This is the mail archive of the
gdb@sourceware.cygnus.com
mailing list for the GDB project.
race condition in solaris pthreads testsuite.
- To: gdb@sourceware.cygnus.com (GDB Developers)
- Subject: race condition in solaris pthreads testsuite.
- From: toddpw@wrs.com (Todd Whitesel)
- Date: Tue, 22 Jun 1999 21:00:55 -0700 (PDT)
Well, I tried running the same 'make check' a few times in a row on solaris
today, and discovered something interesting. There appears to be a race
condition which impacts two tests:
alabama:484% diff gdb.sum.{1st,2nd}
1c1
< Test Run By toddpw on Tue Jun 22 17:36:39 1999
---
> Test Run By toddpw on Tue Jun 22 17:51:00 1999
6249c6249
< FAIL: gdb.threads/pthreads.exp: stopped before calling common_routine 15 times
---
> PASS: gdb.threads/pthreads.exp: All threads running after continuing from ^C stop
6255c6255
< FAIL: gdb.threads/pthreads.exp: stopped in main thread at breakpoint for thread 1
---
> FAIL: gdb.threads/pthreads.exp: stopped in main thread at breakpoint for thread 2
6399,6400c6399,6400
< # of expected passes 5941
< # of unexpected failures 73
---
> # of expected passes 5942
> # of unexpected failures 72
When I ran 'make check' a third time, I got the same results as the first
(but not the second). Weeiirrdd.
--
Todd Whitesel
toddpw @ wrs.com