[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <8512486dc2385043f35c778de7e4bb7b@smtp.hushmail.com>
Date: Sun, 04 Mar 2012 02:26:04 +0100
From: magnum <john.magnum@...hmail.com>
To: john-dev@...ts.openwall.com
Subject: Re: CUDA & OpenCL status
On 03/04/2012 01:44 AM, magnum wrote:
> On 03/04/2012 01:26 AM, Solar Designer wrote:
>> Benchmarking: FreeBSD MD5 [SSE2i 8x]... Segmentation fault (core dumped)
>>
>> Oops. Somehow in the CUDA build, some CPU-only formats fail. This is
>> seen in a -gpu build as well (CUDA and NVidia OpenCL at once), where
>> some CPU-only formats' self-tests fail (IIRC, I was getting not
>> segfaults, but test failures in that build). Has anyone else seen that
>> as well?
>
> I had not, but I do see the exact same problem now. This may be a
> magnum-jumbo specific problem (ie. my fault). I will investigate.
It sure was my fault and it had little to do with GPU, it was the fact I
implicitly use pre-built intrinsics for GPU builds but forgot an
important #define...
Fixed in git trees now. Thanks for reporting. Using gcc 4.6, I never hit 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.
