CVE-2026-68386: bpf, sockmap: Reject unhashed UDP sockets on sockmap update

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

bpf, sockmap: Reject unhashed UDP sockets on sockmap update

UDP sockets get SOCKRCUFREE set when (auto-)bound. This means skisrefcounted(unbound) = true, while skisrefcounted(bound) = false.

Because sockmap accepts unbound UDP sockets, a BPF program can increment a socket's refcount via lookup. If the socket is subsequently bound, the transition from unbound to bound causes bpfskrelease() to skip the decrement of the refcount, causing a memory leak.

unreferenced object 0xffff88810bc2eb40 (size 1984): comm "testprogs", pid 2451, jiffies 4295320596 hex dump (first 32 bytes): 7f 00 00 01 7f 00 00 01 d2 04 1b b7 04 d2 00 00 ................ 02 00 01 40 00 00 00 00 00 00 00 00 00 00 00 00 ...@............ backtrace (crc bdee079d): kmemcacheallocnoprof+0x557/0x660 skprotalloc+0x69/0x240 skalloc+0x30/0x460 inetcreate+0x2ce/0xf80 sockcreate+0x25b/0x5c0 syssocket+0x119/0x1d0 x64syssocket+0x72/0xd0 dosyscall64+0xa1/0x5f0 entrySYSCALL64afterhwframe+0x76/0x7e

Instead of special-casing for refcounted sockets, reject unhashed UDP sockets during sockmap updates, as there is no benefit to supporting those. This effectively reverts the commit under Fixes, with two exceptions:

1. sockmapskstateallowed() maintains a fall-through return true. 2. In the spirit of commit b8b8315e39ff ("bpf, sockmap: Remove unhash handler for BPF sockmap usage"), the proto::unhash BPF handler is not reintroduced.

Historical note: this issue is related to commit 67312adc96b5 ("bpf: reject unhashed sockets in bpfskassign").