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

Newegg Daily Deals

Reply
 
Thread Tools
Old 12-05-06, 10:38 AM   #25
gilboa
Linux addict...
 
Join Date: Jan 2004
Posts: 540
Default Re: Nvidia drivers 9629 slowness issue

Quote:
Originally Posted by netllama
Yes, 1.0-9631 should fix this bug, however, you may need to set the __GL_ALWAYS_HANDLE_FORK environment variable to a non-zero value. See the driver README for more information on its usage.

Thanks,
Lonni
OK. Thanks.
I'll give it a try.

- Gilboa
__________________
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   Reply With Quote
Old 12-08-06, 09:45 AM   #26
brimborium
Registered User
 
Join Date: Dec 2006
Posts: 1
Default Re: Nvidia drivers 9629 slowness issue

Quote:
Originally Posted by gilboa
OK. Thanks.
I'll give it a try.

- Gilboa
does it work?
brimborium is offline   Reply With Quote
Old 12-08-06, 06:03 PM   #27
iki488
Registered User
 
Join Date: Dec 2006
Posts: 13
Default Re: Nvidia drivers 9629 slowness issue

no, I have the same issue with a geforce 4 ti4200 except that the issue appears when I enable compiz... I lose about 50% fps in glxgears.

I have the latest stable driver (9631), and I can't test with 9772 beta driver because it doesn't support my card....
iki488 is offline   Reply With Quote
Old 12-08-06, 06:12 PM   #28
netllama
NVIDIA Corporation
 
Join Date: Dec 2004
Posts: 8,763
Default Re: Nvidia drivers 9629 slowness issue

It is expected that there will always be a performance loss when using a Compositing manager as all the content has to be rendered off screen and then composited onto the desktop. The bug that was resolved for this thread was present even without Composite turned on.
netllama is offline   Reply With Quote
Old 12-08-06, 06:18 PM   #29
iki488
Registered User
 
Join Date: Dec 2006
Posts: 13
Default Re: Nvidia drivers 9629 slowness issue

Quote:
Originally Posted by netllama
Yes, 1.0-9631 should fix this bug, however, you may need to set the __GL_ALWAYS_HANDLE_FORK environment variable to a non-zero value. See the driver README for more information on its usage.

Thanks,
Lonni
I didn't find anything about __GL_ALWAYS_HANDLE_FORK in the readme :s
can you give me a link or explain here ?
iki488 is offline   Reply With Quote
Old 12-08-06, 06:27 PM   #30
netllama
NVIDIA Corporation
 
Join Date: Dec 2004
Posts: 8,763
Default Re: Nvidia drivers 9629 slowness issue

http://us.download.nvidia.com/XFree8...ppendix-e.html
netllama is offline   Reply With Quote
Old 12-08-06, 06:32 PM   #31
iki488
Registered User
 
Join Date: Dec 2006
Posts: 13
Default Re: Nvidia drivers 9629 slowness issue

Quote:
Originally Posted by netllama
ok and how can I set this variable ? and which value do you recommend ?
iki488 is offline   Reply With Quote
Old 12-09-06, 05:13 AM   #32
iki488
Registered User
 
Join Date: Dec 2006
Posts: 13
Default Re: Nvidia drivers 9629 slowness issue

up
iki488 is offline   Reply With Quote

Old 12-09-06, 01:01 PM   #33
fafreeman
Registered User
 
Join Date: Oct 2006
Posts: 23
Default Re: Nvidia drivers 9629 slowness issue

9631 works great for me. Went from the 9742 beta to the stable release since 9742 does not work at with wine(From what the wine devs tell me anways) and works great. No slowdowns or nothing.
__________________
Intel E6600 2.4GHZ with Zalman CNPS9500 LED | Kingston HyperX 2 DDR2 800MHZ 2GB | Asus P5N32 E SLI Plus | Nvidia 8800GTS SOC | Western Digital 320GB Sata 2 | Soundblaster Audigy 2 ZS | ATI TV Wonder Pro TV Card| Kingwin 600 Watt Modular | Viewsonic VX924 19inch LCD | Windows XP Pro

fafreeman is offline   Reply With Quote
Old 12-11-06, 05:06 AM   #34
gilboa
Linux addict...
 
Join Date: Jan 2004
Posts: 540
Default Re: Nvidia drivers 9629 slowness issue

Quote:
Originally Posted by brimborium
does it work?
Seems so.
Performance is back to normal.

- Gilboa
__________________
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   Reply With Quote
Old 12-12-06, 03:50 AM   #35
obmun
Registered User
 
Join Date: Sep 2006
Posts: 3
Default Re: Nvidia drivers 9629 slowness issue

So... apparently the slowdowns I'm experiencing are not related to this nvidia opengl <-> libc fork interaction.

I'm developing a simulation app using Ogre and have found amazing slowdowns (1/10 frame rate) updating from 87.76 drivers to 96.31 or 97.42 (both have been tested). I don't knot exactly which driver introduced this slowdown, but could check. Seems not to be card specific as I've been able to check this on two very different models: 7950GX2 and a old Quadro FX500 AGP. Both were on very different platforms: faster one was on a new Core 2 Duo (64 bits os) and the new one on an old Sempron running in 32 bits mode.

Esentially the problems seems to be an amazing rise of libc calls (theorically from the nvidia ogl lib):

97.42 beta driver (__GL_ALWAYS_HANDLE_FORK enabled):
CPU: Core 2, speed 2671.24 MHz (estimated)
Counted CPU_CLK_UNHALTED events (Clock cycles when not halted) with a unit mask of 0x00 (Unhalted core cycles) count 100000
CPU_CLK_UNHALT...|
samples| %|
------------------
205104 44.8090 libGLcore.so.1.0.9742
118398 25.8664 libc-2.5.so
53012 11.5815 no-vmlinux
48592 10.6159 libode.so

Driver 87.76:
CPU: Core 2, speed 2671.24 MHz (estimated)
Counted CPU_CLK_UNHALTED events (Clock cycles when not halted) with a unit mask of 0x00 (Unhalted core cycles) count 100000
CPU_CLK_UNHALT...|
samples| %|
------------------
223698 45.5036 libOgreMain.so.12.0.0
168020 34.1778 libGLcore.so.1.0.8776
56667 11.5269 libode.so
11182 2.2746 no-vmlinux
7350 1.4951 libc-2.5.so

The only difference between oprofile reports is nvidia driver version. As already mentioned, FORK option was enabled. The main problem is my system has 2.5 glibc nptonly compiled in, so no LinuxThreads; therefore the fork / no fork same results were expected.

Cannot confirm if in the FX500 system the oprofile report is the same as it was not profiled.

Has anyone else seen this behaviour?
obmun 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
NVIDIA Responds to Reports of Kepler V-Sync Stuttering Issue Rieper NVIDIA GeForce 600 Series 13 03-03-13 10:56 PM
xorg locks-up with newest nvidia drivers w/ vdpau. theroot NVIDIA Linux 1 06-24-12 11:04 AM
NVIDIA Drivers Receive Windows 8 Certification News Archived News Items 0 06-01-12 05:30 AM
Radeon 9700 not all that? sancheuz Other Desktop Graphics Cards 200 10-12-02 09:31 PM

All times are GMT -5. The time now is 08:06 PM.


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