From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from smtp.kernel.org (aws-us-west-2-korg-mail-alma10-1.taild15c8.ts.net [100.103.45.18]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 33F2A443C29 for ; Tue, 29 Sep 2026 19:58:47 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=100.103.45.18 ARC-Seal:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1790711928; cv=none; b=N3URJN5h3hCpSwXcaoL9IgA7UH7JkMg2MpSzRTcSJMoGQ716HOkn2JSbGR87lsdLP0xh3D5pVkKtQujyrgUMBu1OskrEVZNGu7ESJIE5UFayBEJeh2JuqNZZxDSvBp3TNFxa6QVt2ilN9j7FgPFhbrXtzEiCBxlNlbkNd1KJaxo= ARC-Message-Signature:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1790711928; c=relaxed/simple; bh=TTaflg99ObC3x+A0xgtRgUqYOrYONrJpqzxYlUdsvvI=; h=Date:To:From:Subject:Message-Id; b=s54Zro/HxyGP+X52ZFvcIBa657knv+GhEGJjm5NoPYvVrjFBAAfue9w8Xw7ynp8/5M5+T9B8wyejfL548YR3AmSRjd3cQVKsTfxPqpUk+e3Z7RgcsbVRDSXCwCJycjFX2Z34spn2xnqbDoajCOoEl3WAW7+JTq7SUsv3dGwwrXk= ARC-Authentication-Results:i=1; smtp.subspace.kernel.org; dkim=pass (1024-bit key) header.d=linux-foundation.org header.i=@linux-foundation.org header.b=OZ4pAMAy; arc=none smtp.client-ip=100.103.45.18 Authentication-Results: smtp.subspace.kernel.org; dkim=pass (1024-bit key) header.d=linux-foundation.org header.i=@linux-foundation.org header.b="OZ4pAMAy" Received: by smtp.kernel.org (Postfix) with ESMTPSA id C783A1F000FF; Tue, 29 Sep 2026 19:58:46 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=linux-foundation.org; s=korg; t=1790711926; bh=Qv9e8KT3pG2eBgcevYQpPPedOvuxaIMy7W/Cv+/5npg=; h=Date:To:From:Subject; b=OZ4pAMAy2eS9dGBpBgmYg0bhgxDTE8dpd/yT57ykDYkoN4MLjpjHVnph9pFnoLKxU +iQLslYABU2BZV7QCRYGUVD23PHNYQpEd79mbQOvO2JGQOiGRSzEzHj2J1AUSKFCpx rBdOVd+N2dHorB2s7yCk50EFXlKiK4v+30gE0lrQ= Date: Tue, 29 Sep 2026 12:58:46 -0700 To: mm-commits@vger.kernel.org,ziy@nvidia.com,vbabka@kernel.org,usama.arif@linux.dev,surenb@google.com,shakeel.butt@linux.dev,mhocko@suse.com,harry@kernel.org,hannes@cmpxchg.org,david@kernel.org,brendan.jackman@linux.dev,kas@kernel.org,akpm@linux-foundation.org From: Andrew Morton Subject: + mm-page_alloc-make-defrag_mode-retries-follow-the-promoted-order.patch added to mm-hotfixes-unstable branch Message-Id: <20260929195846.C783A1F000FF@smtp.kernel.org> Precedence: bulk X-Mailing-List: mm-commits@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: The patch titled Subject: mm: page_alloc: make defrag_mode retries follow the promoted order has been added to the -mm mm-hotfixes-unstable branch. Its filename is mm-page_alloc-make-defrag_mode-retries-follow-the-promoted-order.patch This patch will shortly appear at https://git.kernel.org/pub/scm/linux/kernel/git/akpm/25-new.git/tree/patches/mm-page_alloc-make-defrag_mode-retries-follow-the-promoted-order.patch This patch will later appear in the mm-hotfixes-unstable branch at git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm Before you just go and hit "reply", please: a) Consider who else should be cc'ed b) Prefer to cc a suitable mailing list as well c) Ideally: find the original patch on the mailing list and do a reply-to-all to that, adding suitable additional cc's *** Remember to use Documentation/process/submit-checklist.rst when testing your code *** The -mm tree is included into linux-next via various branches at git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm and is updated there most days ------------------------------------------------------ From: "Kiryl Shutsemau (Meta)" Subject: mm: page_alloc: make defrag_mode retries follow the promoted order Date: Tue, 29 Sep 2026 18:45:51 +0100 Since commit 7e8756d7ad22 ("mm: page_alloc: fix non-movable reclaim storm in defrag_mode"), direct reclaim and compaction for non-movable requests under defrag_mode run at pageblock_order, to produce the whole blocks that ALLOC_NOFRAGMENT needs. The retry decisions that follow still use the request order. An order-0 request can therefore retry indefinitely without ever reaching the ALLOC_NOFRAGMENT fallback: - Reclaim at pageblock_order gives up after one pass as soon as a zone looks compaction_ready(), and do_try_to_free_pages() then returns 1 even though nothing was reclaimed. It returns before the retry that would reclaim memory.low-protected cgroups, so when most memory is protected, the pass that did run finds next to nothing. - Compaction at pageblock_order fails or is deferred. - should_reclaim_retry() takes the reported progress as progress for the order-0 request and resets no_progress_loops. The request retries. Order 1-3 requests loop the same way, and should_compact_retry() also checks their pageblock_order compaction result against the request order. On a production host (64G, defrag_mode, memory.low covering most of the workload), 95% of direct reclaim runs were order-9 runs that returned 1 with nothing reclaimed, at up to 60k runs per second. Across ~200M should_reclaim_retry() calls in a day, no_progress_loops never left 0. The spinning allocations were SLUB slab refills for inode and dentry caches. The time spent registers as memory pressure, and pressure-based OOM killing takes down both workloads and system services. Treat promoted requests like costly orders: - Reclaim progress does not reset no_progress_loops for them. - should_compact_retry() checks the compaction result at the promoted order. It does not retry COMPACT_SKIPPED, since the request can fall back, and it does not escalate compaction to COMPACT_PRIO_SYNC_FULL. When the fallback is taken, reset the retry counters, so that the fallback attempt gets a full retry budget before the OOM killer is considered. In a VM reproducer (32G, defrag_mode, inode churn under memory.low): before after should_reclaim_retry() calls 63M 293k peak memory pressure (PSI some avg10) 99% 12% File creation runs 5.7x faster. Link: https://lore.kernel.org/20260929174553.175333-1-kirill@shutemov.name Fixes: 7e8756d7ad22 ("mm: page_alloc: fix non-movable reclaim storm in defrag_mode") Signed-off-by: Kiryl Shutsemau (Meta) Signed-off-by: Andrew Morton Assisted-by: LLM Cc: Vlastimil Babka Cc: Johannes Weiner Cc: David Hildenbrand Cc: Suren Baghdasaryan Cc: Michal Hocko Cc: Brendan Jackman Cc: Zi Yan Cc: Shakeel Butt Cc: Usama Arif Cc: Harry Yoo --- mm/page_alloc.c | 85 ++++++++++++++++++++++++++++++---------------- 1 file changed, 56 insertions(+), 29 deletions(-) --- a/mm/page_alloc.c~mm-page_alloc-make-defrag_mode-retries-follow-the-promoted-order +++ a/mm/page_alloc.c @@ -4128,6 +4128,31 @@ out: } /* + * 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 of reclaim and compaction to help make + * blocks, instead of spinning in reclaim alone unproductively. Retry + * decisions based on the outcome of that work - reclaim progress and + * compaction results - must account for the promotion as well, see + * should_reclaim_retry() and should_compact_retry(). + */ +static inline unsigned int nofrag_promote_order(unsigned int order, + unsigned int alloc_flags, + const struct alloc_context *ac) +{ + if ((alloc_flags & ALLOC_NOFRAGMENT) && ac->migratetype != MIGRATE_MOVABLE) + return max(order, pageblock_order); + + return order; +} + +/* * Maximum number of compaction retries with a progress before OOM * killer is consider as the only way to move forward. */ @@ -4149,22 +4174,7 @@ __alloc_pages_direct_compact(gfp_t gfp_m .order = order, .page = NULL, }; - int compact_order = 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); + unsigned int compact_order = nofrag_promote_order(order, alloc_flags, ac); if (!compact_order) return NULL; @@ -4256,8 +4266,11 @@ should_compact_retry(gfp_t gfp_mask, str bool ret = false; int retries = *compaction_retries; enum compact_priority priority = *compact_priority; + unsigned int compact_order; - if (!order) + /* Check the compaction result at the order compaction ran at */ + compact_order = nofrag_promote_order(order, alloc_flags, ac); + if (!compact_order) return false; if (fatal_signal_pending(current)) @@ -4266,10 +4279,14 @@ should_compact_retry(gfp_t gfp_mask, str /* * Compaction was skipped due to a lack of free order-0 * migration targets. Continue if reclaim can help. + * + * Promoted requests have exhausted their reclaim retries at + * this point, and they can fall back instead. */ if (compact_result == COMPACT_SKIPPED) { - ret = compaction_zonelist_suitable(ac, order, alloc_flags, - gfp_mask); + if (compact_order == order) + ret = compaction_zonelist_suitable(ac, order, alloc_flags, + gfp_mask); goto out; } @@ -4287,7 +4304,7 @@ should_compact_retry(gfp_t gfp_mask, str * need much more detailed feedback from compaction to * make a better decision. */ - if (order > PAGE_ALLOC_COSTLY_ORDER) + if (compact_order > PAGE_ALLOC_COSTLY_ORDER) max_retries /= 4; if (++(*compaction_retries) <= max_retries) { @@ -4299,7 +4316,7 @@ should_compact_retry(gfp_t gfp_mask, str /* * Compaction failed. Retry with increasing priority. */ - min_priority = (order > PAGE_ALLOC_COSTLY_ORDER) ? + min_priority = (compact_order > PAGE_ALLOC_COSTLY_ORDER) ? MIN_COMPACT_COSTLY_PRIORITY : MIN_COMPACT_PRIORITY; if (*compact_priority > min_priority) { @@ -4468,11 +4485,7 @@ __alloc_pages_direct_reclaim(gfp_t gfp_m 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); + unsigned int reclaim_order = nofrag_promote_order(order, alloc_flags, ac); psi_memstall_enter(&pflags); *did_some_progress = __perform_reclaim(gfp_mask, reclaim_order, ac); @@ -4648,9 +4661,17 @@ should_reclaim_retry(gfp_t gfp_mask, uns /* * Costly allocations might have made a progress but this doesn't mean * their order will become available due to high fragmentation so - * always increment the no progress counter for them + * always increment the no progress counter for them. + * + * The same goes for requests whose reclaim is promoted to make whole + * blocks. At that order, reclaim also reports progress when it backs + * off for compaction without freeing anything. + * + * The watermark check below stays at the request order: it asks + * whether the request itself could succeed after reclaim. */ - if (did_some_progress && order <= PAGE_ALLOC_COSTLY_ORDER) + if (did_some_progress && order <= PAGE_ALLOC_COSTLY_ORDER && + nofrag_promote_order(order, alloc_flags, ac) == order) *no_progress_loops = 0; else (*no_progress_loops)++; @@ -5024,9 +5045,15 @@ retry: &compaction_retries)) goto retry; - /* Reclaim/compaction failed to prevent the fallback */ + /* + * Reclaim/compaction failed to prevent the fallback. The retry + * budget was spent on making blocks, not on the request itself; + * give the fallback a fresh one before considering OOM. + */ if (defrag_mode && (alloc_flags & ALLOC_NOFRAGMENT)) { alloc_flags &= ~ALLOC_NOFRAGMENT; + no_progress_loops = 0; + compaction_retries = 0; goto retry; } _ Patches currently in -mm which might be from kas@kernel.org are mm-page_alloc-make-defrag_mode-retries-follow-the-promoted-order.patch mm-huge_memory-do-not-touch-frozen-folios-in-deferred_split_isolate.patch mm-huge_memory-dequeue-the-deferred-split-after-the-split-freeze.patch mm-huge_memory-add-folio_reset_partially_mapped.patch mm-khugepaged-drop-redundant-mm_struct-pin-in-madvise_collapse.patch mm-khugepaged-count-collapses-where-khugepaged-makes-them.patch mm-khugepaged-rename-mthp_present_ptes-bitmap-to-eligible_ptes.patch mm-collapse-add-collapseh-for-the-collapse-interface.patch mm-collapse-state-what-a-collapse-may-do-in-the-policy.patch mm-collapse-drop-the-collapse_possible-wrapper.patch mm-collapse-name-the-per-table-scan-reset-for-what-it-resets.patch mm-collapse-call-collapse_file-from-collapse_single_pmd.patch mm-collapse-separate-scanning-a-pte-table-from-collapsing-it.patch mm-collapse-open-code-collapse_single_pmd-in-its-two-callers.patch mm-collapse-work-out-the-orders-a-vma-allows-once-per-vma.patch mm-collapse-declare-the-collapse-interface-in-collapseh.patch mm-collapse-implement-madv_collapse-in-madvisec.patch selftests-mm-raise-the-khugepaged-test-case-cap.patch selftests-mm-skip-collapse_compound_extreme-where-the-pmd-is-too-large.patch selftests-mm-scale-khugepageds-collapse-wait-with-the-pmd-size.patch selftests-mm-skip-khugepaged-page-cache-cases-without-a-pmd-folio.patch selftests-mm-make-the-swap-cases-swapout-reliable.patch selftests-mm-stop-khugepaged-during-the-madv_collapse-cases.patch selftests-mm-move-is_backed_by_folio-into-vm_util.patch selftests-mm-add-folio-order-check-for-address-ranges.patch selftests-mm-add-folio-order-detection-self-check.patch selftests-mm-add-khugepaged-completion-barrier-helper.patch selftests-mm-add-order-parameterized-khugepaged-collapse-cases.patch selftests-mm-parameterize-the-mixed-source-collapse-case-by-source-order.patch selftests-mm-cover-a-shared-source-collapse-write-race.patch selftests-mm-run-every-supported-collapse-order-by-default.patch selftests-mm-check-that-one-khugepaged-pass-collapses-one-window.patch selftests-mm-add-khugepaged-race-harness.patch selftests-mm-race-the-collapse-of-windows-with-holes.patch selftests-mm-add-memory-pressure-threads-to-the-khugepaged-race-harness.patch selftests-mm-zap-whole-pte-tables-in-the-khugepaged-race-harness.patch