Does the intel-driver not accalerate bitmask blits?

[Eric Anholt]

On Tue, 2007-06-12 at 20:10 +0200, Clemens Eisserer wrote:
> Hi,
> 
> I tuned a java-benchmark to make better use of hardware-accaleration
> (uses now bitmasks instead of argb images), and saw some improvement,
> but not in the range I expected.
> 
> I used oprofile to see whats going on and got that:
> 85483    51.5340  libfb.so                 libfb.so                 (no symbols)
> 36338    21.9066  java                     (no symbols)
> 10595     6.3873  no-vmlinux               no-vmlinux               (no symbols)
> 10064     6.0671  Xorg                     Xorg                     (no symbols)
> 
> This profile shows that a lot of time is spent inside of libfb,
> however I really wonder why so much time is spent there.
> Does the intel driver simply not accalerate bitmasked blits?

Correct.  If you're attempting to composite through a mask, use an a8
mask.

-- 
Eric Anholt                             anholt at FreeBSD.org
eric at anholt.net                         eric.anholt at intel.com

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 187 bytes
Desc: This is a digitally signed message part
URL: <http://lists.x.org/archives/xorg/attachments/20070612/3b4f052f/attachment.pgp>