gmane.linux.swsusp.devel

Re: current state of nvidia quadro 140m and tuxonice

On Thu, 20 Nov 2008, Nigel Cunningham wrote:

> I've spent a lot of time comparing my code to swsusp code,
> and haven't yet spotted anything that I think should make a difference.

There's a claim on NVNews.net that says it might be related to GPU shared
memory- are we perhaps doing any optimizations in that region somehow?

	-Kenny

--

-- 
Kenneth R. Crudup  Sr. SW Engineer, Scott County Consulting, Los Angeles
O: 3630 S. Sepulveda Blvd. #138, L.A., CA 90034-6809      (888) 454-8181

headers

Nigel Cunningham | 19 Nov 22:41

Re: current state of nvidia quadro 140m and tuxonice

Hi.

On Wed, 2008-11-19 at 12:53 -0800, Kenneth Crudup wrote:
> On Thu, 20 Nov 2008, Nigel Cunningham wrote:
> 
> > I've spent a lot of time comparing my code to swsusp code,
> > and haven't yet spotted anything that I think should make a difference.
> 
> There's a claim on NVNews.net that says it might be related to GPU shared
> memory- are we perhaps doing any optimizations in that region somehow?

TuxOnIce wouldn't know GPU shared memory from ordinary memory. We use
exactly the same data swsusp uses to determine what shouldn't be saved
and the same calls to drivers, afaics. I think it's probably an issue in
TuxOnIce too. I just can't see for the life of me what it is - yet. I
have noticed issues with aborting after doing an atomic copy (as well as
S3 powerdown method). I'm hoping that following that thread will help.

Unfortunately at the moment I just don't have much time to work on all
this.

Nigel

headers

Csaba Keszei | 21 Nov 18:47

Re: current state of nvidia quadro 140m and tuxonice

Hi,

I have migrated to a new laptop with nvidia quadro board:

01:00.0 VGA compatible controller: nVidia Corporation Quadro FX 570M 
(rev a1)

Unfortunately recent tuxonice does not resume. Switching to X or 
starting a new X server locks the machine. This is a soft lockup since 
powr button or crtl-alt-del starts the reboot process. I think it can 
successfully unmount the filesystems but halt/reboot fails. I have 
collected some logs, I hope this helps:

 NVIDIA-Linux-x86-177.82:

success:
Nov 20 15:30:16 kecsalap2 kernel: TuxOnIce debugging info:
Nov 20 15:30:16 kecsalap2 kernel: - TuxOnIce core  : 3.0-rc7
Nov 20 15:30:16 kecsalap2 kernel: - Kernel Version : 2.6.27.6
Nov 20 15:30:16 kecsalap2 kernel: - Compiler vers. : 4.3
Nov 20 15:30:16 kecsalap2 kernel: - Attempt number : 1
Nov 20 15:30:16 kecsalap2 kernel: - Parameters     : 0 1335296 0 0 0 5
Nov 20 15:30:16 kecsalap2 kernel: - Overall expected compression 
percentage: 0.
Nov 20 15:30:16 kecsalap2 kernel: - Checksum method is 'md4'.
Nov 20 15:30:16 kecsalap2 kernel:  0 pages resaved in atomic copy.
Nov 20 15:30:16 kecsalap2 kernel: - Compressor is 'lzf'.
Nov 20 15:30:16 kecsalap2 kernel:  Compressed 11972608 bytes into 
5609668 (53 percent compression).
Nov 20 15:30:16 kecsalap2 kernel: - Max outstanding reads 1055. Max

(Continue reading)

headers

bobzhang | 20 Nov 10:04

in redhat-9.0 , when I execute "hibernate"command , errors


My environment is Redhat9.0 , kernel is 2.4.21 . 

when I executed "hibernate" command , I see :

"hibernate: Suspend was not completed because:
 - Suspend was aborted by user."

 
Following is my detailed environment. 

1, patch kernel :

I downloaded two patch , 

software-suspend-linux-2.4.21-rev2-whole
software-suspend-core-2.0-whole

no any errors when applying patch and compiling kernel

2, hibernate script 

I used the oldest version , hibernate-script-0.95.1.tar.gz  

 
3, my grub.conf is like this:
# grub.conf generated by anaconda
#
# Note that you do not have to rerun grub after making changes to this file
# NOTICE:  You have a /boot partition.  This means that

(Continue reading)