From eae116d1f0449ade3269ca47a67432622f5c6438 Mon Sep 17 00:00:00 2001 From: Gabriel Krisman Bertazi Date: Tue, 25 Feb 2025 22:22:58 -0500 Subject: Revert "mm/page_alloc.c: don't show protection in zone's ->lowmem_reserve[] for empty zone" Commit 96a5c186efff ("mm/page_alloc.c: don't show protection in zone's ->lowmem_reserve[] for empty zone") removes the protection of lower zones from allocations targeting memory-less high zones. This had an unintended impact on the pattern of reclaims because it makes the high-zone-targeted allocation more likely to succeed in lower zones, which adds pressure to said zones. I.e, the following corresponding checks in zone_watermark_ok/zone_watermark_fast are less likely to trigger: if (free_pages <= min + z->lowmem_reserve[highest_zoneidx]) return false; As a result, we are observing an increase in reclaim and kswapd scans, due to the increased pressure. This was initially observed as increased latency in filesystem operations when benchmarking with fio on a machine with some memory-less zones, but it has since been associated with increased contention in locks related to memory reclaim. By reverting this patch, the original performance was recovered on that machine. The original commit was introduced as a clarification of the /proc/zoneinfo output, so it doesn't seem there are usecases depending on it, making the revert a simple solution. For reference, I collected vmstat with and without this patch on a freshly booted system running intensive randread io from an nvme for 5 minutes. I got: rpm-6.12.0-slfo.1.2 -> pgscan_kswapd 5629543865 Patched -> pgscan_kswapd 33580844 33M scans is similar to what we had in kernels predating this patch. These numbers is fairly representative of the workload on this machine, as measured in several runs. So we are talking about a 2-order of magnitude increase. Link: https://lkml.kernel.org/r/20250226032258.234099-1-krisman@suse.de Fixes: 96a5c186efff ("mm/page_alloc.c: don't show protection in zone's ->lowmem_reserve[] for empty zone") Signed-off-by: Gabriel Krisman Bertazi Reviewed-by: Vlastimil Babka Acked-by: Michal Hocko Acked-by: Mel Gorman Cc: Baoquan He Cc: Signed-off-by: Andrew Morton --- mm/page_alloc.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'mm/page_alloc.c') diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 579789600a3c..fe986e6de7a0 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -5849,11 +5849,10 @@ static void setup_per_zone_lowmem_reserve(void) for (j = i + 1; j < MAX_NR_ZONES; j++) { struct zone *upper_zone = &pgdat->node_zones[j]; - bool empty = !zone_managed_pages(upper_zone); managed_pages += zone_managed_pages(upper_zone); - if (clear || empty) + if (clear) zone->lowmem_reserve[j] = 0; else zone->lowmem_reserve[j] = managed_pages / ratio; -- cgit v1.2.3 From 8fe9ed44dc29fba0786b7e956d2e87179e407582 Mon Sep 17 00:00:00 2001 From: Hao Zhang Date: Thu, 27 Feb 2025 11:41:29 +0800 Subject: mm/page_alloc: fix uninitialized variable The variable "compact_result" is not initialized in function __alloc_pages_slowpath(). It causes should_compact_retry() to use an uninitialized value. Initialize variable "compact_result" with the value COMPACT_SKIPPED. BUG: KMSAN: uninit-value in __alloc_pages_slowpath+0xee8/0x16c0 mm/page_alloc.c:4416 __alloc_pages_slowpath+0xee8/0x16c0 mm/page_alloc.c:4416 __alloc_frozen_pages_noprof+0xa4c/0xe00 mm/page_alloc.c:4752 alloc_pages_mpol+0x4cd/0x890 mm/mempolicy.c:2270 alloc_frozen_pages_noprof mm/mempolicy.c:2341 [inline] alloc_pages_noprof mm/mempolicy.c:2361 [inline] folio_alloc_noprof+0x1dc/0x350 mm/mempolicy.c:2371 filemap_alloc_folio_noprof+0xa6/0x440 mm/filemap.c:1019 __filemap_get_folio+0xb9a/0x1840 mm/filemap.c:1970 grow_dev_folio fs/buffer.c:1039 [inline] grow_buffers fs/buffer.c:1105 [inline] __getblk_slow fs/buffer.c:1131 [inline] bdev_getblk+0x2c9/0xab0 fs/buffer.c:1431 getblk_unmovable include/linux/buffer_head.h:369 [inline] ext4_getblk+0x3b7/0xe50 fs/ext4/inode.c:864 ext4_bread_batch+0x9f/0x7d0 fs/ext4/inode.c:933 __ext4_find_entry+0x1ebb/0x36c0 fs/ext4/namei.c:1627 ext4_lookup_entry fs/ext4/namei.c:1729 [inline] ext4_lookup+0x189/0xb40 fs/ext4/namei.c:1797 __lookup_slow+0x538/0x710 fs/namei.c:1793 lookup_slow+0x6a/0xd0 fs/namei.c:1810 walk_component fs/namei.c:2114 [inline] link_path_walk+0xf29/0x1420 fs/namei.c:2479 path_openat+0x30f/0x6250 fs/namei.c:3985 do_filp_open+0x268/0x600 fs/namei.c:4016 do_sys_openat2+0x1bf/0x2f0 fs/open.c:1428 do_sys_open fs/open.c:1443 [inline] __do_sys_openat fs/open.c:1459 [inline] __se_sys_openat fs/open.c:1454 [inline] __x64_sys_openat+0x2a1/0x310 fs/open.c:1454 x64_sys_call+0x36f5/0x3c30 arch/x86/include/generated/asm/syscalls_64.h:258 do_syscall_x64 arch/x86/entry/common.c:52 [inline] do_syscall_64+0xcd/0x1e0 arch/x86/entry/common.c:83 entry_SYSCALL_64_after_hwframe+0x77/0x7f Local variable compact_result created at: __alloc_pages_slowpath+0x66/0x16c0 mm/page_alloc.c:4218 __alloc_frozen_pages_noprof+0xa4c/0xe00 mm/page_alloc.c:4752 Link: https://lkml.kernel.org/r/tencent_ED1032321D6510B145CDBA8CBA0093178E09@qq.com Reported-by: syzbot+0cfd5e38e96a5596f2b6@syzkaller.appspotmail.com Closes: https://syzkaller.appspot.com/bug?extid=0cfd5e38e96a5596f2b6 Signed-off-by: Hao Zhang Reviewed-by: Vlastimil Babka Cc: Michal Hocko Cc: Mel Gorman Cc: Signed-off-by: Andrew Morton --- mm/page_alloc.c | 1 + 1 file changed, 1 insertion(+) (limited to 'mm/page_alloc.c') diff --git a/mm/page_alloc.c b/mm/page_alloc.c index fe986e6de7a0..94917c729120 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -4243,6 +4243,7 @@ __alloc_pages_slowpath(gfp_t gfp_mask, unsigned int order, restart: compaction_retries = 0; no_progress_loops = 0; + compact_result = COMPACT_SKIPPED; compact_priority = DEF_COMPACT_PRIORITY; cpuset_mems_cookie = read_mems_allowed_begin(); zonelist_iter_cookie = zonelist_iter_begin(); -- cgit v1.2.3