[Bug nptl/13690] pthread_mutex_unlock potentially cause invalid access

[bugdal at aerifal dot cx]

https://sourceware.org/bugzilla/show_bug.cgi?id=13690

--- Comment #40 from Rich Felker <bugdal at aerifal dot cx> ---
Created attachment 7653
  --> https://sourceware.org/bugzilla/attachment.cgi?id=7653&action=edit
demonstration of spurious futex wakes from kernel

The attached file demonstrates spurious futex wakes coming from the kernel; it
was designed to mimic the situation I was experiencing where they were breaking
my implementation of pthread_join when it failed to retry the wait. I'm not
sure if they happen or not in situations not connected to CLONE_CHILD_CLEARTID.
I experienced the issue on 3.2 and 3.5 kernels (the latter re-checked just now)
but have not been able to reproduce it on 3.15.

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