[Bug 85505] i965/fs: Better heuristics on when to drop from SIMD16 to SIMD8, on register spilling

[bugzilla-daemon at freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=85505

Matt Turner <mattst88 at gmail.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Assignee|idr at freedesktop.org         |mattst88 at gmail.com
            Summary|Better heuristics on when   |i965/fs: Better heuristics
                   |to drop from SIMD16 to      |on when to drop from SIMD16
                   |SIMD8, on register spilling |to SIMD8, on register
                   |                            |spilling

--- Comment #1 from Matt Turner <mattst88 at gmail.com> ---
I'd be surprised to see a shader that spilled the same number of registers in
SIMD8 and SIMD16. But yes, there are improvements we should make.

I've noticed that we often schedule the SIMD8 program to minimize delays
between dependent instructions at the code of higher register usage, but since
the SIMD16 program uses more registers we're not able to perform the same
schedule. So instead we schedule the SIMD16 program to minimize register usage,
which drastically increases latency. I suspect in cases like this the SIMD16
program is slower.

I've been planning to look at this, so I'll assign this bug to myself.

-- 
You are receiving this mail because:
You are the QA Contact for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.freedesktop.org/archives/intel-3d-bugs/attachments/20141027/1ab8af9d/attachment.html>