android_kernel_motorola_sm6225/arch/i386/mm/pgtable.c
Hugh Dickins 4c21e2f244 [PATCH] mm: split page table lock
Christoph Lameter demonstrated very poor scalability on the SGI 512-way, with
a many-threaded application which concurrently initializes different parts of
a large anonymous area.

This patch corrects that, by using a separate spinlock per page table page, to
guard the page table entries in that page, instead of using the mm's single
page_table_lock.  (But even then, page_table_lock is still used to guard page
table allocation, and anon_vma allocation.)

In this implementation, the spinlock is tucked inside the struct page of the
page table page: with a BUILD_BUG_ON in case it overflows - which it would in
the case of 32-bit PA-RISC with spinlock debugging enabled.

Splitting the lock is not quite for free: another cacheline access.  Ideally,
I suppose we would use split ptlock only for multi-threaded processes on
multi-cpu machines; but deciding that dynamically would have its own costs.
So for now enable it by config, at some number of cpus - since the Kconfig
language doesn't support inequalities, let preprocessor compare that with
NR_CPUS.  But I don't think it's worth being user-configurable: for good
testing of both split and unsplit configs, split now at 4 cpus, and perhaps
change that to 8 later.

There is a benefit even for singly threaded processes: kswapd can be attacking
one part of the mm while another part is busy faulting.

Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-10-29 21:40:42 -07:00

268 lines
6.8 KiB
C

/*
* linux/arch/i386/mm/pgtable.c
*/
#include <linux/config.h>
#include <linux/sched.h>
#include <linux/kernel.h>
#include <linux/errno.h>
#include <linux/mm.h>
#include <linux/swap.h>
#include <linux/smp.h>
#include <linux/highmem.h>
#include <linux/slab.h>
#include <linux/pagemap.h>
#include <linux/spinlock.h>
#include <asm/system.h>
#include <asm/pgtable.h>
#include <asm/pgalloc.h>
#include <asm/fixmap.h>
#include <asm/e820.h>
#include <asm/tlb.h>
#include <asm/tlbflush.h>
void show_mem(void)
{
int total = 0, reserved = 0;
int shared = 0, cached = 0;
int highmem = 0;
struct page *page;
pg_data_t *pgdat;
unsigned long i;
struct page_state ps;
printk(KERN_INFO "Mem-info:\n");
show_free_areas();
printk(KERN_INFO "Free swap: %6ldkB\n", nr_swap_pages<<(PAGE_SHIFT-10));
for_each_pgdat(pgdat) {
for (i = 0; i < pgdat->node_spanned_pages; ++i) {
page = pgdat_page_nr(pgdat, i);
total++;
if (PageHighMem(page))
highmem++;
if (PageReserved(page))
reserved++;
else if (PageSwapCache(page))
cached++;
else if (page_count(page))
shared += page_count(page) - 1;
}
}
printk(KERN_INFO "%d pages of RAM\n", total);
printk(KERN_INFO "%d pages of HIGHMEM\n", highmem);
printk(KERN_INFO "%d reserved pages\n", reserved);
printk(KERN_INFO "%d pages shared\n", shared);
printk(KERN_INFO "%d pages swap cached\n", cached);
get_page_state(&ps);
printk(KERN_INFO "%lu pages dirty\n", ps.nr_dirty);
printk(KERN_INFO "%lu pages writeback\n", ps.nr_writeback);
printk(KERN_INFO "%lu pages mapped\n", ps.nr_mapped);
printk(KERN_INFO "%lu pages slab\n", ps.nr_slab);
printk(KERN_INFO "%lu pages pagetables\n", ps.nr_page_table_pages);
}
/*
* Associate a virtual page frame with a given physical page frame
* and protection flags for that frame.
*/
static void set_pte_pfn(unsigned long vaddr, unsigned long pfn, pgprot_t flags)
{
pgd_t *pgd;
pud_t *pud;
pmd_t *pmd;
pte_t *pte;
pgd = swapper_pg_dir + pgd_index(vaddr);
if (pgd_none(*pgd)) {
BUG();
return;
}
pud = pud_offset(pgd, vaddr);
if (pud_none(*pud)) {
BUG();
return;
}
pmd = pmd_offset(pud, vaddr);
if (pmd_none(*pmd)) {
BUG();
return;
}
pte = pte_offset_kernel(pmd, vaddr);
/* <pfn,flags> stored as-is, to permit clearing entries */
set_pte(pte, pfn_pte(pfn, flags));
/*
* It's enough to flush this one mapping.
* (PGE mappings get flushed as well)
*/
__flush_tlb_one(vaddr);
}
/*
* Associate a large virtual page frame with a given physical page frame
* and protection flags for that frame. pfn is for the base of the page,
* vaddr is what the page gets mapped to - both must be properly aligned.
* The pmd must already be instantiated. Assumes PAE mode.
*/
void set_pmd_pfn(unsigned long vaddr, unsigned long pfn, pgprot_t flags)
{
pgd_t *pgd;
pud_t *pud;
pmd_t *pmd;
if (vaddr & (PMD_SIZE-1)) { /* vaddr is misaligned */
printk(KERN_WARNING "set_pmd_pfn: vaddr misaligned\n");
return; /* BUG(); */
}
if (pfn & (PTRS_PER_PTE-1)) { /* pfn is misaligned */
printk(KERN_WARNING "set_pmd_pfn: pfn misaligned\n");
return; /* BUG(); */
}
pgd = swapper_pg_dir + pgd_index(vaddr);
if (pgd_none(*pgd)) {
printk(KERN_WARNING "set_pmd_pfn: pgd_none\n");
return; /* BUG(); */
}
pud = pud_offset(pgd, vaddr);
pmd = pmd_offset(pud, vaddr);
set_pmd(pmd, pfn_pmd(pfn, flags));
/*
* It's enough to flush this one mapping.
* (PGE mappings get flushed as well)
*/
__flush_tlb_one(vaddr);
}
void __set_fixmap (enum fixed_addresses idx, unsigned long phys, pgprot_t flags)
{
unsigned long address = __fix_to_virt(idx);
if (idx >= __end_of_fixed_addresses) {
BUG();
return;
}
set_pte_pfn(address, phys >> PAGE_SHIFT, flags);
}
pte_t *pte_alloc_one_kernel(struct mm_struct *mm, unsigned long address)
{
return (pte_t *)__get_free_page(GFP_KERNEL|__GFP_REPEAT|__GFP_ZERO);
}
struct page *pte_alloc_one(struct mm_struct *mm, unsigned long address)
{
struct page *pte;
#ifdef CONFIG_HIGHPTE
pte = alloc_pages(GFP_KERNEL|__GFP_HIGHMEM|__GFP_REPEAT|__GFP_ZERO, 0);
#else
pte = alloc_pages(GFP_KERNEL|__GFP_REPEAT|__GFP_ZERO, 0);
#endif
return pte;
}
void pmd_ctor(void *pmd, kmem_cache_t *cache, unsigned long flags)
{
memset(pmd, 0, PTRS_PER_PMD*sizeof(pmd_t));
}
/*
* List of all pgd's needed for non-PAE so it can invalidate entries
* in both cached and uncached pgd's; not needed for PAE since the
* kernel pmd is shared. If PAE were not to share the pmd a similar
* tactic would be needed. This is essentially codepath-based locking
* against pageattr.c; it is the unique case in which a valid change
* of kernel pagetables can't be lazily synchronized by vmalloc faults.
* vmalloc faults work because attached pagetables are never freed.
* The locking scheme was chosen on the basis of manfred's
* recommendations and having no core impact whatsoever.
* -- wli
*/
DEFINE_SPINLOCK(pgd_lock);
struct page *pgd_list;
static inline void pgd_list_add(pgd_t *pgd)
{
struct page *page = virt_to_page(pgd);
page->index = (unsigned long)pgd_list;
if (pgd_list)
set_page_private(pgd_list, (unsigned long)&page->index);
pgd_list = page;
set_page_private(page, (unsigned long)&pgd_list);
}
static inline void pgd_list_del(pgd_t *pgd)
{
struct page *next, **pprev, *page = virt_to_page(pgd);
next = (struct page *)page->index;
pprev = (struct page **)page_private(page);
*pprev = next;
if (next)
set_page_private(next, (unsigned long)pprev);
}
void pgd_ctor(void *pgd, kmem_cache_t *cache, unsigned long unused)
{
unsigned long flags;
if (PTRS_PER_PMD == 1) {
memset(pgd, 0, USER_PTRS_PER_PGD*sizeof(pgd_t));
spin_lock_irqsave(&pgd_lock, flags);
}
clone_pgd_range((pgd_t *)pgd + USER_PTRS_PER_PGD,
swapper_pg_dir + USER_PTRS_PER_PGD,
KERNEL_PGD_PTRS);
if (PTRS_PER_PMD > 1)
return;
pgd_list_add(pgd);
spin_unlock_irqrestore(&pgd_lock, flags);
}
/* never called when PTRS_PER_PMD > 1 */
void pgd_dtor(void *pgd, kmem_cache_t *cache, unsigned long unused)
{
unsigned long flags; /* can be called from interrupt context */
spin_lock_irqsave(&pgd_lock, flags);
pgd_list_del(pgd);
spin_unlock_irqrestore(&pgd_lock, flags);
}
pgd_t *pgd_alloc(struct mm_struct *mm)
{
int i;
pgd_t *pgd = kmem_cache_alloc(pgd_cache, GFP_KERNEL);
if (PTRS_PER_PMD == 1 || !pgd)
return pgd;
for (i = 0; i < USER_PTRS_PER_PGD; ++i) {
pmd_t *pmd = kmem_cache_alloc(pmd_cache, GFP_KERNEL);
if (!pmd)
goto out_oom;
set_pgd(&pgd[i], __pgd(1 + __pa(pmd)));
}
return pgd;
out_oom:
for (i--; i >= 0; i--)
kmem_cache_free(pmd_cache, (void *)__va(pgd_val(pgd[i])-1));
kmem_cache_free(pgd_cache, pgd);
return NULL;
}
void pgd_free(pgd_t *pgd)
{
int i;
/* in the PAE case user pgd entries are overwritten before usage */
if (PTRS_PER_PMD > 1)
for (i = 0; i < USER_PTRS_PER_PGD; ++i)
kmem_cache_free(pmd_cache, (void *)__va(pgd_val(pgd[i])-1));
/* in the non-PAE case, free_pgtables() clears user pgd entries */
kmem_cache_free(pgd_cache, pgd);
}