* + mm-migrate-correct-nr_failed-in-migrate_pages_sync.patch added to mm-unstable branch
@ 2023-10-09 23:30 Andrew Morton
0 siblings, 0 replies; 3+ messages in thread
From: Andrew Morton @ 2023-10-09 23:30 UTC (permalink / raw)
To: mm-commits, ying.huang, willy, david, baolin.wang, ziy, akpm
The patch titled
Subject: mm/migrate: correct nr_failed in migrate_pages_sync()
has been added to the -mm mm-unstable branch. Its filename is
mm-migrate-correct-nr_failed-in-migrate_pages_sync.patch
This patch will shortly appear at
https://git.kernel.org/pub/scm/linux/kernel/git/akpm/25-new.git/tree/patches/mm-migrate-correct-nr_failed-in-migrate_pages_sync.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: Zi Yan <ziy@nvidia.com>
Subject: mm/migrate: correct nr_failed in migrate_pages_sync()
Date: Mon, 9 Oct 2023 16:32:31 -0400
In the upstream tree, large folios can be migrated via move_pages()
syscall and there is no VM_BUG_ON in do_move_pages_to_node() but pages
left in the source page list will not be returned to LRU lists. This is
from my code inspection and I have not reproduced it on upstream tree yet.
nr_failed was missing the rc value from migrate_pages_batch() and can
cause a mismatch between migrate_pages() return value and the number of
not migrated pages, i.e., when the return value of migrate_pages() is 0,
there are still pages left in the from page list. It will happen when a
non-PMD THP large folio fails to migrate due to -ENOMEM and is split
successfully but not all the split pages are not migrated,
migrate_pages_batch() would return non-zero, but astats.nr_thp_split =3D
0. nr_failed would be 0 and returned to the caller of migrate_pages(),
but the not migrated pages are left in the from page list without being
added back to LRU lists.
Link: https://lkml.kernel.org/r/20231009203231.1715845-2-zi.yan@sent.com
Fixes: 2ef7dbb26990 ("migrate_pages: try migrate in batch asynchronously fi=
rstly")
Signed-off-by: Zi Yan <ziy@nvidia.com>
Cc: Baolin Wang <baolin.wang@linux.alibaba.com>
Cc: David Hildenbrand <david@redhat.com>
Cc: Huang Ying <ying.huang@intel.com>
Cc: Matthew Wilcox (Oracle) <willy@infradead.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
---
mm/migrate.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
--- a/mm/migrate.c~mm-migrate-correct-nr_failed-in-migrate_pages_sync
+++ a/mm/migrate.c
@@ -1834,7 +1834,7 @@ static int migrate_pages_sync(struct lis
return rc;
}
stats->nr_thp_failed += astats.nr_thp_split;
- nr_failed += astats.nr_thp_split;
+ nr_failed += rc + astats.nr_thp_split;
/*
* Fall back to migrate all failed folios one by one synchronously. All
* failed folios except split THPs will be retried, so their failure
_
Patches currently in -mm which might be from ziy@nvidia.com are
mm-migrate-correct-nr_failed-in-migrate_pages_sync.patch
^ permalink raw reply [flat|nested] 3+ messages in thread
* + mm-migrate-correct-nr_failed-in-migrate_pages_sync.patch added to mm-unstable branch
@ 2023-10-16 22:26 Andrew Morton
0 siblings, 0 replies; 3+ messages in thread
From: Andrew Morton @ 2023-10-16 22:26 UTC (permalink / raw)
To: mm-commits, ying.huang, willy, david, baolin.wang, ziy, akpm
The patch titled
Subject: mm/migrate: correct nr_failed in migrate_pages_sync()
has been added to the -mm mm-unstable branch. Its filename is
mm-migrate-correct-nr_failed-in-migrate_pages_sync.patch
This patch will shortly appear at
https://git.kernel.org/pub/scm/linux/kernel/git/akpm/25-new.git/tree/patches/mm-migrate-correct-nr_failed-in-migrate_pages_sync.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: Zi Yan <ziy@nvidia.com>
Subject: mm/migrate: correct nr_failed in migrate_pages_sync()
Date: Mon, 16 Oct 2023 11:41:56 -0400
nr_failed was missing the large folio splits from migrate_pages_batch()
and can cause a mismatch between migrate_pages() return value and the
number of not migrated pages, i.e., when the return value of
migrate_pages() is 0, there are still pages left in the from page list.
It will happen when a non-PMD THP large folio fails to migrate due to
-ENOMEM and is split successfully but not all the split pages are not
migrated, migrate_pages_batch() would return non-zero, but
astats.nr_thp_split =3D 0. nr_failed would be 0 and returned to the
caller of migrate_pages(), but the not migrated pages are left in the from
page list without being added back to LRU lists.
Fix it by adding a new nr_split counter for large folio splits and adding
it to nr_failed in migrate_page_sync() after migrate_pages_batch() is
done.
Link: https://lkml.kernel.org/r/20231016154156.1948815-1-zi.yan@sent.com
Fixes: 2ef7dbb26990 ("migrate_pages: try migrate in batch asynchronously fi=
rstly")
Signed-off-by: Zi Yan <ziy@nvidia.com>
Cc: Baolin Wang <baolin.wang@linux.alibaba.com>
Cc: David Hildenbrand <david@redhat.com>
Cc: Huang Ying <ying.huang@intel.com>
Cc: Matthew Wilcox <willy@infradead.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
---
include/trace/events/migrate.h | 24 ++++++++++++++----------
mm/migrate.c | 15 +++++++++++----
2 files changed, 25 insertions(+), 14 deletions(-)
--- a/include/trace/events/migrate.h~mm-migrate-correct-nr_failed-in-migrate_pages_sync
+++ a/include/trace/events/migrate.h
@@ -49,10 +49,11 @@ TRACE_EVENT(mm_migrate_pages,
TP_PROTO(unsigned long succeeded, unsigned long failed,
unsigned long thp_succeeded, unsigned long thp_failed,
- unsigned long thp_split, enum migrate_mode mode, int reason),
+ unsigned long thp_split, unsigned long large_folio_split,
+ enum migrate_mode mode, int reason),
TP_ARGS(succeeded, failed, thp_succeeded, thp_failed,
- thp_split, mode, reason),
+ thp_split, large_folio_split, mode, reason),
TP_STRUCT__entry(
__field( unsigned long, succeeded)
@@ -60,26 +61,29 @@ TRACE_EVENT(mm_migrate_pages,
__field( unsigned long, thp_succeeded)
__field( unsigned long, thp_failed)
__field( unsigned long, thp_split)
+ __field( unsigned long, large_folio_split)
__field( enum migrate_mode, mode)
__field( int, reason)
),
TP_fast_assign(
- __entry->succeeded = succeeded;
- __entry->failed = failed;
- __entry->thp_succeeded = thp_succeeded;
- __entry->thp_failed = thp_failed;
- __entry->thp_split = thp_split;
- __entry->mode = mode;
- __entry->reason = reason;
+ __entry->succeeded = succeeded;
+ __entry->failed = failed;
+ __entry->thp_succeeded = thp_succeeded;
+ __entry->thp_failed = thp_failed;
+ __entry->thp_split = thp_split;
+ __entry->large_folio_split = large_folio_split;
+ __entry->mode = mode;
+ __entry->reason = reason;
),
- TP_printk("nr_succeeded=%lu nr_failed=%lu nr_thp_succeeded=%lu nr_thp_failed=%lu nr_thp_split=%lu mode=%s reason=%s",
+ TP_printk("nr_succeeded=%lu nr_failed=%lu nr_thp_succeeded=%lu nr_thp_failed=%lu nr_thp_split=%lu nr_split=%lu mode=%s reason=%s",
__entry->succeeded,
__entry->failed,
__entry->thp_succeeded,
__entry->thp_failed,
__entry->thp_split,
+ __entry->large_folio_split,
__print_symbolic(__entry->mode, MIGRATE_MODE),
__print_symbolic(__entry->reason, MIGRATE_REASON))
);
--- a/mm/migrate.c~mm-migrate-correct-nr_failed-in-migrate_pages_sync
+++ a/mm/migrate.c
@@ -1495,6 +1495,7 @@ struct migrate_pages_stats {
int nr_thp_succeeded; /* THP migrated successfully */
int nr_thp_failed; /* THP failed to be migrated */
int nr_thp_split; /* THP split before migrating */
+ int nr_split; /* Large folio (include THP) split before migrating */
};
/*
@@ -1614,6 +1615,7 @@ static int migrate_pages_batch(struct li
int nr_retry_pages = 0;
int pass = 0;
bool is_thp = false;
+ bool is_large = false;
struct folio *folio, *folio2, *dst = NULL, *dst2;
int rc, rc_saved = 0, nr_pages;
LIST_HEAD(unmap_folios);
@@ -1629,7 +1631,8 @@ static int migrate_pages_batch(struct li
nr_retry_pages = 0;
list_for_each_entry_safe(folio, folio2, from, lru) {
- is_thp = folio_test_large(folio) && folio_test_pmd_mappable(folio);
+ is_large = folio_test_large(folio);
+ is_thp = is_large && folio_test_pmd_mappable(folio);
nr_pages = folio_nr_pages(folio);
cond_resched();
@@ -1649,6 +1652,7 @@ static int migrate_pages_batch(struct li
stats->nr_thp_failed++;
if (!try_split_folio(folio, split_folios)) {
stats->nr_thp_split++;
+ stats->nr_split++;
continue;
}
stats->nr_failed_pages += nr_pages;
@@ -1677,11 +1681,12 @@ static int migrate_pages_batch(struct li
nr_failed++;
stats->nr_thp_failed += is_thp;
/* Large folio NUMA faulting doesn't split to retry. */
- if (folio_test_large(folio) && !nosplit) {
+ if (is_large && !nosplit) {
int ret = try_split_folio(folio, split_folios);
if (!ret) {
stats->nr_thp_split += is_thp;
+ stats->nr_split++;
break;
} else if (reason == MR_LONGTERM_PIN &&
ret == -EAGAIN) {
@@ -1827,6 +1832,7 @@ static int migrate_pages_sync(struct lis
stats->nr_succeeded += astats.nr_succeeded;
stats->nr_thp_succeeded += astats.nr_thp_succeeded;
stats->nr_thp_split += astats.nr_thp_split;
+ stats->nr_split += astats.nr_split;
if (rc < 0) {
stats->nr_failed_pages += astats.nr_failed_pages;
stats->nr_thp_failed += astats.nr_thp_failed;
@@ -1834,7 +1840,7 @@ static int migrate_pages_sync(struct lis
return rc;
}
stats->nr_thp_failed += astats.nr_thp_split;
- nr_failed += astats.nr_thp_split;
+ nr_failed += astats.nr_split + astats.nr_thp_split;
/*
* Fall back to migrate all failed folios one by one synchronously. All
* failed folios except split THPs will be retried, so their failure
@@ -1969,7 +1975,8 @@ out:
count_vm_events(THP_MIGRATION_SPLIT, stats.nr_thp_split);
trace_mm_migrate_pages(stats.nr_succeeded, stats.nr_failed_pages,
stats.nr_thp_succeeded, stats.nr_thp_failed,
- stats.nr_thp_split, mode, reason);
+ stats.nr_thp_split, stats.nr_split, mode,
+ reason);
if (ret_succeeded)
*ret_succeeded = stats.nr_succeeded;
_
Patches currently in -mm which might be from ziy@nvidia.com are
mm-migrate-correct-nr_failed-in-migrate_pages_sync.patch
^ permalink raw reply [flat|nested] 3+ messages in thread
* + mm-migrate-correct-nr_failed-in-migrate_pages_sync.patch added to mm-unstable branch
@ 2023-10-18 16:18 Andrew Morton
0 siblings, 0 replies; 3+ messages in thread
From: Andrew Morton @ 2023-10-18 16:18 UTC (permalink / raw)
To: mm-commits, ying.huang, willy, david, baolin.wang, ziy, akpm
The patch titled
Subject: mm/migrate: correct nr_failed in migrate_pages_sync()
has been added to the -mm mm-unstable branch. Its filename is
mm-migrate-correct-nr_failed-in-migrate_pages_sync.patch
This patch will shortly appear at
https://git.kernel.org/pub/scm/linux/kernel/git/akpm/25-new.git/tree/patches/mm-migrate-correct-nr_failed-in-migrate_pages_sync.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: Zi Yan <ziy@nvidia.com>
Subject: mm/migrate: correct nr_failed in migrate_pages_sync()
Date: Tue, 17 Oct 2023 12:31:28 -0400
nr_failed was missing the large folio splits from migrate_pages_batch()
and can cause a mismatch between migrate_pages() return value and the
number of not migrated pages, i.e., when the return value of
migrate_pages() is 0, there are still pages left in the from page list.
It will happen when a non-PMD THP large folio fails to migrate due to
-ENOMEM and is split successfully but not all the split pages are not
migrated, migrate_pages_batch() would return non-zero, but
astats.nr_thp_split = 0. nr_failed would be 0 and returned to the caller
of migrate_pages(), but the not migrated pages are left in the from page
list without being added back to LRU lists.
Fix it by adding a new nr_split counter for large folio splits and adding
it to nr_failed in migrate_page_sync() after migrate_pages_batch() is
done.
Link: https://lkml.kernel.org/r/20231017163129.2025214-1-zi.yan@sent.com
Fixes: 2ef7dbb26990 ("migrate_pages: try migrate in batch asynchronously firstly")
Signed-off-by: Zi Yan <ziy@nvidia.com>
Acked-by: Huang Ying <ying.huang@intel.com>
Cc: Baolin Wang <baolin.wang@linux.alibaba.com>
Cc: David Hildenbrand <david@redhat.com>
Cc: Matthew Wilcox <willy@infradead.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
---
mm/migrate.c | 16 +++++++++++++---
1 file changed, 13 insertions(+), 3 deletions(-)
--- a/mm/migrate.c~mm-migrate-correct-nr_failed-in-migrate_pages_sync
+++ a/mm/migrate.c
@@ -1495,6 +1495,7 @@ struct migrate_pages_stats {
int nr_thp_succeeded; /* THP migrated successfully */
int nr_thp_failed; /* THP failed to be migrated */
int nr_thp_split; /* THP split before migrating */
+ int nr_split; /* Large folio (include THP) split before migrating */
};
/*
@@ -1614,6 +1615,7 @@ static int migrate_pages_batch(struct li
int nr_retry_pages = 0;
int pass = 0;
bool is_thp = false;
+ bool is_large = false;
struct folio *folio, *folio2, *dst = NULL, *dst2;
int rc, rc_saved = 0, nr_pages;
LIST_HEAD(unmap_folios);
@@ -1629,7 +1631,8 @@ static int migrate_pages_batch(struct li
nr_retry_pages = 0;
list_for_each_entry_safe(folio, folio2, from, lru) {
- is_thp = folio_test_large(folio) && folio_test_pmd_mappable(folio);
+ is_large = folio_test_large(folio);
+ is_thp = is_large && folio_test_pmd_mappable(folio);
nr_pages = folio_nr_pages(folio);
cond_resched();
@@ -1649,6 +1652,7 @@ static int migrate_pages_batch(struct li
stats->nr_thp_failed++;
if (!try_split_folio(folio, split_folios)) {
stats->nr_thp_split++;
+ stats->nr_split++;
continue;
}
stats->nr_failed_pages += nr_pages;
@@ -1677,11 +1681,12 @@ static int migrate_pages_batch(struct li
nr_failed++;
stats->nr_thp_failed += is_thp;
/* Large folio NUMA faulting doesn't split to retry. */
- if (folio_test_large(folio) && !nosplit) {
+ if (is_large && !nosplit) {
int ret = try_split_folio(folio, split_folios);
if (!ret) {
stats->nr_thp_split += is_thp;
+ stats->nr_split += is_large;
break;
} else if (reason == MR_LONGTERM_PIN &&
ret == -EAGAIN) {
@@ -1827,6 +1832,7 @@ static int migrate_pages_sync(struct lis
stats->nr_succeeded += astats.nr_succeeded;
stats->nr_thp_succeeded += astats.nr_thp_succeeded;
stats->nr_thp_split += astats.nr_thp_split;
+ stats->nr_split += astats.nr_split;
if (rc < 0) {
stats->nr_failed_pages += astats.nr_failed_pages;
stats->nr_thp_failed += astats.nr_thp_failed;
@@ -1834,7 +1840,11 @@ static int migrate_pages_sync(struct lis
return rc;
}
stats->nr_thp_failed += astats.nr_thp_split;
- nr_failed += astats.nr_thp_split;
+ /*
+ * Do not count rc, as pages will be retried below.
+ * Count nr_split only, since it includes nr_thp_split.
+ */
+ nr_failed += astats.nr_split;
/*
* Fall back to migrate all failed folios one by one synchronously. All
* failed folios except split THPs will be retried, so their failure
_
Patches currently in -mm which might be from ziy@nvidia.com are
mm-migrate-correct-nr_failed-in-migrate_pages_sync.patch
mm-migrate-add-nr_split-to-trace_mm_migrate_pages-stats.patch
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2023-10-18 16:18 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2023-10-09 23:30 + mm-migrate-correct-nr_failed-in-migrate_pages_sync.patch added to mm-unstable branch Andrew Morton
-- strict thread matches above, loose matches on Subject: below --
2023-10-16 22:26 Andrew Morton
2023-10-18 16:18 Andrew Morton
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.