* [PATCH] mm/page_owner: use strscpy() instead of strlcpy()
@ 2022-05-09 14:59 Eric Dumazet
2022-05-09 15:48 ` Waiman Long
2022-05-09 15:52 ` Shakeel Butt
0 siblings, 2 replies; 3+ messages in thread
From: Eric Dumazet @ 2022-05-09 14:59 UTC (permalink / raw)
To: Andrew Morton
Cc: linux-kernel, linux-mm, Eric Dumazet, Eric Dumazet, syzbot, Waiman Long
From: Eric Dumazet <edumazet@google.com>
current->comm[] is not a string (no guarantee for
a zero byte in it).
strlcpy(s1, s2, l) is calling strlen(s2), potentially
causing out-of-bound access, as reported by syzbot:
detected buffer overflow in __fortify_strlen
------------[ cut here ]------------
kernel BUG at lib/string_helpers.c:980!
invalid opcode: 0000 [#1] PREEMPT SMP KASAN
CPU: 0 PID: 4087 Comm: dhcpcd-run-hooks Not tainted 5.18.0-rc3-syzkaller-01537-g20b87e7c29df #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
RIP: 0010:fortify_panic+0x18/0x1a lib/string_helpers.c:980
Code: 8c e8 c5 ba e1 fa e9 23 0f bf fa e8 0b 5d 8c f8 eb db 55 48 89 fd e8 e0 49 40 f8 48 89 ee 48 c7 c7 80 f5 26 8a e8 99 09 f1 ff <0f> 0b e8 ca 49 40 f8 48 8b 54 24 18 4c 89 f1 48 c7 c7 00 00 27 8a
RSP: 0018:ffffc900000074a8 EFLAGS: 00010286
RAX: 000000000000002c RBX: ffff88801226b728 RCX: 0000000000000000
RDX: ffff8880198e0000 RSI: ffffffff81600458 RDI: fffff52000000e87
RBP: ffffffff89da2aa0 R08: 000000000000002c R09: 0000000000000000
R10: ffffffff815fae2e R11: 0000000000000000 R12: ffff88801226b700
R13: ffff8880198e0830 R14: 0000000000000000 R15: 0000000000000000
FS: 0000000000000000(0000) GS:ffff8880b9c00000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 00007f5876ad6ff8 CR3: 000000001a48c000 CR4: 00000000003506f0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000600
Call Trace:
<IRQ>
__fortify_strlen include/linux/fortify-string.h:128 [inline]
strlcpy include/linux/fortify-string.h:143 [inline]
__set_page_owner_handle+0x2b1/0x3e0 mm/page_owner.c:171
__set_page_owner+0x3e/0x50 mm/page_owner.c:190
prep_new_page mm/page_alloc.c:2441 [inline]
get_page_from_freelist+0xba2/0x3e00 mm/page_alloc.c:4182
__alloc_pages+0x1b2/0x500 mm/page_alloc.c:5408
alloc_pages+0x1aa/0x310 mm/mempolicy.c:2272
alloc_slab_page mm/slub.c:1799 [inline]
allocate_slab+0x26c/0x3c0 mm/slub.c:1944
new_slab mm/slub.c:2004 [inline]
___slab_alloc+0x8df/0xf20 mm/slub.c:3005
__slab_alloc.constprop.0+0x4d/0xa0 mm/slub.c:3092
slab_alloc_node mm/slub.c:3183 [inline]
slab_alloc mm/slub.c:3225 [inline]
__kmem_cache_alloc_lru mm/slub.c:3232 [inline]
kmem_cache_alloc+0x360/0x3b0 mm/slub.c:3242
dst_alloc+0x146/0x1f0 net/core/dst.c:92
Fixes: 865ed6a32786 ("mm/page_owner: record task command name")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported-by: syzbot <syzkaller@googlegroups.com>
Cc: Waiman Long <longman@redhat.com>
---
mm/page_owner.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/mm/page_owner.c b/mm/page_owner.c
index fb3a05fdebdbf1c3646ba1584cbf06facf0e7a9a..19bc559e49040e60c03a5f4268c89618fa0f1b1c 100644
--- a/mm/page_owner.c
+++ b/mm/page_owner.c
@@ -168,7 +168,7 @@ static inline void __set_page_owner_handle(struct page_ext *page_ext,
page_owner->pid = current->pid;
page_owner->tgid = current->tgid;
page_owner->ts_nsec = local_clock();
- strlcpy(page_owner->comm, current->comm,
+ strscpy(page_owner->comm, current->comm,
sizeof(page_owner->comm));
__set_bit(PAGE_EXT_OWNER, &page_ext->flags);
__set_bit(PAGE_EXT_OWNER_ALLOCATED, &page_ext->flags);
--
2.36.0.512.ge40c2bad7a-goog
^ permalink raw reply related [flat|nested] 3+ messages in thread
* Re: [PATCH] mm/page_owner: use strscpy() instead of strlcpy()
2022-05-09 14:59 [PATCH] mm/page_owner: use strscpy() instead of strlcpy() Eric Dumazet
@ 2022-05-09 15:48 ` Waiman Long
2022-05-09 15:52 ` Shakeel Butt
1 sibling, 0 replies; 3+ messages in thread
From: Waiman Long @ 2022-05-09 15:48 UTC (permalink / raw)
To: Eric Dumazet, Andrew Morton; +Cc: linux-kernel, linux-mm, Eric Dumazet, syzbot
On 5/9/22 10:59, Eric Dumazet wrote:
> From: Eric Dumazet <edumazet@google.com>
>
> current->comm[] is not a string (no guarantee for
> a zero byte in it).
>
> strlcpy(s1, s2, l) is calling strlen(s2), potentially
> causing out-of-bound access, as reported by syzbot:
>
> detected buffer overflow in __fortify_strlen
> ------------[ cut here ]------------
> kernel BUG at lib/string_helpers.c:980!
> invalid opcode: 0000 [#1] PREEMPT SMP KASAN
> CPU: 0 PID: 4087 Comm: dhcpcd-run-hooks Not tainted 5.18.0-rc3-syzkaller-01537-g20b87e7c29df #0
> Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
> RIP: 0010:fortify_panic+0x18/0x1a lib/string_helpers.c:980
> Code: 8c e8 c5 ba e1 fa e9 23 0f bf fa e8 0b 5d 8c f8 eb db 55 48 89 fd e8 e0 49 40 f8 48 89 ee 48 c7 c7 80 f5 26 8a e8 99 09 f1 ff <0f> 0b e8 ca 49 40 f8 48 8b 54 24 18 4c 89 f1 48 c7 c7 00 00 27 8a
> RSP: 0018:ffffc900000074a8 EFLAGS: 00010286
>
> RAX: 000000000000002c RBX: ffff88801226b728 RCX: 0000000000000000
> RDX: ffff8880198e0000 RSI: ffffffff81600458 RDI: fffff52000000e87
> RBP: ffffffff89da2aa0 R08: 000000000000002c R09: 0000000000000000
> R10: ffffffff815fae2e R11: 0000000000000000 R12: ffff88801226b700
> R13: ffff8880198e0830 R14: 0000000000000000 R15: 0000000000000000
> FS: 0000000000000000(0000) GS:ffff8880b9c00000(0000) knlGS:0000000000000000
> CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> CR2: 00007f5876ad6ff8 CR3: 000000001a48c000 CR4: 00000000003506f0
> DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
> DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000600
> Call Trace:
> <IRQ>
> __fortify_strlen include/linux/fortify-string.h:128 [inline]
> strlcpy include/linux/fortify-string.h:143 [inline]
> __set_page_owner_handle+0x2b1/0x3e0 mm/page_owner.c:171
> __set_page_owner+0x3e/0x50 mm/page_owner.c:190
> prep_new_page mm/page_alloc.c:2441 [inline]
> get_page_from_freelist+0xba2/0x3e00 mm/page_alloc.c:4182
> __alloc_pages+0x1b2/0x500 mm/page_alloc.c:5408
> alloc_pages+0x1aa/0x310 mm/mempolicy.c:2272
> alloc_slab_page mm/slub.c:1799 [inline]
> allocate_slab+0x26c/0x3c0 mm/slub.c:1944
> new_slab mm/slub.c:2004 [inline]
> ___slab_alloc+0x8df/0xf20 mm/slub.c:3005
> __slab_alloc.constprop.0+0x4d/0xa0 mm/slub.c:3092
> slab_alloc_node mm/slub.c:3183 [inline]
> slab_alloc mm/slub.c:3225 [inline]
> __kmem_cache_alloc_lru mm/slub.c:3232 [inline]
> kmem_cache_alloc+0x360/0x3b0 mm/slub.c:3242
> dst_alloc+0x146/0x1f0 net/core/dst.c:92
>
> Fixes: 865ed6a32786 ("mm/page_owner: record task command name")
> Signed-off-by: Eric Dumazet <edumazet@google.com>
> Reported-by: syzbot <syzkaller@googlegroups.com>
> Cc: Waiman Long <longman@redhat.com>
> ---
> mm/page_owner.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/mm/page_owner.c b/mm/page_owner.c
> index fb3a05fdebdbf1c3646ba1584cbf06facf0e7a9a..19bc559e49040e60c03a5f4268c89618fa0f1b1c 100644
> --- a/mm/page_owner.c
> +++ b/mm/page_owner.c
> @@ -168,7 +168,7 @@ static inline void __set_page_owner_handle(struct page_ext *page_ext,
> page_owner->pid = current->pid;
> page_owner->tgid = current->tgid;
> page_owner->ts_nsec = local_clock();
> - strlcpy(page_owner->comm, current->comm,
> + strscpy(page_owner->comm, current->comm,
> sizeof(page_owner->comm));
> __set_bit(PAGE_EXT_OWNER, &page_ext->flags);
> __set_bit(PAGE_EXT_OWNER_ALLOCATED, &page_ext->flags);
Yes, I think it is more correct to use strscpy().
Acked-by: Waiman Long <longman@redhat.com>
Thanks,
Longman
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [PATCH] mm/page_owner: use strscpy() instead of strlcpy()
2022-05-09 14:59 [PATCH] mm/page_owner: use strscpy() instead of strlcpy() Eric Dumazet
2022-05-09 15:48 ` Waiman Long
@ 2022-05-09 15:52 ` Shakeel Butt
1 sibling, 0 replies; 3+ messages in thread
From: Shakeel Butt @ 2022-05-09 15:52 UTC (permalink / raw)
To: Eric Dumazet
Cc: Andrew Morton, linux-kernel, linux-mm, Eric Dumazet, syzbot, Waiman Long
On Mon, May 09, 2022 at 07:59:49AM -0700, Eric Dumazet wrote:
> From: Eric Dumazet <edumazet@google.com>
>
> current->comm[] is not a string (no guarantee for
> a zero byte in it).
>
> strlcpy(s1, s2, l) is calling strlen(s2), potentially
> causing out-of-bound access, as reported by syzbot:
>
> detected buffer overflow in __fortify_strlen
> ------------[ cut here ]------------
> kernel BUG at lib/string_helpers.c:980!
> invalid opcode: 0000 [#1] PREEMPT SMP KASAN
> CPU: 0 PID: 4087 Comm: dhcpcd-run-hooks Not tainted 5.18.0-rc3-syzkaller-01537-g20b87e7c29df #0
> Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
> RIP: 0010:fortify_panic+0x18/0x1a lib/string_helpers.c:980
> Code: 8c e8 c5 ba e1 fa e9 23 0f bf fa e8 0b 5d 8c f8 eb db 55 48 89 fd e8 e0 49 40 f8 48 89 ee 48 c7 c7 80 f5 26 8a e8 99 09 f1 ff <0f> 0b e8 ca 49 40 f8 48 8b 54 24 18 4c 89 f1 48 c7 c7 00 00 27 8a
> RSP: 0018:ffffc900000074a8 EFLAGS: 00010286
>
> RAX: 000000000000002c RBX: ffff88801226b728 RCX: 0000000000000000
> RDX: ffff8880198e0000 RSI: ffffffff81600458 RDI: fffff52000000e87
> RBP: ffffffff89da2aa0 R08: 000000000000002c R09: 0000000000000000
> R10: ffffffff815fae2e R11: 0000000000000000 R12: ffff88801226b700
> R13: ffff8880198e0830 R14: 0000000000000000 R15: 0000000000000000
> FS: 0000000000000000(0000) GS:ffff8880b9c00000(0000) knlGS:0000000000000000
> CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> CR2: 00007f5876ad6ff8 CR3: 000000001a48c000 CR4: 00000000003506f0
> DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
> DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000600
> Call Trace:
> <IRQ>
> __fortify_strlen include/linux/fortify-string.h:128 [inline]
> strlcpy include/linux/fortify-string.h:143 [inline]
> __set_page_owner_handle+0x2b1/0x3e0 mm/page_owner.c:171
> __set_page_owner+0x3e/0x50 mm/page_owner.c:190
> prep_new_page mm/page_alloc.c:2441 [inline]
> get_page_from_freelist+0xba2/0x3e00 mm/page_alloc.c:4182
> __alloc_pages+0x1b2/0x500 mm/page_alloc.c:5408
> alloc_pages+0x1aa/0x310 mm/mempolicy.c:2272
> alloc_slab_page mm/slub.c:1799 [inline]
> allocate_slab+0x26c/0x3c0 mm/slub.c:1944
> new_slab mm/slub.c:2004 [inline]
> ___slab_alloc+0x8df/0xf20 mm/slub.c:3005
> __slab_alloc.constprop.0+0x4d/0xa0 mm/slub.c:3092
> slab_alloc_node mm/slub.c:3183 [inline]
> slab_alloc mm/slub.c:3225 [inline]
> __kmem_cache_alloc_lru mm/slub.c:3232 [inline]
> kmem_cache_alloc+0x360/0x3b0 mm/slub.c:3242
> dst_alloc+0x146/0x1f0 net/core/dst.c:92
>
> Fixes: 865ed6a32786 ("mm/page_owner: record task command name")
> Signed-off-by: Eric Dumazet <edumazet@google.com>
> Reported-by: syzbot <syzkaller@googlegroups.com>
> Cc: Waiman Long <longman@redhat.com>
Acked-by: Shakeel Butt <shakeelb@google.com>
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2022-05-09 15:52 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2022-05-09 14:59 [PATCH] mm/page_owner: use strscpy() instead of strlcpy() Eric Dumazet
2022-05-09 15:48 ` Waiman Long
2022-05-09 15:52 ` Shakeel Butt
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).