From 7e8756d7ad22655b935c384f123071aa9de07a27 Mon Sep 17 00:00:00 2001 From: Johannes Weiner Date: Wed, 22 Jul 2026 10:56:47 -0400 Subject: [PATCH] mm: page_alloc: fix non-movable reclaim storm in defrag_mode As we deployed defrag_mode into Meta production, pressure spikes and excessive swapping were observed on some workloads. Tracing confirmed that this is unmovable/reclaimable requests spinning in the allocator and direct reclaim, causing excessive amounts of swap. The initial plan for defrag_mode was to rely on kswapd/kcompactd to produce blocks, and if those are overwhelmed under high pressure, let the allocator fall back (__rmqueue_steal()) after its retry loops. However, that retrying results in more reclaim on some of these workloads than we'd hoped, sometimes excessively so, spurred on by the !costly order conditions in should_reclaim_retry(). The storms are dependent on the request type. Reclaim will inevitably make room in existing movable blocks, since that's where the LRU pages live. So if movable requests retry on reclaim, they make progress. When non-movable requests spin in reclaim that isn't productive. They cannot use the individually freed pages, and the process is unlikely to accidentally free whole blocks to meet the ALLOC_NOFRAGMENT bar. They spin and overreclaim excessively, which tanks performance and triggers userspace guards like swap exhaustion or pressure based OOM. To fix this, send non-movable requests, regardless of order, into pageblock reclaim/compaction. This way, they help move things along to meet the ALLOC_NOFRAGMENT bar. After this patch, the reclaim storms and excess OOM rates are no longer observed in production. The longer-term plan is still to have all requests, including the movable ones, help make blocks to spread the cost of defragmenting more evenly and fairly; combined with proper watermarking to reduce allocation latencies in the common case. However, doing this naively unearths scaling and concurrency limitations in compaction that need to be addressed first. Promoting just non-movables for now is the minimally viable bug fix for the above issue. [brendan.jackman@linux.dev: fix try_to_compact_pages() kerneldoc] Link: https://lore.kernel.org/DK7NM9RPUJOD.11PNJJ5N2OBED@linux.dev Link: https://lore.kernel.org/20260722150006.3848560-5-hannes@cmpxchg.org Fixes: e3aa7df331bc ("mm: page_alloc: defrag_mode") Signed-off-by: Johannes Weiner Signed-off-by: "Brendan Jackman" Reviewed-by: Vlastimil Babka (SUSE) Cc: Brendan Jackman Cc: David Hildenbrand Cc: Gregory Price Cc: Liam R. Howlett Cc: Lorenzo Stoakes Cc: Michal Hocko Cc: Mike Rapoport Cc: Shakeel Butt Cc: Suren Baghdasaryan Cc: Zi Yan Cc: Signed-off-by: Andrew Morton --- mm/compaction.c | 2 +- mm/internal.h | 6 ++++++ mm/page_alloc.c | 31 ++++++++++++++++++++++++++----- 3 files changed, 33 insertions(+), 6 deletions(-) diff --git a/mm/compaction.c b/mm/compaction.c index 0568623d9384..a049415512c6 100644 --- a/mm/compaction.c +++ b/mm/compaction.c @@ -2823,7 +2823,7 @@ static enum compact_result compact_zone_order(struct zone *zone, int order, /** * try_to_compact_pages - Direct compact to satisfy a high-order allocation * @gfp_mask: The GFP mask of the current allocation - * @order: The order of the current allocation + * @order: The order to try to make available * @alloc_flags: The allocation flags of the current allocation * @ac: The context of current allocation * @prio: Determines how hard direct compaction should try to succeed diff --git a/mm/internal.h b/mm/internal.h index f28003f59812..68db5abd0a4c 100644 --- a/mm/internal.h +++ b/mm/internal.h @@ -870,6 +870,12 @@ struct compact_control { struct capture_control { struct zone *zone; int migratetype; + /* + * Allocation request order. May differ from the compaction + * order: defrag_mode promotes sub-block allocations to + * pageblock-order compaction; capture still matches at the + * original allocation order so prep_new_page() is consistent. + */ int order; struct page *page; }; diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 4905f3596998..083cbcb5bdde 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -4149,8 +4149,24 @@ __alloc_pages_direct_compact(gfp_t gfp_mask, unsigned int order, .order = order, .page = NULL, }; + int compact_order = order; - if (!order) + /* + * If fallbacks are not permitted (defrag_mode), we either + * need to reclaim space in a block of matching type, or clear + * out an entire block to allow __rmqueue_claim() to convert. + * + * Reclaim by itself is primarily freeing space in movable + * blocks, since that's where the LRU pages live. So this + * works for movable requests, but not for others. + * + * For those, promote the order to help make blocks, instead + * of spinning in reclaim alone unproductively. + */ + if ((alloc_flags & ALLOC_NOFRAGMENT) && ac->migratetype != MIGRATE_MOVABLE) + compact_order = max(order, pageblock_order); + + if (!compact_order) return NULL; psi_memstall_enter(&pflags); @@ -4166,8 +4182,8 @@ __alloc_pages_direct_compact(gfp_t gfp_mask, unsigned int order, barrier(); WRITE_ONCE(current->capture_control, &capc); - *compact_result = try_to_compact_pages(gfp_mask, order, alloc_flags, ac, - prio, &capc); + *compact_result = try_to_compact_pages(gfp_mask, compact_order, + alloc_flags, ac, prio, &capc); /* * Make sure we hide capture control first before we read the captured @@ -4212,7 +4228,7 @@ __alloc_pages_direct_compact(gfp_t gfp_mask, unsigned int order, struct zone *zone = page_zone(page); zone->compact_blockskip_flush = false; - compaction_defer_reset(zone, order, true); + compaction_defer_reset(zone, compact_order, true); count_vm_event(COMPACTSUCCESS); return page; } @@ -4452,9 +4468,14 @@ __alloc_pages_direct_reclaim(gfp_t gfp_mask, unsigned int order, struct page *page = NULL; unsigned long pflags; bool drained = false; + int reclaim_order = order; + + /* Match the slowpath compaction promotion in __alloc_pages_direct_compact */ + if ((alloc_flags & ALLOC_NOFRAGMENT) && ac->migratetype != MIGRATE_MOVABLE) + reclaim_order = max(order, pageblock_order); psi_memstall_enter(&pflags); - *did_some_progress = __perform_reclaim(gfp_mask, order, ac); + *did_some_progress = __perform_reclaim(gfp_mask, reclaim_order, ac); if (unlikely(!(*did_some_progress))) goto out;