Created attachment 775949 [details] Switch to using PLArenaPool on blapitest.c This patch gets rid of sporadic test core dumps that have been has been observed on AES GCM tests in 64-bit platforms.
This patch is a subset of the last patch from the upstream bug https://bugzilla.mozilla.org/show_bug.cgi?id=802430 https://bug802430.bugzilla.mozilla.org/attachment.cgi?id=744975 See various comments made on Bug 976572.
This patch hasn't solved the problem, nor has a more complete one I tried. Worth noting is the fact that we get the core dumps even though all the tests pass.
Elio, may I ask for a status update? Are you going to update the patch? I am confused by your comment #c10 while switching the bug to the MODIFIED state.
(In reply to Karel Srot from comment #12) > Elio, > may I ask for a status update? Are you going to update the patch? I am > confused by your comment #c10 while switching the bug to the MODIFIED state. I have expanded the patch and it still fails some times. Now that Bob is back from PTO we will be working on this problem together. Yes, it can be confusing and I should have been clearer. I believe that the problem of the sporadic core dumps is orthogonal to the memory leaks we are trying to solve here. I marked it as modified because it was modified. The memory leak patch stands in its own merits and I did not want to get in the way in case you wanted to do some verification of the patch. I understand if you prefer to wait until we have come to a solution of the other problem and have done the respin.
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory, and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. http://rhn.redhat.com/errata/RHBA-2013-1318.html