[cairo] [PATCH] add extents to clone_similar

[Carl Worth]

On Thu, 05 Oct 2006 16:22:53 -0700, Carl Worth wrote:
> [ 30]     xlib-rgb         subimage_copy-1-512     9.518  2.06%   100
> [ 31]     xlib-rgb        subimage_copy-10-512     9.555  2.47%   100
> [ 32]     xlib-rgb       subimage_copy-100-512     9.624  2.03%   100
> [ 33]     xlib-rgb      subimage_copy-1000-512    10.566  2.58%   100
> [ 34]     xlib-rgb     subimage_copy-10000-512    19.594  1.92%   100
> [ 35]     xlib-rgb    subimage_copy-100000-512   110.068  0.38%   100
>
> Wow, look at that. There is some large overhead of about 0.5
> milliseconds here. If we subtract that off from the last three cases
> then we do get times that are consistent with the previous tests:

Eric Anholt suggested that the problem might be due to the (sometimes
undesirable) pixmap migration code in KAA. In particular, the 1x1
XGetImage synchronization barrier we're now using at the beginning of
the test case might be triggering migration from video to system
memory, and the subsequent rendering might trigger migration back.

So I switched over to using an XAA server and indeed got better
results:

[ 30]     xlib-rgb         subimage_copy-1-512     0.124  1.10%   100
[ 31]     xlib-rgb        subimage_copy-10-512     0.132  0.56%   100
[ 32]     xlib-rgb       subimage_copy-100-512     0.211  1.15%   100
[ 33]     xlib-rgb      subimage_copy-1000-512     1.002  0.88%   100
[ 34]     xlib-rgb     subimage_copy-10000-512     9.024  2.74%   100
[ 35]     xlib-rgb    subimage_copy-100000-512    89.725  0.44%   100

So I don't think we have any evidence of a remaining cairo bug here.

Thanks for the help in tracking things down, everybody.

-Carl
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
Url : http://lists.freedesktop.org/archives/cairo/attachments/20061005/c8811747/attachment.pgp