Bug 96959

Summary: nop.sat generated by pow workaround?
Product: Mesa Reporter: Kenneth Graunke <kenneth>
Component: Drivers/DRI/i965Assignee: Matt Turner <mattst88>
Status: RESOLVED FIXED QA Contact: Intel 3D Bugs Mailing List <intel-3d-bugs>
Severity: normal    
Priority: medium    
Version: git   
Hardware: Other   
OS: All   
Whiteboard:
i915 platform: i915 features:

Description Kenneth Graunke 2016-07-16 18:59:12 UTC 
With Mesa 93794145dd5959d905b65234cf87de3bff801aeb (recent master),

INTEL_DEBUG=fs ./run shaders/closed/3dmmes-taiji/42.shader_test produces:

math.sat pow(16) g2<1>F         g9<8,8,1>F      0.15F           { align1 1H };
nop.sat                                                         ;

It's surprising to see a saturated no-op.  It's probably harmless, but I'm not sure why we'd be generating one of those anyway.  Could other things leak in, like predication?  Does it matter?

Matt, I figured I'd mention this in case you were interested.  If you think it's harmless, feel free to close this as NOTABUG.
Comment 1 Matt Turner 2016-12-05 20:25:09 UTC 
I've sent a patch: [PATCH] i965: Emit proper NOPs.
Comment 2 Matt Turner 2016-12-06 18:43:14 UTC 
Fixed by

commit 43cdbb3e6ab7224278a2c68586b8d1a9cb7429a9
Author: Matt Turner <mattst88@gmail.com>
Date:   Sat Dec 3 20:14:55 2016 -0800

    i965: Emit proper NOPs.

Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.