* [PATCH v2] cachefiles: Fix KASAN slab-out-of-bounds in cachefiles_set_volume_xattr
@ 2022-04-05 14:28 Dave Wysochanski
2022-04-05 15:14 ` Jeff Layton
2022-04-08 22:13 ` David Howells
0 siblings, 2 replies; 3+ messages in thread
From: Dave Wysochanski @ 2022-04-05 14:28 UTC (permalink / raw)
To: dhowells; +Cc: linux-cachefs, jlayton, smfrench, linux-cifs
Use the actual length of volume coherency data when setting the
xattr to avoid the following KASAN report.
BUG: KASAN: slab-out-of-bounds in cachefiles_set_volume_xattr+0xa0/0x350 [cachefiles]
Write of size 4 at addr ffff888101e02af4 by task kworker/6:0/1347
CPU: 6 PID: 1347 Comm: kworker/6:0 Kdump: loaded Not tainted 5.18.0-rc1-nfs-fscache-netfs+ #13
Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.14.0-4.fc34 04/01/2014
Workqueue: events fscache_create_volume_work [fscache]
Call Trace:
<TASK>
dump_stack_lvl+0x45/0x5a
print_report.cold+0x5e/0x5db
? __lock_text_start+0x8/0x8
? cachefiles_set_volume_xattr+0xa0/0x350 [cachefiles]
kasan_report+0xab/0x120
? cachefiles_set_volume_xattr+0xa0/0x350 [cachefiles]
kasan_check_range+0xf5/0x1d0
memcpy+0x39/0x60
cachefiles_set_volume_xattr+0xa0/0x350 [cachefiles]
cachefiles_acquire_volume+0x2be/0x500 [cachefiles]
? __cachefiles_free_volume+0x90/0x90 [cachefiles]
fscache_create_volume_work+0x68/0x160 [fscache]
process_one_work+0x3b7/0x6a0
worker_thread+0x2c4/0x650
? process_one_work+0x6a0/0x6a0
kthread+0x16c/0x1a0
? kthread_complete_and_exit+0x20/0x20
ret_from_fork+0x22/0x30
</TASK>
Allocated by task 1347:
kasan_save_stack+0x1e/0x40
__kasan_kmalloc+0x81/0xa0
cachefiles_set_volume_xattr+0x76/0x350 [cachefiles]
cachefiles_acquire_volume+0x2be/0x500 [cachefiles]
fscache_create_volume_work+0x68/0x160 [fscache]
process_one_work+0x3b7/0x6a0
worker_thread+0x2c4/0x650
kthread+0x16c/0x1a0
ret_from_fork+0x22/0x30
The buggy address belongs to the object at ffff888101e02af0
which belongs to the cache kmalloc-8 of size 8
The buggy address is located 4 bytes inside of
8-byte region [ffff888101e02af0, ffff888101e02af8)
The buggy address belongs to the physical page:
page:00000000a2292d70 refcount:1 mapcount:0 mapping:0000000000000000 index:0x0 pfn:0x101e02
flags: 0x17ffffc0000200(slab|node=0|zone=2|lastcpupid=0x1fffff)
raw: 0017ffffc0000200 0000000000000000 dead000000000001 ffff888100042280
raw: 0000000000000000 0000000080660066 00000001ffffffff 0000000000000000
page dumped because: kasan: bad access detected
Memory state around the buggy address:
ffff888101e02980: fc 00 fc fc fc fc 00 fc fc fc fc 00 fc fc fc fc
ffff888101e02a00: 00 fc fc fc fc 00 fc fc fc fc 00 fc fc fc fc 00
>ffff888101e02a80: fc fc fc fc 00 fc fc fc fc 00 fc fc fc fc 04 fc
^
ffff888101e02b00: fc fc fc 00 fc fc fc fc 00 fc fc fc fc 00 fc fc
ffff888101e02b80: fc fc 00 fc fc fc fc 00 fc fc fc fc 00 fc fc fc
==================================================================
Fixes: 413a4a6b0b55 "cachefiles: Fix volume coherency attribute"
Signed-off-by: Dave Wysochanski <dwysocha@redhat.com>
---
fs/cachefiles/xattr.c | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/fs/cachefiles/xattr.c b/fs/cachefiles/xattr.c
index 35465109d9c4..f13b642c1d14 100644
--- a/fs/cachefiles/xattr.c
+++ b/fs/cachefiles/xattr.c
@@ -198,8 +198,7 @@ bool cachefiles_set_volume_xattr(struct cachefiles_volume *volume)
_enter("%x,#%d", volume->vcookie->debug_id, len);
- len += sizeof(*buf);
- buf = kmalloc(len, GFP_KERNEL);
+ buf = kmalloc(sizeof(*buf) + len, GFP_KERNEL);
if (!buf)
return false;
buf->reserved = cpu_to_be32(0);
--
2.27.1
^ permalink raw reply related [flat|nested] 3+ messages in thread
* Re: [PATCH v2] cachefiles: Fix KASAN slab-out-of-bounds in cachefiles_set_volume_xattr
2022-04-05 14:28 [PATCH v2] cachefiles: Fix KASAN slab-out-of-bounds in cachefiles_set_volume_xattr Dave Wysochanski
@ 2022-04-05 15:14 ` Jeff Layton
2022-04-08 22:13 ` David Howells
1 sibling, 0 replies; 3+ messages in thread
From: Jeff Layton @ 2022-04-05 15:14 UTC (permalink / raw)
To: Dave Wysochanski, dhowells; +Cc: linux-cachefs, smfrench, linux-cifs
On Tue, 2022-04-05 at 10:28 -0400, Dave Wysochanski wrote:
> Use the actual length of volume coherency data when setting the
> xattr to avoid the following KASAN report.
>
> BUG: KASAN: slab-out-of-bounds in cachefiles_set_volume_xattr+0xa0/0x350 [cachefiles]
> Write of size 4 at addr ffff888101e02af4 by task kworker/6:0/1347
>
> CPU: 6 PID: 1347 Comm: kworker/6:0 Kdump: loaded Not tainted 5.18.0-rc1-nfs-fscache-netfs+ #13
> Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.14.0-4.fc34 04/01/2014
> Workqueue: events fscache_create_volume_work [fscache]
> Call Trace:
> <TASK>
> dump_stack_lvl+0x45/0x5a
> print_report.cold+0x5e/0x5db
> ? __lock_text_start+0x8/0x8
> ? cachefiles_set_volume_xattr+0xa0/0x350 [cachefiles]
> kasan_report+0xab/0x120
> ? cachefiles_set_volume_xattr+0xa0/0x350 [cachefiles]
> kasan_check_range+0xf5/0x1d0
> memcpy+0x39/0x60
> cachefiles_set_volume_xattr+0xa0/0x350 [cachefiles]
> cachefiles_acquire_volume+0x2be/0x500 [cachefiles]
> ? __cachefiles_free_volume+0x90/0x90 [cachefiles]
> fscache_create_volume_work+0x68/0x160 [fscache]
> process_one_work+0x3b7/0x6a0
> worker_thread+0x2c4/0x650
> ? process_one_work+0x6a0/0x6a0
> kthread+0x16c/0x1a0
> ? kthread_complete_and_exit+0x20/0x20
> ret_from_fork+0x22/0x30
> </TASK>
>
> Allocated by task 1347:
> kasan_save_stack+0x1e/0x40
> __kasan_kmalloc+0x81/0xa0
> cachefiles_set_volume_xattr+0x76/0x350 [cachefiles]
> cachefiles_acquire_volume+0x2be/0x500 [cachefiles]
> fscache_create_volume_work+0x68/0x160 [fscache]
> process_one_work+0x3b7/0x6a0
> worker_thread+0x2c4/0x650
> kthread+0x16c/0x1a0
> ret_from_fork+0x22/0x30
>
> The buggy address belongs to the object at ffff888101e02af0
> which belongs to the cache kmalloc-8 of size 8
> The buggy address is located 4 bytes inside of
> 8-byte region [ffff888101e02af0, ffff888101e02af8)
>
> The buggy address belongs to the physical page:
> page:00000000a2292d70 refcount:1 mapcount:0 mapping:0000000000000000 index:0x0 pfn:0x101e02
> flags: 0x17ffffc0000200(slab|node=0|zone=2|lastcpupid=0x1fffff)
> raw: 0017ffffc0000200 0000000000000000 dead000000000001 ffff888100042280
> raw: 0000000000000000 0000000080660066 00000001ffffffff 0000000000000000
> page dumped because: kasan: bad access detected
>
> Memory state around the buggy address:
> ffff888101e02980: fc 00 fc fc fc fc 00 fc fc fc fc 00 fc fc fc fc
> ffff888101e02a00: 00 fc fc fc fc 00 fc fc fc fc 00 fc fc fc fc 00
> >ffff888101e02a80: fc fc fc fc 00 fc fc fc fc 00 fc fc fc fc 04 fc
> ^
> ffff888101e02b00: fc fc fc 00 fc fc fc fc 00 fc fc fc fc 00 fc fc
> ffff888101e02b80: fc fc 00 fc fc fc fc 00 fc fc fc fc 00 fc fc fc
> ==================================================================
>
> Fixes: 413a4a6b0b55 "cachefiles: Fix volume coherency attribute"
> Signed-off-by: Dave Wysochanski <dwysocha@redhat.com>
> ---
> fs/cachefiles/xattr.c | 3 +--
> 1 file changed, 1 insertion(+), 2 deletions(-)
>
> diff --git a/fs/cachefiles/xattr.c b/fs/cachefiles/xattr.c
> index 35465109d9c4..f13b642c1d14 100644
> --- a/fs/cachefiles/xattr.c
> +++ b/fs/cachefiles/xattr.c
> @@ -198,8 +198,7 @@ bool cachefiles_set_volume_xattr(struct cachefiles_volume *volume)
>
> _enter("%x,#%d", volume->vcookie->debug_id, len);
>
> - len += sizeof(*buf);
> - buf = kmalloc(len, GFP_KERNEL);
> + buf = kmalloc(sizeof(*buf) + len, GFP_KERNEL);
> if (!buf)
> return false;
> buf->reserved = cpu_to_be32(0);
I hit this bug earlier today too.
Reviewed-and-Tested-by: Jeff Layton <jlayton@kernel.org>
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [PATCH v2] cachefiles: Fix KASAN slab-out-of-bounds in cachefiles_set_volume_xattr
2022-04-05 14:28 [PATCH v2] cachefiles: Fix KASAN slab-out-of-bounds in cachefiles_set_volume_xattr Dave Wysochanski
2022-04-05 15:14 ` Jeff Layton
@ 2022-04-08 22:13 ` David Howells
1 sibling, 0 replies; 3+ messages in thread
From: David Howells @ 2022-04-08 22:13 UTC (permalink / raw)
To: Dave Wysochanski; +Cc: dhowells, linux-cachefs, jlayton, smfrench, linux-cifs
Dave Wysochanski <dwysocha@redhat.com> wrote:
> - len += sizeof(*buf);
> - buf = kmalloc(len, GFP_KERNEL);
> + buf = kmalloc(sizeof(*buf) + len, GFP_KERNEL);
Okay, your V2 is wrong and your V1 is correct (len must include the reserverd
word so that it gets saved onto disk).
David
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2022-04-08 22:14 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2022-04-05 14:28 [PATCH v2] cachefiles: Fix KASAN slab-out-of-bounds in cachefiles_set_volume_xattr Dave Wysochanski
2022-04-05 15:14 ` Jeff Layton
2022-04-08 22:13 ` David Howells
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox