|
From: Florian K. <br...@ac...> - 2011-07-28 16:07:19
|
On 07/28/2011 11:11 AM, Bart Van Assche wrote: > > Strange. I've never noticed something similar on any of the systems I > have access to (Linux/i386, Linux/x86_64, Linux/ppc and Darwin). On > all these systems DRD runs fine - not only the regression tests but > also real-world multithreaded applications. > I don't doubt it. The nightly system z build which happens on a z196 (latest model) shows only 3 testcase failures in DRD: drd/tests/tc04_free_lock (stderr) drd/tests/tc09_bad_unlock (stderr) drd/tests/tc23_bogus_condwait (stderr) Not sure what the nature of the failure is. Christian has done more testing with DRD on newer system z models and feels that it runs well on those. This is what we state in README.s390 So the problem I'm seeing is more likely elsewhere. Florian |