Go Back   nV News Forums > Linux Support Forums > NVIDIA Linux

Newegg Daily Deals

Reply
 
Thread Tools
Old 01-08-09, 06:21 PM   #1
ponce
Registered User
 
ponce's Avatar
 
Join Date: Jan 2009
Posts: 29
Default [SOLVED] 180.xx 8600 GT SLI plasma (kde 4.2) crashes at session start

as title, no issues with 177.82. 180.22 seems to be the first 180.xx release that recognize my SLI gpus (before only one is showing in nvidia-settings).
I'm on Slackware 12.2 with kde 4.1.87 packages done using Pat/Robby slackbuilds (built against mesa).

solved changing "startkde" (in .xinitrc) with "startkde & xterm" and then relaunching plasma in a screen session (before that, I generated the attached bug report).

for the rest anything has never been so smooth here, also with effects turned on

tnx in advance for your attention
Attached Files
File Type: gz nvidia-bug-report.log.gz (46.1 KB, 115 views)

Last edited by ponce; 02-02-09 at 02:57 PM. Reason: solved
ponce is offline   Reply With Quote
Old 01-09-09, 01:38 AM   #2
ponce
Registered User
 
ponce's Avatar
 
Join Date: Jan 2009
Posts: 29
Default Re: 180.xx 8600 GT SLI plasma (kde 4.2) crashes at session start

I've managed to redirect x output in a file.

without the xterm trick the result is a black desktop with kwin active but no plasma process (the loading desktop part with icons is fine).
strange thing that relaunching manually make it works too.

it looks like it happens in few card setups, maybe it's something model specific.

if you are interested I can post also x.log and nvidia-bug-report.log with working 177.82.
Attached Files
File Type: gz x.log.gz (39.2 KB, 106 views)
ponce is offline   Reply With Quote
Old 01-09-09, 10:03 AM   #3
ponce
Registered User
 
ponce's Avatar
 
Join Date: Jan 2009
Posts: 29
Default Re: 180.xx 8600 GT SLI plasma (kde 4.2) crashes at session start

here are 177.82 logs
Attached Files
File Type: gz nvidia-bug-report.177.82.log.gz (46.1 KB, 97 views)
File Type: gz x.177.82.log.gz (48.5 KB, 93 views)
ponce is offline   Reply With Quote
Old 01-09-09, 12:06 PM   #4
ponce
Registered User
 
ponce's Avatar
 
Join Date: Jan 2009
Posts: 29
Default Re: 180.xx 8600 GT SLI plasma (kde 4.2) crashes at session start

funny thing, just want to let you know.

was trying to get a strace of plasma at session start so I:

- killed X

- renamed /usr/bin/plasma to /usr/bin/plasma-real

- created an executable script named /usr/bin/plasma with the line

strace plasma-real > /tmp/plasma.log 2>&1

- launched X

- plasma don't die

for the sake of completness I attach plasma.log with new drivers and old ones.
Attached Files
File Type: gz plasma.log.gz (11.1 KB, 106 views)
File Type: gz plasma.177.82.log.gz (11.0 KB, 92 views)

Last edited by ponce; 01-09-09 at 12:24 PM. Reason: added other log
ponce is offline   Reply With Quote
Old 01-09-09, 01:41 PM   #5
hitsov
Registered User
 
Join Date: Nov 2008
Posts: 6
Default Re: 180.xx 8600 GT SLI plasma (kde 4.2) crashes at session start

That solved my problem too.
http://www.nvnews.net/vbulletin/showthread.php?t=123552
But my 3d performance was so poor i reverted to 177.82.
Glxgears got about 4fps. Unreal tournament 2003 and 2004 ran like 1-2 fps. And smplayer got some strange message and coulden't play movies. Funny mplayer played them just fine with just a little bit of lag when toggling full screen.
How is your performance with that driver?
hitsov is offline   Reply With Quote
Old 01-09-09, 06:40 PM   #6
ponce
Registered User
 
ponce's Avatar
 
Join Date: Jan 2009
Posts: 29
Default Re: 180.xx 8600 GT SLI plasma (kde 4.2) crashes at session start

very good (consider I'm an hardcore enemy territory player): try disabling composite in your xorg.conf.
my card section looks like this

Identifier "Geforce 8600"
Driver "nvidia"
Option "NoLogo"
Option "NoBandWidthTest" "true"
Option "NvAGP" "0"
Option "FlatPanel"
Option "DPMS" "true"
Option "SLI" "auto"

but in your case you can also add
Option "RegistryDwords" "RMDisableRenderToSysmem=1"

I use also the nopat option at kernel boot ("append" string in /etc/lilo.conf).
ponce is offline   Reply With Quote
Old 01-22-09, 10:26 AM   #7
ponce
Registered User
 
ponce's Avatar
 
Join Date: Jan 2009
Posts: 29
Default Re: 180.xx 8600 GT SLI plasma (kde 4.2) crashes at session start

kde 4.2 hitted current

http://slackware.osuosl.org/slackwar...packages/kde4/

so I tried again with same results.

I tried also recompiling from slackbuild kdebase-workspace with 180.22 installed but ended with same results, plasma crash at session start.

I would have used the strace solution over but that way hotkeys don't work (even restarting daemon)

again, any hint is appreciated.

if needed this is a valgrind output
http://pastebin.com/f78d97fe5
ponce is offline   Reply With Quote
Old 01-23-09, 06:40 PM   #8
walmartshopper
Registered User
 
Join Date: Dec 2008
Posts: 23
Default Re: 180.xx 8600 GT SLI plasma (kde 4.2) crashes at session start

Quote:
- renamed /usr/bin/plasma to /usr/bin/plasma-real

- created an executable script named /usr/bin/plasma with the line

strace plasma-real > /tmp/plasma.log 2>&1

- launched X
I have the same problem, and this worked for me too, except I didn't use strace. Simply renaming plasma to plasma-real and creating the /usr/bin/plasma script to say...

#!/bin/sh
plasma-real

...was enough to make it work. Same as ponce though... hotkeys still don't work.
walmartshopper is offline   Reply With Quote

Old 01-25-09, 02:09 AM   #9
ponce
Registered User
 
ponce's Avatar
 
Join Date: Jan 2009
Posts: 29
Default Re: 180.xx 8600 GT SLI plasma (kde 4.2) crashes at session start

another interesting thing (maybe for me only cause, as you can see, I'm learning things while reporting):

tried to follow these instructions
http://techbase.kde.org/Development/..._crash_reports
to get a backtrace to gdb so I recompiled kdebase-workspace with -g flag and disabled stripping.
if I quit plasma and relaunch in gdb with 177.82 i can run it, pause it, and I got this backtrace
Code:
(gdb) thread apply all backtrace

Thread 2 (Thread 0xa8deeb90 (LWP 27299)):
#0  0xb654e6f0 in pthread_cond_wait@@GLIBC_2.3.2 () from /lib/libpthread.so.0
#1  0xb65e5ae8 in ?? () from /usr/lib/libQtCore.so.4
#2  0xb65e55e7 in QWaitCondition::wait () from /usr/lib/libQtCore.so.4
#3  0xb7a97488 in ?? () from /usr/lib/libQtNetwork.so.4
#4  0xb65e4fb3 in ?? () from /usr/lib/libQtCore.so.4
#5  0xb654a369 in start_thread () from /lib/libpthread.so.0
#6  0xb63b5cfe in clone () from /lib/libc.so.6

Thread 1 (Thread 0xb450c6f0 (LWP 27295)):
#0  0xb63ac01d in poll () from /lib/libc.so.6
#1  0xb55a8526 in ?? () from /usr/lib/libglib-2.0.so.0
#2  0x089f3f90 in ?? ()
#3  0x00000008 in ?? ()
#4  0x00001daa in ?? ()
#5  0x089f3f90 in ?? ()
#6  0x0000000a in ?? ()
#7  0xb563b5f8 in ?? () from /usr/lib/libglib-2.0.so.0
#8  0xb563b620 in ?? () from /usr/lib/libglib-2.0.so.0
#9  0xbfd9d494 in ?? ()
#10 0x00000001 in ?? ()
#11 0x00000001 in ?? ()
#12 0x0808c858 in ?? ()
#13 0x089f3f90 in ?? ()
#14 0xb63abfb0 in ?? () from /lib/libc.so.6
#15 0xb654da90 in ?? () from /lib/libpthread.so.0
#16 0xb654c200 in ?? () from /lib/libpthread.so.0
#17 0xb55cc7e2 in g_thread_self () from /usr/lib/libglib-2.0.so.0
#18 0xb55a8aae in g_main_context_iteration () from /usr/lib/libglib-2.0.so.0
#19 0xb6721266 in QEventDispatcherGlib::processEvents () from /usr/lib/libQtCore.so.4
#20 0xb6a5d00c in ?? () from /usr/lib/libQtGui.so.4
#21 0xb66eac76 in QEventLoop::processEvents () from /usr/lib/libQtCore.so.4
#22 0xb66eaeb5 in QEventLoop::exec () from /usr/lib/libQtCore.so.4
#23 0xb66eebea in QCoreApplication::exec () from /usr/lib/libQtCore.so.4
#24 0xb69a4a18 in QApplication::exec () from /usr/lib/libQtGui.so.4
#25 0xb804254b in kdemain (argc=2, argv=0xbfd9d784) at /tmp/kdebase-workspace-4.2.0/workspace/plasma/shells/desktop/main.cpp:54
#26 0x080486b2 in main (argc=Cannot access memory at address 0x8
) at /tmp/kdebase-workspace-4.2.0/build/workspace/plasma/shells/desktop/plasma_qgv_dummy.cpp:3
if I do the same thing with 180.22 program exits with code 01 and this is the output
Code:
(gdb) run --nofork
Starting program: /usr/bin/plasma --nofork
[Thread debugging using libthread_db enabled]
[New Thread 0xb43d36f0 (LWP 28601)]
dlsym vsnprintf error:(null)

Program exited with code 01.
(gdb) thread apply all backtrace
No registers.
I think I'm missing something?
ponce is offline   Reply With Quote
Old 01-25-09, 03:30 AM   #10
progger1986
Registered User
 
Join Date: Aug 2008
Posts: 19
Default Re: 180.xx 8600 GT SLI plasma (kde 4.2) crashes at session start

I had the same problem with the 180.xx drivers.
The problem maybe the wallpaper-position:
With 180.xx plasma crashes the first startup when the wallpaper-position is "scaled and cutten" (Sorry: I don't know the correct english term).
When using "scaled" as wallpaper-position it should work.
progger1986 is offline   Reply With Quote
Old 01-25-09, 03:35 AM   #11
ponce
Registered User
 
ponce's Avatar
 
Join Date: Jan 2009
Posts: 29
Default Re: 180.xx 8600 GT SLI plasma (kde 4.2) crashes at session start

i think is a different problem: mine has always been scaled only.

also I noted this messages in syslog, don't know if related (probably not)
Code:
Jan 25 09:01:42 toscibo kernel: IRQ 16/nvidia: IRQF_DISABLED is not guaranteed on shared IRQs
Jan 25 09:01:44 toscibo kernel: IRQ 17/nvidia: IRQF_DISABLED is not guaranteed on shared IRQs
ponce is offline   Reply With Quote
Old 01-26-09, 04:53 AM   #12
ponce
Registered User
 
ponce's Avatar
 
Join Date: Jan 2009
Posts: 29
Default Re: 180.xx 8600 GT SLI plasma (kde 4.2) crashes at session start

Quote:
Originally Posted by ponce View Post
another interesting thing (maybe for me only cause, as you can see, I'm learning things while reporting):

tried to follow these instructions
http://techbase.kde.org/Development/..._crash_reports
to get a backtrace to gdb so I recompiled kdebase-workspace with -g flag and disabled stripping.
if I quit plasma and relaunch in gdb with 177.82 i can run it, pause it, and I got this backtrace
Code:
(gdb) thread apply all backtrace

Thread 2 (Thread 0xa8deeb90 (LWP 27299)):
#0  0xb654e6f0 in pthread_cond_wait@@GLIBC_2.3.2 () from /lib/libpthread.so.0
#1  0xb65e5ae8 in ?? () from /usr/lib/libQtCore.so.4
#2  0xb65e55e7 in QWaitCondition::wait () from /usr/lib/libQtCore.so.4
#3  0xb7a97488 in ?? () from /usr/lib/libQtNetwork.so.4
#4  0xb65e4fb3 in ?? () from /usr/lib/libQtCore.so.4
#5  0xb654a369 in start_thread () from /lib/libpthread.so.0
#6  0xb63b5cfe in clone () from /lib/libc.so.6

Thread 1 (Thread 0xb450c6f0 (LWP 27295)):
#0  0xb63ac01d in poll () from /lib/libc.so.6
#1  0xb55a8526 in ?? () from /usr/lib/libglib-2.0.so.0
#2  0x089f3f90 in ?? ()
#3  0x00000008 in ?? ()
#4  0x00001daa in ?? ()
#5  0x089f3f90 in ?? ()
#6  0x0000000a in ?? ()
#7  0xb563b5f8 in ?? () from /usr/lib/libglib-2.0.so.0
#8  0xb563b620 in ?? () from /usr/lib/libglib-2.0.so.0
#9  0xbfd9d494 in ?? ()
#10 0x00000001 in ?? ()
#11 0x00000001 in ?? ()
#12 0x0808c858 in ?? ()
#13 0x089f3f90 in ?? ()
#14 0xb63abfb0 in ?? () from /lib/libc.so.6
#15 0xb654da90 in ?? () from /lib/libpthread.so.0
#16 0xb654c200 in ?? () from /lib/libpthread.so.0
#17 0xb55cc7e2 in g_thread_self () from /usr/lib/libglib-2.0.so.0
#18 0xb55a8aae in g_main_context_iteration () from /usr/lib/libglib-2.0.so.0
#19 0xb6721266 in QEventDispatcherGlib::processEvents () from /usr/lib/libQtCore.so.4
#20 0xb6a5d00c in ?? () from /usr/lib/libQtGui.so.4
#21 0xb66eac76 in QEventLoop::processEvents () from /usr/lib/libQtCore.so.4
#22 0xb66eaeb5 in QEventLoop::exec () from /usr/lib/libQtCore.so.4
#23 0xb66eebea in QCoreApplication::exec () from /usr/lib/libQtCore.so.4
#24 0xb69a4a18 in QApplication::exec () from /usr/lib/libQtGui.so.4
#25 0xb804254b in kdemain (argc=2, argv=0xbfd9d784) at /tmp/kdebase-workspace-4.2.0/workspace/plasma/shells/desktop/main.cpp:54
#26 0x080486b2 in main (argc=Cannot access memory at address 0x8
) at /tmp/kdebase-workspace-4.2.0/build/workspace/plasma/shells/desktop/plasma_qgv_dummy.cpp:3
if I do the same thing with 180.22 program exits with code 01 and this is the output
Code:
(gdb) run --nofork
Starting program: /usr/bin/plasma --nofork
[Thread debugging using libthread_db enabled]
[New Thread 0xb43d36f0 (LWP 28601)]
dlsym vsnprintf error:(null)

Program exited with code 01.
(gdb) thread apply all backtrace
No registers.
I think I'm missing something?
sorry if I quote myself but I asked for this also on kde forums but it looks like is not kde related.
ponce is offline   Reply With Quote
Reply


Thread Tools

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Forum Jump


All times are GMT -5. The time now is 03:40 AM.


Powered by vBulletin® Version 3.7.1
Copyright ©2000 - 2014, Jelsoft Enterprises Ltd.
Copyright 1998 - 2014, nV News.