2005-06-25 17:58:02 -04:00
|
|
|
/*
|
|
|
|
* 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/string.h>
|
|
|
|
#include <linux/reboot.h>
|
|
|
|
#include <asm/pgtable.h>
|
|
|
|
#include <asm/tlbflush.h>
|
|
|
|
#include <asm/mmu_context.h>
|
|
|
|
#include <asm/io.h>
|
2005-07-29 15:25:28 -04:00
|
|
|
|
|
|
|
static void init_level2_page(pmd_t *level2p, unsigned long addr)
|
2005-06-25 17:58:02 -04:00
|
|
|
{
|
|
|
|
unsigned long end_addr;
|
2005-06-25 17:58:28 -04:00
|
|
|
|
2005-06-25 17:58:02 -04:00
|
|
|
addr &= PAGE_MASK;
|
2005-07-29 15:25:28 -04:00
|
|
|
end_addr = addr + PUD_SIZE;
|
2005-06-25 17:58:28 -04:00
|
|
|
while (addr < end_addr) {
|
2005-07-29 15:25:28 -04:00
|
|
|
set_pmd(level2p++, __pmd(addr | __PAGE_KERNEL_LARGE_EXEC));
|
|
|
|
addr += PMD_SIZE;
|
2005-06-25 17:58:02 -04:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2005-07-29 15:25:28 -04:00
|
|
|
static int init_level3_page(struct kimage *image, pud_t *level3p,
|
2005-06-25 17:58:28 -04:00
|
|
|
unsigned long addr, unsigned long last_addr)
|
2005-06-25 17:58:02 -04:00
|
|
|
{
|
|
|
|
unsigned long end_addr;
|
|
|
|
int result;
|
2005-06-25 17:58:28 -04:00
|
|
|
|
2005-06-25 17:58:02 -04:00
|
|
|
result = 0;
|
|
|
|
addr &= PAGE_MASK;
|
2005-07-29 15:25:28 -04:00
|
|
|
end_addr = addr + PGDIR_SIZE;
|
2005-06-25 17:58:28 -04:00
|
|
|
while ((addr < last_addr) && (addr < end_addr)) {
|
2005-06-25 17:58:02 -04:00
|
|
|
struct page *page;
|
2005-07-29 15:25:28 -04:00
|
|
|
pmd_t *level2p;
|
2005-06-25 17:58:28 -04:00
|
|
|
|
2005-06-25 17:58:02 -04:00
|
|
|
page = kimage_alloc_control_pages(image, 0);
|
|
|
|
if (!page) {
|
|
|
|
result = -ENOMEM;
|
|
|
|
goto out;
|
|
|
|
}
|
2005-07-29 15:25:28 -04:00
|
|
|
level2p = (pmd_t *)page_address(page);
|
2005-06-25 17:58:02 -04:00
|
|
|
init_level2_page(level2p, addr);
|
2005-07-29 15:25:28 -04:00
|
|
|
set_pud(level3p++, __pud(__pa(level2p) | _KERNPG_TABLE));
|
|
|
|
addr += PUD_SIZE;
|
2005-06-25 17:58:02 -04:00
|
|
|
}
|
|
|
|
/* clear the unused entries */
|
2005-06-25 17:58:28 -04:00
|
|
|
while (addr < end_addr) {
|
2005-07-29 15:25:28 -04:00
|
|
|
pud_clear(level3p++);
|
|
|
|
addr += PUD_SIZE;
|
2005-06-25 17:58:02 -04:00
|
|
|
}
|
|
|
|
out:
|
|
|
|
return result;
|
|
|
|
}
|
|
|
|
|
|
|
|
|
2005-07-29 15:25:28 -04:00
|
|
|
static int init_level4_page(struct kimage *image, pgd_t *level4p,
|
2005-06-25 17:58:28 -04:00
|
|
|
unsigned long addr, unsigned long last_addr)
|
2005-06-25 17:58:02 -04:00
|
|
|
{
|
|
|
|
unsigned long end_addr;
|
|
|
|
int result;
|
2005-06-25 17:58:28 -04:00
|
|
|
|
2005-06-25 17:58:02 -04:00
|
|
|
result = 0;
|
|
|
|
addr &= PAGE_MASK;
|
2005-07-29 15:25:28 -04:00
|
|
|
end_addr = addr + (PTRS_PER_PGD * PGDIR_SIZE);
|
2005-06-25 17:58:28 -04:00
|
|
|
while ((addr < last_addr) && (addr < end_addr)) {
|
2005-06-25 17:58:02 -04:00
|
|
|
struct page *page;
|
2005-07-29 15:25:28 -04:00
|
|
|
pud_t *level3p;
|
2005-06-25 17:58:28 -04:00
|
|
|
|
2005-06-25 17:58:02 -04:00
|
|
|
page = kimage_alloc_control_pages(image, 0);
|
|
|
|
if (!page) {
|
|
|
|
result = -ENOMEM;
|
|
|
|
goto out;
|
|
|
|
}
|
2005-07-29 15:25:28 -04:00
|
|
|
level3p = (pud_t *)page_address(page);
|
2005-06-25 17:58:02 -04:00
|
|
|
result = init_level3_page(image, level3p, addr, last_addr);
|
|
|
|
if (result) {
|
|
|
|
goto out;
|
|
|
|
}
|
2005-07-29 15:25:28 -04:00
|
|
|
set_pgd(level4p++, __pgd(__pa(level3p) | _KERNPG_TABLE));
|
|
|
|
addr += PGDIR_SIZE;
|
2005-06-25 17:58:02 -04:00
|
|
|
}
|
|
|
|
/* clear the unused entries */
|
2005-06-25 17:58:28 -04:00
|
|
|
while (addr < end_addr) {
|
2005-07-29 15:25:28 -04:00
|
|
|
pgd_clear(level4p++);
|
|
|
|
addr += PGDIR_SIZE;
|
2005-06-25 17:58:02 -04:00
|
|
|
}
|
2005-06-25 17:58:28 -04:00
|
|
|
out:
|
2005-06-25 17:58:02 -04:00
|
|
|
return result;
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
static int init_pgtable(struct kimage *image, unsigned long start_pgtable)
|
|
|
|
{
|
2005-07-29 15:25:28 -04:00
|
|
|
pgd_t *level4p;
|
|
|
|
level4p = (pgd_t *)__va(start_pgtable);
|
2005-06-25 17:58:28 -04:00
|
|
|
return init_level4_page(image, level4p, 0, end_pfn << PAGE_SHIFT);
|
2005-06-25 17:58:02 -04:00
|
|
|
}
|
|
|
|
|
|
|
|
static void set_idt(void *newidt, u16 limit)
|
|
|
|
{
|
2005-07-29 15:02:09 -04:00
|
|
|
struct desc_ptr curidt;
|
2005-06-25 17:58:02 -04:00
|
|
|
|
|
|
|
/* x86-64 supports unaliged loads & stores */
|
2005-07-29 15:02:09 -04:00
|
|
|
curidt.size = limit;
|
|
|
|
curidt.address = (unsigned long)newidt;
|
2005-06-25 17:58:02 -04:00
|
|
|
|
|
|
|
__asm__ __volatile__ (
|
2005-07-29 15:02:09 -04:00
|
|
|
"lidtq %0\n"
|
|
|
|
: : "m" (curidt)
|
2005-06-25 17:58:02 -04:00
|
|
|
);
|
|
|
|
};
|
|
|
|
|
|
|
|
|
|
|
|
static void set_gdt(void *newgdt, u16 limit)
|
|
|
|
{
|
2005-07-29 15:02:09 -04:00
|
|
|
struct desc_ptr curgdt;
|
2005-06-25 17:58:02 -04:00
|
|
|
|
|
|
|
/* x86-64 supports unaligned loads & stores */
|
2005-07-29 15:02:09 -04:00
|
|
|
curgdt.size = limit;
|
|
|
|
curgdt.address = (unsigned long)newgdt;
|
2005-06-25 17:58:02 -04:00
|
|
|
|
|
|
|
__asm__ __volatile__ (
|
2005-07-29 15:02:09 -04:00
|
|
|
"lgdtq %0\n"
|
|
|
|
: : "m" (curgdt)
|
2005-06-25 17:58:02 -04:00
|
|
|
);
|
|
|
|
};
|
|
|
|
|
|
|
|
static void load_segments(void)
|
|
|
|
{
|
|
|
|
__asm__ __volatile__ (
|
2005-07-29 15:02:09 -04:00
|
|
|
"\tmovl %0,%%ds\n"
|
|
|
|
"\tmovl %0,%%es\n"
|
|
|
|
"\tmovl %0,%%ss\n"
|
|
|
|
"\tmovl %0,%%fs\n"
|
|
|
|
"\tmovl %0,%%gs\n"
|
|
|
|
: : "a" (__KERNEL_DS)
|
2005-06-25 17:58:02 -04:00
|
|
|
);
|
|
|
|
}
|
|
|
|
|
2005-06-25 17:58:28 -04:00
|
|
|
typedef NORET_TYPE void (*relocate_new_kernel_t)(unsigned long indirection_page,
|
|
|
|
unsigned long control_code_buffer,
|
|
|
|
unsigned long start_address,
|
|
|
|
unsigned long pgtable) ATTRIB_NORET;
|
2005-06-25 17:58:02 -04:00
|
|
|
|
|
|
|
const extern unsigned char relocate_new_kernel[];
|
|
|
|
const extern unsigned long relocate_new_kernel_size;
|
|
|
|
|
|
|
|
int machine_kexec_prepare(struct kimage *image)
|
|
|
|
{
|
|
|
|
unsigned long start_pgtable, control_code_buffer;
|
|
|
|
int result;
|
|
|
|
|
|
|
|
/* Calculate the offsets */
|
2005-06-25 17:58:28 -04:00
|
|
|
start_pgtable = page_to_pfn(image->control_code_page) << PAGE_SHIFT;
|
2005-07-29 15:25:28 -04:00
|
|
|
control_code_buffer = start_pgtable + PAGE_SIZE;
|
2005-06-25 17:58:02 -04:00
|
|
|
|
|
|
|
/* Setup the identity mapped 64bit page table */
|
|
|
|
result = init_pgtable(image, start_pgtable);
|
2005-06-25 17:58:28 -04:00
|
|
|
if (result)
|
2005-06-25 17:58:02 -04:00
|
|
|
return result;
|
|
|
|
|
|
|
|
/* Place the code in the reboot code buffer */
|
2005-06-25 17:58:28 -04:00
|
|
|
memcpy(__va(control_code_buffer), relocate_new_kernel,
|
|
|
|
relocate_new_kernel_size);
|
2005-06-25 17:58:02 -04:00
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
void machine_kexec_cleanup(struct kimage *image)
|
|
|
|
{
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* 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;
|
|
|
|
unsigned long control_code_buffer;
|
|
|
|
unsigned long start_pgtable;
|
|
|
|
relocate_new_kernel_t rnk;
|
|
|
|
|
|
|
|
/* Interrupts aren't acceptable while we reboot */
|
|
|
|
local_irq_disable();
|
|
|
|
|
|
|
|
/* Calculate the offsets */
|
2005-06-25 17:58:28 -04:00
|
|
|
page_list = image->head;
|
|
|
|
start_pgtable = page_to_pfn(image->control_code_page) << PAGE_SHIFT;
|
2005-07-29 15:25:28 -04:00
|
|
|
control_code_buffer = start_pgtable + PAGE_SIZE;
|
2005-06-25 17:58:02 -04:00
|
|
|
|
|
|
|
/* Set the low half of the page table to my identity mapped
|
|
|
|
* page table for kexec. Leave the high half pointing at the
|
|
|
|
* kernel pages. Don't bother to flush the global pages
|
|
|
|
* as that will happen when I fully switch to my identity mapped
|
|
|
|
* page table anyway.
|
|
|
|
*/
|
|
|
|
memcpy(__va(read_cr3()), __va(start_pgtable), PAGE_SIZE/2);
|
|
|
|
__flush_tlb();
|
|
|
|
|
|
|
|
|
|
|
|
/* The segment registers are funny things, they are
|
|
|
|
* automatically loaded from a table, in memory wherever you
|
|
|
|
* set them to a specific selector, but this table is never
|
|
|
|
* accessed again unless you set the segment to a different selector.
|
|
|
|
*
|
|
|
|
* The more common model are caches where the behide
|
|
|
|
* the scenes work is done, but is also dropped at arbitrary
|
|
|
|
* times.
|
|
|
|
*
|
|
|
|
* 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 */
|
|
|
|
rnk = (relocate_new_kernel_t) control_code_buffer;
|
|
|
|
(*rnk)(page_list, control_code_buffer, image->start, start_pgtable);
|
|
|
|
}
|