From nobody Fri Dec 19 04:52:59 2025 Received: from wout3-smtp.messagingengine.com (wout3-smtp.messagingengine.com [64.147.123.19]) (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 D16E3127B6C for ; Tue, 20 Feb 2024 18:32:41 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=64.147.123.19 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1708453963; cv=none; b=JwGFq4/r7y9VBjT4JktNJj0vLqiJI8kP5M1kKLtfVl8INaJGzE8KQNM/mvNRSWdxaQwLOoI/Xrk2F7ZGgoC8LocYRz/lkKrJOpKzh/xqg7EEH4SeyfF2sp8CRWqO99wAb0vIXp0Kck08SwXcFNYtXrJMq/Nd/zhwWm4LOGHy/WI= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1708453963; c=relaxed/simple; bh=jysQqj4P6OHjX2f8XiwuDNsMZWqBz8wpGxcTRhCdhAw=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version:Content-Type; b=cj89DY+MBOajkwQjWjJ/OrQQeSMhg6ebBsoriOJmrne80Nqvll+/8h28kspZdH/tGXJvVpRpmOyWJnF1p+TKXqLHnKg5p0/iKuLWUceWXID5mBRigv6juibGDl1mv6f8lFT8OMt5efUKkXbPXzYKKdp2SbVMvWoy1HuuLGBkMd4= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=sent.com; spf=pass smtp.mailfrom=sent.com; dkim=pass (2048-bit key) header.d=sent.com header.i=@sent.com header.b=OSmmEdRo; dkim=pass (2048-bit key) header.d=messagingengine.com header.i=@messagingengine.com header.b=gmSswlMA; arc=none smtp.client-ip=64.147.123.19 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=sent.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=sent.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=sent.com header.i=@sent.com header.b="OSmmEdRo"; dkim=pass (2048-bit key) header.d=messagingengine.com header.i=@messagingengine.com header.b="gmSswlMA" Received: from compute7.internal (compute7.nyi.internal [10.202.2.48]) by mailout.west.internal (Postfix) with ESMTP id 4F7983200495; Tue, 20 Feb 2024 13:32:39 -0500 (EST) Received: from mailfrontend2 ([10.202.2.163]) by compute7.internal (MEProxy); Tue, 20 Feb 2024 13:32:41 -0500 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=sent.com; h=cc :cc:content-transfer-encoding:content-type:content-type:date :date:from:from:in-reply-to:in-reply-to:message-id:mime-version :references:reply-to:reply-to:subject:subject:to:to; s=fm1; t= 1708453958; x=1708540358; bh=3te7tA2WvcLuVx5XlSmApKI15+NRwijvQju oj0usTfY=; b=OSmmEdRoDCwAgPBnLo0uZuQmHYzka++qvdQaXCk44ZDkenqRSb/ K0hZmACbnPKTnIsRqCesEe/C/TxyJW1JtH3gDiOYSbyF/lPmgVgJHa74i9MPm7Jh +xvtuWp0gV4+1HEK/3fO/FEoeQIJE7pkTCxHyF7EjWGEzWwX2+e43B7Z5mqLpcxV CAtDT76PMAUS67DsKKkWC8s/WhKVzmpArFJn5fQwR/PbJICXsLUVT1mK7haRlUCz q20JRQS7zbZ2s/C4ugZifll3pLFFfeH/uY8Xy/zbFahW1D/qOhX4BAPxHlVBbf8x UiEcd+fQdY2ss9rZHvh9zePsQIzaKA7AxVA== DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d= messagingengine.com; h=cc:cc:content-transfer-encoding :content-type:content-type:date:date:feedback-id:feedback-id :from:from:in-reply-to:in-reply-to:message-id:mime-version :references:reply-to:reply-to:subject:subject:to:to:x-me-proxy :x-me-proxy:x-me-sender:x-me-sender:x-sasl-enc; s=fm1; t= 1708453958; x=1708540358; bh=3te7tA2WvcLuVx5XlSmApKI15+NRwijvQju oj0usTfY=; b=gmSswlMAp/+WPxW+MYBqcg9EJgcY/p1MxHMZAoruS9GrwGchYPy 3Ous9gaijN0aVRwVNWXOKzWiekjpb3GPsNmM27lXwaOzEATaSZRln67Z2gwdpcwH P8tDUsZ96dxTJg1enKijxxxmJ7EY3Ulh9wCBwheJRscZJuQ4kP7yQCOhmIZ3qo1R 2+zvY0cUT0NAAgl3tWqEjFvleWyVa6PXV6s85zwbJsbCzKG+yuGZEa9yJxxc/kaK LKJTYsEgjYX/aYYH27H9VUZDnpg8UqgWttfzdBVGAFgx2ECmXKM/N2iv4mKc2hQN STYviACeyGO3U9Pdx7uwp/A5LSjRlCNGLLQ== X-ME-Sender: X-ME-Received: X-ME-Proxy-Cause: gggruggvucftvghtrhhoucdtuddrgedvledrfedtgdduudehucetufdoteggodetrfdotf fvucfrrhhofhhilhgvmecuhfgrshhtofgrihhlpdfqfgfvpdfurfetoffkrfgpnffqhgen uceurghilhhouhhtmecufedttdenucesvcftvggtihhpihgvnhhtshculddquddttddmne cujfgurhephffvvefufffkofgjfhhrgggtgfesthekredtredtjeenucfhrhhomhepkghi ucgjrghnuceoiihirdihrghnsehsvghnthdrtghomheqnecuggftrfgrthhtvghrnhepje ekteekffelleekudfftdefvddtjeejuedtuedtteegjefgvedtfedujeekieevnecuvehl uhhsthgvrhfuihiivgeptdenucfrrghrrghmpehmrghilhhfrhhomhepiihirdihrghnse hsvghnthdrtghomh X-ME-Proxy: Feedback-ID: iccd040f4:Fastmail Received: by mail.messagingengine.com (Postfix) with ESMTPA; Tue, 20 Feb 2024 13:32:37 -0500 (EST) From: Zi Yan To: linux-mm@kvack.org, linux-kernel@vger.kernel.org Cc: Zi Yan , "Huang, Ying" , Ryan Roberts , Andrew Morton , "Matthew Wilcox (Oracle)" , David Hildenbrand , "Yin, Fengwei" , Yu Zhao , Vlastimil Babka , "Kirill A . Shutemov" , Johannes Weiner , Baolin Wang , Kemeng Shi , Mel Gorman , Rohan Puri , Mcgrof Chamberlain , Adam Manzanares , "Vishal Moola (Oracle)" Subject: [PATCH v7 1/4] mm/page_alloc: remove unused fpi_flags in free_pages_prepare() Date: Tue, 20 Feb 2024 13:32:17 -0500 Message-ID: <20240220183220.1451315-2-zi.yan@sent.com> X-Mailer: git-send-email 2.43.0 In-Reply-To: <20240220183220.1451315-1-zi.yan@sent.com> References: <20240220183220.1451315-1-zi.yan@sent.com> Reply-To: Zi Yan Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: quoted-printable From: Zi Yan Commit 0a54864f8dfb ("kasan: remove PG_skip_kasan_poison flag") removes the use of fpi_flags in should_skip_kasan_poison() and fpi_flags is only passed to should_skip_kasan_poison() in free_pages_prepare(). Remove the unused parameter. Signed-off-by: Zi Yan Reviewed-by: Vlastimil Babka Reviewed-by: David Hildenbrand --- mm/page_alloc.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 7ae4b74c9e5c..70c1ed3addf3 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -1061,7 +1061,7 @@ static int free_tail_page_prepare(struct page *head_p= age, struct page *page) * on-demand allocation and then freed again before the deferred pages * initialization is done, but this is not likely to happen. */ -static inline bool should_skip_kasan_poison(struct page *page, fpi_t fpi_f= lags) +static inline bool should_skip_kasan_poison(struct page *page) { if (IS_ENABLED(CONFIG_KASAN_GENERIC)) return deferred_pages_enabled(); @@ -1081,10 +1081,10 @@ static void kernel_init_pages(struct page *page, in= t numpages) } =20 static __always_inline bool free_pages_prepare(struct page *page, - unsigned int order, fpi_t fpi_flags) + unsigned int order) { int bad =3D 0; - bool skip_kasan_poison =3D should_skip_kasan_poison(page, fpi_flags); + bool skip_kasan_poison =3D should_skip_kasan_poison(page); bool init =3D want_init_on_free(); bool compound =3D PageCompound(page); =20 @@ -1266,7 +1266,7 @@ static void __free_pages_ok(struct page *page, unsign= ed int order, unsigned long pfn =3D page_to_pfn(page); struct zone *zone =3D page_zone(page); =20 - if (!free_pages_prepare(page, order, fpi_flags)) + if (!free_pages_prepare(page, order)) return; =20 /* @@ -2379,7 +2379,7 @@ static bool free_unref_page_prepare(struct page *page= , unsigned long pfn, { int migratetype; =20 - if (!free_pages_prepare(page, order, FPI_NONE)) + if (!free_pages_prepare(page, order)) return false; =20 migratetype =3D get_pfnblock_migratetype(page, pfn); --=20 2.43.0 From nobody Fri Dec 19 04:52:59 2025 Received: from wout3-smtp.messagingengine.com (wout3-smtp.messagingengine.com [64.147.123.19]) (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 1A54D1292FD for ; Tue, 20 Feb 2024 18:32:44 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=64.147.123.19 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1708453966; cv=none; b=lCq4m7TrcSOIpPbmJv2tvSw8PDb3hyV4QpHMjDBQmt0Hs5YxM9tsxDCeBjRm+IJfQffna4Gq0p06THPrj+FSa0FbSfyah4MnyoRhjpMDmYJh0XX39A0jUfWAMPWrH2Omu4ZOMbMJ5klnBVCg3aTVC1maWTAXgeuTjWRpwJqpHlc= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1708453966; c=relaxed/simple; bh=HhOvTaFPif92cm6vV+oSKDGsnXOY2r1rAN3bwiOwq/I=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version:Content-Type; b=rpRIW/wVumZT9te2cIEFKLsAQZb50qfklcaAIv/mVXy8LcPvUhrGJHRkQWWkXykEbUadY8DnOw9YnT/M9HLGIjKadNfDW+C4rmFjSmcdN+O7Qd+kk0Uqy7g5ijV5OaOREJ6AcXmPgx7j+p76APGTfjsO5PJkTkGJvuNj38Ezu7E= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=sent.com; spf=pass smtp.mailfrom=sent.com; dkim=pass (2048-bit key) header.d=sent.com header.i=@sent.com header.b=IBCxnz5k; dkim=pass (2048-bit key) header.d=messagingengine.com header.i=@messagingengine.com header.b=jiSueTP2; arc=none smtp.client-ip=64.147.123.19 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=sent.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=sent.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=sent.com header.i=@sent.com header.b="IBCxnz5k"; dkim=pass (2048-bit key) header.d=messagingengine.com header.i=@messagingengine.com header.b="jiSueTP2" Received: from compute2.internal (compute2.nyi.internal [10.202.2.46]) by mailout.west.internal (Postfix) with ESMTP id 8B5B03200A4D; Tue, 20 Feb 2024 13:32:42 -0500 (EST) Received: from mailfrontend2 ([10.202.2.163]) by compute2.internal (MEProxy); Tue, 20 Feb 2024 13:32:44 -0500 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=sent.com; h=cc :cc:content-transfer-encoding:content-type:content-type:date :date:from:from:in-reply-to:in-reply-to:message-id:mime-version :references:reply-to:reply-to:subject:subject:to:to; s=fm1; t= 1708453962; x=1708540362; bh=VthfB9xHWXamr5LNX5aI7SP9sOuq2T7O1Uc OskPjsOI=; b=IBCxnz5kEzrcnVVh8byadrCYLGeY+gCPJu4KYwlIh6EYBa+DLmh DVgGL5okBSyRGQJM8oS2cHgWVRx3A/VIJsXjlm3/Q8xVAl81p8ZTP/MDsKI2ilKg j1Ao/xGia5AXXNoAg2GqbL8XA2QOpbmWkUYhgoERs1srQZHv3C2sJWikcSOgJwz/ +rJtvtpe0YpSvNLJwH2TfKlclX3dzkRoKZiE4kGoAG4IF8+gMNdMFu8x2Hi8ckn9 gW/FXHoVwtF++TWclK+cV7TKWHLboYQV2MVqYg/G0zDsRVdgwpivGy01VjhcrLvX Bh05vMbRhOlvB5aLeaXKrxXzQ4cKhU/uynQ== DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d= messagingengine.com; h=cc:cc:content-transfer-encoding :content-type:content-type:date:date:feedback-id:feedback-id :from:from:in-reply-to:in-reply-to:message-id:mime-version :references:reply-to:reply-to:subject:subject:to:to:x-me-proxy :x-me-proxy:x-me-sender:x-me-sender:x-sasl-enc; s=fm1; t= 1708453962; x=1708540362; bh=VthfB9xHWXamr5LNX5aI7SP9sOuq2T7O1Uc OskPjsOI=; b=jiSueTP2NFAW16xmDieh0jQv1EMcSklBlVsdVriYGBp7o94Ggcq fl0/Y10zf0QltMFk5nK18Kf7eVAmp5HMBYWlwcMMrbzm2fCsUiP4W6EWAgsjePP1 pugKuaGP0Dg72DLTzagqAwEVofFHzqE8dUqb7Ia413zAYAzGmQ/RmNmNMqu80iK8 dpyfuQmLBOu24LqQBr77Xce4SnEjXhA2bFEF/1r0RToVXdwMDXckGJibLq+U8kYX /vNtbBD9Is8Z+R/sn+DyixhqNPCqW9vOfAJ/u/DEMTKXT0EFLnPVqCUjBp/RfVJ4 2jWosqtyzTbBoRRq+5SVmMelvVS5BbOYhzg== X-ME-Sender: X-ME-Received: X-ME-Proxy-Cause: gggruggvucftvghtrhhoucdtuddrgedvledrfedtgdduudehucetufdoteggodetrfdotf fvucfrrhhofhhilhgvmecuhfgrshhtofgrihhlpdfqfgfvpdfurfetoffkrfgpnffqhgen uceurghilhhouhhtmecufedttdenucesvcftvggtihhpihgvnhhtshculddquddttddmne cujfgurhephffvvefufffkofgjfhhrgggtgfesthekredtredtjeenucfhrhhomhepkghi ucgjrghnuceoiihirdihrghnsehsvghnthdrtghomheqnecuggftrfgrthhtvghrnhepje ekteekffelleekudfftdefvddtjeejuedtuedtteegjefgvedtfedujeekieevnecuvehl uhhsthgvrhfuihiivgeptdenucfrrghrrghmpehmrghilhhfrhhomhepiihirdihrghnse hsvghnthdrtghomh X-ME-Proxy: Feedback-ID: iccd040f4:Fastmail Received: by mail.messagingengine.com (Postfix) with ESMTPA; Tue, 20 Feb 2024 13:32:41 -0500 (EST) From: Zi Yan To: linux-mm@kvack.org, linux-kernel@vger.kernel.org Cc: Zi Yan , "Huang, Ying" , Ryan Roberts , Andrew Morton , "Matthew Wilcox (Oracle)" , David Hildenbrand , "Yin, Fengwei" , Yu Zhao , Vlastimil Babka , "Kirill A . Shutemov" , Johannes Weiner , Baolin Wang , Kemeng Shi , Mel Gorman , Rohan Puri , Mcgrof Chamberlain , Adam Manzanares , "Vishal Moola (Oracle)" Subject: [PATCH v7 2/4] mm/compaction: enable compacting >0 order folios. Date: Tue, 20 Feb 2024 13:32:18 -0500 Message-ID: <20240220183220.1451315-3-zi.yan@sent.com> X-Mailer: git-send-email 2.43.0 In-Reply-To: <20240220183220.1451315-1-zi.yan@sent.com> References: <20240220183220.1451315-1-zi.yan@sent.com> Reply-To: Zi Yan Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: quoted-printable From: Zi Yan migrate_pages() supports >0 order folio migration and during compaction, even if compaction_alloc() cannot provide >0 order free pages, migrate_pages() can split the source page and try to migrate the base pages from the split. It can be a baseline and start point for adding support for compacting >0 order folios. Signed-off-by: Zi Yan Suggested-by: Huang Ying Reviewed-by: Baolin Wang Reviewed-by: Vlastimil Babka Tested-by: Baolin Wang Tested-by: Yu Zhao Cc: Adam Manzanares Cc: David Hildenbrand Cc: Johannes Weiner Cc: Kemeng Shi Cc: Kirill A. Shutemov Cc: Luis Chamberlain Cc: Matthew Wilcox (Oracle) Cc: Mel Gorman Cc: Ryan Roberts Cc: Vishal Moola (Oracle) Cc: Vlastimil Babka Cc: Yin Fengwei --- mm/compaction.c | 101 ++++++++++++++++++++++++++++++++++++------------ 1 file changed, 76 insertions(+), 25 deletions(-) diff --git a/mm/compaction.c b/mm/compaction.c index ccd447282665..6509fab27be9 100644 --- a/mm/compaction.c +++ b/mm/compaction.c @@ -40,9 +40,22 @@ static inline void count_compact_events(enum vm_event_it= em item, long delta) { count_vm_events(item, delta); } + +/* + * order =3D=3D -1 is expected when compacting proactively via + * 1. /proc/sys/vm/compact_memory + * 2. /sys/devices/system/node/nodex/compact + * 3. /proc/sys/vm/compaction_proactiveness + */ +static inline bool is_via_compact_memory(int order) +{ + return order =3D=3D -1; +} + #else #define count_compact_event(item) do { } while (0) #define count_compact_events(item, delta) do { } while (0) +static inline bool is_via_compact_memory(int order) { return false; } #endif =20 #if defined CONFIG_COMPACTION || defined CONFIG_CMA @@ -816,6 +829,32 @@ static bool too_many_isolated(struct compact_control *= cc) return too_many; } =20 +/** + * skip_isolation_on_order() - determine when to skip folio isolation base= d on + * folio order and compaction target order + * @order: to-be-isolated folio order + * @target_order: compaction target order + * + * This avoids unnecessary folio isolations during compaction. + */ +static bool skip_isolation_on_order(int order, int target_order) +{ + /* + * Unless we are performing global compaction (i.e., + * is_via_compact_memory), skip any folios that are larger than the + * target order: we wouldn't be here if we'd have a free folio with + * the desired target_order, so migrating this folio would likely fail + * later. + */ + if (!is_via_compact_memory(target_order) && order >=3D target_order) + return true; + /* + * We limit memory compaction to pageblocks and won't try + * creating free blocks of memory that are larger than that. + */ + return order >=3D pageblock_order; +} + /** * isolate_migratepages_block() - isolate all migrate-able pages within * a single pageblock @@ -947,7 +986,22 @@ isolate_migratepages_block(struct compact_control *cc,= unsigned long low_pfn, valid_page =3D page; } =20 - if (PageHuge(page) && cc->alloc_contig) { + if (PageHuge(page)) { + /* + * skip hugetlbfs if we are not compacting for pages + * bigger than its order. THPs and other compound pages + * are handled below. + */ + if (!cc->alloc_contig) { + const unsigned int order =3D compound_order(page); + + if (order <=3D MAX_PAGE_ORDER) { + low_pfn +=3D (1UL << order) - 1; + nr_scanned +=3D (1UL << order) - 1; + } + goto isolate_fail; + } + /* for alloc_contig case */ if (locked) { unlock_page_lruvec_irqrestore(locked, flags); locked =3D NULL; @@ -1008,21 +1062,24 @@ isolate_migratepages_block(struct compact_control *= cc, unsigned long low_pfn, } =20 /* - * Regardless of being on LRU, compound pages such as THP and - * hugetlbfs are not to be compacted unless we are attempting - * an allocation much larger than the huge page size (eg CMA). - * We can potentially save a lot of iterations if we skip them - * at once. The check is racy, but we can consider only valid - * values and the only danger is skipping too much. + * Regardless of being on LRU, compound pages such as THP + * (hugetlbfs is handled above) are not to be compacted unless + * we are attempting an allocation larger than the compound + * page size. We can potentially save a lot of iterations if we + * skip them at once. The check is racy, but we can consider + * only valid values and the only danger is skipping too much. */ if (PageCompound(page) && !cc->alloc_contig) { const unsigned int order =3D compound_order(page); =20 - if (likely(order <=3D MAX_PAGE_ORDER)) { - low_pfn +=3D (1UL << order) - 1; - nr_scanned +=3D (1UL << order) - 1; + /* Skip based on page order and compaction target order. */ + if (skip_isolation_on_order(order, cc->order)) { + if (order <=3D MAX_PAGE_ORDER) { + low_pfn +=3D (1UL << order) - 1; + nr_scanned +=3D (1UL << order) - 1; + } + goto isolate_fail; } - goto isolate_fail; } =20 /* @@ -1165,10 +1222,11 @@ isolate_migratepages_block(struct compact_control *= cc, unsigned long low_pfn, } =20 /* - * folio become large since the non-locked check, - * and it's on LRU. + * Check LRU folio order under the lock */ - if (unlikely(folio_test_large(folio) && !cc->alloc_contig)) { + if (unlikely(skip_isolation_on_order(folio_order(folio), + cc->order) && + !cc->alloc_contig)) { low_pfn +=3D folio_nr_pages(folio) - 1; nr_scanned +=3D folio_nr_pages(folio) - 1; folio_set_lru(folio); @@ -1788,6 +1846,10 @@ static struct folio *compaction_alloc(struct folio *= src, unsigned long data) struct compact_control *cc =3D (struct compact_control *)data; struct folio *dst; =20 + /* this makes migrate_pages() split the source page and retry */ + if (folio_test_large(src)) + return NULL; + if (list_empty(&cc->freepages)) { isolate_freepages(cc); =20 @@ -2090,17 +2152,6 @@ static isolate_migrate_t isolate_migratepages(struct= compact_control *cc) return cc->nr_migratepages ? ISOLATE_SUCCESS : ISOLATE_NONE; } =20 -/* - * order =3D=3D -1 is expected when compacting proactively via - * 1. /proc/sys/vm/compact_memory - * 2. /sys/devices/system/node/nodex/compact - * 3. /proc/sys/vm/compaction_proactiveness - */ -static inline bool is_via_compact_memory(int order) -{ - return order =3D=3D -1; -} - /* * Determine whether kswapd is (or recently was!) running on this node. * --=20 2.43.0 From nobody Fri Dec 19 04:52:59 2025 Received: from wout3-smtp.messagingengine.com (wout3-smtp.messagingengine.com [64.147.123.19]) (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 3462012FF78 for ; Tue, 20 Feb 2024 18:32:48 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=64.147.123.19 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1708453970; cv=none; b=IBasMUAqpCTBLNgrksGObum2dpPfyDuw1sQ5/51GeUt/CC5awM0dactvDAO05+75d6hlQv7ZdkLvvloVLr/KvtM9WWPKfmj2chD06IsW1DAS+gYKu20pAVEjLvyt35OsD9pqLy9ILbIDbeymz9bVy7cYh2KeOgPRku7sOqPkm2Y= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1708453970; c=relaxed/simple; bh=5oHdqzGZOzpdqVXwMvOTRo5TusqeNuCtG62sP5UYHzw=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version:Content-Type; b=Ljzuy0UIgNm9FgqgSiEkfBYT44o9a/48ZG2vemVvh+sZTtwDKw/Wvl1juOSwLSySzIyRtp8+P99yBg/I91JQ/QlFa5qJMs9aHVFuCaKMM58pbQCvruiQXVB4/ZInz9bziUEtsp4Xs3ekryg3BWVJx1dRmgZWR7tIDIiYZ5zzBw4= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=sent.com; spf=pass smtp.mailfrom=sent.com; dkim=pass (2048-bit key) header.d=sent.com header.i=@sent.com header.b=VRjtlA7G; dkim=pass (2048-bit key) header.d=messagingengine.com header.i=@messagingengine.com header.b=C2yAIoGa; arc=none smtp.client-ip=64.147.123.19 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=sent.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=sent.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=sent.com header.i=@sent.com header.b="VRjtlA7G"; dkim=pass (2048-bit key) header.d=messagingengine.com header.i=@messagingengine.com header.b="C2yAIoGa" Received: from compute6.internal (compute6.nyi.internal [10.202.2.47]) by mailout.west.internal (Postfix) with ESMTP id C6B773200A20; Tue, 20 Feb 2024 13:32:45 -0500 (EST) Received: from mailfrontend2 ([10.202.2.163]) by compute6.internal (MEProxy); Tue, 20 Feb 2024 13:32:47 -0500 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=sent.com; h=cc :cc:content-transfer-encoding:content-type:content-type:date :date:from:from:in-reply-to:in-reply-to:message-id:mime-version :references:reply-to:reply-to:subject:subject:to:to; s=fm1; t= 1708453965; x=1708540365; bh=JtFLx5yaOzF263m9p88TzPIu/+lPjxUEnJ1 pW1s4Puw=; b=VRjtlA7G82KrDKIIKo1Y7Zm97nEw7/o4k6umQLb1da6cGh3M+DT dDklWKU4Qnp80XU0OSfJi1o0G3qh1YtEelMeKfa4B33OTTwJtSHiYM8DVnlDmn8g Z5BP+vNbdbToMdlPDRnNnn3la5ZE8QWsK+Yf2vrpejqOTnQGctc/Whg7sOoP78ER b56jS6EaXfeolzC9F70DvpOVLNcxAkjOCZG0h1XQ9aalybgCYVdA9dESNQ0dbKkP KQc/zTaG/s1s0pidf0IfHrnMSZpXbAfNZx1TOyw//4rdjL0zU56XJoLmAOgmbcUN G05WBCWSJSgjWzQJCplCGSKAOU3fPe8vhag== DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d= messagingengine.com; h=cc:cc:content-transfer-encoding :content-type:content-type:date:date:feedback-id:feedback-id :from:from:in-reply-to:in-reply-to:message-id:mime-version :references:reply-to:reply-to:subject:subject:to:to:x-me-proxy :x-me-proxy:x-me-sender:x-me-sender:x-sasl-enc; s=fm1; t= 1708453965; x=1708540365; bh=JtFLx5yaOzF263m9p88TzPIu/+lPjxUEnJ1 pW1s4Puw=; b=C2yAIoGaOWw1Un3N2oDdKfx/corJNnTKh2TK3UwOjIelhk7AwYS ivuzr2s0kc9BMZQxzvvBuOf4evdDB97D1V9Mj9d7YfxGKausSen2CKgwe4pjbCcu TX+jzENAUiQ16cmb+/JIbWJM5jHQXtx82EJpqQju1t6EqnfzvK/NlTtZV41ly+/6 U+paWmsReq8kIm/4wA+4jEqAARCaH0UdqpvpgtgFG/dbGVNmsLwhJNxS6m4ZX8t8 BN1c5q9E51SH9cFkOq6JqPkDUKTeEBVMFBiTy/38Gdd61FoqFlF4qOchBtLcDd0U 7qja5yTTCOorXbwGs+JZ/sgvbky845lsD+g== X-ME-Sender: X-ME-Received: X-ME-Proxy-Cause: gggruggvucftvghtrhhoucdtuddrgedvledrfedtgdduudehucetufdoteggodetrfdotf fvucfrrhhofhhilhgvmecuhfgrshhtofgrihhlpdfqfgfvpdfurfetoffkrfgpnffqhgen uceurghilhhouhhtmecufedttdenucesvcftvggtihhpihgvnhhtshculddquddttddmne cujfgurhephffvvefufffkofgjfhhrgggtgfesthekredtredtjeenucfhrhhomhepkghi ucgjrghnuceoiihirdihrghnsehsvghnthdrtghomheqnecuggftrfgrthhtvghrnhepje ekteekffelleekudfftdefvddtjeejuedtuedtteegjefgvedtfedujeekieevnecuvehl uhhsthgvrhfuihiivgeptdenucfrrghrrghmpehmrghilhhfrhhomhepiihirdihrghnse hsvghnthdrtghomh X-ME-Proxy: Feedback-ID: iccd040f4:Fastmail Received: by mail.messagingengine.com (Postfix) with ESMTPA; Tue, 20 Feb 2024 13:32:44 -0500 (EST) From: Zi Yan To: linux-mm@kvack.org, linux-kernel@vger.kernel.org Cc: Zi Yan , "Huang, Ying" , Ryan Roberts , Andrew Morton , "Matthew Wilcox (Oracle)" , David Hildenbrand , "Yin, Fengwei" , Yu Zhao , Vlastimil Babka , "Kirill A . Shutemov" , Johannes Weiner , Baolin Wang , Kemeng Shi , Mel Gorman , Rohan Puri , Mcgrof Chamberlain , Adam Manzanares , "Vishal Moola (Oracle)" Subject: [PATCH v7 3/4] mm/compaction: add support for >0 order folio memory compaction. Date: Tue, 20 Feb 2024 13:32:19 -0500 Message-ID: <20240220183220.1451315-4-zi.yan@sent.com> X-Mailer: git-send-email 2.43.0 In-Reply-To: <20240220183220.1451315-1-zi.yan@sent.com> References: <20240220183220.1451315-1-zi.yan@sent.com> Reply-To: Zi Yan Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: quoted-printable From: Zi Yan Before last commit, memory compaction only migrates order-0 folios and skips >0 order folios. Last commit splits all >0 order folios during compaction. This commit migrates >0 order folios during compaction by keeping isolated free pages at their original size without splitting them into order-0 pages and using them directly during migration process. What is different from the prior implementation: 1. All isolated free pages are kept in a NR_PAGE_ORDERS array of page lists, where each page list stores free pages in the same order. 2. All free pages are not post_alloc_hook() processed nor buddy pages, although their orders are stored in first page's private like buddy pages. 3. During migration, in new page allocation time (i.e., in compaction_alloc()), free pages are then processed by post_alloc_hook(). When migration fails and a new page is returned (i.e., in compaction_free()), free pages are restored by reversing the post_alloc_hook() operations using newly added free_pages_prepare_fpi_none(). Step 3 is done for a latter optimization that splitting and/or merging free pages during compaction becomes easier. Note: without splitting free pages, compaction can end prematurely due to migration will return -ENOMEM even if there is free pages. This happens when no order-0 free page exist and compaction_alloc() return NULL. Signed-off-by: Zi Yan Reviewed-by: Baolin Wang Reviewed-by: Vlastimil Babka Tested-by: Baolin Wang Tested-by: Yu Zhao Cc: Adam Manzanares Cc: David Hildenbrand Cc: Huang Ying Cc: Johannes Weiner Cc: Kemeng Shi Cc: Kirill A. Shutemov Cc: Luis Chamberlain Cc: Matthew Wilcox (Oracle) Cc: Mel Gorman Cc: Ryan Roberts Cc: Vishal Moola (Oracle) Cc: Vlastimil Babka Cc: Yin Fengwei --- mm/compaction.c | 140 +++++++++++++++++++++++++++--------------------- mm/internal.h | 4 +- mm/page_alloc.c | 2 +- 3 files changed, 83 insertions(+), 63 deletions(-) diff --git a/mm/compaction.c b/mm/compaction.c index 6509fab27be9..112711752321 100644 --- a/mm/compaction.c +++ b/mm/compaction.c @@ -79,45 +79,56 @@ static inline bool is_via_compact_memory(int order) { r= eturn false; } #define COMPACTION_HPAGE_ORDER (PMD_SHIFT - PAGE_SHIFT) #endif =20 -static unsigned long release_freepages(struct list_head *freelist) +static void split_map_pages(struct list_head *freepages) { + unsigned int i, order; struct page *page, *next; - unsigned long high_pfn =3D 0; + LIST_HEAD(tmp_list); =20 - list_for_each_entry_safe(page, next, freelist, lru) { - unsigned long pfn =3D page_to_pfn(page); - list_del(&page->lru); - __free_page(page); - if (pfn > high_pfn) - high_pfn =3D pfn; - } + for (order =3D 0; order < NR_PAGE_ORDERS; order++) { + list_for_each_entry_safe(page, next, &freepages[order], lru) { + unsigned int nr_pages; =20 - return high_pfn; + list_del(&page->lru); + + nr_pages =3D 1 << order; + + post_alloc_hook(page, order, __GFP_MOVABLE); + if (order) + split_page(page, order); + + for (i =3D 0; i < nr_pages; i++) { + list_add(&page->lru, &tmp_list); + page++; + } + } + list_splice_init(&tmp_list, &freepages[0]); + } } =20 -static void split_map_pages(struct list_head *list) +static unsigned long release_free_list(struct list_head *freepages) { - unsigned int i, order, nr_pages; - struct page *page, *next; - LIST_HEAD(tmp_list); - - list_for_each_entry_safe(page, next, list, lru) { - list_del(&page->lru); + int order; + unsigned long high_pfn =3D 0; =20 - order =3D page_private(page); - nr_pages =3D 1 << order; + for (order =3D 0; order < NR_PAGE_ORDERS; order++) { + struct page *page, *next; =20 - post_alloc_hook(page, order, __GFP_MOVABLE); - if (order) - split_page(page, order); + list_for_each_entry_safe(page, next, &freepages[order], lru) { + unsigned long pfn =3D page_to_pfn(page); =20 - for (i =3D 0; i < nr_pages; i++) { - list_add(&page->lru, &tmp_list); - page++; + list_del(&page->lru); + /* + * Convert free pages into post allocation pages, so + * that we can free them via __free_page. + */ + post_alloc_hook(page, order, __GFP_MOVABLE); + __free_pages(page, order); + if (pfn > high_pfn) + high_pfn =3D pfn; } } - - list_splice(&tmp_list, list); + return high_pfn; } =20 #ifdef CONFIG_COMPACTION @@ -670,7 +681,7 @@ static unsigned long isolate_freepages_block(struct com= pact_control *cc, nr_scanned +=3D isolated - 1; total_isolated +=3D isolated; cc->nr_freepages +=3D isolated; - list_add_tail(&page->lru, freelist); + list_add_tail(&page->lru, &freelist[order]); =20 if (!strict && cc->nr_migratepages <=3D cc->nr_freepages) { blockpfn +=3D isolated; @@ -735,7 +746,11 @@ isolate_freepages_range(struct compact_control *cc, unsigned long start_pfn, unsigned long end_pfn) { unsigned long isolated, pfn, block_start_pfn, block_end_pfn; - LIST_HEAD(freelist); + int order; + struct list_head tmp_freepages[NR_PAGE_ORDERS]; + + for (order =3D 0; order < NR_PAGE_ORDERS; order++) + INIT_LIST_HEAD(&tmp_freepages[order]); =20 pfn =3D start_pfn; block_start_pfn =3D pageblock_start_pfn(pfn); @@ -766,7 +781,7 @@ isolate_freepages_range(struct compact_control *cc, break; =20 isolated =3D isolate_freepages_block(cc, &isolate_start_pfn, - block_end_pfn, &freelist, 0, true); + block_end_pfn, tmp_freepages, 0, true); =20 /* * In strict mode, isolate_freepages_block() returns 0 if @@ -783,15 +798,15 @@ isolate_freepages_range(struct compact_control *cc, */ } =20 - /* __isolate_free_page() does not map the pages */ - split_map_pages(&freelist); - if (pfn < end_pfn) { /* Loop terminated early, cleanup. */ - release_freepages(&freelist); + release_free_list(tmp_freepages); return 0; } =20 + /* __isolate_free_page() does not map the pages */ + split_map_pages(tmp_freepages); + /* We don't use freelists for anything. */ return pfn; } @@ -1518,7 +1533,7 @@ fast_isolate_around(struct compact_control *cc, unsig= ned long pfn) if (!page) return; =20 - isolate_freepages_block(cc, &start_pfn, end_pfn, &cc->freepages, 1, false= ); + isolate_freepages_block(cc, &start_pfn, end_pfn, cc->freepages, 1, false); =20 /* Skip this pageblock in the future as it's full or nearly full */ if (start_pfn =3D=3D end_pfn && !cc->no_set_skip_hint) @@ -1647,7 +1662,7 @@ static void fast_isolate_freepages(struct compact_con= trol *cc) nr_scanned +=3D nr_isolated - 1; total_isolated +=3D nr_isolated; cc->nr_freepages +=3D nr_isolated; - list_add_tail(&page->lru, &cc->freepages); + list_add_tail(&page->lru, &cc->freepages[order]); count_compact_events(COMPACTISOLATED, nr_isolated); } else { /* If isolation fails, abort the search */ @@ -1724,13 +1739,12 @@ static void isolate_freepages(struct compact_contro= l *cc) unsigned long isolate_start_pfn; /* exact pfn we start at */ unsigned long block_end_pfn; /* end of current pageblock */ unsigned long low_pfn; /* lowest pfn scanner is able to scan */ - struct list_head *freelist =3D &cc->freepages; unsigned int stride; =20 /* Try a small search of the free lists for a candidate */ fast_isolate_freepages(cc); if (cc->nr_freepages) - goto splitmap; + return; =20 /* * Initialise the free scanner. The starting point is where we last @@ -1790,7 +1804,7 @@ static void isolate_freepages(struct compact_control = *cc) =20 /* Found a block suitable for isolating free pages from. */ nr_isolated =3D isolate_freepages_block(cc, &isolate_start_pfn, - block_end_pfn, freelist, stride, false); + block_end_pfn, cc->freepages, stride, false); =20 /* Update the skip hint if the full pageblock was scanned */ if (isolate_start_pfn =3D=3D block_end_pfn) @@ -1831,10 +1845,6 @@ static void isolate_freepages(struct compact_control= *cc) * and the loop terminated due to isolate_start_pfn < low_pfn */ cc->free_pfn =3D isolate_start_pfn; - -splitmap: - /* __isolate_free_page() does not map the pages */ - split_map_pages(freelist); } =20 /* @@ -1845,24 +1855,22 @@ static struct folio *compaction_alloc(struct folio = *src, unsigned long data) { struct compact_control *cc =3D (struct compact_control *)data; struct folio *dst; + int order =3D folio_order(src); =20 - /* this makes migrate_pages() split the source page and retry */ - if (folio_test_large(src)) - return NULL; - - if (list_empty(&cc->freepages)) { + if (list_empty(&cc->freepages[order])) { isolate_freepages(cc); - - if (list_empty(&cc->freepages)) + if (list_empty(&cc->freepages[order])) return NULL; } =20 - dst =3D list_entry(cc->freepages.next, struct folio, lru); + dst =3D list_first_entry(&cc->freepages[order], struct folio, lru); list_del(&dst->lru); - cc->nr_freepages--; - cc->nr_migratepages--; - - return dst; + post_alloc_hook(&dst->page, order, __GFP_MOVABLE); + if (order) + prep_compound_page(&dst->page, order); + cc->nr_freepages -=3D 1 << order; + cc->nr_migratepages -=3D 1 << order; + return page_rmappable_folio(&dst->page); } =20 /* @@ -1873,10 +1881,19 @@ static struct folio *compaction_alloc(struct folio = *src, unsigned long data) static void compaction_free(struct folio *dst, unsigned long data) { struct compact_control *cc =3D (struct compact_control *)data; + int order =3D folio_order(dst); + struct page *page =3D &dst->page; =20 - list_add(&dst->lru, &cc->freepages); - cc->nr_freepages++; - cc->nr_migratepages++; + if (folio_put_testzero(dst)) { + free_pages_prepare(page, order); + list_add(&dst->lru, &cc->freepages[order]); + cc->nr_freepages +=3D 1 << order; + } + cc->nr_migratepages +=3D 1 << order; + /* + * someone else has referenced the page, we cannot take it back to our + * free list. + */ } =20 /* possible outcome of isolate_migratepages */ @@ -2489,6 +2506,7 @@ compact_zone(struct compact_control *cc, struct captu= re_control *capc) const bool sync =3D cc->mode !=3D MIGRATE_ASYNC; bool update_cached; unsigned int nr_succeeded =3D 0; + int order; =20 /* * These counters track activities during zone compaction. Initialize @@ -2498,7 +2516,8 @@ compact_zone(struct compact_control *cc, struct captu= re_control *capc) cc->total_free_scanned =3D 0; cc->nr_migratepages =3D 0; cc->nr_freepages =3D 0; - INIT_LIST_HEAD(&cc->freepages); + for (order =3D 0; order < NR_PAGE_ORDERS; order++) + INIT_LIST_HEAD(&cc->freepages[order]); INIT_LIST_HEAD(&cc->migratepages); =20 cc->migratetype =3D gfp_migratetype(cc->gfp_mask); @@ -2684,7 +2703,7 @@ compact_zone(struct compact_control *cc, struct captu= re_control *capc) * so we don't leave any returned pages behind in the next attempt. */ if (cc->nr_freepages > 0) { - unsigned long free_pfn =3D release_freepages(&cc->freepages); + unsigned long free_pfn =3D release_free_list(cc->freepages); =20 cc->nr_freepages =3D 0; VM_BUG_ON(free_pfn =3D=3D 0); @@ -2703,7 +2722,6 @@ compact_zone(struct compact_control *cc, struct captu= re_control *capc) =20 trace_mm_compaction_end(cc, start_pfn, end_pfn, sync, ret); =20 - VM_BUG_ON(!list_empty(&cc->freepages)); VM_BUG_ON(!list_empty(&cc->migratepages)); =20 return ret; diff --git a/mm/internal.h b/mm/internal.h index 1e29c5821a1d..93e229112045 100644 --- a/mm/internal.h +++ b/mm/internal.h @@ -447,6 +447,8 @@ extern void prep_compound_page(struct page *page, unsig= ned int order); =20 extern void post_alloc_hook(struct page *page, unsigned int order, gfp_t gfp_flags); +extern bool free_pages_prepare(struct page *page, unsigned int order); + extern int user_min_free_kbytes; =20 extern void free_unref_page(struct page *page, unsigned int order); @@ -481,7 +483,7 @@ int split_free_page(struct page *free_page, * completes when free_pfn <=3D migrate_pfn */ struct compact_control { - struct list_head freepages; /* List of free pages to migrate to */ + struct list_head freepages[NR_PAGE_ORDERS]; /* List of free pages to migr= ate to */ struct list_head migratepages; /* List of pages being migrated */ unsigned int nr_freepages; /* Number of isolated free pages */ unsigned int nr_migratepages; /* Number of pages to migrate */ diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 70c1ed3addf3..b0b92ce997dc 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -1080,7 +1080,7 @@ static void kernel_init_pages(struct page *page, int = numpages) kasan_enable_current(); } =20 -static __always_inline bool free_pages_prepare(struct page *page, +__always_inline bool free_pages_prepare(struct page *page, unsigned int order) { int bad =3D 0; --=20 2.43.0 From nobody Fri Dec 19 04:52:59 2025 Received: from wout3-smtp.messagingengine.com (wout3-smtp.messagingengine.com [64.147.123.19]) (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 90C31130E32 for ; Tue, 20 Feb 2024 18:32:51 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=64.147.123.19 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1708453973; cv=none; b=mxkn0cIe1fmhBPmQwIkMKb4FMQVaK+ipXetUADXf5587Q3KZUnRtmaectxeieSSdU7JOpsyEepKbwB+BFUkxj32BpuN2HynqRNSoR4Gw5E993TogzfifHdiEFtKBUFwSrDN0YEXMIKX/noDoztJog3hcVKYkxDGK+mBQ7oB+aOg= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1708453973; c=relaxed/simple; bh=peQzf9WHqEOx4IpzLUnSKnIIBdZa/+nwsL6Us3kBuKA=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version:Content-Type; b=qomuw/L3jARpDKBYz8QaCjT2LDJyrape440gTp0G7ylGgiyL3W/+dvSAWYcsTXSaHWCWQQCdRYfefm3eyfwpTVQQdK1bAlYtZBIvzvagT/EDCcR8FGnXRhmsyrJcTqCs5Do1IBPtA8ziuwNW4XM6uwrSvFATRlCaxTAZZr6Id4k= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=sent.com; spf=pass smtp.mailfrom=sent.com; dkim=pass (2048-bit key) header.d=sent.com header.i=@sent.com header.b=C/BfFgk3; dkim=pass (2048-bit key) header.d=messagingengine.com header.i=@messagingengine.com header.b=Udaf3BFo; arc=none smtp.client-ip=64.147.123.19 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=sent.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=sent.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=sent.com header.i=@sent.com header.b="C/BfFgk3"; dkim=pass (2048-bit key) header.d=messagingengine.com header.i=@messagingengine.com header.b="Udaf3BFo" Received: from compute4.internal (compute4.nyi.internal [10.202.2.44]) by mailout.west.internal (Postfix) with ESMTP id 1454C3200A54; Tue, 20 Feb 2024 13:32:48 -0500 (EST) Received: from mailfrontend2 ([10.202.2.163]) by compute4.internal (MEProxy); Tue, 20 Feb 2024 13:32:50 -0500 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=sent.com; h=cc :cc:content-transfer-encoding:content-type:content-type:date :date:from:from:in-reply-to:in-reply-to:message-id:mime-version :references:reply-to:reply-to:subject:subject:to:to; s=fm1; t= 1708453968; x=1708540368; bh=2I9yVROKB+W15mI+UpYHE5WhFE4737+G/6r 8N16qHZk=; b=C/BfFgk39mofWJiQq1RU1DWERdAHjGqZ9CYe67lUaTnv4DkXexC JUgAvnAb0Geveu0vCR8Yc9SLyPMqyN/tNho4hPNJx4U/8p4TzIyaIc5OktmhEKCr br24aOTURhAPSOHXhberlecYH0+s+agkxnTKkHWM7EBjj9LkiJlmyvYIg9lYsb39 k1MC00poth8h21dF4zg8DknBjbZouWKEK9G0gXzDmEqPy/WYgcb5cbhSnBK3PxEU ezq4uAP4EnS/t7OdwAvx1gFmLzKX6+0I4E9AaHC+dQFzsxKKnCmseSrNxt7+pneX K8rOQO1BMHCdml1nFSal2TuN6nsalawNrtQ== DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d= messagingengine.com; h=cc:cc:content-transfer-encoding :content-type:content-type:date:date:feedback-id:feedback-id :from:from:in-reply-to:in-reply-to:message-id:mime-version :references:reply-to:reply-to:subject:subject:to:to:x-me-proxy :x-me-proxy:x-me-sender:x-me-sender:x-sasl-enc; s=fm1; t= 1708453968; x=1708540368; bh=2I9yVROKB+W15mI+UpYHE5WhFE4737+G/6r 8N16qHZk=; b=Udaf3BFoiTXOq1ww31/D3N6rdbCeObxT1gXG2hXa5NBnQUEJ+p1 FrTz7sVN7TfskcfLfC8lvIg7uWLwAM4gKoGLGq5PDq1H3TzHRCOWn5eABFHaUPWP hRcLPmnjhUdWr7YgxaQgBvgIUqBEFq8sUmNuaeI6CvWUaUlfr3eaIw45CIHpesDM Na1wRUkDZmVVdMvqma3brYOhrSvHyQgVMba8K8fiGYgkM1qfP2v3zaZpIaOApfNj XL6QJvDFvWitaPcbo8lJ/oJtMrlxWiFEHifseopH2PWBynBXM9l0160t+59lzuIK WRXy6u9dxKaJeE4fNamBvXRDfWPNYrzk+uA== X-ME-Sender: X-ME-Received: X-ME-Proxy-Cause: gggruggvucftvghtrhhoucdtuddrgedvledrfedtgdduudehucetufdoteggodetrfdotf fvucfrrhhofhhilhgvmecuhfgrshhtofgrihhlpdfqfgfvpdfurfetoffkrfgpnffqhgen uceurghilhhouhhtmecufedttdenucesvcftvggtihhpihgvnhhtshculddquddttddmne cujfgurhephffvvefufffkofgjfhhrgggtgfesthekredtredtjeenucfhrhhomhepkghi ucgjrghnuceoiihirdihrghnsehsvghnthdrtghomheqnecuggftrfgrthhtvghrnhepje ekteekffelleekudfftdefvddtjeejuedtuedtteegjefgvedtfedujeekieevnecuvehl uhhsthgvrhfuihiivgeptdenucfrrghrrghmpehmrghilhhfrhhomhepiihirdihrghnse hsvghnthdrtghomh X-ME-Proxy: Feedback-ID: iccd040f4:Fastmail Received: by mail.messagingengine.com (Postfix) with ESMTPA; Tue, 20 Feb 2024 13:32:47 -0500 (EST) From: Zi Yan To: linux-mm@kvack.org, linux-kernel@vger.kernel.org Cc: Zi Yan , "Huang, Ying" , Ryan Roberts , Andrew Morton , "Matthew Wilcox (Oracle)" , David Hildenbrand , "Yin, Fengwei" , Yu Zhao , Vlastimil Babka , "Kirill A . Shutemov" , Johannes Weiner , Baolin Wang , Kemeng Shi , Mel Gorman , Rohan Puri , Mcgrof Chamberlain , Adam Manzanares , "Vishal Moola (Oracle)" Subject: [PATCH v7 4/4] mm/compaction: optimize >0 order folio compaction with free page split. Date: Tue, 20 Feb 2024 13:32:20 -0500 Message-ID: <20240220183220.1451315-5-zi.yan@sent.com> X-Mailer: git-send-email 2.43.0 In-Reply-To: <20240220183220.1451315-1-zi.yan@sent.com> References: <20240220183220.1451315-1-zi.yan@sent.com> Reply-To: Zi Yan Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: quoted-printable From: Zi Yan During migration in a memory compaction, free pages are placed in an array of page lists based on their order. But the desired free page order (i.e., the order of a source page) might not be always present, thus leading to migration failures and premature compaction termination. Split a high order free pages when source migration page has a lower order to increase migration successful rate. Note: merging free pages when a migration fails and a lower order free page is returned via compaction_free() is possible, but there is too much work. Since the free pages are not buddy pages, it is hard to identify these free pages using existing PFN-based page merging algorithm. Signed-off-by: Zi Yan Reviewed-by: Baolin Wang Reviewed-by: Vlastimil Babka Tested-by: Baolin Wang Tested-by: Yu Zhao Cc: Adam Manzanares Cc: David Hildenbrand Cc: Huang Ying Cc: Johannes Weiner Cc: Kemeng Shi Cc: Kirill A. Shutemov Cc: Luis Chamberlain Cc: Matthew Wilcox (Oracle) Cc: Mel Gorman Cc: Ryan Roberts Cc: Vishal Moola (Oracle) Cc: Vlastimil Babka Cc: Yin Fengwei --- mm/compaction.c | 35 ++++++++++++++++++++++++++++++----- 1 file changed, 30 insertions(+), 5 deletions(-) diff --git a/mm/compaction.c b/mm/compaction.c index 112711752321..e70309674262 100644 --- a/mm/compaction.c +++ b/mm/compaction.c @@ -1856,15 +1856,40 @@ static struct folio *compaction_alloc(struct folio = *src, unsigned long data) struct compact_control *cc =3D (struct compact_control *)data; struct folio *dst; int order =3D folio_order(src); + bool has_isolated_pages =3D false; + int start_order; + struct page *freepage; + unsigned long size; + +again: + for (start_order =3D order; start_order < NR_PAGE_ORDERS; start_order++) + if (!list_empty(&cc->freepages[start_order])) + break; =20 - if (list_empty(&cc->freepages[order])) { - isolate_freepages(cc); - if (list_empty(&cc->freepages[order])) + /* no free pages in the list */ + if (start_order =3D=3D NR_PAGE_ORDERS) { + if (has_isolated_pages) return NULL; + isolate_freepages(cc); + has_isolated_pages =3D true; + goto again; + } + + freepage =3D list_first_entry(&cc->freepages[start_order], struct page, + lru); + size =3D 1 << start_order; + + list_del(&freepage->lru); + + while (start_order > order) { + start_order--; + size >>=3D 1; + + list_add(&freepage[size].lru, &cc->freepages[start_order]); + set_page_private(&freepage[size], start_order); } + dst =3D (struct folio *)freepage; =20 - dst =3D list_first_entry(&cc->freepages[order], struct folio, lru); - list_del(&dst->lru); post_alloc_hook(&dst->page, order, __GFP_MOVABLE); if (order) prep_compound_page(&dst->page, order); --=20 2.43.0