From patchwork Wed Apr 20 09:59:02 2022 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Mel Gorman X-Patchwork-Id: 12820009 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from kanga.kvack.org (kanga.kvack.org [205.233.56.17]) by smtp.lore.kernel.org (Postfix) with ESMTP id C52D4C433EF for ; Wed, 20 Apr 2022 09:59:41 +0000 (UTC) Received: by kanga.kvack.org (Postfix) id 39FE06B007E; Wed, 20 Apr 2022 05:59:41 -0400 (EDT) Received: by kanga.kvack.org (Postfix, from userid 40) id 34E8E6B0081; Wed, 20 Apr 2022 05:59:41 -0400 (EDT) X-Delivered-To: int-list-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix, from userid 63042) id 2170B6B0082; Wed, 20 Apr 2022 05:59:41 -0400 (EDT) X-Delivered-To: linux-mm@kvack.org Received: from relay.hostedemail.com (relay.a.hostedemail.com [64.99.140.24]) by kanga.kvack.org (Postfix) with ESMTP id 13FD16B007E for ; Wed, 20 Apr 2022 05:59:41 -0400 (EDT) Received: from smtpin11.hostedemail.com (a10.router.float.18 [10.200.18.1]) by unirelay11.hostedemail.com (Postfix) with ESMTP id D7EFF813F0 for ; Wed, 20 Apr 2022 09:59:40 +0000 (UTC) X-FDA: 79376810520.11.DD25948 Received: from outbound-smtp63.blacknight.com (outbound-smtp63.blacknight.com [46.22.136.252]) by imf18.hostedemail.com (Postfix) with ESMTP id 020E41C000D for ; Wed, 20 Apr 2022 09:59:38 +0000 (UTC) Received: from mail.blacknight.com (pemlinmail04.blacknight.ie [81.17.254.17]) by outbound-smtp63.blacknight.com (Postfix) with ESMTPS id 3A7BFFAE19 for ; Wed, 20 Apr 2022 10:59:39 +0100 (IST) Received: (qmail 10849 invoked from network); 20 Apr 2022 09:59:39 -0000 Received: from unknown (HELO morpheus.112glenside.lan) (mgorman@techsingularity.net@[84.203.198.246]) by 81.17.254.9 with ESMTPA; 20 Apr 2022 09:59:38 -0000 From: Mel Gorman To: Nicolas Saenz Julienne Cc: Marcelo Tosatti , Vlastimil Babka , Michal Hocko , LKML , Linux-MM , Mel Gorman Subject: [PATCH 2/6] mm/page_alloc: Use only one PCP list for THP-sized allocations Date: Wed, 20 Apr 2022 10:59:02 +0100 Message-Id: <20220420095906.27349-3-mgorman@techsingularity.net> X-Mailer: git-send-email 2.34.1 In-Reply-To: <20220420095906.27349-1-mgorman@techsingularity.net> References: <20220420095906.27349-1-mgorman@techsingularity.net> MIME-Version: 1.0 Authentication-Results: imf18.hostedemail.com; dkim=none; dmarc=none; spf=pass (imf18.hostedemail.com: domain of mgorman@techsingularity.net designates 46.22.136.252 as permitted sender) smtp.mailfrom=mgorman@techsingularity.net X-Rspam-User: X-Rspamd-Server: rspam08 X-Rspamd-Queue-Id: 020E41C000D X-Stat-Signature: 6kdokznkzzey9m1m894d57bjybitnf4e X-HE-Tag: 1650448778-295524 X-Bogosity: Ham, tests=bogofilter, spamicity=0.000000, version=1.2.4 Sender: owner-linux-mm@kvack.org Precedence: bulk X-Loop: owner-majordomo@kvack.org List-ID: The per_cpu_pages is cache-aligned on a standard x86-64 distribution configuration but a later patch will add a new field which would push the structure into the next cache line. Use only one list to store THP-sized pages on the per-cpu list. This assumes that the vast majority of THP-sized allocations are GFP_MOVABLE but even if it was another type, it would not contribute to serious fragmentation that potentially causes a later THP allocation failure. Align per_cpu_pages on the cacheline boundary to ensure there is no false cache sharing. After this patch, the structure sizing is; struct per_cpu_pages { int count; /* 0 4 */ int high; /* 4 4 */ int batch; /* 8 4 */ short int free_factor; /* 12 2 */ short int expire; /* 14 2 */ struct list_head lists[13]; /* 16 208 */ /* size: 256, cachelines: 4, members: 6 */ /* padding: 32 */ } __attribute__((__aligned__(64))); Signed-off-by: Mel Gorman --- include/linux/mmzone.h | 11 +++++++---- mm/page_alloc.c | 4 ++-- 2 files changed, 9 insertions(+), 6 deletions(-) diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h index 962b14d403e8..abe530748de6 100644 --- a/include/linux/mmzone.h +++ b/include/linux/mmzone.h @@ -358,15 +358,18 @@ enum zone_watermarks { }; /* - * One per migratetype for each PAGE_ALLOC_COSTLY_ORDER plus one additional - * for pageblock size for THP if configured. + * One per migratetype for each PAGE_ALLOC_COSTLY_ORDER. One additional list + * for THP which will usually be GFP_MOVABLE. Even if it is another type, + * it should not contribute to serious fragmentation causing THP allocation + * failures. */ #ifdef CONFIG_TRANSPARENT_HUGEPAGE #define NR_PCP_THP 1 #else #define NR_PCP_THP 0 #endif -#define NR_PCP_LISTS (MIGRATE_PCPTYPES * (PAGE_ALLOC_COSTLY_ORDER + 1 + NR_PCP_THP)) +#define NR_LOWORDER_PCP_LISTS (MIGRATE_PCPTYPES * (PAGE_ALLOC_COSTLY_ORDER + 1)) +#define NR_PCP_LISTS (NR_LOWORDER_PCP_LISTS + NR_PCP_THP) /* * Shift to encode migratetype and order in the same integer, with order @@ -392,7 +395,7 @@ struct per_cpu_pages { /* Lists of pages, one per migrate type stored on the pcp-lists */ struct list_head lists[NR_PCP_LISTS]; -}; +} ____cacheline_aligned_in_smp; struct per_cpu_zonestat { #ifdef CONFIG_SMP diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 63976ad4b7f1..ed2deb93a758 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -648,7 +648,7 @@ static inline unsigned int order_to_pindex(int migratetype, int order) #ifdef CONFIG_TRANSPARENT_HUGEPAGE if (order > PAGE_ALLOC_COSTLY_ORDER) { VM_BUG_ON(order != pageblock_order); - base = PAGE_ALLOC_COSTLY_ORDER + 1; + return NR_LOWORDER_PCP_LISTS; } #else VM_BUG_ON(order > PAGE_ALLOC_COSTLY_ORDER); @@ -662,7 +662,7 @@ static inline int pindex_to_order(unsigned int pindex) int order = pindex / MIGRATE_PCPTYPES; #ifdef CONFIG_TRANSPARENT_HUGEPAGE - if (order > PAGE_ALLOC_COSTLY_ORDER) + if (pindex == NR_LOWORDER_PCP_LISTS) order = pageblock_order; #else VM_BUG_ON(order > PAGE_ALLOC_COSTLY_ORDER);