mm/page_alloc: keep track of free highatomic
Subsystems (4)
PATCH v3 1/1 mm/page_alloc: keep track of free highatomic
OOM kills due to vastly overestimated free highatomic reserves were observed:
... invoked oom-killer: gfp_mask=0x100cca(GFP_HIGHUSER_MOVABLE), order=0 ...
Node 0 Normal free:1482936kB boost:0kB min:410416kB low:739404kB high:1068392kB reserved_highatomic:1073152KB ...
Node 0 Normal: 1292*4kB (ME) 1920*8kB (E) 383*16kB (UE) 220*32kB (ME) 340*64kB (E) 2155*128kB (UE) 3243*256kB (UE) 615*512kB (U) 1*1024kB (M) 0*2048kB 0*4096kB = 1477408kB
The second line above shows that the OOM kill was due to the following condition:
free (1482936kB) - reserved_highatomic (1073152kB) = 409784KB < min (410416kB)
And the third line shows there were no free pages in any MIGRATE_HIGHATOMIC pageblocks, which otherwise would show up as type 'H'. Therefore __zone_watermark_unusable_free() underestimated the usable free memory by over 1GB, which resulted in the unnecessary OOM kill above.
The comments in __zone_watermark_unusable_free() warns about the potential risk, i.e.,
If the caller does not have rights to reserves below the min watermark then subtract the high-atomic reserves. This will over-estimate the size of the atomic reserve but it avoids a search.
However, it is possible to keep track of free pages in reserved highatomic pageblocks with a new per-zone counter nr_free_highatomic protected by the zone lock, to avoid a search when calculating the usable free memory. And the cost would be minimal, i.e., simple arithmetics in the highatomic alloc/free/move paths.
Note that since nr_free_highatomic can be relatively small, using a per-cpu counter might cause too much drift and defeat its purpose, in addition to the extra memory overhead.
--- include/linux/mmzone.h | 1 + mm/page_alloc.c | 10 +++++++--- 2 files changed, 8 insertions(+), 3 deletions(-)
include/linux/mmzone.h +1 -0
@@ -825,6 +825,7 @@ struct zone {unsigned long watermark_boost;unsigned long nr_reserved_highatomic; +unsigned long nr_free_highatomic;/* * We don't know if the memory that we're going to allocate will be
mm/page_alloc.c +7 -3
@@ -635,6 +635,8 @@ compaction_capture(struct capture_control *capc, struct page *page,static inline void account_freepages (struct zone *zone,int nr_pages,int migratetype) { +lockdep_assert_held (&zone->lock); +if (is_migrate_isolate (migratetype))return ;
@@ -642,6 +644,9 @@ static inline void account_freepages(struct zone *zone, int nr_pages,if (is_migrate_cma (migratetype))__mod_zone_page_state (zone,NR_FREE_CMA_PAGES , nr_pages); + +if (is_migrate_highatomic (migratetype)) +WRITE_ONCE (zone->nr_free_highatomic, zone->nr_free_highatomic + nr_pages); }/* Used for pages not on another list */
@@ -3117,11 +3122,10 @@ static inline long __zone_watermark_unusable_free(struct zone *z,/* * If the caller does not have rights to reserves below the min -* watermark then subtract the high-atomic reserves. This will -* over-estimate the size of the atomic reserve but it avoids a search. +* watermark then subtract the free pages reserved for highatomic. */ if (likely (!(alloc_flags &ALLOC_RESERVES ))) - unusable_free += z->nr_reserved_highatomic; + unusable_free +=READ_ONCE (z->nr_free_highatomic);#ifdef CONFIG_CMA /* If allocation can't use CMA areas don't use free CMA pages */
Signature · 1 line
2.47.0.163.g1226f6d8fa-goog
Quoted 31 lines · Yu Zhao
OOM kills due to vastly overestimated free highatomic reserves were observed:
... invoked oom-killer: gfp_mask=0x100cca(GFP_HIGHUSER_MOVABLE), order=0 ...
Node 0 Normal free:1482936kB boost:0kB min:410416kB low:739404kB high:1068392kB reserved_highatomic:1073152KB ...
Node 0 Normal: 1292*4kB (ME) 1920*8kB (E) 383*16kB (UE) 220*32kB (ME) 340*64kB (E) 2155*128kB (UE) 3243*256kB (UE) 615*512kB (U) 1*1024kB (M) 0*2048kB 0*4096kB = 1477408kBThe second line above shows that the OOM kill was due to the following condition:
free (1482936kB) -
reserved_highatomic(1073152kB) = 409784KB < min (410416kB)And the third line shows there were no free pages in any
MIGRATE_HIGHATOMICpageblocks, which otherwise would show up as type 'H'. Therefore__zone_watermark_unusable_free()underestimated the usable free memory by over 1GB, which resulted in the unnecessary OOM kill above.The comments in
__zone_watermark_unusable_free()warns about the potential risk, i.e.,If the caller does not have rights to reserves below the min watermark then subtract the high-atomic reserves. This will over-estimate the size of the atomic reserve but it avoids a search.
However, it is possible to keep track of free pages in reserved highatomic pageblocks with a new per-zone counter
nr_free_highatomicprotected by the zone lock, to avoid a search when calculating the usable free memory. And the cost would be minimal, i.e., simple arithmetics in the highatomic alloc/free/move paths.
Quoted 7 lines
Note that since
nr_free_highatomiccan be relatively small, using a per-cpu counter might cause too much drift and defeat its purpose, in addition to the extra memory overhead.Reported-by Link Lin <[email protected]>Signed-off-by Yu Zhao <[email protected]>Acked-by David Rientjes <[email protected]>
For LTS benefit I'd also add:
Quoted 54 lines
--- include/linux/mmzone.h | 1 + mm/page_alloc.c | 10 +++++++--- 2 files changed, 8 insertions(+), 3 deletions(-)include/linux/mmzone.h +1
-0@@ -825,6 +825,7 @@ struct zone {unsigned long watermark_boost;unsigned long nr_reserved_highatomic; +unsigned long nr_free_highatomic;/* * We don't know if the memory that we're going to allocate will be mm/page_alloc.c +7
-3@@ -635,6 +635,8 @@ compaction_capture(struct capture_control *capc, struct page *page,static inline void account_freepages (struct zone *zone,int nr_pages,int migratetype) { +lockdep_assert_held (&zone->lock); +if (is_migrate_isolate (migratetype))return ;@@ -642,6 +644,9 @@ static inline void account_freepages(struct zone *zone, int nr_pages,if (is_migrate_cma (migratetype))__mod_zone_page_state (zone,NR_FREE_CMA_PAGES , nr_pages); + +if (is_migrate_highatomic (migratetype)) +WRITE_ONCE (zone->nr_free_highatomic, zone->nr_free_highatomic + nr_pages); }/* Used for pages not on another list */ @@ -3117,11 +3122,10 @@ static inline long __zone_watermark_unusable_free(struct zone *z,/* * If the caller does not have rights to reserves below the min -* watermark then subtract the high-atomic reserves. This will -* over-estimate the size of the atomic reserve but it avoids a search. +* watermark then subtract the free pages reserved for highatomic. */ if (likely (!(alloc_flags &ALLOC_RESERVES ))) - unusable_free += z->nr_reserved_highatomic; + unusable_free +=READ_ONCE (z->nr_free_highatomic);#ifdef CONFIG_CMA /* If allocation can't use CMA areas don't use free CMA pages */
Quoted 31 lines · Yu Zhao
OOM kills due to vastly overestimated free highatomic reserves were observed:
... invoked oom-killer: gfp_mask=0x100cca(GFP_HIGHUSER_MOVABLE), order=0 ...
Node 0 Normal free:1482936kB boost:0kB min:410416kB low:739404kB high:1068392kB reserved_highatomic:1073152KB ...
Node 0 Normal: 1292*4kB (ME) 1920*8kB (E) 383*16kB (UE) 220*32kB (ME) 340*64kB (E) 2155*128kB (UE) 3243*256kB (UE) 615*512kB (U) 1*1024kB (M) 0*2048kB 0*4096kB = 1477408kBThe second line above shows that the OOM kill was due to the following condition:
free (1482936kB) -
reserved_highatomic(1073152kB) = 409784KB < min (410416kB)And the third line shows there were no free pages in any
MIGRATE_HIGHATOMICpageblocks, which otherwise would show up as type 'H'. Therefore__zone_watermark_unusable_free()underestimated the usable free memory by over 1GB, which resulted in the unnecessary OOM kill above.The comments in
__zone_watermark_unusable_free()warns about the potential risk, i.e.,If the caller does not have rights to reserves below the min watermark then subtract the high-atomic reserves. This will over-estimate the size of the atomic reserve but it avoids a search.
However, it is possible to keep track of free pages in reserved highatomic pageblocks with a new per-zone counter
nr_free_highatomicprotected by the zone lock, to avoid a search when calculating the usable free memory. And the cost would be minimal, i.e., simple arithmetics in the highatomic alloc/free/move paths.
Quoted 7 lines
Note that since
nr_free_highatomiccan be relatively small, using a per-cpu counter might cause too much drift and defeat its purpose, in addition to the extra memory overhead.Reported-by Link Lin <[email protected]>Signed-off-by Yu Zhao <[email protected]>Acked-by David Rientjes <[email protected]>
mm/page_alloc.c static inline void account_freepages(struct zone *zone, int nr_pages, · line 649
Quoted 7 lines
mm/page_alloc.c +3
-0@@ -642,6 +644,9 @@ static inline void account_freepages(struct zone *zone, int nr_pages,if (is_migrate_cma (migratetype))__mod_zone_page_state (zone,NR_FREE_CMA_PAGES , nr_pages); + +if (is_migrate_highatomic (migratetype)) +WRITE_ONCE (zone->nr_free_highatomic, zone->nr_free_highatomic + nr_pages);
Minor nit, the page can only be of one migratetype, so `else if' would be better.
Quoted 45 lines · Johannes Weiner
Quoted 31 lines · Yu Zhao
OOM kills due to vastly overestimated free highatomic reserves were observed:
... invoked oom-killer: gfp_mask=0x100cca(GFP_HIGHUSER_MOVABLE), order=0 ...
Node 0 Normal free:1482936kB boost:0kB min:410416kB low:739404kB high:1068392kB reserved_highatomic:1073152KB ...
Node 0 Normal: 1292*4kB (ME) 1920*8kB (E) 383*16kB (UE) 220*32kB (ME) 340*64kB (E) 2155*128kB (UE) 3243*256kB (UE) 615*512kB (U) 1*1024kB (M) 0*2048kB 0*4096kB = 1477408kBThe second line above shows that the OOM kill was due to the following condition:
free (1482936kB) -
reserved_highatomic(1073152kB) = 409784KB < min (410416kB)And the third line shows there were no free pages in any
MIGRATE_HIGHATOMICpageblocks, which otherwise would show up as type 'H'. Therefore__zone_watermark_unusable_free()underestimated the usable free memory by over 1GB, which resulted in the unnecessary OOM kill above.The comments in
__zone_watermark_unusable_free()warns about the potential risk, i.e.,If the caller does not have rights to reserves below the min watermark then subtract the high-atomic reserves. This will over-estimate the size of the atomic reserve but it avoids a search.
However, it is possible to keep track of free pages in reserved highatomic pageblocks with a new per-zone counter
nr_free_highatomicprotected by the zone lock, to avoid a search when calculating the usable free memory. And the cost would be minimal, i.e., simple arithmetics in the highatomic alloc/free/move paths.Quoted 7 lines
Note that since
nr_free_highatomiccan be relatively small, using a per-cpu counter might cause too much drift and defeat its purpose, in addition to the extra memory overhead.Reported-by Link Lin <[email protected]>Signed-off-by Yu Zhao <[email protected]>Acked-by David Rientjes <[email protected]>Acked-by Johannes Weiner <[email protected]>Quoted 2 lines
(quoted diff) +0
-0@@ -642,6 +644,9 @@ static inline void account_freepages(struct zone *zone, int nr_pages,
Quoted 8 lines
mm/page_alloc.c static inline void account_freepages(struct zone *zone, int nr_pages, · line 649
Quoted 5 lines
mm/page_alloc.c +3
-0if (is_migrate_cma (migratetype))__mod_zone_page_state (zone,NR_FREE_CMA_PAGES , nr_pages); + +if (is_migrate_highatomic (migratetype)) +WRITE_ONCE (zone->nr_free_highatomic, zone->nr_free_highatomic + nr_pages);Minor nit, the page can only be of one migratetype, so `else if' would be better.
Right (copied and pasted without thinking).
Andrew, could you please fix this up in place? Thank you!