This is the mail archive of the
glibc-bugs@sourceware.org
mailing list for the glibc project.
[Bug nptl/18463] pthread_cond_broadcast issue when surrounded by PTHREAD_PRIO_INHERIT mutex on ARM
- From: "mkl at pengutronix dot de" <sourceware-bugzilla at sourceware dot org>
- To: glibc-bugs at sourceware dot org
- Date: Sat, 14 May 2016 18:41:18 +0000
- Subject: [Bug nptl/18463] pthread_cond_broadcast issue when surrounded by PTHREAD_PRIO_INHERIT mutex on ARM
- Auto-submitted: auto-generated
- References: <bug-18463-131 at http dot sourceware dot org/bugzilla/>
https://sourceware.org/bugzilla/show_bug.cgi?id=18463
--- Comment #12 from Marc Kleine-Budde <mkl at pengutronix dot de> ---
(In reply to Adhemerval Zanella from comment #9)
> Now that minimum kernel version support is 3.2 we can assume that all PI
> futuxes operations as covered with cmpxchg enabled check
> (29bfcea07745737f385b0d092e71527051c29029 [1]), so calling futexes with
> FUTEX_WAIT_REQUEUE_PI or FUTEX_CMP_REQUEUE_PI will result in ENOSYS if cmpxchg
> is not enabled.
A mutex with PTHREAD_PRIO_INHERIT setup will fail on kernel if
futex_cmpxchg_enabled is false - as it calls FUTEX_UNLOCK_PI to test if the
kernel supports it. So a FUTEX_WAIT_REQUEUE_PI or FUTEX_CMP_REQUEUE_PI will
_not_ be used later, because the mutex cannot be initialed in the first place.
--
You are receiving this mail because:
You are on the CC list for the bug.