From 7aad25b4b47ea5b67e1eb8be0db211b899dce60d Mon Sep 17 00:00:00 2001 From: Kairui Song Date: Wed, 22 May 2024 01:58:53 +0800 Subject: mm/swap: reduce swap cache search space Currently we use one swap_address_space for every 64M chunk to reduce lock contention, this is like having a set of smaller swap files inside one swap device. But when doing swap cache look up or insert, we are still using the offset of the whole large swap device. This is OK for correctness, as the offset (key) is unique. But Xarray is specially optimized for small indexes, it creates the radix tree levels lazily to be just enough to fit the largest key stored in one Xarray. So we are wasting tree nodes unnecessarily. For 64M chunk it should only take at most 3 levels to contain everything. But if we are using the offset from the whole swap device, the offset (key) value will be way beyond 64M, and so will the tree level. Optimize this by using a new helper swap_cache_index to get a swap entry's unique offset in its own 64M swap_address_space. I see a ~1% performance gain in benchmark and actual workload with high memory pressure. Test with `time memhog 128G` inside a 8G memcg using 128G swap (ramdisk with SWP_SYNCHRONOUS_IO dropped, tested 3 times, results are stable. The test result is similar but the improvement is smaller if SWP_SYNCHRONOUS_IO is enabled, as swap out path can never skip swap cache): Before: 6.07user 250.74system 4:17.26elapsed 99%CPU (0avgtext+0avgdata 8373376maxresident)k 0inputs+0outputs (55major+33555018minor)pagefaults 0swaps After (1.8% faster): 6.08user 246.09system 4:12.58elapsed 99%CPU (0avgtext+0avgdata 8373248maxresident)k 0inputs+0outputs (54major+33555027minor)pagefaults 0swaps Similar result with MySQL and sysbench using swap: Before: 94055.61 qps After (0.8% faster): 94834.91 qps Radix tree slab usage is also very slightly lower. Link: https://lkml.kernel.org/r/20240521175854.96038-12-ryncsn@gmail.com Signed-off-by: Kairui Song Reviewed-by: "Huang, Ying" Cc: Anna Schumaker Cc: Barry Song Cc: Chao Yu Cc: Chris Li Cc: David Hildenbrand Cc: David Howells Cc: Hugh Dickins Cc: Ilya Dryomov Cc: Jaegeuk Kim Cc: Jeff Layton Cc: Marc Dionne Cc: Matthew Wilcox (Oracle) Cc: Minchan Kim Cc: NeilBrown Cc: Ryan Roberts Cc: Ryusuke Konishi Cc: Trond Myklebust Cc: Xiubo Li Cc: Yosry Ahmed Signed-off-by: Andrew Morton --- mm/memcontrol.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'mm/memcontrol.c') diff --git a/mm/memcontrol.c b/mm/memcontrol.c index 71fe2a95b8bd..86a741d8a37d 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -6146,7 +6146,7 @@ static struct page *mc_handle_swap_pte(struct vm_area_struct *vma, * Because swap_cache_get_folio() updates some statistics counter, * we call find_get_page() with swapper_space directly. */ - page = find_get_page(swap_address_space(ent), swp_offset(ent)); + page = find_get_page(swap_address_space(ent), swap_cache_index(ent)); entry->val = ent.val; return page; -- cgit v1.2.3-73-gaa49b