diff options
| author | Shengming Hu <hu.shengming@zte.com.cn> | 2026-06-04 15:27:32 +0300 |
|---|---|---|
| committer | Vlastimil Babka (SUSE) <vbabka@kernel.org> | 2026-06-05 14:47:08 +0300 |
| commit | 19b206b9534a85266efa78febeb4ae185e75bccd (patch) | |
| tree | 423cac8a5aaad7422a4eef145555271740b2b2e4 | |
| parent | 0fc52deec1068ea3cc8eaa6e045c96fbf73f20e2 (diff) | |
| download | linux-19b206b9534a85266efa78febeb4ae185e75bccd.tar.xz | |
mm/slub: preserve original size in _kmalloc_nolock_noprof retry path
_kmalloc_nolock_noprof() retries from the next kmalloc bucket when the
initial allocation fails. The retry currently reuses `size` as the
bucket selector and overwrites it with s->object_size + 1.
That value is later passed as the original allocation size to
__slab_alloc_node(), slab_post_alloc_hook() and kasan_kmalloc(). On a
successful retry this makes KASAN/slub-debug observe the retry bucket
selector rather than the caller requested size, potentially widening the
valid kmalloc range and hiding overflows.
Keep the caller requested size separately as orig_size and pass it to
the allocation/debug/KASAN paths. Continue using `size` as the retry cache
selector.
Fixes: af92793e52c3 ("slab: Introduce kmalloc_nolock() and kfree_nolock()")
Signed-off-by: Shengming Hu <hu.shengming@zte.com.cn>
Reviewed-by: Harry Yoo (Oracle) <harry@kernel.org>
Reviewed-by: Hao Li <hao.li@linux.dev>
Link: https://patch.msgid.link/202606042027323804pk3MRY42Jy7y42OHAhQZ@zte.com.cn
Signed-off-by: Vlastimil Babka (SUSE) <vbabka@kernel.org>
| -rw-r--r-- | mm/slub.c | 7 |
1 files changed, 4 insertions, 3 deletions
diff --git a/mm/slub.c b/mm/slub.c index 492128ae3af9..ff83345d7f95 100644 --- a/mm/slub.c +++ b/mm/slub.c @@ -5359,6 +5359,7 @@ EXPORT_SYMBOL(__kmalloc_noprof); void *kmalloc_nolock_noprof(size_t size, gfp_t gfp_flags, int node) { gfp_t alloc_gfp = __GFP_NOWARN | __GFP_NOMEMALLOC | gfp_flags; + size_t orig_size = size; struct kmem_cache *s; bool can_retry = true; void *ret; @@ -5407,7 +5408,7 @@ retry: * kfence_alloc. Hence call __slab_alloc_node() (at most twice) * and slab_post_alloc_hook() directly. */ - ret = __slab_alloc_node(s, alloc_gfp, node, _RET_IP_, size); + ret = __slab_alloc_node(s, alloc_gfp, node, _RET_IP_, orig_size); /* * It's possible we failed due to trylock as we preempted someone with @@ -5431,9 +5432,9 @@ retry: success: maybe_wipe_obj_freeptr(s, ret); slab_post_alloc_hook(s, NULL, alloc_gfp, 1, &ret, - slab_want_init_on_alloc(alloc_gfp, s), size); + slab_want_init_on_alloc(alloc_gfp, s), orig_size); - ret = kasan_kmalloc(s, ret, size, alloc_gfp); + ret = kasan_kmalloc(s, ret, orig_size, alloc_gfp); return ret; } EXPORT_SYMBOL_GPL(kmalloc_nolock_noprof); |
