Date: Fri, 18 Sep 2015 20:30:16 +0200 From: magnum <john.magnum@...hmail.com> To: john-dev@...ts.openwall.com Subject: Re: multi-threaded hash table initialization On 2015-09-18 19:46, jfoug wrote: > When I was having every single format fail in the TS, running in -v -v > showed that it was crashing (I was getting ASAN errors once I built with > that enabled). All of that went away when I reverted that commit (I > don't have the number in front of me right now, I am away from the > system I worked on to find it). But even without the asan, I still am > getting 'some' failures, but these could be long standing problems that > have to do with the encoding problems we have had with the TS. We might have a brand new 32-bit and/or non-SIMD problem in raw-md5. The remaining TS issues are, I believe, neither TS problems nor have anything to do with all recent changes. I think you will find that 0ae850f1^ (version prior to all of Solar's recent changes) and d5d4edb (current HEAD) gets mostly the same TS results. There are a few problems, nothing dramatic but I think they are real bugs in JtR and not TS. Don't get overwhelmed and start running around in circles, just calmly pick a format at a time and analyze it ;-) magnum
Powered by blists - more mailing lists
Confused about mailing lists and their use? Read about mailing lists on Wikipedia and check out these guidelines on proper formatting of your messages.