The patch titled
Subject: mm: compaction: update the cc->nr_migratepages when allocating or freeing the freepages
has been added to the -mm mm-unstable branch. Its filename is
mm-compaction-update-the-cc-nr_migratepages-when-allocating-or-freeing-the-freepages.patch
This patch will shortly appear at
https://git.kernel.org/pub/scm/linux/kernel/git/akpm/25-new.git/tree/patches/mm-compaction-update-the-cc-nr_migratepages-when-allocating-or-freeing-the-freepages.patch
This patch will later appear in the mm-unstable branch at
git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm
Before you just go and hit "reply", please:
a) Consider who else should be cc'ed
b) Prefer to cc a suitable mailing list as well
c) Ideally: find the original patch on the mailing list and do a
reply-to-all to that, adding suitable additional cc's
*** Remember to use Documentation/process/submit-checklist.rst when testing your code ***
The -mm tree is included into linux-next via the mm-everything
branch at git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm
and is updated there every 2-3 working days
------------------------------------------------------
From: Baolin Wang <baolin.wang@xxxxxxxxxxxxxxxxx>
Subject: mm: compaction: update the cc->nr_migratepages when allocating or freeing the freepages
Date: Tue, 20 Feb 2024 14:16:31 +0800
Currently we will use 'cc->nr_freepages >= cc->nr_migratepages' comparison
to ensure that enough freepages are isolated in isolate_freepages(),
however it just decreases the cc->nr_freepages without updating
cc->nr_migratepages in compaction_alloc(), which will waste more CPU
cycles and cause too many freepages to be isolated.
So we should also update the cc->nr_migratepages when allocating or
freeing the freepages to avoid isolating excess freepages. And I can see
fewer free pages are scanned and isolated when running thpcompact on my
Arm64 server:
k6.7 k6.7_patched
Ops Compaction pages isolated 120692036.00 118160797.00
Ops Compaction migrate scanned 131210329.00 154093268.00
Ops Compaction free scanned 1090587971.00 1080632536.00
Ops Compact scan efficiency 12.03 14.26
Moreover, I did not see an obvious latency improvements, this is likely
because isolating freepages is not the bottleneck in the thpcompact test
case.
k6.7 k6.7_patched
Amean fault-both-1 1089.76 ( 0.00%) 1080.16 * 0.88%*
Amean fault-both-3 1616.48 ( 0.00%) 1636.65 * -1.25%*
Amean fault-both-5 2266.66 ( 0.00%) 2219.20 * 2.09%*
Amean fault-both-7 2909.84 ( 0.00%) 2801.90 * 3.71%*
Amean fault-both-12 4861.26 ( 0.00%) 4733.25 * 2.63%*
Amean fault-both-18 7351.11 ( 0.00%) 6950.51 * 5.45%*
Amean fault-both-24 9059.30 ( 0.00%) 9159.99 * -1.11%*
Amean fault-both-30 10685.68 ( 0.00%) 11399.02 * -6.68%*
Link: https://lkml.kernel.org/r/6440493f18da82298152b6305d6b41c2962a3ce6.1708409245.git.baolin.wang@xxxxxxxxxxxxxxxxx
Signed-off-by: Baolin Wang <baolin.wang@xxxxxxxxxxxxxxxxx>
Acked-by: Mel Gorman <mgorman@xxxxxxxxxxxxxxxxxxx>
Reviewed-by: Vlastimil Babka <vbabka@xxxxxxx>
Cc: Masami Hiramatsu <mhiramat@xxxxxxxxxx>
Cc: Mathieu Desnoyers <mathieu.desnoyers@xxxxxxxxxxxx>
Cc: Steven Rostedt <rostedt@xxxxxxxxxxx>
Cc: Zi Yan <ziy@xxxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---
include/trace/events/compaction.h | 6 +++---
mm/compaction.c | 12 ++++++++++--
2 files changed, 13 insertions(+), 5 deletions(-)
--- a/include/trace/events/compaction.h~mm-compaction-update-the-cc-nr_migratepages-when-allocating-or-freeing-the-freepages
+++ a/include/trace/events/compaction.h
@@ -78,10 +78,10 @@ DEFINE_EVENT(mm_compaction_isolate_templ
#ifdef CONFIG_COMPACTION
TRACE_EVENT(mm_compaction_migratepages,
- TP_PROTO(struct compact_control *cc,
+ TP_PROTO(unsigned int nr_migratepages,
unsigned int nr_succeeded),
- TP_ARGS(cc, nr_succeeded),
+ TP_ARGS(nr_migratepages, nr_succeeded),
TP_STRUCT__entry(
__field(unsigned long, nr_migrated)
@@ -90,7 +90,7 @@ TRACE_EVENT(mm_compaction_migratepages,
TP_fast_assign(
__entry->nr_migrated = nr_succeeded;
- __entry->nr_failed = cc->nr_migratepages - nr_succeeded;
+ __entry->nr_failed = nr_migratepages - nr_succeeded;
),
TP_printk("nr_migrated=%lu nr_failed=%lu",
--- a/mm/compaction.c~mm-compaction-update-the-cc-nr_migratepages-when-allocating-or-freeing-the-freepages
+++ a/mm/compaction.c
@@ -1798,6 +1798,7 @@ static struct folio *compaction_alloc(st
dst = list_entry(cc->freepages.next, struct folio, lru);
list_del(&dst->lru);
cc->nr_freepages--;
+ cc->nr_migratepages--;
return dst;
}
@@ -1813,6 +1814,7 @@ static void compaction_free(struct folio
list_add(&dst->lru, &cc->freepages);
cc->nr_freepages++;
+ cc->nr_migratepages++;
}
/* possible outcome of isolate_migratepages */
@@ -2435,7 +2437,7 @@ compact_zone(struct compact_control *cc,
unsigned long last_migrated_pfn;
const bool sync = cc->mode != MIGRATE_ASYNC;
bool update_cached;
- unsigned int nr_succeeded = 0;
+ unsigned int nr_succeeded = 0, nr_migratepages;
/*
* These counters track activities during zone compaction. Initialize
@@ -2553,11 +2555,17 @@ rescan:
pageblock_start_pfn(cc->migrate_pfn - 1));
}
+ /*
+ * Record the number of pages to migrate since the
+ * compaction_alloc/free() will update cc->nr_migratepages
+ * properly.
+ */
+ nr_migratepages = cc->nr_migratepages;
err = migrate_pages(&cc->migratepages, compaction_alloc,
compaction_free, (unsigned long)cc, cc->mode,
MR_COMPACTION, &nr_succeeded);
- trace_mm_compaction_migratepages(cc, nr_succeeded);
+ trace_mm_compaction_migratepages(nr_migratepages, nr_succeeded);
/* All pages were either migrated or will be released */
cc->nr_migratepages = 0;
_
Patches currently in -mm which might be from baolin.wang@xxxxxxxxxxxxxxxxx are
mm-compaction-limit-the-suitable-target-page-order-to-be-less-than-cc-order.patch
mm-compaction-update-the-cc-nr_migratepages-when-allocating-or-freeing-the-freepages.patch
mm-hugetlb-improve-the-handling-of-hugetlb-allocation-failure-for-freed-or-in-use-hugetlb.patch
