[PATCH 0/4] memcg: shrink obj_stock_pcp and cache multiple objcgs

[PATCH 0/4] memcg: shrink obj_stock_pcp and cache multiple objcgs

[Shakeel Butt]

Commit 01b9da291c49 ("mm: memcontrol: convert objcg to be per-memcg
per-node type") split a memcg's single obj_cgroup into one per NUMA
node so that reparenting LRU folios can take per-node lru locks. As a
side effect, the per-CPU obj_stock_pcp -- which caches a single
cached_objcg pointer -- thrashes on workloads where threads of the
same memcg run on different NUMA nodes. The kernel test robot reported
a 67.7% regression on stress-ng.switch.ops_per_sec from this pattern.

Commit d0211878ce06 ("memcg: cache obj_stock by memcg, not by objcg
pointer") landed as a temporary fix by treating sibling per-node
objcgs as equivalent for the cache lookup, intended to be reverted
once per-node kmem accounting is introduced. This series takes a more
general approach: cache multiple objcgs per CPU using the multi-slot
pattern memcg_stock_pcp already uses, so the per-node objcg variants
of one memcg can all coexist in the stock without ever forcing a
drain. The temporary fix can then be reverted.

To avoid increasing the per-CPU cache footprint, the first three
patches shrink the existing single-slot obj_stock_pcp fields.
The final patch converts cached_objcg and nr_bytes into
NR_OBJ_STOCK=5 slot arrays and reorders the struct so the entire
consume/refill/account hot path fits within a single 64-byte cache
line on non-debug 64-bit builds (verified with pahole).

Reported-by: kernel test robot <oliver.sang@intel.com>
Closes: https://lore.kernel.org/oe-lkp/202605121641.b6a60cb0-lkp@intel.com
Fixes: 01b9da291c49 ("mm: memcontrol: convert objcg to be per-memcg per-node type")
Tested-by: kernel test robot <oliver.sang@intel.com>

Shakeel Butt (4):
  memcg: store node_id instead of pglist_data pointer
  memcg: uint16_t for nr_bytes in obj_stock_pcp
  memcg: int16_t for cached slab stats
  memcg: multi objcg charge support

 mm/memcontrol.c | 214 +++++++++++++++++++++++++++++++++++-------------
 1 file changed, 157 insertions(+), 57 deletions(-)

--
2.53.0-Meta

[PATCH 1/4] memcg: store node_id instead of pglist_data pointer

[Shakeel Butt]

The struct obj_stock_pcp stores a pointer to pglist_data for the slab
stats cached on the cpu. On 64-bit machines, this costs 8 bytes. The
pointer is not strictly required: NODE_DATA() can recover it from the
node id. Replace cached_pgdat with int16_t node_id and use NUMA_NO_NODE
as the "no stats cached" sentinel.

At the moment all the archs limit MAX_NUMNODES to 1024 so int16_t is
plenty; a BUILD_BUG_ON() makes sure we notice if that ever changes.

Signed-off-by: Shakeel Butt <shakeel.butt@linux.dev>
Tested-by: kernel test robot <oliver.sang@intel.com>
---
 mm/memcontrol.c | 26 +++++++++++++++++++-------
 1 file changed, 19 insertions(+), 7 deletions(-)

diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index b8caeb7ccaa3..d7c162946719 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -2021,7 +2021,7 @@ struct obj_stock_pcp {
 	local_trylock_t lock;
 	unsigned int nr_bytes;
 	struct obj_cgroup *cached_objcg;
-	struct pglist_data *cached_pgdat;
+	int16_t node_id;
 	int nr_slab_reclaimable_b;
 	int nr_slab_unreclaimable_b;
 
@@ -2031,6 +2031,7 @@ struct obj_stock_pcp {
 
 static DEFINE_PER_CPU_ALIGNED(struct obj_stock_pcp, obj_stock) = {
 	.lock = INIT_LOCAL_TRYLOCK(lock),
+	.node_id = NUMA_NO_NODE,
 };
 
 static DEFINE_MUTEX(percpu_charge_mutex);
@@ -3159,6 +3160,13 @@ static void __account_obj_stock(struct obj_cgroup *objcg,
 {
 	int *bytes;
 
+	/*
+	 * Though at the moment MAX_NUMNODES <= 1024 in all archs but let's make
+	 * sure it does not exceed S16_MAX otherwise we need to fix node_id type
+	 * in struct obj_stock_pcp.
+	 */
+	BUILD_BUG_ON(MAX_NUMNODES >= S16_MAX);
+
 	if (!stock || READ_ONCE(stock->cached_objcg) != objcg)
 		goto direct;
 
@@ -3166,9 +3174,11 @@ static void __account_obj_stock(struct obj_cgroup *objcg,
 	 * Save vmstat data in stock and skip vmstat array update unless
 	 * accumulating over a page of vmstat data or when pgdat changes.
 	 */
-	if (stock->cached_pgdat != pgdat) {
+	if (stock->node_id == NUMA_NO_NODE) {
+		stock->node_id = pgdat->node_id;
+	} else if (stock->node_id != pgdat->node_id) {
 		/* Flush the existing cached vmstat data */
-		struct pglist_data *oldpg = stock->cached_pgdat;
+		struct pglist_data *oldpg = NODE_DATA(stock->node_id);
 
 		if (stock->nr_slab_reclaimable_b) {
 			mod_objcg_mlstate(objcg, oldpg, NR_SLAB_RECLAIMABLE_B,
@@ -3180,7 +3190,7 @@ static void __account_obj_stock(struct obj_cgroup *objcg,
 					  stock->nr_slab_unreclaimable_b);
 			stock->nr_slab_unreclaimable_b = 0;
 		}
-		stock->cached_pgdat = pgdat;
+		stock->node_id = pgdat->node_id;
 	}
 
 	bytes = (idx == NR_SLAB_RECLAIMABLE_B) ? &stock->nr_slab_reclaimable_b
@@ -3276,19 +3286,21 @@ static void drain_obj_stock(struct obj_stock_pcp *stock)
 	 * Flush the vmstat data in current stock
 	 */
 	if (stock->nr_slab_reclaimable_b || stock->nr_slab_unreclaimable_b) {
+		struct pglist_data *oldpg = NODE_DATA(stock->node_id);
+
 		if (stock->nr_slab_reclaimable_b) {
-			mod_objcg_mlstate(old, stock->cached_pgdat,
+			mod_objcg_mlstate(old, oldpg,
 					  NR_SLAB_RECLAIMABLE_B,
 					  stock->nr_slab_reclaimable_b);
 			stock->nr_slab_reclaimable_b = 0;
 		}
 		if (stock->nr_slab_unreclaimable_b) {
-			mod_objcg_mlstate(old, stock->cached_pgdat,
+			mod_objcg_mlstate(old, oldpg,
 					  NR_SLAB_UNRECLAIMABLE_B,
 					  stock->nr_slab_unreclaimable_b);
 			stock->nr_slab_unreclaimable_b = 0;
 		}
-		stock->cached_pgdat = NULL;
+		stock->node_id = NUMA_NO_NODE;
 	}
 
 	WRITE_ONCE(stock->cached_objcg, NULL);
-- 
2.53.0-Meta

Re: [PATCH 1/4] memcg: store node_id instead of pglist_data pointer

[Harry Yoo]

On 5/20/26 2:31 PM, Shakeel Butt wrote:
> The struct obj_stock_pcp stores a pointer to pglist_data for the slab
> stats cached on the cpu. On 64-bit machines, this costs 8 bytes. The
> pointer is not strictly required: NODE_DATA() can recover it from the
> node id. Replace cached_pgdat with int16_t node_id and use NUMA_NO_NODE
> as the "no stats cached" sentinel.
> 
> At the moment all the archs limit MAX_NUMNODES to 1024 so int16_t is
> plenty; a BUILD_BUG_ON() makes sure we notice if that ever changes.
> 
> Signed-off-by: Shakeel Butt <shakeel.butt@linux.dev>
> Tested-by: kernel test robot <oliver.sang@intel.com>
> ---

Reviewed-by: Harry Yoo (Oracle) <harry@kernel.org>

-- 
Cheers,
Harry / Hyeonggon

Re: [PATCH 1/4] memcg: store node_id instead of pglist_data pointer

[Muchun Song]

> On May 20, 2026, at 13:31, Shakeel Butt <shakeel.butt@linux.dev> wrote:
> 
> The struct obj_stock_pcp stores a pointer to pglist_data for the slab
> stats cached on the cpu. On 64-bit machines, this costs 8 bytes. The
> pointer is not strictly required: NODE_DATA() can recover it from the
> node id. Replace cached_pgdat with int16_t node_id and use NUMA_NO_NODE
> as the "no stats cached" sentinel.
> 
> At the moment all the archs limit MAX_NUMNODES to 1024 so int16_t is
> plenty; a BUILD_BUG_ON() makes sure we notice if that ever changes.
> 
> Signed-off-by: Shakeel Butt <shakeel.butt@linux.dev>
> Tested-by: kernel test robot <oliver.sang@intel.com>

Acked-by: Muchun Song <muchun.song@linux.dev>

Thanks.

[PATCH 2/4] memcg: uint16_t for nr_bytes in obj_stock_pcp

[Shakeel Butt]

Currently struct obj_stock_pcp stores nr_bytes in an 'unsigned int'
which is 4 bytes on 64-bit machines. Switch the field to uint16_t to
shrink the per-CPU cache.

The kernel supports PAGE_SIZE_4KB, _8KB, _16KB, _32KB, _64KB and
_256KB (see HAVE_PAGE_SIZE_* in arch/Kconfig). After the
PAGE_SIZE-aligned flush in __refill_obj_stock(), the sub-page
remainder fits in uint16_t up through 64KiB pages where PAGE_SIZE - 1
== U16_MAX, but on 256KiB pages PAGE_SIZE - 1 == 0x3FFFF exceeds
U16_MAX. The accumulator also needs to stay within uint16_t between
page-aligned flushes on 64KiB pages where PAGE_SIZE itself is
U16_MAX + 1.

Accumulate the new total in an 'unsigned int' local, then:

  1. Flush whenever the accumulator would hit U16_MAX. Together with
     the existing allow_uncharge flush at PAGE_SIZE, this keeps the
     uint16_t safe on PAGE_SIZE <= 64KiB.

  2. On configs with PAGE_SHIFT > 16 (PAGE_SIZE_256KB on hexagon and
     powerpc 44x), push any sub-page remainder above U16_MAX into
     objcg->nr_charged_bytes via atomic_add before storing back, so
     the store cannot silently truncate. The PAGE_SHIFT > 16 guard
     folds the branch out at compile time on smaller page sizes.

Signed-off-by: Shakeel Butt <shakeel.butt@linux.dev>
Tested-by: kernel test robot <oliver.sang@intel.com>
---
 mm/memcontrol.c | 33 +++++++++++++++++++++++++++------
 1 file changed, 27 insertions(+), 6 deletions(-)

diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index d7c162946719..b3d63d9f267c 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -2019,7 +2019,7 @@ static DEFINE_PER_CPU_ALIGNED(struct memcg_stock_pcp, memcg_stock) = {
 
 struct obj_stock_pcp {
 	local_trylock_t lock;
-	unsigned int nr_bytes;
+	uint16_t nr_bytes;
 	struct obj_cgroup *cached_objcg;
 	int16_t node_id;
 	int nr_slab_reclaimable_b;
@@ -3331,6 +3331,7 @@ static void __refill_obj_stock(struct obj_cgroup *objcg,
 			       bool allow_uncharge)
 {
 	unsigned int nr_pages = 0;
+	unsigned int stock_nr_bytes;
 
 	if (!stock) {
 		nr_pages = nr_bytes >> PAGE_SHIFT;
@@ -3339,21 +3340,41 @@ static void __refill_obj_stock(struct obj_cgroup *objcg,
 		goto out;
 	}
 
+	stock_nr_bytes = stock->nr_bytes;
 	if (READ_ONCE(stock->cached_objcg) != objcg) { /* reset if necessary */
 		drain_obj_stock(stock);
 		obj_cgroup_get(objcg);
-		stock->nr_bytes = atomic_read(&objcg->nr_charged_bytes)
+		stock_nr_bytes = atomic_read(&objcg->nr_charged_bytes)
 				? atomic_xchg(&objcg->nr_charged_bytes, 0) : 0;
 		WRITE_ONCE(stock->cached_objcg, objcg);
 
 		allow_uncharge = true;	/* Allow uncharge when objcg changes */
 	}
-	stock->nr_bytes += nr_bytes;
+	stock_nr_bytes += nr_bytes;
+
+	/* Since stock->nr_bytes is uint16_t, don't refill >= U16_MAX */
+	if ((allow_uncharge && (stock_nr_bytes > PAGE_SIZE)) ||
+	    stock_nr_bytes >= U16_MAX) {
+		nr_pages = stock_nr_bytes >> PAGE_SHIFT;
+		stock_nr_bytes &= (PAGE_SIZE - 1);
+
+		/*
+		 * On configs with PAGE_SHIFT > 16 (PAGE_SIZE_256KB on
+		 * hexagon and powerpc 44x), the sub-page remainder can
+		 * still exceed U16_MAX. Push the excess back to
+		 * objcg->nr_charged_bytes so the store into uint16_t
+		 * cannot silently truncate; folded out at compile time
+		 * on smaller page sizes.
+		 */
+		if (PAGE_SHIFT > 16 && stock_nr_bytes > U16_MAX) {
+			unsigned int kept = stock_nr_bytes & U16_MAX;
 
-	if (allow_uncharge && (stock->nr_bytes > PAGE_SIZE)) {
-		nr_pages = stock->nr_bytes >> PAGE_SHIFT;
-		stock->nr_bytes &= (PAGE_SIZE - 1);
+			atomic_add(stock_nr_bytes - kept,
+				   &objcg->nr_charged_bytes);
+			stock_nr_bytes = kept;
+		}
 	}
+	stock->nr_bytes = stock_nr_bytes;
 
 out:
 	if (nr_pages)
-- 
2.53.0-Meta

Re: [PATCH 2/4] memcg: uint16_t for nr_bytes in obj_stock_pcp

[Harry Yoo]

On 5/20/26 2:31 PM, Shakeel Butt wrote:
> Currently struct obj_stock_pcp stores nr_bytes in an 'unsigned int'
> which is 4 bytes on 64-bit machines. Switch the field to uint16_t to
> shrink the per-CPU cache.
> 
> The kernel supports PAGE_SIZE_4KB, _8KB, _16KB, _32KB, _64KB and
> _256KB (see HAVE_PAGE_SIZE_* in arch/Kconfig). After the
> PAGE_SIZE-aligned flush in __refill_obj_stock(), the sub-page
> remainder fits in uint16_t up through 64KiB pages where PAGE_SIZE - 1
> == U16_MAX, but on 256KiB pages PAGE_SIZE - 1 == 0x3FFFF exceeds
> U16_MAX. The accumulator also needs to stay within uint16_t between
> page-aligned flushes on 64KiB pages where PAGE_SIZE itself is
> U16_MAX + 1.
> 
> Accumulate the new total in an 'unsigned int' local, then:
> 
>    1. Flush whenever the accumulator would hit U16_MAX. Together with
>       the existing allow_uncharge flush at PAGE_SIZE, this keeps the
>       uint16_t safe on PAGE_SIZE <= 64KiB.
> 
>    2. On configs with PAGE_SHIFT > 16 (PAGE_SIZE_256KB on hexagon and
>       powerpc 44x), push any sub-page remainder above U16_MAX into
>       objcg->nr_charged_bytes via atomic_add before storing back, so
>       the store cannot silently truncate. The PAGE_SHIFT > 16 guard
>       folds the branch out at compile time on smaller page sizes.
> 
> Signed-off-by: Shakeel Butt <shakeel.butt@linux.dev>
> Tested-by: kernel test robot <oliver.sang@intel.com>
> ---

Reviewed-by: Harry Yoo (Oracle) <harry@kernel.org>

-- 
Cheers,
Harry / Hyeonggon

Re: [PATCH 2/4] memcg: uint16_t for nr_bytes in obj_stock_pcp

[Harry Yoo]

On 5/20/26 2:31 PM, Shakeel Butt wrote:
> Currently struct obj_stock_pcp stores nr_bytes in an 'unsigned int'
> which is 4 bytes on 64-bit machines. Switch the field to uint16_t to
> shrink the per-CPU cache.
> 
> The kernel supports PAGE_SIZE_4KB, _8KB, _16KB, _32KB, _64KB and
> _256KB (see HAVE_PAGE_SIZE_* in arch/Kconfig). After the
> PAGE_SIZE-aligned flush in __refill_obj_stock(), the sub-page
> remainder fits in uint16_t up through 64KiB pages where PAGE_SIZE - 1
> == U16_MAX, but on 256KiB pages PAGE_SIZE - 1 == 0x3FFFF exceeds
> U16_MAX. The accumulator also needs to stay within uint16_t between
> page-aligned flushes on 64KiB pages where PAGE_SIZE itself is
> U16_MAX + 1.
> 
> Accumulate the new total in an 'unsigned int' local, then:
> 
>    1. Flush whenever the accumulator would hit U16_MAX. Together with
>       the existing allow_uncharge flush at PAGE_SIZE, this keeps the
>       uint16_t safe on PAGE_SIZE <= 64KiB.
> 
>    2. On configs with PAGE_SHIFT > 16 (PAGE_SIZE_256KB on hexagon and
>       powerpc 44x), push any sub-page remainder above U16_MAX into
>       objcg->nr_charged_bytes via atomic_add before storing back, so
>       the store cannot silently truncate. The PAGE_SHIFT > 16 guard
>       folds the branch out at compile time on smaller page sizes.
> 
> Signed-off-by: Shakeel Butt <shakeel.butt@linux.dev>
> Tested-by: kernel test robot <oliver.sang@intel.com>
> ---
>   mm/memcontrol.c | 33 +++++++++++++++++++++++++++------
>   1 file changed, 27 insertions(+), 6 deletions(-)
> 
> diff --git a/mm/memcontrol.c b/mm/memcontrol.c
> index d7c162946719..b3d63d9f267c 100644
> --- a/mm/memcontrol.c
> +++ b/mm/memcontrol.c
> @@ -3339,21 +3340,41 @@ static void __refill_obj_stock(struct obj_cgroup *objcg,
>   		goto out;
>   	}
>   
> +	stock_nr_bytes = stock->nr_bytes;
>   	if (READ_ONCE(stock->cached_objcg) != objcg) { /* reset if necessary */
>   		drain_obj_stock(stock);
>   		obj_cgroup_get(objcg);
> -		stock->nr_bytes = atomic_read(&objcg->nr_charged_bytes)
> +		stock_nr_bytes = atomic_read(&objcg->nr_charged_bytes)
>   				? atomic_xchg(&objcg->nr_charged_bytes, 0) : 0;
>   		WRITE_ONCE(stock->cached_objcg, objcg);
>   
>   		allow_uncharge = true;	/* Allow uncharge when objcg changes */
>   	}
> -	stock->nr_bytes += nr_bytes;
> +	stock_nr_bytes += nr_bytes;
> +
> +	/* Since stock->nr_bytes is uint16_t, don't refill >= U16_MAX */
> +	if ((allow_uncharge && (stock_nr_bytes > PAGE_SIZE)) ||
> +	    stock_nr_bytes >= U16_MAX) {

nit: This should be > U16_MAX?

> +		nr_pages = stock_nr_bytes >> PAGE_SHIFT;
> +		stock_nr_bytes &= (PAGE_SIZE - 1);
> +
> +		/*
> +		 * On configs with PAGE_SHIFT > 16 (PAGE_SIZE_256KB on
> +		 * hexagon and powerpc 44x), the sub-page remainder can
> +		 * still exceed U16_MAX. Push the excess back to
> +		 * objcg->nr_charged_bytes so the store into uint16_t
> +		 * cannot silently truncate; folded out at compile time
> +		 * on smaller page sizes.
> +		 */
> +		if (PAGE_SHIFT > 16 && stock_nr_bytes > U16_MAX) {
> +			unsigned int kept = stock_nr_bytes & U16_MAX;
>   
> -	if (allow_uncharge && (stock->nr_bytes > PAGE_SIZE)) {
> -		nr_pages = stock->nr_bytes >> PAGE_SHIFT;
> -		stock->nr_bytes &= (PAGE_SIZE - 1);
> +			atomic_add(stock_nr_bytes - kept,
> +				   &objcg->nr_charged_bytes);
> +			stock_nr_bytes = kept;
> +		}
>   	}
> +	stock->nr_bytes = stock_nr_bytes;
>   
>   out:
>   	if (nr_pages)

-- 
Cheers,
Harry / Hyeonggon

Re: [PATCH 2/4] memcg: uint16_t for nr_bytes in obj_stock_pcp

[Shakeel Butt]

On Wed, May 20, 2026 at 04:01:45PM +0900, Harry Yoo wrote:
> 
> 
> On 5/20/26 2:31 PM, Shakeel Butt wrote:
> > Currently struct obj_stock_pcp stores nr_bytes in an 'unsigned int'
> > which is 4 bytes on 64-bit machines. Switch the field to uint16_t to
> > shrink the per-CPU cache.
> > 
> > The kernel supports PAGE_SIZE_4KB, _8KB, _16KB, _32KB, _64KB and
> > _256KB (see HAVE_PAGE_SIZE_* in arch/Kconfig). After the
> > PAGE_SIZE-aligned flush in __refill_obj_stock(), the sub-page
> > remainder fits in uint16_t up through 64KiB pages where PAGE_SIZE - 1
> > == U16_MAX, but on 256KiB pages PAGE_SIZE - 1 == 0x3FFFF exceeds
> > U16_MAX. The accumulator also needs to stay within uint16_t between
> > page-aligned flushes on 64KiB pages where PAGE_SIZE itself is
> > U16_MAX + 1.
> > 
> > Accumulate the new total in an 'unsigned int' local, then:
> > 
> >    1. Flush whenever the accumulator would hit U16_MAX. Together with
> >       the existing allow_uncharge flush at PAGE_SIZE, this keeps the
> >       uint16_t safe on PAGE_SIZE <= 64KiB.
> > 
> >    2. On configs with PAGE_SHIFT > 16 (PAGE_SIZE_256KB on hexagon and
> >       powerpc 44x), push any sub-page remainder above U16_MAX into
> >       objcg->nr_charged_bytes via atomic_add before storing back, so
> >       the store cannot silently truncate. The PAGE_SHIFT > 16 guard
> >       folds the branch out at compile time on smaller page sizes.
> > 
> > Signed-off-by: Shakeel Butt <shakeel.butt@linux.dev>
> > Tested-by: kernel test robot <oliver.sang@intel.com>
> > ---
> >   mm/memcontrol.c | 33 +++++++++++++++++++++++++++------
> >   1 file changed, 27 insertions(+), 6 deletions(-)
> > 
> > diff --git a/mm/memcontrol.c b/mm/memcontrol.c
> > index d7c162946719..b3d63d9f267c 100644
> > --- a/mm/memcontrol.c
> > +++ b/mm/memcontrol.c
> > @@ -3339,21 +3340,41 @@ static void __refill_obj_stock(struct obj_cgroup *objcg,
> >   		goto out;
> >   	}
> > +	stock_nr_bytes = stock->nr_bytes;
> >   	if (READ_ONCE(stock->cached_objcg) != objcg) { /* reset if necessary */
> >   		drain_obj_stock(stock);
> >   		obj_cgroup_get(objcg);
> > -		stock->nr_bytes = atomic_read(&objcg->nr_charged_bytes)
> > +		stock_nr_bytes = atomic_read(&objcg->nr_charged_bytes)
> >   				? atomic_xchg(&objcg->nr_charged_bytes, 0) : 0;
> >   		WRITE_ONCE(stock->cached_objcg, objcg);
> >   		allow_uncharge = true;	/* Allow uncharge when objcg changes */
> >   	}
> > -	stock->nr_bytes += nr_bytes;
> > +	stock_nr_bytes += nr_bytes;
> > +
> > +	/* Since stock->nr_bytes is uint16_t, don't refill >= U16_MAX */
> > +	if ((allow_uncharge && (stock_nr_bytes > PAGE_SIZE)) ||
> > +	    stock_nr_bytes >= U16_MAX) {
> 
> nit: This should be > U16_MAX?

Ack.

Re: [PATCH 2/4] memcg: uint16_t for nr_bytes in obj_stock_pcp

[David Laight]

On Tue, 19 May 2026 22:31:20 -0700
Shakeel Butt <shakeel.butt@linux.dev> wrote:

> Currently struct obj_stock_pcp stores nr_bytes in an 'unsigned int'
> which is 4 bytes on 64-bit machines. Switch the field to uint16_t to
> shrink the per-CPU cache.
> 
> The kernel supports PAGE_SIZE_4KB, _8KB, _16KB, _32KB, _64KB and
> _256KB (see HAVE_PAGE_SIZE_* in arch/Kconfig). After the
> PAGE_SIZE-aligned flush in __refill_obj_stock(), the sub-page
> remainder fits in uint16_t up through 64KiB pages where PAGE_SIZE - 1
> == U16_MAX, but on 256KiB pages PAGE_SIZE - 1 == 0x3FFFF exceeds
> U16_MAX. The accumulator also needs to stay within uint16_t between
> page-aligned flushes on 64KiB pages where PAGE_SIZE itself is
> U16_MAX + 1.
> 
> Accumulate the new total in an 'unsigned int' local, then:
> 
>   1. Flush whenever the accumulator would hit U16_MAX. Together with
>      the existing allow_uncharge flush at PAGE_SIZE, this keeps the
>      uint16_t safe on PAGE_SIZE <= 64KiB.
> 
>   2. On configs with PAGE_SHIFT > 16 (PAGE_SIZE_256KB on hexagon and
>      powerpc 44x), push any sub-page remainder above U16_MAX into
>      objcg->nr_charged_bytes via atomic_add before storing back, so
>      the store cannot silently truncate. The PAGE_SHIFT > 16 guard
>      folds the branch out at compile time on smaller page sizes.
> 
> Signed-off-by: Shakeel Butt <shakeel.butt@linux.dev>
> Tested-by: kernel test robot <oliver.sang@intel.com>
> ---
>  mm/memcontrol.c | 33 +++++++++++++++++++++++++++------
>  1 file changed, 27 insertions(+), 6 deletions(-)
> 
> diff --git a/mm/memcontrol.c b/mm/memcontrol.c
> index d7c162946719..b3d63d9f267c 100644
> --- a/mm/memcontrol.c
> +++ b/mm/memcontrol.c
> @@ -2019,7 +2019,7 @@ static DEFINE_PER_CPU_ALIGNED(struct memcg_stock_pcp, memcg_stock) = {
>  
>  struct obj_stock_pcp {
>  	local_trylock_t lock;
> -	unsigned int nr_bytes;
> +	uint16_t nr_bytes;
>  	struct obj_cgroup *cached_objcg;
>  	int16_t node_id;

You might want to move it to this hole.
The size of 'lock' depends on kernel build options.

-- David

>  	int nr_slab_reclaimable_b;
> @@ -3331,6 +3331,7 @@ static void __refill_obj_stock(struct obj_cgroup *objcg,
>  			       bool allow_uncharge)
>  {
>  	unsigned int nr_pages = 0;
> +	unsigned int stock_nr_bytes;
>  
>  	if (!stock) {
>  		nr_pages = nr_bytes >> PAGE_SHIFT;
> @@ -3339,21 +3340,41 @@ static void __refill_obj_stock(struct obj_cgroup *objcg,
>  		goto out;
>  	}
>  
> +	stock_nr_bytes = stock->nr_bytes;
>  	if (READ_ONCE(stock->cached_objcg) != objcg) { /* reset if necessary */
>  		drain_obj_stock(stock);
>  		obj_cgroup_get(objcg);
> -		stock->nr_bytes = atomic_read(&objcg->nr_charged_bytes)
> +		stock_nr_bytes = atomic_read(&objcg->nr_charged_bytes)
>  				? atomic_xchg(&objcg->nr_charged_bytes, 0) : 0;
>  		WRITE_ONCE(stock->cached_objcg, objcg);
>  
>  		allow_uncharge = true;	/* Allow uncharge when objcg changes */
>  	}
> -	stock->nr_bytes += nr_bytes;
> +	stock_nr_bytes += nr_bytes;
> +
> +	/* Since stock->nr_bytes is uint16_t, don't refill >= U16_MAX */
> +	if ((allow_uncharge && (stock_nr_bytes > PAGE_SIZE)) ||
> +	    stock_nr_bytes >= U16_MAX) {
> +		nr_pages = stock_nr_bytes >> PAGE_SHIFT;
> +		stock_nr_bytes &= (PAGE_SIZE - 1);
> +
> +		/*
> +		 * On configs with PAGE_SHIFT > 16 (PAGE_SIZE_256KB on
> +		 * hexagon and powerpc 44x), the sub-page remainder can
> +		 * still exceed U16_MAX. Push the excess back to
> +		 * objcg->nr_charged_bytes so the store into uint16_t
> +		 * cannot silently truncate; folded out at compile time
> +		 * on smaller page sizes.
> +		 */
> +		if (PAGE_SHIFT > 16 && stock_nr_bytes > U16_MAX) {
> +			unsigned int kept = stock_nr_bytes & U16_MAX;
>  
> -	if (allow_uncharge && (stock->nr_bytes > PAGE_SIZE)) {
> -		nr_pages = stock->nr_bytes >> PAGE_SHIFT;
> -		stock->nr_bytes &= (PAGE_SIZE - 1);
> +			atomic_add(stock_nr_bytes - kept,
> +				   &objcg->nr_charged_bytes);
> +			stock_nr_bytes = kept;
> +		}
>  	}
> +	stock->nr_bytes = stock_nr_bytes;
>  
>  out:
>  	if (nr_pages)

Re: [PATCH 2/4] memcg: uint16_t for nr_bytes in obj_stock_pcp

[Shakeel Butt]

On Wed, May 20, 2026 at 02:20:23PM +0100, David Laight wrote:
> On Tue, 19 May 2026 22:31:20 -0700
> Shakeel Butt <shakeel.butt@linux.dev> wrote:
> 
> > Currently struct obj_stock_pcp stores nr_bytes in an 'unsigned int'
> > which is 4 bytes on 64-bit machines. Switch the field to uint16_t to
> > shrink the per-CPU cache.
> > 
> > The kernel supports PAGE_SIZE_4KB, _8KB, _16KB, _32KB, _64KB and
> > _256KB (see HAVE_PAGE_SIZE_* in arch/Kconfig). After the
> > PAGE_SIZE-aligned flush in __refill_obj_stock(), the sub-page
> > remainder fits in uint16_t up through 64KiB pages where PAGE_SIZE - 1
> > == U16_MAX, but on 256KiB pages PAGE_SIZE - 1 == 0x3FFFF exceeds
> > U16_MAX. The accumulator also needs to stay within uint16_t between
> > page-aligned flushes on 64KiB pages where PAGE_SIZE itself is
> > U16_MAX + 1.
> > 
> > Accumulate the new total in an 'unsigned int' local, then:
> > 
> >   1. Flush whenever the accumulator would hit U16_MAX. Together with
> >      the existing allow_uncharge flush at PAGE_SIZE, this keeps the
> >      uint16_t safe on PAGE_SIZE <= 64KiB.
> > 
> >   2. On configs with PAGE_SHIFT > 16 (PAGE_SIZE_256KB on hexagon and
> >      powerpc 44x), push any sub-page remainder above U16_MAX into
> >      objcg->nr_charged_bytes via atomic_add before storing back, so
> >      the store cannot silently truncate. The PAGE_SHIFT > 16 guard
> >      folds the branch out at compile time on smaller page sizes.
> > 
> > Signed-off-by: Shakeel Butt <shakeel.butt@linux.dev>
> > Tested-by: kernel test robot <oliver.sang@intel.com>
> > ---
> >  mm/memcontrol.c | 33 +++++++++++++++++++++++++++------
> >  1 file changed, 27 insertions(+), 6 deletions(-)
> > 
> > diff --git a/mm/memcontrol.c b/mm/memcontrol.c
> > index d7c162946719..b3d63d9f267c 100644
> > --- a/mm/memcontrol.c
> > +++ b/mm/memcontrol.c
> > @@ -2019,7 +2019,7 @@ static DEFINE_PER_CPU_ALIGNED(struct memcg_stock_pcp, memcg_stock) = {
> >  
> >  struct obj_stock_pcp {
> >  	local_trylock_t lock;
> > -	unsigned int nr_bytes;
> > +	uint16_t nr_bytes;
> >  	struct obj_cgroup *cached_objcg;
> >  	int16_t node_id;
> 
> You might want to move it to this hole.
> The size of 'lock' depends on kernel build options.

Thanks. In the final patch, I am rearranging the fields for better packing.
Please take a look at 4th patch and see if it still need fixing.

[PATCH 3/4] memcg: int16_t for cached slab stats

[Shakeel Butt]

Currently struct obj_stock_pcp stores cached slab stats in 'int' which
is 4 bytes per counter on 64-bit machines. Switch them to int16_t to
shrink the cached metadata.

The existing PAGE_SIZE flush in __account_obj_stock() bounds *bytes at
PAGE_SIZE on 4KiB and 16KiB page archs, well within int16_t. On 64KiB
pages PAGE_SIZE is well above S16_MAX so that flush never fires, and a
sufficiently long run of accumulations would overflow the cache. Add
an explicit S16_MAX guard before each add: when the next add would
push abs(*bytes) past S16_MAX, fold the cached value into @nr and
flush directly via mod_objcg_mlstate() before the accumulation.

Signed-off-by: Shakeel Butt <shakeel.butt@linux.dev>
Tested-by: kernel test robot <oliver.sang@intel.com>
---
 mm/memcontrol.c | 16 +++++++++++++---
 1 file changed, 13 insertions(+), 3 deletions(-)

diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index b3d63d9f267c..1ed27fd06850 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -2022,8 +2022,8 @@ struct obj_stock_pcp {
 	uint16_t nr_bytes;
 	struct obj_cgroup *cached_objcg;
 	int16_t node_id;
-	int nr_slab_reclaimable_b;
-	int nr_slab_unreclaimable_b;
+	int16_t nr_slab_reclaimable_b;
+	int16_t nr_slab_unreclaimable_b;
 
 	struct work_struct work;
 	unsigned long flags;
@@ -3158,7 +3158,7 @@ static void __account_obj_stock(struct obj_cgroup *objcg,
 				struct obj_stock_pcp *stock, int nr,
 				struct pglist_data *pgdat, enum node_stat_item idx)
 {
-	int *bytes;
+	int16_t *bytes;
 
 	/*
 	 * Though at the moment MAX_NUMNODES <= 1024 in all archs but let's make
@@ -3195,6 +3195,16 @@ static void __account_obj_stock(struct obj_cgroup *objcg,
 
 	bytes = (idx == NR_SLAB_RECLAIMABLE_B) ? &stock->nr_slab_reclaimable_b
 					       : &stock->nr_slab_unreclaimable_b;
+	/*
+	 * To avoid overflow or underflow, flush directly if accumulating @nr
+	 * would push the cached value past S16_MAX.
+	 */
+	if (abs(nr + *bytes) >= S16_MAX) {
+		nr += *bytes;
+		*bytes = 0;
+		goto direct;
+	}
+
 	/*
 	 * Even for large object >= PAGE_SIZE, the vmstat data will still be
 	 * cached locally at least once before pushing it out.
-- 
2.53.0-Meta

Re: [PATCH 3/4] memcg: int16_t for cached slab stats

[Harry Yoo]

On 5/20/26 2:31 PM, Shakeel Butt wrote:
> Currently struct obj_stock_pcp stores cached slab stats in 'int' which
> is 4 bytes per counter on 64-bit machines. Switch them to int16_t to
> shrink the cached metadata.
> 
> The existing PAGE_SIZE flush in __account_obj_stock() bounds *bytes at
> PAGE_SIZE on 4KiB and 16KiB page archs, well within int16_t. On 64KiB
> pages PAGE_SIZE is well above S16_MAX so that flush never fires, and a
> sufficiently long run of accumulations would overflow the cache. Add
> an explicit S16_MAX guard before each add: when the next add would
> push abs(*bytes) past S16_MAX, fold the cached value into @nr and
> flush directly via mod_objcg_mlstate() before the accumulation.
> 
> Signed-off-by: Shakeel Butt <shakeel.butt@linux.dev>
> Tested-by: kernel test robot <oliver.sang@intel.com>
> ---

So arches with 64KiB sizes won't benefit from

"Even for large object >= PAGE_SIZE, the vmstat data will still be 
cached locally at least once before pushing it out" case.

But its benefit is questionable (to me), might be ok.

>   mm/memcontrol.c | 16 +++++++++++++---
>   1 file changed, 13 insertions(+), 3 deletions(-)
> 
> diff --git a/mm/memcontrol.c b/mm/memcontrol.c
> index b3d63d9f267c..1ed27fd06850 100644
> --- a/mm/memcontrol.c
> +++ b/mm/memcontrol.c
> @@ -3158,7 +3158,7 @@ static void __account_obj_stock(struct obj_cgroup *objcg,
>   				struct obj_stock_pcp *stock, int nr,
>   				struct pglist_data *pgdat, enum node_stat_item idx)
>   {
> -	int *bytes;
> +	int16_t *bytes;
>   
>   	/*
>   	 * Though at the moment MAX_NUMNODES <= 1024 in all archs but let's make
> @@ -3195,6 +3195,16 @@ static void __account_obj_stock(struct obj_cgroup *objcg,
>   
>   	bytes = (idx == NR_SLAB_RECLAIMABLE_B) ? &stock->nr_slab_reclaimable_b
>   					       : &stock->nr_slab_unreclaimable_b;
> +	/*
> +	 * To avoid overflow or underflow, flush directly if accumulating @nr
> +	 * would push the cached value past S16_MAX.
> +	 */
> +	if (abs(nr + *bytes) >= S16_MAX) {
nit: should be > S16_MAX?

> +		nr += *bytes;
> +		*bytes = 0;
> +		goto direct;
> +	}
> +
>   	/*
>   	 * Even for large object >= PAGE_SIZE, the vmstat data will still be
>   	 * cached locally at least once before pushing it out.

FWIW:
Reviewed-by: Harry Yoo (Oracle) <harry@kernel.org>

-- 
Cheers,
Harry / Hyeonggon

[PATCH 4/4] memcg: multi objcg charge support

[Shakeel Butt]

Commit 01b9da291c49 ("mm: memcontrol: convert objcg to be per-memcg
per-node type") split a memcg's single obj_cgroup into one per NUMA
node so that reparenting LRU folios can take per-node lru locks. As a
side effect, the per-CPU obj_stock_pcp -- which caches exactly one
cached_objcg -- thrashes on workloads where threads of the same memcg
run on different NUMA nodes. The kernel test robot reported a 67.7%
regression on stress-ng.switch.ops_per_sec from this pattern.

Mirror the multi-slot pattern already used by memcg_stock_pcp: turn
nr_bytes and cached_objcg into NR_OBJ_STOCK-element arrays, scan all
slots on consume/refill/account, prefer empty slots when inserting,
and evict a random slot only when full. With multiple slots a CPU can
hold the per-node objcg variants of one memcg plus a few siblings
without ever forcing a drain.

A single int8_t index records which slot the cached slab stats belong
to; the stats are flushed on slot or pgdat change. With NR_OBJ_STOCK
= 5 the layout (verified with pahole) is:

  offset 0  : lock(1) + index(1) + node_id(2) + slab stats(4) = 8B
  offset 8  : nr_bytes[5]                                     = 10B
  offset 18 : padding                                         = 6B
  offset 24 : cached[5]                                       = 40B
  offset 64 : (line 2) work_struct + flags (cold)

so consume_obj_stock, refill_obj_stock and the slab account path each
touch exactly one 64-byte cache line on non-debug 64-bit builds.

Reported-by: kernel test robot <oliver.sang@intel.com>
Closes: https://lore.kernel.org/oe-lkp/202605121641.b6a60cb0-lkp@intel.com
Fixes: 01b9da291c49 ("mm: memcontrol: convert objcg to be per-memcg per-node type")
Signed-off-by: Shakeel Butt <shakeel.butt@linux.dev>
Tested-by: kernel test robot <oliver.sang@intel.com>
---
 mm/memcontrol.c | 185 ++++++++++++++++++++++++++++++++++--------------
 1 file changed, 133 insertions(+), 52 deletions(-)

diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index 1ed27fd06850..52104cbb8e7c 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -150,14 +150,14 @@ static void obj_cgroup_release(struct percpu_ref *ref)
 	 * However, it can be PAGE_SIZE or (x * PAGE_SIZE).
 	 *
 	 * The following sequence can lead to it:
-	 * 1) CPU0: objcg == stock->cached_objcg
+	 * 1) CPU0: objcg cached in one of stock->cached[i]
 	 * 2) CPU1: we do a small allocation (e.g. 92 bytes),
 	 *          PAGE_SIZE bytes are charged
 	 * 3) CPU1: a process from another memcg is allocating something,
 	 *          the stock if flushed,
 	 *          objcg->nr_charged_bytes = PAGE_SIZE - 92
 	 * 5) CPU0: we do release this object,
-	 *          92 bytes are added to stock->nr_bytes
+	 *          92 bytes are added to stock->nr_bytes[i]
 	 * 6) CPU0: stock is flushed,
 	 *          92 bytes are added to objcg->nr_charged_bytes
 	 *
@@ -2017,13 +2017,25 @@ static DEFINE_PER_CPU_ALIGNED(struct memcg_stock_pcp, memcg_stock) = {
 	.lock = INIT_LOCAL_TRYLOCK(lock),
 };
 
+/*
+ * NR_OBJ_STOCK is sized so the entire hot path of obj_stock_pcp
+ * (lock, accounting metadata, nr_bytes[] and cached[]) fits within a
+ * single 64-byte cache line on non-debug 64-bit builds. With 5 slots:
+ *   lock(1) + index(1) + node_id(2) + slab stats(4) + nr_bytes(10)
+ *   + pad(6) + cached(40) == 64 bytes.
+ * A CPU can thus consume/refill/account against five different objcgs
+ * (typically per-node variants of the same memcg) while incurring at
+ * most one cache miss on the stock.
+ */
+#define NR_OBJ_STOCK 5
 struct obj_stock_pcp {
 	local_trylock_t lock;
-	uint16_t nr_bytes;
-	struct obj_cgroup *cached_objcg;
+	int8_t index;
 	int16_t node_id;
 	int16_t nr_slab_reclaimable_b;
 	int16_t nr_slab_unreclaimable_b;
+	uint16_t nr_bytes[NR_OBJ_STOCK];
+	struct obj_cgroup *cached[NR_OBJ_STOCK];
 
 	struct work_struct work;
 	unsigned long flags;
@@ -2031,11 +2043,13 @@ struct obj_stock_pcp {
 
 static DEFINE_PER_CPU_ALIGNED(struct obj_stock_pcp, obj_stock) = {
 	.lock = INIT_LOCAL_TRYLOCK(lock),
+	.index = -1,
 	.node_id = NUMA_NO_NODE,
 };
 
 static DEFINE_MUTEX(percpu_charge_mutex);
 
+static void drain_obj_stock_slot(struct obj_stock_pcp *stock, int i);
 static void drain_obj_stock(struct obj_stock_pcp *stock);
 static bool obj_stock_flush_required(struct obj_stock_pcp *stock,
 				     struct mem_cgroup *root_memcg);
@@ -3153,12 +3167,13 @@ static void unlock_stock(struct obj_stock_pcp *stock)
 		local_unlock(&obj_stock.lock);
 }
 
-/* Call after __refill_obj_stock() to ensure stock->cached_objg == objcg */
+/* Call after __refill_obj_stock() so a slot for objcg exists in the stock */
 static void __account_obj_stock(struct obj_cgroup *objcg,
 				struct obj_stock_pcp *stock, int nr,
 				struct pglist_data *pgdat, enum node_stat_item idx)
 {
 	int16_t *bytes;
+	int i;
 
 	/*
 	 * Though at the moment MAX_NUMNODES <= 1024 in all archs but let's make
@@ -3167,29 +3182,39 @@ static void __account_obj_stock(struct obj_cgroup *objcg,
 	 */
 	BUILD_BUG_ON(MAX_NUMNODES >= S16_MAX);
 
-	if (!stock || READ_ONCE(stock->cached_objcg) != objcg)
+	if (!stock)
+		goto direct;
+
+	for (i = 0; i < NR_OBJ_STOCK; ++i) {
+		if (READ_ONCE(stock->cached[i]) == objcg)
+			break;
+	}
+	if (i == NR_OBJ_STOCK)
 		goto direct;
 
 	/*
 	 * Save vmstat data in stock and skip vmstat array update unless
-	 * accumulating over a page of vmstat data or when pgdat changes.
+	 * accumulating over a page of vmstat data or when the objcg slot or
+	 * pgdat the stats belong to changes.
 	 */
-	if (stock->node_id == NUMA_NO_NODE) {
+	if (stock->index < 0) {
+		stock->index = i;
 		stock->node_id = pgdat->node_id;
-	} else if (stock->node_id != pgdat->node_id) {
-		/* Flush the existing cached vmstat data */
+	} else if (stock->index != i || stock->node_id != pgdat->node_id) {
+		struct obj_cgroup *old = READ_ONCE(stock->cached[stock->index]);
 		struct pglist_data *oldpg = NODE_DATA(stock->node_id);
 
 		if (stock->nr_slab_reclaimable_b) {
-			mod_objcg_mlstate(objcg, oldpg, NR_SLAB_RECLAIMABLE_B,
+			mod_objcg_mlstate(old, oldpg, NR_SLAB_RECLAIMABLE_B,
 					  stock->nr_slab_reclaimable_b);
 			stock->nr_slab_reclaimable_b = 0;
 		}
 		if (stock->nr_slab_unreclaimable_b) {
-			mod_objcg_mlstate(objcg, oldpg, NR_SLAB_UNRECLAIMABLE_B,
+			mod_objcg_mlstate(old, oldpg, NR_SLAB_UNRECLAIMABLE_B,
 					  stock->nr_slab_unreclaimable_b);
 			stock->nr_slab_unreclaimable_b = 0;
 		}
+		stock->index = i;
 		stock->node_id = pgdat->node_id;
 	}
 
@@ -3230,10 +3255,16 @@ static bool __consume_obj_stock(struct obj_cgroup *objcg,
 				struct obj_stock_pcp *stock,
 				unsigned int nr_bytes)
 {
-	if (objcg == READ_ONCE(stock->cached_objcg) &&
-	    stock->nr_bytes >= nr_bytes) {
-		stock->nr_bytes -= nr_bytes;
-		return true;
+	int i;
+
+	for (i = 0; i < NR_OBJ_STOCK; ++i) {
+		if (READ_ONCE(stock->cached[i]) != objcg)
+			continue;
+		if (stock->nr_bytes[i] >= nr_bytes) {
+			stock->nr_bytes[i] -= nr_bytes;
+			return true;
+		}
+		return false;
 	}
 
 	return false;
@@ -3254,16 +3285,42 @@ static bool consume_obj_stock(struct obj_cgroup *objcg, unsigned int nr_bytes)
 	return ret;
 }
 
-static void drain_obj_stock(struct obj_stock_pcp *stock)
+/* Flush the cached slab stats (if any) back to their owning objcg/pgdat. */
+static void drain_obj_stock_stats(struct obj_stock_pcp *stock)
 {
-	struct obj_cgroup *old = READ_ONCE(stock->cached_objcg);
+	struct obj_cgroup *old;
+	struct pglist_data *oldpg;
+
+	if (stock->index < 0)
+		return;
+
+	old = READ_ONCE(stock->cached[stock->index]);
+	oldpg = NODE_DATA(stock->node_id);
+
+	if (stock->nr_slab_reclaimable_b) {
+		mod_objcg_mlstate(old, oldpg, NR_SLAB_RECLAIMABLE_B,
+				  stock->nr_slab_reclaimable_b);
+		stock->nr_slab_reclaimable_b = 0;
+	}
+	if (stock->nr_slab_unreclaimable_b) {
+		mod_objcg_mlstate(old, oldpg, NR_SLAB_UNRECLAIMABLE_B,
+				  stock->nr_slab_unreclaimable_b);
+		stock->nr_slab_unreclaimable_b = 0;
+	}
+	stock->index = -1;
+	stock->node_id = NUMA_NO_NODE;
+}
+
+static void drain_obj_stock_slot(struct obj_stock_pcp *stock, int i)
+{
+	struct obj_cgroup *old = READ_ONCE(stock->cached[i]);
 
 	if (!old)
 		return;
 
-	if (stock->nr_bytes) {
-		unsigned int nr_pages = stock->nr_bytes >> PAGE_SHIFT;
-		unsigned int nr_bytes = stock->nr_bytes & (PAGE_SIZE - 1);
+	if (stock->nr_bytes[i]) {
+		unsigned int nr_pages = stock->nr_bytes[i] >> PAGE_SHIFT;
+		unsigned int nr_bytes = stock->nr_bytes[i] & (PAGE_SIZE - 1);
 
 		if (nr_pages) {
 			struct mem_cgroup *memcg;
@@ -3289,46 +3346,43 @@ static void drain_obj_stock(struct obj_stock_pcp *stock)
 		 * so it might be changed in the future.
 		 */
 		atomic_add(nr_bytes, &old->nr_charged_bytes);
-		stock->nr_bytes = 0;
+		stock->nr_bytes[i] = 0;
 	}
 
-	/*
-	 * Flush the vmstat data in current stock
-	 */
-	if (stock->nr_slab_reclaimable_b || stock->nr_slab_unreclaimable_b) {
-		struct pglist_data *oldpg = NODE_DATA(stock->node_id);
-
-		if (stock->nr_slab_reclaimable_b) {
-			mod_objcg_mlstate(old, oldpg,
-					  NR_SLAB_RECLAIMABLE_B,
-					  stock->nr_slab_reclaimable_b);
-			stock->nr_slab_reclaimable_b = 0;
-		}
-		if (stock->nr_slab_unreclaimable_b) {
-			mod_objcg_mlstate(old, oldpg,
-					  NR_SLAB_UNRECLAIMABLE_B,
-					  stock->nr_slab_unreclaimable_b);
-			stock->nr_slab_unreclaimable_b = 0;
-		}
-		stock->node_id = NUMA_NO_NODE;
-	}
+	/* Flush vmstat data when its owning slot is being drained. */
+	if (stock->index == i)
+		drain_obj_stock_stats(stock);
 
-	WRITE_ONCE(stock->cached_objcg, NULL);
+	WRITE_ONCE(stock->cached[i], NULL);
 	obj_cgroup_put(old);
 }
 
+static void drain_obj_stock(struct obj_stock_pcp *stock)
+{
+	int i;
+
+	for (i = 0; i < NR_OBJ_STOCK; ++i)
+		drain_obj_stock_slot(stock, i);
+}
+
 static bool obj_stock_flush_required(struct obj_stock_pcp *stock,
 				     struct mem_cgroup *root_memcg)
 {
-	struct obj_cgroup *objcg = READ_ONCE(stock->cached_objcg);
+	struct obj_cgroup *objcg;
 	struct mem_cgroup *memcg;
 	bool flush = false;
+	int i;
 
 	rcu_read_lock();
-	if (objcg) {
+	for (i = 0; i < NR_OBJ_STOCK; ++i) {
+		objcg = READ_ONCE(stock->cached[i]);
+		if (!objcg)
+			continue;
 		memcg = obj_cgroup_memcg(objcg);
-		if (memcg && mem_cgroup_is_descendant(memcg, root_memcg))
+		if (memcg && mem_cgroup_is_descendant(memcg, root_memcg)) {
 			flush = true;
+			break;
+		}
 	}
 	rcu_read_unlock();
 
@@ -3342,6 +3396,7 @@ static void __refill_obj_stock(struct obj_cgroup *objcg,
 {
 	unsigned int nr_pages = 0;
 	unsigned int stock_nr_bytes;
+	int i, slot = -1, empty_slot = -1;
 
 	if (!stock) {
 		nr_pages = nr_bytes >> PAGE_SHIFT;
@@ -3350,19 +3405,45 @@ static void __refill_obj_stock(struct obj_cgroup *objcg,
 		goto out;
 	}
 
-	stock_nr_bytes = stock->nr_bytes;
-	if (READ_ONCE(stock->cached_objcg) != objcg) { /* reset if necessary */
-		drain_obj_stock(stock);
+	for (i = 0; i < NR_OBJ_STOCK; ++i) {
+		struct obj_cgroup *cached = READ_ONCE(stock->cached[i]);
+
+		if (!cached) {
+			if (empty_slot == -1)
+				empty_slot = i;
+			continue;
+		}
+		if (cached == objcg) {
+			slot = i;
+			break;
+		}
+	}
+
+	if (slot == -1) {
+		slot = empty_slot;
+		if (slot == -1) {
+			slot = get_random_u32_below(NR_OBJ_STOCK);
+			drain_obj_stock_slot(stock, slot);
+		}
 		obj_cgroup_get(objcg);
+		/*
+		 * Keep the xchg result in the unsigned int local; storing
+		 * it directly into stock->nr_bytes[slot] (uint16_t) would
+		 * silently truncate values >= U16_MAX and bypass the flush
+		 * guard below, leaking page-counter charges.
+		 */
 		stock_nr_bytes = atomic_read(&objcg->nr_charged_bytes)
 				? atomic_xchg(&objcg->nr_charged_bytes, 0) : 0;
-		WRITE_ONCE(stock->cached_objcg, objcg);
+		WRITE_ONCE(stock->cached[slot], objcg);
 
 		allow_uncharge = true;	/* Allow uncharge when objcg changes */
+	} else {
+		stock_nr_bytes = stock->nr_bytes[slot];
 	}
+
 	stock_nr_bytes += nr_bytes;
 
-	/* Since stock->nr_bytes is uint16_t, don't refill >= U16_MAX */
+	/* nr_bytes[] is uint16_t; flush if we would refill >= U16_MAX. */
 	if ((allow_uncharge && (stock_nr_bytes > PAGE_SIZE)) ||
 	    stock_nr_bytes >= U16_MAX) {
 		nr_pages = stock_nr_bytes >> PAGE_SHIFT;
@@ -3384,7 +3465,7 @@ static void __refill_obj_stock(struct obj_cgroup *objcg,
 			stock_nr_bytes = kept;
 		}
 	}
-	stock->nr_bytes = stock_nr_bytes;
+	stock->nr_bytes[slot] = stock_nr_bytes;
 
 out:
 	if (nr_pages)
-- 
2.53.0-Meta

Re: [PATCH 4/4] memcg: multi objcg charge support

[Harry Yoo]

On 5/20/26 2:31 PM, Shakeel Butt wrote:
> Commit 01b9da291c49 ("mm: memcontrol: convert objcg to be per-memcg
> per-node type") split a memcg's single obj_cgroup into one per NUMA
> node so that reparenting LRU folios can take per-node lru locks. As a
> side effect, the per-CPU obj_stock_pcp -- which caches exactly one
> cached_objcg -- thrashes on workloads where threads of the same memcg
> run on different NUMA nodes. The kernel test robot reported a 67.7%
> regression on stress-ng.switch.ops_per_sec from this pattern.
> 
> Mirror the multi-slot pattern already used by memcg_stock_pcp: turn
> nr_bytes and cached_objcg into NR_OBJ_STOCK-element arrays, scan all
> slots on consume/refill/account, prefer empty slots when inserting,
> and evict a random slot only when full. With multiple slots a CPU can
> hold the per-node objcg variants of one memcg plus a few siblings
> without ever forcing a drain.
> 
> A single int8_t index records which slot the cached slab stats belong
> to; the stats are flushed on slot or pgdat change. With NR_OBJ_STOCK
> = 5 the layout (verified with pahole) is:
> 
>    offset 0  : lock(1) + index(1) + node_id(2) + slab stats(4) = 8B
>    offset 8  : nr_bytes[5]                                     = 10B
>    offset 18 : padding                                         = 6B
>    offset 24 : cached[5]                                       = 40B
>    offset 64 : (line 2) work_struct + flags (cold)
> 
> so consume_obj_stock, refill_obj_stock and the slab account path each
> touch exactly one 64-byte cache line on non-debug 64-bit builds.
> 
> Reported-by: kernel test robot <oliver.sang@intel.com>
> Closes: https://lore.kernel.org/oe-lkp/202605121641.b6a60cb0-lkp@intel.com
> Fixes: 01b9da291c49 ("mm: memcontrol: convert objcg to be per-memcg per-node type")
> Signed-off-by: Shakeel Butt <shakeel.butt@linux.dev>
> Tested-by: kernel test robot <oliver.sang@intel.com>
> ---
> @@ -3350,19 +3405,45 @@ static void __refill_obj_stock(struct obj_cgroup *objcg,
>   		goto out;
>   	}
>   
> -	stock_nr_bytes = stock->nr_bytes;
> -	if (READ_ONCE(stock->cached_objcg) != objcg) { /* reset if necessary */
> -		drain_obj_stock(stock);
> +	for (i = 0; i < NR_OBJ_STOCK; ++i) {
> +		struct obj_cgroup *cached = READ_ONCE(stock->cached[i]);
> +
> +		if (!cached) {
> +			if (empty_slot == -1)
> +				empty_slot = i;
> +			continue;
> +		}
> +		if (cached == objcg) {
> +			slot = i;
> +			break;
> +		}
> +	}
> +
> +	if (slot == -1) {
> +		slot = empty_slot;
> +		if (slot == -1) {
> +			slot = get_random_u32_below(NR_OBJ_STOCK);

It would break kmalloc_nolock() because _get_random_bytes() uses a 
spinlock. perhaps prandom_u32_state() should be sufficient in this case.

Is there a reason why it uses random eviction, unlike multi-memcg percpu 
charge cache?

Otherwise LGTM!

-- 
Cheers,
Harry / Hyeonggon

Re: [PATCH 4/4] memcg: multi objcg charge support

[Shakeel Butt]

On Wed, May 20, 2026 at 06:35:30PM +0900, Harry Yoo wrote:
> 
> 
> On 5/20/26 2:31 PM, Shakeel Butt wrote:
> > Commit 01b9da291c49 ("mm: memcontrol: convert objcg to be per-memcg
> > per-node type") split a memcg's single obj_cgroup into one per NUMA
> > node so that reparenting LRU folios can take per-node lru locks. As a
> > side effect, the per-CPU obj_stock_pcp -- which caches exactly one
> > cached_objcg -- thrashes on workloads where threads of the same memcg
> > run on different NUMA nodes. The kernel test robot reported a 67.7%
> > regression on stress-ng.switch.ops_per_sec from this pattern.
> > 
> > Mirror the multi-slot pattern already used by memcg_stock_pcp: turn
> > nr_bytes and cached_objcg into NR_OBJ_STOCK-element arrays, scan all
> > slots on consume/refill/account, prefer empty slots when inserting,
> > and evict a random slot only when full. With multiple slots a CPU can
> > hold the per-node objcg variants of one memcg plus a few siblings
> > without ever forcing a drain.
> > 
> > A single int8_t index records which slot the cached slab stats belong
> > to; the stats are flushed on slot or pgdat change. With NR_OBJ_STOCK
> > = 5 the layout (verified with pahole) is:
> > 
> >    offset 0  : lock(1) + index(1) + node_id(2) + slab stats(4) = 8B
> >    offset 8  : nr_bytes[5]                                     = 10B
> >    offset 18 : padding                                         = 6B
> >    offset 24 : cached[5]                                       = 40B
> >    offset 64 : (line 2) work_struct + flags (cold)
> > 
> > so consume_obj_stock, refill_obj_stock and the slab account path each
> > touch exactly one 64-byte cache line on non-debug 64-bit builds.
> > 
> > Reported-by: kernel test robot <oliver.sang@intel.com>
> > Closes: https://lore.kernel.org/oe-lkp/202605121641.b6a60cb0-lkp@intel.com
> > Fixes: 01b9da291c49 ("mm: memcontrol: convert objcg to be per-memcg per-node type")
> > Signed-off-by: Shakeel Butt <shakeel.butt@linux.dev>
> > Tested-by: kernel test robot <oliver.sang@intel.com>
> > ---
> > @@ -3350,19 +3405,45 @@ static void __refill_obj_stock(struct obj_cgroup *objcg,
> >   		goto out;
> >   	}
> > -	stock_nr_bytes = stock->nr_bytes;
> > -	if (READ_ONCE(stock->cached_objcg) != objcg) { /* reset if necessary */
> > -		drain_obj_stock(stock);
> > +	for (i = 0; i < NR_OBJ_STOCK; ++i) {
> > +		struct obj_cgroup *cached = READ_ONCE(stock->cached[i]);
> > +
> > +		if (!cached) {
> > +			if (empty_slot == -1)
> > +				empty_slot = i;
> > +			continue;
> > +		}
> > +		if (cached == objcg) {
> > +			slot = i;
> > +			break;
> > +		}
> > +	}
> > +
> > +	if (slot == -1) {
> > +		slot = empty_slot;
> > +		if (slot == -1) {
> > +			slot = get_random_u32_below(NR_OBJ_STOCK);
> 
> It would break kmalloc_nolock() because _get_random_bytes() uses a spinlock.
> perhaps prandom_u32_state() should be sufficient in this case.
> 
> Is there a reason why it uses random eviction, unlike multi-memcg percpu
> charge cache?

Oh I didn't know and actually we are already using get_random_u32_below() in
refill_stock(). So, it need fixing as well. That would be a separate patch.

I will explore prandom_u32_state().

Re: [PATCH 4/4] memcg: multi objcg charge support

[Harry Yoo]

On 5/21/26 10:05 AM, Shakeel Butt wrote:
> On Wed, May 20, 2026 at 06:35:30PM +0900, Harry Yoo wrote:
>>> @@ -3350,19 +3405,45 @@ static void __refill_obj_stock(struct obj_cgroup *objcg,
>>>    		goto out;
>>>    	}
>>> -	stock_nr_bytes = stock->nr_bytes;
>>> -	if (READ_ONCE(stock->cached_objcg) != objcg) { /* reset if necessary */
>>> -		drain_obj_stock(stock);
>>> +	for (i = 0; i < NR_OBJ_STOCK; ++i) {
>>> +		struct obj_cgroup *cached = READ_ONCE(stock->cached[i]);
>>> +
>>> +		if (!cached) {
>>> +			if (empty_slot == -1)
>>> +				empty_slot = i;
>>> +			continue;
>>> +		}
>>> +		if (cached == objcg) {
>>> +			slot = i;
>>> +			break;
>>> +		}
>>> +	}
>>> +
>>> +	if (slot == -1) {
>>> +		slot = empty_slot;
>>> +		if (slot == -1) {
>>> +			slot = get_random_u32_below(NR_OBJ_STOCK);
>>
>> It would break kmalloc_nolock() because _get_random_bytes() uses a spinlock.
>> perhaps prandom_u32_state() should be sufficient in this case.

s/spinlock/local_lock/

>> Is there a reason why it uses random eviction, unlike multi-memcg percpu
>> charge cache?
> 
> Oh I didn't know and actually we are already using get_random_u32_below() in
> refill_stock(). So, it need fixing as well. That would be a separate patch.

Ouch, I see.
> I will explore prandom_u32_state().

Thanks!

FYI, SLUB had a similar issue that was recently fixed:
commit a1e244a9f1778 ("mm/slab: use prandom if !allow_spin").

It uses prandom if spinning is not allowed when shuffling slab freelist.

-- 
Cheers,
Harry / Hyeonggon

Re: [PATCH 4/4] memcg: multi objcg charge support

[Shakeel Butt]

On Thu, May 21, 2026 at 10:43:11AM +0900, Harry Yoo wrote:
> 
> 
> On 5/21/26 10:05 AM, Shakeel Butt wrote:
> > On Wed, May 20, 2026 at 06:35:30PM +0900, Harry Yoo wrote:
> > > > @@ -3350,19 +3405,45 @@ static void __refill_obj_stock(struct obj_cgroup *objcg,
> > > >    		goto out;
> > > >    	}
> > > > -	stock_nr_bytes = stock->nr_bytes;
> > > > -	if (READ_ONCE(stock->cached_objcg) != objcg) { /* reset if necessary */
> > > > -		drain_obj_stock(stock);
> > > > +	for (i = 0; i < NR_OBJ_STOCK; ++i) {
> > > > +		struct obj_cgroup *cached = READ_ONCE(stock->cached[i]);
> > > > +
> > > > +		if (!cached) {
> > > > +			if (empty_slot == -1)
> > > > +				empty_slot = i;
> > > > +			continue;
> > > > +		}
> > > > +		if (cached == objcg) {
> > > > +			slot = i;
> > > > +			break;
> > > > +		}
> > > > +	}
> > > > +
> > > > +	if (slot == -1) {
> > > > +		slot = empty_slot;
> > > > +		if (slot == -1) {
> > > > +			slot = get_random_u32_below(NR_OBJ_STOCK);
> > > 
> > > It would break kmalloc_nolock() because _get_random_bytes() uses a spinlock.
> > > perhaps prandom_u32_state() should be sufficient in this case.
> 
> s/spinlock/local_lock/

I do see spinlock in crng_make_state() for some code paths.

> 
> > > Is there a reason why it uses random eviction, unlike multi-memcg percpu
> > > charge cache?
> > 
> > Oh I didn't know and actually we are already using get_random_u32_below() in
> > refill_stock(). So, it need fixing as well. That would be a separate patch.
> 
> Ouch, I see.
> > I will explore prandom_u32_state().
> 
> Thanks!
> 
> FYI, SLUB had a similar issue that was recently fixed:
> commit a1e244a9f1778 ("mm/slab: use prandom if !allow_spin").
> 
> It uses prandom if spinning is not allowed when shuffling slab freelist.

The drain does not really need a random number. Fixing an index like 0 makes it
much simpler but it might expose some corner cases. Round robin might be enough
for this though. I will see what would be the easiest way forward.

Re: [PATCH 4/4] memcg: multi objcg charge support

[Joshua Hahn]

On Wed, 20 May 2026 18:05:23 -0700 Shakeel Butt <shakeel.butt@linux.dev> wrote:

> On Wed, May 20, 2026 at 06:35:30PM +0900, Harry Yoo wrote:
> > 
> > 
> > On 5/20/26 2:31 PM, Shakeel Butt wrote:
> > > Commit 01b9da291c49 ("mm: memcontrol: convert objcg to be per-memcg
> > > per-node type") split a memcg's single obj_cgroup into one per NUMA
> > > node so that reparenting LRU folios can take per-node lru locks. As a
> > > side effect, the per-CPU obj_stock_pcp -- which caches exactly one
> > > cached_objcg -- thrashes on workloads where threads of the same memcg
> > > run on different NUMA nodes. The kernel test robot reported a 67.7%
> > > regression on stress-ng.switch.ops_per_sec from this pattern.
> > > 
> > > Mirror the multi-slot pattern already used by memcg_stock_pcp: turn
> > > nr_bytes and cached_objcg into NR_OBJ_STOCK-element arrays, scan all
> > > slots on consume/refill/account, prefer empty slots when inserting,
> > > and evict a random slot only when full. With multiple slots a CPU can
> > > hold the per-node objcg variants of one memcg plus a few siblings
> > > without ever forcing a drain.
> > > 
> > > A single int8_t index records which slot the cached slab stats belong
> > > to; the stats are flushed on slot or pgdat change. With NR_OBJ_STOCK
> > > = 5 the layout (verified with pahole) is:
> > > 
> > >    offset 0  : lock(1) + index(1) + node_id(2) + slab stats(4) = 8B
> > >    offset 8  : nr_bytes[5]                                     = 10B
> > >    offset 18 : padding                                         = 6B
> > >    offset 24 : cached[5]                                       = 40B
> > >    offset 64 : (line 2) work_struct + flags (cold)
> > > 
> > > so consume_obj_stock, refill_obj_stock and the slab account path each
> > > touch exactly one 64-byte cache line on non-debug 64-bit builds.
> > > 
> > > Reported-by: kernel test robot <oliver.sang@intel.com>
> > > Closes: https://lore.kernel.org/oe-lkp/202605121641.b6a60cb0-lkp@intel.com
> > > Fixes: 01b9da291c49 ("mm: memcontrol: convert objcg to be per-memcg per-node type")
> > > Signed-off-by: Shakeel Butt <shakeel.butt@linux.dev>
> > > Tested-by: kernel test robot <oliver.sang@intel.com>
> > > ---
> > > @@ -3350,19 +3405,45 @@ static void __refill_obj_stock(struct obj_cgroup *objcg,
> > >   		goto out;
> > >   	}
> > > -	stock_nr_bytes = stock->nr_bytes;
> > > -	if (READ_ONCE(stock->cached_objcg) != objcg) { /* reset if necessary */
> > > -		drain_obj_stock(stock);
> > > +	for (i = 0; i < NR_OBJ_STOCK; ++i) {
> > > +		struct obj_cgroup *cached = READ_ONCE(stock->cached[i]);
> > > +
> > > +		if (!cached) {
> > > +			if (empty_slot == -1)
> > > +				empty_slot = i;
> > > +			continue;
> > > +		}
> > > +		if (cached == objcg) {
> > > +			slot = i;
> > > +			break;
> > > +		}
> > > +	}
> > > +
> > > +	if (slot == -1) {
> > > +		slot = empty_slot;
> > > +		if (slot == -1) {
> > > +			slot = get_random_u32_below(NR_OBJ_STOCK);
> > 
> > It would break kmalloc_nolock() because _get_random_bytes() uses a spinlock.
> > perhaps prandom_u32_state() should be sufficient in this case.
> 
> > Is there a reason why it uses random eviction, unlike multi-memcg percpu
> > charge cache?
> 
> Oh I didn't know and actually we are already using get_random_u32_below() in
> refill_stock(). So, it need fixing as well. That would be a separate patch.

Hello Harry and Shakeel!

I just wanted to note that my work to push the memcg stock to page_counter [1]
actually gets rid of the random eviction and also the get_random_u32_below.
So the problem will no longer exist, at least for memcg/page_counter stock.

I was also thinking whether there was any way we can abstract or push down the
objcg stock to a different per_X slot (with finer granularity than per-node)
so we don't do random evictions, but it doesn't seem like there's a natural
point to do this from a quick glance.

The fix for get_random_32_below to use prandom_u32_state seems quite simple
though, so if either of you would like to send that in as a small hotfix
it shouldn't be a big deal to rebase on top of that instead too.

(Sorry for joining the conversation late too. I'll take a closer look at this
entire series tomorrow as well).

Thank you both. Have a great day!
Joshua

[1] https://lore.kernel.org/all/20260410210742.550489-1-joshua.hahnjy@gmail.com/

[PATCH 0/4] memcg: shrink obj_stock_pcp and cache multiple objcgs

[Shakeel Butt]

Commit 01b9da291c49 ("mm: memcontrol: convert objcg to be per-memcg
per-node type") split a memcg's single obj_cgroup into one per NUMA
node so that reparenting LRU folios can take per-node lru locks. As a
side effect, the per-CPU obj_stock_pcp -- which caches a single
cached_objcg pointer -- thrashes on workloads where threads of the
same memcg run on different NUMA nodes. The kernel test robot reported
a 67.7% regression on stress-ng.switch.ops_per_sec from this pattern.

Commit d0211878ce06 ("memcg: cache obj_stock by memcg, not by objcg
pointer") landed as a temporary fix by treating sibling per-node
objcgs as equivalent for the cache lookup, intended to be reverted
once per-node kmem accounting is introduced. This series takes a more
general approach: cache multiple objcgs per CPU using the multi-slot
pattern memcg_stock_pcp already uses, so the per-node objcg variants
of one memcg can all coexist in the stock without ever forcing a
drain. The temporary fix can then be reverted.

To avoid increasing the per-CPU cache footprint, the first three
patches shrink the existing single-slot obj_stock_pcp fields.
The final patch converts cached_objcg and nr_bytes into
NR_OBJ_STOCK=5 slot arrays and reorders the struct so the entire
consume/refill/account hot path fits within a single 64-byte cache
line on non-debug 64-bit builds (verified with pahole).

Reported-by: kernel test robot <oliver.sang@intel.com>
Closes: https://lore.kernel.org/oe-lkp/202605121641.b6a60cb0-lkp@intel.com
Fixes: 01b9da291c49 ("mm: memcontrol: convert objcg to be per-memcg per-node type")
Tested-by: kernel test robot <oliver.sang@intel.com>

Shakeel Butt (4):
  memcg: store node_id instead of pglist_data pointer
  memcg: uint16_t for nr_bytes in obj_stock_pcp
  memcg: int16_t for cached slab stats
  memcg: multi objcg charge support

 mm/memcontrol.c | 214 +++++++++++++++++++++++++++++++++++-------------
 1 file changed, 157 insertions(+), 57 deletions(-)

--
2.53.0-Meta