Skip to contentSkip to conversation

Resuming from hibernation on T61 with nvidia

DISCUSSION dienet 4 participants 10 messages Nov 15, 2011 – Nov 16, 2011 ACPI

lore
Share & tools
lore

dienet

author
#0 lore raw

Hello

As far I can remember (since I have my ThinkPad T61), there is problem with resuming from hibernation (suspend-to-disk) on this system. And I know that there are lots of people with the same problem. I was checking lots of kernel versions but with no luck.

Was is the problem? In my T61 I have 01:00.0 VGA compatible controller: nVidia Corporation G84M [Quadro NVS 140M] (rev a1) and I am using nvidia drivers (also checked quite a few versions without any luck). Suspending to disk is working without any problem. The problem is when I want to resume. The resume process is stopping at some point (I think after the pages are red) for about 60 seconds. After that time, the resume process is continuing, and after few sec, I am back in X11.

I have motivation to fix it, but:

1. How to trace it?
2. Is it possible to fix this? Maybe the fix should take place in nvidia driver (nvidia.o), and that can not be done. But maybe the fix can be done in kernel?

I want to notice that, if the fix will only work for me, then that is fine with me.

I know that answer to 2. might be hard, so maybe some guide how to trace the problem?

Signature · 8 lines
pozdr0
dienet

"Old C programmers never die. They're just cast into void."

----------------------------------------------------------------
Nie zmieniaj opon, zmien auto!
http://linkint.pl/f2a7c
#1 lore raw
Quoted 22 lines · dienet

Hello

As far I can remember (since I have my ThinkPad T61), there is problem with resuming from hibernation (suspend-to-disk) on this system. And I know that there are lots of people with the same problem.
I was checking lots of kernel versions but with no luck.

Was is the problem? In my T61 I have 01:00.0 VGA compatible controller:
nVidia Corporation G84M [Quadro NVS 140M] (rev a1) and I am using nvidia drivers (also checked quite a few versions without any luck). Suspending to disk is working without any problem. The problem is when I want to resume. The resume process is stopping at some point (I think after the pages are red) for about 60 seconds. After that time, the resume process is continuing, and after few sec, I am back in X11.

I have motivation to fix it, but:

1. How to trace it?
2. Is it possible to fix this? Maybe the fix should take place in nvidia driver (nvidia.o), and that can not be done. But maybe the fix can be done in kernel?

Quoted 6 lines

I want to notice that, if the fix will only work for me, then that is fine with me.

I know that answer to 2. might be hard, so maybe some guide how to trace the problem?

Hi,

[Adding linux-pm and linux-kernel mailing lists to CC.]

You can try using the pm_test framework for debugging this, to a certain extent. See Documentation/power/basic-pm-debugging.txt for details on how to use it.

Also, if you could post any logs that indicate the problem, that would be great!

Thanks,
Srivatsa S. Bhat

author
#2 lore raw
Quoted 3 lines · Srivatsa S. Bhat

You can try using the pm_test framework for debugging this, to a certain extent. See Documentation/power/basic-pm-debugging.txt for details on how to use it.

I will look into that.

Quoted 2 lines

Also, if you could post any logs that indicate the problem, that would be great!

At this point "the logs" are camera shoots of my monitor. I really do not know if there is some think else that can be useful.

Signature · 8 lines
pozdr0
dienet

"Old C programmers never die. They're just cast into void."

----------------------------------------------------------------
Andrzejki tuz tuz ...
Zapytaj wrozke >> http://linkint.pl/f2a8e
author
#3 lore raw
In reply to Srivatsa S. Bhat
Quoted 3 lines · Srivatsa S. Bhat

You can try using the pm_test framework for debugging this, to a certain extent. See Documentation/power/basic-pm-debugging.txt for details on how to use it.

So I run some test:

Using platform in state file:

1. echo freezer > pm_test - no problems.
2. echo devices > pm_test - problems: no picture (LCD is up and shining, but there is only black picture [even after 60 seconds]). I can use power button to shut down the system. Other keyboard keys are not working (I was trying to type reboot with no luck).
3. echo processors > pm_test - the same as above
4. echo core > pm_test - the same as above
5. I turn off the X11 (telinit 2), but the nvidia module is still loaded - no problems.
6. I left X11 working on f7 (7th tty) and switch to f1 (1st tty) - echo core > pm_test works OK, but switching back to X11 (alt+f7) will give me the same as in 2. I can change the brightness of black screen, but I can not go back to tty1 (ctrl+alt+f1).

Is there more I can do?

Signature · 8 lines
pozdr0
dienet

"Old C programmers never die. They're just cast into void."

----------------------------------------------------------------
Mamy mieszkanie wlasnie dla Ciebie! Sprawdz >>
http://linkint.pl/f2a80
#4 lore raw
In reply to dienet
Quoted 8 lines · dienet

Was is the problem? In my T61 I have 01:00.0 VGA compatible
controller: nVidia Corporation G84M [Quadro NVS 140M] (rev a1) and I am using nvidia drivers (also checked quite a few versions without any luck). Suspending to disk is working without any problem. The problem is when I want to resume. The resume process is stopping at some point (I think after the pages are red) for about 60 seconds. After that time, the resume process is continuing, and after few sec, I am back in X11.

Does the same thing happen if you use nouveau rather than the nvidia drivers?

Signature · 1 line
Matthew Garrett | [email protected]
author
#5 lore raw
Quoted 2 lines · Matthew Garrett

Does the same thing happen if you use nouveau rather than the nvidia drivers?

Can not answer that, but this is not happening when using nv X11 driver.

Signature · 8 lines
pozdr0
dienet

"Old C programmers never die. They're just cast into void."

-------------------------------------
Sprawdz za darmo zdolnosc kredytowa!	
http://linkint.pl/f2a77
#6 lore raw
Quoted 7 lines · dienet
Quoted 2 lines · Matthew Garrett

Does the same thing happen if you use nouveau rather than the nvidia drivers?

Can not answer that, but this is not happening when using nv X11 driver.

In that case it's almost certainly something you need to bring up with nvidia.

Signature · 1 line
Matthew Garrett | [email protected]
author
#7 lore raw
Quoted 2 lines · Matthew Garrett

In that case it's almost certainly something you need to bring up with nvidia.

Yes, but that might be hard, or even impossible. Maybe there is something that I can change in kernel to fix it, without breaking too much or not breaking anything ;)

Signature · 8 lines
pozdr0
dienet

"Old C programmers never die. They're just cast into void."

----------------------------------------------------------------
Mamy mieszkanie wlasnie dla Ciebie! Sprawdz >>
http://linkint.pl/f2a80
#8 lore raw

Don't forget that you can use nvidia-reportbug.sh...can't remember exactly what its called but
I know that nvidia does actually look at those generated bug reports.

-Josh

Quoted 16 lines · dienet
Quoted 2 lines · Matthew Garrett

In that case it's almost certainly something you need to bring up with nvidia.

Yes, but that might be hard, or even impossible. Maybe there is something that I can change in kernel to fix it, without breaking too much or not breaking anything ;)

--
pozdr0
dienet

"Old C programmers never die. They're just cast into void."

Quoted 9 lines

----------------------------------------------------------------
Mamy mieszkanie wlasnie dla Ciebie! Sprawdz >>
http://linkint.pl/f2a80
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected]
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

--
To unsubscribe from this list: send the line "unsubscribe linux-acpi" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

author
#9 lore raw
Quoted 3 lines · Joshua Scoggins

Don't forget that you can use nvidia-reportbug.sh...can't remember exactly what its called but
I know that nvidia does actually look at those generated bug reports.

OK ;) Got the motivation, I did send them the report, will see...

But I am still gonna check linux kernel trace methods, to know more.

Signature · 8 lines
pozdr0
dienet

"Old C programmers never die. They're just cast into void."

----------------------------------------------------------------
Jesienne, zimowe wieczory? Zagraj!
Sprawdz >>>  http://linkint.pl/f2a85