<div dir="ltr"><div class="gmail_default" style="font-family:verdana,sans-serif;font-size:small"><br></div><div class="gmail_extra"><br><div class="gmail_quote">On Wed, Feb 7, 2018 at 2:46 PM, Kenneth Graunke <span dir="ltr"><<a href="mailto:kenneth@whitecape.org" target="_blank">kenneth@whitecape.org</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">On Wednesday, February 7, 2018 1:21:53 PM PST Anuj Phogat wrote:<br>
> On Wed, Feb 7, 2018 at 12:24 PM, Francisco Jerez wrote:<br>
> > Anuj Phogat writes:<br>
<span class="">> > > On Wed, Feb 7, 2018 at 11:55 AM, Francisco Jerez wrote:<br>
> > > > You can also just drop the patch, unless you want to change the<br>
> > > > front-end in addition to stop allocating a destination for memory fences<br>
> > > > on HSW-SKL in order to save a small amount of register pressure?<br>
> > > ><br>
> > > It also fixes a fulsim error other than reducing register pressure.<br>
> ><br>
> > On what platform?<br>
><br>
> I noticed this on Icelake. But, I'm sure it exists on CNL as well.<br>
<br>
</span>But in patch 1, you always set commit enable on Gen10+. So you wouldn't<br>
hit the "no commit" case that wants a null destination.<br>
<br>
I'm guessing you saw that before you wrote patch 1 (and reordered them)?<br></blockquote><div><div class="gmail_default" style="font-family:verdana,sans-serif;font-size:small;display:inline">I think that's what happened. I'll test it again.</div> </div></div><br></div></div>