mm/slab: extract __free_to_pcs_batch() from free_to_pcs_bulk()

It has been noted that free_to_pcs_bulk() is difficult to follow, with a
number of goto labels, and this has contributed to two memory leak bugs
in there.

Extract part of the code to __free_to_pcs_batch(), which focuses only on
freeing free-hook-processed local objects to a percpu sheaf, and
returning how many were freed. Zero means a trylock failure or no empty
sheaf available, and thus the caller should fallback to
__kmem_cache_free_bulk().

Make free_to_pcs_bulk() call this in a while loop, removing all goto
labels from the function. __free_to_pcs_batch() retains two rather
straightforward ones.

Reviewed-by: Shengming Hu <hu.shengming@zte.com.cn>
Link: https://patch.msgid.link/20260707-slab-simplify-bulk-pcs-v1-1-4850dbe0d904@kernel.org
Reviewed-by: Hao Li <hao.li@linux.dev>
Reviewed-by: Harry Yoo (Oracle) <harry@kernel.org>
Signed-off-by: Vlastimil Babka (SUSE) <vbabka@kernel.org>
This commit is contained in:
Vlastimil Babka (SUSE) 2026-07-07 14:16:00 +02:00
parent bfec5d0b39
commit e1fa264890

116
mm/slub.c
View File

@ -6179,51 +6179,21 @@ static __always_inline bool can_free_to_pcs(struct slab *slab)
}
/*
* Bulk free objects to the percpu sheaves.
* Unlike free_to_pcs() this includes the calls to all necessary hooks
* and the fallback to freeing to slab pages.
* Try to free as many objects (already processed by free hooks) as possible to
* a single per-cpu sheaf.
*
* Returns how many objects were freed. Zero means failure and the caller should
* fall back to __kmem_cache_free_bulk().
*/
static void free_to_pcs_bulk(struct kmem_cache *s, size_t size, void **p)
static unsigned int __free_to_pcs_batch(struct kmem_cache *s, size_t size, void **p)
{
struct slub_percpu_sheaves *pcs;
struct slab_sheaf *main, *empty;
bool init = slab_want_init_on_free(s);
unsigned int batch, i = 0;
struct node_barn *barn;
void *remote_objects[PCS_BATCH_MAX];
unsigned int remote_nr = 0;
unsigned int batch;
while (i < size) {
struct slab *slab = virt_to_slab(p[i]);
memcg_slab_free_hook(s, slab, p + i, 1);
alloc_tagging_slab_free_hook(s, slab, p + i, 1);
if (unlikely(!slab_free_hook(s, p[i], init, false))) {
p[i] = p[--size];
continue;
}
if (unlikely(!can_free_to_pcs(slab))) {
remote_objects[remote_nr] = p[i];
p[i] = p[--size];
if (++remote_nr >= PCS_BATCH_MAX) {
__kmem_cache_free_bulk(s, remote_nr, &remote_objects[0]);
stat_add(s, FREE_SLOWPATH, remote_nr);
remote_nr = 0;
}
continue;
}
i++;
}
if (!size)
goto flush_remote;
next_batch:
if (!local_trylock(&s->cpu_sheaves->lock))
goto fallback;
return 0;
pcs = this_cpu_ptr(s->cpu_sheaves);
@ -6269,29 +6239,63 @@ static void free_to_pcs_bulk(struct kmem_cache *s, size_t size, void **p)
stat_add(s, FREE_FASTPATH, batch);
if (batch < size) {
p += batch;
size -= batch;
goto next_batch;
}
if (remote_nr)
goto flush_remote;
return;
return batch;
no_empty:
local_unlock(&s->cpu_sheaves->lock);
/*
* if we depleted all empty sheaves in the barn or there are too
* many full sheaves, free the rest to slab pages
*/
fallback:
__kmem_cache_free_bulk(s, size, p);
stat_add(s, FREE_SLOWPATH, size);
return 0;
}
/*
* Bulk free objects to the percpu sheaves.
* Unlike free_to_pcs() this includes the calls to all necessary hooks
* and the fallback to freeing to slab pages.
*/
static void free_to_pcs_bulk(struct kmem_cache *s, size_t size, void **p)
{
bool init = slab_want_init_on_free(s);
void *remote_objects[PCS_BATCH_MAX];
unsigned int remote_nr = 0;
for (unsigned int i = 0; i < size;) {
struct slab *slab = virt_to_slab(p[i]);
memcg_slab_free_hook(s, slab, p + i, 1);
alloc_tagging_slab_free_hook(s, slab, p + i, 1);
if (unlikely(!slab_free_hook(s, p[i], init, false))) {
p[i] = p[--size];
continue;
}
if (unlikely(!can_free_to_pcs(slab))) {
remote_objects[remote_nr] = p[i];
p[i] = p[--size];
if (++remote_nr >= PCS_BATCH_MAX) {
__kmem_cache_free_bulk(s, remote_nr, &remote_objects[0]);
stat_add(s, FREE_SLOWPATH, remote_nr);
remote_nr = 0;
}
continue;
}
i++;
}
while (size) {
unsigned int batch_freed = __free_to_pcs_batch(s, size, p);
if (!batch_freed) {
__kmem_cache_free_bulk(s, size, p);
stat_add(s, FREE_SLOWPATH, size);
break;
}
p += batch_freed;
size -= batch_freed;
}
flush_remote:
if (remote_nr) {
__kmem_cache_free_bulk(s, remote_nr, &remote_objects[0]);
stat_add(s, FREE_SLOWPATH, remote_nr);