Re: [PATCH bpf v2 3/4] bpf, sockmap: Adapt for the af_unix-specific lock

[Michal Luczaj]

On 2/7/26 23:00, Kuniyuki Iwashima wrote:
> On Sat, Feb 7, 2026 at 6:35 AM Michal Luczaj <mhal@xxxxxxx> wrote:
>> This patch also happens to fix a deadlock that may occur when
>> bpf_iter_unix_seq_show()'s lock_sock_fast() takes the fast path and the
>> iter prog attempts to update a sockmap. Which ends up spinning at
>> sock_map_update_elem()'s bh_lock_sock():
> 
> Hmm.. this seems to be a more general problem for
> bpf iter vs sockmap.  bpf_iter_{tcp,udp}_seq_show() also
> hold lock_sock(),  where this patch's solution does not help.
> We need to resolve this regardless of socket family.

I don't see any deadlocks there. Note that I've mentioned lock_sock_fast()
fast path was a problem, not lock_sock().

> Also, I feel lock_sock() should be outside of unix_state_lock()
> since the former is usually sleepable.  If we used such locking
> order in the future, it would trigger ABBA deadlock and we would
> have to revisit this problem.

Do I understand correctly you're suggesting taking lock_sock() for af_unix
just as well?