Follow @Openwall on Twitter for new release announcements and other news
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20150813190423.GA21231@openwall.com>
Date: Thu, 13 Aug 2015 22:04:23 +0300
From: Solar Designer <solar@...nwall.com>
To: john-dev@...ts.openwall.com
Subject: Re: --test-full=0 crashes the Bitcoin format

Hi Kai,

On Fri, Aug 07, 2015 at 08:14:11AM +0800, Kai Zhao wrote:
> I will debug this crash.

Can you explain what you're trying to achieve by running:

kai      11232 1700  0.0 2413320 7436 ?        Sl   Aug12 25756:30 ./john --test-full=0 --format=bitcoin

for a day now?  Are you hoping to trigger this crash?  If so, no, I
don't expect you will: I only saw it happen (twice) when running
"--test-full=0" without "--format=bitcoin".  It is quite clear that
having other formats test-full'ed prior to Bitcoin is a prerequisite to
triggering this crash.

Also, how did you achieve that this process is running for a day and is
not terminating?  Normally, even a full test of the Bitcoin format won't
be taking this long, would it?

And, do you have an idea why this process tends to consume exactly 1700%
of CPU (17 active threads)?  I checked /proc/11232/environ and to my
surprise I did not find OMP_NUM_THREADS=17 in there.  Can you explain?

I think that the only thing you're achieving so far is preventing others
from using super for benchmarking.  So please stop.

Thanks,

Alexander

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.