<html>
<head>
<meta content="text/html; charset=UTF-8" http-equiv="Content-Type">
</head>
<body bgcolor="#FFFFFF" text="#000000">
I figured your hands are tied.<br>
<br>
So far, it seems GTK+ themes turn the Xorg server into a wait
station. I switched out the GTK+ Theme for Chrome to it's default
one and the system stop laging for Chrome.<br>
<br>
However, Iceweasel, Thunderbird [while typing this] is like walking
through a time field where the echo to the screen is a solid 1- 2
second delay that scales up to 2 seconds with several keys types in
rapid succession.<br>
<br>
Unfortunately, I think XUL/MozJS and the Theme engine don't like the
latest Xorg/Nvidia combo.<br>
<br>
Nvidia thinks it has to do with GPU acceleration verses software
acceleration with trapezoids but after discovering the GTK+ Theme
lag it's clear that how GTK+ is handling primitive window drawing
with GDK and rasterizing fonts to the screen it's clear it's just
not a rendering via the Driver side, but how GTK+/GDK, XOrg and
Nvidia all work together.<br>
<br>
I haven't been the most cordial with Nvidia but it appears as I'm
typing this that there are quite a few memory leaks in GTK+ Themes,
caching issues and much more.<br>
<br>
- Marc<br>
<br>
On 10/04/2011 02:42 PM, Russ Allbery wrote:
<blockquote cite="mid:87aa9gcu1y.fsf@windlord.stanford.edu"
type="cite">
<pre wrap="">"Marc J. Driftmeyer" <a class="moz-txt-link-rfc2396E" href="mailto:mjd@reanimality.com"><mjd@reanimality.com></a> writes:
</pre>
<blockquote type="cite">
<pre wrap="">I don't know if it's a timeout issue or what [I'm more well-versed in
WindowServer of OS X and Openstep] but this latest fix from Nvidia is
nothing but a bag of hurt.
</pre>
</blockquote>
<pre wrap="">
</pre>
<blockquote type="cite">
<pre wrap="">Several comments are building already on nvnews.net's Linux Forum.
</pre>
</blockquote>
<pre wrap="">
That would indeed be the appropriate place to discuss it. There's
literally nothing that we as NVIDIA package maintainers can do, since this
is a closed-source driver handed to us as a compiled binary from NVIDIA.
They're the only ones who can fix it; there really isn't anything we can
do from a packaging perspective.
That's not to say that your bug filing is a bad idea. It's good to have
it recorded in the BTS so that other people know it's still a problem when
they see the same thing. But unfortunately we're pretty much in a waiting
game for NVIDIA to figure out what the problem is and release updated
drivers.
</pre>
</blockquote>
<br>
<div class="moz-signature">-- <br>
Marc J. Driftmeyer<br>
Email :: <a href="mailto:mjd@reanimality.com">mjd@reanimality.com</a><br>
Web :: <a href="http://www.reanimality.com">http://www.reanimality.com</a><br>
Cell :: (509) 435-5212
</div>
</body>
</html>