linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: Alex Shi <alex.shi@linux.alibaba.com>
To: Anshuman Khandual <anshuman.khandual@arm.com>,
	David Hildenbrand <david@redhat.com>,
	Matthew Wilcox <willy@infradead.org>,
	Vlastimil Babka <vbabka@suse.cz>,
	Alexander Duyck <alexander.duyck@gmail.com>
Cc: Andrew Morton <akpm@linux-foundation.org>,
	Mel Gorman <mgorman@techsingularity.net>,
	linux-kernel@vger.kernel.org, linux-mm@kvack.org
Subject: [PATCH v4 2/4] mm/pageblock: remove false sharing in pageblock_flags
Date: Thu,  3 Sep 2020 15:01:21 +0800	[thread overview]
Message-ID: <1599116482-7410-2-git-send-email-alex.shi@linux.alibaba.com> (raw)
In-Reply-To: <1599116482-7410-1-git-send-email-alex.shi@linux.alibaba.com>

Current pageblock_flags is only 4 bits, so it has to share a char size
in cmpxchg when get set, the false sharing cause perf drop.

If we incrase the bits up to 8, false sharing would gone in cmpxchg. and
the only cost is half char per pageblock, which is half char per 128MB
on x86, 4 chars in 1 GB.

Signed-off-by: Alex Shi <alex.shi@linux.alibaba.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Mel Gorman <mgorman@techsingularity.net>
Cc: linux-kernel@vger.kernel.org
Cc: linux-mm@kvack.org
---
 include/linux/page-isolation.h  |  2 +-
 include/linux/pageblock-flags.h |  2 +-
 mm/page_alloc.c                 | 10 +++-------
 3 files changed, 5 insertions(+), 9 deletions(-)

diff --git a/include/linux/page-isolation.h b/include/linux/page-isolation.h
index 572458016331..baa2e1ecc134 100644
--- a/include/linux/page-isolation.h
+++ b/include/linux/page-isolation.h
@@ -35,7 +35,7 @@ static inline bool is_migrate_isolate(int migratetype)
 
 struct page *has_unmovable_pages(struct zone *zone, struct page *page,
 				 int migratetype, int flags);
-void set_pageblock_migratetype(struct page *page, int migratetype);
+void set_pageblock_migratetype(struct page *page, unsigned char migratetype);
 int move_freepages_block(struct zone *zone, struct page *page,
 				int migratetype, int *num_movable);
 
diff --git a/include/linux/pageblock-flags.h b/include/linux/pageblock-flags.h
index d189441568eb..f785c9d6d68c 100644
--- a/include/linux/pageblock-flags.h
+++ b/include/linux/pageblock-flags.h
@@ -25,7 +25,7 @@ enum pageblock_bits {
 	 * Assume the bits will always align on a word. If this assumption
 	 * changes then get/set pageblock needs updating.
 	 */
-	NR_PAGEBLOCK_BITS
+	NR_PAGEBLOCK_BITS = BITS_PER_BYTE
 };
 
 #ifdef CONFIG_HUGETLB_PAGE
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 81e96d4d9c42..3688e6b83318 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -517,19 +517,15 @@ void set_pfnblock_flags_mask(struct page *page, unsigned long flags,
 	unsigned long bitidx, byte_bitidx;
 	unsigned char old_byte, byte;
 
-	BUILD_BUG_ON(NR_PAGEBLOCK_BITS != 4);
+	BUILD_BUG_ON(NR_PAGEBLOCK_BITS != BITS_PER_BYTE);
 	BUILD_BUG_ON(MIGRATE_TYPES > (1 << PB_migratetype_bits));
 
 	bitmap = get_pageblock_bitmap(page, pfn);
 	bitidx = pfn_to_bitidx(page, pfn);
 	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;
-
 	byte = READ_ONCE(bitmap[byte_bitidx]);
 	for (;;) {
 		old_byte = cmpxchg(&bitmap[byte_bitidx], byte, (byte & ~mask) | flags);
@@ -539,13 +535,13 @@ void set_pfnblock_flags_mask(struct page *page, unsigned long flags,
 	}
 }
 
-void set_pageblock_migratetype(struct page *page, int migratetype)
+void set_pageblock_migratetype(struct page *page, unsigned char migratetype)
 {
 	if (unlikely(page_group_by_mobility_disabled &&
 		     migratetype < MIGRATE_PCPTYPES))
 		migratetype = MIGRATE_UNMOVABLE;
 
-	set_pfnblock_flags_mask(page, (unsigned long)migratetype,
+	set_pfnblock_flags_mask(page, migratetype,
 				page_to_pfn(page), MIGRATETYPE_MASK);
 }
 
-- 
1.8.3.1



  reply	other threads:[~2020-09-03  7:01 UTC|newest]

Thread overview: 13+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-09-03  7:01 [PATCH v4 1/4] mm/pageblock: mitigation cmpxchg false sharing in pageblock flags Alex Shi
2020-09-03  7:01 ` Alex Shi [this message]
2020-09-03  7:01 ` [PATCH v4 3/4] mm/pageblock: work around multiple arch's cmpxchg support issue Alex Shi
2020-09-03  7:29   ` Max Filippov
2020-09-03  8:50     ` Alex Shi
2020-09-10  5:51   ` Christoph Hellwig
2020-09-03  7:24 ` [PATCH v4 1/4] mm/pageblock: mitigation cmpxchg false sharing in pageblock flags Mel Gorman
2020-09-03  8:32   ` Alex Shi
2020-09-03  8:40     ` Alex Shi
2020-09-03  9:00       ` Vlastimil Babka
2020-09-03  9:31     ` Mel Gorman
2020-09-03  8:19 ` David Hildenbrand
2020-09-03  9:14   ` Alex Shi

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1599116482-7410-2-git-send-email-alex.shi@linux.alibaba.com \
    --to=alex.shi@linux.alibaba.com \
    --cc=akpm@linux-foundation.org \
    --cc=alexander.duyck@gmail.com \
    --cc=anshuman.khandual@arm.com \
    --cc=david@redhat.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=mgorman@techsingularity.net \
    --cc=vbabka@suse.cz \
    --cc=willy@infradead.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox