[Beignet] [PATCH V2 2/2] utests: add an utest for mix

Pan, Xiuli xiuli.pan at intel.com
Mon Nov 2 17:58:13 PST 2015


Thanks for you remind about the INF problem, the spec does not talk about the <x> and <y>, so I think we may consider it more.
This is surely a problem, so how about mad? It can also calculate the mix but may need some more instruction, can it handle the INF problem?

-----Original Message-----
From: Matt Turner [mailto:mattst88 at gmail.com] 
Sent: Tuesday, November 3, 2015 1:47 AM
To: Pan, Xiuli <xiuli.pan at intel.com>
Cc: beignet at lists.freedesktop.org
Subject: Re: [Beignet] [PATCH V2 2/2] utests: add an utest for mix

On Wed, Oct 21, 2015 at 8:21 PM, Pan Xiuli <xiuli.pan at intel.com> wrote:
> Add a testcase for compiler mix. Since mix will have error, we take 
> err limit as 1e-3 and print the max err.

I don't know what OpenCL's spec says about this issue, but you should be aware that Gen's LRP instruction does not return INFINITY when <x> or <y> is INFINITY.


More information about the Beignet mailing list