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

Newegg Daily Deals

Closed Thread
 
Thread Tools
Old 04-14-05, 02:15 PM   #13
nagual.hsu
Registered User
 
Join Date: Feb 2005
Posts: 19
Default Re: Trying to make headway into finding the Xid crashes source...

Quote:
Originally Posted by gilboa
What I don't get is this: If indeed the driver fails in user-land context, how come the user-land application doesn't get stuck (be that in the first function, or the proceeding poll attempts) in an un-interruptible sleep? (Zombie)
Cheers,
If I understood you correctly, user-land AP would become Zombile if
it can't return from its kernel space part( ie. in this case, somewhere
within the ioctl() call).

However user-land application (ie. X) can be killed. So this is my blind
guess:

User-land application does some kind of 'poll' on hardware back and
forth within a loop. This kind of 'poll' would wait a definately
very small mount of time and then return, no matter the result is
acceptable or not. Then the user-land application check
the 'poll' result and finds that is unacceptable, so the user-land
application decide to continue loops with great confidence
of getting an acceptable result in the very short future.
But the acceptable result never comes.

When the user-land application 'poll/ioctl', it must have resched
happened. In one side I guess this user-land application(X) must
have 'sched_setscheduler' to some real time scheduling policy
(Round-robin or FIFO) in order to offer services to many
Xwindow APs. In the other side, when the user-land application
using ioctl() to poll the hardware, a resched happens and that's
why we can remotely ssh/telnet to the 'freezing' PC.

Basically, my guess is like yours. I prefer they add some sorts
of flags in nvidia's driver/lib (Maybe an big array that can be accessed
and logged by another program or module). I believe sometimes
printk is not that synchronous.

Sincerely, nagual
nagual.hsu is offline  
Old 04-14-05, 03:48 PM   #14
LubosD
Registered User
 
Join Date: Jan 2005
Location: Czech Republic
Posts: 451
Send a message via ICQ to LubosD
Default Re: Trying to make headway into finding the Xid crashes source...

I've turned on "Prefetch caching" in my BIOS and it seems to stabilize the system a bit.
LubosD is offline  
Old 04-14-05, 05:20 PM   #15
gilboa
Linux addict...
 
Join Date: Jan 2004
Posts: 540
Default Re: Trying to make headway into finding the Xid crashes source...

Quote:
Originally Posted by LubosD
I've turned on "Prefetch caching" in my BIOS and it seems to stabilize the system a bit.
How stable? rock solid or only crashes once every 5 minutes...

Quote:
Originally Posted by nagual.hsu
User-land application does some kind of 'poll' on hardware back and
forth within a loop. This kind of 'poll' would wait a definately
very small mount of time and then return, no matter the result is
acceptable or not. Then the user-land application check
the 'poll' result and finds that is unacceptable, so the user-land
application decide to continue loops with great confidence
of getting an acceptable result in the very short future.
But the acceptable result never comes.

When the user-land application 'poll/ioctl', it must have resched
happened. In one side I guess this user-land application(X) must
have 'sched_setscheduler' to some real time scheduling policy
(Round-robin or FIFO) in order to offer services to many
Xwindow APs. In the other side, when the user-land application
using ioctl() to poll the hardware, a resched happens and that's
why we can remotely ssh/telnet to the 'freezing' PC.
I agree with the first part... the second is more problematic.
If you check my glgears, you'll see that the GL indeed yields the CPU between polls. However, because I ran the software as user, it had no access to SCHED_FIFO or SCHED_RR (both require root priv'); however, it may not matter; even with SCHED_NORMAL, as long as you don't have one OpenGL application, you should be able to poll the device periodically without a problem.

Never the less, both our guesses are very close; it'll be nice if someone in nVidia is actually reading this...


BTW, did anyone try the debug flag (NVreg_ResmanDebugLevel=0) I mentioned in my first post?
__________________
DEV-NG: Intel S2600C0, 2xE52658V2, 32GB, 4x2TB, GTX680, F19/x86_64, Dell U2711.
DEV: Intel S5520SC, 2xX5680, 36GB, 5x320GB, GTX550, F19/x86_64, Dell U2711 (^).
SRV: Tyan Tempest i5400XT, 2xE5335, 8GB, 4x2TB, 9800GTX, F19/x86-64, Dell U2412.
LAP: ASUS N56VJ, i7-3630QM, 16GB, 1TB, 635M, F19/x86_64.
gilboa is offline  
Old 04-14-05, 11:47 PM   #16
nagual.hsu
Registered User
 
Join Date: Feb 2005
Posts: 19
Default Re: Trying to make headway into finding the Xid crashes source...

Quote:
Originally Posted by gilboa
I agree with the first part... the second is more problematic.
If you check my glgears, you'll see that the GL indeed yields the CPU between polls. However, because I ran the software as user, it had no access to SCHED_FIFO or SCHED_RR (both require root priv'); however, it may not matter; even with SCHED_NORMAL, as long as you don't have one OpenGL application, you should be able to poll the device periodically without a problem.
What I mean is not your glgears but X. I run programs using Xvideo extension, which
is 'Video Blitter' actually(a false 2D function, actually 3D to imitate 2D, thus using 3D
functions of the hardware, instead of real XV). OpenGL does not go through X but
may need some cooperation with X. When I run my programs, after
a long time, most computers freeze with X eating 100% cpu usage. But few computers
freeze within 2 hours. They have all kinds of motherboards/chipsets.
This problem really drives us nuts. I just had to write a small program to
monitor X and resstart X automatically when sh*t happens.
nagual.hsu is offline  
Old 04-15-05, 12:34 AM   #17
nukem
Registered User
 
Join Date: Dec 2004
Posts: 226
Default Re: Trying to make headway into finding the Xid crashes source...

I have the exact same problem. What I have found is that it happens on some games(doom3 nwn or anything using cedega) and others work fine(q3a et glxgears). The solution Ive figured out for this is to run glxgears before running the game. This enables me to play any game just fine. If I go from one game right to the other my system does lock. I have to let it sit for awhile before I can switch games. The reason for this is that the card screws up switching from 2D mode to semi-3D mode. This is best explained here. Ive also read that some windows users are seeing this problem.

My system specs:

OS: Gentoo Linux 2005.0 x86_64
CPU: AMD64 3500+
Video Card: eVGA nVidia FX 5900 Ultra
Mother Board: ASUS A8V Delux
nukem is offline  
Old 04-15-05, 03:41 AM   #18
LubosD
Registered User
 
Join Date: Jan 2005
Location: Czech Republic
Posts: 451
Send a message via ICQ to LubosD
Default Re: Trying to make headway into finding the Xid crashes source...

Quote:
Originally Posted by nukem
Ive also read that some windows users are seeing this problem.
Yes, I sometimes have colourful crashes in Windows when switching to 3D. The solution is to have the 3D frequency and 2D frequency the same.

Then I sometimes get blue screen in Half Life 2 - the driver is in infinite loop.
LubosD is offline  
Old 04-15-05, 05:18 AM   #19
gilboa
Linux addict...
 
Join Date: Jan 2004
Posts: 540
Default Re: Trying to make headway into finding the Xid crashes source...

Now... the only thing missing in this thread is an nVidia rep :/
__________________
DEV-NG: Intel S2600C0, 2xE52658V2, 32GB, 4x2TB, GTX680, F19/x86_64, Dell U2711.
DEV: Intel S5520SC, 2xX5680, 36GB, 5x320GB, GTX550, F19/x86_64, Dell U2711 (^).
SRV: Tyan Tempest i5400XT, 2xE5335, 8GB, 4x2TB, 9800GTX, F19/x86-64, Dell U2412.
LAP: ASUS N56VJ, i7-3630QM, 16GB, 1TB, 635M, F19/x86_64.
gilboa is offline  
Old 04-15-05, 07:51 AM   #20
nukem
Registered User
 
Join Date: Dec 2004
Posts: 226
Default Re: Trying to make headway into finding the Xid crashes source...

Quote:
Originally Posted by LubosD
Yes, I sometimes have colourful crashes in Windows when switching to 3D. The solution is to have the 3D frequency and 2D frequency the same.

Then I sometimes get blue screen in Half Life 2 - the driver is in infinite loop.
I would think that nvidia would take this problem more seriously since it seems that its a problem in there drive on every platform.
nukem is offline  

Old 04-15-05, 03:27 PM   #21
Ironi
Registered User
 
Join Date: Dec 2004
Posts: 86
Default Re: Trying to make headway into finding the Xid crashes source...

Quote:
Originally Posted by nukem
I would think that nvidia would take this problem more seriously since it seems that its a problem in there drive on every platform.
"Colorful crashes when switching to 3D" is not the same problem being discussed in this thread, AFAIK. The "active mouse pointer freeze" (where X uses 100% CPU but SysRq and SSHing in still work) is the issue that I thought people were trying to shed more light on, and it happens even without running a 3D app.

As for nVidia taking the latter issue seriously, I'm not convinced that the fault lies with driver: 6629 worked fine for me up until a certain point, after which 6629 and later drivers would freeze up eventually. I hadn't touched Xorg's config, so nothing changed there.
Ironi is offline  
Old 04-15-05, 06:00 PM   #22
nukem
Registered User
 
Join Date: Dec 2004
Posts: 226
Default Re: Trying to make headway into finding the Xid crashes source...

I think this is all related in some way. The bug shows itself different ways but its all the same.
nukem is offline  
Old 04-15-05, 07:13 PM   #23
technikolor
Registered User
 
Join Date: Jul 2003
Posts: 18
Default Re: Trying to make headway into finding the Xid crashes source...

More and more I'm wondering why I pay $200+ for a card that gives me nothing but headaches. I miss my Matrox. I remember when I could use a console framebuffer, it was nice.
__________________
Tyan ThunderK7
Dual AthlonMP 1.2G
1GB PC2100 Reg ECC
XFX 6600GT AGP
TrueCombat: Q3A Mod - truecombat.com
technikolor is offline  
Old 04-15-05, 07:22 PM   #24
nukem
Registered User
 
Join Date: Dec 2004
Posts: 226
Default Re: Trying to make headway into finding the Xid crashes source...

You can still use a console framebuffer with a nvidia card its just that you wont be able to run any X apps. My nvidia card(FX 5900 ultra) works without any problems on my P4 when I upgraded it works well for a few months then I get the XiD error. Anyway try running glxgears before you run a game, it might "fix" it.
nukem 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 12:42 AM.


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