* [PATCH] slab: Saturate to SIZE_MAX for allocation size overflows
@ 2026-02-25 1:40 Kees Cook
2026-02-25 2:36 ` Harry Yoo
0 siblings, 1 reply; 2+ messages in thread
From: Kees Cook @ 2026-02-25 1:40 UTC (permalink / raw)
To: Vlastimil Babka
Cc: Kees Cook, Vlastimil Babka, Andrew Morton, Christoph Lameter,
David Rientjes, Roman Gushchin, Harry Yoo, linux-mm,
linux-kernel, linux-hardening
Instead of silently returning NULL on size overflows from array
allocations, saturate the request to SIZE_MAX so the error will be
surfaced to the allocator (and still return NULL).
Suggested-by: Vlastimil Babka <vbabka@suse.cz>
Link: https://lore.kernel.org/lkml/a144cd1e-8bfc-4380-8f1b-071db0af0b2c@suse.cz/
Signed-off-by: Kees Cook <kees@kernel.org>
---
Cc: Vlastimil Babka <vbabka@kernel.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Christoph Lameter <cl@gentwo.org>
Cc: David Rientjes <rientjes@google.com>
Cc: Roman Gushchin <roman.gushchin@linux.dev>
Cc: Harry Yoo <harry.yoo@oracle.com>
Cc: <linux-mm@kvack.org>
---
include/linux/slab.h | 8 ++++----
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/include/linux/slab.h b/include/linux/slab.h
index a5a5e4108ae5..8453c81c75c3 100644
--- a/include/linux/slab.h
+++ b/include/linux/slab.h
@@ -1105,7 +1105,7 @@ static inline __alloc_size(1, 2) void *kmalloc_array_noprof(size_t n, size_t siz
size_t bytes;
if (unlikely(check_mul_overflow(n, size, &bytes)))
- return NULL;
+ bytes = SIZE_MAX;
return kmalloc_noprof(bytes, flags);
}
#define kmalloc_array(...) alloc_hooks(kmalloc_array_noprof(__VA_ARGS__))
@@ -1135,7 +1135,7 @@ static inline __realloc_size(2, 3) void * __must_check krealloc_array_noprof(voi
size_t bytes;
if (unlikely(check_mul_overflow(new_n, new_size, &bytes)))
- return NULL;
+ bytes = SIZE_MAX;
return krealloc_noprof(p, bytes, flags);
}
@@ -1175,7 +1175,7 @@ static inline __alloc_size(1, 2) void *kmalloc_array_node_noprof(size_t n, size_
size_t bytes;
if (unlikely(check_mul_overflow(n, size, &bytes)))
- return NULL;
+ bytes = SIZE_MAX;
if (__builtin_constant_p(n) && __builtin_constant_p(size))
return kmalloc_node_noprof(bytes, flags, node);
return __kmalloc_node_noprof(PASS_BUCKET_PARAMS(bytes, NULL), flags, node);
@@ -1223,7 +1223,7 @@ kvmalloc_array_node_noprof(size_t n, size_t size, gfp_t flags, int node)
size_t bytes;
if (unlikely(check_mul_overflow(n, size, &bytes)))
- return NULL;
+ bytes = SIZE_MAX;
return kvmalloc_node_align_noprof(bytes, 1, flags, node);
}
--
2.34.1
^ permalink raw reply [flat|nested] 2+ messages in thread
* Re: [PATCH] slab: Saturate to SIZE_MAX for allocation size overflows
2026-02-25 1:40 [PATCH] slab: Saturate to SIZE_MAX for allocation size overflows Kees Cook
@ 2026-02-25 2:36 ` Harry Yoo
0 siblings, 0 replies; 2+ messages in thread
From: Harry Yoo @ 2026-02-25 2:36 UTC (permalink / raw)
To: Kees Cook
Cc: Vlastimil Babka, Vlastimil Babka, Andrew Morton,
Christoph Lameter, David Rientjes, Roman Gushchin, linux-mm,
linux-kernel, linux-hardening
On Tue, Feb 24, 2026 at 05:40:02PM -0800, Kees Cook wrote:
> Instead of silently returning NULL on size overflows from array
> allocations, saturate the request to SIZE_MAX so the error will be
> surfaced to the allocator (and still return NULL).
>
> Suggested-by: Vlastimil Babka <vbabka@suse.cz>
> Link: https://lore.kernel.org/lkml/a144cd1e-8bfc-4380-8f1b-071db0af0b2c@suse.cz/
> Signed-off-by: Kees Cook <kees@kernel.org>
> ---
Yes, since it's larger than order-1, it'll be passed to
the page allocator and will hit the order > MAX_PAGE_ORDER warning in
__alloc_frozen_pages_noprof().
Looks good to me,
Reviewed-by: Harry Yoo <harry.yoo@oracle.com>
--
Cheers,
Harry / Hyeonggon
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2026-02-25 2:36 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2026-02-25 1:40 [PATCH] slab: Saturate to SIZE_MAX for allocation size overflows Kees Cook
2026-02-25 2:36 ` Harry Yoo
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox