Subject: kexec: Move asm segment handling code to the assembly file (i386)
From: http://xenbits.xensource.com/xen-unstable.hg (tip 13816)
Patch-mainline: n/a

This patch moves the idt, gdt, and segment handling code from machine_kexec.c
to relocate_kernel.S. The main reason behind this move is to avoid code 
duplication in the Xen hypervisor. With this patch all code required to kexec
is put on the control page.

On top of that this patch also counts as a cleanup - I think it is much
nicer to write assembly directly in assembly files than wrap inline assembly
in C functions for no apparent reason.

Signed-off-by: Magnus Damm <magnus@valinux.co.jp>
Acked-by: jbeulich@novell.com

 Applies to 2.6.19-rc1.
 jb: fixed up register usage (paralleling what's needed for 2.6.30 on x86-64)

--- head-2010-01-19.orig/arch/x86/kernel/machine_kexec_32.c	2010-01-19 13:26:10.000000000 +0100
+++ head-2010-01-19/arch/x86/kernel/machine_kexec_32.c	2010-01-19 14:51:07.000000000 +0100
@@ -27,48 +27,6 @@
 #include <asm/cacheflush.h>
 #include <asm/debugreg.h>
 
-static void set_idt(void *newidt, __u16 limit)
-{
-	struct desc_ptr curidt;
-
-	/* ia32 supports unaliged loads & stores */
-	curidt.size    = limit;
-	curidt.address = (unsigned long)newidt;
-
-	load_idt(&curidt);
-}
-
-
-static void set_gdt(void *newgdt, __u16 limit)
-{
-	struct desc_ptr curgdt;
-
-	/* ia32 supports unaligned loads & stores */
-	curgdt.size    = limit;
-	curgdt.address = (unsigned long)newgdt;
-
-	load_gdt(&curgdt);
-}
-
-static void load_segments(void)
-{
-#define __STR(X) #X
-#define STR(X) __STR(X)
-
-	__asm__ __volatile__ (
-		"\tljmp $"STR(__KERNEL_CS)",$1f\n"
-		"\t1:\n"
-		"\tmovl $"STR(__KERNEL_DS)",%%eax\n"
-		"\tmovl %%eax,%%ds\n"
-		"\tmovl %%eax,%%es\n"
-		"\tmovl %%eax,%%fs\n"
-		"\tmovl %%eax,%%gs\n"
-		"\tmovl %%eax,%%ss\n"
-		: : : "eax", "memory");
-#undef STR
-#undef __STR
-}
-
 static void machine_kexec_free_page_tables(struct kimage *image)
 {
 	free_page((unsigned long)image->arch.pgd);
@@ -228,24 +186,6 @@ void machine_kexec(struct kimage *image)
 		page_list[PA_SWAP_PAGE] = (page_to_pfn(image->swap_page)
 						<< PAGE_SHIFT);
 
-	/*
-	 * The segment registers are funny things, they have both a
-	 * visible and an invisible part.  Whenever the visible part is
-	 * set to a specific selector, the invisible part is loaded
-	 * with from a table in memory.  At no other time is the
-	 * descriptor table in memory accessed.
-	 *
-	 * I take advantage of this here by force loading the
-	 * segments, before I zap the gdt with an invalid value.
-	 */
-	load_segments();
-	/*
-	 * The gdt & idt are now invalid.
-	 * If you want to load them you must set up your own idt & gdt.
-	 */
-	set_gdt(phys_to_virt(0), 0);
-	set_idt(phys_to_virt(0), 0);
-
 	/* now call it */
 	image->start = relocate_kernel_ptr((unsigned long)image->head,
 					   (unsigned long)page_list,
--- head-2010-01-19.orig/arch/x86/kernel/relocate_kernel_32.S	2009-06-10 05:05:27.000000000 +0200
+++ head-2010-01-19/arch/x86/kernel/relocate_kernel_32.S	2010-01-19 14:51:07.000000000 +0100
@@ -87,14 +87,32 @@ relocate_kernel:
 	movl	PTR(PA_PGD)(%ebp), %eax
 	movl	%eax, %cr3
 
+	/* setup idt */
+	lidtl	idt_48 - relocate_kernel(%edi)
+
+	/* setup gdt */
+	leal	gdt - relocate_kernel(%edi), %eax
+	movl	%eax, (gdt_48 - relocate_kernel) + 2(%edi)
+	lgdtl	gdt_48 - relocate_kernel(%edi)
+
+	/* setup data segment registers */
+	mov	$(gdt_ds - gdt), %eax
+	mov	%eax, %ds
+	mov	%eax, %es
+	mov	%eax, %fs
+	mov	%eax, %gs
+	mov	%eax, %ss
+
 	/* setup a new stack at the end of the physical control page */
 	lea	PAGE_SIZE(%edi), %esp
 
-	/* jump to identity mapped page */
+	/* load new code segment and jump to identity mapped page */
+	pushl	$0
+	pushl	$(gdt_cs - gdt)
 	movl    %edi, %eax
 	addl    $(identity_mapped - relocate_kernel), %eax
 	pushl   %eax
-	ret
+	iretl
 
 identity_mapped:
 	/* store the start address on the stack */
@@ -271,5 +289,22 @@ swap_pages:
 	popl	%ebp
 	ret
 
+	.align	16
+gdt:
+	.quad	0x0000000000000000	/* NULL descriptor */
+gdt_cs:
+	.quad	0x00cf9a000000ffff	/* kernel 4GB code at 0x00000000 */
+gdt_ds:
+	.quad	0x00cf92000000ffff	/* kernel 4GB data at 0x00000000 */
+gdt_end:
+
+gdt_48:
+	.word	gdt_end - gdt - 1	/* limit */
+	.long	0			/* base - filled in by code above */
+
+idt_48:
+	.word	0			/* limit */
+	.long	0			/* base */
+
 	.globl kexec_control_code_size
 .set kexec_control_code_size, . - relocate_kernel