtype == type) && + (mp_irqs[i].mp_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) + for (apic = 0; apic < nr_ioapics; apic++) { + if (mp_ioapics[apic].mp_apicid == mp_irqs[i].mp_dstapic) return apic; } } @@ -918,28 +926,28 @@ int IO_APIC_get_PCI_irq_vector(int bus, 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) { + if (test_bit(bus, mp_bus_not_pci)) { 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; + int lbus = mp_irqs[i].mp_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) + if (mp_ioapics[apic].mp_apicid == mp_irqs[i].mp_dstapic || + mp_irqs[i].mp_dstapic == MP_APIC_ALL) break; if (!test_bit(lbus, mp_bus_not_pci) && - !mp_irqs[i].mpc_irqtype && + !mp_irqs[i].mp_irqtype && (bus == lbus) && - (slot == ((mp_irqs[i].mpc_srcbusirq >> 2) & 0x1f))) { - int irq = pin_2_irq(i,apic,mp_irqs[i].mpc_dstirq); + (slot == ((mp_irqs[i].mp_srcbusirq >> 2) & 0x1f))) { + int irq = pin_2_irq(i, apic, mp_irqs[i].mp_dstirq); if (!(apic || IO_APIC_IRQ(irq))) continue; - if (pin == (mp_irqs[i].mpc_srcbusirq & 3)) + if (pin == (mp_irqs[i].mp_srcbusirq & 3)) return irq; /* * Use the first all-but-pin matching entry as a @@ -954,7 +962,7 @@ int IO_APIC_get_PCI_irq_vector(int bus, EXPORT_SYMBOL(IO_APIC_get_PCI_irq_vector); /* - * This function currently is only a helper for the i386 smp boot process where + * 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 */ @@ -1008,7 +1016,7 @@ static int EISA_ELCR(unsigned int irq) * 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_trigger(idx) (EISA_ELCR(mp_irqs[idx].mp_srcbusirq)) #define default_EISA_polarity(idx) default_ISA_polarity(idx) /* PCI interrupts are always polarity one level triggered, @@ -1025,118 +1033,115 @@ static int EISA_ELCR(unsigned int irq) static int MPBIOS_polarity(int idx) { - int bus = mp_irqs[idx].mpc_srcbus; + int bus = mp_irqs[idx].mp_srcbus; int polarity; /* * Determine IRQ line polarity (high active or low active): */ - switch (mp_irqs[idx].mpc_irqflag & 3) + switch (mp_irqs[idx].mp_irqflag & 3) { + case 0: /* conforms, ie. bus-type dependent polarity */ { - case 0: /* conforms, ie. bus-type dependent polarity */ - { - polarity = test_bit(bus, mp_bus_not_pci)? - default_ISA_polarity(idx): - default_PCI_polarity(idx); - 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; - } + polarity = test_bit(bus, mp_bus_not_pci)? + default_ISA_polarity(idx): + default_PCI_polarity(idx); + 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 bus = mp_irqs[idx].mp_srcbus; int trigger; /* * Determine IRQ trigger mode (edge or level sensitive): */ - switch ((mp_irqs[idx].mpc_irqflag>>2) & 3) + switch ((mp_irqs[idx].mp_irqflag>>2) & 3) { + case 0: /* conforms, ie. bus-type dependent */ { - case 0: /* conforms, ie. bus-type dependent */ - { - trigger = test_bit(bus, mp_bus_not_pci)? - default_ISA_trigger(idx): - default_PCI_trigger(idx); + trigger = test_bit(bus, mp_bus_not_pci)? + default_ISA_trigger(idx): + default_PCI_trigger(idx); #if defined(CONFIG_EISA) || defined(CONFIG_MCA) - switch (mp_bus_id_to_type[bus]) - { - case MP_BUS_ISA: /* ISA pin */ - { - /* set before the switch */ - break; - } - case MP_BUS_EISA: /* EISA pin */ - { - trigger = default_EISA_trigger(idx); - break; - } - case MP_BUS_PCI: /* PCI pin */ - { - /* set before the switch */ - break; - } - case MP_BUS_MCA: /* MCA pin */ - { - trigger = default_MCA_trigger(idx); - break; - } - default: - { - printk(KERN_WARNING "broken BIOS!!\n"); - trigger = 1; - break; - } - } -#endif + switch (mp_bus_id_to_type[bus]) { + case MP_BUS_ISA: /* ISA pin */ + { + /* set before the switch */ break; } - case 1: /* edge */ + case MP_BUS_EISA: /* EISA pin */ { - trigger = 0; + trigger = default_EISA_trigger(idx); break; } - case 2: /* reserved */ + case MP_BUS_PCI: /* PCI pin */ { - printk(KERN_WARNING "broken BIOS!!\n"); - trigger = 1; + /* set before the switch */ break; } - case 3: /* level */ + case MP_BUS_MCA: /* MCA pin */ { - trigger = 1; + trigger = default_MCA_trigger(idx); break; } - default: /* invalid */ + default: { printk(KERN_WARNING "broken BIOS!!\n"); - trigger = 0; + trigger = 1; break; } } +#endif + 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; } @@ -1153,16 +1158,16 @@ static inline int irq_trigger(int idx) static int pin_2_irq(int idx, int apic, int pin) { int irq, i; - int bus = mp_irqs[idx].mpc_srcbus; + int bus = mp_irqs[idx].mp_srcbus; /* * Debugging check, we are in big trouble if this message pops up! */ - if (mp_irqs[idx].mpc_dstirq != pin) + if (mp_irqs[idx].mp_dstirq != pin) printk(KERN_ERR "broken BIOS or MPTABLE parser, ayiee!!\n"); if (test_bit(bus, mp_bus_not_pci)) - irq = mp_irqs[idx].mpc_srcbusirq; + irq = mp_irqs[idx].mp_srcbusirq; else { /* * PCI IRQs are mapped in order @@ -1204,8 +1209,8 @@ static inline int IO_APIC_irq_trigger(in 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))) + idx = find_irq_entry(apic, pin, mp_INT); + if ((idx != -1) && (irq == pin_2_irq(idx, apic, pin))) return irq_trigger(idx); } } @@ -1291,25 +1296,25 @@ static void __init setup_IO_APIC_irqs(vo /* * add it to the IO-APIC irq-routing table: */ - memset(&entry,0,sizeof(entry)); + 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 = + entry.dest.logical.logical_dest = cpu_mask_to_apicid(TARGET_CPUS); - idx = find_irq_entry(apic,pin,mp_INT); + 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, + mp_ioapics[apic].mp_apicid, pin); first_notcon = 0; } else apic_printk(APIC_VERBOSE, ", %d-%d", - mp_ioapics[apic].mpc_apicid, pin); + mp_ioapics[apic].mp_apicid, pin); continue; } @@ -1343,7 +1348,7 @@ static void __init setup_IO_APIC_irqs(vo vector = assign_irq_vector(irq); entry.vector = vector; ioapic_register_intr(irq, vector, IOAPIC_AUTO); - + if (!apic && (irq < 16)) disable_8259A_irq(irq); } @@ -1355,27 +1360,23 @@ static void __init setup_IO_APIC_irqs(vo apic_printk(APIC_VERBOSE, " not connected.\n"); } +#ifndef CONFIG_XEN /* - * Set up the 8259A-master output pin: + * Set up the timer pin, possibly with the 8259A-master behind. */ -#ifndef CONFIG_XEN -static void __init setup_ExtINT_IRQ0_pin(unsigned int apic, unsigned int pin, int vector) +static void __init setup_timer_IRQ0_pin(unsigned int apic, unsigned int pin, + int vector) { struct IO_APIC_route_entry entry; - memset(&entry,0,sizeof(entry)); - - disable_8259A_irq(0); - - /* mask LVT0 */ - apic_write_around(APIC_LVT0, APIC_LVT_MASKED | APIC_DM_EXTINT); + memset(&entry, 0, sizeof(entry)); /* * 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.mask = 1; /* mask IRQ now */ entry.dest.logical.logical_dest = cpu_mask_to_apicid(TARGET_CPUS); entry.delivery_mode = INT_DELIVERY_MODE; entry.polarity = 0; @@ -1384,17 +1385,14 @@ static void __init setup_ExtINT_IRQ0_pin /* * The timer IRQ doesn't have to know that behind the - * scene we have a 8259A-master in AEOI mode ... + * scene we may have a 8259A-master in AEOI mode ... */ - irq_desc[0].chip = &ioapic_chip; - set_irq_handler(0, handle_edge_irq); + ioapic_register_intr(0, vector, IOAPIC_EDGE); /* * Add it to the IO-APIC irq-routing table: */ ioapic_write_entry(apic, pin, entry); - - enable_8259A_irq(0); } void __init print_IO_APIC(void) @@ -1409,10 +1407,10 @@ void __init print_IO_APIC(void) if (apic_verbosity == APIC_QUIET) return; - printk(KERN_DEBUG "number of MP IRQ sources: %d.\n", mp_irq_entries); + 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]); + mp_ioapics[i].mp_apicid, nr_ioapic_registers[i]); /* * We are a bit conservative about what we expect. We have to @@ -1431,7 +1429,7 @@ void __init print_IO_APIC(void) 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 "IO APIC #%d......\n", mp_ioapics[apic].mp_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); @@ -1512,7 +1510,7 @@ void __init print_IO_APIC(void) return; } -static void print_APIC_bitfield (int base) +static void print_APIC_bitfield(int base) { unsigned int v; int i, j; @@ -1533,7 +1531,7 @@ static void print_APIC_bitfield (int bas } } -void /*__init*/ print_local_APIC(void * dummy) +void /*__init*/ print_local_APIC(void *dummy) { unsigned int v, ver, maxlvt; @@ -1542,6 +1540,7 @@ void /*__init*/ print_local_APIC(void * 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(read_apic_id())); v = apic_read(APIC_LVR); @@ -1616,9 +1615,9 @@ void /*__init*/ print_local_APIC(void * printk("\n"); } -void print_all_local_APICs (void) +void print_all_local_APICs(void) { - on_each_cpu(print_local_APIC, NULL, 1, 1); + on_each_cpu(print_local_APIC, NULL, 1); } void /*__init*/ print_PIC(void) @@ -1639,11 +1638,11 @@ void /*__init*/ print_PIC(void) v = inb(0xa0) << 8 | inb(0x20); printk(KERN_DEBUG "... PIC IRR: %04x\n", v); - outb(0x0b,0xa0); - outb(0x0b,0x20); + outb(0x0b, 0xa0); + outb(0x0b, 0x20); v = inb(0xa0) << 8 | inb(0x20); - outb(0x0a,0xa0); - outb(0x0a,0x20); + outb(0x0a, 0xa0); + outb(0x0a, 0x20); spin_unlock_irqrestore(&i8259A_lock, flags); @@ -1652,6 +1651,8 @@ void /*__init*/ print_PIC(void) v = inb(0x4d1) << 8 | inb(0x4d0); printk(KERN_DEBUG "... PIC ELCR: %04x\n", v); } +#else +void __init print_IO_APIC(void) {} #endif /* !CONFIG_XEN */ static void __init enable_IO_APIC(void) @@ -1681,7 +1682,7 @@ static void __init enable_IO_APIC(void) nr_ioapic_registers[apic] = reg_01.bits.entries+1; } #ifndef CONFIG_XEN - for(apic = 0; apic < nr_ioapics; apic++) { + 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++) { @@ -1774,7 +1775,7 @@ void disable_IO_APIC(void) * by Matt Domsch Tue Dec 21 12:25:05 CST 1999 */ -#if !defined(CONFIG_XEN) && !defined(CONFIG_X86_NUMAQ) +#ifndef CONFIG_XEN static void __init setup_ioapic_ids_from_mpc(void) { union IO_APIC_reg_00 reg_00; @@ -1784,6 +1785,11 @@ static void __init setup_ioapic_ids_from unsigned char old_id; unsigned long flags; +#ifdef CONFIG_X86_NUMAQ + if (found_numaq) + return; +#endif + /* * Don't check I/O APIC IDs for xAPIC systems. They have * no meaning without the serial APIC bus. @@ -1806,15 +1812,15 @@ static void __init setup_ioapic_ids_from 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()) { + old_id = mp_ioapics[apic].mp_apicid; + + if (mp_ioapics[apic].mp_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); + apic, mp_ioapics[apic].mp_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; + mp_ioapics[apic].mp_apicid = reg_00.bits.ID; } /* @@ -1823,9 +1829,9 @@ static void __init setup_ioapic_ids_from * 'stuck on smp_invalidate_needed IPI wait' messages. */ if (check_apicid_used(phys_id_present_map, - mp_ioapics[apic].mpc_apicid)) { + mp_ioapics[apic].mp_apicid)) { printk(KERN_ERR "BIOS bug, IO-APIC#%d ID %d is already used!...\n", - apic, mp_ioapics[apic].mpc_apicid); + apic, mp_ioapics[apic].mp_apicid); for (i = 0; i < get_physical_broadcast(); i++) if (!physid_isset(i, phys_id_present_map)) break; @@ -1834,13 +1840,13 @@ static void __init setup_ioapic_ids_from 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; + mp_ioapics[apic].mp_apicid = i; } else { physid_mask_t tmp; - tmp = apicid_to_cpu_present(mp_ioapics[apic].mpc_apicid); + tmp = apicid_to_cpu_present(mp_ioapics[apic].mp_apicid); apic_printk(APIC_VERBOSE, "Setting %d in the " "phys_id_present_map\n", - mp_ioapics[apic].mpc_apicid); + mp_ioapics[apic].mp_apicid); physids_or(phys_id_present_map, phys_id_present_map, tmp); } @@ -1849,21 +1855,21 @@ static void __init setup_ioapic_ids_from * We need to adjust the IRQ routing table * if the ID changed. */ - if (old_id != mp_ioapics[apic].mpc_apicid) + if (old_id != mp_ioapics[apic].mp_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; + if (mp_irqs[i].mp_dstapic == old_id) + mp_irqs[i].mp_dstapic + = mp_ioapics[apic].mp_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); + mp_ioapics[apic].mp_apicid); - reg_00.bits.ID = mp_ioapics[apic].mpc_apicid; + reg_00.bits.ID = mp_ioapics[apic].mp_apicid; spin_lock_irqsave(&ioapic_lock, flags); io_apic_write(apic, 0, reg_00.raw); spin_unlock_irqrestore(&ioapic_lock, flags); @@ -1874,17 +1880,13 @@ static void __init setup_ioapic_ids_from 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) + if (reg_00.bits.ID != mp_ioapics[apic].mp_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 int no_timer_check __initdata; static int __init notimercheck(char *s) @@ -2077,45 +2079,53 @@ static inline void init_IO_APIC_traps(vo * The local APIC irq-chip implementation: */ -static void ack_apic(unsigned int irq) +static void ack_lapic_irq(unsigned int irq) { ack_APIC_irq(); } -static void mask_lapic_irq (unsigned int irq) +static void mask_lapic_irq(unsigned int irq) { unsigned long v; v = apic_read(APIC_LVT0); - apic_write_around(APIC_LVT0, v | APIC_LVT_MASKED); + apic_write(APIC_LVT0, v | APIC_LVT_MASKED); } -static void unmask_lapic_irq (unsigned int irq) +static void unmask_lapic_irq(unsigned int irq) { unsigned long v; v = apic_read(APIC_LVT0); - apic_write_around(APIC_LVT0, v & ~APIC_LVT_MASKED); + apic_write(APIC_LVT0, v & ~APIC_LVT_MASKED); } static struct irq_chip lapic_chip __read_mostly = { - .name = "local-APIC-edge", + .name = "local-APIC", .mask = mask_lapic_irq, .unmask = unmask_lapic_irq, - .eoi = ack_apic, + .ack = ack_lapic_irq, }; +static void lapic_register_intr(int irq, int vector) +{ + irq_desc[irq].status &= ~IRQ_LEVEL; + set_irq_chip_and_handler_name(irq, &lapic_chip, handle_edge_irq, + "edge"); + set_intr_gate(vector, interrupt[irq]); +} + static void __init setup_nmi(void) { /* - * Dirty trick to enable the NMI watchdog ... + * 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 ..."); enable_NMI_through_LVT0(); @@ -2191,11 +2201,16 @@ static inline void __init unlock_ExtINT_ static inline void __init check_timer(void) { int apic1, pin1, apic2, pin2; + int no_pin1 = 0; int vector; + unsigned int ver; unsigned long flags; local_irq_save(flags); + ver = apic_read(APIC_LVR); + ver = GET_APIC_VERSION(ver); + /* * get/set the timer IRQ vector: */ @@ -2204,34 +2219,54 @@ static inline void __init check_timer(vo 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. + * As IRQ0 is to be enabled in the 8259A, the virtual + * wire has to be disabled in the local APIC. Also + * timer interrupts need to be acknowledged manually in + * the 8259A for the i82489DX when using the NMI + * watchdog as that APIC treats NMIs as level-triggered. + * The AEOI mode will finish them in the 8259A + * automatically. */ - apic_write_around(APIC_LVT0, APIC_LVT_MASKED | APIC_DM_EXTINT); + apic_write(APIC_LVT0, APIC_LVT_MASKED | APIC_DM_EXTINT); init_8259A(1); - timer_ack = 1; - if (timer_over_8254 > 0) - enable_8259A_irq(0); + timer_ack = (nmi_watchdog == NMI_IO_APIC && !APIC_INTEGRATED(ver)); pin1 = find_isa_irq_pin(0, mp_INT); apic1 = find_isa_irq_apic(0, mp_INT); pin2 = ioapic_i8259.pin; apic2 = ioapic_i8259.apic; - printk(KERN_INFO "..TIMER: vector=0x%02X apic1=%d pin1=%d apic2=%d pin2=%d\n", - vector, apic1, pin1, apic2, pin2); + apic_printk(APIC_QUIET, KERN_INFO "..TIMER: vector=0x%02X " + "apic1=%d pin1=%d apic2=%d pin2=%d\n", + vector, apic1, pin1, apic2, pin2); + + /* + * Some BIOS writers are clueless and report the ExtINTA + * I/O APIC input from the cascaded 8259A as the timer + * interrupt input. So just in case, if only one pin + * was found above, try it both directly and through the + * 8259A. + */ + if (pin1 == -1) { + pin1 = pin2; + apic1 = apic2; + no_pin1 = 1; + } else if (pin2 == -1) { + pin2 = pin1; + apic2 = apic1; + } if (pin1 != -1) { /* * Ok, does IRQ0 through the IOAPIC work? */ + if (no_pin1) { + add_pin_to_irq(0, apic1, pin1); + setup_timer_IRQ0_pin(apic1, pin1, vector); + } 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); } @@ -2240,71 +2275,77 @@ static inline void __init check_timer(vo goto out; } 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); + if (!no_pin1) + apic_printk(APIC_QUIET, KERN_ERR "..MP-BIOS bug: " + "8254 timer not connected to IO-APIC\n"); + + apic_printk(APIC_QUIET, KERN_INFO "...trying to set up timer " + "(IRQ0) through the 8259A ...\n"); + apic_printk(APIC_QUIET, KERN_INFO + "..... (found apic %d pin %d) ...\n", apic2, pin2); /* * legacy devices should be connected to IO APIC #0 */ - setup_ExtINT_IRQ0_pin(apic2, pin2, vector); + replace_pin_at_irq(0, apic1, pin1, apic2, pin2); + setup_timer_IRQ0_pin(apic2, pin2, vector); + unmask_IO_APIC_irq(0); + enable_8259A_irq(0); 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); + apic_printk(APIC_QUIET, KERN_INFO "....... works.\n"); + timer_through_8259 = 1; if (nmi_watchdog == NMI_IO_APIC) { + disable_8259A_irq(0); setup_nmi(); + enable_8259A_irq(0); } goto out; } /* * Cleanup, just in case ... */ + disable_8259A_irq(0); clear_IO_APIC_pin(apic2, pin2); + apic_printk(APIC_QUIET, KERN_INFO "....... failed.\n"); } - 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; + apic_printk(APIC_QUIET, KERN_WARNING "timer doesn't work " + "through the IO-APIC - disabling NMI Watchdog!\n"); + nmi_watchdog = NMI_NONE; } + timer_ack = 0; - printk(KERN_INFO "...trying to set up timer as Virtual Wire IRQ..."); + apic_printk(APIC_QUIET, KERN_INFO + "...trying to set up timer as Virtual Wire IRQ...\n"); - disable_8259A_irq(0); - set_irq_chip_and_handler_name(0, &lapic_chip, handle_fasteoi_irq, - "fasteoi"); - apic_write_around(APIC_LVT0, APIC_DM_FIXED | vector); /* Fixed mode */ + lapic_register_intr(0, vector); + apic_write(APIC_LVT0, APIC_DM_FIXED | vector); /* Fixed mode */ enable_8259A_irq(0); if (timer_irq_works()) { - printk(" works.\n"); + apic_printk(APIC_QUIET, KERN_INFO "..... works.\n"); goto out; } - apic_write_around(APIC_LVT0, APIC_LVT_MASKED | APIC_DM_FIXED | vector); - printk(" failed.\n"); + disable_8259A_irq(0); + apic_write(APIC_LVT0, APIC_LVT_MASKED | APIC_DM_FIXED | vector); + apic_printk(APIC_QUIET, KERN_INFO "..... failed.\n"); - printk(KERN_INFO "...trying to set up timer as ExtINT IRQ..."); + apic_printk(APIC_QUIET, KERN_INFO + "...trying to set up timer as ExtINT IRQ...\n"); - timer_ack = 0; init_8259A(0); make_8259A_irq(0); - apic_write_around(APIC_LVT0, APIC_DM_EXTINT); + apic_write(APIC_LVT0, APIC_DM_EXTINT); unlock_ExtINT_logic(); if (timer_irq_works()) { - printk(" works.\n"); + apic_printk(APIC_QUIET, KERN_INFO "..... works.\n"); goto out; } - printk(" failed :(.\n"); + apic_printk(APIC_QUIET, KERN_INFO "..... failed :(.\n"); panic("IO-APIC + timer doesn't work! Boot with apic=debug and send a " - "report. Then try booting with the 'noapic' option"); + "report. Then try booting with the 'noapic' option.\n"); out: local_irq_restore(flags); } @@ -2314,11 +2355,21 @@ int timer_uses_ioapic_pin_0 = 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. + * Traditionally ISA IRQ2 is the cascade IRQ, and is not available + * to devices. However there may be an I/O APIC pin available for + * this interrupt regardless. The pin may be left unconnected, but + * typically it will be reused as an ExtINT cascade interrupt for + * the master 8259A. In the MPS case such a pin will normally be + * reported as an ExtINT interrupt in the MP table. With ACPI + * there is no provision for ExtINT interrupts, and in the absence + * of an override it would be treated as an ordinary ISA I/O APIC + * interrupt, that is edge-triggered and unmasked by default. We + * used to do this, but it caused problems on some systems because + * of the NMI watchdog and sometimes IRQ0 of the 8254 timer using + * the same ExtINT cascade interrupt to drive the local APIC of the + * bootstrap processor. Therefore we refrain from routing IRQ2 to + * the I/O APIC in all cases now. No actual device should request + * it anyway. --macro */ #define PIC_IRQS (1 << PIC_CASCADE_IR) @@ -2328,25 +2379,22 @@ void __init setup_IO_APIC(void) int i; /* Reserve all the system vectors. */ - for (i = FIRST_SYSTEM_VECTOR; i < NR_VECTORS; i++) + for (i = first_system_vector; i < NR_VECTORS; i++) set_bit(i, used_vectors); #endif enable_IO_APIC(); - if (acpi_ioapic) - io_apic_irqs = ~0; /* all IRQs go through IOAPIC */ - else - io_apic_irqs = ~PIC_IRQS; + io_apic_irqs = ~PIC_IRQS; printk("ENABLING IO-APIC IRQs\n"); +#ifndef CONFIG_XEN /* * 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(); @@ -2356,28 +2404,14 @@ void __init setup_IO_APIC(void) 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) + if (sis_apic_bug == -1) sis_apic_bug = 0; if (is_initial_xendomain()) { struct xen_platform_op op = { .cmd = XENPF_platform_quirk }; @@ -2396,17 +2430,17 @@ 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 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; int i; - + data = container_of(dev, struct sysfs_ioapic_data, dev); entry = data->entry; - for (i = 0; i < nr_ioapic_registers[dev->id]; i ++) + for (i = 0; i < nr_ioapic_registers[dev->id]; i++) entry[i] = ioapic_read_entry(dev->id, i); return 0; @@ -2419,18 +2453,18 @@ static int ioapic_resume(struct sys_devi 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; + if (reg_00.bits.ID != mp_ioapics[dev->id].mp_apicid) { + reg_00.bits.ID = mp_ioapics[dev->id].mp_apicid; io_apic_write(dev->id, 0, reg_00.raw); } spin_unlock_irqrestore(&ioapic_lock, flags); - for (i = 0; i < nr_ioapic_registers[dev->id]; i ++) + for (i = 0; i < nr_ioapic_registers[dev->id]; i++) ioapic_write_entry(dev->id, i, entry[i]); return 0; @@ -2444,24 +2478,23 @@ static struct sysdev_class ioapic_sysdev static int __init ioapic_init_sysfs(void) { - struct sys_device * dev; + 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] + 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); + mp_ioapic_data[i] = kzalloc(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->id = i; dev->cls = &ioapic_sysdev_class; error = sysdev_register(dev); if (error) { @@ -2538,7 +2571,7 @@ static int msi_compose_msg(struct pci_de msg->address_lo = MSI_ADDR_BASE_LO | ((INT_DEST_MODE == 0) ? - MSI_ADDR_DEST_MODE_PHYSICAL: +MSI_ADDR_DEST_MODE_PHYSICAL: MSI_ADDR_DEST_MODE_LOGICAL) | ((INT_DELIVERY_MODE != dest_LowestPrio) ? MSI_ADDR_REDIRECTION_CPU: @@ -2549,7 +2582,7 @@ static int msi_compose_msg(struct pci_de MSI_DATA_TRIGGER_EDGE | MSI_DATA_LEVEL_ASSERT | ((INT_DELIVERY_MODE != dest_LowestPrio) ? - MSI_DATA_DELIVERY_FIXED: +MSI_DATA_DELIVERY_FIXED: MSI_DATA_DELIVERY_LOWPRI) | MSI_DATA_VECTOR(vector); } @@ -2720,13 +2753,13 @@ int arch_setup_ht_irq(unsigned int irq, #endif /* CONFIG_HT_IRQ */ /* -------------------------------------------------------------------------- - ACPI-based IOAPIC Configuration + ACPI-based IOAPIC Configuration -------------------------------------------------------------------------- */ #ifdef CONFIG_ACPI #ifndef CONFIG_XEN -int __init io_apic_get_unique_id (int ioapic, int apic_id) +int __init io_apic_get_unique_id(int ioapic, int apic_id) { union IO_APIC_reg_00 reg_00; static physid_mask_t apic_id_map = PHYSID_MASK_NONE; @@ -2735,10 +2768,10 @@ int __init io_apic_get_unique_id (int io 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 + * 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. */ @@ -2757,7 +2790,7 @@ int __init io_apic_get_unique_id (int io } /* - * Every APIC in a system must have a unique ID or we get lots of nice + * 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)) { @@ -2774,7 +2807,7 @@ int __init io_apic_get_unique_id (int io "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); @@ -2802,7 +2835,7 @@ int __init io_apic_get_unique_id (int io #endif /* !CONFIG_XEN */ -int __init io_apic_get_version (int ioapic) +int __init io_apic_get_version(int ioapic) { union IO_APIC_reg_01 reg_01; unsigned long flags; @@ -2815,7 +2848,7 @@ int __init io_apic_get_version (int ioap } -int __init io_apic_get_redir_entries (int ioapic) +int __init io_apic_get_redir_entries(int ioapic) { union IO_APIC_reg_01 reg_01; unsigned long flags; @@ -2828,7 +2861,7 @@ int __init io_apic_get_redir_entries (in } -int io_apic_set_pci_routing (int ioapic, int pin, int irq, int edge_level, int active_high_low) +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; @@ -2844,7 +2877,7 @@ int io_apic_set_pci_routing (int ioapic, * corresponding device driver registers for this IRQ. */ - memset(&entry,0,sizeof(entry)); + memset(&entry, 0, sizeof(entry)); entry.delivery_mode = INT_DELIVERY_MODE; entry.dest_mode = INT_DEST_MODE; @@ -2863,7 +2896,7 @@ int io_apic_set_pci_routing (int ioapic, 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, + mp_ioapics[ioapic].mp_apicid, pin, entry.vector, irq, edge_level, active_high_low); ioapic_register_intr(irq, entry.vector, edge_level); @@ -2884,8 +2917,8 @@ int acpi_get_override_irq(int bus_irq, i return -1; for (i = 0; i < mp_irq_entries; i++) - if (mp_irqs[i].mpc_irqtype == mp_INT && - mp_irqs[i].mpc_srcbusirq == bus_irq) + if (mp_irqs[i].mp_irqtype == mp_INT && + mp_irqs[i].mp_srcbusirq == bus_irq) break; if (i >= mp_irq_entries) return -1; @@ -2918,3 +2951,35 @@ static int __init parse_noapic(char *arg return 0; } early_param("noapic", parse_noapic); + +#ifndef CONFIG_XEN +void __init ioapic_init_mappings(void) +{ + unsigned long ioapic_phys, idx = FIX_IO_APIC_BASE_0; + int i; + + for (i = 0; i < nr_ioapics; i++) { + if (smp_found_config) { + ioapic_phys = mp_ioapics[i].mp_apicaddr; + if (!ioapic_phys) { + printk(KERN_ERR + "WARNING: bogus zero IO-APIC " + "address found in MPTABLE, " + "disabling IO/APIC support!\n"); + smp_found_config = 0; + skip_ioapic_setup = 1; + goto fake_ioapic_page; + } + } else { +fake_ioapic_page: + ioapic_phys = (unsigned long) + alloc_bootmem_pages(PAGE_SIZE); + ioapic_phys = __pa(ioapic_phys); + } + set_fixmap_nocache(idx, ioapic_phys); + printk(KERN_DEBUG "mapped IOAPIC to %08lx (%08lx)\n", + __fix_to_virt(idx), ioapic_phys); + idx++; + } +} +#endif --- head-2011-03-11.orig/arch/x86/kernel/io_apic_64-xen.c 2011-01-31 18:07:35.000000000 +0100 +++ head-2011-03-11/arch/x86/kernel/io_apic_64-xen.c 2011-02-01 14:38:38.000000000 +0100 @@ -45,6 +45,7 @@ #include #include #include +#include #include #include #include @@ -63,10 +64,16 @@ struct irq_cfg { }; /* irq_cfg is indexed by the sum of all RTEs in all I/O APICs. */ -struct irq_cfg irq_cfg[NR_IRQS] __read_mostly; +static struct irq_cfg irq_cfg[NR_IRQS] __read_mostly; static int assign_irq_vector(int irq, cpumask_t mask); +#ifndef CONFIG_XEN +int first_system_vector = 0xfe; + +char system_vectors[NR_VECTORS] = { [0 ... NR_VECTORS-1] = SYS_VECTOR_FREE}; +#endif + #define __apicdebuginit __init int sis_apic_bug; /* not actually supported, dummy for compile */ @@ -89,14 +96,14 @@ unsigned long io_apic_irqs; #define clear_IO_APIC() ((void)0) #else -int timer_over_8254 __initdata = 1; +int timer_through_8259 __initdata; /* 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); -DEFINE_SPINLOCK(vector_lock); +static DEFINE_SPINLOCK(vector_lock); /* * # of IRQ routing registers @@ -104,15 +111,17 @@ DEFINE_SPINLOCK(vector_lock); int nr_ioapic_registers[MAX_IO_APICS]; /* I/O APIC entries */ -struct mpc_config_ioapic mp_ioapics[MAX_IO_APICS]; +struct mp_config_ioapic mp_ioapics[MAX_IO_APICS]; int nr_ioapics; /* MP IRQ source entries */ -struct mpc_config_intsrc mp_irqs[MAX_IRQ_SOURCES]; +struct mp_config_intsrc mp_irqs[MAX_IRQ_SOURCES]; /* # of MP IRQ source entries */ int mp_irq_entries; +DECLARE_BITMAP(mp_bus_not_pci, MAX_MP_BUSSES); + /* * Rough estimation of how many shared IRQs there are, can * be changed anytime. @@ -141,7 +150,7 @@ struct io_apic { static __attribute_const__ struct io_apic __iomem *io_apic_base(int idx) { return (void __iomem *) __fix_to_virt(FIX_IO_APIC_BASE_0 + idx) - + (mp_ioapics[idx].mpc_apicaddr & ~PAGE_MASK); + + (mp_ioapics[idx].mp_apicaddr & ~PAGE_MASK); } #endif @@ -155,7 +164,7 @@ static inline unsigned int io_apic_read( struct physdev_apic apic_op; int ret; - apic_op.apic_physbase = mp_ioapics[apic].mpc_apicaddr; + apic_op.apic_physbase = mp_ioapics[apic].mp_apicaddr; apic_op.reg = reg; ret = HYPERVISOR_physdev_op(PHYSDEVOP_apic_read, &apic_op); if (ret) @@ -173,7 +182,7 @@ static inline void io_apic_write(unsigne #else struct physdev_apic apic_op; - apic_op.apic_physbase = mp_ioapics[apic].mpc_apicaddr; + apic_op.apic_physbase = mp_ioapics[apic].mp_apicaddr; apic_op.reg = reg; apic_op.value = value; WARN_ON(HYPERVISOR_physdev_op(PHYSDEVOP_apic_write, &apic_op)); @@ -209,7 +218,7 @@ static bool io_apic_level_ack_pending(un break; reg = io_apic_read(entry->apic, 0x10 + pin*2); /* Is the remote IRR bit set? */ - if ((reg >> 14) & 1) { + if (reg & IO_APIC_REDIR_REMOTE_IRR) { spin_unlock_irqrestore(&ioapic_lock, flags); return true; } @@ -308,7 +317,7 @@ static void __target_IO_APIC_irq(unsigne break; io_apic_write(apic, 0x11 + pin*2, dest); reg = io_apic_read(apic, 0x10 + pin*2); - reg &= ~0x000000ff; + reg &= ~IO_APIC_REDIR_VECTOR_MASK; reg |= vector; io_apic_modify(apic, reg); if (!entry->next) @@ -372,6 +381,26 @@ static void add_pin_to_irq(unsigned int } #ifndef CONFIG_XEN +/* + * 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; + } +} + #define __DO_ACTION(R, ACTION, FINAL) \ \ { \ @@ -399,10 +428,11 @@ static void add_pin_to_irq(unsigned int 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 */ +/* mask = 1 */ +DO_ACTION(__mask, 0, |= IO_APIC_REDIR_MASKED, io_apic_sync(entry->apic)) + +/* mask = 0 */ +DO_ACTION(__unmask, 0, &= ~IO_APIC_REDIR_MASKED, ) static void mask_IO_APIC_irq (unsigned int irq) { @@ -465,22 +495,6 @@ static int __init disable_timer_pin_setu } __setup("disable_timer_pin_1", disable_timer_pin_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 */ - /* * Find the IRQ entry number of a certain pin. @@ -490,10 +504,10 @@ static int find_irq_entry(int apic, int 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) + if (mp_irqs[i].mp_irqtype == type && + (mp_irqs[i].mp_dstapic == mp_ioapics[apic].mp_apicid || + mp_irqs[i].mp_dstapic == MP_APIC_ALL) && + mp_irqs[i].mp_dstirq == pin) return i; return -1; @@ -508,13 +522,13 @@ static int __init find_isa_irq_pin(int i int i; for (i = 0; i < mp_irq_entries; i++) { - int lbus = mp_irqs[i].mpc_srcbus; + int lbus = mp_irqs[i].mp_srcbus; if (test_bit(lbus, mp_bus_not_pci) && - (mp_irqs[i].mpc_irqtype == type) && - (mp_irqs[i].mpc_srcbusirq == irq)) + (mp_irqs[i].mp_irqtype == type) && + (mp_irqs[i].mp_srcbusirq == irq)) - return mp_irqs[i].mpc_dstirq; + return mp_irqs[i].mp_dstirq; } return -1; } @@ -524,17 +538,17 @@ static int __init find_isa_irq_apic(int int i; for (i = 0; i < mp_irq_entries; i++) { - int lbus = mp_irqs[i].mpc_srcbus; + int lbus = mp_irqs[i].mp_srcbus; if (test_bit(lbus, mp_bus_not_pci) && - (mp_irqs[i].mpc_irqtype == type) && - (mp_irqs[i].mpc_srcbusirq == irq)) + (mp_irqs[i].mp_irqtype == type) && + (mp_irqs[i].mp_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) + if (mp_ioapics[apic].mp_apicid == mp_irqs[i].mp_dstapic) return apic; } } @@ -555,28 +569,28 @@ int IO_APIC_get_PCI_irq_vector(int bus, 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) { + if (test_bit(bus, mp_bus_not_pci)) { 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; + int lbus = mp_irqs[i].mp_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) + if (mp_ioapics[apic].mp_apicid == mp_irqs[i].mp_dstapic || + mp_irqs[i].mp_dstapic == MP_APIC_ALL) break; if (!test_bit(lbus, mp_bus_not_pci) && - !mp_irqs[i].mpc_irqtype && + !mp_irqs[i].mp_irqtype && (bus == lbus) && - (slot == ((mp_irqs[i].mpc_srcbusirq >> 2) & 0x1f))) { - int irq = pin_2_irq(i,apic,mp_irqs[i].mpc_dstirq); + (slot == ((mp_irqs[i].mp_srcbusirq >> 2) & 0x1f))) { + int irq = pin_2_irq(i,apic,mp_irqs[i].mp_dstirq); if (!(apic || IO_APIC_IRQ(irq))) continue; - if (pin == (mp_irqs[i].mpc_srcbusirq & 3)) + if (pin == (mp_irqs[i].mp_srcbusirq & 3)) return irq; /* * Use the first all-but-pin matching entry as a @@ -604,13 +618,13 @@ int IO_APIC_get_PCI_irq_vector(int bus, static int MPBIOS_polarity(int idx) { - int bus = mp_irqs[idx].mpc_srcbus; + int bus = mp_irqs[idx].mp_srcbus; int polarity; /* * Determine IRQ line polarity (high active or low active): */ - switch (mp_irqs[idx].mpc_irqflag & 3) + switch (mp_irqs[idx].mp_irqflag & 3) { case 0: /* conforms, ie. bus-type dependent polarity */ if (test_bit(bus, mp_bus_not_pci)) @@ -646,13 +660,13 @@ static int MPBIOS_polarity(int idx) static int MPBIOS_trigger(int idx) { - int bus = mp_irqs[idx].mpc_srcbus; + int bus = mp_irqs[idx].mp_srcbus; int trigger; /* * Determine IRQ trigger mode (edge or level sensitive): */ - switch ((mp_irqs[idx].mpc_irqflag>>2) & 3) + switch ((mp_irqs[idx].mp_irqflag>>2) & 3) { case 0: /* conforms, ie. bus-type dependent */ if (test_bit(bus, mp_bus_not_pci)) @@ -699,16 +713,16 @@ static inline int irq_trigger(int idx) static int pin_2_irq(int idx, int apic, int pin) { int irq, i; - int bus = mp_irqs[idx].mpc_srcbus; + int bus = mp_irqs[idx].mp_srcbus; /* * Debugging check, we are in big trouble if this message pops up! */ - if (mp_irqs[idx].mpc_dstirq != pin) + if (mp_irqs[idx].mp_dstirq != pin) printk(KERN_ERR "broken BIOS or MPTABLE parser, ayiee!!\n"); if (test_bit(bus, mp_bus_not_pci)) { - irq = mp_irqs[idx].mpc_srcbusirq; + irq = mp_irqs[idx].mp_srcbusirq; } else { /* * PCI IRQs are mapped in order @@ -722,6 +736,19 @@ static int pin_2_irq(int idx, int apic, return irq; } +void lock_vector_lock(void) +{ + /* Used to the online set of cpus does not change + * during assign_irq_vector. + */ + spin_lock(&vector_lock); +} + +void unlock_vector_lock(void) +{ + spin_unlock(&vector_lock); +} + static int __assign_irq_vector(int irq, cpumask_t mask) { struct physdev_irq irq_op; @@ -773,7 +800,7 @@ static void __clear_irq_vector(int irq) vector = cfg->vector; cpus_and(mask, cfg->domain, cpu_online_map); - for_each_cpu_mask(cpu, mask) + for_each_cpu_mask_nr(cpu, mask) per_cpu(vector_irq, cpu)[vector] = -1; cfg->vector = 0; @@ -842,7 +869,7 @@ static void setup_IO_APIC_irq(int apic, apic_printk(APIC_VERBOSE,KERN_DEBUG "IOAPIC[%d]: Set routing entry (%d-%d -> 0x%x -> " "IRQ %d Mode:%i Active:%i)\n", - apic, mp_ioapics[apic].mpc_apicid, pin, cfg->vector, + apic, mp_ioapics[apic].mp_apicid, pin, cfg->vector, irq, trigger, polarity); /* @@ -883,10 +910,10 @@ static void __init setup_IO_APIC_irqs(vo 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); + apic_printk(APIC_VERBOSE, KERN_DEBUG " IO-APIC (apicid-pin) %d-%d", mp_ioapics[apic].mp_apicid, pin); first_notcon = 0; } else - apic_printk(APIC_VERBOSE, ", %d-%d", mp_ioapics[apic].mpc_apicid, pin); + apic_printk(APIC_VERBOSE, ", %d-%d", mp_ioapics[apic].mp_apicid, pin); continue; } if (!first_notcon) { @@ -908,26 +935,21 @@ static void __init setup_IO_APIC_irqs(vo #ifndef CONFIG_XEN /* - * Set up the 8259A-master output pin as broadcast to all - * CPUs. + * Set up the timer pin, possibly with the 8259A-master behind. */ -static void __init setup_ExtINT_IRQ0_pin(unsigned int apic, unsigned int pin, int vector) +static void __init setup_timer_IRQ0_pin(unsigned int apic, unsigned int pin, + int vector) { struct IO_APIC_route_entry entry; 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.mask = 1; /* mask IRQ now */ entry.dest = cpu_mask_to_apicid(TARGET_CPUS); entry.delivery_mode = INT_DELIVERY_MODE; entry.polarity = 0; @@ -936,7 +958,7 @@ static void __init setup_ExtINT_IRQ0_pin /* * The timer IRQ doesn't have to know that behind the - * scene we have a 8259A-master in AEOI mode ... + * scene we may have a 8259A-master in AEOI mode ... */ set_irq_chip_and_handler_name(0, &ioapic_chip, handle_edge_irq, "edge"); @@ -944,8 +966,6 @@ static void __init setup_ExtINT_IRQ0_pin * Add it to the IO-APIC irq-routing table: */ ioapic_write_entry(apic, pin, entry); - - enable_8259A_irq(0); } void __apicdebuginit print_IO_APIC(void) @@ -962,7 +982,7 @@ void __apicdebuginit print_IO_APIC(void) 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]); + mp_ioapics[i].mp_apicid, nr_ioapic_registers[i]); /* * We are a bit conservative about what we expect. We have to @@ -980,7 +1000,7 @@ void __apicdebuginit print_IO_APIC(void) spin_unlock_irqrestore(&ioapic_lock, flags); printk("\n"); - printk(KERN_DEBUG "IO APIC #%d......\n", mp_ioapics[apic].mpc_apicid); + printk(KERN_DEBUG "IO APIC #%d......\n", mp_ioapics[apic].mp_apicid); printk(KERN_DEBUG ".... register #00: %08X\n", reg_00.raw); printk(KERN_DEBUG "....... : physical APIC id: %02X\n", reg_00.bits.ID); @@ -1072,6 +1092,7 @@ void __apicdebuginit print_local_APIC(vo 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(read_apic_id())); v = apic_read(APIC_LVR); printk(KERN_INFO "... APIC VERSION: %08x\n", v); @@ -1141,7 +1162,7 @@ void __apicdebuginit print_local_APIC(vo void print_all_local_APICs (void) { - on_each_cpu(print_local_APIC, NULL, 1, 1); + on_each_cpu(print_local_APIC, NULL, 1); } void __apicdebuginit print_PIC(void) @@ -1175,6 +1196,8 @@ void __apicdebuginit print_PIC(void) v = inb(0x4d1) << 8 | inb(0x4d0); printk(KERN_DEBUG "... PIC ELCR: %04x\n", v); } +#else +void __apicdebuginit print_IO_APIC(void) {} #endif /* !CONFIG_XEN */ void __init enable_IO_APIC(void) @@ -1359,12 +1382,10 @@ static unsigned int startup_ioapic_irq(u static int ioapic_retrigger_irq(unsigned int irq) { struct irq_cfg *cfg = &irq_cfg[irq]; - cpumask_t mask; unsigned long flags; spin_lock_irqsave(&vector_lock, flags); - mask = cpumask_of_cpu(first_cpu(cfg->domain)); - send_IPI_mask(mask, cfg->vector); + send_IPI_mask(cpumask_of_cpu(first_cpu(cfg->domain)), cfg->vector); spin_unlock_irqrestore(&vector_lock, flags); return 1; @@ -1545,7 +1566,7 @@ static inline void init_IO_APIC_traps(vo } #ifndef CONFIG_XEN -static void enable_lapic_irq (unsigned int irq) +static void unmask_lapic_irq(unsigned int irq) { unsigned long v; @@ -1553,7 +1574,7 @@ static void enable_lapic_irq (unsigned i apic_write(APIC_LVT0, v & ~APIC_LVT_MASKED); } -static void disable_lapic_irq (unsigned int irq) +static void mask_lapic_irq(unsigned int irq) { unsigned long v; @@ -1566,19 +1587,20 @@ static void ack_lapic_irq (unsigned int ack_APIC_irq(); } -static void end_lapic_irq (unsigned int i) { /* nothing */ } - -static struct hw_interrupt_type lapic_irq_type __read_mostly = { - .name = "local-APIC", - .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 struct irq_chip lapic_chip __read_mostly = { + .name = "local-APIC", + .mask = mask_lapic_irq, + .unmask = unmask_lapic_irq, + .ack = ack_lapic_irq, }; +static void lapic_register_intr(int irq) +{ + irq_desc[irq].status &= ~IRQ_LEVEL; + set_irq_chip_and_handler_name(irq, &lapic_chip, handle_edge_irq, + "edge"); +} + static void __init setup_nmi(void) { /* @@ -1664,6 +1686,7 @@ static inline void __init check_timer(vo struct irq_cfg *cfg = irq_cfg + 0; int apic1, pin1, apic2, pin2; unsigned long flags; + int no_pin1 = 0; local_irq_save(flags); @@ -1674,34 +1697,48 @@ static inline void __init check_timer(vo assign_irq_vector(0, TARGET_CPUS); /* - * 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. + * As IRQ0 is to be enabled in the 8259A, 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; - apic_printk(APIC_VERBOSE,KERN_INFO "..TIMER: vector=0x%02X apic1=%d pin1=%d apic2=%d pin2=%d\n", - cfg->vector, apic1, pin1, apic2, pin2); + apic_printk(APIC_QUIET, KERN_INFO "..TIMER: vector=0x%02X " + "apic1=%d pin1=%d apic2=%d pin2=%d\n", + cfg->vector, apic1, pin1, apic2, pin2); + + /* + * Some BIOS writers are clueless and report the ExtINTA + * I/O APIC input from the cascaded 8259A as the timer + * interrupt input. So just in case, if only one pin + * was found above, try it both directly and through the + * 8259A. + */ + if (pin1 == -1) { + pin1 = pin2; + apic1 = apic2; + no_pin1 = 1; + } else if (pin2 == -1) { + pin2 = pin1; + apic2 = apic1; + } if (pin1 != -1) { /* * Ok, does IRQ0 through the IOAPIC work? */ + if (no_pin1) { + add_pin_to_irq(0, apic1, pin1); + setup_timer_IRQ0_pin(apic1, pin1, cfg->vector); + } 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); } @@ -1710,54 +1747,62 @@ static inline void __init check_timer(vo goto out; } 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); + if (!no_pin1) + apic_printk(APIC_QUIET, KERN_ERR "..MP-BIOS bug: " + "8254 timer not connected to IO-APIC\n"); + + apic_printk(APIC_QUIET, KERN_INFO "...trying to set up timer " + "(IRQ0) through the 8259A ...\n"); + apic_printk(APIC_QUIET, KERN_INFO + "..... (found apic %d pin %d) ...\n", apic2, pin2); /* * legacy devices should be connected to IO APIC #0 */ - setup_ExtINT_IRQ0_pin(apic2, pin2, cfg->vector); + replace_pin_at_irq(0, apic1, pin1, apic2, pin2); + setup_timer_IRQ0_pin(apic2, pin2, cfg->vector); + unmask_IO_APIC_irq(0); + enable_8259A_irq(0); if (timer_irq_works()) { - apic_printk(APIC_VERBOSE," works.\n"); - nmi_watchdog_default(); + apic_printk(APIC_QUIET, KERN_INFO "....... works.\n"); + timer_through_8259 = 1; if (nmi_watchdog == NMI_IO_APIC) { + disable_8259A_irq(0); setup_nmi(); + enable_8259A_irq(0); } goto out; } /* * Cleanup, just in case ... */ + disable_8259A_irq(0); clear_IO_APIC_pin(apic2, pin2); + apic_printk(APIC_QUIET, KERN_INFO "....... failed.\n"); } - 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_QUIET, KERN_WARNING "timer doesn't work " + "through the IO-APIC - disabling NMI Watchdog!\n"); + nmi_watchdog = NMI_NONE; } - apic_printk(APIC_VERBOSE, KERN_INFO "...trying to set up timer as Virtual Wire IRQ..."); + apic_printk(APIC_QUIET, KERN_INFO + "...trying to set up timer as Virtual Wire IRQ...\n"); - disable_8259A_irq(0); - irq_desc[0].chip = &lapic_irq_type; + lapic_register_intr(0); apic_write(APIC_LVT0, APIC_DM_FIXED | cfg->vector); /* Fixed mode */ enable_8259A_irq(0); if (timer_irq_works()) { - apic_printk(APIC_VERBOSE," works.\n"); + apic_printk(APIC_QUIET, KERN_INFO "..... works.\n"); goto out; } + disable_8259A_irq(0); apic_write(APIC_LVT0, APIC_LVT_MASKED | APIC_DM_FIXED | cfg->vector); - apic_printk(APIC_VERBOSE," failed.\n"); + apic_printk(APIC_QUIET, KERN_INFO "..... failed.\n"); - apic_printk(APIC_VERBOSE, KERN_INFO "...trying to set up timer as ExtINT IRQ..."); + apic_printk(APIC_QUIET, KERN_INFO + "...trying to set up timer as ExtINT IRQ...\n"); init_8259A(0); make_8259A_irq(0); @@ -1766,11 +1811,12 @@ static inline void __init check_timer(vo unlock_ExtINT_logic(); if (timer_irq_works()) { - apic_printk(APIC_VERBOSE," works.\n"); + apic_printk(APIC_QUIET, KERN_INFO "..... works.\n"); goto out; } - apic_printk(APIC_VERBOSE," failed :(.\n"); - panic("IO-APIC + timer doesn't work! Try using the 'noapic' kernel parameter\n"); + apic_printk(APIC_QUIET, KERN_INFO "..... failed :(.\n"); + panic("IO-APIC + timer doesn't work! Boot with apic=debug and send a " + "report. Then try booting with the 'noapic' option.\n"); out: local_irq_restore(flags); } @@ -1788,10 +1834,21 @@ __setup("no_timer_check", notimercheck); /* * - * IRQs 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. + * Traditionally ISA IRQ2 is the cascade IRQ, and is not available + * to devices. However there may be an I/O APIC pin available for + * this interrupt regardless. The pin may be left unconnected, but + * typically it will be reused as an ExtINT cascade interrupt for + * the master 8259A. In the MPS case such a pin will normally be + * reported as an ExtINT interrupt in the MP table. With ACPI + * there is no provision for ExtINT interrupts, and in the absence + * of an override it would be treated as an ordinary ISA I/O APIC + * interrupt, that is edge-triggered and unmasked by default. We + * used to do this, but it caused problems on some systems because + * of the NMI watchdog and sometimes IRQ0 of the 8254 timer using + * the same ExtINT cascade interrupt to drive the local APIC of the + * bootstrap processor. Therefore we refrain from routing IRQ2 to + * the I/O APIC in all cases now. No actual device should request + * it anyway. --macro */ #define PIC_IRQS (1<<2) @@ -1799,10 +1856,7 @@ 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; + io_apic_irqs = ~PIC_IRQS; apic_printk(APIC_VERBOSE, "ENABLING IO-APIC IRQs\n"); @@ -1851,8 +1905,8 @@ static int ioapic_resume(struct sys_devi 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; + if (reg_00.bits.ID != mp_ioapics[dev->id].mp_apicid) { + reg_00.bits.ID = mp_ioapics[dev->id].mp_apicid; io_apic_write(dev->id, 0, reg_00.raw); } spin_unlock_irqrestore(&ioapic_lock, flags); @@ -2254,8 +2308,8 @@ int acpi_get_override_irq(int bus_irq, i return -1; for (i = 0; i < mp_irq_entries; i++) - if (mp_irqs[i].mpc_irqtype == mp_INT && - mp_irqs[i].mpc_srcbusirq == bus_irq) + if (mp_irqs[i].mp_irqtype == mp_INT && + mp_irqs[i].mp_srcbusirq == bus_irq) break; if (i >= mp_irq_entries) return -1; @@ -2349,7 +2403,7 @@ void __init ioapic_init_mappings(void) ioapic_res = ioapic_setup_resources(); for (i = 0; i < nr_ioapics; i++) { if (smp_found_config) { - ioapic_phys = mp_ioapics[i].mpc_apicaddr; + ioapic_phys = mp_ioapics[i].mp_apicaddr; } else { ioapic_phys = (unsigned long) alloc_bootmem_pages(PAGE_SIZE); --- head-2011-03-11.orig/arch/x86/kernel/ldt-xen.c 2011-01-31 18:01:51.000000000 +0100 +++ head-2011-03-11/arch/x86/kernel/ldt-xen.c 2011-02-01 14:38:38.000000000 +0100 @@ -20,9 +20,9 @@ #include #ifdef CONFIG_SMP -static void flush_ldt(void *null) +static void flush_ldt(void *current_mm) { - if (current->active_mm) + if (current->active_mm == current_mm) load_LDT(¤t->active_mm->context); } #endif @@ -62,8 +62,6 @@ static int alloc_ldt(mm_context_t *pc, i if (reload) { #ifdef CONFIG_SMP - cpumask_t mask; - preempt_disable(); #endif make_pages_readonly(newldt, @@ -71,9 +69,9 @@ static int alloc_ldt(mm_context_t *pc, i 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); + if (!cpus_equal(current->mm->cpu_vm_mask, + cpumask_of_cpu(smp_processor_id()))) + smp_call_function(flush_ldt, current->mm, 1); preempt_enable(); #endif } --- head-2011-03-11.orig/arch/x86/kernel/machine_kexec_32.c 2011-01-31 14:54:00.000000000 +0100 +++ head-2011-03-11/arch/x86/kernel/machine_kexec_32.c 2011-02-01 14:38:38.000000000 +0100 @@ -131,6 +131,8 @@ void machine_kexec_setup_load_arg(xen_ke xki->page_list[PA_PTE_0] = __ma(kexec_pte0); xki->page_list[PA_PTE_1] = __ma(kexec_pte1); + if (image->type == KEXEC_TYPE_DEFAULT) + xki->page_list[PA_SWAP_PAGE] = page_to_phys(image->swap_page); } int __init machine_kexec_setup_resources(struct resource *hypervisor, --- head-2011-03-11.orig/arch/x86/kernel/microcode-xen.c 2011-01-31 18:07:35.000000000 +0100 +++ head-2011-03-11/arch/x86/kernel/microcode-xen.c 2011-02-01 14:38:38.000000000 +0100 @@ -5,13 +5,14 @@ * 2006 Shaohua Li * * This driver allows to upgrade microcode on Intel processors - * belonging to IA-32 family - PentiumPro, Pentium II, + * 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 + * Reference: Section 8.11 of Volume 3a, IA-32 Intel? Architecture + * Software Developer's Manual + * Order Number 253668 or free download from: + * + * http://developer.intel.com/design/pentium4/manuals/253668.htm * * For more information, go to http://www.urbanmyth.org/microcode * @@ -26,6 +27,7 @@ #include #include #include +#include #include #include #include @@ -86,6 +88,7 @@ static int do_microcode_update (const vo static int microcode_open (struct inode *unused1, struct file *unused2) { + cycle_kernel_lock(); return capable(CAP_SYS_RAWIO) ? 0 : -EPERM; } @@ -162,7 +165,7 @@ static int request_microcode(void) c->x86, c->x86_model, c->x86_mask); error = request_firmware(&firmware, name, µcode_pdev->dev); if (error) { - pr_debug("microcode: ucode data file %s load failed\n", name); + pr_debug("microcode: data file %s load failed\n", name); return error; } @@ -183,6 +186,9 @@ static int __init microcode_init (void) { int error; + printk(KERN_INFO + "IA-32 Microcode Update Driver: v" MICROCODE_VERSION " \n"); + error = microcode_dev_init(); if (error) return error; @@ -195,8 +201,6 @@ static int __init microcode_init (void) request_microcode(); - printk(KERN_INFO - "IA-32 Microcode Update Driver: v" MICROCODE_VERSION " \n"); return 0; } --- head-2011-03-11.orig/arch/x86/kernel/mpparse-xen.c 2011-01-31 18:07:35.000000000 +0100 +++ head-2011-03-11/arch/x86/kernel/mpparse-xen.c 2011-02-01 14:38:38.000000000 +0100 @@ -25,6 +25,9 @@ #include #include #include +#include +#include +#include #include #ifdef CONFIG_X86_32 @@ -32,27 +35,10 @@ #include #endif -/* Have we found an MP table */ -int smp_found_config; - -/* - * Various Linux-internal data structures created from the - * MP-table. - */ -#if defined (CONFIG_MCA) || defined (CONFIG_EISA) -int mp_bus_id_to_type[MAX_MP_BUSSES]; -#endif - -DECLARE_BITMAP(mp_bus_not_pci, MAX_MP_BUSSES); -int mp_bus_id_to_pci_bus[MAX_MP_BUSSES] = {[0 ... MAX_MP_BUSSES - 1] = -1 }; - -static int mp_current_pci_id; - -int pic_mode; - -/* - * Intel MP BIOS table parsing routines: - */ +static void *_bus_to_virt(unsigned long ma) +{ + return is_ISA_range(ma, ma) ? isa_bus_to_virt(ma) : bus_to_virt(ma); +} /* * Checksum an MP configuration block. @@ -68,19 +54,7 @@ static int __init mpf_checksum(unsigned return sum & 0xFF; } -#ifdef CONFIG_X86_NUMAQ -/* - * 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] - __cpuinitdata; -#endif - -static void __cpuinit MP_processor_info(struct mpc_config_processor *m) +static void __init MP_processor_info(struct mpc_config_processor *m) { #ifndef CONFIG_XEN int apicid; @@ -90,11 +64,12 @@ static void __cpuinit MP_processor_info( disabled_cpus++; return; } -#ifdef CONFIG_X86_NUMAQ - apicid = mpc_apic_id(m, translation_table[mpc_record]); -#else - apicid = m->mpc_apicid; -#endif + + if (x86_quirks->mpc_apic_id) + apicid = x86_quirks->mpc_apic_id(m); + else + apicid = m->mpc_apicid; + if (m->mpc_cpuflag & CPU_BOOTPROCESSOR) { bootup_cpu = " (Bootup-CPU)"; boot_cpu_physical_apicid = m->mpc_apicid; @@ -107,18 +82,17 @@ static void __cpuinit MP_processor_info( #endif } +#ifdef CONFIG_X86_IO_APIC static void __init MP_bus_info(struct mpc_config_bus *m) { char str[7]; - memcpy(str, m->mpc_bustype, 6); str[6] = 0; -#ifdef CONFIG_X86_NUMAQ - mpc_oem_bus_info(m, str, translation_table[mpc_record]); -#else - Dprintk("Bus #%d is %s\n", m->mpc_busid, str); -#endif + if (x86_quirks->mpc_oem_bus_info) + x86_quirks->mpc_oem_bus_info(m, str); + else + apic_printk(APIC_VERBOSE, "Bus #%d is %s\n", m->mpc_busid, str); #if MAX_MP_BUSSES < 256 if (m->mpc_busid >= MAX_MP_BUSSES) { @@ -135,12 +109,10 @@ static void __init MP_bus_info(struct mp mp_bus_id_to_type[m->mpc_busid] = MP_BUS_ISA; #endif } else if (strncmp(str, BUSTYPE_PCI, sizeof(BUSTYPE_PCI) - 1) == 0) { -#ifdef CONFIG_X86_NUMAQ - mpc_oem_pci_bus(m, translation_table[mpc_record]); -#endif + if (x86_quirks->mpc_oem_pci_bus) + x86_quirks->mpc_oem_pci_bus(m); + clear_bit(m->mpc_busid, mp_bus_not_pci); - mp_bus_id_to_pci_bus[m->mpc_busid] = mp_current_pci_id; - mp_current_pci_id++; #if defined(CONFIG_EISA) || defined (CONFIG_MCA) mp_bus_id_to_type[m->mpc_busid] = MP_BUS_PCI; } else if (strncmp(str, BUSTYPE_EISA, sizeof(BUSTYPE_EISA) - 1) == 0) { @@ -151,6 +123,7 @@ static void __init MP_bus_info(struct mp } else printk(KERN_WARNING "Unknown bustype %s - ignoring\n", str); } +#endif #ifdef CONFIG_X86_IO_APIC @@ -180,117 +153,111 @@ static void __init MP_ioapic_info(struct if (bad_ioapic(m->mpc_apicaddr)) return; - mp_ioapics[nr_ioapics] = *m; + mp_ioapics[nr_ioapics].mp_apicaddr = m->mpc_apicaddr; + mp_ioapics[nr_ioapics].mp_apicid = m->mpc_apicid; + mp_ioapics[nr_ioapics].mp_type = m->mpc_type; + mp_ioapics[nr_ioapics].mp_apicver = m->mpc_apicver; + mp_ioapics[nr_ioapics].mp_flags = m->mpc_flags; nr_ioapics++; } -static void __init MP_intsrc_info(struct mpc_config_intsrc *m) +static void print_MP_intsrc_info(struct mpc_config_intsrc *m) { - mp_irqs[mp_irq_entries] = *m; - Dprintk("Int: type %d, pol %d, trig %d, bus %d," + apic_printk(APIC_VERBOSE, "Int: type %d, pol %d, trig %d, bus %02x," " 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"); } -#endif - -static void __init MP_lintsrc_info(struct mpc_config_lintsrc *m) +static void __init print_mp_irq_info(struct mp_config_intsrc *mp_irq) { - 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); + apic_printk(APIC_VERBOSE, "Int: type %d, pol %d, trig %d, bus %02x," + " IRQ %02x, APIC ID %x, APIC INT %02x\n", + mp_irq->mp_irqtype, mp_irq->mp_irqflag & 3, + (mp_irq->mp_irqflag >> 2) & 3, mp_irq->mp_srcbus, + mp_irq->mp_srcbusirq, mp_irq->mp_dstapic, mp_irq->mp_dstirq); } -#ifdef CONFIG_X86_NUMAQ -static void __init MP_translation_info(struct mpc_config_translation *m) +static void __init assign_to_mp_irq(struct mpc_config_intsrc *m, + struct mp_config_intsrc *mp_irq) { - 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); + mp_irq->mp_dstapic = m->mpc_dstapic; + mp_irq->mp_type = m->mpc_type; + mp_irq->mp_irqtype = m->mpc_irqtype; + mp_irq->mp_irqflag = m->mpc_irqflag; + mp_irq->mp_srcbus = m->mpc_srcbus; + mp_irq->mp_srcbusirq = m->mpc_srcbusirq; + mp_irq->mp_dstirq = m->mpc_dstirq; +} - 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); +static void __init assign_to_mpc_intsrc(struct mp_config_intsrc *mp_irq, + struct mpc_config_intsrc *m) +{ + m->mpc_dstapic = mp_irq->mp_dstapic; + m->mpc_type = mp_irq->mp_type; + m->mpc_irqtype = mp_irq->mp_irqtype; + m->mpc_irqflag = mp_irq->mp_irqflag; + m->mpc_srcbus = mp_irq->mp_srcbus; + m->mpc_srcbusirq = mp_irq->mp_srcbusirq; + m->mpc_dstirq = mp_irq->mp_dstirq; } -/* - * Read/parse the MPC oem tables - */ +static int __init mp_irq_mpc_intsrc_cmp(struct mp_config_intsrc *mp_irq, + struct mpc_config_intsrc *m) +{ + if (mp_irq->mp_dstapic != m->mpc_dstapic) + return 1; + if (mp_irq->mp_type != m->mpc_type) + return 2; + if (mp_irq->mp_irqtype != m->mpc_irqtype) + return 3; + if (mp_irq->mp_irqflag != m->mpc_irqflag) + return 4; + if (mp_irq->mp_srcbus != m->mpc_srcbus) + return 5; + if (mp_irq->mp_srcbusirq != m->mpc_srcbusirq) + return 6; + if (mp_irq->mp_dstirq != m->mpc_dstirq) + return 7; + + return 0; +} -static void __init smp_read_mpc_oem(struct mp_config_oemtable *oemtable, - unsigned short oemsize) +static void __init MP_intsrc_info(struct mpc_config_intsrc *m) { - int count = sizeof(*oemtable); /* the header size */ - unsigned char *oemptr = ((unsigned char *)oemtable) + count; + int i; - 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; - } - } + print_MP_intsrc_info(m); + + for (i = 0; i < mp_irq_entries; i++) { + if (!mp_irq_mpc_intsrc_cmp(&mp_irqs[i], m)) + return; } + + assign_to_mp_irq(m, &mp_irqs[mp_irq_entries]); + if (++mp_irq_entries == MAX_IRQ_SOURCES) + panic("Max # of irq sources exceeded!!\n"); } -static inline void mps_oem_check(struct mp_config_table *mpc, char *oem, - char *productid) +#endif + +static void __init MP_lintsrc_info(struct mpc_config_lintsrc *m) { - 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); + apic_printk(APIC_VERBOSE, "Lint: type %d, pol %d, trig %d, bus %02x," + " 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); } -#endif /* CONFIG_X86_NUMAQ */ /* * Read/parse the MPC */ -static int __init smp_read_mpc(struct mp_config_table *mpc, unsigned early) +static int __init smp_check_mpc(struct mp_config_table *mpc, char *oem, + char *str) { - 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 "MPTABLE: bad signature [%c%c%c%c]!\n", @@ -313,20 +280,44 @@ static int __init smp_read_mpc(struct mp } memcpy(oem, mpc->mpc_oem, 8); oem[8] = 0; - printk(KERN_INFO "MPTABLE: OEM ID: %s ", oem); + printk(KERN_INFO "MPTABLE: OEM ID: %s\n", oem); memcpy(str, mpc->mpc_productid, 12); str[12] = 0; - printk("Product ID: %s ", str); -#ifdef CONFIG_X86_32 - mps_oem_check(mpc, oem, str); -#endif - printk(KERN_INFO "MPTABLE: Product ID: %s ", str); + printk(KERN_INFO "MPTABLE: Product ID: %s\n", str); #ifndef CONFIG_XEN printk(KERN_INFO "MPTABLE: APIC at: 0x%X\n", mpc->mpc_lapic); +#endif + + return 1; +} + +static int __init smp_read_mpc(struct mp_config_table *mpc, unsigned early) +{ + char str[16]; + char oem[10]; + + int count = sizeof(*mpc); + unsigned char *mpt = ((unsigned char *)mpc) + count; + + if (!smp_check_mpc(mpc, oem, str)) + return 0; +#ifdef CONFIG_X86_32 + /* + * need to make sure summit and es7000's mps_oem_check is safe to be + * called early via genericarch 's mps_oem_check + */ + if (early) { +#ifdef CONFIG_X86_NUMAQ + numaq_mps_oem_check(mpc, oem, str); +#endif + } else + mps_oem_check(mpc, oem, str); +#endif +#ifndef CONFIG_XEN /* save the local APIC address, it might be non-default */ if (!acpi_lapic) mp_lapic_addr = mpc->mpc_lapic; @@ -335,12 +326,17 @@ static int __init smp_read_mpc(struct mp if (early) return 1; + if (mpc->mpc_oemptr && x86_quirks->smp_read_mpc_oem) { + struct mp_config_oemtable *oem_table = (struct mp_config_oemtable *)(unsigned long)mpc->mpc_oemptr; + x86_quirks->smp_read_mpc_oem(oem_table, mpc->mpc_oemsize); + } + /* * Now process the configuration blocks. */ -#ifdef CONFIG_X86_NUMAQ - mpc_record = 0; -#endif + if (x86_quirks->mpc_record) + *x86_quirks->mpc_record = 0; + while (count < mpc->mpc_length) { switch (*mpt) { case MP_PROCESSOR: @@ -358,7 +354,9 @@ static int __init smp_read_mpc(struct mp { struct mpc_config_bus *m = (struct mpc_config_bus *)mpt; +#ifdef CONFIG_X86_IO_APIC MP_bus_info(m); +#endif mpt += sizeof(*m); count += sizeof(*m); break; @@ -404,10 +402,14 @@ static int __init smp_read_mpc(struct mp count = mpc->mpc_length; break; } -#ifdef CONFIG_X86_NUMAQ - ++mpc_record; -#endif + if (x86_quirks->mpc_record) + (*x86_quirks->mpc_record)++; } + +#ifdef CONFIG_X86_GENERICARCH + generic_bigsmp_probe(); +#endif + setup_apic_routing(); if (!num_processors) printk(KERN_ERR "MPTABLE: no processors registered!\n"); @@ -433,7 +435,7 @@ static void __init construct_default_ioi intsrc.mpc_type = MP_INTSRC; intsrc.mpc_irqflag = 0; /* conforming */ intsrc.mpc_srcbus = 0; - intsrc.mpc_dstapic = mp_ioapics[0].mpc_apicid; + intsrc.mpc_dstapic = mp_ioapics[0].mp_apicid; intsrc.mpc_irqtype = mp_INT; @@ -494,42 +496,11 @@ static void __init construct_default_ioi MP_intsrc_info(&intsrc); } -#endif -static inline void __init construct_default_ISA_mptable(int mpc_default_type) +static void __init construct_ioapic_table(int mpc_default_type) { - struct mpc_config_processor processor; - struct mpc_config_bus bus; -#ifdef CONFIG_X86_IO_APIC struct mpc_config_ioapic ioapic; -#endif - struct mpc_config_lintsrc lintsrc; - int linttypes[2] = { mp_ExtINT, mp_NMI }; - int i; - -#ifndef CONFIG_XEN - /* - * local APIC has default address - */ - mp_lapic_addr = APIC_DEFAULT_PHYS_BASE; -#endif - - /* - * 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); - } + struct mpc_config_bus bus; bus.mpc_type = MP_BUS; bus.mpc_busid = 0; @@ -558,7 +529,6 @@ static inline void __init construct_defa MP_bus_info(&bus); } -#ifdef CONFIG_X86_IO_APIC ioapic.mpc_type = MP_IOAPIC; ioapic.mpc_apicid = 2; ioapic.mpc_apicver = mpc_default_type > 4 ? 0x10 : 0x01; @@ -570,7 +540,44 @@ static inline void __init construct_defa * We set up most of the low 16 IO-APIC pins according to MPS rules. */ construct_default_ioirq_mptable(mpc_default_type); +} +#else +static inline void __init construct_ioapic_table(int mpc_default_type) { } +#endif + +static inline void __init construct_default_ISA_mptable(int mpc_default_type) +{ + struct mpc_config_processor processor; + struct mpc_config_lintsrc lintsrc; + int linttypes[2] = { mp_ExtINT, mp_NMI }; + int i; + +#ifndef CONFIG_XEN + /* + * local APIC has default address + */ + mp_lapic_addr = APIC_DEFAULT_PHYS_BASE; #endif + + /* + * 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); + } + + construct_ioapic_table(mpc_default_type); + lintsrc.mpc_type = MP_LINTSRC; lintsrc.mpc_irqflag = 0; /* conforming */ lintsrc.mpc_srcbusid = 0; @@ -589,7 +596,7 @@ static struct intel_mp_floating *mpf_fou * Scan the memory blocks for an SMP configuration block. */ #ifndef CONFIG_XEN -static void __init __get_smp_config(unsigned early) +static void __init __get_smp_config(unsigned int early) #else void __init get_smp_config(void) #define early 0 @@ -597,6 +604,10 @@ void __init get_smp_config(void) { struct intel_mp_floating *mpf = mpf_found; + if (x86_quirks->mach_get_smp_config) { + if (x86_quirks->mach_get_smp_config(early)) + return; + } if (acpi_lapic && early) return; /* @@ -613,7 +624,7 @@ void __init get_smp_config(void) printk(KERN_INFO "Intel MultiProcessor Specification v1.%d\n", mpf->mpf_specification); -#ifdef CONFIG_X86_32 +#if defined(CONFIG_X86_LOCAL_APIC) && defined(CONFIG_X86_32) && !defined(CONFIG_XEN) if (mpf->mpf_feature2 & (1 << 7)) { printk(KERN_INFO " IMCR and PIC compatibility mode.\n"); pic_mode = 1; @@ -646,8 +657,10 @@ void __init get_smp_config(void) * Read the physical hardware table. Anything here will * override the defaults. */ - if (!smp_read_mpc(isa_bus_to_virt(mpf->mpf_physptr), early)) { + if (!smp_read_mpc(_bus_to_virt(mpf->mpf_physptr), early)) { +#ifdef CONFIG_X86_LOCAL_APIC smp_found_config = 0; +#endif printk(KERN_ERR "BIOS bug, MP table errors detected!...\n"); printk(KERN_ERR "... disabling SMP support. " @@ -704,10 +717,11 @@ void __init get_smp_config(void) static int __init smp_scan_config(unsigned long base, unsigned long length, unsigned reserve) { - unsigned int *bp = isa_bus_to_virt(base); + unsigned int *bp = _bus_to_virt(base); struct intel_mp_floating *mpf; - Dprintk("Scan SMP from %p for %ld bytes.\n", bp, length); + apic_printk(APIC_VERBOSE, "Scan SMP from %p for %ld bytes.\n", + bp, length); BUILD_BUG_ON(sizeof(*mpf) != 16); while (length > 0) { @@ -717,16 +731,22 @@ static int __init smp_scan_config(unsign !mpf_checksum((unsigned char *)bp, 16) && ((mpf->mpf_specification == 1) || (mpf->mpf_specification == 4))) { - +#ifdef CONFIG_X86_LOCAL_APIC smp_found_config = 1; +#endif mpf_found = mpf; -#ifdef CONFIG_X86_32 + #ifndef CONFIG_XEN printk(KERN_INFO "found SMP MP-table at [%p] %08lx\n", mpf, virt_to_phys(mpf)); - reserve_bootmem(virt_to_phys(mpf), PAGE_SIZE, + + if (!reserve) + return 1; + reserve_bootmem_generic(virt_to_phys(mpf), PAGE_SIZE, BOOTMEM_DEFAULT); if (mpf->mpf_physptr) { + unsigned long size = PAGE_SIZE; +#ifdef CONFIG_X86_32 /* * We cannot access to MPC table to compute * table size yet, as only few megabytes from @@ -736,27 +756,18 @@ static int __init smp_scan_config(unsign * 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, +#endif + reserve_bootmem_generic(mpf->mpf_physptr, size, BOOTMEM_DEFAULT); } #else printk(KERN_INFO "found SMP MP-table at [%p] %08lx\n", - mpf, ((void *)bp - isa_bus_to_virt(base)) + base); -#endif -#elif !defined(CONFIG_XEN) - if (!reserve) - return 1; - - reserve_bootmem_generic(virt_to_phys(mpf), PAGE_SIZE); - if (mpf->mpf_physptr) - reserve_bootmem_generic(mpf->mpf_physptr, - PAGE_SIZE); + mpf, ((void *)bp - _bus_to_virt(base)) + base); #endif - return 1; + return 1; } bp += 4; length -= 16; @@ -764,12 +775,16 @@ static int __init smp_scan_config(unsign return 0; } -static void __init __find_smp_config(unsigned reserve) +static void __init __find_smp_config(unsigned int reserve) { #ifndef CONFIG_XEN unsigned int address; #endif + if (x86_quirks->mach_find_smp_config) { + if (x86_quirks->mach_find_smp_config(reserve)) + return; + } /* * FIXME: Linux assumes you have 640K of base ram.. * this continues the error... @@ -816,302 +831,297 @@ void __init find_smp_config(void) __find_smp_config(1); } -/* -------------------------------------------------------------------------- - ACPI-based MP Configuration - -------------------------------------------------------------------------- */ - -/* - * Keep this outside and initialized to 0, for !CONFIG_ACPI builds: - */ -int es7000_plat; - -#ifdef CONFIG_ACPI +#ifdef CONFIG_X86_IO_APIC +static u8 __initdata irq_used[MAX_IRQ_SOURCES]; -#ifdef CONFIG_X86_IO_APIC +static int __init get_MP_intsrc_index(struct mpc_config_intsrc *m) +{ + int i; -#define MP_ISA_BUS 0 + if (m->mpc_irqtype != mp_INT) + return 0; -extern struct mp_ioapic_routing mp_ioapic_routing[MAX_IO_APICS]; + if (m->mpc_irqflag != 0x0f) + return 0; -static int mp_find_ioapic(int gsi) -{ - int i = 0; + /* not legacy */ - /* 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; + for (i = 0; i < mp_irq_entries; i++) { + if (mp_irqs[i].mp_irqtype != mp_INT) + continue; + + if (mp_irqs[i].mp_irqflag != 0x0f) + continue; + + if (mp_irqs[i].mp_srcbus != m->mpc_srcbus) + continue; + if (mp_irqs[i].mp_srcbusirq != m->mpc_srcbusirq) + continue; + if (irq_used[i]) { + /* already claimed */ + return -2; + } + irq_used[i] = 1; + return i; } - printk(KERN_ERR "ERROR: Unable to locate IOAPIC for GSI %d\n", gsi); + /* not found */ return -1; } -static u8 __init uniq_ioapic_id(u8 id) -{ -#ifdef CONFIG_X86_32 -#ifndef CONFIG_XEN - if ((boot_cpu_data.x86_vendor == X86_VENDOR_INTEL) && - !APIC_XAPIC(apic_version[boot_cpu_physical_apicid])) - return io_apic_get_unique_id(nr_ioapics, id); - else -#endif - return id; -#else - int i; - DECLARE_BITMAP(used, 256); - bitmap_zero(used, 256); - for (i = 0; i < nr_ioapics; i++) { - struct mpc_config_ioapic *ia = &mp_ioapics[i]; - __set_bit(ia->mpc_apicid, used); - } - if (!test_bit(id, used)) - return id; - return find_first_zero_bit(used, 256); +#define SPARE_SLOT_NUM 20 + +static struct mpc_config_intsrc __initdata *m_spare[SPARE_SLOT_NUM]; #endif -} -void __init mp_register_ioapic(int id, u32 address, u32 gsi_base) +static int __init replace_intsrc_all(struct mp_config_table *mpc, + unsigned long mpc_new_phys, + unsigned long mpc_new_length) { - int idx = 0; - - if (bad_ioapic(address)) - return; +#ifdef CONFIG_X86_IO_APIC + int i; + int nr_m_spare = 0; +#endif - idx = nr_ioapics; + int count = sizeof(*mpc); + unsigned char *mpt = ((unsigned char *)mpc) + count; - mp_ioapics[idx].mpc_type = MP_IOAPIC; - mp_ioapics[idx].mpc_flags = MPC_APIC_USABLE; - mp_ioapics[idx].mpc_apicaddr = address; + printk(KERN_INFO "mpc_length %x\n", mpc->mpc_length); + while (count < mpc->mpc_length) { + switch (*mpt) { + case MP_PROCESSOR: + { + struct mpc_config_processor *m = + (struct mpc_config_processor *)mpt; + mpt += sizeof(*m); + count += sizeof(*m); + break; + } + case MP_BUS: + { + struct mpc_config_bus *m = + (struct mpc_config_bus *)mpt; + mpt += sizeof(*m); + count += sizeof(*m); + break; + } + case MP_IOAPIC: + { + mpt += sizeof(struct mpc_config_ioapic); + count += sizeof(struct mpc_config_ioapic); + break; + } + case MP_INTSRC: + { +#ifdef CONFIG_X86_IO_APIC + struct mpc_config_intsrc *m = + (struct mpc_config_intsrc *)mpt; -#ifndef CONFIG_XEN - set_fixmap_nocache(FIX_IO_APIC_BASE_0 + idx, address); + apic_printk(APIC_VERBOSE, "OLD "); + print_MP_intsrc_info(m); + i = get_MP_intsrc_index(m); + if (i > 0) { + assign_to_mpc_intsrc(&mp_irqs[i], m); + apic_printk(APIC_VERBOSE, "NEW "); + print_mp_irq_info(&mp_irqs[i]); + } else if (!i) { + /* legacy, do nothing */ + } else if (nr_m_spare < SPARE_SLOT_NUM) { + /* + * not found (-1), or duplicated (-2) + * are invalid entries, + * we need to use the slot later + */ + m_spare[nr_m_spare] = m; + nr_m_spare++; + } #endif - mp_ioapics[idx].mpc_apicid = uniq_ioapic_id(id); -#ifdef CONFIG_X86_32 - mp_ioapics[idx].mpc_apicver = io_apic_get_version(idx); -#else - mp_ioapics[idx].mpc_apicver = 0; + mpt += sizeof(struct mpc_config_intsrc); + count += sizeof(struct mpc_config_intsrc); + break; + } + case MP_LINTSRC: + { + struct mpc_config_lintsrc *m = + (struct mpc_config_lintsrc *)mpt; + mpt += sizeof(*m); + count += sizeof(*m); + break; + } + default: + /* wrong mptable */ + printk(KERN_ERR "Your mptable is wrong, contact your HW vendor!\n"); + printk(KERN_ERR "type %x\n", *mpt); + print_hex_dump(KERN_ERR, " ", DUMP_PREFIX_ADDRESS, 16, + 1, mpc, mpc->mpc_length, 1); + goto out; + } + } + +#ifdef CONFIG_X86_IO_APIC + for (i = 0; i < mp_irq_entries; i++) { + if (irq_used[i]) + continue; + + if (mp_irqs[i].mp_irqtype != mp_INT) + continue; + + if (mp_irqs[i].mp_irqflag != 0x0f) + continue; + + if (nr_m_spare > 0) { + apic_printk(APIC_VERBOSE, "*NEW* found\n"); + nr_m_spare--; + assign_to_mpc_intsrc(&mp_irqs[i], m_spare[nr_m_spare]); + m_spare[nr_m_spare] = NULL; + } else { + struct mpc_config_intsrc *m = + (struct mpc_config_intsrc *)mpt; + count += sizeof(struct mpc_config_intsrc); + if (!mpc_new_phys) { + printk(KERN_INFO "No spare slots, try to append...take your risk, new mpc_length %x\n", count); + } else { + if (count <= mpc_new_length) + printk(KERN_INFO "No spare slots, try to append..., new mpc_length %x\n", count); + else { + printk(KERN_ERR "mpc_new_length %lx is too small\n", mpc_new_length); + goto out; + } + } + assign_to_mpc_intsrc(&mp_irqs[i], m); + mpc->mpc_length = count; + mpt += sizeof(struct mpc_config_intsrc); + } + print_mp_irq_info(&mp_irqs[i]); + } #endif - /* - * 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(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_base, mp_ioapic_routing[idx].gsi_end); +out: + /* update checksum */ + mpc->mpc_checksum = 0; + mpc->mpc_checksum -= mpf_checksum((unsigned char *)mpc, + mpc->mpc_length); - nr_ioapics++; + return 0; } -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; +static int __initdata enable_update_mptable; - /* - * 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; +static int __init update_mptable_setup(char *str) +{ + enable_update_mptable = 1; + return 0; +} +early_param("update_mptable", update_mptable_setup); - 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# */ +static unsigned long __initdata mpc_new_phys; +static unsigned long mpc_new_length __initdata = 4096; - MP_intsrc_info(&intsrc); +/* alloc_mptable or alloc_mptable=4k */ +static int __initdata alloc_mptable; +static int __init parse_alloc_mptable_opt(char *p) +{ + enable_update_mptable = 1; + alloc_mptable = 1; + if (!p) + return 0; + mpc_new_length = memparse(p, &p); + return 0; } +early_param("alloc_mptable", parse_alloc_mptable_opt); -void __init mp_config_acpi_legacy_irqs(void) +void __init early_reserve_e820_mpc_new(void) { - struct mpc_config_intsrc intsrc; - int i = 0; - int ioapic = -1; - -#if defined (CONFIG_MCA) || defined (CONFIG_EISA) - /* - * Fabricate the legacy ISA bus (bus #31). - */ - mp_bus_id_to_type[MP_ISA_BUS] = MP_BUS_ISA; -#endif - set_bit(MP_ISA_BUS, mp_bus_not_pci); - 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; -#ifdef CONFIG_X86_IO_APIC - intsrc.mpc_dstapic = mp_ioapics[ioapic].mpc_apicid; + if (enable_update_mptable && alloc_mptable) { + u64 startt = 0; +#ifdef CONFIG_X86_TRAMPOLINE + startt = TRAMPOLINE_BASE; #endif - /* - * 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; - - MP_intsrc_info(&intsrc); + mpc_new_phys = early_reserve_e820(startt, mpc_new_length, 4); } } -int mp_register_gsi(u32 gsi, int triggering, int polarity) +static int __init update_mp_table(void) { - int ioapic; - int ioapic_pin; -#if defined(CONFIG_X86_32) && !defined(CONFIG_XEN) -#define MAX_GSI_NUM 4096 -#define IRQ_COMPRESSION_START 64 + char str[16]; + char oem[10]; + struct intel_mp_floating *mpf; + struct mp_config_table *mpc; + struct mp_config_table *mpc_new; + + if (!enable_update_mptable) + return 0; + + mpf = mpf_found; + if (!mpf) + return 0; - static int pci_irq = IRQ_COMPRESSION_START; /* - * Mapping between Global System Interrupts, which - * represent all possible interrupts, and IRQs - * assigned to actual devices. + * Now see if we need to go further. */ - static int gsi_to_irq[MAX_GSI_NUM]; -#else + if (mpf->mpf_feature1 != 0) + return 0; - if (acpi_irq_model != ACPI_IRQ_MODEL_IOAPIC) - return gsi; -#endif + if (!mpf->mpf_physptr) + return 0; - /* Don't set up the ACPI SCI because it's already set up */ - if (acpi_gbl_FADT.sci_interrupt == gsi) - return gsi; + mpc = _bus_to_virt(mpf->mpf_physptr); - ioapic = mp_find_ioapic(gsi); - if (ioapic < 0) { - printk(KERN_WARNING "No IOAPIC for GSI %u\n", gsi); - return gsi; - } + if (!smp_check_mpc(mpc, oem, str)) + return 0; - ioapic_pin = gsi - mp_ioapic_routing[ioapic].gsi_base; + printk(KERN_INFO "mpf: %lx\n", (long)arbitrary_virt_to_machine(mpf)); + printk(KERN_INFO "mpf_physptr: %x\n", mpf->mpf_physptr); -#ifndef CONFIG_X86_32 - if (ioapic_renumber_irq) - gsi = ioapic_renumber_irq(ioapic, gsi); -#endif + if (mpc_new_phys && mpc->mpc_length > mpc_new_length) { + mpc_new_phys = 0; + printk(KERN_INFO "mpc_new_length is %ld, please use alloc_mptable=8k\n", + mpc_new_length); + } + + if (!mpc_new_phys) { + unsigned char old, new; + /* check if we can change the postion */ + mpc->mpc_checksum = 0; + old = mpf_checksum((unsigned char *)mpc, mpc->mpc_length); + mpc->mpc_checksum = 0xff; + new = mpf_checksum((unsigned char *)mpc, mpc->mpc_length); + if (old == new) { + printk(KERN_INFO "mpc is readonly, please try alloc_mptable instead\n"); + return 0; + } + printk(KERN_INFO "use in-positon replacing\n"); + } else { + maddr_t mpc_new_bus; - /* - * Avoid pin reprogramming. PRTs typically include entries - * with redundant pin->gsi mappings (but unique PCI devices); - * we only program the IOAPIC on the first. - */ - if (ioapic_pin > MP_MAX_IOAPIC_PIN) { - printk(KERN_ERR "Invalid reference to IOAPIC pin " - "%d-%d\n", mp_ioapic_routing[ioapic].apic_id, - ioapic_pin); - return gsi; - } - if (test_bit(ioapic_pin, mp_ioapic_routing[ioapic].pin_programmed)) { - Dprintk(KERN_DEBUG "Pin %d-%d already programmed\n", - mp_ioapic_routing[ioapic].apic_id, ioapic_pin); -#if defined(CONFIG_X86_32) && !defined(CONFIG_XEN) - return (gsi < IRQ_COMPRESSION_START ? gsi : gsi_to_irq[gsi]); -#else - return gsi; -#endif + mpc_new_bus = phys_to_machine(mpc_new_phys); + mpf->mpf_physptr = mpc_new_bus; + mpc_new = phys_to_virt(mpc_new_phys); + memcpy(mpc_new, mpc, mpc->mpc_length); + mpc = mpc_new; + /* check if we can modify that */ + if (mpc_new_bus - mpf->mpf_physptr) { + struct intel_mp_floating *mpf_new; + /* steal 16 bytes from [0, 1k) */ + printk(KERN_INFO "mpf new: %x\n", 0x400 - 16); + mpf_new = isa_bus_to_virt(0x400 - 16); + memcpy(mpf_new, mpf, 16); + mpf = mpf_new; + mpf->mpf_physptr = mpc_new_bus; + } + mpf->mpf_checksum = 0; + mpf->mpf_checksum -= mpf_checksum((unsigned char *)mpf, 16); + printk(KERN_INFO "mpf_physptr new: %x\n", mpf->mpf_physptr); } - set_bit(ioapic_pin, mp_ioapic_routing[ioapic].pin_programmed); -#if defined(CONFIG_X86_32) && !defined(CONFIG_XEN) /* - * For GSI >= 64, use IRQ compression + * only replace the one with mp_INT and + * MP_IRQ_TRIGGER_LEVEL|MP_IRQ_POLARITY_LOW, + * already in mp_irqs , stored by ... and mp_config_acpi_gsi, + * may need pci=routeirq for all coverage */ - if ((gsi >= IRQ_COMPRESSION_START) - && (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. - */ - gsi = pci_irq++; - /* - * Don't assign IRQ used by ACPI SCI - */ - if (gsi == acpi_gbl_FADT.sci_interrupt) - gsi = pci_irq++; - gsi_to_irq[irq] = gsi; - } else { - printk(KERN_ERR "GSI %u is too high\n", gsi); - return gsi; - } - } -#endif - io_apic_set_pci_routing(ioapic, ioapic_pin, gsi, - triggering == ACPI_EDGE_SENSITIVE ? 0 : 1, - polarity == ACPI_ACTIVE_HIGH ? 0 : 1); - return gsi; + replace_intsrc_all(mpc, mpc_new_phys, mpc_new_length); + + return 0; } -#endif /* CONFIG_X86_IO_APIC */ -#endif /* CONFIG_ACPI */ +late_initcall(update_mp_table); --- head-2011-03-11.orig/arch/x86/kernel/pci-dma-xen.c 2011-01-31 18:07:35.000000000 +0100 +++ head-2011-03-11/arch/x86/kernel/pci-dma-xen.c 2011-02-01 14:38:38.000000000 +0100 @@ -5,13 +5,13 @@ #include #include -#include +#include #include +#include -int forbid_dac __read_mostly; -EXPORT_SYMBOL(forbid_dac); +static int forbid_dac __read_mostly; -const struct dma_mapping_ops *dma_ops; +struct dma_mapping_ops *dma_ops; EXPORT_SYMBOL(dma_ops); static int iommu_sac_force __read_mostly; @@ -74,13 +74,17 @@ early_param("dma32_size", parse_dma32_si void __init dma32_reserve_bootmem(void) { unsigned long size, align; - if (end_pfn <= MAX_DMA32_PFN) + if (max_pfn <= MAX_DMA32_PFN) return; + /* + * check aperture_64.c allocate_aperture() for reason about + * using 512M as goal + */ align = 64ULL<<20; size = round_up(dma32_bootmem_size, align); dma32_bootmem_ptr = __alloc_bootmem_nopanic(size, align, - __pa(MAX_DMA_ADDRESS)); + 512ULL<<20); if (dma32_bootmem_ptr) dma32_bootmem_size = size; else @@ -88,17 +92,14 @@ void __init dma32_reserve_bootmem(void) } static void __init dma32_free_bootmem(void) { - int node; - if (end_pfn <= MAX_DMA32_PFN) + if (max_pfn <= MAX_DMA32_PFN) return; if (!dma32_bootmem_ptr) return; - for_each_online_node(node) - free_bootmem_node(NODE_DATA(node), __pa(dma32_bootmem_ptr), - dma32_bootmem_size); + free_bootmem(__pa(dma32_bootmem_ptr), dma32_bootmem_size); dma32_bootmem_ptr = NULL; dma32_bootmem_size = 0; @@ -107,7 +108,7 @@ static void __init dma32_free_bootmem(vo #define dma32_free_bootmem() ((void)0) #endif -static const struct dma_mapping_ops swiotlb_dma_ops = { +static struct dma_mapping_ops swiotlb_dma_ops = { .mapping_error = swiotlb_dma_mapping_error, .map_single = swiotlb_map_single_phys, .unmap_single = swiotlb_unmap_single, @@ -130,25 +131,31 @@ void __init pci_iommu_alloc(void) * The order of these functions is important for * fall-back/fail-over reasons */ -#ifdef CONFIG_GART_IOMMU gart_iommu_hole_init(); -#endif -#ifdef CONFIG_CALGARY_IOMMU detect_calgary(); -#endif detect_intel_iommu(); -#ifdef CONFIG_SWIOTLB + amd_iommu_detect(); + swiotlb_init(); if (swiotlb) { printk(KERN_INFO "PCI-DMA: Using software bounce buffering for IO (SWIOTLB)\n"); dma_ops = &swiotlb_dma_ops; } -#endif } +#ifndef CONFIG_XEN +unsigned long iommu_num_pages(unsigned long addr, unsigned long len) +{ + unsigned long size = roundup((addr & ~PAGE_MASK) + len, PAGE_SIZE); + + return size >> PAGE_SHIFT; +} +EXPORT_SYMBOL(iommu_num_pages); +#endif + /* * See for the iommu kernel parameter * documentation. @@ -201,9 +208,7 @@ static __init int iommu_setup(char *p) swiotlb = 1; #endif -#ifdef CONFIG_GART_IOMMU gart_parse_options(p); -#endif #ifdef CONFIG_CALGARY_IOMMU if (!strncmp(p, "calgary", 7)) @@ -245,136 +250,19 @@ int range_straddles_page_boundary(paddr_ !check_pages_physically_contiguous(pfn, offset, size)); } -#ifdef CONFIG_X86_32 -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 = NULL; - int pages = size >> PAGE_SHIFT; - int bitmap_size = BITS_TO_LONGS(pages) * sizeof(long); - - 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 = kzalloc(sizeof(struct dma_coherent_mem), GFP_KERNEL); - if (!dev->dma_mem) - goto out; - dev->dma_mem->bitmap = kzalloc(bitmap_size, GFP_KERNEL); - if (!dev->dma_mem->bitmap) - goto free1_out; - - 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); - out: - if (mem_base) - iounmap(mem_base); - 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 pos, err; - int pages = (size + (device_addr & ~PAGE_MASK) + PAGE_SIZE - 1); - - pages >>= PAGE_SHIFT; - - 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); - -static int dma_alloc_from_coherent_mem(struct device *dev, ssize_t size, - dma_addr_t *dma_handle, void **ret) -{ - struct dma_coherent_mem *mem = dev ? dev->dma_mem : NULL; - int order = get_order(size); - - 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); - } - if (mem->flags & DMA_MEMORY_EXCLUSIVE) - *ret = NULL; - } - return (mem != NULL); -} - -static int dma_release_coherent(struct device *dev, int order, void *vaddr) -{ - struct dma_coherent_mem *mem = dev ? dev->dma_mem : NULL; - - 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); - return 1; - } - return 0; -} -#else -#define dma_alloc_from_coherent_mem(dev, size, handle, ret) (0) -#define dma_release_coherent(dev, order, vaddr) (0) -#endif /* CONFIG_X86_32 */ - int dma_supported(struct device *dev, u64 mask) { + struct dma_mapping_ops *ops = get_dma_ops(dev); + #ifdef CONFIG_PCI if (mask > 0xffffffff && forbid_dac > 0) { - printk(KERN_INFO "PCI: Disallowing DAC for device %s\n", - dev->bus_id); + dev_info(dev, "PCI: Disallowing DAC for device\n"); return 0; } #endif - if (dma_ops->dma_supported) - return dma_ops->dma_supported(dev, mask); + if (ops->dma_supported) + return ops->dma_supported(dev, mask); /* Copied from i386. Doesn't make much sense, because it will only work for pci_alloc_coherent. @@ -395,8 +283,7 @@ int dma_supported(struct device *dev, u6 type. Normally this doesn't make any difference, but gives more gentle handling of IOMMU overflow. */ if (iommu_sac_force && (mask >= DMA_40BIT_MASK)) { - printk(KERN_INFO "%s: Force SAC with mask %Lx\n", - dev->bus_id, mask); + dev_info(dev, "Force SAC with mask %Lx\n", mask); return 0; } @@ -422,6 +309,9 @@ void * dma_alloc_coherent(struct device *dev, size_t size, dma_addr_t *dma_handle, gfp_t gfp) { +#ifndef CONFIG_XEN + struct dma_mapping_ops *ops = get_dma_ops(dev); +#endif void *memory = NULL; struct page *page; unsigned long dma_mask = 0; @@ -431,7 +321,7 @@ dma_alloc_coherent(struct device *dev, s /* ignore region specifiers */ gfp &= ~(__GFP_DMA | __GFP_HIGHMEM | __GFP_DMA32); - if (dma_alloc_from_coherent_mem(dev, size, dma_handle, &memory)) + if (dma_alloc_from_coherent(dev, size, dma_handle, &memory)) return memory; if (!dev) { @@ -491,8 +381,8 @@ dma_alloc_coherent(struct device *dev, s /* Let low level make its own zone decisions */ gfp &= ~(GFP_DMA32|GFP_DMA); - if (dma_ops->alloc_coherent) - return dma_ops->alloc_coherent(dev, size, + if (ops->alloc_coherent) + return ops->alloc_coherent(dev, size, dma_handle, gfp); return NULL; } @@ -504,14 +394,14 @@ dma_alloc_coherent(struct device *dev, s } } - if (dma_ops->alloc_coherent) { + if (ops->alloc_coherent) { free_pages((unsigned long)memory, order); gfp &= ~(GFP_DMA|GFP_DMA32); - return dma_ops->alloc_coherent(dev, size, dma_handle, gfp); + return ops->alloc_coherent(dev, size, dma_handle, gfp); } - if (dma_ops->map_simple) { - *dma_handle = dma_ops->map_simple(dev, virt_to_bus(memory), + if (ops->map_simple) { + *dma_handle = ops->map_simple(dev, virt_to_bus(memory), size, PCI_DMA_BIDIRECTIONAL); if (*dma_handle != bad_dma_address) @@ -542,13 +432,17 @@ EXPORT_SYMBOL(dma_alloc_coherent); void dma_free_coherent(struct device *dev, size_t size, void *vaddr, dma_addr_t bus) { +#ifndef CONFIG_XEN + struct dma_mapping_ops *ops = get_dma_ops(dev); +#endif + int order = get_order(size); WARN_ON(irqs_disabled()); /* for portability */ - if (dma_release_coherent(dev, order, vaddr)) + if (dma_release_from_coherent(dev, order, vaddr)) return; #ifndef CONFIG_XEN - if (dma_ops->unmap_single) - dma_ops->unmap_single(dev, bus, size, 0); + if (ops->unmap_single) + ops->unmap_single(dev, bus, size, 0); #endif xen_destroy_contiguous_region((unsigned long)vaddr, order); free_pages((unsigned long)vaddr, order); @@ -557,15 +451,13 @@ EXPORT_SYMBOL(dma_free_coherent); static int __init pci_iommu_init(void) { -#ifdef CONFIG_CALGARY_IOMMU calgary_iommu_init(); -#endif intel_iommu_init(); -#ifdef CONFIG_GART_IOMMU + amd_iommu_init(); + gart_iommu_init(); -#endif no_iommu_init(); return 0; --- head-2011-03-11.orig/arch/x86/kernel/pci-nommu-xen.c 2011-01-31 18:07:35.000000000 +0100 +++ head-2011-03-11/arch/x86/kernel/pci-nommu-xen.c 2011-02-01 14:38:38.000000000 +0100 @@ -84,18 +84,12 @@ static int nommu_dma_supported(struct de return 1; } -static int nommu_mapping_error(dma_addr_t dma_addr) -{ - return (dma_addr == bad_dma_address); -} - -static const struct dma_mapping_ops nommu_dma_ops = { +static struct dma_mapping_ops nommu_dma_ops = { .map_single = gnttab_map_single, .unmap_single = gnttab_unmap_single, .map_sg = gnttab_map_sg, .unmap_sg = gnttab_unmap_sg, .dma_supported = nommu_dma_supported, - .mapping_error = nommu_mapping_error }; void __init no_iommu_init(void) --- head-2011-03-11.orig/arch/x86/kernel/probe_roms_32.c 2011-03-15 16:45:55.000000000 +0100 +++ head-2011-03-11/arch/x86/kernel/probe_roms_32.c 2011-02-01 14:38:38.000000000 +0100 @@ -131,7 +131,7 @@ void __init probe_roms(void) upper = system_rom_resource.start; /* check for extension rom (ignore length byte!) */ - rom = isa_bus_to_virt(extension_rom_resource.start); + rom = isa_bus_to_virt((unsigned long)extension_rom_resource.start); if (romsignature(rom)) { length = extension_rom_resource.end - extension_rom_resource.start + 1; if (romchecksum(rom, length)) { --- head-2011-03-11.orig/arch/x86/kernel/process-xen.c 2011-01-31 18:07:35.000000000 +0100 +++ head-2011-03-11/arch/x86/kernel/process-xen.c 2011-03-03 15:59:49.000000000 +0100 @@ -6,6 +6,13 @@ #include #include #include +#include +#include + +unsigned long idle_halt; +EXPORT_SYMBOL(idle_halt); +unsigned long idle_nomwait; +EXPORT_SYMBOL(idle_nomwait); struct kmem_cache *task_xstate_cachep; @@ -45,6 +52,41 @@ void arch_task_cache_init(void) SLAB_PANIC, NULL); } +/* + * Idle related variables and functions + */ +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); + +/* + * We use this if we don't have any better + * idle routine.. + */ +void xen_idle(void) +{ + current_thread_info()->status &= ~TS_POLLING; + /* + * TS_POLLING-cleared state must be visible before we + * test NEED_RESCHED: + */ + smp_mb(); + + if (!need_resched()) + safe_halt(); /* enables interrupts racelessly */ + else + local_irq_enable(); + current_thread_info()->status |= TS_POLLING; +} +#ifdef CONFIG_APM_MODULE +EXPORT_SYMBOL(default_idle); +#endif + static void do_nothing(void *unused) { } @@ -61,7 +103,7 @@ void cpu_idle_wait(void) { smp_mb(); /* kick all the CPUs so that they exit out of pm_idle */ - smp_call_function(do_nothing, NULL, 0, 1); + smp_call_function(do_nothing, NULL, 1); } EXPORT_SYMBOL_GPL(cpu_idle_wait); @@ -125,60 +167,175 @@ static void poll_idle(void) * * idle=mwait overrides this decision and forces the usage of mwait. */ +static int __cpuinitdata force_mwait; + +#define MWAIT_INFO 0x05 +#define MWAIT_ECX_EXTENDED_INFO 0x01 +#define MWAIT_EDX_C1 0xf0 + static int __cpuinit mwait_usable(const struct cpuinfo_x86 *c) { + u32 eax, ebx, ecx, edx; + if (force_mwait) return 1; - if (c->x86_vendor == X86_VENDOR_AMD) { - switch(c->x86) { - case 0x10: - case 0x11: - return 0; + if (c->cpuid_level < MWAIT_INFO) + return 0; + + cpuid(MWAIT_INFO, &eax, &ebx, &ecx, &edx); + /* Check, whether EDX has extended info about MWAIT */ + if (!(ecx & MWAIT_ECX_EXTENDED_INFO)) + return 1; + + /* + * edx enumeratios MONITOR/MWAIT extensions. Check, whether + * C1 supports MWAIT + */ + return (edx & MWAIT_EDX_C1); +} + +/* + * Check for AMD CPUs, which have potentially C1E support + */ +static int __cpuinit check_c1e_idle(const struct cpuinfo_x86 *c) +{ + if (c->x86_vendor != X86_VENDOR_AMD) + return 0; + + if (c->x86 < 0x0F) + return 0; + + /* Family 0x0f models < rev F do not have C1E */ + if (c->x86 == 0x0f && c->x86_model < 0x40) + return 0; + + return 1; +} + +static cpumask_t c1e_mask = CPU_MASK_NONE; +static int c1e_detected; + +void c1e_remove_cpu(int cpu) +{ + cpu_clear(cpu, c1e_mask); +} + +/* + * C1E aware idle routine. We check for C1E active in the interrupt + * pending message MSR. If we detect C1E, then we handle it the same + * way as C3 power states (local apic timer and TSC stop) + */ +static void c1e_idle(void) +{ + if (need_resched()) + return; + + if (!c1e_detected) { + u32 lo, hi; + + rdmsr(MSR_K8_INT_PENDING_MSG, lo, hi); + if (lo & K8_INTP_C1E_ACTIVE_MASK) { + c1e_detected = 1; + if (!boot_cpu_has(X86_FEATURE_CONSTANT_TSC)) + mark_tsc_unstable("TSC halt in AMD C1E"); + printk(KERN_INFO "System has AMD C1E enabled\n"); + set_cpu_cap(&boot_cpu_data, X86_FEATURE_AMDC1E); } } - return 1; + + if (c1e_detected) { + int cpu = smp_processor_id(); + + if (!cpu_isset(cpu, c1e_mask)) { + cpu_set(cpu, c1e_mask); + /* + * Force broadcast so ACPI can not interfere. Needs + * to run with interrupts enabled as it uses + * smp_function_call. + */ + local_irq_enable(); + clockevents_notify(CLOCK_EVT_NOTIFY_BROADCAST_FORCE, + &cpu); + printk(KERN_INFO "Switch to broadcast mode on CPU%d\n", + cpu); + local_irq_disable(); + } + clockevents_notify(CLOCK_EVT_NOTIFY_BROADCAST_ENTER, &cpu); + + default_idle(); + + /* + * The switch back from broadcast mode needs to be + * called with interrupts disabled. + */ + local_irq_disable(); + clockevents_notify(CLOCK_EVT_NOTIFY_BROADCAST_EXIT, &cpu); + local_irq_enable(); + } else + default_idle(); } #endif void __cpuinit select_idle_routine(const struct cpuinfo_x86 *c) { #ifndef CONFIG_XEN - static int selected; - - if (selected) - return; #ifdef CONFIG_X86_SMP if (pm_idle == poll_idle && smp_num_siblings > 1) { printk(KERN_WARNING "WARNING: polling idle and HT enabled," " performance may degrade.\n"); } #endif + if (pm_idle) + return; + if (cpu_has(c, X86_FEATURE_MWAIT) && mwait_usable(c)) { /* - * Skip, if setup has overridden idle. * One CPU supports mwait => All CPUs supports mwait */ - if (!pm_idle) { - printk(KERN_INFO "using mwait in idle threads.\n"); - pm_idle = mwait_idle; - } - } - selected = 1; + printk(KERN_INFO "using mwait in idle threads.\n"); + pm_idle = mwait_idle; + } else if (check_c1e_idle(c)) { + printk(KERN_INFO "using C1E aware idle routine\n"); + pm_idle = c1e_idle; + } else + pm_idle = default_idle; #endif } static int __init idle_setup(char *str) { + if (!str) + return -EINVAL; + if (!strcmp(str, "poll")) { printk("using polling idle threads.\n"); pm_idle = poll_idle; - } #ifndef CONFIG_XEN - else if (!strcmp(str, "mwait")) + } else if (!strcmp(str, "mwait")) force_mwait = 1; + else if (!strcmp(str, "halt")) { + /* + * When the boot option of idle=halt is added, halt is + * forced to be used for CPU idle. In such case CPU C2/C3 + * won't be used again. + * To continue to load the CPU idle driver, don't touch + * the boot_option_idle_override. + */ + pm_idle = default_idle; + idle_halt = 1; + return 0; + } else if (!strcmp(str, "nomwait")) { + /* + * If the boot option of "idle=nomwait" is added, + * it means that mwait will be disabled for CPU C2/C3 + * states. In such case it won't touch the variable + * of boot_option_idle_override. + */ + idle_nomwait = 1; + return 0; #endif - else + } else return -1; boot_option_idle_override = 1; --- head-2011-03-11.orig/arch/x86/kernel/process_32-xen.c 2011-01-31 18:07:35.000000000 +0100 +++ head-2011-03-11/arch/x86/kernel/process_32-xen.c 2011-02-01 14:38:38.000000000 +0100 @@ -59,15 +59,11 @@ #include #include #include +#include asmlinkage void ret_from_fork(void) __asm__("ret_from_fork"); asmlinkage void cstar_ret_from_fork(void) __asm__("cstar_ret_from_fork"); -static int hlt_counter; - -unsigned long boot_option_idle_override = 0; -EXPORT_SYMBOL(boot_option_idle_override); - DEFINE_PER_CPU(struct task_struct *, current_task) = &init_task; EXPORT_PER_CPU_SYMBOL(current_task); @@ -82,46 +78,27 @@ unsigned long thread_saved_pc(struct tas return ((unsigned long *)tsk->thread.sp)[3]; } -/* - * Powermanagement idle function, if any.. - */ -void (*pm_idle)(void); -EXPORT_SYMBOL(pm_idle); +#ifdef CONFIG_HOTPLUG_CPU +#ifndef CONFIG_XEN +#include -void disable_hlt(void) +static void cpu_exit_clear(void) { - hlt_counter++; -} + int cpu = raw_smp_processor_id(); -EXPORT_SYMBOL(disable_hlt); - -void enable_hlt(void) -{ - hlt_counter--; -} + idle_task_exit(); -EXPORT_SYMBOL(enable_hlt); + cpu_uninit(); + irq_ctx_exit(cpu); -static void xen_idle(void) -{ - current_thread_info()->status &= ~TS_POLLING; - /* - * TS_POLLING-cleared state must be visible before we - * test NEED_RESCHED: - */ - smp_mb(); + cpu_clear(cpu, cpu_callout_map); + cpu_clear(cpu, cpu_callin_map); - if (!need_resched()) - safe_halt(); /* enables interrupts racelessly */ - else - local_irq_enable(); - current_thread_info()->status |= TS_POLLING; + numa_remove_cpu(cpu); + c1e_remove_cpu(cpu); } -#ifdef CONFIG_APM_MODULE -EXPORT_SYMBOL(default_idle); #endif -#ifdef CONFIG_HOTPLUG_CPU static inline void play_dead(void) { idle_task_exit(); @@ -152,13 +129,11 @@ void cpu_idle(void) /* endless idle loop with no priority at all */ while (1) { - tick_nohz_stop_sched_tick(); + tick_nohz_stop_sched_tick(1); while (!need_resched()) { - void (*idle)(void); check_pgt_cache(); rmb(); - idle = xen_idle; /* no alternatives */ if (rcu_pending(cpu)) rcu_check_callbacks(cpu, 0); @@ -168,7 +143,10 @@ void cpu_idle(void) local_irq_disable(); __get_cpu_var(irq_stat).idle_timestamp = jiffies; - idle(); + /* Don't trace irqs off for idle */ + stop_critical_timings(); + xen_idle(); + start_critical_timings(); } tick_nohz_restart_sched_tick(); preempt_enable_no_resched(); --- head-2011-03-11.orig/arch/x86/kernel/process_64-xen.c 2011-02-02 08:32:46.000000000 +0100 +++ head-2011-03-11/arch/x86/kernel/process_64-xen.c 2011-02-02 08:34:01.000000000 +0100 @@ -64,15 +64,6 @@ asmlinkage extern void ret_from_fork(voi 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 ATOMIC_NOTIFIER_HEAD(idle_notifier); void idle_notifier_register(struct notifier_block *n) @@ -102,25 +93,13 @@ void exit_idle(void) __exit_idle(); } -static void xen_idle(void) -{ - current_thread_info()->status &= ~TS_POLLING; - /* - * TS_POLLING-cleared state must be visible before we - * test NEED_RESCHED: - */ - smp_mb(); - if (!need_resched()) - safe_halt(); /* enables interrupts racelessly */ - else - local_irq_enable(); - current_thread_info()->status |= TS_POLLING; -} - #ifdef CONFIG_HOTPLUG_CPU static inline void play_dead(void) { idle_task_exit(); +#ifndef CONFIG_XEN + c1e_remove_cpu(raw_smp_processor_id()); +#endif local_irq_disable(); cpu_clear(smp_processor_id(), cpu_initialized); preempt_enable_no_resched(); @@ -145,12 +124,11 @@ void cpu_idle(void) current_thread_info()->status |= TS_POLLING; /* endless idle loop with no priority at all */ while (1) { - tick_nohz_stop_sched_tick(); + tick_nohz_stop_sched_tick(1); while (!need_resched()) { - void (*idle)(void); rmb(); - idle = xen_idle; /* no alternatives */ + if (cpu_is_offline(smp_processor_id())) play_dead(); /* @@ -160,7 +138,10 @@ void cpu_idle(void) */ local_irq_disable(); enter_idle(); - idle(); + /* Don't trace irqs off for idle */ + stop_critical_timings(); + xen_idle(); + start_critical_timings(); /* In many cases the interrupt that ended idle has already called exit_idle. But some idle loops can be woken up without interrupt. */ @@ -270,7 +251,7 @@ void exit_thread(void) } } -void load_gs_index(unsigned gs) +void xen_load_gs_index(unsigned gs) { WARN_ON(HYPERVISOR_set_segment_base(SEGBASE_GS_USER_SEL, gs)); } @@ -371,10 +352,10 @@ int copy_thread(int nr, unsigned long cl 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)); + savesegment(gs, p->thread.gsindex); + savesegment(fs, p->thread.fsindex); + savesegment(es, p->thread.es); + savesegment(ds, p->thread.ds); if (unlikely(test_tsk_thread_flag(me, TIF_IO_BITMAP))) { p->thread.io_bitmap_ptr = kmalloc(IO_BITMAP_BYTES, GFP_KERNEL); @@ -415,7 +396,9 @@ out: void start_thread(struct pt_regs *regs, unsigned long new_ip, unsigned long new_sp) { - asm volatile("movl %0, %%fs; movl %0, %%es; movl %0, %%ds" :: "r"(0)); + loadsegment(fs, 0); + loadsegment(es, 0); + loadsegment(ds, 0); load_gs_index(0); regs->ip = new_ip; regs->sp = new_sp; @@ -554,8 +537,8 @@ static inline void __switch_to_xtra(stru 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; + struct thread_struct *prev = &prev_p->thread; + struct thread_struct *next = &next_p->thread; int cpu = smp_processor_id(); #ifndef CONFIG_X86_NO_TSS struct tss_struct *tss = &per_cpu(init_tss, cpu); @@ -658,12 +641,25 @@ __switch_to(struct task_struct *prev_p, */ if (unlikely(next->es)) loadsegment(es, next->es); - + if (unlikely(next->ds)) loadsegment(ds, next->ds); + /* + * Leave lazy mode, flushing any hypercalls made here. + * This must be done before restoring TLS segments so + * the GDT and LDT are properly updated, and must be + * done before math_state_restore, so the TS bit is up + * to date. + */ + arch_leave_lazy_cpu_mode(); + /* * Switch FS and GS. + * + * Segment register != 0 always requires a reload. Also + * reload when it has changed. When prev process used 64bit + * base always reload to avoid an information leak. */ if (unlikely(next->fsindex)) loadsegment(fs, next->fsindex); @@ -682,7 +678,8 @@ __switch_to(struct task_struct *prev_p, */ write_pda(pcurrent, next_p); write_pda(kernelstack, - (unsigned long)task_stack_page(next_p) + THREAD_SIZE - PDA_STACKOFFSET); + (unsigned long)task_stack_page(next_p) + + THREAD_SIZE - PDA_STACKOFFSET); #ifdef CONFIG_CC_STACKPROTECTOR write_pda(stack_canary, next_p->stack_canary); @@ -843,7 +840,7 @@ long do_arch_prctl(struct task_struct *t set_32bit_tls(task, FS_TLS, addr); if (doit) { load_TLS(&task->thread, cpu); - asm volatile("movl %0,%%fs" :: "r"(FS_TLS_SEL)); + loadsegment(fs, FS_TLS_SEL); } task->thread.fsindex = FS_TLS_SEL; task->thread.fs = 0; @@ -853,7 +850,7 @@ long do_arch_prctl(struct task_struct *t if (doit) { /* set the selector to 0 to not confuse __switch_to */ - asm volatile("movl %0,%%fs" :: "r" (0)); + loadsegment(fs, 0); ret = HYPERVISOR_set_segment_base(SEGBASE_FS, addr); } @@ -877,7 +874,7 @@ long do_arch_prctl(struct task_struct *t if (task->thread.gsindex == GS_TLS_SEL) base = read_32bit_tls(task, GS_TLS); else if (doit) { - asm("movl %%gs,%0" : "=r" (gsindex)); + savesegment(gs, gsindex); if (gsindex) rdmsrl(MSR_KERNEL_GS_BASE, base); else --- head-2011-03-11.orig/arch/x86/kernel/setup-xen.c 2011-01-31 18:07:35.000000000 +0100 +++ head-2011-03-11/arch/x86/kernel/setup-xen.c 2011-03-04 15:09:03.000000000 +0100 @@ -1,143 +1,1132 @@ -#include +/* + * Copyright (C) 1995 Linus Torvalds + * + * Support of BIGMEM added by Gerhard Wichert, Siemens AG, July 1999 + * + * Memory region support + * David Parsons , July-August 1999 + * + * Added E820 sanitization routine (removes overlapping memory regions); + * Brian Moyle , February 2001 + * + * Moved CPU detection code to cpu/${cpu}.c + * Patrick Mochel , March 2002 + * + * Provisions for empty E820 memory regions (reported by certain BIOSes). + * Alex Achenbach , December 2002. + * + */ + +/* + * This file handles the architecture-dependent parts of initialization + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include #include +#include #include -#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include +#include + #include -#include -#include +#include + +#include