linux/arch/x86/boot/compressed/pagetable.c
Xunlei Pang 66aad4fdf2 x86/mm: Add support for gbpages to kernel_ident_mapping_init()
Kernel identity mappings on x86-64 kernels are created in two
ways: by the early x86 boot code, or by kernel_ident_mapping_init().

Native kernels (which is the dominant usecase) use the former,
but the kexec and the hibernation code uses kernel_ident_mapping_init().

There's a subtle difference between these two ways of how identity
mappings are created, the current kernel_ident_mapping_init() code
creates identity mappings always using 2MB page(PMD level) - while
the native kernel boot path also utilizes gbpages where available.

This difference is suboptimal both for performance and for memory
usage: kernel_ident_mapping_init() needs to allocate pages for the
page tables when creating the new identity mappings.

This patch adds 1GB page(PUD level) support to kernel_ident_mapping_init()
to address these concerns.

The primary advantage would be better TLB coverage/performance,
because we'd utilize 1GB TLBs instead of 2MB ones.

It is also useful for machines with large number of memory to
save paging structure allocations(around 4MB/TB using 2MB page)
when setting identity mappings for all the memory, after using
1GB page it will consume only 8KB/TB.

( Note that this change alone does not activate gbpages in kexec,
  we are doing that in a separate patch. )

Signed-off-by: Xunlei Pang <xlpang@redhat.com>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Brian Gerst <brgerst@gmail.com>
Cc: Dave Young <dyoung@redhat.com>
Cc: Denys Vlasenko <dvlasenk@redhat.com>
Cc: Eric Biederman <ebiederm@xmission.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Josh Poimboeuf <jpoimboe@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Yinghai Lu <yinghai@kernel.org>
Cc: akpm@linux-foundation.org
Cc: kexec@lists.infradead.org
Link: http://lkml.kernel.org/r/1493862171-8799-1-git-send-email-xlpang@redhat.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
2017-05-08 08:28:40 +02:00

139 lines
4.0 KiB
C

/*
* This code is used on x86_64 to create page table identity mappings on
* demand by building up a new set of page tables (or appending to the
* existing ones), and then switching over to them when ready.
*
* Copyright (C) 2015-2016 Yinghai Lu
* Copyright (C) 2016 Kees Cook
*/
/*
* Since we're dealing with identity mappings, physical and virtual
* addresses are the same, so override these defines which are ultimately
* used by the headers in misc.h.
*/
#define __pa(x) ((unsigned long)(x))
#define __va(x) ((void *)((unsigned long)(x)))
#include "misc.h"
/* These actually do the work of building the kernel identity maps. */
#include <asm/init.h>
#include <asm/pgtable.h>
/* Use the static base for this part of the boot process */
#undef __PAGE_OFFSET
#define __PAGE_OFFSET __PAGE_OFFSET_BASE
#include "../../mm/ident_map.c"
/* Used by pgtable.h asm code to force instruction serialization. */
unsigned long __force_order;
/* Used to track our page table allocation area. */
struct alloc_pgt_data {
unsigned char *pgt_buf;
unsigned long pgt_buf_size;
unsigned long pgt_buf_offset;
};
/*
* Allocates space for a page table entry, using struct alloc_pgt_data
* above. Besides the local callers, this is used as the allocation
* callback in mapping_info below.
*/
static void *alloc_pgt_page(void *context)
{
struct alloc_pgt_data *pages = (struct alloc_pgt_data *)context;
unsigned char *entry;
/* Validate there is space available for a new page. */
if (pages->pgt_buf_offset >= pages->pgt_buf_size) {
debug_putstr("out of pgt_buf in " __FILE__ "!?\n");
debug_putaddr(pages->pgt_buf_offset);
debug_putaddr(pages->pgt_buf_size);
return NULL;
}
entry = pages->pgt_buf + pages->pgt_buf_offset;
pages->pgt_buf_offset += PAGE_SIZE;
return entry;
}
/* Used to track our allocated page tables. */
static struct alloc_pgt_data pgt_data;
/* The top level page table entry pointer. */
static unsigned long level4p;
/*
* Mapping information structure passed to kernel_ident_mapping_init().
* Due to relocation, pointers must be assigned at run time not build time.
*/
static struct x86_mapping_info mapping_info = {
.page_flag = __PAGE_KERNEL_LARGE_EXEC,
};
/* Locates and clears a region for a new top level page table. */
void initialize_identity_maps(void)
{
/* Init mapping_info with run-time function/buffer pointers. */
mapping_info.alloc_pgt_page = alloc_pgt_page;
mapping_info.context = &pgt_data;
/*
* It should be impossible for this not to already be true,
* but since calling this a second time would rewind the other
* counters, let's just make sure this is reset too.
*/
pgt_data.pgt_buf_offset = 0;
/*
* If we came here via startup_32(), cr3 will be _pgtable already
* and we must append to the existing area instead of entirely
* overwriting it.
*/
level4p = read_cr3();
if (level4p == (unsigned long)_pgtable) {
debug_putstr("booted via startup_32()\n");
pgt_data.pgt_buf = _pgtable + BOOT_INIT_PGT_SIZE;
pgt_data.pgt_buf_size = BOOT_PGT_SIZE - BOOT_INIT_PGT_SIZE;
memset(pgt_data.pgt_buf, 0, pgt_data.pgt_buf_size);
} else {
debug_putstr("booted via startup_64()\n");
pgt_data.pgt_buf = _pgtable;
pgt_data.pgt_buf_size = BOOT_PGT_SIZE;
memset(pgt_data.pgt_buf, 0, pgt_data.pgt_buf_size);
level4p = (unsigned long)alloc_pgt_page(&pgt_data);
}
}
/*
* Adds the specified range to what will become the new identity mappings.
* Once all ranges have been added, the new mapping is activated by calling
* finalize_identity_maps() below.
*/
void add_identity_map(unsigned long start, unsigned long size)
{
unsigned long end = start + size;
/* Align boundary to 2M. */
start = round_down(start, PMD_SIZE);
end = round_up(end, PMD_SIZE);
if (start >= end)
return;
/* Build the mapping. */
kernel_ident_mapping_init(&mapping_info, (pgd_t *)level4p,
start, end);
}
/*
* This switches the page tables to the new level4 that has been built
* via calls to add_identity_map() above. If booted via startup_32(),
* this is effectively a no-op.
*/
void finalize_identity_maps(void)
{
write_cr3(level4p);
}