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

Newegg Daily Deals

Reply
 
Thread Tools
Old 02-04-04, 04:09 PM   #1
Brian Ellis
The Linux Guy
 
Join Date: Aug 2003
Location: Kingsport Tn
Posts: 6
Send a message via Yahoo to Brian Ellis
Default Clock Skew Anomaly from GFFX 5600 + 1.0.5336 + 2.6.0

I have just noticed a clock skew that occurs under periods of high texture traffic caused by loading textures at a rate of about 240 times per second in blocks of 256x256x16-bit. I examined the clocks of two systems concurrently and in a period of 10 min, the one running my gl app had fallen behind by about 21% (2:08)! This may not be caused by software alone, so I am listing my entire hardware and software spec as best as I can. I will post more information about this as soon as I run some more tests. There were lots of kernel messages to the likes of:

0: nvidia: trying to map 0xd79f1000 to kernel space, but we're in an interrupt or holding a spinlock

This may possibly explain missing timer interrupts.

Hardware Spec:
MB: Biostar M7VIB v1.2 - VIA KT133 chipset
CPU: AMD Athlon 1.2Ghz
RAM: 512MB Kingston PC133
GPU: nVidia GeForceFX 5600 - MicroStar Inc MSI5600-VD-128

Software Spec:
Linux Kernel: 2.6.0 compiled with gcc-2.95.3
XFree86: 4.3.0.1 compiled with gcc-3.3.1
nVidia-drivers: 1.0-5336 AGP Disabled

I thank you for your linux support,
Brian

EDIT: I had mistakenly put M7-VIP as the motherboard, I was looking at the documentation for the wrong system. The correct MB is the M7-VIB

EDIT: More corrections : AGP is actually disabled (I should probably enable it), and the video card has 128M not 256 (my aspirations got the better of me)

Last edited by Brian Ellis; 02-04-04 at 04:31 PM.
Brian Ellis is offline   Reply With Quote
Old 02-04-04, 05:41 PM   #2
Brian Ellis
The Linux Guy
 
Join Date: Aug 2003
Location: Kingsport Tn
Posts: 6
Send a message via Yahoo to Brian Ellis
Default Follow up

After enabling AGP, performance is back to expected (go figure). There are still some kernel messages about mapping interrupts durring spinlock, but they are far less frequent. This may still be causing a slight clock drift. I think the only redeeming thing about my posting this thread is to indicate that under terrible conditions, there can be a bit of a clock drift caused by this problem.
Brian Ellis 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


Similar Threads
Thread Thread Starter Forum Replies Last Post
Base mosaic and rotation? happyfunbot NVIDIA Linux 5 07-13-12 01:23 PM

All times are GMT -5. The time now is 07:41 PM.


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