mm, compaction: fix 'limit' in fast_isolate_freepages
Subsystems (2)
PATCH 1/1 mm, compaction: fix 'limit' in fast_isolate_freepages
Because of 'min(1, ...)', fast_isolate_freepages set 'limit' to 0 or 1. This takes away the opportunities of find candinate pages. Also, even if 'limit' reaches zero, it scan once. It is not consistent. So, modify the minimum value of 'limit' to 1.
--- mm/compaction.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-)
mm/compaction.c +3 -3
@@ -1380,7 +1380,7 @@ static int next_search_order(struct compact_control *cc, int order)static unsigned long fast_isolate_freepages (struct compact_control *cc) { -unsigned int limit =min (1U ,freelist_scan_limit (cc) >>1 ); +unsigned int limit =max (1U ,freelist_scan_limit (cc) >>1 );unsigned int nr_scanned =0 ;unsigned long low_pfn, min_pfn, highest =0 ;unsigned long nr_isolated =0 ;
@@ -1456,7 +1456,7 @@ fast_isolate_freepages(struct compact_control *cc) high_pfn = pfn;/* Shorten the scan if a candidate is found */ - limit >>=1 ; + limit =max (1U , limit >>1 ); }if (order_scanned >= limit)
@@ -1496,7 +1496,7 @@ fast_isolate_freepages(struct compact_control *cc)* to freelist_scan_limit. */ if (order_scanned >= limit) - limit =min (1U , limit >>1 ); + limit =max (1U , limit >>1 ); }if (!page) {
Signature · 1 line
2.30.2
Quoted 5 lines · Wonhyuk Yang
Because of '
min(1, ...)',fast_isolate_freepagesset 'limit' to 0 or 1. This takes away the opportunities of find candinate pages. Also, even if 'limit' reaches zero, it scan once. It is not consistent. So, modify the minimum value of 'limit' to 1.
The changelog could do with a little polish.
In addition, what were the effects of this and what load did you use to evaluate it? While your patch is mostly correct, it has the potential side-effect of increasing system CPU usage in some cases and I'm curious to hear what you observed. Minimally it is worth noting in the changelog that there is a risk of increasing system CPU usage but that there are advantages too. Describe them in the changelog in case a regression bisects to your patch.
Quoted 12 lines
Fixes 5a811889de10f ("mm, compaction: use free lists to quickly locate a migration target")Signed-off-by Wonhyuk Yang <[email protected]>--- mm/compaction.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-)mm/compaction.c +0
-0
mm/compaction.c static int next_search_order(struct compact_control *cc, int order) · line 1386
Quoted 8 lines
mm/compaction.c +1
-1@@ -1380,7 +1380,7 @@ static int next_search_order(struct compact_control *cc, int order)static unsigned long fast_isolate_freepages (struct compact_control *cc) { -unsigned int limit =min (1U ,freelist_scan_limit (cc) >>1 ); +unsigned int limit =max (1U ,freelist_scan_limit (cc) >>1 );unsigned int nr_scanned =0 ;unsigned long low_pfn, min_pfn, highest =0 ;unsigned long nr_isolated =0 ;
Ok.
mm/compaction.c fast_isolate_freepages(struct compact_control *cc) · line 1462
Quoted 9 lines
mm/compaction.c +1
-1@@ -1456,7 +1456,7 @@ fast_isolate_freepages(struct compact_control *cc) high_pfn = pfn;/* Shorten the scan if a candidate is found */ - limit >>=1 ; + limit =max (1U , limit >>1 ); }if (order_scanned >= limit)
This hunk should be dropped. Once a candidate free page has been identified, it's ok to decay the limit to 0. This hunk introduces a risk of increasing system CPU usage unnecessarily.
mm/compaction.c fast_isolate_freepages(struct compact_control *cc) · line 1500
Quoted 7 lines
mm/compaction.c +1
-1@@ -1496,7 +1496,7 @@ fast_isolate_freepages(struct compact_control *cc)* to freelist_scan_limit. */ if (order_scanned >= limit) - limit =min (1U , limit >>1 ); + limit =max (1U , limit >>1 ); }
The change is fine but I have a minor nitpick that you are free to ignore. The comment above this block has a typo.
s/scan ig related/scan is related/
Ordinarily patches to fix spelling are ignored but you are altering this area anyway and it's helpful to see the full comment when reviewing this patch. I think it would be harmless to fix the spelling in the context of this patch.
Thanks.
Signature · 2 lines
Mel Gorman SUSE Labs
Quoted 10 lines · Mel Gorman
Quoted 5 lines · Wonhyuk Yang
Because of '
min(1, ...)',fast_isolate_freepagesset 'limit' to 0 or 1. This takes away the opportunities of find candinate pages. Also, even if 'limit' reaches zero, it scan once. It is not consistent. So, modify the minimum value of 'limit' to 1.The changelog could do with a little polish.
Quoted 7 lines
In addition, what were the effects of this and what load did you use to evaluate it? While your patch is mostly correct, it has the potential side-effect of increasing system CPU usage in some cases and I'm curious to hear what you observed. Minimally it is worth noting in the changelog that there is a risk of increasing system CPU usage but that there are advantages too. Describe them in the changelog in case a regression bisects to your patch.
I tested it on the thpscale and the results are as follows.
5.12
5.12
vanilla
patched
Amean fault-both-1 598.15 ( 0.00%) 592.56 ( 0.93%)
Amean fault-both-3 1494.47 ( 0.00%) 1514.35 ( -1.33%)
Amean fault-both-5 2519.48 ( 0.00%) 2471.76 ( 1.89%)
Amean fault-both-7 3173.85 ( 0.00%) 3079.19 ( 2.98%)
Amean fault-both-12 8063.83 ( 0.00%) 7858.29 ( 2.55%)
Amean fault-both-18 8781.20 ( 0.00%) 7827.70 * 10.86%*
Amean fault-both-24 12576.44 ( 0.00%) 12250.20 ( 2.59%)
Amean fault-both-30 18503.27 ( 0.00%) 17528.11 * 5.27%*
Amean fault-both-32 16133.69 ( 0.00%) 13874.24 * 14.00%*5.12 5.12
vanilla patched
Ops Compaction migrate scanned 6547133.00 5963901.00
Ops Compaction free scanned 32452453.00 26609101.00
One thing to worry about is that the results are very different every time. Is there any precise way to measure this patch?
Quoted 8 lines
Quoted 8 lines
mm/compaction.c +1
-1@@ -1456,7 +1456,7 @@ fast_isolate_freepages(struct compact_control *cc) high_pfn = pfn;/* Shorten the scan if a candidate is found */ - limit >>=1 ; + limit =max (1U , limit >>1 ); }
Quoted 5 lines
Review fast_isolate_freepages(struct compact_control *cc) · line 1462
Quoted 1 line
(quoted diff) +0
-0if (order_scanned >= limit)This hunk should be dropped. Once a candidate free page has been identified, it's ok to decay the limit to 0. This hunk introduces a risk of increasing system CPU usage unnecessarily.
Yes, you are right. I'll take your opinion.
Quoted 11 lines
mm/compaction.c fast_isolate_freepages(struct compact_control *cc) · line 1500
Quoted 7 lines
mm/compaction.c +1
-1@@ -1496,7 +1496,7 @@ fast_isolate_freepages(struct compact_control *cc)* to freelist_scan_limit. */ if (order_scanned >= limit) - limit =min (1U , limit >>1 ); + limit =max (1U , limit >>1 ); }The change is fine but I have a minor nitpick that you are free to ignore. The comment above this block has a typo.
Quoted 6 lines
s/scan ig related/scan is related/
Ordinarily patches to fix spelling are ignored but you are altering this area anyway and it's helpful to see the full comment when reviewing this patch. I think it would be harmless to fix the spelling in the context of this patch.
Okay, I'll fix this as well.
Thank you for your review.
Quoted 36 lines · Wonhyuk Yang
Quoted 10 lines · Mel Gorman
Quoted 5 lines · Wonhyuk Yang
Because of '
min(1, ...)',fast_isolate_freepagesset 'limit' to 0 or 1. This takes away the opportunities of find candinate pages. Also, even if 'limit' reaches zero, it scan once. It is not consistent. So, modify the minimum value of 'limit' to 1.The changelog could do with a little polish.
Quoted 7 lines
In addition, what were the effects of this and what load did you use to evaluate it? While your patch is mostly correct, it has the potential side-effect of increasing system CPU usage in some cases and I'm curious to hear what you observed. Minimally it is worth noting in the changelog that there is a risk of increasing system CPU usage but that there are advantages too. Describe them in the changelog in case a regression bisects to your patch.
I tested it on the thpscale and the results are as follows.
5.12 5.12 vanilla patched Amean fault-both-1 598.15 ( 0.00%) 592.56 ( 0.93%) Amean fault-both-3 1494.47 ( 0.00%) 1514.35 ( -1.33%) Amean fault-both-5 2519.48 ( 0.00%) 2471.76 ( 1.89%) Amean fault-both-7 3173.85 ( 0.00%) 3079.19 ( 2.98%) Amean fault-both-12 8063.83 ( 0.00%) 7858.29 ( 2.55%) Amean fault-both-18 8781.20 ( 0.00%) 7827.70 * 10.86%* Amean fault-both-24 12576.44 ( 0.00%) 12250.20 ( 2.59%) Amean fault-both-30 18503.27 ( 0.00%) 17528.11 * 5.27%* Amean fault-both-32 16133.69 ( 0.00%) 13874.24 * 14.00%*
Quoted 6 lines
5.12 5.12
vanilla patched
Ops Compaction migrate scanned 6547133.00 5963901.00
Ops Compaction free scanned 32452453.00 26609101.00
Ok, mention this in the changelog and maybe include the overall system CPU usage as well. It will be higher but should be acceptable.
Quoted 3 lines
One thing to worry about is that the results are very different every time. Is there any precise way to measure this patch?
Not with this workload, it was designed to simply hammer compaction heavily to see if latencies were unacceptably high and also for tracing various compaction corner cases.
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
mm/compaction.c +1
-1@@ -1456,7 +1456,7 @@ fast_isolate_freepages(struct compact_control *cc) high_pfn = pfn;/* Shorten the scan if a candidate is found */ - limit >>=1 ; + limit =max (1U , limit >>1 ); }
Quoted 8 lines
Quoted 5 lines
Review fast_isolate_freepages(struct compact_control *cc) · line 1462
Quoted 1 line
(quoted diff) +0
-0if (order_scanned >= limit)This hunk should be dropped. Once a candidate free page has been identified, it's ok to decay the limit to 0. This hunk introduces a risk of increasing system CPU usage unnecessarily.
Yes, you are right. I'll take your opinion.
Thanks.
Quoted 13 lines
Quoted 13 lines
mm/compaction.c fast_isolate_freepages(struct compact_control *cc) · line 1500
Quoted 7 lines
mm/compaction.c +1
-1@@ -1496,7 +1496,7 @@ fast_isolate_freepages(struct compact_control *cc)* to freelist_scan_limit. */ if (order_scanned >= limit) - limit =min (1U , limit >>1 ); + limit =max (1U , limit >>1 ); }The change is fine but I have a minor nitpick that you are free to ignore. The comment above this block has a typo.
s/scan ig related/scan is related/
Quoted 8 lines
Quoted 4 lines
Ordinarily patches to fix spelling are ignored but you are altering this area anyway and it's helpful to see the full comment when reviewing this patch. I think it would be harmless to fix the spelling in the context of this patch.
Okay, I'll fix this as well.
Thank you for your review.
No problem, thank you for the patch. Please cc me on v2 and I'll rerun some tests just to be sure before acking it.
Signature · 2 lines
Mel Gorman SUSE Labs
Quoted 92 lines · Mel Gorman
Quoted 36 lines · Wonhyuk Yang
Quoted 10 lines · Mel Gorman
Quoted 5 lines · Wonhyuk Yang
Because of '
min(1, ...)',fast_isolate_freepagesset 'limit' to 0 or 1. This takes away the opportunities of find candinate pages. Also, even if 'limit' reaches zero, it scan once. It is not consistent. So, modify the minimum value of 'limit' to 1.The changelog could do with a little polish.
Quoted 7 lines
In addition, what were the effects of this and what load did you use to evaluate it? While your patch is mostly correct, it has the potential side-effect of increasing system CPU usage in some cases and I'm curious to hear what you observed. Minimally it is worth noting in the changelog that there is a risk of increasing system CPU usage but that there are advantages too. Describe them in the changelog in case a regression bisects to your patch.
I tested it on the thpscale and the results are as follows.
5.12 5.12 vanilla patched Amean fault-both-1 598.15 ( 0.00%) 592.56 ( 0.93%) Amean fault-both-3 1494.47 ( 0.00%) 1514.35 ( -1.33%) Amean fault-both-5 2519.48 ( 0.00%) 2471.76 ( 1.89%) Amean fault-both-7 3173.85 ( 0.00%) 3079.19 ( 2.98%) Amean fault-both-12 8063.83 ( 0.00%) 7858.29 ( 2.55%) Amean fault-both-18 8781.20 ( 0.00%) 7827.70 * 10.86%* Amean fault-both-24 12576.44 ( 0.00%) 12250.20 ( 2.59%) Amean fault-both-30 18503.27 ( 0.00%) 17528.11 * 5.27%* Amean fault-both-32 16133.69 ( 0.00%) 13874.24 * 14.00%*Quoted 6 lines
5.12 5.12
vanilla patched
Ops Compaction migrate scanned 6547133.00 5963901.00
Ops Compaction free scanned 32452453.00 26609101.00Ok, mention this in the changelog and maybe include the overall system CPU usage as well. It will be higher but should be acceptable.
Quoted 3 lines
One thing to worry about is that the results are very different every time. Is there any precise way to measure this patch?
Not with this workload, it was designed to simply hammer compaction heavily to see if latencies were unacceptably high and also for tracing various compaction corner cases.
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
mm/compaction.c +1
-1@@ -1456,7 +1456,7 @@ fast_isolate_freepages(struct compact_control *cc) high_pfn = pfn;/* Shorten the scan if a candidate is found */ - limit >>=1 ; + limit =max (1U , limit >>1 ); }Quoted 8 lines
Quoted 5 lines
Review fast_isolate_freepages(struct compact_control *cc) · line 1462
Quoted 1 line
(quoted diff) +0
-0if (order_scanned >= limit)This hunk should be dropped. Once a candidate free page has been identified, it's ok to decay the limit to 0. This hunk introduces a risk of increasing system CPU usage unnecessarily.
Yes, you are right. I'll take your opinion.
Thanks.
Quoted 18 lines
Quoted 17 lines
mm/compaction.c fast_isolate_freepages(struct compact_control *cc) · line 1500
Quoted 7 lines
mm/compaction.c +1
-1@@ -1496,7 +1496,7 @@ fast_isolate_freepages(struct compact_control *cc)* to freelist_scan_limit. */ if (order_scanned >= limit) - limit =min (1U , limit >>1 ); + limit =max (1U , limit >>1 ); }The change is fine but I have a minor nitpick that you are free to ignore. The comment above this block has a typo.
s/scan ig related/scan is related/
Ordinarily patches to fix spelling are ignored but you are altering this area anyway and it's helpful to see the full comment when reviewing this patch. I think it would be harmless to fix the spelling in the context of this patch.
Quoted 7 lines
Quoted 3 lines
Okay, I'll fix this as well.
Thank you for your review.
No problem, thank you for the patch. Please cc me on v2 and I'll rerun some tests just to be sure before acking it.
Okay, I'll do that.