This is the mail archive of the glibc-bugs@sourceware.org mailing list for the glibc project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

[Bug nptl/21422] Deadlock in 2.25 pthread_cond_broadcast after process abort


https://sourceware.org/bugzilla/show_bug.cgi?id=21422

Florian Schmidt <florian.schmidt at dlr dot de> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |florian.schmidt at dlr dot de

--- Comment #12 from Florian Schmidt <florian.schmidt at dlr dot de> ---
i also stumbled upon this, and missed this report before reporting to my
distribution: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=884776

i changed our communication middleware to now use futex-syscalls on linux.

this lib is used cross-platform. i already had my own (hacky) condition
variable impl for win32 and vxworks. the QNX pshared condvar worked as expected
out-of-the box (they don't mention robust variants at all...).

i did had some strange behaviour with the old pshared condvar on linux (i think
it was related to pthread_cond_destroy()'ing them...) 

so our code base grew another special case for signaling waiters. i like the
futex based special case, i hope it will behave much better than the old impl.

-- 
You are receiving this mail because:
You are on the CC list for the bug.

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]