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

Newegg Daily Deals

Reply
 
Thread Tools
Old 01-11-07, 06:18 AM   #1
colchaodemola
Registered User
 
Join Date: Jan 2007
Posts: 18
Default Find out is my nvidia is turbo cache

Is there a way to find out if my nvidia card that came with my notebook is turbo cache or not ? Some linux tool for example. nvidia-settings shows 512Mb but the seller told me it was 128MB so i am not sure.
colchaodemola is offline   Reply With Quote
Old 01-11-07, 03:55 PM   #2
whig
550Ti
 
Join Date: Jan 2004
Location: New Zealand
Posts: 854
Default Re: Find out is my nvidia is turbo cache

Get two pieces of info:

Code:
$ lspci -v
...
01:00.0 VGA compatible controller: nVidia Corporation Unknown device 02e0 (rev a2) (prog-if 00 [VGA])
        Subsystem: LeadTek Research Inc. Unknown device 2a57
        Flags: bus master, 66MHz, medium devsel, latency 248, IRQ 16
        Memory at e8000000 (32-bit, non-prefetchable) [size=16M]
        Memory at d0000000 (32-bit, prefetchable) [size=256M]
        Memory at e9000000 (32-bit, non-prefetchable) [size=16M]
        [virtual] Expansion ROM at ea000000 [disabled] [size=128K]
        Capabilities: <access denied>
Compare it to /var/log/Xorg.0.log:
Code:
(II) NVIDIA(0): NVIDIA GPU GeForce 7600 GT at PCI:1:0:0 (GPU-0)
(--) NVIDIA(0): Memory: 262144 kBytes
If the X log is greater than lspci, turbocache is being used. (Correct me if wrong...)
whig is offline   Reply With Quote
Old 01-11-07, 04:21 PM   #3
colchaodemola
Registered User
 
Join Date: Jan 2007
Posts: 18
Default Re: Find out is my nvidia is turbo cache

Code:
01:00.0 VGA compatible controller: nVidia Corporation GeForce Go 7400 (rev a1) (prog-if 00 [VGA])
        Subsystem: Hewlett-Packard Company Unknown device 30a5
        Flags: bus master, fast devsel, latency 0, IRQ 16
        Memory at d1000000 (32-bit, non-prefetchable) [size=16M]
        Memory at c0000000 (64-bit, prefetchable) [size=256M]
        Memory at d0000000 (64-bit, non-prefetchable) [size=16M]
        Capabilities: [60] Power Management version 2
        Capabilities: [68] Message Signalled Interrupts: 64bit+ Queue=0/0 Enable-
        Capabilities: [78] Express Endpoint IRQ 0
        Capabilities: [100] Virtual Channel
        Capabilities: [128] Power Budgeting

Code:
(II) NVIDIA(0): NVIDIA GPU GeForce Go 7400 at PCI:1:0:0 (GPU-0)
(--) NVIDIA(0): Memory: 524288 kBytes
Does that mean that i have 256Mb dedicated and 256Mb shared [total 512Mb]?
colchaodemola is offline   Reply With Quote
Old 01-11-07, 05:40 PM   #4
ioannis
Registered User
 
Join Date: Nov 2006
Posts: 21
Default Re: Find out is my nvidia is turbo cache

what happens if it's smaller ?
ioannis is offline   Reply With Quote
Old 01-11-07, 07:01 PM   #5
whig
550Ti
 
Join Date: Jan 2004
Location: New Zealand
Posts: 854
Default Re: Find out is my nvidia is turbo cache

I suspect so... benchmark with and without shared memory (if that's possible).

Also do
Code:
$ free -m
             total       used       free     shared    buffers     cached
Mem:          1004        992         12          0        182        423
-/+ buffers/cache:        386        618
Swap:         1058          0       1058
I know I've got 1GB of ram, and the 1004 total is close to that. If your free -m is missing 256MB we know where it's gone...
whig is offline   Reply With Quote
Old 01-11-07, 07:13 PM   #6
colchaodemola
Registered User
 
Join Date: Jan 2007
Posts: 18
Default Re: Find out is my nvidia is turbo cache

Well , i have 1009Mb
Now i really do not know how many dedicated memory i have.


Code:
#free -m
             total       used       free     shared    buffers     cached
Mem:          1009        821        188          0         70        446
-/+ buffers/cache:        304        704
Swap:         1027          0       1027
colchaodemola is offline   Reply With Quote
Old 01-11-07, 09:17 PM   #7
whig
550Ti
 
Join Date: Jan 2004
Location: New Zealand
Posts: 854
Default Re: Find out is my nvidia is turbo cache

Maybe turbocache grabs system memory on demand? If you have a app/game which uses 512 vram run it and run free -m while it's still active.
whig is offline   Reply With Quote
Old 01-12-07, 04:05 AM   #8
lloeki
Arch Linux
 
lloeki's Avatar
 
Join Date: Oct 2006
Posts: 122
Default Re: Find out is my nvidia is turbo cache

Code:
01:00.0 VGA compatible controller: nVidia Corporation GeForce Go 7400 (rev a1) (prog-if 00 [VGA])
        Subsystem: Dell Unknown device 01d7
        Flags: bus master, fast devsel, latency 0, IRQ 169
        Memory at dd000000 (32-bit, non-prefetchable) [size=16M]
        Memory at c0000000 (64-bit, prefetchable) [size=256M]
        Memory at de000000 (64-bit, non-prefetchable) [size=16M]
        [virtual] Expansion ROM at df000000 [disabled] [size=128K]
        Capabilities: [60] Power Management version 2
        Capabilities: [68] Message Signalled Interrupts: 64bit+ Queue=0/0 Enable-
        Capabilities: [78] Express Endpoint IRQ 0
        Capabilities: [100] Virtual Channel
        Capabilities: [128] Power Budgeting
but I have 64MB on-board only , and a total of 256 with turbocache, correctly reported by X
Code:
(II) NVIDIA(0): NVIDIA GPU GeForce Go 7400 at PCI:1:0:0 (GPU-0)
(--) NVIDIA(0): Memory: 262144 kBytes
so I don't think lspci is relevant. I think this only give the on-board memory the card could have.

turbocache will certainly not reduce the 'total' field (this would ipmly some form of memory hotplugging, and memory hotplugging is not a feature I know of on x86), so turbocache certainly allocates memory via the kernel.
I think turbocache allocates memory at X startup, which could be revealed by a +200~250MB used memory once I start X (usually a trivial X would take ~30MB, plus the 192MB lacking to reach 256MB). I'll have to test to confirm that.
A second thing is that vram is calculated at X startup, so either:
- it is allocated on X startup, and the number told to X is always correct.
- the driver is lying to X with a highest vram possibly available, and it is then allocated on demand. but then, what would happen when running out of ram(+swap) and vram needs more allocation?
so, really, I think allocation on X startup is correct.

additionally, you can read here: http://www.nvidia.com/page/turbocache.html and find out that it seems to be dynamically allocated. but does dynamically means 'on demand' or 'on driver initialization' (thus it is dynamic, as opposed to a setting in the system bios, like for old shared ram). plus that marketing may apply to windows only, and behavior under linux may just as well be different.

Last edited by lloeki; 01-12-07 at 04:27 AM.
lloeki 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
Need Help Installing NVIDIA Tesla M2070Q in Linux RHEL5 Ferianto85 NVIDIA Linux 0 05-18-12 09:35 PM
nvidia: What makes our new Kepler GPU architecture tick? Find out at #GTC12 http://t. News Archived News Items 0 05-15-12 06:30 PM
Rumor regarding lack of 680 availability ViN86 Rumor Mill 6 05-09-12 05:48 PM
Error! Linux cannot find my nVidia integrated sound card! Please help! targatop NVIDIA Linux 1 09-08-02 05:55 PM

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


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