[linux-2.6-block.git] / arch / x86 / kernel / machine_kexec_32.c

/*
 * machine_kexec.c - handle transition of Linux booting another kernel
 * Copyright (C) 2002-2005 Eric Biederman  <ebiederm@xmission.com>
 *
 * This source code is licensed under the GNU General Public License,
 * Version 2.  See the file COPYING for more details.
 */

#include <linux/mm.h>
#include <linux/kexec.h>
#include <linux/delay.h>
#include <linux/init.h>
#include <asm/pgtable.h>
#include <asm/pgalloc.h>
#include <asm/tlbflush.h>
#include <asm/mmu_context.h>
#include <asm/io.h>
#include <asm/apic.h>
#include <asm/cpufeature.h>
#include <asm/desc.h>
#include <asm/system.h>

#define PAGE_ALIGNED __attribute__ ((__aligned__(PAGE_SIZE)))
static u32 kexec_pgd[1024] PAGE_ALIGNED;
#ifdef CONFIG_X86_PAE
static u32 kexec_pmd0[1024] PAGE_ALIGNED;
static u32 kexec_pmd1[1024] PAGE_ALIGNED;
#endif
static u32 kexec_pte0[1024] PAGE_ALIGNED;
static u32 kexec_pte1[1024] PAGE_ALIGNED;

static void set_idt(void *newidt, __u16 limit)
{
	struct Xgt_desc_struct 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 Xgt_desc_struct 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
}

/*
 * A architecture hook called to validate the
 * proposed image and prepare the control pages
 * as needed.  The pages for KEXEC_CONTROL_CODE_SIZE
 * have been allocated, but the segments have yet
 * been copied into the kernel.
 *
 * Do what every setup is needed on image and the
 * reboot code buffer to allow us to avoid allocations
 * later.
 *
 * Currently nothing.
 */
int machine_kexec_prepare(struct kimage *image)
{
	return 0;
}

/*
 * Undo anything leftover by machine_kexec_prepare
 * when an image is freed.
 */
void machine_kexec_cleanup(struct kimage *image)
{
}

/*
 * Do not allocate memory (or fail in any way) in machine_kexec().
 * We are past the point of no return, committed to rebooting now.
 */
NORET_TYPE void machine_kexec(struct kimage *image)
{
	unsigned long page_list[PAGES_NR];
	void *control_page;

	/* Interrupts aren't acceptable while we reboot */
	local_irq_disable();

	control_page = page_address(image->control_code_page);
	memcpy(control_page, relocate_kernel, PAGE_SIZE);

	page_list[PA_CONTROL_PAGE] = __pa(control_page);
	page_list[VA_CONTROL_PAGE] = (unsigned long)relocate_kernel;
	page_list[PA_PGD] = __pa(kexec_pgd);
	page_list[VA_PGD] = (unsigned long)kexec_pgd;
#ifdef CONFIG_X86_PAE
	page_list[PA_PMD_0] = __pa(kexec_pmd0);
	page_list[VA_PMD_0] = (unsigned long)kexec_pmd0;
	page_list[PA_PMD_1] = __pa(kexec_pmd1);
	page_list[VA_PMD_1] = (unsigned long)kexec_pmd1;
#endif
	page_list[PA_PTE_0] = __pa(kexec_pte0);
	page_list[VA_PTE_0] = (unsigned long)kexec_pte0;
	page_list[PA_PTE_1] = __pa(kexec_pte1);
	page_list[VA_PTE_1] = (unsigned long)kexec_pte1;

	/* 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 */
	relocate_kernel((unsigned long)image->head, (unsigned long)page_list,
			image->start, cpu_has_pae);
}

/* crashkernel=size@addr specifies the location to reserve for
 * a crash kernel.  By reserving this memory we guarantee
 * that linux never sets it up as a DMA target.
 * Useful for holding code to do something appropriate
 * after a kernel panic.
 */
static int __init parse_crashkernel(char *arg)
{
	unsigned long size, base;
	size = memparse(arg, &arg);
	if (*arg == '@') {
		base = memparse(arg+1, &arg);
		/* FIXME: Do I want a sanity check
		 * to validate the memory range?
		 */
		crashk_res.start = base;
		crashk_res.end   = base + size - 1;
	}
	return 0;
}
early_param("crashkernel", parse_crashkernel);
Commit	Line	Data
5033cba0 EB	1	/*
	2	* machine_kexec.c - handle transition of Linux booting another kernel
	3	* Copyright (C) 2002-2005 Eric Biederman <ebiederm@xmission.com>
	4	*
	5	* This source code is licensed under the GNU General Public License,
	6	* Version 2. See the file COPYING for more details.
	7	*/
	8
	9	#include <linux/mm.h>
	10	#include <linux/kexec.h>
	11	#include <linux/delay.h>
1a3f239d	12	#include <linux/init.h>
5033cba0 EB	13	#include <asm/pgtable.h>
	14	#include <asm/pgalloc.h>
	15	#include <asm/tlbflush.h>
	16	#include <asm/mmu_context.h>
	17	#include <asm/io.h>
	18	#include <asm/apic.h>
	19	#include <asm/cpufeature.h>
e7b47cca	20	#include <asm/desc.h>
4bb0d3ec	21	#include <asm/system.h>
5033cba0 EB	22
5033cba0 EB	23	#define PAGE_ALIGNED __attribute__ ((__aligned__(PAGE_SIZE)))
3566561b MD	24	static u32 kexec_pgd[1024] PAGE_ALIGNED;
	25	#ifdef CONFIG_X86_PAE
	26	static u32 kexec_pmd0[1024] PAGE_ALIGNED;
	27	static u32 kexec_pmd1[1024] PAGE_ALIGNED;
5033cba0	28	#endif
3566561b MD	29	static u32 kexec_pte0[1024] PAGE_ALIGNED;
3566561b MD	30	static u32 kexec_pte1[1024] PAGE_ALIGNED;
5033cba0	31
5033cba0 EB	32	static void set_idt(void *newidt, __u16 limit)
5033cba0 EB	33	{
e7b47cca	34	struct Xgt_desc_struct curidt;
5033cba0 EB	35
5033cba0 EB	36	/* ia32 supports unaliged loads & stores */
e7b47cca EB	37	curidt.size = limit;
e7b47cca EB	38	curidt.address = (unsigned long)newidt;
5033cba0	39
f2ab4461	40	load_idt(&curidt);
5033cba0 EB	41	};
	42
	43
	44	static void set_gdt(void *newgdt, __u16 limit)
	45	{
e7b47cca	46	struct Xgt_desc_struct curgdt;
5033cba0 EB	47
5033cba0 EB	48	/* ia32 supports unaligned loads & stores */
e7b47cca EB	49	curgdt.size = limit;
e7b47cca EB	50	curgdt.address = (unsigned long)newgdt;
5033cba0	51
f2ab4461	52	load_gdt(&curgdt);
5033cba0 EB	53	};
	54
	55	static void load_segments(void)
	56	{
	57	#define __STR(X) #X
	58	#define STR(X) __STR(X)
	59
	60	__asm__ __volatile__ (
	61	"\tljmp $"STR(__KERNEL_CS)",$1f\n"
	62	"\t1:\n"
2ec5e3a8 MM	63	"\tmovl $"STR(__KERNEL_DS)",%%eax\n"
	64	"\tmovl %%eax,%%ds\n"
	65	"\tmovl %%eax,%%es\n"
	66	"\tmovl %%eax,%%fs\n"
	67	"\tmovl %%eax,%%gs\n"
	68	"\tmovl %%eax,%%ss\n"
	69	::: "eax", "memory");
5033cba0 EB	70	#undef STR
	71	#undef __STR
	72	}
	73
5033cba0 EB	74	/*
	75	* A architecture hook called to validate the
	76	* proposed image and prepare the control pages
	77	* as needed. The pages for KEXEC_CONTROL_CODE_SIZE
	78	* have been allocated, but the segments have yet
	79	* been copied into the kernel.
	80	*
	81	* Do what every setup is needed on image and the
	82	* reboot code buffer to allow us to avoid allocations
	83	* later.
	84	*
	85	* Currently nothing.
	86	*/
	87	int machine_kexec_prepare(struct kimage *image)
	88	{
	89	return 0;
	90	}
	91
	92	/*
	93	* Undo anything leftover by machine_kexec_prepare
	94	* when an image is freed.
	95	*/
	96	void machine_kexec_cleanup(struct kimage *image)
	97	{
	98	}
	99
	100	/*
	101	* Do not allocate memory (or fail in any way) in machine_kexec().
	102	* We are past the point of no return, committed to rebooting now.
	103	*/
	104	NORET_TYPE void machine_kexec(struct kimage *image)
	105	{
3566561b MD	106	unsigned long page_list[PAGES_NR];
3566561b MD	107	void *control_page;
5033cba0 EB	108
	109	/* Interrupts aren't acceptable while we reboot */
	110	local_irq_disable();
	111
3566561b MD	112	control_page = page_address(image->control_code_page);
	113	memcpy(control_page, relocate_kernel, PAGE_SIZE);
	114
	115	page_list[PA_CONTROL_PAGE] = __pa(control_page);
	116	page_list[VA_CONTROL_PAGE] = (unsigned long)relocate_kernel;
	117	page_list[PA_PGD] = __pa(kexec_pgd);
	118	page_list[VA_PGD] = (unsigned long)kexec_pgd;
	119	#ifdef CONFIG_X86_PAE
	120	page_list[PA_PMD_0] = __pa(kexec_pmd0);
	121	page_list[VA_PMD_0] = (unsigned long)kexec_pmd0;
	122	page_list[PA_PMD_1] = __pa(kexec_pmd1);
	123	page_list[VA_PMD_1] = (unsigned long)kexec_pmd1;
	124	#endif
	125	page_list[PA_PTE_0] = __pa(kexec_pte0);
	126	page_list[VA_PTE_0] = (unsigned long)kexec_pte0;
	127	page_list[PA_PTE_1] = __pa(kexec_pte1);
	128	page_list[VA_PTE_1] = (unsigned long)kexec_pte1;
5033cba0	129
2a8a3d5b EB	130	/* The segment registers are funny things, they have both a
	131	* visible and an invisible part. Whenever the visible part is
	132	* set to a specific selector, the invisible part is loaded
	133	* with from a table in memory. At no other time is the
	134	* descriptor table in memory accessed.
5033cba0 EB	135	*
	136	* I take advantage of this here by force loading the
	137	* segments, before I zap the gdt with an invalid value.
	138	*/
	139	load_segments();
	140	/* The gdt & idt are now invalid.
	141	* If you want to load them you must set up your own idt & gdt.
	142	*/
	143	set_gdt(phys_to_virt(0),0);
	144	set_idt(phys_to_virt(0),0);
	145
	146	/* now call it */
3566561b MD	147	relocate_kernel((unsigned long)image->head, (unsigned long)page_list,
3566561b MD	148	image->start, cpu_has_pae);
5033cba0	149	}
1a3f239d RR	150
	151	/* crashkernel=size@addr specifies the location to reserve for
	152	* a crash kernel. By reserving this memory we guarantee
	153	* that linux never sets it up as a DMA target.
	154	* Useful for holding code to do something appropriate
	155	* after a kernel panic.
	156	*/
	157	static int __init parse_crashkernel(char *arg)
	158	{
	159	unsigned long size, base;
	160	size = memparse(arg, &arg);
	161	if (*arg == '@') {
	162	base = memparse(arg+1, &arg);
	163	/* FIXME: Do I want a sanity check
	164	* to validate the memory range?
	165	*/
	166	crashk_res.start = base;
	167	crashk_res.end = base + size - 1;
	168	}
	169	return 0;
	170	}
	171	early_param("crashkernel", parse_crashkernel);