ANNOUNCEMENT: Live Wireshark University & Allegro Packets online APAC Wireshark Training Session
July 17th, 2024 | 10:00am-11:55am SGT (UTC+8) | Online

Wireshark-bugs: [Wireshark-bugs] [Bug 10122] Buildbot crash output: fuzz-2014-05-20-5906.pcap

Date: Thu, 04 Sep 2014 02:24:36 +0000

Comment # 23 on bug 10122 from
(In reply to Evan Huus from comment #19)
> I don't mind bumping the valgrind time limit even more - it's there
> primarily to catch infinite-loops, and those aren't valgrind-specific so
> it's very unlikely that we'd actually find an infinite-loop case during a
> valgrind run that wouldn't be caught by the normal fuzz run first.
> 
> Currently, normal fuzzing limits to 5min and valgrind fuzzing to 12.5min.
> Gerald, do we know how long these captures would actually take when run
> under valgrind on the current fuzz-bot?

It took 9.5 minutes on my i7-3770S CPU @ 3.10GHz.  If the fuzz-bot is running
closer to 2GHz it should be on the order of 14-15 minutes.

I'd agree that upping the limit is reasonable: the loops will be caught by the
regular fuzz testing.  The only other alternative is to remove all the large
files from the menagerie.


You are receiving this mail because:
  • You are watching all bug changes.