The branch main has been updated by pkubaj: URL: https://cgit.FreeBSD.org/src/commit/?id=c5d9485e7d7faeba1019301ed83a781dac2d0cb0
commit c5d9485e7d7faeba1019301ed83a781dac2d0cb0 Author: Piotr Kubaj <[email protected]> AuthorDate: 2026-09-03 14:08:45 +0000 Commit: Piotr Kubaj <[email protected]> CommitDate: 2026-09-04 09:27:49 +0000 umtx: do not sleep on an unowned mutex after a spurious CAS failure On ll/sc architectures casueword32() may report a spurious store-conditional failure (reservation lost to an interrupt, preemption, or another CPU touching the same reservation granule), and this is indistinguishable from a genuine comparison mismatch: both return 1. That is intentional since D20772 and documented in casueword(9) ("The store can fail on load-linked/store-conditional architectures."), so callers must cope. do_lock_normal() does not fully cope. When the initial UMUTEX_UNOWNED -> id acquire CAS fails spuriously, the observed owner is still UMUTEX_UNOWNED, so neither the UMUTEX_CONTESTED branch nor the real-owner case applies, and execution falls through past the "rv == 1 but not contested, likely store failure" comment into the sleep path. There, the contested-bit CAS (expecting the observed owner, i.e. UMUTEX_UNOWNED) succeeds because the mutex really is unowned, stamping m_owner = UMUTEX_CONTESTED with no owner tid, and the thread sleeps on "umtxn" forever: nobody owns the mutex, so no unlock and no wakeup ever arrive. In _UMUTEX_TRY mode the same situation returns a false EBUSY for a free mutex. Treat an observed owner of UMUTEX_UNOWNED like UMUTEX_CONTESTED: try to acquire the mutex, setting the contested bit, instead of falling through to the sleep path. rv == 1 with the observed value equal to the expected value can only mean a spurious store failure, so the mutex is free. If the acquire CAS fails again, the outer loop restarts and re-evaluates ownership. The contested bit set with no waiters present only costs the matching unlock one trip through the kernel. This was hit in practice on powerpc64le (POWER9): the Swift runtime's Synchronization.Mutex issues _umtx_op(UMTX_OP_MUTEX_LOCK) directly with no userspace fast path, so an uncontended lock of an unowned mutex runs the kernel CAS exactly where a spurious failure deadlocks (single-threaded process parked on "umtxn" with m_owner == 0x80000000, observed as Foundation.Process.run() hanging). libthr mostly masks the bug because pthread_mutex_lock() enters the kernel only when there is a real owner that will eventually issue a wakeup. The mechanism was confirmed with an experimental powerpc kernel that instead retried the ll/sc sequence inside casueword32()/casueword(); that also eliminated the hang, but is not proposed here since the single-attempt semantics of casueword(9) are intentional. Reviewed by: kib MFC after: 2 weeks Differential Revision: https://reviews.freebsd.org/D59338 --- sys/kern/kern_umtx.c | 20 ++++++++++++-------- 1 file changed, 12 insertions(+), 8 deletions(-) diff --git a/sys/kern/kern_umtx.c b/sys/kern/kern_umtx.c index ac2469379690..55ef536d5b98 100644 --- a/sys/kern/kern_umtx.c +++ b/sys/kern/kern_umtx.c @@ -1423,19 +1423,23 @@ do_lock_normal(struct thread *td, struct umutex *m, uint32_t flags, } /* - * If no one owns it but it is contested try - * to acquire it. + * If no one owns it, but it is contested or + * the CAS above failed spuriously (possible + * on ll/sc architectures), try to acquire it. + * Sleeping would be forever in the spurious + * case: no owner exists to wake us. */ MPASS(rv == 1); - if (owner == UMUTEX_CONTESTED) { - rv = casueword32(&m->m_owner, - UMUTEX_CONTESTED, &owner, - id | UMUTEX_CONTESTED); + if (owner == UMUTEX_CONTESTED || + owner == UMUTEX_UNOWNED) { + rv = casueword32(&m->m_owner, owner, + &owner, id | UMUTEX_CONTESTED); /* The address was invalid. */ if (rv == -1) return (EFAULT); if (rv == 0) { - MPASS(owner == UMUTEX_CONTESTED); + MPASS(owner == UMUTEX_CONTESTED || + owner == UMUTEX_UNOWNED); return (0); } if (rv == 1) { @@ -1451,7 +1455,7 @@ do_lock_normal(struct thread *td, struct umutex *m, uint32_t flags, continue; } - /* rv == 1 but not contested, likely store failure */ + /* rv == 1 with a real owner, fall through to sleep. */ rv = thread_check_susp(td, false); if (rv != 0) return (rv);
