When the head of the wait queue claims the lock and its cmpxchg from n,0,0 to 0,0,1 fails, it sets the locked bit and waits for node->next, with no timeout. The comment, taken from qspinlock, says the cmpxchg can only fail if somebody queued behind us, or if a waiter set PENDING, which will then see our tail and queue behind us. In rqspinlock that waiter does not always queue. After clear_pending() it goes to queue:, and two paths return -EDEADLK there before xchg_tail(): - the trylock fallback for an NMI waiter whose per-CPU node is busy (in_nmi() && idx > 0), or when idx >= _Q_MAX_NODES; - check_deadlock_AA(), when this CPU already holds an entry for the lock. For example, with the queue head H alone in the queue: CPU 0 (H, head) CPU 1 (NMI, idx 1 on another lock) reads lock->val == H,0,0 queued_fetch_set_pending_acquire() sees tail H try_cmpxchg(H,0,0 -> 0,0,1) fails set_locked() -> H,1,1 clear_pending() -> H,0,1 in_nmi() && idx > 0, trylock fails return -EDEADLK waits for node->next ... H now owns the lock and spins until another context queues on the same lock. For a lock that is rarely taken, such as a hash map bucket lock, that can be a long time, with interrupts disabled. The same happens on v6.17 and v6.18 for any NMI waiter, as the trylock fallback was used for all of them. The AA path needs the interrupted context to have taken its entry for the lock and not to try the lock again, so it is much narrower. Only a different tail shows that a waiter did xchg_tail() and will link into our node. So while the tail is still ours, retry the claim and keep the PENDING bit of a waiter that may still be backing out, and set the locked bit and wait for node->next only once the tail has moved. The loop that waits for node->next is unchanged. Found with a TLA+ model of the slow path. With this change the model has no violation in the configurations where the head hangs without it (an NMI waiter using the trylock fallback, and the AA bail-out). A userspace port of the slow path, driven through the interleaving above, hangs without this change and completes with the lock word clean with it. I have not reproduced the hang on a running kernel. Fixes: 0d80e7f951be ("rqspinlock: Choose trylock fallback for NMI waiters") Fixes: 087849cca31d ("rqspinlock: Precede non-head waiter queueing with AA check") Suggested-by: Kumar Kartikeya Dwivedi Assisted-by: LLM Signed-off-by: Shashank Mohan Jain --- Changes in v2: - Use Kumar's approach: retry the claim while the tail is still ours, keeping PENDING, instead of resetting the tail in the loop that waits for node->next, so that loop is not slowed down for this rare case. The diff is Kumar's from the v1 review. - Target bpf-next and drop Cc: stable (Kumar). - Checked the new approach in the TLA+ model (no violation in the NMI trylock and AA configurations that hang without a fix) and in the userspace replay of the interleaving (completes, lock word clean). v1: https://lore.kernel.org/all/20261117225023.36660-1-jain.sm@gmail.com/ Prepared with Claude Code (Anthropic), model Claude Opus 5.5 (claude-opus-5-5). Based on bpf-next 32b66b94c86f. kernel/bpf/rqspinlock.c | 23 +++++++++++++---------- 1 file changed, 13 insertions(+), 10 deletions(-) diff --git a/kernel/bpf/rqspinlock.c b/kernel/bpf/rqspinlock.c index 111ec80..f6a938e 100644 --- a/kernel/bpf/rqspinlock.c +++ b/kernel/bpf/rqspinlock.c @@ -624,20 +624,23 @@ waitq_timeout: * Otherwise, we only need to grab the lock. */ - /* - * Note: at this point: (val & _Q_PENDING_MASK) == 0, because of the - * above wait condition, therefore any concurrent setting of - * PENDING will make the uncontended transition fail. - */ if ((val & _Q_TAIL_MASK) == tail) { - if (atomic_try_cmpxchg_relaxed(&lock->val, &val, _Q_LOCKED_VAL)) - goto release; /* No contention */ + u32 new = _Q_LOCKED_VAL; + + for (;;) { + if (atomic_try_cmpxchg_relaxed(&lock->val, &val, new)) + goto release; /* No queued successor */ + if ((val & _Q_TAIL_MASK) != tail) + break; + /* A pending claimant can abandon before publishing a tail. */ + new = _Q_LOCKED_VAL | (val & _Q_PENDING_MASK); + } } /* - * Either somebody is queued behind us or _Q_PENDING_VAL got set - * which will then detect the remaining tail and queue behind us - * ensuring we'll see a @next. + * A different tail proves that a successor will publish node->next. + * Preserve pending while installing ownership; a racing claimant may + * still be undoing its speculative reservation against this queue. */ set_locked(lock); base-commit: 32b66b94c86f009b49668de7340f2ed65aad2dc5 -- 2.54.0 (Apple Git-157)