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

Newegg Daily Deals

Closed Thread
 
Thread Tools
Old 01-28-06, 07:39 PM   #301
chunkey
#!/?*
 
Join Date: Oct 2004
Posts: 662
Default Re: Trying to make headway into finding the Xid crashes source...

or:

c) Hardwarebugs... But nooo, there's nothing on nVidias page that talks about these _design_limitations, in contrast to Intel or AMD, where you just _download_ erratas without signing an NDA or anything else...
chunkey is offline  
Old 01-29-06, 04:13 AM   #302
Dabljuh
Registered User
 
Join Date: Apr 2005
Posts: 27
Default Re: Trying to make headway into finding the Xid crashes source...

Er, that's almost what I meant saying "A massive proportion of nVidia cards are faulty".

Its probably not a Design specific limitation, as for many people, their GF6800GTs and GF6600GTs run just fine. But for a huge number of people, they produce erratic XIDs (crashes) that can only be circumvented by massively reducing the speed of the card, e.g. by disabling AGP support.

Because it's been about a year now, NO RELIABLE WORKS-FOR-EVERYONE SOLUTION IS KNOWN other than "get a replacement card"
Dabljuh is offline  
Old 01-29-06, 07:39 PM   #303
cleverhandle
Registered User
 
Join Date: Jul 2004
Posts: 10
Default Re: Trying to make headway into finding the Xid crashes source...

Just to add another data point to this whole affair...

I was having problems pretty typical to those reported in this thread - X hanging by gobbling up CPU, sometimes killable via ssh. Problems started after replacing a dead 5700Ultra with a 6600, like several others. Hangs manifested themselves apparently randomly while browsing, switching between windows, etc. but were inevitable when playing games via XMAME. UT2003 worked fine, so it wasn't a GL issue per se.

After trying many, many things my current solution has been lowering my AGP aperture BIOS setting to 32MB from 64MB - I don't remember seeing that mentioned here, so maybe that will be of some help to someone. Also, I haven't really run the system that long yet, but I was able to play some XMAME games for extended periods, so it's at least a step in the right direction. Finally, I can't yet tell whether there's any performance hit from changing the aperture - it does seem rather small by current standards (this is with a 6600 w/256MB and 512MB of system RAM). Most of the comments I see about aperture size on the web appear to be a lot of people guessing and talking out of their rear ends.

Here's the summary of my current config - I attached a bug report with all the details as well:

Debian Stable (Sarge)
Kernel 2.6.7
NVidia Driver 6629
NVAGP
RenderAccel and HWCursor Off
APIC and ACPI On

All of those things (and many more) I had tested various ways before I changed the aperture setting, to no effect. If I have the energy, I may go back and fiddle with them some more if my current config does turn out to be truly stable, rather than just less unstable.

Finally, for those just now encountering the whole "X hangs" issue and skipping to this part of the thread, be assured that this is not just an nVidia problem. If you search the web, you'll see this happening with both ATI and nVidia users (though, perhaps, more frequently with nVidia), across distributions and hardware eras, and with both XFree/XOrg. So don't bawl out nVidia alone. There's clearly a very low level instability lurking here, probably touching the kernel, X, drivers, and hardware implementations among other things. Since so many of the "works-for-me solutions" have to do with low-level timings of hardware and busses, I suspect that all this is just the consequence of manufacturers pushing their hardware closer and closer to the limits. Hopefully, the people responsible for all of these low-level Linux components will be able to tighten everything up and eliminate the crashes.
Attached Files
File Type: log nvidia-bug-report.log (68.5 KB, 214 views)
cleverhandle is offline  
Old 01-30-06, 11:18 AM   #304
Dabljuh
Registered User
 
Join Date: Apr 2005
Posts: 27
Default Re: Trying to make headway into finding the Xid crashes source...

Well, you see, there have been approximately 140 thousand "possible" solutions posted, but not the first commend from any nvidia rep. Most people on the internet are talking out of their asses, especially when it comes to lower level technical stuff, so your solution to decrease agp aperture doesn't exactly have to work either. Another fun detail about that: the 8xxx drivers have some bug in them that only allows 24 mb of agp aperture, (actually a regression, the same bug has been in various older versions as well) which DOES have detrimental effect on performance of certain games (especially in wine emulated DXGames). nVidia's linux driver team isn't exactly up to par. And this may also mean that if you experience crashes with the 8xxx drivers, changing the bios agp aperture won't change anything about the XIDs either.

Oh, and one way to reduce the occurence of the XIDs has always been to decrease the performance of the card. By disabling AGP, using slower drivers, underclocking it etc...

If it isn't nvidia's fault, why didn't they show up meanwhile and tell us what the problem is? If it is something else (kernel etc) they'd certainly happily point at that direction, shifting the blame. But kernels come and go, the problems with nvidia cards persist. the high number of "works for me" posts indicates all but one thing: Some people are not affected by the xids. Some people are affected only by 3d related xids. some people have crashes all over the place. Reason? We don't know. Software solutions don't work, hardware solutions don't work, the only reasonable possibility that's left is that a huge portion of nvidia cards are faulty.

Last edited by Dabljuh; 01-30-06 at 11:30 AM.
Dabljuh is offline  
Old 01-30-06, 02:46 PM   #305
chunkey
#!/?*
 
Join Date: Oct 2004
Posts: 662
Default Re: Trying to make headway into finding the Xid crashes source...

Fine... but what do we do with our faulty cards? I mean it costs me $200 when I bought it in january last year? (Yes, I tried RMA, but they send me the same card back,...) Or get a new one and hope there's a REV 2.00 that doesn't have these bugs anymore?
chunkey is offline  
Old 01-31-06, 08:49 AM   #306
Dabljuh
Registered User
 
Join Date: Apr 2005
Posts: 27
Default Re: Trying to make headway into finding the Xid crashes source...

Warranty case. You got a faulty product. You deserve your money back. Also, write the magazines about the issue. Maybe if there's some sort of media echo, nVidia launches a recall.
Dabljuh is offline  
Old 02-02-06, 06:28 AM   #307
yain
Registered User
 
Join Date: Feb 2006
Posts: 1
Default Re: Trying to make headway into finding the Xid crashes source...

FWIW, it's not a Linux specific problem despite it being discussed in this part of the forum. I have exactly the same effects on FreeBSD 6 with GeForce2 MX/MX 400, the only difference, perhaps, is that I can more or less consistently reproduce this probem with some GLUT code.

Also I have very hard time believing that this is hardware fault since my card worked on "the other OS" for several years just fine.
yain is offline  
Old 02-06-06, 10:46 AM   #308
chunkey
#!/?*
 
Join Date: Oct 2004
Posts: 662
Default Re: Trying to make headway into finding the Xid crashes source...

there's also an other pov? is the board(chipset) faulty?
since, many people had problems with NF3 & GeForce 6(check this thread)... but not with a KT800 & GeForce 6.
chunkey is offline  

Old 02-06-06, 11:01 AM   #309
ticoli
Registered User
 
Join Date: Jan 2006
Posts: 13
Default Re: Trying to make headway into finding the Xid crashes source...

Perhaps ... i'vnt got via motherboard for the test ... but if somebody can do it (it's the job of nvidia, but i'm fed up praying), i could be very informative!
ticoli is offline  
Old 02-16-06, 09:50 AM   #310
cococr
Registered User
 
Join Date: Jan 2006
Posts: 2
Default Re: Trying to make headway into finding the Xid crashes source...

Bizarre... I haven't had my PC lock up in over a week, where normally it would within a few hours. The only difference is that I had to switch to a new mouse (I had an old PS/2 IntelliMouse, and now I have a USB IntelliMouse Explorer (optical)).
cococr is offline  
Old 02-16-06, 09:58 AM   #311
ticoli
Registered User
 
Join Date: Jan 2006
Posts: 13
Default Re: Trying to make headway into finding the Xid crashes source...

uh?
a few hours you was lucky yet!... I can't start xwindow at all...

For the mouse, i think it could be bad spirits ... perhaps you have see a marabout or something else?
ticoli is offline  
Old 02-16-06, 10:32 AM   #312
JaXXoN
Registered User
 
Join Date: Jul 2005
Location: Munich
Posts: 910
Default Re: Trying to make headway into finding the Xid crashes source...

BTW.: after updating to a more recent kernel, i also had system
freezes or silent reboots within minutes. After doing a lot of
experimentation, i finally figured out that applying "acpi=off"
did the trick for me. I have the impression that there are a lot
of mainboards with broken ACPI BIOSes out there. It seems,
that for some of them, the kernel does better work-arounds than
for others (or in my case even worse after updating the kernel).

regards

Bernhard
JaXXoN is offline  
Closed Thread


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


Similar Threads
Thread Thread Starter Forum Replies Last Post
Random crashes, NVRM Xid messages Iesos NVIDIA Linux 90 10-04-12 04:27 AM

All times are GMT -5. The time now is 07:33 PM.


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