linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: Andrea Arcangeli <aarcange@redhat.com>
To: linux-mm@kvack.org, Andrew Morton <akpm@linux-foundation.org>
Cc: Marcelo Tosatti <mtosatti@redhat.com>,
	Adam Litke <agl@us.ibm.com>, Avi Kivity <avi@redhat.com>,
	Izik Eidus <ieidus@redhat.com>,
	Hugh Dickins <hugh.dickins@tiscali.co.uk>,
	Nick Piggin <npiggin@suse.de>, Rik van Riel <riel@redhat.com>,
	Mel Gorman <mel@csn.ul.ie>, Dave Hansen <dave@linux.vnet.ibm.com>,
	Benjamin Herrenschmidt <benh@kernel.crashing.org>,
	Ingo Molnar <mingo@elte.hu>, Mike Travis <travis@sgi.com>,
	KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>,
	Christoph Lameter <cl@linux-foundation.org>,
	Chris Wright <chrisw@sous-sol.org>,
	bpicco@redhat.com,
	KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>,
	Balbir Singh <balbir@linux.vnet.ibm.com>,
	Arnd Bergmann <arnd@arndb.de>,
	"Michael S. Tsirkin" <mst@redhat.com>,
	Peter Zijlstra <peterz@infradead.org>,
	Johannes Weiner <hannes@cmpxchg.org>,
	Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp>,
	Chris Mason <chris.mason@oracle.com>
Subject: [PATCH 62 of 67] do not display compaction-related stats when !CONFIG_COMPACTION
Date: Thu, 08 Apr 2010 03:51:45 +0200	[thread overview]
Message-ID: <13a20baba2f4a4fcd166.1270691505@v2.random> (raw)
In-Reply-To: <patchbomb.1270691443@v2.random>

From: Mel Gorman <mel@csn.ul.ie>

Although compaction can be disabled from .config, the vmstat entries still
exist.  This patch removes the vmstat entries.  As page_alloc.c refers
directly to the counters, the patch introduces
__alloc_pages_direct_compact() to isolate use of the counters.

Signed-off-by: Mel Gorman <mel@csn.ul.ie>
Cc: Minchan Kim <minchan.kim@gmail.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Christoph Lameter <cl@linux-foundation.org>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Rik van Riel <riel@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
---

diff --git a/include/linux/vmstat.h b/include/linux/vmstat.h
--- a/include/linux/vmstat.h
+++ b/include/linux/vmstat.h
@@ -43,8 +43,10 @@ enum vm_event_item { PGPGIN, PGPGOUT, PS
 		KSWAPD_LOW_WMARK_HIT_QUICKLY, KSWAPD_HIGH_WMARK_HIT_QUICKLY,
 		KSWAPD_SKIP_CONGESTION_WAIT,
 		PAGEOUTRUN, ALLOCSTALL, PGROTATED,
+#ifdef CONFIG_COMPACTION
 		COMPACTBLOCKS, COMPACTPAGES, COMPACTPAGEFAILED,
 		COMPACTSTALL, COMPACTFAIL, COMPACTSUCCESS,
+#endif
 #ifdef CONFIG_HUGETLB_PAGE
 		HTLB_BUDDY_PGALLOC, HTLB_BUDDY_PGALLOC_FAIL,
 #endif
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -1736,6 +1736,59 @@ out:
 	return page;
 }
 
+#ifdef CONFIG_COMPACTION
+/* Try memory compaction for high-order allocations before reclaim */
+static struct page *
+__alloc_pages_direct_compact(gfp_t gfp_mask, unsigned int order,
+	struct zonelist *zonelist, enum zone_type high_zoneidx,
+	nodemask_t *nodemask, int alloc_flags, struct zone *preferred_zone,
+	int migratetype, unsigned long *did_some_progress)
+{
+	struct page *page;
+
+	if (!order)
+		return NULL;
+
+	*did_some_progress = try_to_compact_pages(zonelist, order, gfp_mask,
+								nodemask);
+	if (*did_some_progress != COMPACT_SKIPPED) {
+
+		/* Page migration frees to the PCP lists but we want merging */
+		drain_pages(get_cpu());
+		put_cpu();
+
+		page = get_page_from_freelist(gfp_mask, nodemask,
+				order, zonelist, high_zoneidx,
+				alloc_flags, preferred_zone,
+				migratetype);
+		if (page) {
+			__count_vm_event(COMPACTSUCCESS);
+			return page;
+		}
+
+		/*
+		 * It's bad if compaction run occurs and fails.
+		 * The most likely reason is that pages exist,
+		 * but not enough to satisfy watermarks.
+		 */
+		count_vm_event(COMPACTFAIL);
+
+		cond_resched();
+	}
+
+	return NULL;
+}
+#else
+static inline struct page *
+__alloc_pages_direct_compact(gfp_t gfp_mask, unsigned int order,
+	struct zonelist *zonelist, enum zone_type high_zoneidx,
+	nodemask_t *nodemask, int alloc_flags, struct zone *preferred_zone,
+	int migratetype, unsigned long *did_some_progress)
+{
+	return NULL;
+}
+#endif /* CONFIG_COMPACTION */
+
 /* The really slow allocator path where we enter direct reclaim */
 static inline struct page *
 __alloc_pages_direct_reclaim(gfp_t gfp_mask, unsigned int order,
@@ -1957,6 +2010,15 @@ rebalance:
 	if (test_thread_flag(TIF_MEMDIE) && !(gfp_mask & __GFP_NOFAIL))
 		goto nopage;
 
+	/* Try direct compaction */
+	page = __alloc_pages_direct_compact(gfp_mask, order,
+					zonelist, high_zoneidx,
+					nodemask,
+					alloc_flags, preferred_zone,
+					migratetype, &did_some_progress);
+	if (page)
+		goto got_pg;
+
 	/* Try direct reclaim and then allocating */
 	page = __alloc_pages_direct_reclaim(gfp_mask, order,
 					zonelist, high_zoneidx,
diff --git a/mm/vmstat.c b/mm/vmstat.c
--- a/mm/vmstat.c
+++ b/mm/vmstat.c
@@ -905,12 +905,14 @@ static const char * const vmstat_text[] 
 
 	"pgrotated",
 
+#ifdef CONFIG_COMPACTION
 	"compact_blocks_moved",
 	"compact_pages_moved",
 	"compact_pagemigrate_failed",
 	"compact_stall",
 	"compact_fail",
 	"compact_success",
+#endif
 
 #ifdef CONFIG_HUGETLB_PAGE
 	"htlb_buddy_alloc_success",

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

  parent reply	other threads:[~2010-04-08  2:57 UTC|newest]

Thread overview: 95+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-04-08  1:50 [PATCH 00 of 67] Transparent Hugepage Support #18 Andrea Arcangeli
2010-04-08  1:50 ` [PATCH 01 of 67] define MADV_HUGEPAGE Andrea Arcangeli
2010-04-08  1:50 ` [PATCH 02 of 67] compound_lock Andrea Arcangeli
2010-04-08  1:50 ` [PATCH 03 of 67] alter compound get_page/put_page Andrea Arcangeli
2010-04-08  1:50 ` [PATCH 04 of 67] update futex compound knowledge Andrea Arcangeli
2010-04-08  1:50 ` [PATCH 05 of 67] fix bad_page to show the real reason the page is bad Andrea Arcangeli
2010-04-08  1:50 ` [PATCH 06 of 67] clear compound mapping Andrea Arcangeli
2010-04-08  1:50 ` [PATCH 07 of 67] add native_set_pmd_at Andrea Arcangeli
2010-04-08  1:50 ` [PATCH 08 of 67] add pmd paravirt ops Andrea Arcangeli
2010-04-08  1:50 ` [PATCH 09 of 67] no paravirt version of pmd ops Andrea Arcangeli
2010-04-08  1:50 ` [PATCH 10 of 67] export maybe_mkwrite Andrea Arcangeli
2010-04-08  1:50 ` [PATCH 11 of 67] comment reminder in destroy_compound_page Andrea Arcangeli
2010-04-08  1:50 ` [PATCH 12 of 67] config_transparent_hugepage Andrea Arcangeli
2010-04-08  1:50 ` [PATCH 13 of 67] special pmd_trans_* functions Andrea Arcangeli
2010-04-08  1:50 ` [PATCH 14 of 67] add pmd mangling generic functions Andrea Arcangeli
2010-04-08  1:50 ` [PATCH 15 of 67] add pmd mangling functions to x86 Andrea Arcangeli
2010-04-08  1:50 ` [PATCH 16 of 67] bail out gup_fast on splitting pmd Andrea Arcangeli
2010-04-08  1:51 ` [PATCH 17 of 67] pte alloc trans splitting Andrea Arcangeli
2010-04-08  1:51 ` [PATCH 18 of 67] add pmd mmu_notifier helpers Andrea Arcangeli
2010-04-08  1:51 ` [PATCH 19 of 67] clear page compound Andrea Arcangeli
2010-04-08  1:51 ` [PATCH 20 of 67] add pmd_huge_pte to mm_struct Andrea Arcangeli
2010-04-08  1:51 ` [PATCH 21 of 67] This fixes some minor issues that bugged me while going over the code: Andrea Arcangeli
2010-04-08  1:51 ` [PATCH 22 of 67] Split out functions to handle hugetlb ranges, pte ranges and unmapped Andrea Arcangeli
2010-04-08  1:51 ` [PATCH 23 of 67] Instead of passing a start address and a number of pages into the helper Andrea Arcangeli
2010-04-08  1:51 ` [PATCH 24 of 67] Do page table walks with the well-known nested loops we use in several Andrea Arcangeli
2010-04-08  1:51 ` [PATCH 25 of 67] split_huge_page_mm/vma Andrea Arcangeli
2010-04-08  1:51 ` [PATCH 26 of 67] split_huge_page paging Andrea Arcangeli
2010-04-08  1:51 ` [PATCH 27 of 67] clear_copy_huge_page Andrea Arcangeli
2010-04-08  1:51 ` [PATCH 28 of 67] kvm mmu transparent hugepage support Andrea Arcangeli
2010-04-08  1:51 ` [PATCH 29 of 67] _GFP_NO_KSWAPD Andrea Arcangeli
2010-04-08  1:51 ` [PATCH 30 of 67] don't alloc harder for gfp nomemalloc even if nowait Andrea Arcangeli
2010-04-08  1:51 ` [PATCH 31 of 67] transparent hugepage core Andrea Arcangeli
2010-04-08  1:51 ` [PATCH 32 of 67] verify pmd_trans_huge isn't leaking Andrea Arcangeli
2010-04-08  1:51 ` [PATCH 33 of 67] madvise(MADV_HUGEPAGE) Andrea Arcangeli
2010-04-08  1:51 ` [PATCH 34 of 67] pmd_trans_huge migrate bugcheck Andrea Arcangeli
2010-04-08  1:51 ` [PATCH 35 of 67] memcg compound Andrea Arcangeli
2010-04-08  1:51 ` [PATCH 36 of 67] memcg huge memory Andrea Arcangeli
2010-04-08  1:51 ` [PATCH 37 of 67] transparent hugepage vmstat Andrea Arcangeli
2010-04-08 11:53   ` Avi Kivity
2010-04-08  1:51 ` [PATCH 38 of 67] khugepaged Andrea Arcangeli
2010-04-08  1:51 ` [PATCH 39 of 67] don't leave orhpaned swap cache after ksm merging Andrea Arcangeli
2010-04-08  1:51 ` [PATCH 40 of 67] skip transhuge pages in ksm for now Andrea Arcangeli
2010-04-08  1:51 ` [PATCH 41 of 67] remove PG_buddy Andrea Arcangeli
2010-04-08  1:51 ` [PATCH 42 of 67] add x86 32bit support Andrea Arcangeli
2010-04-08  1:51 ` [PATCH 43 of 67] mincore transparent hugepage support Andrea Arcangeli
2010-04-08  1:51 ` [PATCH 44 of 67] add pmd_modify Andrea Arcangeli
2010-04-08  1:51 ` [PATCH 45 of 67] mprotect: pass vma down to page table walkers Andrea Arcangeli
2010-04-08  1:51 ` [PATCH 46 of 67] mprotect: transparent huge page support Andrea Arcangeli
2010-04-08  1:51 ` [PATCH 47 of 67] set recommended min free kbytes Andrea Arcangeli
2010-04-08  1:51 ` [PATCH 48 of 67] remove lumpy_reclaim Andrea Arcangeli
2010-04-08  1:51 ` [PATCH 49 of 67] Take a reference to the anon_vma before migrating Andrea Arcangeli
2010-04-08  1:51 ` [PATCH 50 of 67] Do not try to migrate unmapped anonymous pages Andrea Arcangeli
2010-04-08  1:51 ` [PATCH 51 of 67] Share the anon_vma ref counts between KSM and page migration Andrea Arcangeli
2010-04-08  1:51 ` [PATCH 52 of 67] Allow CONFIG_MIGRATION to be set without CONFIG_NUMA or memory hot-remove Andrea Arcangeli
2010-04-08  1:51 ` [PATCH 53 of 67] Export unusable free space index via /proc/unusable_index Andrea Arcangeli
2010-04-08  1:51 ` [PATCH 54 of 67] Export fragmentation index via /proc/extfrag_index Andrea Arcangeli
2010-04-08  1:51 ` [PATCH 55 of 67] Move definition for LRU isolation modes to a header Andrea Arcangeli
2010-04-08  1:51 ` [PATCH 56 of 67] Memory compaction core Andrea Arcangeli
2010-04-08 16:18   ` Johannes Weiner
2010-04-08 16:46     ` Andrea Arcangeli
2010-04-08 17:09       ` Andrea Arcangeli
2010-04-08 17:14         ` Andrea Arcangeli
2010-04-08 17:56           ` Johannes Weiner
2010-04-08 17:58             ` Andrea Arcangeli
2010-04-08 18:48               ` Johannes Weiner
2010-04-08 21:23                 ` Andrea Arcangeli
2010-04-08 21:32                   ` Andrea Arcangeli
2010-04-09 10:51                   ` Mel Gorman
2010-04-09 15:37                     ` Andrea Arcangeli
2010-04-08  1:51 ` [PATCH 57 of 67] Add /proc trigger for memory compaction Andrea Arcangeli
2010-04-08  1:51 ` [PATCH 58 of 67] Add /sys trigger for per-node " Andrea Arcangeli
2010-04-08  1:51 ` [PATCH 59 of 67] Direct compact when a high-order allocation fails Andrea Arcangeli
2010-04-08  1:51 ` [PATCH 60 of 67] Add a tunable that decides when memory should be compacted and when it should be reclaimed Andrea Arcangeli
2010-04-08  1:51 ` [PATCH 61 of 67] Allow the migration of PageSwapCache pages Andrea Arcangeli
2010-04-08  1:51 ` Andrea Arcangeli [this message]
2010-04-08  1:51 ` [PATCH 63 of 67] disable migreate_prep() Andrea Arcangeli
2010-04-08  1:51 ` [PATCH 64 of 67] page buddy can go away before reading page_order Andrea Arcangeli
2010-04-08  1:51 ` [PATCH 65 of 67] select CONFIG_COMPACTION if TRANSPARENT_HUGEPAGE enabled Andrea Arcangeli
2010-04-08  1:51 ` [PATCH 66 of 67] enable direct defrag Andrea Arcangeli
2010-04-08  1:51 ` [PATCH 67 of 67] memcg fix prepare migration Andrea Arcangeli
2010-04-08  3:57   ` Daisuke Nishimura
2010-04-13  1:29     ` Andrew Morton
2010-04-09  8:13   ` KAMEZAWA Hiroyuki
2010-04-08  9:39 ` [PATCH 00 of 67] Transparent Hugepage Support #18 Avi Kivity
2010-04-08 11:44   ` Avi Kivity
2010-04-08 15:23     ` Andrea Arcangeli
2010-04-08 15:27       ` Avi Kivity
2010-04-08 16:02         ` Andrea Arcangeli
2010-04-08 15:32       ` Christoph Lameter
2010-04-08 23:17         ` Andrea Arcangeli
2010-04-09  8:45     ` Avi Kivity
2010-04-09 15:50       ` Andrea Arcangeli
2010-04-09 17:44         ` Avi Kivity
2010-04-09  2:05 ` Transparent Hugepage Support #19 Andrea Arcangeli
2010-04-09 15:43   ` Andrea Arcangeli

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=13a20baba2f4a4fcd166.1270691505@v2.random \
    --to=aarcange@redhat.com \
    --cc=agl@us.ibm.com \
    --cc=akpm@linux-foundation.org \
    --cc=arnd@arndb.de \
    --cc=avi@redhat.com \
    --cc=balbir@linux.vnet.ibm.com \
    --cc=benh@kernel.crashing.org \
    --cc=bpicco@redhat.com \
    --cc=chris.mason@oracle.com \
    --cc=chrisw@sous-sol.org \
    --cc=cl@linux-foundation.org \
    --cc=dave@linux.vnet.ibm.com \
    --cc=hannes@cmpxchg.org \
    --cc=hugh.dickins@tiscali.co.uk \
    --cc=ieidus@redhat.com \
    --cc=kamezawa.hiroyu@jp.fujitsu.com \
    --cc=kosaki.motohiro@jp.fujitsu.com \
    --cc=linux-mm@kvack.org \
    --cc=mel@csn.ul.ie \
    --cc=mingo@elte.hu \
    --cc=mst@redhat.com \
    --cc=mtosatti@redhat.com \
    --cc=nishimura@mxp.nes.nec.co.jp \
    --cc=npiggin@suse.de \
    --cc=peterz@infradead.org \
    --cc=riel@redhat.com \
    --cc=travis@sgi.com \
    /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;
as well as URLs for NNTP newsgroup(s).