Follow @Openwall on Twitter for new release announcements and other news
[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Date: Tue, 17 Sep 2013 18:30:37 +0200
From: magnum <john.magnum@...hmail.com>
To: john-users@...ts.openwall.com
Subject: Re: --fork using different OpenCL devices

On 17 sep 2013, at 02:14, Rick Zero_Chaos Farina <zerochaos@...too.org> wrote:
> On 09/16/2013 07:41 PM, Solar Designer wrote:
>> On Tue, Sep 17, 2013 at 01:24:31AM +0200, magnum wrote:
>>> On 17 sep 2013, at 00:55, Solar Designer <solar@...nwall.com>
>>> wrote:
>>>> On Tue, Sep 17, 2013 at 12:03:56AM +0200, magnum wrote:
>>>>> Apparently, both Bull's 13.4 and Well's 13.6beta work fine
>>>>> with timers, signals and keypresses. Maybe we don't even have
>>>>> to use OS_TIMER=0 anymore, or is that unrelated?
>>>> 
>>>> It is related.  Yes, maybe we can set OS_TIMER = 1 now.  Please
>>>> try.
>>> 
>>> Yep, that works fine on both. So should we stick to OS_TIMER = 0
>>> anyway for now, or should we drop that clause from os.h and
>>> require a recent driver?
>> 
>> This is up to you.  As an option, can we possibly have a
>> compile-time check of Catalyst version (even though it might differ
>> from the runtime version)?

Not sure how to detect it other than fugly backticks in Makefile, and we have too much of that already (btw we should probably comment out the ones for NSS, GMP and KRB5 for releases).


> One vote for just setting a recent driver minimum and calling it good.

I dropped it now and added a comment in README-OPENCL about how/when to put that back.

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.