Follow @Openwall on Twitter for new release announcements and other news
[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-ID: <CA+TsHUBhiQwKdoDxuYw-Rex-LhFA0GU=gMmoeZkhHy3Cvry-vQ@mail.gmail.com>
Date: Sun, 8 Jul 2012 10:51:54 +0530
From: Sayantan Datta <std2048@...il.com>
To: john-dev@...ts.openwall.com
Subject: Rotate and bitselect investigation

Alexander and all,

I have investigated the rotate and bitselct issue on 7970.

Both type of rotate(manual and inbuilt opencl function) use bitalign
instruction.  I investigated using rotate(x,(uint)30) and ((x << 30) | ((x
) >> 2)). Also the values loaded in the bitalign instructions are exactly
same except they  operate on different registers. So you won't see any
performance increase in this case.

However with bitselct the situation is different. The inbuilt function uses
an alien bfi instruction which I couldn't find anywhere in the docs. The
manual version uses ixor and iand.

Regards,
Sayantan

Content of type "text/html" skipped

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.