Follow @Openwall on Twitter for new release announcements and other news
[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Date: Thu, 17 May 2012 19:36:08 +0800
From: myrice <>
Subject: Re: Myrice: Weekly Report #5

On Thu, May 17, 2012 at 8:29 AM, Solar Designer <> wrote:
> By "violate", do you mean "volatile"?

Yes, I misspell this..

> Like Frank suggested, can you generate a simple testcase?  Not only for
> posting to AMD forums, but also to let others on our team test it on
> other GPUs.  From Samuele's test of your entire XSHA512 kernel, it is
> unclear whether the self-test failed for the same reason or maybe for
> another reason.  A smaller testcase with some debugging output would
> make this clearer.

The bug is so strange. I have two printf(SWAP64(a)) before[1] the 80
loops and at the first line[2] inside the 80 loops of SHA512 and . The
[1] gave the right result. The [2] lost high 32bit with 0. I will try
to reproduce the bug.


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.