From: Chen Yucong <slaoub@gmail.com>
To: akpm@linux-foundation.org
Cc: minchan@kernel.org, mgorman@suse.de, hannes@cmpxchg.org,
mhocko@suse.cz, riel@redhat.com, linux-mm@kvack.org,
linux-kernel@vger.kernel.org, Chen Yucong <slaoub@gmail.com>
Subject: [PATCH] mm/vmscan.c: fix an implementation flaw in proportional scanning
Date: Tue, 17 Jun 2014 12:55:02 +0800 [thread overview]
Message-ID: <1402980902-6345-1-git-send-email-slaoub@gmail.com> (raw)
Via https://lkml.org/lkml/2013/4/10/897, we can know that the relative design
idea is to keep
scan_target[anon] : scan_target[file]
== really_scanned_num[anon] : really_scanned_num[file]
But we can find the following snippet in shrink_lruvec():
if (nr_file > nr_anon) {
...
} else {
...
}
However, the above code fragment broke the design idea. We can assume:
nr[LRU_ACTIVE_FILE] = 30
nr[LRU_INACTIVE_FILE] = 30
nr[LRU_ACTIVE_ANON] = 0
nr[LRU_INACTIVE_ANON] = 40
When the value of (nr_reclaimed < nr_to_reclaim) become false, there are
the following results:
nr[LRU_ACTIVE_FILE] = 15
nr[LRU_INACTIVE_FILE] = 15
nr[LRU_ACTIVE_ANON] = 0
nr[LRU_INACTIVE_ANON] = 25
nr_file = 30
nr_anon = 25
file_percent = 30 / 60 = 0.5
anon_percent = 25 / 40 = 0.65
According to the above design idea, we should scan some pages from ANON,
but in fact we execute the an error code path due to "if (nr_file > nr_anon)".
In this way, nr[lru] is likely to be a negative number. Luckily,
"nr[lru] -= min(nr[lru], nr_scanned)" can help us to filter this situation,
but it has rebelled against our design idea.
Signed-off-by: Chen Yucong <slaoub@gmail.com>
---
mm/vmscan.c | 39 ++++++++++++++++++---------------------
1 file changed, 18 insertions(+), 21 deletions(-)
diff --git a/mm/vmscan.c b/mm/vmscan.c
index a8ffe4e..2c35e34 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -2057,8 +2057,7 @@ out:
static void shrink_lruvec(struct lruvec *lruvec, struct scan_control *sc)
{
unsigned long nr[NR_LRU_LISTS];
- unsigned long targets[NR_LRU_LISTS];
- unsigned long nr_to_scan;
+ unsigned long file_target, anon_target;
enum lru_list lru;
unsigned long nr_reclaimed = 0;
unsigned long nr_to_reclaim = sc->nr_to_reclaim;
@@ -2067,8 +2066,8 @@ static void shrink_lruvec(struct lruvec *lruvec, struct scan_control *sc)
get_scan_count(lruvec, sc, nr);
- /* Record the original scan target for proportional adjustments later */
- memcpy(targets, nr, sizeof(nr));
+ file_target = nr[LRU_INACTIVE_FILE] + nr[LRU_ACTIVE_FILE];
+ anon_target = nr[LRU_INACTIVE_ANON] + nr[LRU_ACTIVE_ANON];
/*
* Global reclaiming within direct reclaim at DEF_PRIORITY is a normal
@@ -2087,8 +2086,8 @@ static void shrink_lruvec(struct lruvec *lruvec, struct scan_control *sc)
blk_start_plug(&plug);
while (nr[LRU_INACTIVE_ANON] || nr[LRU_ACTIVE_FILE] ||
nr[LRU_INACTIVE_FILE]) {
- unsigned long nr_anon, nr_file, percentage;
- unsigned long nr_scanned;
+ unsigned long nr_anon, nr_file, file_percent, anon_percent;
+ unsigned long nr_to_scan, nr_scanned, percentage;
for_each_evictable_lru(lru) {
if (nr[lru]) {
@@ -2122,16 +2121,19 @@ static void shrink_lruvec(struct lruvec *lruvec, struct scan_control *sc)
if (!nr_file || !nr_anon)
break;
- if (nr_file > nr_anon) {
- unsigned long scan_target = targets[LRU_INACTIVE_ANON] +
- targets[LRU_ACTIVE_ANON] + 1;
+ file_percent = nr_file * 100 / file_target;
+ anon_percent = nr_anon * 100 / anon_target;
+
+ if (file_percent > anon_percent) {
lru = LRU_BASE;
- percentage = nr_anon * 100 / scan_target;
+ nr_scanned = file_target - nr_file;
+ nr_to_scan = file_target * (100 - anon_percent) / 100;
+ percentage = nr[LRU_FILE] * 100 / nr_file;
} else {
- unsigned long scan_target = targets[LRU_INACTIVE_FILE] +
- targets[LRU_ACTIVE_FILE] + 1;
lru = LRU_FILE;
- percentage = nr_file * 100 / scan_target;
+ nr_scanned = anon_target - nr_anon;
+ nr_to_scan = anon_target * (100 - file_percent) / 100;
+ percentage = nr[LRU_BASE] * 100 / nr_anon;
}
/* Stop scanning the smaller of the LRU */
@@ -2143,14 +2145,9 @@ static void shrink_lruvec(struct lruvec *lruvec, struct scan_control *sc)
* scan target and the percentage scanning already complete
*/
lru = (lru == LRU_FILE) ? LRU_BASE : LRU_FILE;
- nr_scanned = targets[lru] - nr[lru];
- nr[lru] = targets[lru] * (100 - percentage) / 100;
- nr[lru] -= min(nr[lru], nr_scanned);
-
- lru += LRU_ACTIVE;
- nr_scanned = targets[lru] - nr[lru];
- nr[lru] = targets[lru] * (100 - percentage) / 100;
- nr[lru] -= min(nr[lru], nr_scanned);
+ nr_to_scan -= min(nr_to_scan, nr_scanned);
+ nr[lru] = nr_to_scan * percentage / 100;
+ nr[lru + LRU_ACTIVE] = nr_to_scan - nr[lru];
scan_adjusted = true;
}
--
1.7.10.4
--
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>
WARNING: multiple messages have this Message-ID (diff)
From: Chen Yucong <slaoub@gmail.com>
To: akpm@linux-foundation.org
Cc: minchan@kernel.org, mgorman@suse.de, hannes@cmpxchg.org,
mhocko@suse.cz, riel@redhat.com, linux-mm@kvack.org,
linux-kernel@vger.kernel.org, Chen Yucong <slaoub@gmail.com>
Subject: [PATCH] mm/vmscan.c: fix an implementation flaw in proportional scanning
Date: Tue, 17 Jun 2014 12:55:02 +0800 [thread overview]
Message-ID: <1402980902-6345-1-git-send-email-slaoub@gmail.com> (raw)
Via https://lkml.org/lkml/2013/4/10/897, we can know that the relative design
idea is to keep
scan_target[anon] : scan_target[file]
== really_scanned_num[anon] : really_scanned_num[file]
But we can find the following snippet in shrink_lruvec():
if (nr_file > nr_anon) {
...
} else {
...
}
However, the above code fragment broke the design idea. We can assume:
nr[LRU_ACTIVE_FILE] = 30
nr[LRU_INACTIVE_FILE] = 30
nr[LRU_ACTIVE_ANON] = 0
nr[LRU_INACTIVE_ANON] = 40
When the value of (nr_reclaimed < nr_to_reclaim) become false, there are
the following results:
nr[LRU_ACTIVE_FILE] = 15
nr[LRU_INACTIVE_FILE] = 15
nr[LRU_ACTIVE_ANON] = 0
nr[LRU_INACTIVE_ANON] = 25
nr_file = 30
nr_anon = 25
file_percent = 30 / 60 = 0.5
anon_percent = 25 / 40 = 0.65
According to the above design idea, we should scan some pages from ANON,
but in fact we execute the an error code path due to "if (nr_file > nr_anon)".
In this way, nr[lru] is likely to be a negative number. Luckily,
"nr[lru] -= min(nr[lru], nr_scanned)" can help us to filter this situation,
but it has rebelled against our design idea.
Signed-off-by: Chen Yucong <slaoub@gmail.com>
---
mm/vmscan.c | 39 ++++++++++++++++++---------------------
1 file changed, 18 insertions(+), 21 deletions(-)
diff --git a/mm/vmscan.c b/mm/vmscan.c
index a8ffe4e..2c35e34 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -2057,8 +2057,7 @@ out:
static void shrink_lruvec(struct lruvec *lruvec, struct scan_control *sc)
{
unsigned long nr[NR_LRU_LISTS];
- unsigned long targets[NR_LRU_LISTS];
- unsigned long nr_to_scan;
+ unsigned long file_target, anon_target;
enum lru_list lru;
unsigned long nr_reclaimed = 0;
unsigned long nr_to_reclaim = sc->nr_to_reclaim;
@@ -2067,8 +2066,8 @@ static void shrink_lruvec(struct lruvec *lruvec, struct scan_control *sc)
get_scan_count(lruvec, sc, nr);
- /* Record the original scan target for proportional adjustments later */
- memcpy(targets, nr, sizeof(nr));
+ file_target = nr[LRU_INACTIVE_FILE] + nr[LRU_ACTIVE_FILE];
+ anon_target = nr[LRU_INACTIVE_ANON] + nr[LRU_ACTIVE_ANON];
/*
* Global reclaiming within direct reclaim at DEF_PRIORITY is a normal
@@ -2087,8 +2086,8 @@ static void shrink_lruvec(struct lruvec *lruvec, struct scan_control *sc)
blk_start_plug(&plug);
while (nr[LRU_INACTIVE_ANON] || nr[LRU_ACTIVE_FILE] ||
nr[LRU_INACTIVE_FILE]) {
- unsigned long nr_anon, nr_file, percentage;
- unsigned long nr_scanned;
+ unsigned long nr_anon, nr_file, file_percent, anon_percent;
+ unsigned long nr_to_scan, nr_scanned, percentage;
for_each_evictable_lru(lru) {
if (nr[lru]) {
@@ -2122,16 +2121,19 @@ static void shrink_lruvec(struct lruvec *lruvec, struct scan_control *sc)
if (!nr_file || !nr_anon)
break;
- if (nr_file > nr_anon) {
- unsigned long scan_target = targets[LRU_INACTIVE_ANON] +
- targets[LRU_ACTIVE_ANON] + 1;
+ file_percent = nr_file * 100 / file_target;
+ anon_percent = nr_anon * 100 / anon_target;
+
+ if (file_percent > anon_percent) {
lru = LRU_BASE;
- percentage = nr_anon * 100 / scan_target;
+ nr_scanned = file_target - nr_file;
+ nr_to_scan = file_target * (100 - anon_percent) / 100;
+ percentage = nr[LRU_FILE] * 100 / nr_file;
} else {
- unsigned long scan_target = targets[LRU_INACTIVE_FILE] +
- targets[LRU_ACTIVE_FILE] + 1;
lru = LRU_FILE;
- percentage = nr_file * 100 / scan_target;
+ nr_scanned = anon_target - nr_anon;
+ nr_to_scan = anon_target * (100 - file_percent) / 100;
+ percentage = nr[LRU_BASE] * 100 / nr_anon;
}
/* Stop scanning the smaller of the LRU */
@@ -2143,14 +2145,9 @@ static void shrink_lruvec(struct lruvec *lruvec, struct scan_control *sc)
* scan target and the percentage scanning already complete
*/
lru = (lru == LRU_FILE) ? LRU_BASE : LRU_FILE;
- nr_scanned = targets[lru] - nr[lru];
- nr[lru] = targets[lru] * (100 - percentage) / 100;
- nr[lru] -= min(nr[lru], nr_scanned);
-
- lru += LRU_ACTIVE;
- nr_scanned = targets[lru] - nr[lru];
- nr[lru] = targets[lru] * (100 - percentage) / 100;
- nr[lru] -= min(nr[lru], nr_scanned);
+ nr_to_scan -= min(nr_to_scan, nr_scanned);
+ nr[lru] = nr_to_scan * percentage / 100;
+ nr[lru + LRU_ACTIVE] = nr_to_scan - nr[lru];
scan_adjusted = true;
}
--
1.7.10.4
next reply other threads:[~2014-06-17 4:58 UTC|newest]
Thread overview: 22+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-06-17 4:55 Chen Yucong [this message]
2014-06-17 4:55 ` [PATCH] mm/vmscan.c: fix an implementation flaw in proportional scanning Chen Yucong
2014-06-18 9:00 ` Jerome Marchand
2014-06-18 9:00 ` Jerome Marchand
2014-06-18 9:08 ` Chen Yucong
2014-06-18 9:08 ` Chen Yucong
2014-06-18 9:54 ` Jerome Marchand
2014-06-18 9:54 ` Jerome Marchand
2014-06-18 22:27 ` Andrew Morton
2014-06-18 22:27 ` Andrew Morton
2014-06-19 0:04 ` Chen Yucong
2014-06-19 0:04 ` Chen Yucong
2014-06-19 0:40 ` Andrew Morton
2014-06-19 0:40 ` Andrew Morton
2014-06-19 1:02 ` Minchan Kim
2014-06-19 1:02 ` Minchan Kim
2014-06-19 20:13 ` Andrew Morton
2014-06-19 20:13 ` Andrew Morton
2014-06-23 10:00 ` Mel Gorman
2014-06-23 10:00 ` Mel Gorman
2014-06-23 11:02 ` Mel Gorman
2014-06-23 11:02 ` Mel Gorman
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=1402980902-6345-1-git-send-email-slaoub@gmail.com \
--to=slaoub@gmail.com \
--cc=akpm@linux-foundation.org \
--cc=hannes@cmpxchg.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mgorman@suse.de \
--cc=mhocko@suse.cz \
--cc=minchan@kernel.org \
--cc=riel@redhat.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 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.