CVE-2026-64109: af_unix: Fix UAF read of tail->len in unix_stream_data_wait()

In the Linux kernel, the following vulnerability has been resolved:

afunix: Fix UAF read of tail->len in unixstreamdatawait()

unixstreamdatawait() does skbpeektail(&sk->skreceivequeue) without holding any lock that prevents SKBs on that queue from being dequeued and freed. This has been the case since commit 79f632c71bea ("unix/stream: fix peeking with an offset larger than data in queue"). The first consequence of this is that the pointer comparison tail != last can be false even if last semantically refers to an already-freed SKB while tail is a new SKB allocated at the same address; which can cause unixstreamdatawait() to wrongly keep blocking after new data has arrived, but only in a weird scenario where a peeking recv() and a normal recv() on the same socket are racing, which is probably not a real problem.

But since commit 2b514574f7e8 ("net: afunix: implement splice for stream afunix sockets"), tail is actually dereferenced, which can cause UAF in the following race scenario (where testsetup() runs single-threaded, and afterwards, testthread1() and testthread2() run concurrently in two threads: static int socks[2]; void testsetup(void) { socketpair(AFUNIX, SOCKSTREAM, 0, socks); send(socks[1], "A", 1, 0); int peekoff = 1; setsockopt(socks[0], SOLSOCKET, SOPEEKOFF, &peekoff, sizeof(peekoff)); } void testthread1(void) { char dummy; recv(socks[0], &dummy, 1, MSGPEEK); } void testthread2(void) { char dummy; recv(socks[0], &dummy, 1, 0); shutdown(socks[1], SHUTWR); }

when racing like this: thread1 thread2 unixstreamreadgeneric mutexlock(&u->iolock) skbpeek(&sk->skreceivequeue) skbpeeknext(skb, &sk->skreceivequeue) mutexunlock(&u->iolock) unixstreamreadgeneric unixstatelock(sk) skbpeek(&sk->skreceivequeue) unixstateunlock(sk) unixstreamdatawait unixstatelock(sk) tail = skbpeektail(&sk->skreceivequeue) spinlock(&sk->skreceivequeue.lock) skbunlink(skb, &sk->skreceivequeue) spinunlock(&sk->skreceivequeue.lock) consumeskb(skb) [frees the SKB] tail != last: false tail: true tail->len != lastlen UAF

Fix the UAF by removing the read of tail->len; checking tail->len would only make sense if SKBs in the receive queue of a UNIX socket could grow, which can no longer happen.

Kuniyuki explained:

When commit 869e7c62486e ("net: afunix: implement stream sendpage support") added sendpage() support, data could be appended to the last skb in the receiver's queue. That's why we needed to check if the length of the last skb was changed while waiting for new data in unixstreamdatawait(). However, commit a0dbf5f818f9 ("afunix: Support MSGSPLICEPAGES") and commit 57d44a354a43 ("unix: Convert unixstreamsendpage() to use MSGSPLICEPAGES") refactored sendmsg(), and now data is always added to a new skb.

That means this fix is not suitable for kernels before 6.5.