[PATCH 1/7] render: Inline common FindGlyph case

[Keith Packard]

Eric Anholt <eric at anholt.net> writes:

> Keith Packard <keithp at keithp.com> writes:
>
>> Provide the common path for FindGlyph as an inline function, falling
>> back to a function for the slow case.
>
> From the end of your current branch, reverting this commit gives me this
> on aa10text:

> Difference at 95.0% confidence
> 	240333 +/- 53225.9
> 	3.41191% +/- 0.755628%
> 	(Student's t, pooled s = 266420)

> So I don't think we should land this.  The prep patch of clearing
> signature would still be fine with me.

Weird. I'm getting the opposite result (oddly), which makes sense given
the amount of code I'm not running in an inner loop. Let's pend this
(it's just an optimization) until we get some better statistics running
on a native X server instead of Xephyr, and perhaps on a machine less
subject to thermal throttling.

Thanks for evaluating the effect though; makes it pretty clear that more
study is needed.

-- 
-keith
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 810 bytes
Desc: not available
URL: <http://lists.x.org/archives/xorg-devel/attachments/20150514/54734263/attachment.sig>