Catalyst 9.8 - "compiz default maximize" = effect significant lag

[fermulator]

Does anyone else experience a HUGE lag/stutter when you maximize a window with compiz enabled? Other effects such as "wobbly windows" is rendered without problems. But the 'maxmize/unmaximize" effect is terribly laggy.

Config:
Ubuntu 9.04
01:00.0 VGA compatible controller: ATI Technologies Inc RV770 [Radeon HD 4870]
Linux fermmy 2.6.28-15-server #49-Ubuntu SMP Tue Aug 18 19:30:06 UTC 2009 i686 GNU/Linux
Latest Catalyst 9.8 fglrx driver:

fglrx-amdcccle_8.640-0ubuntu1_i386.deb
fglrx-kernel-source_8.640-0ubuntu1_i386.deb
fglrx-modaliases_8.640-0ubuntu1_i386.deb
libamdxvba1_8.640-0ubuntu1_i386.deb
xorg-driver-fglrx_8.640-0ubuntu1_i386.deb

[RealNC]

Yes, everyone has that. It's an old problem. And there's no work-around other that patching xorg with 107_fedora_dont_backfill_bg_none.patch (which will result in some temporary screen garbage while opening and maximizing windows/menus/lists/etc.

[fermulator]

Oh I see. Good to know.

So what is the problem then? A bug in X (which we need to wait to be fixed and come downstream to Ubuntu), or a bug in fglrx (which we need to wait for ATI to fix).

[nanonyme]

By RealNC's answer it sounds like the real answer is "developers don't yet know where the real problem is". You can't fix what you cannot understand.

[fermulator]

Interesting. Well, I can confirm that the problem does not exist with the open source radeon driver ... so that might narrow it down to a problem with fglrx, (maybe). heh

[RealNC]

fglrx doesn't seem to accelerate something in X that assumes the driver offers EXA support (fglrx uses the old XAA acceleration). If you take the open drivers and switch to XAA, you will have the same problem.

[fermulator]

So is there a bug report submitted to XServer, or should we expect a workaround from fglrx?

[RealNC]

I've no idea what the situation is. There was a lot of "ping-pong" the last time I looked it up (X saying it's the driver's problem, Ubuntu saying it's an X problem, ATI saying they have no idea and don't really care go patch your X since they don't really support upstream X...)

It isn't pretty

[energyman]

the problem is known. But ati devs say it's X fault, X devs say it is ati's fault.

Interessting tidbit: nvidia has/had the same problem, also by that change that is reversed with the patch. And this change was introduced to make intel less horrible. Most X devs have something to do with Intel.

You can think about that for a moment.

[BlackStar]

the problem is known. But ati devs say it's X fault, X devs say it is ati's fault.

Interessting tidbit: nvidia has/had the same problem, also by that change that is reversed with the patch. And this change was introduced to make intel less horrible. Most X devs have something to do with Intel.

You can think about that for a moment.

Interesting tidbit: nvidia drivers seem to do something similar to the "no backfill" patch, where some windows show garbage for a split second when maximized.

Fglrx without "no backfill" = no garbage but slow maximization
Fglrx with "no backfill" = garbage for a split second but fast maximization
Nvidia without "no backfill" = garbage for a split secont but fast maximization

I see behavior this on both a Quadro NVS135M (190.xx drivers) and a 9500GT (185.xx drivers). Obviously, I cannot be 100% certain, but it seems that nvidia drivers have some kind of hack for the slow maximization issue.