[Intel-gfx] An apparent large performance regression of 3D display over the network

[Alan W. Irwin]

On 2011-02-14 15:34+0100 Sven Arvidsson wrote:

> Hi,
>
> Not sure if you have seen this patch (I just stumbled upon it myself)
> but it sounds like it makes quite a difference with indirect rendering:
> http://lists.x.org/archives/xorg-devel/2011-January/018623.html

Yes, I noticed that patch from Chris Wilson as well and look forward
to when it is propagated to Debian.

Chris, does that patch make the etracer initial menu usable with

export LIBGL_ALWAYS_INDIRECT=1

? That menu is almost unusable (much flickering, extremely slow
response to cursor motions) here for LIBGL_ALWAYS_INDIRECT=1.  That's
for the case of running etracer directly on the machine where the X
server is located.  For those same conditions with
LIBGL_ALWAYS_INDIRECT not set, the etracer initial menu is fine.

As an X-terminal user for the last decade, I am glad to see that there
is at least some attention being paid to efficiency issues with the
LIBGL_ALWAYS_INDIRECT=1 case.

Alan
__________________________
Alan W. Irwin

Astronomical research affiliation with Department of Physics and Astronomy,
University of Victoria (astrowww.phys.uvic.ca).

Programming affiliations with the FreeEOS equation-of-state implementation
for stellar interiors (freeeos.sf.net); PLplot scientific plotting software
package (plplot.org); the libLASi project (unifont.org/lasi); the Loads of
Linux Links project (loll.sf.net); and the Linux Brochure Project
(lbproject.sf.net).
__________________________

Linux-powered Science
__________________________