Not solved!

[Nix]

On 6 Feb 2009, Michel Dänzer told this:
> On Thu, 2009-02-05 at 18:43 +0000, Nix wrote:
>> A kludgy approach would be to do whatever gets done on VT switch (dump
>> and repopulate the entirety of EXA offscreen memory?) whenever a
>> significant fraction of allocation requests start to fail due to
>> fragmentation.
>
> Yeah, but I implemented something fancier. :) Care to try the attached
> patch?

Testing now. I suppose it'll be rather obvious if it fubars the glyph
cache :) so if it's still OK in half a day or so things are probably
fine, and if it's still fast the patch is presumably doing something as
well! I looked over the most failure-and-typo-prone area (linked list
maintenance, of course) and nothing obviously wrong leapt out at me.

> P.S. Moving to the xorg-devel list.

(not subscribed, please keep Cc:s)