mm: memcg/slab: Fix objcg pointer array handling problem
Subsystems (4)
v3 cover mm: memcg/slab: Fix objcg pointer array handling problem
v3:
- Update patch 2 commit log and rework kmalloc_type() to make it easier to read.
v2:
- Take suggestion from Vlastimil to use a new set of kmalloc-cg-* to handle the objcg pointer array allocation and freeing problems.
Since the merging of the new slab memory controller in v5.9, the page structure stores a pointer to objcg pointer array for slab pages. When the slab has no used objects, it can be freed in free_slab() which will call kfree() to free the objcg pointer array in memcg_alloc_page_obj_cgroups(). If it happens that the objcg pointer array is the last used object in its slab, that slab may then be freed which may caused kfree() to be called again.
With the right workload, the slab cache may be set up in a way that allows the recursive kfree() calling loop to nest deep enough to cause a kernel stack overflow and panic the system. In fact, we have a reproducer that can cause kernel stack overflow on a s390 system involving kmalloc-rcl-256 and kmalloc-rcl-128 slabs with the following kfree() loop recursively called 74 times:
[ 285.520739] [<000000000ec432fc>] kfree+0x4bc/0x560 [ 285.520740] [<000000000ec43466>] __free_slab+0xc6/0x228 [ 285.520741] [<000000000ec41fc2>] __slab_free+0x3c2/0x3e0 [ 285.520742] [<000000000ec432fc>] kfree+0x4bc/0x560 :
While investigating this issue, I also found an issue on the allocation side. If the objcg pointer array happen to come from the same slab or a circular dependency linkage is formed with multiple slabs, those affected slabs can never be freed again.
This patch series addresses these two issues by introducing a new set of kmalloc-cg-<n> caches split from kmalloc-<n> caches. The new set will only contain non-reclaimable and non-dma objects that are accounted in memory cgroups whereas the old set are now for unaccounted objects only. By making this split, all the objcg pointer arrays will come from the kmalloc-<n> caches, but those caches will never hold any objcg pointer array. As a result, deeply nested kfree() call and the unfreeable slab problems are now gone.
Waiman Long (2):
mm: memcg/slab: Properly set up gfp flags for objcg pointer array
mm: memcg/slab: Create a new set of kmalloc-cg-<n> caches
include/linux/slab.h | 42 ++++++++++++++++++++++++++++++++++-------- mm/memcontrol.c | 8 ++++++++ mm/slab.h | 1 - mm/slab_common.c | 23 +++++++++++++++-------- 4 files changed, 57 insertions(+), 17 deletions(-)
Signature · 1 line
2.18.1
PATCH v3 1/2 mm: memcg/slab: Properly set up gfp flags for objcg pointer array
Since the merging of the new slab memory controller in v5.9, the page structure may store a pointer to obj_cgroup pointer array for slab pages. Currently, only the __GFP_ACCOUNT bit is masked off. However, the array is not readily reclaimable and doesn't need to come from the DMA buffer. So those GFP bits should be masked off as well.
Do the flag bit clearing at memcg_alloc_page_obj_cgroups() to make sure that it is consistently applied no matter where it is called.
--- mm/memcontrol.c | 8 ++++++++ mm/slab.h | 1 - 2 files changed, 8 insertions(+), 1 deletion(-)
mm/memcontrol.c +8 -0
@@ -2863,6 +2863,13 @@ static struct mem_cgroup *get_mem_cgroup_from_objcg(struct obj_cgroup *objcg) }#ifdef CONFIG_MEMCG_KMEM +/* +* The allocated objcg pointers array is not accounted directly. +* Moreover, it should not come from DMA buffer and is not readily +* reclaimable. So those GFP bits should be masked off. +*/ +#define OBJCGS_CLEAR_MASK (__GFP_DMA | __GFP_RECLAIMABLE | __GFP_ACCOUNT) +int memcg_alloc_page_obj_cgroups (struct page *page,struct kmem_cache *s,gfp_t gfp,bool new_page) {
@@ -2870,6 +2877,7 @@ int memcg_alloc_page_obj_cgroups(struct page *page, struct kmem_cache *s,unsigned long memcg_data;void *vec; + gfp &= ~OBJCGS_CLEAR_MASK ; vec =kcalloc_node (objects,sizeof (struct obj_cgroup *), gfp,page_to_nid (page));if (!vec)
mm/slab.h +0 -1
@@ -309,7 +309,6 @@ static inline void memcg_slab_post_alloc_hook(struct kmem_cache *s,if (!memcg_kmem_enabled () || !objcg)return ; - flags &= ~__GFP_ACCOUNT;for (i =0 ; i < size; i++) {if (likely (p[i])) { page =virt_to_head_page (p[i]);
Signature · 1 line
2.18.1
Re: [PATCH v3 1/2] mm: memcg/slab: Properly set up gfp flags for objcg pointer array
Quoted 7 lines · Waiman Long
Since the merging of the new slab memory controller in v5.9, the page structure may store a pointer to
obj_cgrouppointer array for slab pages. Currently, only the__GFP_ACCOUNTbit is masked off. However, the array is not readily reclaimable and doesn't need to come from the DMA buffer. So those GFP bits should be masked off as well.
Quoted 5 lines
Do the flag bit clearing at
memcg_alloc_page_obj_cgroups()to make sure that it is consistently applied no matter where it is called.Fixes 286e04b8ed7a ("mm: memcg/slab: allocate obj_cgroups for non-root slab pages")Signed-off-by Waiman Long <[email protected]>
Re: [PATCH v3 1/2] mm: memcg/slab: Properly set up gfp flags for objcg pointer array
In reply to Waiman LongQuoted 9 lines · Waiman Long
Since the merging of the new slab memory controller in v5.9, the page structure may store a pointer to
obj_cgrouppointer array for slab pages. Currently, only the__GFP_ACCOUNTbit is masked off. However, the array is not readily reclaimable and doesn't need to come from the DMA buffer. So those GFP bits should be masked off as well.Do the flag bit clearing at
memcg_alloc_page_obj_cgroups()to make sure that it is consistently applied no matter where it is called.
Quoted 6 lines
Fixes 286e04b8ed7a ("mm: memcg/slab: allocate obj_cgroups for non-root slab pages")Signed-off-by Waiman Long <[email protected]>--- mm/memcontrol.c | 8 ++++++++ mm/slab.h | 1 - 2 files changed, 8 insertions(+), 1 deletion(-)
Thanks!
PATCH v3 2/2 mm: memcg/slab: Create a new set of kmalloc-cg-<n> caches
In reply to Waiman LongThere are currently two problems in the way the objcg pointer array (memcg_data) in the page structure is being allocated and freed.
On its allocation, it is possible that the allocated objcg pointer array comes from the same slab that requires memory accounting. If this happens, the slab will never become empty again as there is at least one object left (the obj_cgroup array) in the slab.
When it is freed, the objcg pointer array object may be the last one in its slab and hence causes kfree() to be called again. With the right workload, the slab cache may be set up in a way that allows the recursive kfree() calling loop to nest deep enough to cause a kernel stack overflow and panic the system.
One way to solve this problem is to split the kmalloc-<n> caches (KMALLOC_NORMAL) into two separate sets - a new set of kmalloc-<n> (KMALLOC_NORMAL) caches for non-accounted objects only and a new set of kmalloc-cg-<n> (KMALLOC_CGROUP) caches for accounted objects only. All the other caches can still allow a mix of accounted and non-accounted objects.
With this change, all the objcg pointer array objects will come from KMALLOC_NORMAL caches which won't have their objcg pointer arrays. So both the recursive kfree() problem and non-freeable slab problem are gone. Since both the KMALLOC_NORMAL and KMALLOC_CGROUP caches no longer have mixed accounted and unaccounted objects, this will slightly reduce the number of objcg pointer arrays that need to be allocated and save a bit of memory.
The new KMALLOC_CGROUP is added between KMALLOC_NORMAL and KMALLOC_RECLAIM so that the first for loop in create_kmalloc_caches() will include the newly added caches without change.
--- include/linux/slab.h | 42 ++++++++++++++++++++++++++++++++++-------- mm/slab_common.c | 23 +++++++++++++++-------- 2 files changed, 49 insertions(+), 16 deletions(-)
include/linux/slab.h +34 -8
@@ -305,9 +305,16 @@ static inline void __check_heap_object(const void *ptr, unsigned long n,/* * Whenever changing this, take care of that kmalloc_type() and * create_kmalloc_caches() still work as intended. +* +* KMALLOC_NORMAL is for non-accounted objects only whereas KMALLOC_CGROUP +* is for accounted objects only. All the other kmem caches can have both +* accounted and non-accounted objects. */ enum kmalloc_cache_type {KMALLOC_NORMAL =0 , +#ifdef CONFIG_MEMCG_KMEM +KMALLOC_CGROUP , +#endif KMALLOC_RECLAIM ,#ifdef CONFIG_ZONE_DMA KMALLOC_DMA ,
@@ -315,28 +322,47 @@ enum kmalloc_cache_type {NR_KMALLOC_TYPES }; +#ifndef CONFIG_MEMCG_KMEM +#define KMALLOC_CGROUP KMALLOC_NORMAL +#endif +#ifndef CONFIG_ZONE_DMA +#define KMALLOC_DMA KMALLOC_NORMAL +#endif +#ifndef CONFIG_SLOB extern struct kmem_cache * kmalloc_caches[NR_KMALLOC_TYPES ][KMALLOC_SHIFT_HIGH +1 ]; +/* +* Define gfp bits that should not be set for KMALLOC_NORMAL. +*/ +#define KMALLOC_NOT_NORMAL_BITS \ + (__GFP_RECLAIMABLE | \ + (IS_ENABLED(CONFIG_ZONE_DMA) ? __GFP_DMA : 0) | \ + (IS_ENABLED(CONFIG_MEMCG_KMEM) ? __GFP_ACCOUNT : 0)) + static __always_inline enum kmalloc_cache_type kmalloc_type(gfp_t flags) { -#ifdef CONFIG_ZONE_DMA /* * The most common case is KMALLOC_NORMAL, so test for it * with a single branch for both flags. */ -if (likely ((flags & (__GFP_DMA | __GFP_RECLAIMABLE)) ==0 )) +if (likely ((flags &KMALLOC_NOT_NORMAL_BITS ) ==0 ))return KMALLOC_NORMAL ;/* -* At least one of the flags has to be set. If both are, __GFP_DMA -* is more important. +* At least one of the flags has to be set. Their priorities in +* decreasing order are: +* 1) __GFP_DMA +* 2) __GFP_RECLAIMABLE +* 3) __GFP_ACCOUNT */ -return flags & __GFP_DMA ?KMALLOC_DMA :KMALLOC_RECLAIM ; -#else -return flags & __GFP_RECLAIMABLE ?KMALLOC_RECLAIM :KMALLOC_NORMAL ; -#endif +if (IS_ENABLED (CONFIG_ZONE_DMA ) && (flags & __GFP_DMA)) +return KMALLOC_DMA ; +if (!IS_ENABLED (CONFIG_MEMCG_KMEM ) || (flags & __GFP_RECLAIMABLE)) +return KMALLOC_RECLAIM ; +else +return KMALLOC_CGROUP ; }/*
mm/slab_common.c +15 -8
@@ -727,21 +727,25 @@ struct kmem_cache *kmalloc_slab(size_t size, gfp_t flags) }#ifdef CONFIG_ZONE_DMA -#define INIT_KMALLOC_INFO (__size, __short_size) \ -{ \ - .name[KMALLOC_NORMAL] = "kmalloc-" #__short_size, \ - .name[KMALLOC_RECLAIM] = "kmalloc-rcl-" #__short_size, \ - .name[KMALLOC_DMA] = "dma-kmalloc-" #__short_size, \ - .size = __size, \ -} +#define KMALLOC_DMA_NAME (sz) .name[KMALLOC_DMA] = "dma-kmalloc-" #sz, +#else +#define KMALLOC_DMA_NAME (sz) +#endif + +#ifdef CONFIG_MEMCG_KMEM +#define KMALLOC_CGROUP_NAME (sz) .name[KMALLOC_CGROUP] = "kmalloc-cg-" #sz,#else +#define KMALLOC_CGROUP_NAME (sz) +#endif +#define INIT_KMALLOC_INFO (__size, __short_size) \ { \ .name[KMALLOC_NORMAL] = "kmalloc-" #__short_size, \ .name[KMALLOC_RECLAIM] = "kmalloc-rcl-" #__short_size, \ + KMALLOC_CGROUP_NAME(__short_size) \ + KMALLOC_DMA_NAME(__short_size) \ .size = __size, \ } -#endif /* * kmalloc_info[] is to make slub_debug=,kmalloc-xx option work at boot time.
@@ -847,6 +851,9 @@ void __init create_kmalloc_caches(slab_flags_t flags)int i;enum kmalloc_cache_type type; +/* +* Including KMALLOC_CGROUP if CONFIG_MEMCG_KMEM defined +*/ for (type =KMALLOC_NORMAL ; type <=KMALLOC_RECLAIM ; type++) {for (i =KMALLOC_SHIFT_LOW ; i <=KMALLOC_SHIFT_HIGH ; i++) {if (!kmalloc_caches[type][i])
Signature · 1 line
2.18.1
Re: [PATCH v3 2/2] mm: memcg/slab: Create a new set of kmalloc-cg-<n> caches
Quoted 62 lines · Waiman Long
There are currently two problems in the way the objcg pointer array (
memcg_data) in the page structure is being allocated and freed.On its allocation, it is possible that the allocated objcg pointer array comes from the same slab that requires memory accounting. If this happens, the slab will never become empty again as there is at least one object left (the
obj_cgrouparray) in the slab.When it is freed, the objcg pointer array object may be the last one in its slab and hence causes
kfree()to be called again. With the right workload, the slab cache may be set up in a way that allows the recursivekfree()calling loop to nest deep enough to cause a kernel stack overflow and panic the system.One way to solve this problem is to split the kmalloc-<n> caches (
KMALLOC_NORMAL) into two separate sets - a new set of kmalloc-<n> (KMALLOC_NORMAL) caches for non-accounted objects only and a new set of kmalloc-cg-<n> (KMALLOC_CGROUP) caches for accounted objects only. All the other caches can still allow a mix of accounted and non-accounted objects.With this change, all the objcg pointer array objects will come from
KMALLOC_NORMALcaches which won't have their objcg pointer arrays. So both the recursivekfree()problem and non-freeable slab problem are gone. Since both theKMALLOC_NORMALandKMALLOC_CGROUPcaches no longer have mixed accounted and unaccounted objects, this will slightly reduce the number of objcg pointer arrays that need to be allocated and save a bit of memory.The new
KMALLOC_CGROUPis added betweenKMALLOC_NORMALandKMALLOC_RECLAIMso that the first for loop increate_kmalloc_caches()will include the newly added caches without change.Suggested-by Vlastimil Babka <[email protected]>Signed-off-by Waiman Long <[email protected]>--- include/linux/slab.h | 42 ++++++++++++++++++++++++++++++++++-------- mm/slab_common.c | 23 +++++++++++++++-------- 2 files changed, 49 insertions(+), 16 deletions(-)include/linux/slab.h +7
-0@@ -305,9 +305,16 @@ static inline void __check_heap_object(const void *ptr, unsigned long n,/* * Whenever changing this, take care of that kmalloc_type() and * create_kmalloc_caches() still work as intended. +* +* KMALLOC_NORMAL is for non-accounted objects only whereas KMALLOC_CGROUP +* is for accounted objects only. All the other kmem caches can have both +* accounted and non-accounted objects. */ enum kmalloc_cache_type {KMALLOC_NORMAL =0 , +#ifdef CONFIG_MEMCG_KMEM +KMALLOC_CGROUP , +#endif KMALLOC_RECLAIM ,#ifdef CONFIG_ZONE_DMA KMALLOC_DMA ,
include/linux/slab.h enum kmalloc_cache_type { · line 330
Quoted 9 lines
include/linux/slab.h +6
-0@@ -315,28 +322,47 @@ enum kmalloc_cache_type { NR_KMALLOC_TYPES }; +#ifndef CONFIG_MEMCG_KMEM +#define KMALLOC_CGROUP KMALLOC_NORMAL +#endif +#ifndef CONFIG_ZONE_DMA +#define KMALLOC_DMA KMALLOC_NORMAL +#endif
You could move this to the enum definition itself? E.g.:
#ifdef CONFIG_MEMCG_KMEM
KMALLOC_CGROUP,
#else
KMALLOC_CGROUP = KMALLOC_NORMAL,
#endif
Quoted 9 lines
include/linux/slab.h +5
-0+#ifndef CONFIG_SLOB extern struct kmem_cache * kmalloc_caches[NR_KMALLOC_TYPES ][KMALLOC_SHIFT_HIGH +1 ]; +/* +* Define gfp bits that should not be set for KMALLOC_NORMAL. +*/ +#define KMALLOC_NOT_NORMAL_BITS \
include/linux/slab.h enum kmalloc_cache_type { · line 348
Quoted 10 lines
include/linux/slab.h +4
-1+ (__GFP_RECLAIMABLE | \ + (IS_ENABLED (CONFIG_ZONE_DMA ) ? __GFP_DMA :0 ) | \ + (IS_ENABLED(CONFIG_MEMCG_KMEM) ? __GFP_ACCOUNT : 0)) + static __always_inline enum kmalloc_cache_type kmalloc_type(gfp_t flags) { -#ifdef CONFIG_ZONE_DMA /* * The most common case is KMALLOC_NORMAL, so test for it * with a single branch for both flags.
Not "both flags" anymore. Something like "so test with a single branch that there are none of the flags that would select a different type"
Quoted 15 lines
include/linux/slab.h +6
-4*/ -if (likely ((flags & (__GFP_DMA | __GFP_RECLAIMABLE)) ==0 )) +if (likely ((flags &KMALLOC_NOT_NORMAL_BITS ) ==0 ))return KMALLOC_NORMAL ;/* -* At least one of the flags has to be set. If both are, __GFP_DMA -* is more important. +* At least one of the flags has to be set. Their priorities in +* decreasing order are: +* 1) __GFP_DMA +* 2) __GFP_RECLAIMABLE +* 3) __GFP_ACCOUNT */ -return flags & __GFP_DMA ?KMALLOC_DMA :KMALLOC_RECLAIM ;
include/linux/slab.h enum kmalloc_cache_type { · line 366
Quoted 10 lines
include/linux/slab.h +6
-3-#else -return flags & __GFP_RECLAIMABLE ?KMALLOC_RECLAIM :KMALLOC_NORMAL ; -#endif +if (IS_ENABLED (CONFIG_ZONE_DMA ) && (flags & __GFP_DMA)) +return KMALLOC_DMA ; +if (!IS_ENABLED (CONFIG_MEMCG_KMEM ) || (flags & __GFP_RECLAIMABLE)) +return KMALLOC_RECLAIM ; +else +return KMALLOC_CGROUP ; }
Works for me this way, thanks.
Quoted 51 lines
include/linux/slab.h +0
-0/* mm/slab_common.c +15
-8@@ -727,21 +727,25 @@ struct kmem_cache *kmalloc_slab(size_t size, gfp_t flags) }#ifdef CONFIG_ZONE_DMA -#define INIT_KMALLOC_INFO (__size, __short_size) \ -{ \ - .name[KMALLOC_NORMAL] = "kmalloc-" #__short_size, \ - .name[KMALLOC_RECLAIM] = "kmalloc-rcl-" #__short_size, \ - .name[KMALLOC_DMA] = "dma-kmalloc-" #__short_size, \ - .size = __size, \ -} +#define KMALLOC_DMA_NAME (sz) .name[KMALLOC_DMA] = "dma-kmalloc-" #sz, +#else +#define KMALLOC_DMA_NAME (sz) +#endif + +#ifdef CONFIG_MEMCG_KMEM +#define KMALLOC_CGROUP_NAME (sz) .name[KMALLOC_CGROUP] = "kmalloc-cg-" #sz,#else +#define KMALLOC_CGROUP_NAME (sz) +#endif +#define INIT_KMALLOC_INFO (__size, __short_size) \ { \ .name[KMALLOC_NORMAL] = "kmalloc-" #__short_size, \ .name[KMALLOC_RECLAIM] = "kmalloc-rcl-" #__short_size, \ + KMALLOC_CGROUP_NAME(__short_size) \ + KMALLOC_DMA_NAME(__short_size) \ .size = __size, \ } -#endif /* * kmalloc_info[] is to make slub_debug=,kmalloc-xx option work at boot time. @@ -847,6 +851,9 @@ void __init create_kmalloc_caches(slab_flags_t flags)int i;enum kmalloc_cache_type type; +/* +* Including KMALLOC_CGROUP if CONFIG_MEMCG_KMEM defined +*/ for (type =KMALLOC_NORMAL ; type <=KMALLOC_RECLAIM ; type++) {for (i =KMALLOC_SHIFT_LOW ; i <=KMALLOC_SHIFT_HIGH ; i++) {if (!kmalloc_caches[type][i])
Re: [PATCH v3 2/2] mm: memcg/slab: Create a new set of kmalloc-cg-<n> caches
Quoted 85 lines · Vlastimil Babka
Quoted 62 lines · Waiman Long
There are currently two problems in the way the objcg pointer array (
memcg_data) in the page structure is being allocated and freed.On its allocation, it is possible that the allocated objcg pointer array comes from the same slab that requires memory accounting. If this happens, the slab will never become empty again as there is at least one object left (the
obj_cgrouparray) in the slab.When it is freed, the objcg pointer array object may be the last one in its slab and hence causes
kfree()to be called again. With the right workload, the slab cache may be set up in a way that allows the recursivekfree()calling loop to nest deep enough to cause a kernel stack overflow and panic the system.One way to solve this problem is to split the kmalloc-<n> caches (
KMALLOC_NORMAL) into two separate sets - a new set of kmalloc-<n> (KMALLOC_NORMAL) caches for non-accounted objects only and a new set of kmalloc-cg-<n> (KMALLOC_CGROUP) caches for accounted objects only. All the other caches can still allow a mix of accounted and non-accounted objects.With this change, all the objcg pointer array objects will come from
KMALLOC_NORMALcaches which won't have their objcg pointer arrays. So both the recursivekfree()problem and non-freeable slab problem are gone. Since both theKMALLOC_NORMALandKMALLOC_CGROUPcaches no longer have mixed accounted and unaccounted objects, this will slightly reduce the number of objcg pointer arrays that need to be allocated and save a bit of memory.The new
KMALLOC_CGROUPis added betweenKMALLOC_NORMALandKMALLOC_RECLAIMso that the first for loop increate_kmalloc_caches()will include the newly added caches without change.Suggested-by Vlastimil Babka <[email protected]>Signed-off-by Waiman Long <[email protected]>--- include/linux/slab.h | 42 ++++++++++++++++++++++++++++++++++-------- mm/slab_common.c | 23 +++++++++++++++-------- 2 files changed, 49 insertions(+), 16 deletions(-)include/linux/slab.h +7
-0@@ -305,9 +305,16 @@ static inline void __check_heap_object(const void *ptr, unsigned long n,/* * Whenever changing this, take care of that kmalloc_type() and * create_kmalloc_caches() still work as intended. +* +* KMALLOC_NORMAL is for non-accounted objects only whereas KMALLOC_CGROUP +* is for accounted objects only. All the other kmem caches can have both +* accounted and non-accounted objects. */ enum kmalloc_cache_type {KMALLOC_NORMAL =0 , +#ifdef CONFIG_MEMCG_KMEM +KMALLOC_CGROUP , +#endif KMALLOC_RECLAIM ,#ifdef CONFIG_ZONE_DMA KMALLOC_DMA ,include/linux/slab.h enum kmalloc_cache_type { · line 330
Quoted 9 lines
include/linux/slab.h +6
-0@@ -315,28 +322,47 @@ enum kmalloc_cache_type { NR_KMALLOC_TYPES }; +#ifndef CONFIG_MEMCG_KMEM +#define KMALLOC_CGROUP KMALLOC_NORMAL +#endif +#ifndef CONFIG_ZONE_DMA +#define KMALLOC_DMA KMALLOC_NORMAL +#endif You could move this to the enum definition itself? E.g.:
#ifdef
CONFIG_MEMCG_KMEM
KMALLOC_CGROUP,
#else
KMALLOC_CGROUP=KMALLOC_NORMAL,
#endifQuoted 5 lines
include/linux/slab.h +1
-0+#ifndef CONFIG_SLOB extern struct kmem_cache * kmalloc_caches[NR_KMALLOC_TYPES ][KMALLOC_SHIFT_HIGH +1 ];
Quoted 16 lines
Quoted 4 lines
include/linux/slab.h +4
-0+/* +* Define gfp bits that should not be set for KMALLOC_NORMAL. +*/ +#define KMALLOC_NOT_NORMAL_BITS \include/linux/slab.h enum kmalloc_cache_type { · line 348
Quoted 10 lines
include/linux/slab.h +4
-1+ (__GFP_RECLAIMABLE | \ + (IS_ENABLED (CONFIG_ZONE_DMA ) ? __GFP_DMA :0 ) | \ + (IS_ENABLED(CONFIG_MEMCG_KMEM) ? __GFP_ACCOUNT : 0)) + static __always_inline enum kmalloc_cache_type kmalloc_type(gfp_t flags) { -#ifdef CONFIG_ZONE_DMA /* * The most common case is KMALLOC_NORMAL, so test for it * with a single branch for both flags. Not "both flags" anymore. Something like "so test with a single branch that there are none of the flags that would select a different type"
Right. I just left the comment there without taking a deeper look. My bad.
Cheers,
Longman
Re: [PATCH v3 2/2] mm: memcg/slab: Create a new set of kmalloc-cg-<n> caches
In reply to Waiman LongQuoted 30 lines · Waiman Long
There are currently two problems in the way the objcg pointer array (
memcg_data) in the page structure is being allocated and freed.On its allocation, it is possible that the allocated objcg pointer array comes from the same slab that requires memory accounting. If this happens, the slab will never become empty again as there is at least one object left (the
obj_cgrouparray) in the slab.When it is freed, the objcg pointer array object may be the last one in its slab and hence causes
kfree()to be called again. With the right workload, the slab cache may be set up in a way that allows the recursivekfree()calling loop to nest deep enough to cause a kernel stack overflow and panic the system.One way to solve this problem is to split the kmalloc-<n> caches (
KMALLOC_NORMAL) into two separate sets - a new set of kmalloc-<n> (KMALLOC_NORMAL) caches for non-accounted objects only and a new set of kmalloc-cg-<n> (KMALLOC_CGROUP) caches for accounted objects only. All the other caches can still allow a mix of accounted and non-accounted objects.With this change, all the objcg pointer array objects will come from
KMALLOC_NORMALcaches which won't have their objcg pointer arrays. So both the recursivekfree()problem and non-freeable slab problem are gone. Since both theKMALLOC_NORMALandKMALLOC_CGROUPcaches no longer have mixed accounted and unaccounted objects, this will slightly reduce the number of objcg pointer arrays that need to be allocated and save a bit of memory.
Quoted 6 lines
The new
KMALLOC_CGROUPis added betweenKMALLOC_NORMALandKMALLOC_RECLAIMso that the first for loop increate_kmalloc_caches()will include the newly added caches without change.Suggested-by Vlastimil Babka <[email protected]>Signed-off-by Waiman Long <[email protected]>
One nit below and after incorporating Vlastimil's suggestions:
Quoted 10 lines
--- include/linux/slab.h | 42 ++++++++++++++++++++++++++++++++++-------- mm/slab_common.c | 23 +++++++++++++++-------- 2 files changed, 49 insertions(+), 16 deletions(-)include/linux/slab.h +0
-0
include/linux/slab.h static inline void __check_heap_object(const void *ptr, unsigned long n, · line 310
Quoted 6 lines
include/linux/slab.h +3
-0@@ -305,9 +305,16 @@ static inline void __check_heap_object(const void *ptr, unsigned long n,/* * Whenever changing this, take care of that kmalloc_type() and * create_kmalloc_caches() still work as intended. +* +* KMALLOC_NORMAL is for non-accounted objects only whereas KMALLOC_CGROUP +* is for accounted objects only.
I think you can say "KMALLOC_CGROUP is for accounted and unreclaimable objects only".
Re: [PATCH v3 2/2] mm: memcg/slab: Create a new set of kmalloc-cg-<n> caches
Quoted 45 lines · Shakeel Butt
Quoted 29 lines · Waiman Long
There are currently two problems in the way the objcg pointer array (
memcg_data) in the page structure is being allocated and freed.On its allocation, it is possible that the allocated objcg pointer array comes from the same slab that requires memory accounting. If this happens, the slab will never become empty again as there is at least one object left (the
obj_cgrouparray) in the slab.When it is freed, the objcg pointer array object may be the last one in its slab and hence causes
kfree()to be called again. With the right workload, the slab cache may be set up in a way that allows the recursivekfree()calling loop to nest deep enough to cause a kernel stack overflow and panic the system.One way to solve this problem is to split the kmalloc-<n> caches (
KMALLOC_NORMAL) into two separate sets - a new set of kmalloc-<n> (KMALLOC_NORMAL) caches for non-accounted objects only and a new set of kmalloc-cg-<n> (KMALLOC_CGROUP) caches for accounted objects only. All the other caches can still allow a mix of accounted and non-accounted objects.With this change, all the objcg pointer array objects will come from
KMALLOC_NORMALcaches which won't have their objcg pointer arrays. So both the recursivekfree()problem and non-freeable slab problem are gone. Since both theKMALLOC_NORMALandKMALLOC_CGROUPcaches no longer have mixed accounted and unaccounted objects, this will slightly reduce the number of objcg pointer arrays that need to be allocated and save a bit of memory.Quoted 6 lines
The new
KMALLOC_CGROUPis added betweenKMALLOC_NORMALandKMALLOC_RECLAIMso that the first for loop increate_kmalloc_caches()will include the newly added caches without change.Suggested-by Vlastimil Babka <[email protected]>Signed-off-by Waiman Long <[email protected]>One nit below and after incorporating Vlastimil's suggestions:
Reviewed-by Shakeel Butt <[email protected]>Quoted 5 lines
--- include/linux/slab.h | 42 ++++++++++++++++++++++++++++++++++-------- mm/slab_common.c | 23 +++++++++++++++-------- 2 files changed, 49 insertions(+), 16 deletions(-)
Quoted 14 lines
include/linux/slab.h static inline void __check_heap_object(const void *ptr, unsigned long n, · line 310
Quoted 11 lines
include/linux/slab.h +3
-0@@ -305,9 +305,16 @@ static inline void __check_heap_object(const void *ptr, unsigned long n,/* * Whenever changing this, take care of that kmalloc_type() and * create_kmalloc_caches() still work as intended. +* +* KMALLOC_NORMAL is for non-accounted objects only whereas KMALLOC_CGROUP +* is for accounted objects only. I think you can say "
KMALLOC_CGROUPis for accounted and unreclaimable objects only".
Thanks for the suggestion. Will incorporate that.
Cheers,
Longman
Re: [PATCH v3 2/2] mm: memcg/slab: Create a new set of kmalloc-cg-<n> caches
In reply to Waiman LongQuoted 14 lines · Waiman Long
There are currently two problems in the way the objcg pointer array (
memcg_data) in the page structure is being allocated and freed.On its allocation, it is possible that the allocated objcg pointer array comes from the same slab that requires memory accounting. If this happens, the slab will never become empty again as there is at least one object left (the
obj_cgrouparray) in the slab.When it is freed, the objcg pointer array object may be the last one in its slab and hence causes
kfree()to be called again. With the right workload, the slab cache may be set up in a way that allows the recursivekfree()calling loop to nest deep enough to cause a kernel stack overflow and panic the system.
Quoted 6 lines
One way to solve this problem is to split the kmalloc-<n> caches (
KMALLOC_NORMAL) into two separate sets - a new set of kmalloc-<n> (KMALLOC_NORMAL) caches for non-accounted objects only and a new set of kmalloc-cg-<n> (KMALLOC_CGROUP) caches for accounted objects only. All the other caches can still allow a mix of accounted and non-accounted objects.
I agree that it's likely the best approach here. Thanks for discovering and fixing the problem!
Quoted 8 lines
With this change, all the objcg pointer array objects will come from
KMALLOC_NORMALcaches which won't have their objcg pointer arrays. So both the recursivekfree()problem and non-freeable slab problem are gone. Since both theKMALLOC_NORMALandKMALLOC_CGROUPcaches no longer have mixed accounted and unaccounted objects, this will slightly reduce the number of objcg pointer arrays that need to be allocated and save a bit of memory.
Unfortunately the positive effect of this change will be likely reversed by a lower utilization due to a larger number of caches.
Btw, I wonder if we also need a change in the slab caches merging procedure? KMALLOC_NORMAL caches should not be merged with caches which can potentially include accounted objects.
Quoted 56 lines
The new
KMALLOC_CGROUPis added betweenKMALLOC_NORMALandKMALLOC_RECLAIMso that the first for loop increate_kmalloc_caches()will include the newly added caches without change.Suggested-by Vlastimil Babka <[email protected]>Signed-off-by Waiman Long <[email protected]>--- include/linux/slab.h | 42 ++++++++++++++++++++++++++++++++++-------- mm/slab_common.c | 23 +++++++++++++++-------- 2 files changed, 49 insertions(+), 16 deletions(-)include/linux/slab.h +22
-0@@ -305,9 +305,16 @@ static inline void __check_heap_object(const void *ptr, unsigned long n,/* * Whenever changing this, take care of that kmalloc_type() and * create_kmalloc_caches() still work as intended. +* +* KMALLOC_NORMAL is for non-accounted objects only whereas KMALLOC_CGROUP +* is for accounted objects only. All the other kmem caches can have both +* accounted and non-accounted objects. */ enum kmalloc_cache_type {KMALLOC_NORMAL =0 , +#ifdef CONFIG_MEMCG_KMEM +KMALLOC_CGROUP , +#endif KMALLOC_RECLAIM ,#ifdef CONFIG_ZONE_DMA KMALLOC_DMA ,@@ -315,28 +322,47 @@ enum kmalloc_cache_type { NR_KMALLOC_TYPES }; +#ifndef CONFIG_MEMCG_KMEM +#define KMALLOC_CGROUP KMALLOC_NORMAL +#endif +#ifndef CONFIG_ZONE_DMA +#define KMALLOC_DMA KMALLOC_NORMAL +#endif +#ifndef CONFIG_SLOB extern struct kmem_cache * kmalloc_caches[NR_KMALLOC_TYPES ][KMALLOC_SHIFT_HIGH +1 ]; +/* +* Define gfp bits that should not be set for KMALLOC_NORMAL. +*/ +#define KMALLOC_NOT_NORMAL_BITS \ + (__GFP_RECLAIMABLE | \ + (IS_ENABLED(CONFIG_ZONE_DMA) ? __GFP_DMA : 0) | \ + (IS_ENABLED(CONFIG_MEMCG_KMEM) ? __GFP_ACCOUNT : 0)) +
include/linux/slab.h enum kmalloc_cache_type { · line 351
Quoted 10 lines
include/linux/slab.h +1
-2static __always_inline enum kmalloc_cache_type kmalloc_type(gfp_t flags) { -#ifdef CONFIG_ZONE_DMA /* * The most common case is KMALLOC_NORMAL, so test for it * with a single branch for both flags. */ -if (likely ((flags & (__GFP_DMA | __GFP_RECLAIMABLE)) ==0 )) +if (likely ((flags &KMALLOC_NOT_NORMAL_BITS ) ==0 ))return KMALLOC_NORMAL ;
Likely KMALLOC_CGROUP is also very popular, so maybe we want to change the optimization here a bit.
Quoted 74 lines
include/linux/slab.h +11
-6/* -* At least one of the flags has to be set. If both are, __GFP_DMA -* is more important. +* At least one of the flags has to be set. Their priorities in +* decreasing order are: +* 1) __GFP_DMA +* 2) __GFP_RECLAIMABLE +* 3) __GFP_ACCOUNT */ -return flags & __GFP_DMA ?KMALLOC_DMA :KMALLOC_RECLAIM ; -#else -return flags & __GFP_RECLAIMABLE ?KMALLOC_RECLAIM :KMALLOC_NORMAL ; -#endif +if (IS_ENABLED (CONFIG_ZONE_DMA ) && (flags & __GFP_DMA)) +return KMALLOC_DMA ; +if (!IS_ENABLED (CONFIG_MEMCG_KMEM ) || (flags & __GFP_RECLAIMABLE)) +return KMALLOC_RECLAIM ; +else +return KMALLOC_CGROUP ; }/* mm/slab_common.c +15
-8@@ -727,21 +727,25 @@ struct kmem_cache *kmalloc_slab(size_t size, gfp_t flags) }#ifdef CONFIG_ZONE_DMA -#define INIT_KMALLOC_INFO (__size, __short_size) \ -{ \ - .name[KMALLOC_NORMAL] = "kmalloc-" #__short_size, \ - .name[KMALLOC_RECLAIM] = "kmalloc-rcl-" #__short_size, \ - .name[KMALLOC_DMA] = "dma-kmalloc-" #__short_size, \ - .size = __size, \ -} +#define KMALLOC_DMA_NAME (sz) .name[KMALLOC_DMA] = "dma-kmalloc-" #sz, +#else +#define KMALLOC_DMA_NAME (sz) +#endif + +#ifdef CONFIG_MEMCG_KMEM +#define KMALLOC_CGROUP_NAME (sz) .name[KMALLOC_CGROUP] = "kmalloc-cg-" #sz,#else +#define KMALLOC_CGROUP_NAME (sz) +#endif +#define INIT_KMALLOC_INFO (__size, __short_size) \ { \ .name[KMALLOC_NORMAL] = "kmalloc-" #__short_size, \ .name[KMALLOC_RECLAIM] = "kmalloc-rcl-" #__short_size, \ + KMALLOC_CGROUP_NAME(__short_size) \ + KMALLOC_DMA_NAME(__short_size) \ .size = __size, \ } -#endif /* * kmalloc_info[] is to make slub_debug=,kmalloc-xx option work at boot time. @@ -847,6 +851,9 @@ void __init create_kmalloc_caches(slab_flags_t flags)int i;enum kmalloc_cache_type type; +/* +* Including KMALLOC_CGROUP if CONFIG_MEMCG_KMEM defined +*/ for (type =KMALLOC_NORMAL ; type <=KMALLOC_RECLAIM ; type++) {for (i =KMALLOC_SHIFT_LOW ; i <=KMALLOC_SHIFT_HIGH ; i++) {if (!kmalloc_caches[type][i])Signature · 1 line
2.18.1
Re: [PATCH v3 2/2] mm: memcg/slab: Create a new set of kmalloc-cg-<n> caches
Quoted 10 lines · Roman Gushchin
Quoted 8 lines · Waiman Long
With this change, all the objcg pointer array objects will come from
KMALLOC_NORMALcaches which won't have their objcg pointer arrays. So both the recursivekfree()problem and non-freeable slab problem are gone. Since both theKMALLOC_NORMALandKMALLOC_CGROUPcaches no longer have mixed accounted and unaccounted objects, this will slightly reduce the number of objcg pointer arrays that need to be allocated and save a bit of memory.
Quoted 6 lines
Unfortunately the positive effect of this change will be likely reversed by a lower utilization due to a larger number of caches.
Btw, I wonder if we also need a change in the slab caches merging procedure?
KMALLOC_NORMALcaches should not be merged with caches which can potentially include accounted objects.
Good point. But looks like kmalloc* caches are extempt from all merging in create_boot_cache() via
s->refcount = -1; /* Exempt from merging for now */
It wouldn't hurt though to create the kmalloc-cg-* caches with SLAB_ACCOUNT flag to prevent accidental merging in case the above is ever removed. It would also better reflect reality, and ensure that the array is allocated immediately with the page, AFAICS.
Re: [PATCH v3 2/2] mm: memcg/slab: Create a new set of kmalloc-cg-<n> caches
Quoted 14 lines · Vlastimil Babka
Quoted 13 lines · Roman Gushchin
Quoted 8 lines · Waiman Long
With this change, all the objcg pointer array objects will come from
KMALLOC_NORMALcaches which won't have their objcg pointer arrays. So both the recursivekfree()problem and non-freeable slab problem are gone. Since both theKMALLOC_NORMALandKMALLOC_CGROUPcaches no longer have mixed accounted and unaccounted objects, this will slightly reduce the number of objcg pointer arrays that need to be allocated and save a bit of memory.Unfortunately the positive effect of this change will be likely reversed by a lower utilization due to a larger number of caches.
Quoted 8 lines
Quoted 3 lines
Btw, I wonder if we also need a change in the slab caches merging procedure?
KMALLOC_NORMALcaches should not be merged with caches which can potentially include accounted objects.Good point. But looks like kmalloc* caches are extempt from all merging in
create_boot_cache()vias->refcount = -1; /* Exempt from merging for now */
Oh, interesting... I wonder if there is (still) a good reason for that? Maybe we can remove this limitation and save some memory?
Quoted 5 lines
It wouldn't hurt though to create the kmalloc-cg-* caches with
SLAB_ACCOUNTflag to prevent accidental merging in case the above is ever removed. It would also better reflect reality, and ensure that the array is allocated immediately with the page, AFAICS.
That wouldn't be enough, because a !SLAB_ACCOUNT cache can still have accounted allocations and be merged with kmalloc-* cache. What we might wanna do is to keep the no-merging rule for kmalloc-*, but relax it for kmalloc-cg-* caches.
But we can do it later, as a separate change.
Thanks!
Re: [PATCH v3 2/2] mm: memcg/slab: Create a new set of kmalloc-cg-<n> caches
In reply to Vlastimil BabkaQuoted 18 lines · Vlastimil Babka
Quoted 8 lines · Roman Gushchin
Quoted 7 lines · Waiman Long
With this change, all the objcg pointer array objects will come from
KMALLOC_NORMALcaches which won't have their objcg pointer arrays. So both the recursivekfree()problem and non-freeable slab problem are gone. Since both theKMALLOC_NORMALandKMALLOC_CGROUPcaches no longer have mixed accounted and unaccounted objects, this will slightly reduce the number of objcg pointer arrays that need to be allocated and save a bit of memory.Quoted 6 lines
Unfortunately the positive effect of this change will be likely reversed by a lower utilization due to a larger number of caches.
Btw, I wonder if we also need a change in the slab caches merging procedure?
KMALLOC_NORMALcaches should not be merged with caches which can potentially include accounted objects.Good point. But looks like kmalloc* caches are extempt from all merging in
create_boot_cache()via
Quoted 7 lines
s->refcount = -1; /* Exempt from merging for now */It wouldn't hurt though to create the kmalloc-cg-* caches with
SLAB_ACCOUNTflag to prevent accidental merging in case the above is ever removed. It would also better reflect reality, and ensure that the array is allocated immediately with the page, AFAICS.
I am not sure if this is really true.
struct kmem_cache *__init create_kmalloc_cache(const char *name, unsigned int size, slab_flags_t flags, unsigned int useroffset, unsigned int usersize) {
struct kmem_cache *s = kmem_cache_zalloc(kmem_cache, GFP_NOWAIT);
if (!s)
panic("Out of memory when creating slab %s\n", name);
create_boot_cache(s, name, size, flags, useroffset, usersize); kasan_cache_create_kmalloc(s);
list_add(&s->list, &slab_caches);
s->refcount = 1;
return s;
}
Even though refcount is set to -1 initially, it is set back to 1 afterward. So merging can still happen AFAICS.
Cheers,
Longman
Re: [PATCH v3 2/2] mm: memcg/slab: Create a new set of kmalloc-cg-<n> caches
Quoted 44 lines · Waiman Long
Quoted 18 lines · Vlastimil Babka
Quoted 8 lines · Roman Gushchin
Quoted 7 lines · Waiman Long
With this change, all the objcg pointer array objects will come from
KMALLOC_NORMALcaches which won't have their objcg pointer arrays. So both the recursivekfree()problem and non-freeable slab problem are gone. Since both theKMALLOC_NORMALandKMALLOC_CGROUPcaches no longer have mixed accounted and unaccounted objects, this will slightly reduce the number of objcg pointer arrays that need to be allocated and save a bit of memory.Quoted 6 lines
Unfortunately the positive effect of this change will be likely reversed by a lower utilization due to a larger number of caches.
Btw, I wonder if we also need a change in the slab caches merging procedure?
KMALLOC_NORMALcaches should not be merged with caches which can potentially include accounted objects.Good point. But looks like kmalloc* caches are extempt from all merging in
create_boot_cache()viaQuoted 7 lines
s->refcount = -1; /* Exempt from merging for now */It wouldn't hurt though to create the kmalloc-cg-* caches with
SLAB_ACCOUNTflag to prevent accidental merging in case the above is ever removed. It would also better reflect reality, and ensure that the array is allocated immediately with the page, AFAICS.I am not sure if this is really true.
struct
kmem_cache*__initcreate_kmalloc_cache(const char *name, unsigned int size,slab_flags_tflags, unsigned int useroffset, unsigned int usersize) {
structkmem_cache*s =kmem_cache_zalloc(kmem_cache, GFP_NOWAIT);if (!s)
panic("Out of memory when creating slab %s\n", name);
create_boot_cache(s, name, size, flags, useroffset, usersize);kasan_cache_create_kmalloc(s);
list_add(&s->list, &slab_caches);
s->refcount= 1;
return s;
}
Quoted 2 lines
Even though refcount is set to -1 initially, it is set back to 1 afterward. So merging can still happen AFAICS.
Right, thanks, I already noticed it. Then yeah, we should make sure we're not merging KMALLOC_NORMAL caches with any others.
Re: [PATCH v3 2/2] mm: memcg/slab: Create a new set of kmalloc-cg-<n> caches
Quoted 45 lines · Roman Gushchin
Quoted 44 lines · Waiman Long
Quoted 18 lines · Vlastimil Babka
Quoted 8 lines · Roman Gushchin
Quoted 7 lines · Waiman Long
With this change, all the objcg pointer array objects will come from
KMALLOC_NORMALcaches which won't have their objcg pointer arrays. So both the recursivekfree()problem and non-freeable slab problem are gone. Since both theKMALLOC_NORMALandKMALLOC_CGROUPcaches no longer have mixed accounted and unaccounted objects, this will slightly reduce the number of objcg pointer arrays that need to be allocated and save a bit of memory.Quoted 6 lines
Unfortunately the positive effect of this change will be likely reversed by a lower utilization due to a larger number of caches.
Btw, I wonder if we also need a change in the slab caches merging procedure?
KMALLOC_NORMALcaches should not be merged with caches which can potentially include accounted objects.Good point. But looks like kmalloc* caches are extempt from all merging in
create_boot_cache()viaQuoted 7 lines
s->refcount = -1; /* Exempt from merging for now */It wouldn't hurt though to create the kmalloc-cg-* caches with
SLAB_ACCOUNTflag to prevent accidental merging in case the above is ever removed. It would also better reflect reality, and ensure that the array is allocated immediately with the page, AFAICS.I am not sure if this is really true.
struct
kmem_cache*__initcreate_kmalloc_cache(const char *name,
unsigned int size,slab_flags_tflags, unsigned int useroffset, unsigned int usersize)
{
structkmem_cache*s =kmem_cache_zalloc(kmem_cache, GFP_NOWAIT);if (!s)
panic("Out of memory when creating slab %s\n", name);
create_boot_cache(s, name, size, flags, useroffset, usersize);kasan_cache_create_kmalloc(s);
list_add(&s->list, &slab_caches);
s->refcount= 1;
return s;
}
Quoted 5 lines
Quoted 2 lines
Even though refcount is set to -1 initially, it is set back to 1 afterward. So merging can still happen AFAICS.
Right, thanks, I already noticed it. Then yeah, we should make sure we're not merging
KMALLOC_NORMALcaches with any others.
That should be easy. We just set the refcount to -1 for the KMALLOC_NORMAL caches right after its creation then.
Cheers,
Longman
Re: [PATCH v3 2/2] mm: memcg/slab: Create a new set of kmalloc-cg-<n> caches
In reply to Vlastimil BabkaQuoted 14 lines · Vlastimil Babka
Quoted 13 lines · Roman Gushchin
Quoted 8 lines · Waiman Long
With this change, all the objcg pointer array objects will come from
KMALLOC_NORMALcaches which won't have their objcg pointer arrays. So both the recursivekfree()problem and non-freeable slab problem are gone. Since both theKMALLOC_NORMALandKMALLOC_CGROUPcaches no longer have mixed accounted and unaccounted objects, this will slightly reduce the number of objcg pointer arrays that need to be allocated and save a bit of memory.Unfortunately the positive effect of this change will be likely reversed by a lower utilization due to a larger number of caches.
Quoted 8 lines
Quoted 3 lines
Btw, I wonder if we also need a change in the slab caches merging procedure?
KMALLOC_NORMALcaches should not be merged with caches which can potentially include accounted objects.Good point. But looks like kmalloc* caches are extempt from all merging in
create_boot_cache()vias->refcount = -1; /* Exempt from merging for now */
Wait, s->refcount is adjusted to 1 in create_kmalloc_cache() after calling into create_boot_cache?
It means they are not exempt actually.
Re: [PATCH v3 2/2] mm: memcg/slab: Create a new set of kmalloc-cg-<n> caches
Quoted 19 lines · Roman Gushchin
Quoted 18 lines · Vlastimil Babka
Quoted 16 lines · Roman Gushchin
Quoted 8 lines · Waiman Long
With this change, all the objcg pointer array objects will come from
KMALLOC_NORMALcaches which won't have their objcg pointer arrays. So both the recursivekfree()problem and non-freeable slab problem are gone. Since both theKMALLOC_NORMALandKMALLOC_CGROUPcaches no longer have mixed accounted and unaccounted objects, this will slightly reduce the number of objcg pointer arrays that need to be allocated and save a bit of memory.Unfortunately the positive effect of this change will be likely reversed by a lower utilization due to a larger number of caches.
Btw, I wonder if we also need a change in the slab caches merging procedure?
KMALLOC_NORMALcaches should not be merged with caches which can potentially include accounted objects.
Quoted 7 lines
Quoted 4 lines
Good point. But looks like kmalloc* caches are extempt from all merging in
create_boot_cache()vias->refcount = -1; /* Exempt from merging for now */Wait,
s->refcountis adjusted to 1 increate_kmalloc_cache()after calling intocreate_boot_cache?
Hmm I missed that
Now I wonder why all kmalloc caches on my system have 0 aliases :) cat /sys/kernel/slab/kmalloc-*/aliases
Quoted 2 lines
It means they are not exempt actually.
Re: [PATCH v3 2/2] mm: memcg/slab: Create a new set of kmalloc-cg-<n> caches
Quoted 25 lines · Vlastimil Babka
Quoted 24 lines · Roman Gushchin
Quoted 22 lines · Vlastimil Babka
Quoted 10 lines · Roman Gushchin
Quoted 8 lines · Waiman Long
With this change, all the objcg pointer array objects will come from
KMALLOC_NORMALcaches which won't have their objcg pointer arrays. So both the recursivekfree()problem and non-freeable slab problem are gone. Since both theKMALLOC_NORMALandKMALLOC_CGROUPcaches no longer have mixed accounted and unaccounted objects, this will slightly reduce the number of objcg pointer arrays that need to be allocated and save a bit of memory.Quoted 6 lines
Unfortunately the positive effect of this change will be likely reversed by a lower utilization due to a larger number of caches.
Btw, I wonder if we also need a change in the slab caches merging procedure?
KMALLOC_NORMALcaches should not be merged with caches which can potentially include accounted objects.Good point. But looks like kmalloc* caches are extempt from all merging in
create_boot_cache()vias->refcount = -1; /* Exempt from merging for now */
Quoted 7 lines
Quoted 2 lines
Wait,
s->refcountis adjusted to 1 increate_kmalloc_cache()after calling intocreate_boot_cache?Hmm I missed that
Now I wonder why all kmalloc caches on my system have 0 aliases :) cat
/sys/kernel/slab/kmalloc-*/aliases
Yeah, I noticed it too, it's a good question. And I remember a case from the past when it wasn't true (kmalloc-32 was shared with something else).
Re: [PATCH v3 2/2] mm: memcg/slab: Create a new set of kmalloc-cg-<n> caches
Quoted 22 lines · Roman Gushchin
Quoted 21 lines · Vlastimil Babka
Quoted 20 lines · Roman Gushchin
Quoted 19 lines · Vlastimil Babka
Quoted 8 lines · Roman Gushchin
Quoted 7 lines · Waiman Long
With this change, all the objcg pointer array objects will come from
KMALLOC_NORMALcaches which won't have their objcg pointer arrays. So both the recursivekfree()problem and non-freeable slab problem are gone. Since both theKMALLOC_NORMALandKMALLOC_CGROUPcaches no longer have mixed accounted and unaccounted objects, this will slightly reduce the number of objcg pointer arrays that need to be allocated and save a bit of memory.Quoted 6 lines
Unfortunately the positive effect of this change will be likely reversed by a lower utilization due to a larger number of caches.
Btw, I wonder if we also need a change in the slab caches merging procedure?
KMALLOC_NORMALcaches should not be merged with caches which can potentially include accounted objects.Good point. But looks like kmalloc* caches are extempt from all merging in
create_boot_cache()vias->refcount = -1; /* Exempt from merging for now */
Quoted 9 lines
Quoted 6 lines
Quoted 2 lines
Wait,
s->refcountis adjusted to 1 increate_kmalloc_cache()after calling intocreate_boot_cache?Hmm I missed that
Now I wonder why all kmalloc caches on my system have 0 aliases :) cat
/sys/kernel/slab/kmalloc-*/aliasesYeah, I noticed it too, it's a good question. And I remember a case from the past when it wasn't true (kmalloc-32 was shared with something else).
The criteria for cache merging require close to exact match in all attributes with a size difference of no more than sizeof(void *). So it is not easy to find a close match.
Cheers,
Longman
Re: [PATCH v3 2/2] mm: memcg/slab: Create a new set of kmalloc-cg-<n> caches
In reply to Roman GushchinQuoted 31 lines · Roman Gushchin
Quoted 14 lines · Waiman Long
There are currently two problems in the way the objcg pointer array (
memcg_data) in the page structure is being allocated and freed.On its allocation, it is possible that the allocated objcg pointer array comes from the same slab that requires memory accounting. If this happens, the slab will never become empty again as there is at least one object left (the
obj_cgrouparray) in the slab.When it is freed, the objcg pointer array object may be the last one in its slab and hence causes
kfree()to be called again. With the right workload, the slab cache may be set up in a way that allows the recursivekfree()calling loop to nest deep enough to cause a kernel stack overflow and panic the system.Quoted 6 lines
One way to solve this problem is to split the kmalloc-<n> caches (
KMALLOC_NORMAL) into two separate sets - a new set of kmalloc-<n> (KMALLOC_NORMAL) caches for non-accounted objects only and a new set of kmalloc-cg-<n> (KMALLOC_CGROUP) caches for accounted objects only. All the other caches can still allow a mix of accounted and non-accounted objects.I agree that it's likely the best approach here. Thanks for discovering and fixing the problem!
Quoted 7 lines
With this change, all the objcg pointer array objects will come from
KMALLOC_NORMALcaches which won't have their objcg pointer arrays. So both the recursivekfree()problem and non-freeable slab problem are gone. Since both theKMALLOC_NORMALandKMALLOC_CGROUPcaches no longer have mixed accounted and unaccounted objects, this will slightly reduce the number of objcg pointer arrays that need to be allocated and save a bit of memory.
Quoted 2 lines
Unfortunately the positive effect of this change will be likely reversed by a lower utilization due to a larger number of caches.
That is also true, will mention that.
Quoted 4 lines
Btw, I wonder if we also need a change in the slab caches merging procedure?
KMALLOC_NORMALcaches should not be merged with caches which can potentially include accounted objects.
Thank for catching this omission.
I will take a look and modify the merging procedure in a new patch. Accounting is usually specified at kmem_cache_create() time. Though, I did find one instance of setting ACCOUNT flag in kmem_cache_alloc(), I will ignore this case and merge accounted, but unreclaimable caches to KMALLOC_CGROUP.
Quoted 60 lines
Quoted 60 lines
The new
KMALLOC_CGROUPis added betweenKMALLOC_NORMALandKMALLOC_RECLAIMso that the first for loop increate_kmalloc_caches()will include the newly added caches without change.Suggested-by Vlastimil Babka <[email protected]>Signed-off-by Waiman Long <[email protected]>--- include/linux/slab.h | 42 ++++++++++++++++++++++++++++++++++-------- mm/slab_common.c | 23 +++++++++++++++-------- 2 files changed, 49 insertions(+), 16 deletions(-)include/linux/slab.h +22
-1@@ -305,9 +305,16 @@ static inline void __check_heap_object(const void *ptr, unsigned long n,/* * Whenever changing this, take care of that kmalloc_type() and * create_kmalloc_caches() still work as intended. +* +* KMALLOC_NORMAL is for non-accounted objects only whereas KMALLOC_CGROUP +* is for accounted objects only. All the other kmem caches can have both +* accounted and non-accounted objects. */ enum kmalloc_cache_type {KMALLOC_NORMAL =0 , +#ifdef CONFIG_MEMCG_KMEM +KMALLOC_CGROUP , +#endif KMALLOC_RECLAIM ,#ifdef CONFIG_ZONE_DMA KMALLOC_DMA ,@@ -315,28 +322,47 @@ enum kmalloc_cache_type {NR_KMALLOC_TYPES }; +#ifndef CONFIG_MEMCG_KMEM +#define KMALLOC_CGROUP KMALLOC_NORMAL +#endif +#ifndef CONFIG_ZONE_DMA +#define KMALLOC_DMA KMALLOC_NORMAL +#endif +#ifndef CONFIG_SLOB extern struct kmem_cache * kmalloc_caches[NR_KMALLOC_TYPES ][KMALLOC_SHIFT_HIGH +1 ]; +/* +* Define gfp bits that should not be set for KMALLOC_NORMAL. +*/ +#define KMALLOC_NOT_NORMAL_BITS \ + (__GFP_RECLAIMABLE | \ + (IS_ENABLED(CONFIG_ZONE_DMA) ? __GFP_DMA : 0) | \ + (IS_ENABLED(CONFIG_MEMCG_KMEM) ? __GFP_ACCOUNT : 0)) + static __always_inline enum kmalloc_cache_type kmalloc_type(gfp_t flags) { -#ifdef CONFIG_ZONE_DMA /*
Quoted 8 lines
include/linux/slab.h enum kmalloc_cache_type { · line 351
Quoted 6 lines
include/linux/slab.h +1
-1* The most common case is KMALLOC_NORMAL, so test for it * with a single branch for both flags. */ -if (likely ((flags & (__GFP_DMA | __GFP_RECLAIMABLE)) ==0 )) +if (likely ((flags &KMALLOC_NOT_NORMAL_BITS ) ==0 ))return KMALLOC_NORMAL ;Likely
KMALLOC_CGROUPis also very popular, so maybe we want to change the optimization here a bit.
I doubt this optimization is really noticeable and whether KMALLOC_CGROUP is really popular will depend on the workloads. I am not planning to spend additional time to micro-optimize this part of the code.
Cheers,
Longman
Re: [PATCH v3 2/2] mm: memcg/slab: Create a new set of kmalloc-cg-<n> caches
Quoted 32 lines · Waiman Long
Quoted 31 lines · Roman Gushchin
Quoted 14 lines · Waiman Long
There are currently two problems in the way the objcg pointer array (
memcg_data) in the page structure is being allocated and freed.On its allocation, it is possible that the allocated objcg pointer array comes from the same slab that requires memory accounting. If this happens, the slab will never become empty again as there is at least one object left (the
obj_cgrouparray) in the slab.When it is freed, the objcg pointer array object may be the last one in its slab and hence causes
kfree()to be called again. With the right workload, the slab cache may be set up in a way that allows the recursivekfree()calling loop to nest deep enough to cause a kernel stack overflow and panic the system.Quoted 6 lines
One way to solve this problem is to split the kmalloc-<n> caches (
KMALLOC_NORMAL) into two separate sets - a new set of kmalloc-<n> (KMALLOC_NORMAL) caches for non-accounted objects only and a new set of kmalloc-cg-<n> (KMALLOC_CGROUP) caches for accounted objects only. All the other caches can still allow a mix of accounted and non-accounted objects.I agree that it's likely the best approach here. Thanks for discovering and fixing the problem!
Quoted 7 lines
With this change, all the objcg pointer array objects will come from
KMALLOC_NORMALcaches which won't have their objcg pointer arrays. So both the recursivekfree()problem and non-freeable slab problem are gone. Since both theKMALLOC_NORMALandKMALLOC_CGROUPcaches no longer have mixed accounted and unaccounted objects, this will slightly reduce the number of objcg pointer arrays that need to be allocated and save a bit of memory.
Quoted 4 lines
Quoted 2 lines
Unfortunately the positive effect of this change will be likely reversed by a lower utilization due to a larger number of caches.
That is also true, will mention that.
Thanks!
Quoted 6 lines
Quoted 5 lines
Btw, I wonder if we also need a change in the slab caches merging procedure?
KMALLOC_NORMALcaches should not be merged with caches which can potentially include accounted objects.
Quoted 7 lines
Thank for catching this omission.
I will take a look and modify the merging procedure in a new patch. Accounting is usually specified at
kmem_cache_create()time. Though, I did find one instance of setting ACCOUNT flag inkmem_cache_alloc(), I will ignore this case and merge accounted, but unreclaimable caches toKMALLOC_CGROUP.
Vlastimil pointed out that it's not an actual problem, because kmalloc caches are exempt from the merging. Please, add a comment about it into the commit log/code. We might wanna relax this rule for kmalloc-cg-*, but we can do it later.
Quoted 68 lines
Quoted 59 lines
Quoted 59 lines
The new
KMALLOC_CGROUPis added betweenKMALLOC_NORMALandKMALLOC_RECLAIMso that the first for loop increate_kmalloc_caches()will include the newly added caches without change.Suggested-by Vlastimil Babka <[email protected]>Signed-off-by Waiman Long <[email protected]>--- include/linux/slab.h | 42 ++++++++++++++++++++++++++++++++++-------- mm/slab_common.c | 23 +++++++++++++++-------- 2 files changed, 49 insertions(+), 16 deletions(-)include/linux/slab.h +22
-1@@ -305,9 +305,16 @@ static inline void __check_heap_object(const void *ptr, unsigned long n,/* * Whenever changing this, take care of that kmalloc_type() and * create_kmalloc_caches() still work as intended. +* +* KMALLOC_NORMAL is for non-accounted objects only whereas KMALLOC_CGROUP +* is for accounted objects only. All the other kmem caches can have both +* accounted and non-accounted objects. */ enum kmalloc_cache_type {KMALLOC_NORMAL =0 , +#ifdef CONFIG_MEMCG_KMEM +KMALLOC_CGROUP , +#endif KMALLOC_RECLAIM ,#ifdef CONFIG_ZONE_DMA KMALLOC_DMA ,@@ -315,28 +322,47 @@ enum kmalloc_cache_type {NR_KMALLOC_TYPES }; +#ifndef CONFIG_MEMCG_KMEM +#define KMALLOC_CGROUP KMALLOC_NORMAL +#endif +#ifndef CONFIG_ZONE_DMA +#define KMALLOC_DMA KMALLOC_NORMAL +#endif +#ifndef CONFIG_SLOB extern struct kmem_cache * kmalloc_caches[NR_KMALLOC_TYPES ][KMALLOC_SHIFT_HIGH +1 ]; +/* +* Define gfp bits that should not be set for KMALLOC_NORMAL. +*/ +#define KMALLOC_NOT_NORMAL_BITS \ + (__GFP_RECLAIMABLE | \ + (IS_ENABLED(CONFIG_ZONE_DMA) ? __GFP_DMA : 0) | \ + (IS_ENABLED(CONFIG_MEMCG_KMEM) ? __GFP_ACCOUNT : 0)) + static __always_inline enum kmalloc_cache_type kmalloc_type(gfp_t flags) { -#ifdef CONFIG_ZONE_DMA /* Quoted 8 lines
include/linux/slab.h enum kmalloc_cache_type { · line 351
Quoted 6 lines
include/linux/slab.h +1
-1* The most common case is KMALLOC_NORMAL, so test for it * with a single branch for both flags. */ -if (likely ((flags & (__GFP_DMA | __GFP_RECLAIMABLE)) ==0 )) +if (likely ((flags &KMALLOC_NOT_NORMAL_BITS ) ==0 ))return KMALLOC_NORMAL ;Likely
KMALLOC_CGROUPis also very popular, so maybe we want to change the optimization here a bit.
Quoted 3 lines
I doubt this optimization is really noticeable and whether
KMALLOC_CGROUPis really popular will depend on the workloads. I am not planning to spend additional time to micro-optimize this part of the code.
Ok.
Thanks!
Re: [PATCH v3 2/2] mm: memcg/slab: Create a new set of kmalloc-cg-<n> caches
In reply to Waiman LongQuoted 10 lines · Waiman Long
Quoted 4 lines · Roman Gushchin
Btw, I wonder if we also need a change in the slab caches merging procedure?
KMALLOC_NORMALcaches should not be merged with caches which can potentially include accounted objects.Thank for catching this omission.
I will take a look and modify the merging procedure in a new patch. Accounting is usually specified at
kmem_cache_create()time. Though, I did find one instance of setting ACCOUNT flag inkmem_cache_alloc(), I will ignore this case and merge accounted, but unreclaimable caches toKMALLOC_CGROUP.
In mm/slab_common.c:
#define SLAB_MERGE_SAME (SLAB_RECLAIM_ACCOUNT | SLAB_CACHE_DMA | \
SLAB_CACHE_DMA32 | SLAB_ACCOUNT)
struct kmem_cache *find_mergeable(unsigned int size, unsigned int align, :
if ((flags & SLAB_MERGE_SAME) != (s->flags & SLAB_MERGE_SAME)) continue;
By making sure kmalloc-cg-* has SLAB_ACCOUNT bit set, a kmemcache created with with SLAB_ACCOUNT may merge with kmalloc-cg-* whereas one without SLAB_ACCOUNT may merge with kmalloc-* for now. So the current code should work fine for most cases. Though, if the ACCOUNT flag is set at kmem_cache_alloc() and the cache happens to be merged into kmalloc-*, we will have the rare case that an objcg pointer array may have to be added to a kmalloc-* cache. However, this is not a common practice, and the three cases (not one, sorry) that I found so far is in
arch/x86/kvm/x86.c: ctxt = kmem_cache_zalloc(x86_emulator_cache, GFP_KERNEL_ACCOUNT); fs/hostfs/hostfs_kern.c: hi = kmem_cache_alloc(hostfs_inode_cache, GFP_KERNEL_ACCOUNT); virt/kvm/kvm_main.c: vcpu = kmem_cache_zalloc(kvm_vcpu_cache, GFP_KERNEL_ACCOUNT);
We will have to advise against doing that.
Cheers,
Longman