[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20120626021711.GC11840@openwall.com>
Date: Tue, 26 Jun 2012 06:17:11 +0400
From: Solar Designer <solar@...nwall.com>
To: john-dev@...ts.openwall.com
Subject: Re: OpenCL kernel max running time vs. "ASIC hang"
On Tue, Jun 26, 2012 at 07:34:42AM +0530, SAYANTAN DATTA wrote:
> Wouldn't calling clEnqueNDRangeKernel too many times will cause a
> performance hit?
Yes, this is definitely a tradeoff. Reliability and interactive
response time vs. speed.
> What about pausing the execution for some time as
> requested by the user. Say we press 'P' which will pasue the execution so
> that the user can perform the graphic oriented tasks and then resume the
> execution when he is finished.
Yes, we may support something like that. We may also have a john.conf
or/and command-line option to specify duty cycle (e.g., as 90%).
Currently the process may be kill -STOP'ed and then kill -CONT'ed, but
that's not pretty and it might trigger issues in the SDK.
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.
