From patchwork Tue Sep 1 02:50:10 2020 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Alex Shi X-Patchwork-Id: 11747255 Return-Path: Received: from mail.kernel.org (pdx-korg-mail-1.web.codeaurora.org [172.30.200.123]) by pdx-korg-patchwork-2.web.codeaurora.org (Postfix) with ESMTP id 07C9414E5 for ; Tue, 1 Sep 2020 02:50:34 +0000 (UTC) Received: from kanga.kvack.org (kanga.kvack.org [205.233.56.17]) by mail.kernel.org (Postfix) with ESMTP id D303E207EA for ; Tue, 1 Sep 2020 02:50:33 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.3.2 mail.kernel.org D303E207EA Authentication-Results: mail.kernel.org; dmarc=fail (p=none dis=none) header.from=linux.alibaba.com Authentication-Results: mail.kernel.org; spf=pass smtp.mailfrom=owner-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix) id ABF456B005A; Mon, 31 Aug 2020 22:50:30 -0400 (EDT) Delivered-To: linux-mm-outgoing@kvack.org Received: by kanga.kvack.org (Postfix, from userid 40) id 916F56B005D; Mon, 31 Aug 2020 22:50:30 -0400 (EDT) X-Original-To: int-list-linux-mm@kvack.org X-Delivered-To: int-list-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix, from userid 63042) id 8046F6B005A; Mon, 31 Aug 2020 22:50:30 -0400 (EDT) X-Original-To: linux-mm@kvack.org X-Delivered-To: linux-mm@kvack.org Received: from forelay.hostedemail.com (smtprelay0183.hostedemail.com [216.40.44.183]) by kanga.kvack.org (Postfix) with ESMTP id 671386B0037 for ; Mon, 31 Aug 2020 22:50:30 -0400 (EDT) Received: from smtpin03.hostedemail.com (10.5.19.251.rfc1918.com [10.5.19.251]) by forelay02.hostedemail.com (Postfix) with ESMTP id 269ED3636 for ; Tue, 1 Sep 2020 02:50:30 +0000 (UTC) X-FDA: 77212964220.03.spade23_03075fd27094 Received: from filter.hostedemail.com (10.5.16.251.rfc1918.com [10.5.16.251]) by smtpin03.hostedemail.com (Postfix) with ESMTP id ECFBA28A4E8 for ; Tue, 1 Sep 2020 02:50:29 +0000 (UTC) X-Spam-Summary: 1,0,0,e51e5ae7fc84da53,d41d8cd98f00b204,alex.shi@linux.alibaba.com,,RULES_HIT:2:41:69:355:379:541:800:960:973:988:989:1260:1261:1345:1431:1437:1535:1605:1606:1730:1747:1777:1792:1801:2393:2559:2562:2693:2742:3138:3139:3140:3141:3142:3865:3866:3867:3868:3870:3871:3872:3874:4118:4321:4605:5007:6117:6261:7903:7904:8531:8603:9036:9121:10004:11026:11473:11658:11914:12043:12296:12297:12438:12555:12683:12895:12986:13138:13231:13255:13846:14394:21060:21080:21324:21451:21627:21795:21990:30051:30054:30070,0,RBL:115.124.30.132:@linux.alibaba.com:.lbl8.mailshell.net-62.20.2.100 64.201.201.201;04yf7fzmufqm1jf1ytogz5xfizk5jocw7i4ni4a48i39n4oxnqzgtggbrjbf1ct.wuqkdf387ezyz7bq6aatt4gazjbcmm4sbwdqtdwfbbfm3g7ff443hdckgbyn3yi.1-lbl8.mailshell.net-223.238.255.100,CacheIP:none,Bayesian:0.5,0.5,0.5,Netcheck:none,DomainCache:0,MSF:not bulk,SPF:fp,MSBL:0,DNSBL:none,Custom_rules:0:0:0,LFtime:23,LUA_SUMMARY:none X-HE-Tag: spade23_03075fd27094 X-Filterd-Recvd-Size: 7144 Received: from out30-132.freemail.mail.aliyun.com (out30-132.freemail.mail.aliyun.com [115.124.30.132]) by imf43.hostedemail.com (Postfix) with ESMTP for ; Tue, 1 Sep 2020 02:50:27 +0000 (UTC) X-Alimail-AntiSpam: AC=PASS;BC=-1|-1;BR=01201311R141e4;CH=green;DM=||false|;DS=||;FP=0|-1|-1|-1|0|-1|-1|-1;HT=e01e01355;MF=alex.shi@linux.alibaba.com;NM=1;PH=DS;RN=7;SR=0;TI=SMTPD_---0U7Txrtg_1598928621; Received: from aliy80.localdomain(mailfrom:alex.shi@linux.alibaba.com fp:SMTPD_---0U7Txrtg_1598928621) by smtp.aliyun-inc.com(127.0.0.1); Tue, 01 Sep 2020 10:50:22 +0800 From: Alex Shi To: Anshuman Khandual , David Hildenbrand , Matthew Wilcox Cc: Andrew Morton , Mel Gorman , linux-mm@kvack.org, linux-kernel@vger.kernel.org Subject: [PATCH v3 1/3] mm/pageblock: mitigation cmpxchg false sharing in pageblock flags Date: Tue, 1 Sep 2020 10:50:10 +0800 Message-Id: <1598928612-68996-1-git-send-email-alex.shi@linux.alibaba.com> X-Mailer: git-send-email 1.8.3.1 X-Rspamd-Queue-Id: ECFBA28A4E8 X-Spamd-Result: default: False [0.00 / 100.00] X-Rspamd-Server: rspam02 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: pageblock_flags is used as long, since every pageblock_flags is just 4 bits, 'long' size will include 8(32bit machine) or 16 pageblocks' flags, that flag setting has to sync in cmpxchg with 7 or 15 other pageblock flags. It would cause long waiting for sync. If we could change the pageblock_flags variable as char, we could use char size cmpxchg, which just sync up with 2 pageblock flags. it could relief much false sharing in cmpxchg. With this and next patch, we could see mmtests/thpscale get slight fast on my 4 cores box, and cmpxchg retry times is reduced. pageblock pageblock pageblock rc2 rc2 rc2 16 16-2 16-3 a b c Duration User 14.81 15.24 14.55 14.76 14.97 14.38 Duration System 84.44 88.38 90.64 100.43 89.15 88.89 Duration Elapsed 98.83 99.06 99.81 100.30 99.24 99.14 rc2 is 5.9-rc2 kernel, pageblock is 5.9-rc2 + this patchset Signed-off-by: Alex Shi Cc: Andrew Morton Cc: Mel Gorman Cc: linux-mm@kvack.org Cc: linux-kernel@vger.kernel.org --- include/linux/mmzone.h | 6 +++--- include/linux/pageblock-flags.h | 2 +- mm/page_alloc.c | 38 +++++++++++++++++++------------------- 3 files changed, 23 insertions(+), 23 deletions(-) diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h index 8379432f4f2f..be676e659fb7 100644 --- a/include/linux/mmzone.h +++ b/include/linux/mmzone.h @@ -437,7 +437,7 @@ struct zone { * Flags for a pageblock_nr_pages block. See pageblock-flags.h. * In SPARSEMEM, this map is stored in struct mem_section */ - unsigned long *pageblock_flags; + unsigned char *pageblock_flags; #endif /* CONFIG_SPARSEMEM */ /* zone_start_pfn == zone_start_paddr >> PAGE_SHIFT */ @@ -1159,7 +1159,7 @@ struct mem_section_usage { DECLARE_BITMAP(subsection_map, SUBSECTIONS_PER_SECTION); #endif /* See declaration of similar field in struct zone */ - unsigned long pageblock_flags[0]; + unsigned char pageblock_flags[0]; }; void subsection_map_init(unsigned long pfn, unsigned long nr_pages); @@ -1212,7 +1212,7 @@ struct mem_section { extern struct mem_section mem_section[NR_SECTION_ROOTS][SECTIONS_PER_ROOT]; #endif -static inline unsigned long *section_to_usemap(struct mem_section *ms) +static inline unsigned char *section_to_usemap(struct mem_section *ms) { return ms->usage->pageblock_flags; } diff --git a/include/linux/pageblock-flags.h b/include/linux/pageblock-flags.h index fff52ad370c1..d189441568eb 100644 --- a/include/linux/pageblock-flags.h +++ b/include/linux/pageblock-flags.h @@ -54,7 +54,7 @@ enum pageblock_bits { /* Forward declaration */ struct page; -unsigned long get_pfnblock_flags_mask(struct page *page, +unsigned char get_pfnblock_flags_mask(struct page *page, unsigned long pfn, unsigned long mask); diff --git a/mm/page_alloc.c b/mm/page_alloc.c index fab5e97dc9ca..81e96d4d9c42 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -445,7 +445,7 @@ static inline bool defer_init(int nid, unsigned long pfn, unsigned long end_pfn) #endif /* Return a pointer to the bitmap storing bits affecting a block of pages */ -static inline unsigned long *get_pageblock_bitmap(struct page *page, +static inline unsigned char *get_pageblock_bitmap(struct page *page, unsigned long pfn) { #ifdef CONFIG_SPARSEMEM @@ -474,24 +474,24 @@ static inline int pfn_to_bitidx(struct page *page, unsigned long pfn) * Return: pageblock_bits flags */ static __always_inline -unsigned long __get_pfnblock_flags_mask(struct page *page, +unsigned char __get_pfnblock_flags_mask(struct page *page, unsigned long pfn, unsigned long mask) { - unsigned long *bitmap; - unsigned long bitidx, word_bitidx; - unsigned long word; + unsigned char *bitmap; + unsigned long bitidx, byte_bitidx; + unsigned char byte; bitmap = get_pageblock_bitmap(page, pfn); bitidx = pfn_to_bitidx(page, pfn); - word_bitidx = bitidx / BITS_PER_LONG; - bitidx &= (BITS_PER_LONG-1); + byte_bitidx = bitidx / BITS_PER_BYTE; + bitidx &= (BITS_PER_BYTE-1); - word = bitmap[word_bitidx]; - return (word >> bitidx) & mask; + byte = bitmap[byte_bitidx]; + return (byte >> bitidx) & mask; } -unsigned long get_pfnblock_flags_mask(struct page *page, unsigned long pfn, +unsigned char get_pfnblock_flags_mask(struct page *page, unsigned long pfn, unsigned long mask) { return __get_pfnblock_flags_mask(page, pfn, mask); @@ -513,29 +513,29 @@ void set_pfnblock_flags_mask(struct page *page, unsigned long flags, unsigned long pfn, unsigned long mask) { - unsigned long *bitmap; - unsigned long bitidx, word_bitidx; - unsigned long old_word, word; + unsigned char *bitmap; + unsigned long bitidx, byte_bitidx; + unsigned char old_byte, byte; BUILD_BUG_ON(NR_PAGEBLOCK_BITS != 4); BUILD_BUG_ON(MIGRATE_TYPES > (1 << PB_migratetype_bits)); bitmap = get_pageblock_bitmap(page, pfn); bitidx = pfn_to_bitidx(page, pfn); - word_bitidx = bitidx / BITS_PER_LONG; - bitidx &= (BITS_PER_LONG-1); + byte_bitidx = bitidx / BITS_PER_BYTE; + bitidx &= (BITS_PER_BYTE-1); VM_BUG_ON_PAGE(!zone_spans_pfn(page_zone(page), pfn), page); mask <<= bitidx; flags <<= bitidx; - word = READ_ONCE(bitmap[word_bitidx]); + byte = READ_ONCE(bitmap[byte_bitidx]); for (;;) { - old_word = cmpxchg(&bitmap[word_bitidx], word, (word & ~mask) | flags); - if (word == old_word) + old_byte = cmpxchg(&bitmap[byte_bitidx], byte, (byte & ~mask) | flags); + if (byte == old_byte) break; - word = old_word; + byte = old_byte; } }