[neomagic] Big slowdown after starting gtk2 application
Richard Mittendorfer
delist at gmx.net
Mon Jun 26 07:27:29 PDT 2006
Hello *!
I have some trouble here on an Libretto 110ct/266MHz mPmmx/64M:
Everytime I start an gtk-2 application (gtkcookie, gftp, ..) displaying
lists is painfully slow. Ok, this might be a performance issue with
gtk-2. (List)Entries get created one in one second _but_ at the same
time scrolling, e.g. ls -l in an [ax]term or reading a manpage becomes
jerky and the xorgprocess is taking much more cpu as usual for such
operations. This behavior _stays_ even if I end that gtk2-app until the
Xserver is restarted. I'm running fluxbox as WM, but others show the
same effect.
<snip Xorg.0.log>
X Window System Version 7.0.0
Release Date: 21 December 2005
X Protocol Version 11, Revision 0, Release 7.0
Build Operating System:Linux 2.6.12-1-686 i686
Current Operating System: Linux libre 2.6.17-ck1 #1 PREEMPT Mon ...
[...]
(II) NEOMAGIC: Driver for Neomagic chipsets: neo2070, neo2090, neo2093,
neo2097, neo2160, neo2200, neo2230, neo2360, neo2380
(II) Primary Device is: PCI 00:04:0
(--) Chipset neo2160 found
[...]
</snip>
<snip xorg.conf>
[...]
Section "Module"
# Load "bitmap" # angeblich automatisch
Load "dbe" # conky
# Load "ddc"
# Load "dri"
Load "extmod"
Load "freetype"
# Load "glx"
# Load "GLcore"
# Load "int10"
# Load "record"
# Load "type1"
# Load "v4l"
# Load "vbe" # xfree86
# Load "xtt"
# Load "speedo"
EndSection
[...]
Section "Device"
Identifier "NeoMagic"
Driver "neomagic"
Option "overridevalidatemode"
Option "strangelockups" "false"
# Option "BackingStore" "true"
# Option "ShadowFB" "true"
# Option "OverlayMem" "0"
# Option "NoStretch" "true"
# Option "SWcursor" "true"
# Option "XaaNoScanlineImageWriteRect"
# Option "XaaNoScanlineCPUToScreenColorExpandFill"
BusID "PCI:0:4:0"
EndSection
[...]
</snip>
I've tried various settings in the xorg.conf, but none showed
improvemend. It doesn't matter if I start that gtk-2 app localy or get
it from a remote box. I have not found errors in .xsession-errors or any
logfile about this issue. IIRC I see this behavior since I switched from
Xfree86 to Xorg, Debian testing is running.
The full Xorg.0.log and xorg.conf can be found here:
http://www.mittendorfer.com/rm/temp/libre_xorg_log.txt
http://www.mittendorfer.com/rm/temp/libre_xorg_conf.txt
I'd be happy to hear some ideas. I'm after this for quite a while now.
THX, ritch
More information about the xorg
mailing list