6200Go Turbo Cache (128MB)

[cellstije]

HI folks.

I'm having some troubles with a nvidia 6200Go Turbo Cache (128MB) on a sony vaio fs115m.
The os is (k)ubuntu hoary, with the latest drivers 1-76.76

the card is working so far in 2d and 3d, but:

the card is recognised as a 6200Go with 64MB (nvidia-settings)

an lspci gives:
VGA compatible controller: nVidia Corporation: Unknown device 0167 (rev a1)

and everytime i start a 3d application i get this message in /var/log/kern.log:
NVRM: bad caching on address 0xc9056000: actual 0x163 != expected 0x173
NVRM: please see the README section on Cache Aliasing for more information
NVRM: bad caching on address 0xc9056000: actual 0x163 != expected 0x173

and the laptop temperature rises very quickly even if the cpu is idle most of the time (and at minimum of the freq. scaling - powersave governor).

I tried to down clock in the pannel but the

Option "Coolbits" "1"

in xorg.conf does not work...

I used to deal with ati cards... so completely newbie here.

Suggestions?

[netllama]

cellstije,
lspci output is obtained from your kernel's PCI ID database. The fact that lspci shows your card as being 'Unknown' only means that your kernel's PCI ID database may not be up to date. There isn't normally any functionality or performance problems related to this.

The bad caching warnings are a kernel bug that is resolved in 2.6.11 and later kernels. Cache aliasing can result in instability, including lockups, so it is advisable to upgrade your kernel if at all possible.

When you say the laptop temperature is rising, do you actually mean the CPU temperature, or the temperature of your videocard? Where are you seeing this temperature reading?

Thanks,
Lonni

[Thunderbird]

Can't be the gpu temperature as the sensor is locked, guess it is some i2c sensor.

[cellstije]

Hi folks

first of all thanks for replying!

I did not find much about those warnings on the net ... i'll upgrade the kernel (i'm on the standard ubuntu 2.6.10-5-686) asap!

about the temperature: i do not read it anywhere, just the fan(s) starts working hard even if the cpu (centrino) is in powersave mode and its idle is 80%.
Btw, is there a way to make the coolbits option working?

Is thre anything i can do to make the turbo cache working?
I suppose it is not working since the driver does not recognise the card as such ...

Cheers
c.

[netllama]

cellstije,
Fan speeds (external to a videocard) are controlled by ACPI, which is managed by your kernel & your SBIOS. If your laptop's cooling fan is always spinning, that's a kernel and/or BIOS issue.

What leads you to believe that turbocache is not working?

The Coolbits option is not supported on mobile (Go) GPUs, and only on NV3x and higher GPUs.

Thanks,
Lonni

[cellstije]

Originally Posted by netllama

cellstije,
Fan speeds (external to a videocard) are controlled by ACPI, which is managed by your kernel & your SBIOS. If your laptop's cooling fan is always spinning, that's a kernel and/or BIOS issue.

What leads you to believe that turbocache is not working?

The Coolbits option is not supported on mobile (Go) GPUs, and only on NV3x and higher GPUs.

Thanks,
Lonni

Lonni,

I presume that the turbocache is not activated because every information i have (nvidia-settings utility) does not mention it anywhere. In fact the card is recognized as a standard 6200go with 64MB.
But I could be wrong...

I know that the system fans are controlled by the BIOS but if the video card overheats (for some reason unknown to me) I espect the whole system (it's a laptop and very thin as well) to get hot and the fans to kick in... and (un)fortunately I don't have Windows installed to check what is the "normal" behaviour of the laptop.

... so that's basicly why i started checking the logs of the system and found those error messages in /var/log/kern.log

But, if u say that everything is ok (after upgrading the kernel)...

Cheers
c.

[netllama]

cellstije,
Please generate and post an nvidia-bug-report.log and I can determine whether your GPU is being correctly detected.

Thanks,
Lonni

[cellstije]

here you go!

P.S.
I updated the kernel and the messeges disappeared!

Cheers!
c.

[netllama]

cellstije,
Can you post the output from "lspci -n"? Your kernel doesn't have a /proc/pci, so the bug report didn't contain the info I needed.

Thanks,
Lonni

[cellstije]

Originally Posted by netllama

cellstije,
Can you post the output from "lspci -n"? Your kernel doesn't have a /proc/pci, so the bug report didn't contain the info I needed.

Thanks,
Lonni

Lonni,

this is the output of 'lspci -n'

0000:00:00.0 0600: 8086:2590 (rev 03)
0000:00:01.0 0604: 8086:2591 (rev 03)
0000:00:1b.0 0403: 8086:2668 (rev 04)
0000:00:1d.0 0c03: 8086:2658 (rev 04)
0000:00:1d.1 0c03: 8086:2659 (rev 04)
0000:00:1d.2 0c03: 8086:265a (rev 04)
0000:00:1d.3 0c03: 8086:265b (rev 04)
0000:00:1d.7 0c03: 8086:265c (rev 04)
0000:00:1e.0 0604: 8086:2448 (rev d4)
0000:00:1f.0 0601: 8086:2641 (rev 04)
0000:00:1f.1 0101: 8086:266f (rev 04)
0000:00:1f.3 0c05: 8086:266a (rev 04)
0000:01:00.0 0300: 10de:0167 (rev a1)
0000:06:03.0 0607: 104c:ac8e
0000:06:03.2 0c00: 104c:802e
0000:06:03.3 0180: 104c:ac8f
0000:06:04.0 0280: 8086:4220 (rev 05)
0000:06:08.0 0200: 8086:1068 (rev 04)

However i guess i have to revert back to the previous (and live for a while with the nvrm error messages) 'couse the new one (2.6.11-1-686) gives me a kernel panic everytime i turn off the pc ...

Tha'
c.

[netllama]

cellstije,
Thanks for that output. It looks like there may be a bug here. In your bug report I see:
MemTotal: 515664 kB

Should your laptop have a total of 512MB of RAM?

Thanks,
Lonni

[cellstije]

Originally Posted by netllama

cellstije,
Thanks for that output. It looks like there may be a bug here. In your bug report I see:
MemTotal: 515664 kB

Should your laptop have a total of 512MB of RAM?

Thanks,
Lonni

Yep, the laptop has 512MB of ram - it is a sony vaio VGN-FS115M:

512MB Ram
80GB HD
nvidia 6200go turbocache (128MB)

... i guess it is the standard configuration for such a laptop...

I reverted to the old kernel because the new one was "panicking" ()
So attached you find the bug report obtained with the "standard" ubuntu kernel (2.6.10-5-686) ... just in case...

btw.
I just corrected 3 errors in the acpi DSDT of the laptop and added it to the the initrd.img file ... i don't know if this is going to add confusion to the nvidia bug report .... let me know...

cheers
c.