In the Linux kernel, the following vulnerability has been resolved:
9p: set req refcount to zero to avoid uninitialized usage
When a new request is allocated, the refcount will be zero if it is reused, but if the request is newly allocated from slab, it is not fully initialized before being added to idr.
If the p9readwork got a response before the refcount initiated. It will use a uninitialized req, which will result in a bad request data struct.
Here is the logs from syzbot.
Corrupted memory at 0xffff88807eade00b [ 0xff 0x07 0x00 0x00 0x00 0x00 0x00 0x00 . . . . . . . . ] (in kfence-#110): p9fcallfini net/9p/client.c:248 [inline] p9reqput net/9p/client.c:396 [inline] p9reqput+0x208/0x250 net/9p/client.c:390 p9clientwalk+0x247/0x540 net/9p/client.c:1165 clonefid fs/9p/fid.h:21 [inline] v9fsfidxattrset+0xe4/0x2b0 fs/9p/xattr.c:118 v9fsxattrset fs/9p/xattr.c:100 [inline] v9fsxattrhandler_set+0x6f/0x120 fs/9p/xattr.c:159 __vfs_setxattr+0x119/0x180 fs/xattr.c:182 __vfssetxattrnoperm+0x129/0x5f0 fs/xattr.c:216 __vfssetxattrlocked+0x1d3/0x260 fs/xattr.c:277 vfssetxattr+0x143/0x340 fs/xattr.c:309 setxattr+0x146/0x160 fs/xattr.c:617 pathsetxattr+0x197/0x1c0 fs/xattr.c:636 __dosyssetxattr fs/xattr.c:652 [inline] __sesyssetxattr fs/xattr.c:648 [inline] __ia32syssetxattr+0xc0/0x160 fs/xattr.c:648 dosyscall32irqson arch/x86/entry/common.c:112 [inline] __dofastsyscall32+0x65/0xf0 arch/x86/entry/common.c:178 dofastsyscall32+0x33/0x70 arch/x86/entry/common.c:203 entrySYSENTERcompatafterhwframe+0x70/0x82
Below is a similar scenario, the scenario in the syzbot log looks more complicated than this one, but this patch can fix it.
T21124 p9_read_work
======================== second trans ================================= p9clientwalk p9clientrpc p9clientpreparereq p9tagalloc req = kmemcachealloc(p9reqcache, GFPNOFS); tag = idralloc << preempted >> req->tc.tag = tag; /* req->[refcount/tag] == uninitialized */ m->rreq = p9tag_lookup(m->client, m->rc.tag); /* increments uninitalized refcount */
refcount_set(&req->refcount, 2);
/* cb drops one ref */
p9_client_cb(req)
/* reader thread drops its ref:
request is incorrectly freed */
p9_req_put(req)
/* use after free and ref underflow */
p9_req_put(req)
To fix it, we can initialize the refcount to zero before add to idr.
{
"osv_generated_from": "https://github.com/CVEProject/cvelistV5/tree/main/cves/2022/50xxx/CVE-2022-50335.json",
"cna_assigner": "Linux"
}