New implementation of pthread condition variables in rawhide

[Siddhesh Poyarekar]

Hi,

The latest glibc in rawhide (2.21.90-13.fc23) has a new implementation
for pthread condition variable functions (pthread_cond_wait,
pthread_cond_timedwait, etc.).  If you experience any weird locking
issues or bugs that can be attributed to pthread_cond_* functions
potentially misbehaving, please cc one of us (Carlos, Torvald or me)
in the bug so that we can check if it is a bug in the new
implementation.

Thanks,
Siddhesh

[Dominik 'Rathann' Mierzejewski]

On Friday, 22 May 2015 at 06:57, Siddhesh Poyarekar wrote:
>  Hi,
>  
>  The latest glibc in rawhide (2.21.90-13.fc23) has a new implementation
>  for pthread condition variable functions (pthread_cond_wait,
>  pthread_cond_timedwait, etc.).  If you experience any weird locking
>  issues or bugs that can be attributed to pthread_cond_* functions
>  potentially misbehaving, please cc one of us (Carlos, Torvald or me)
>  in the bug so that we can check if it is a bug in the new
>  implementation. 
Could you take a look at:
https://bugzilla.redhat.com/show_bug.cgi?id=1219646
https://bugzilla.redhat.com/show_bug.cgi?id=1223055
https://bugzilla.redhat.com/show_bug.cgi?id=1222551
These seem to be related.

Regards,
Dominik
-- 
Fedora http://fedoraproject.org/wiki/User:Rathann
RPMFusion http://rpmfusion.org | MPlayer http://mplayerhq.hu
"Faith manages."
        -- Delenn to Lennier in Babylon 5:"Confessions and Lamentations"

[Siddhesh Poyarekar]

On Fri, May 22, 2015 at 02:39:11PM +0200, Dominik 'Rathann' Mierzejewski wrote:
>  On Friday, 22 May 2015 at 06:57, Siddhesh Poyarekar wrote:
>  > Hi,
>  > 
>  > The latest glibc in rawhide (2.21.90-13.fc23) has a new implementation
>  > for pthread condition variable functions (pthread_cond_wait,
>  > pthread_cond_timedwait, etc.).  If you experience any weird locking
>  > issues or bugs that can be attributed to pthread_cond_* functions
>  > potentially misbehaving, please cc one of us (Carlos, Torvald or me)
>  > in the bug so that we can check if it is a bug in the new
>  > implementation.
>  
>  Could you take a look at:
>  https://bugzilla.redhat.com/show_bug.cgi?id=1219646
>  https://bugzilla.redhat.com/show_bug.cgi?id=1223055
>  https://bugzilla.redhat.com/show_bug.cgi?id=1222551
>  These seem to be related. 
No, these are different.  This problem was introduced in
2.21.90-7.fc23.  The actual bug is much older and was hidden and only
made obvious by a recent fix.  I'm working on it.

Siddhesh

[Siddhesh Poyarekar]

On Fri, May 22, 2015 at 06:21:43PM +0530, Siddhesh Poyarekar wrote:
>  No, these are different.  This problem was introduced in
>  2.21.90-7.fc23.  The actual bug is much older and was hidden and only 
Of course, I meant 2.21.90-8.fc23.  It worked correctly on 2.21.90-7.

Siddhesh