weird behavior with and without nvidia driversand a memory hungry X
Hi all, I run Suse 7.2 prof with kde2.2.2 and qt2.3.1 on a MSI K7T Turbo RAID motherboard with an NVIDIA 2 MX 200 32 MB AGP Graphics Card. RAM is 512 MB. X as installed with 7.2 prof. not modified. Originally I used the nv driver, but, to gain some speed, especially because changing the desktops was quite slow, I installed the new nvidia drivers NVIDIA_kernel-1.0-2313.suse72.i386.rpm and NVIDIA_GLX-1.0-2313.suse72.i386.rpm, using sax2. Before the update I was seeing a memory leak killing X slowly. When I start working top show X using Memory just under 40%. While working (kmail, quanta, opera, konqueror) this increased steadily to over 100 % (after a couple of hours) and then the swapping takes forever, I have to log out and at least restart X. I blamed it on Quanta, since that was the prog always used, while the others are not used at all the time, however I did not test working a few hours without quanta. (Question: How could I check what is causing X to use more and more memory?) Then, after switching from nv to the nvidia driver, first no change. desktop updates had been as slow as before. X did eat the memory away as before. But now, when I log out from kde, the screen freezes with a nice black-white stripe pattern. Nothing works, no reaction on any keyboard input. I have to hard-reboot. After this happened a couple of times, and I had been ready to reinstall the nv drivers, suddenly kde comes up with weird fonts (I have anti-aliasing and a lot of tt fonts installed). I cannot see most of the fonts. I blame the hard reboot and try to do a clean reboot. Next time the weird font setup is still there, but all fonts are back, and I can reconfigure the fonts again. -> But now the desktop changes are really FAST in one quick step a desktop changes and not slow in two steps, as before. -> And now, which miracle, X stopped eating all memory, I can work for hours with the same progs (quanta etc.) and X stays always around 40%! My guess is that the nvidia drivers had been installed, but for a special reason not used properly at the beginning. The crash did delete something, which then allowed the nvidia drivers to really work. But what? It would be fine now, but the crashes after logging out from kde are NOT gone, and they surely will cause some damage soon. So I'm stuck between the crashes at every log out and going back to nv, and possibly re-introducing the memory hunger of X, which makes the system quickly slow and also forces reboot regularly. Well, you understand, I would appreciate any ideas and hints to address this problem, if possible staying with the nvidia drivers and stopping the crashes, and, if that is not possible, to go back to nv but without getting a overly memory hungry X again.
On Sat, 5 Jan 2002, Matt wrote: m> Hi all, m> m> I run Suse 7.2 prof with kde2.2.2 and qt2.3.1 on a MSI K7T Turbo RAID m> motherboard with an NVIDIA 2 MX 200 32 MB AGP Graphics Card. RAM is 512 MB. m> X as installed with 7.2 prof. not modified. m> I'm also running SuSE 7.2 professional, with KDE-2.2.1, QT-2.3.1 on an iWill motherboard and a Viper 16 MB graphics card. Ram is at 768MB and nothing has changed with X accept what YOU and yast update would do. m> Originally I used the nv driver, but, to gain some speed, especially because m> changing the desktops was quite slow, I installed the new nvidia drivers m> NVIDIA_kernel-1.0-2313.suse72.i386.rpm and m> NVIDIA_GLX-1.0-2313.suse72.i386.rpm, using sax2. The drivers I'm using are the same, installed via YOU m> m> Then, after switching from nv to the nvidia driver, first no change. desktop m> updates had been as slow as before. X did eat the memory away as before. But m> now, when I log out from kde, the screen freezes with a nice black-white m> stripe pattern. Nothing works, no reaction on any keyboard input. I have to m> hard-reboot. m> I have noticed this ame thing, accept mine is colored as green and blue stripes. Exiting X via CTRL+Alt+1-6 does nothing. Only options I have are to do a hard reset or to logon through one o the other computers with SSH and do a 'shutdown -r' there. I also tried doing an 'init 3' but nothing happens, it shows X isn't running when I check with top or ps -ax but it appears and acts locks up from the actual workstation. m> m> My guess is that the nvidia drivers had been installed, but for a special m> reason not used properly at the beginning. The crash did delete something, m> which then allowed the nvidia drivers to really work. But what? m> I suspect the drivers are causing some problem, but I also suspect the the KDE logout has a problem. I have been debating about going back to the SuSE brand of the drivers until I tried doing a quick logout from KDE by pressing Ctrl+Alt+Shift+Del which seems to exit safetly back to the KDM logon screen. The reason I suspect KDE as being part of the problem is that when hitting the logout button or pressing Ctrl+Alt+Del it asks to save the current configuration which I'm skipping with the above key sequence. m> It would be fine now, but the crashes after logging out from kde are NOT m> gone, and they surely will cause some damage soon. m> m> So I'm stuck between the crashes at every log out and going back to nv, and m> possibly re-introducing the memory hunger of X, which makes the system m> quickly slow and also forces reboot regularly. m> m> Well, you understand, I would appreciate any ideas and hints to address this m> problem, if possible staying with the nvidia drivers and stopping the m> crashes, and, if that is not possible, to go back to nv but without getting a m> overly memory hungry X again. m> m> -- S.Toms - smotrs@mindspring.com - www.mindspring.com/~smotrs SuSE Linux v7.2+ - Kernel 2.4.4 Sorry, no fortune this time.
On Sunday 06 January 2002 02:37, S.Toms wrote: [...]
I suspect the drivers are causing some problem, but I also suspect the the KDE logout has a problem. I have been debating about going back to the SuSE brand of the drivers until I tried doing a quick logout from KDE by pressing Ctrl+Alt+Shift+Del which seems to exit safetly back to the KDM logon screen.
Well, for me Ctrl+Alt+Shift+Del gives the same crash. However Ctrl+Alt+Backspace brings me back to the KDM logon screen. Great, at least a workaround for now!
The reason I suspect KDE as being part of the problem is that when hitting the logout button or pressing Ctrl+Alt+Del it asks to save the current configuration which I'm skipping with the above key sequence.
Did you try what happens if you logout under Gnome? For me it is exactly the same problem - the same crash just before the KDM logon screen should appear. So at least in my installation it is not a kde specific logout problem, however it could of course be a kdm logon problem. I did not switch off the nvidia logo, so I see it coming up nicely just before the crash. Does anyone know what happens differently at a logout compared to the first login, or the one forced with Ctrl+Alt+Backspace ? Thanks, Matt On Sunday 06 January 2002 02:37, S.Toms wrote:
On Sat, 5 Jan 2002, Matt wrote:
m> Hi all, m> m> I run Suse 7.2 prof with kde2.2.2 and qt2.3.1 on a MSI K7T Turbo RAID m> motherboard with an NVIDIA 2 MX 200 32 MB AGP Graphics Card. RAM is 512 MB. m> X as installed with 7.2 prof. not modified. m> I'm also running SuSE 7.2 professional, with KDE-2.2.1, QT-2.3.1 on an iWill motherboard and a Viper 16 MB graphics card. Ram is at 768MB and nothing has changed with X accept what YOU and yast update would do.
m> Originally I used the nv driver, but, to gain some speed, especially because m> changing the desktops was quite slow, I installed the new nvidia drivers m> NVIDIA_kernel-1.0-2313.suse72.i386.rpm and m> NVIDIA_GLX-1.0-2313.suse72.i386.rpm, using sax2.
The drivers I'm using are the same, installed via YOU
m> m> Then, after switching from nv to the nvidia driver, first no change. desktop m> updates had been as slow as before. X did eat the memory away as before. But m> now, when I log out from kde, the screen freezes with a nice black-white m> stripe pattern. Nothing works, no reaction on any keyboard input. I have to m> hard-reboot. m>
I have noticed this ame thing, accept mine is colored as green and blue stripes. Exiting X via CTRL+Alt+1-6 does nothing. Only options I have are to do a hard reset or to logon through one o the other computers with SSH and do a 'shutdown -r' there. I also tried doing an 'init 3' but nothing happens, it shows X isn't running when I check with top or ps -ax but it appears and acts locks up from the actual workstation.
m> m> My guess is that the nvidia drivers had been installed, but for a special m> reason not used properly at the beginning. The crash did delete something, m> which then allowed the nvidia drivers to really work. But what? m>
I suspect the drivers are causing some problem, but I also suspect the the KDE logout has a problem. I have been debating about going back to the SuSE brand of the drivers until I tried doing a quick logout from KDE by pressing Ctrl+Alt+Shift+Del which seems to exit safetly back to the KDM logon screen. The reason I suspect KDE as being part of the problem is that when hitting the logout button or pressing Ctrl+Alt+Del it asks to save the current configuration which I'm skipping with the above key sequence.
m> It would be fine now, but the crashes after logging out from kde are NOT m> gone, and they surely will cause some damage soon. m> m> So I'm stuck between the crashes at every log out and going back to nv, and m> possibly re-introducing the memory hunger of X, which makes the system m> quickly slow and also forces reboot regularly. m> m> Well, you understand, I would appreciate any ideas and hints to address this m> problem, if possible staying with the nvidia drivers and stopping the m> crashes, and, if that is not possible, to go back to nv but without getting a m> overly memory hungry X again. m> m>
On Sun, 6 Jan 2002, Matt T. wrote: mt> mt> On Sunday 06 January 2002 02:37, S.Toms wrote: mt> [...] mt> > I suspect the drivers are causing some problem, but I also suspect mt> > the the KDE logout has a problem. I have been debating about going back mt> > to the SuSE brand of the drivers until I tried doing a quick logout from mt> > KDE by pressing Ctrl+Alt+Shift+Del which seems to exit safetly back to the mt> > KDM logon screen. mt> mt> Well, for me Ctrl+Alt+Shift+Del gives the same crash. However mt> Ctrl+Alt+Backspace brings me back to the KDM logon screen. mt> Great, at least a workaround for now! mt> I was hoping it would have worked for you, that was theonly workaround I could come up with. Using the Ctrl+Alt+Backspace I mine actually causes it to freeze whereas the Ctrl+Alt+Shift+Del will safetly log me out. The Ctrl+Alt+Backspace doesn't really logout though, I explain it more below. mt> > The reason I suspect KDE as being part of the problem is that when mt> > hitting the logout button or pressing Ctrl+Alt+Del it asks to save the mt> > current configuration which I'm skipping with the above key sequence. mt> mt> Did you try what happens if you logout under Gnome? For me it is exactly the mt> same problem - the same crash just before the KDM logon screen should appear. mt> So at least in my installation it is not a kde specific logout problem, mt> however it could of course be a kdm logon problem. mt> No, I'm not using gnome on my systems so I havn't even tried it. mt> I did not switch off the nvidia logo, so I see it coming up nicely just mt> before the crash. mt> I don't even get the logo before it crashes. The only time I see it is during the next restart after that crash. mt> Does anyone know what happens differently at a logout compared to the first mt> login, or the one forced with Ctrl+Alt+Backspace ? mt> The Ctrl+Alt+Backspace isn't really a logout, it actually forcibly kills the X session. The only two methods of logging out supported by KDE are the Ctrl+Alt+Del and the Ctrl+Alt+Shift+Del. The former will ask for user confirmation whereas the latter will just exit quickly. mt> Thanks, mt> Matt mt> mt> -- S.Toms - smotrs@mindspring.com - www.mindspring.com/~smotrs SuSE Linux v7.2+ - Kernel 2.4.4 Going to church does not make a person religious, nor does going to school make a person educated, any more than going to a garage makes a person a car.
Hi all! Hidden th the thread "Slightly baffled with the Euro symbol" Bill Margolis did send with the subject "Geforce3 Ti and the Black Screen of Death" a solution to the problem, which worked also for me with my GeForce2 MX DDR. No more frozen patterns when logging out from kde/gnome. The guilty ones are AGPART and Nvidia, who do not like to work together, at least on Bill's and on my system. Nvidia prefers NvAGP, which has to be enabled as an option. Before you do so, you should however check with nvidia if your AGP chipset is supported (my Via KT133 is). You find this info in "NVidia-LinuxNotes2313.pdf" on their website. How to enable NvAGP (and disable AGPGART): =========================================== You can do this in or with SAX 2. - With Sax 2: Under Graphics-Adapter->Properties->Advanced->Raw Definition enter Option "NvAGP" "1" in the field Raw Definition. Then test and save and reboot. - in etc/X11/XF86Config add the line Option "NvAGP" "1" in the Section "Device" Mine looks now like this (created by Sax2): Section "Device" BoardName "GeForce2 MX DDR" BusID "1:0:0" Driver "nvidia" Identifier "Device[0]" Option "NvAGP" "1" Screen 0 VendorName "Nvidia" EndSection After reboot NvAGP should get loaded instead of AGPART. Thanks to Bill Margolis for showing the way to fix this! At the end of this message you can see the logout-from-kde sequence in /var/log/messages with not-crashing NvAGP and with freezing AGPART. HTH. After examining the /var/log/XFree86.0.log file there is just one line left which irritates me, maybe someone can explain/correct this: (WW) NVIDIA(0): Failed to allocate a DMA push buffer using AGP memory... (WW) NVIDIA(0): attempting to use PCI memory I guess the xconsole messages, which I get now, have to do with this: Jan 9 16:14:19 penguin kernel: mtrr: no MTRR for d8000000,1000 found Jan 9 16:14:19 penguin kernel: unset mem range for 0xd8000000 0x2000000 Jan 9 16:14:20 penguin kdm[632]: Server for display :0 terminated unexpectedly Jan 9 16:14:22 penguin kdm[1081]: session start failed Jan 9 16:14:22 penguin kernel: mtrr: no MTRR for d8000000,1000 found Jan 9 16:14:22 penguin kernel: unset mem range for 0xd8000000 0x2000000 Looks like something taking away AGP-Performance, what a pity. However at least it works fine now. Have Fun, Matt On Thursday 10 January 2002 02:09, Bill Margolis wrote:
Howdy All, As a Suse newbie, allow me to record some partial successes:
Step 1: Acquired Suse 7.3 Pro, plus Athlon XP DDR with MSI GeForce3 Ti200 card, (plus Audigy sound card, but that is still an open issue...).
Do Quick Install : <really impressive with Reiserfs! ... nice graphics ... default installation ... until...> ends abruptly (presumably at first graphics board probe) with those lovely, retro, X crosshairs on a Black Screen of Death! (Cold start required.)
Step 2: Examined and saved xf86 graphics logfile.
Step 3: Do another Quick Install this time in Text Mode. <Need to remember to custom the partitions to retain the Reiserfs> . Hand link the xf86.4 X server and turn the xf86 graphics logfile into a XF86Config.
Step 4: Do startx and find myself in graphics mode. Very nice. But, exit keystrokes result in ... Black Screen of Death (oops, but as expected).
Step 5: So, modify driver section of XF86Config to Driver "nv" Option "NvAgp" "0" Exit keystrokes ... are normal!
Step 6: Modified to Driver "nvidia" Option "NvAgp" "1"
Exit keystrokes ...are normal!
Step 7: Configured modem'd network, YOU'd to download NVIDIA modules, restarted with newly Nvidia'd startx.
Exit keystrokes ... normal!
Hmmm: Appears that using NVAGP, the NVIDIA driver, works, but AGPGART, the AGP module that comes with the linux kernel, freezes at exit !??
-Bill "Sticking to Kde at the moment for the Anti-Aliasing" Margolis
No more crash at kde/gnome logouts, now with NvAGP - /var/log/messages: ======================================================================= Jan 9 16:13:31 penguin PAM-unix2[708]: session started for user matt, service xdm Jan 9 16:14:19 penguin kernel: mtrr: no MTRR for d8000000,1000 found Jan 9 16:14:19 penguin kernel: unset mem range for 0xd8000000 0x2000000 Jan 9 16:14:19 penguin PAM-unix2[708]: session finished for user matt, service xdm Jan 9 16:14:20 penguin kdm[632]: Server for display :0 terminated unexpectedly Jan 9 16:14:22 penguin kdm[1081]: session start failed Jan 9 16:14:22 penguin kernel: mtrr: no MTRR for d8000000,1000 found Jan 9 16:14:22 penguin kernel: unset mem range for 0xd8000000 0x2000000 Jan 9 16:14:55 penguin PAM-unix2[1084]: session started for user matt, service xdm When crashing at kde/gnome logouts, with AGPART - /var/log/messages: ==================================================================== Jan 7 16:20:34 penguin kernel: NVRM: AGPGART: freed 16 pages Jan 7 16:20:34 penguin kernel: NVRM: AGPGART: backend released Jan 7 16:20:34 penguin PAM-unix2[690]: session finished for user matt, service xdm Jan 7 16:20:34 penguin kernel: NVRM: AGPGART: VIA Apollo KT133 chipset Jan 7 16:20:34 penguin kernel: NVRM: AGPGART: aperture: 64M @ 0xd8000000 Jan 7 16:20:34 penguin kernel: NVRM: AGPGART: aperture mapped from 0xd8000000 to 0xe7ad7000 Jan 7 16:20:34 penguin kernel: NVRM: AGPGART: mode 4x Jan 7 16:20:34 penguin kernel: NVRM: AGPGART: allocated 16 pages Jan 7 16:20:34 penguin kdm[641]: Server for display :0 terminated unexpectedly Jan 7 16:20:34 penguin kernel: NVRM: AGPGART: freed 16 pages Jan 7 16:20:35 penguin kernel: NVRM: AGPGART: allocated 16 pages Jan 7 16:20:36 penguin kdm[1491]: session start failed Jan 7 16:20:36 penguin kernel: NVRM: AGPGART: freed 16 pages Jan 7 16:20:36 penguin kernel: NVRM: AGPGART: backend released Jan 7 16:20:36 penguin kernel: Unable to handle kernel NULL pointer dereference at virtual address 00000000 Jan 7 16:20:36 penguin kernel: printing eip: Jan 7 16:20:36 penguin kernel: e29d420a Jan 7 16:20:36 penguin kernel: *pde = 00000000 Jan 7 16:20:36 penguin kernel: Oops: 0000 Jan 7 16:20:36 penguin kernel: CPU: 0 Jan 7 16:20:36 penguin kernel: EIP: 0010:[<e29d420a>] Jan 7 16:20:36 penguin kernel: EFLAGS: 00013056 Jan 7 16:20:36 penguin kernel: eax: 00000000 ebx: 00000000 ecx: 00000000 edx: dd906924 Jan 7 16:20:36 penguin kernel: esi: e7a98004 edi: 00000000 ebp: ddc5db24 esp: ddc5db10 Jan 7 16:20:36 penguin kernel: ds: 0018 es: 0018 ss: 0018 Jan 7 16:20:36 penguin kernel: Process X (pid: 1490, stackpage=ddc5d000) Jan 7 16:20:36 penguin kernel: Stack: 00000000 e7a98004 00000010 00004000 00000000 ddc5db4c e29d46bd e7a98004 Jan 7 16:20:36 penguin kernel: 00000010 ddc5db44 d83264e0 ddc5dc78 e2a67720 d83264e0 dd906924 ddc5db8c Jan 7 16:20:36 penguin kernel: e29c868f e7a98004 ddc5dc60 00000010 00000041 ddc5dc10 00000000 00000028 Jan 7 16:20:36 penguin kernel: Call Trace: [<e7a98004>] [<e29d46bd>] [<e7a98004>] [<e2a67720>] [<e29c868f>] [<e7a98004>] [<e29caafd>] Jan 7 16:20:36 penguin kernel: [<e2a67720>] [<e29cf8db>] [<e7a98004>] [<e29cf6f9>] [<e7a98004>] [<e7a98004>] [<e7a98004>] [<e29cc37c>] Jan 7 16:20:36 penguin kernel: [<e7ab3004>] [<e7a98004>] [<e2a2203c>] [<e7a98004>] [<e7a98004>] [<e29fa701>] [<e7a98004>] [<e7a98004>] Jan 7 16:20:36 penguin kernel: [<e7a980a6>] [<e29faa20>] [<e7a98004>] [<e7a98004>] [<e7a98004>] [<e29fbcc8>] [<e7a98004>] [<e7a98004>] Jan 7 16:20:36 penguin kernel: [<e2a637d3>] [<e7a9c348>] [<e7a98004>] [printer:__insmod_printer_O/lib/modules/2.4.4-4GB/kernel/drivers/usb+-1352381/96] [printer:__insmod_printer_O/lib/modules/2.4.4-4GB/kernel/drivers/usb+-1350289/96] [printer:__insmod_printer_O/lib/modules/2.4.4-4GB/kernel/drivers/usb+-1307136/96] [printer:__insmod_printer_O/lib/modules/2.4.4-4GB/kernel/drivers/usb+-1269692/96] [printer:__insmod_printer_O/lib/modules/2.4.4-4GB/kernel/drivers/usb+-1320112/96] Jan 7 16:20:36 penguin kernel: [printer:__insmod_printer_O/lib/modules/2.4.4-4GB/kernel/drivers/usb+-1317923/96] [printer:__insmod_printer_O/lib/modules/2.4.4-4GB/kernel/drivers/usb+-1321722/96] [printer:__insmod_printer_O/lib/modules/2.4.4-4GB/kernel/drivers/usb+-1402365/96] [printer:__insmod_printer_O/lib/modules/2.4.4-4GB/kernel/drivers/usb+-1288305/96] [__mark_inode_dirty+42/116] [generic_commit_write+135/148] [printer:__insmod_printer_O/lib/modules/2.4.4-4GB/kernel/drivers/usb+-1396592/96] [printer:__insmod_printer_O/lib/modules/2.4.4-4GB/kernel/drivers/usb+-1396477/96] Jan 7 16:20:36 penguin kernel: [<e29c77d9>] [<e2a67720>] [generic_file_write+1219/1584] [sys_ioctl+491/516] [system_call+51/64] Jan 7 16:20:36 penguin kernel: Jan 7 16:20:36 penguin kernel: Code: 80 3c 38 00 75 08 83 c3 07 e9 88 00 00 00 89 d8 c1 e8 03 8b Jan 7 16:20:36 penguin kdm[1494]: IO Error in XOpenDisplay Jan 7 16:20:36 penguin kdm[641]: Server for display :0 terminated unexpectedly Jan 7 16:20:36 penguin kdm[641]: Display :0 cannot be opened Jan 7 16:20:37 penguin kdm[1496]: IO Error in XOpenDisplay Jan 7 16:20:37 penguin kdm[641]: Display :0 cannot be opened Jan 7 16:20:37 penguin kernel: Unable to handle kernel NULL pointer dereference at virtual address 00000000 Jan 7 16:20:37 penguin kernel: printing eip: Jan 7 16:20:37 penguin kernel: e29d420a Jan 7 16:20:37 penguin kernel: *pde = 00000000 Jan 7 16:20:37 penguin kernel: Oops: 0000 Jan 7 16:20:37 penguin kernel: CPU: 0 Jan 7 16:20:37 penguin kernel: EIP: 0010:[<e29d420a>] Jan 7 16:20:37 penguin kernel: EFLAGS: 00013056 Jan 7 16:20:37 penguin kernel: eax: 00000000 ebx: 00000000 ecx: 00000000 edx: dd906884 Jan 7 16:20:37 penguin kernel: esi: e7a98004 edi: 00000000 ebp: dd9a3b24 esp: dd9a3b10 Jan 7 16:20:37 penguin kernel: ds: 0018 es: 0018 ss: 0018 Jan 7 16:20:37 penguin kernel: Process X (pid: 1495, stackpage=dd9a3000) Jan 7 16:20:37 penguin kernel: Stack: 00000000 e7a98004 00000010 00004000 00000000 dd9a3b4c e29d46bd e7a98004 Jan 7 16:20:37 penguin kernel: 00000010 dd9a3b44 d8326360 dd9a3c78 e2a67720 d8326360 dd906884 dd9a3b8c Jan 7 16:20:37 penguin kernel: e29c868f e7a98004 dd9a3c60 00000010 00000041 dd9a3c10 00000000 00000028 Jan 7 16:20:37 penguin kernel: Call Trace: [<e7a98004>] [<e29d46bd>] [<e7a98004>] [<e2a67720>] [<e29c868f>] [<e7a98004>] [<e29caafd>] Jan 7 16:20:37 penguin kernel: [<e2a67720>] [<e29cf8db>] [<e7a98004>] [<e29cf6f9>] [<e7a98004>] [<e7a98004>] [<e7a98004>] [<e29cc37c>] Jan 7 16:20:37 penguin kernel: [<e7abf004>] [printer:__insmod_printer_O/lib/modules/2.4.4-4GB/kernel/drivers/usb+-1352381/96] [printer:__insmod_printer_O/lib/modules/2.4.4-4GB/kernel/drivers/usb+-1350289/96] [printer:__insmod_printer_O/lib/modules/2.4.4-4GB/kernel/drivers/usb+-1307136/96] [printer:__insmod_printer_O/lib/modules/2.4.4-4GB/kernel/drivers/usb+-1269692/96] [printer:__insmod_printer_O/lib/modules/2.4.4-4GB/kernel/drivers/usb+-1320112/96] [printer:__insmod_printer_O/lib/modules/2.4.4-4GB/kernel/drivers/usb+-51371/96] [printer:__insmod_printer_O/lib/modules/2.4.4-4GB/kernel/drivers/usb+-51463/96] Jan 7 16:20:37 penguin kernel: [timer_bh+574/636] [printer:__insmod_printer_O/lib/modules/2.4.4-4GB/kernel/drivers/usb+-1352381/96] [printer:__insmod_printer_O/lib/modules/2.4.4-4GB/kernel/drivers/usb+-1350289/96] [printer:__insmod_printer_O/lib/modules/2.4.4-4GB/kernel/drivers/usb+-1307136/96] [printer:__insmod_printer_O/lib/modules/2.4.4-4GB/kernel/drivers/usb+-1269692/96] [printer:__insmod_printer_O/lib/modules/2.4.4-4GB/kernel/drivers/usb+-1320112/96] [printer:__insmod_printer_O/lib/modules/2.4.4-4GB/kernel/drivers/usb+-1317923/96] [printer:__insmod_printer_O/lib/modules/2.4.4-4GB/kernel/drivers/usb+-1321722/96] Jan 7 16:20:37 penguin kernel: [printer:__insmod_printer_O/lib/modules/2.4.4-4GB/kernel/drivers/usb+-1402365/96] [printer:__insmod_printer_O/lib/modules/2.4.4-4GB/kernel/drivers/usb+-1288305/96] [__mark_inode_dirty+42/116] [generic_commit_write+135/148] [printer:__insmod_printer_O/lib/modules/2.4.4-4GB/kernel/drivers/usb+-1396592/96] [printer:__insmod_printer_O/lib/modules/2.4.4-4GB/kernel/drivers/usb+-1396477/96] [<e29c77d9>] [<e2a67720>] Jan 7 16:20:37 penguin kernel: [generic_file_write+1219/1584] [sys_ioctl+491/516] [system_call+51/64] Jan 7 16:20:37 penguin kernel: Jan 7 16:20:37 penguin kernel: Code: 80 3c 38 00 75 08 83 c3 07 e9 88 00 00 00 89 d8 c1 e8 03 8b Jan 7 16:20:37 penguin kdm[1498]: IO Error in XOpenDisplay Jan 7 16:20:37 penguin kdm[641]: Display :0 cannot be opened Jan 7 16:20:37 penguin kernel: Unable to handle kernel NULL pointer dereference at virtual address 00000000 Jan 7 16:20:37 penguin kernel: printing eip: Jan 7 16:20:37 penguin kernel: e29d420a Jan 7 16:20:37 penguin kernel: *pde = 00000000 Jan 7 16:20:37 penguin kernel: Oops: 0000 Jan 7 16:20:37 penguin kernel: CPU: 0 Jan 7 16:20:37 penguin kernel: EIP: 0010:[<e29d420a>] Jan 7 16:20:37 penguin kernel: EFLAGS: 00013056 Jan 7 16:20:37 penguin kernel: eax: 00000000 ebx: 00000000 ecx: 00000000 edx: dd906704 Jan 7 16:20:37 penguin kernel: esi: e7a98004 edi: 00000000 ebp: dd9a3b24 esp: dd9a3b10 Jan 7 16:20:37 penguin kernel: ds: 0018 es: 0018 ss: 0018 Jan 7 16:20:37 penguin kernel: Process X (pid: 1497, stackpage=dd9a3000) Jan 7 16:20:37 penguin kdm[641]: Display :0 is being disabled (restarting too fast) Jan 7 16:20:37 penguin kernel: Stack: 00000000 e7a98004 00000010 00004000 00000000 dd9a3b4c e29d46bd e7a98004 Jan 7 16:20:37 penguin kernel: 00000010 dd9a3b44 d8326420 dd9a3c78 e2a67720 d8326420 dd906704 dd9a3b8c Jan 7 16:20:37 penguin kernel: e29c868f e7a98004 dd9a3c60 00000010 00000041 dd9a3c10 00000000 00000028 Jan 7 16:20:37 penguin kernel: Call Trace: [<e7a98004>] [<e29d46bd>] [<e7a98004>] [<e2a67720>] [<e29c868f>] [<e7a98004>] [<e29caafd>] Jan 7 16:20:37 penguin kernel: [<e2a67720>] [<e29cf8db>] [<e7a98004>] [<e29cf6f9>] [<e7a98004>] [<e7a98004>] [<e7a98004>] [<e29cc37c>] Jan 7 16:20:37 penguin kernel: [<e7ac1004>] [printer:__insmod_printer_O/lib/modules/2.4.4-4GB/kernel/drivers/usb+-1352381/96] [printer:__insmod_printer_O/lib/modules/2.4.4-4GB/kernel/drivers/usb+-1350289/96] [printer:__insmod_printer_O/lib/modules/2.4.4-4GB/kernel/drivers/usb+-1307136/96] [printer:__insmod_printer_O/lib/modules/2.4.4-4GB/kernel/drivers/usb+-1269692/96] [printer:__insmod_printer_O/lib/modules/2.4.4-4GB/kernel/drivers/usb+-1320112/96] [printer:__insmod_printer_O/lib/modules/2.4.4-4GB/kernel/drivers/usb+-1317923/96] [printer:__insmod_printer_O/lib/modules/2.4.4-4GB/kernel/drivers/usb+-1321722/96] Jan 7 16:20:37 penguin kernel: [printer:__insmod_printer_O/lib/modules/2.4.4-4GB/kernel/drivers/usb+-1352381/96] [printer:__insmod_printer_O/lib/modules/2.4.4-4GB/kernel/drivers/usb+-1350289/96] [printer:__insmod_printer_O/lib/modules/2.4.4-4GB/kernel/drivers/usb+-1307136/96] [printer:__insmod_printer_O/lib/modules/2.4.4-4GB/kernel/drivers/usb+-1269692/96] [printer:__insmod_printer_O/lib/modules/2.4.4-4GB/kernel/drivers/usb+-1320112/96] [printer:__insmod_printer_O/lib/modules/2.4.4-4GB/kernel/drivers/usb+-1317923/96] [printer:__insmod_printer_O/lib/modules/2.4.4-4GB/kernel/drivers/usb+-1321722/96] [printer:__insmod_printer_O/lib/modules/2.4.4-4GB/kernel/drivers/usb+-1402365/96] Jan 7 16:20:37 penguin kernel: [printer:__insmod_printer_O/lib/modules/2.4.4-4GB/kernel/drivers/usb+-1288305/96] [__mark_inode_dirty+42/116] [generic_commit_write+135/148] [printer:__insmod_printer_O/lib/modules/2.4.4-4GB/kernel/drivers/usb+-1396592/96] [printer:__insmod_printer_O/lib/modules/2.4.4-4GB/kernel/drivers/usb+-1396477/96] [<e29c77d9>] [<e2a67720>] [generic_file_write+1219/1584] Jan 7 16:20:37 penguin kernel: [sys_ioctl+491/516] [system_call+51/64] Jan 7 16:20:37 penguin kernel: Jan 7 16:20:37 penguin kernel: Code: 80 3c 38 00 75 08 83 c3 07 e9 88 00 00 00 89 d8 c1 e8 03 8b
Hi, from the nvidia mailing list I took another idea to fix the random freezes: It is to disable AGP 4x in the bios. For me it does fix one of the error messages below. On Wednesday 09 January 2002 17:27, Matt T. wrote: [...]
After examining the /var/log/XFree86.0.log file there is just one line left which irritates me, maybe someone can explain/correct this:
(WW) NVIDIA(0): Failed to allocate a DMA push buffer using AGP memory... (WW) NVIDIA(0): attempting to use PCI memory
After switching off AGP4x in the bios this error did go away. Now there is: (II) NVIDIA(0): AGP 2X successfully initialized HTH, Matt
participants (3)
-
Matt -
Matt T.
-
S.Toms