diff options
| author | Kirill A. Shutemov <kirill.shutemov@linux.intel.com> | 2013-11-14 14:31:53 -0800 |
|---|---|---|
| committer | Linus Torvalds <torvalds@linux-foundation.org> | 2013-11-15 09:32:20 +0900 |
| commit | ea1e7ed33708c7a760419ff9ded0a6cb90586a50 (patch) | |
| tree | 553c0c4024639bf2b1706d144fd6bb8ada63861e /mm/memory.c | |
| parent | 539edb5846c740d78a8b6c2e43a99ca4323df68f (diff) | |
| download | linux-ea1e7ed33708c7a760419ff9ded0a6cb90586a50.tar.gz linux-ea1e7ed33708c7a760419ff9ded0a6cb90586a50.tar.bz2 linux-ea1e7ed33708c7a760419ff9ded0a6cb90586a50.zip | |
mm: create a separate slab for page->ptl allocation
If DEBUG_SPINLOCK and DEBUG_LOCK_ALLOC are enabled spinlock_t on x86_64
is 72 bytes. For page->ptl they will be allocated from kmalloc-96 slab,
so we loose 24 on each. An average system can easily allocate few tens
thousands of page->ptl and overhead is significant.
Let's create a separate slab for page->ptl allocation to solve this.
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Diffstat (limited to 'mm/memory.c')
| -rw-r--r-- | mm/memory.c | 7 |
1 files changed, 7 insertions, 0 deletions
diff --git a/mm/memory.c b/mm/memory.c index 5d9025f3b3e1..0409e8f43fa0 100644 --- a/mm/memory.c +++ b/mm/memory.c @@ -4272,6 +4272,13 @@ void copy_user_huge_page(struct page *dst, struct page *src, #endif /* CONFIG_TRANSPARENT_HUGEPAGE || CONFIG_HUGETLBFS */ #if USE_SPLIT_PTE_PTLOCKS && BLOATED_SPINLOCKS +static struct kmem_cache *page_ptl_cachep; +void __init ptlock_cache_init(void) +{ + page_ptl_cachep = kmem_cache_create("page->ptl", sizeof(spinlock_t), 0, + SLAB_PANIC, NULL); +} + bool ptlock_alloc(struct page *page) { spinlock_t *ptl; |