Openwall GNU/*/Linux - a small security-enhanced Linux distro for servers
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Mon, 21 Oct 2013 21:33:47 +0200
From: magnum <john.magnum@...hmail.com>
To: "john-dev@...ts.openwall.com" <john-dev@...ts.openwall.com>
Subject: Re: OpenCL vectorizing how-to.

On 2013-10-19 00:42, magnum wrote:
> I'm looking forward to see how eg. wpapsk-opencl performs now on the
> Xeon Phi. Unless its driver rather auto-vectorizes of course :-/

While doing a more or less unrelated search on coalescing I stumbled 
upon this:
http://software.intel.com/en-us/articles/opencl-design-and-programming-guide-for-the-intel-xeon-phi-coprocessor

Xeon Phi's compiler indeed wants scalar code. BTW I thought only some 
models of Xeon Phi were 512-bit width but this paper seem to indicate 
they all are (but may be wider in future).

One thing I can't understand is why pre-vectorized code with the correct 
width is not used "as-is" by these compilers. Apparently the compiler 
first scalarizes it and then re-vectorizes it - with very poor results, 
at least on Well. OTOH this isn't a problem now that we can supply the 
requested [lack of] width.

magnum


Powered by blists - more mailing lists

Your e-mail address:

Powered by Openwall GNU/*/Linux - Powered by OpenVZ