[Bug 34156] r600g performance regression between 780c183b and 862ebb41

[bugzilla-daemon at freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=34156

--- Comment #3 from Dave Witbrodt <dawitbro at sbcglobal.net> 2011-02-12 12:56:02 PST ---
Created an attachment (id=43304)
 --> (https://bugs.freedesktop.org/attachment.cgi?id=43304)
Final output of 'git bisect log':  a77e813d is first bad commit -- r600g: Split
r600_bc_alu_src [Henri Verbeet]

After bisecting this regression, the last run of 'git bisect' gave this output:

a77e813de32643ae2dfffd7ad12abed596172cab is first bad commit
commit a77e813de32643ae2dfffd7ad12abed596172cab
Author: Henri Verbeet <hverbeet at gmail.com>
Date:   Mon Feb 7 15:22:07 2011 +0100

    r600g: Split r600_bc_alu_src.

    The r600_bc_alu_src structure is used in two different ways, as a vector
and
    for the individual channels of that same vector. This is somewhat fragile,
    and probably confusing.

:040000 040000 d8beff954fcb75012395bb5068c10d14088150bd
0456b8700e9170a15554f2a425374b17b891c485 M    src


Before this commit, I can get min fps of 30+ on the 'torcs' track called
"E-Road", with an upper bound of 60+ fps occasionally.  At, or after, this
commit the upper bound is about 33 fps, roughly the same as the minimum frame
rate I was getting before.

I made two naive attempts to revert a77e813d in a local branch based on the
HEAD commit I had available (a6b7393e... of Feb. 12), but neither would build
because the resulting conflict requires more massaging than simply keeping the
"before" or "after" choices offered by git.

At any rate, I'll be sticking with the commit before the one causing the
regression (3b1c1f02) until it is decided by the devs whether this is a bug or
a feature.

-- 
Configure bugmail: https://bugs.freedesktop.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.