Bug 15485 - amd_adac error
Summary: amd_adac error
Status: RESOLVED INVALID
Alias: None
Product: Memory Management
Classification: Unclassified
Component: Other (show other bugs)
Hardware: All Linux
: P1 normal
Assignee: Andrew Morton
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2010-03-09 11:50 UTC by masoud
Modified: 2010-03-26 08:33 UTC (History)
1 user (show)

See Also:
Kernel Version: 2.6.31
Subsystem:
Regression: No
Bisected commit-id:


Attachments

Description masoud 2010-03-09 11:50:22 UTC
when I execute a software (XCrysden) that has a graphical face, I see the graphical face only for less than one second and then disappears. the operating system is SUSE 11.2 (64 bit), and Hardwares :
AMD 3 cores 2.8 GHz (64 bit)
RAM : DDR3 8 Gig
Motherboard: Asus
Graphics: on board 

the error:
 > drmRadeonCmdBuffer: -22. Kernel failed to parse or rejected command > stream. See dmesg for more info

I executed "dmesg" in the /bin for finding the probelm and in the results there is a line included an error : 

 [ 8.255075] amd64_edac: probe of 0000:00:18.2 failed with error -22
Comment 1 Borislav Petkov 2010-03-18 22:10:40 UTC
Hi,

do you have by chance something like the error message below also in dmesg?

[    4.977521] EDAC amd64: This node reports that Memory ECC is currently disabled, set F3x44[22] (0000:00:18.3).
[    4.977781] EDAC amd64: ECC disabled in the BIOS or no ECC capability, module will not load.
[    4.977782]  Either enable ECC checking or force module loading by setting 'ecc_enable_override'.

If yes, then the message says what the problem is and if not, please
send me your whole dmesg output. Also, what is the exact board model and
that of the DIMMs?

Thanks,
Boris.
Comment 2 masoud 2010-03-23 19:29:29 UTC
Dear Borislav Petkov

Thanks for your kind attention and help. I had enabled ECC erstwhile.
Main Board:  ASUS, M4A78T-E that is optimized for AM3 CPU and the graphics is on board (ATi RADEON 3300)
RAM: Corsair DDR3 XMS3-1333 PC3 10700H (667 MHz) 8 Gigs (4 x 2 GB, Dual) 
CPU: AMD 3 cores 2.8 GHz

Attached please find the dmesg file. 

Many Thanks 
Masoud Nahali
Sharif University of Technology



> Date: Thu, 18 Mar 2010 22:10:45 +0000
> From: bugzilla-daemon@bugzilla.kernel.org
> To: masoudnahali@live.com
> Subject: [Bug 15485] amd_adac error
> 
> http://bugzilla.kernel.org/show_bug.cgi?id=15485
> 
> 
> Borislav Petkov <bbpetkov@yahoo.de> changed:
> 
>            What    |Removed                     |Added
> ----------------------------------------------------------------------------
>                  CC|                            |bbpetkov@yahoo.de
> 
> 
> 
> 
> --- Comment #1 from Borislav Petkov <bbpetkov@yahoo.de>  2010-03-18 22:10:40
> ---
> Hi,
> 
> do you have by chance something like the error message below also in dmesg?
> 
> [    4.977521] EDAC amd64: This node reports that Memory ECC is currently
> disabled, set F3x44[22] (0000:00:18.3).
> [    4.977781] EDAC amd64: ECC disabled in the BIOS or no ECC capability,
> module will not load.
> [    4.977782]  Either enable ECC checking or force module loading by setting
> 'ecc_enable_override'.
> 
> If yes, then the message says what the problem is and if not, please
> send me your whole dmesg output. Also, what is the exact board model and
> that of the DIMMs?
> 
> Thanks,
> Boris.
> 
> -- 
> Configure bugmail: http://bugzilla.kernel.org/userprefs.cgi?tab=email
> ------- You are receiving this mail because: -------
> You reported the bug.
 		 	   		  
_________________________________________________________________
Hotmail: Free, trusted and rich email service.
https://signup.live.com/signup.aspx?id=60969
Comment 3 Borislav Petkov 2010-03-23 22:29:17 UTC
> Thanks for your kind attention and help. I had enabled ECC erstwhile.
> Main Board:  ASUS, M4A78T-E that is optimized for AM3 CPU and the graphics is
> on board (ATi RADEON 3300)
> RAM: Corsair DDR3 XMS3-1333 PC3 10700H (667 MHz) 8 Gigs (4 x 2 GB, Dual) 

Are you sure your DIMMs support ECC? I still cannot pinpoint them exactly since we need the part number but if it is any of those here: http://www.corsair.com/products/xms3/default.aspx, it looks like they're all non-ECC... And you need ECC modules for doing DRAM error detection...
Comment 4 masoud 2010-03-24 18:06:53 UTC
Dear  Borislav Petkov

Many thanks for your prompt answer and attention. I think that the type of the RAM is non-ECC. 
the Part number is : " CM3X2G1333C9DHX " . If the Ram is non-ECC , can I solve the problem or not?! Is the problem related to the RAM only?
I appreciate if one help me to solve this persecutor problem. 

Many Thanks
Masoud Nahali
Sharif University of Technology

  

> Date: Tue, 23 Mar 2010 22:29:22 +0000
> From: bugzilla-daemon@bugzilla.kernel.org
> To: masoudnahali@live.com
> Subject: [Bug 15485] amd_adac error
> 
> https://bugzilla.kernel.org/show_bug.cgi?id=15485
> 
> 
> 
> 
> 
> --- Comment #3 from Borislav Petkov <bbpetkov@yahoo.de>  2010-03-23 22:29:17
> ---
> > Thanks for your kind attention and help. I had enabled ECC erstwhile.
> > Main Board:  ASUS, M4A78T-E that is optimized for AM3 CPU and the graphics
> is
> > on board (ATi RADEON 3300)
> > RAM: Corsair DDR3 XMS3-1333 PC3 10700H (667 MHz) 8 Gigs (4 x 2 GB, Dual) 
> 
> Are you sure your DIMMs support ECC? I still cannot pinpoint them exactly
> since
> we need the part number but if it is any of those here:
> http://www.corsair.com/products/xms3/default.aspx, it looks like they're all
> non-ECC... And you need ECC modules for doing DRAM error detection...
> 
> -- 
> Configure bugmail: https://bugzilla.kernel.org/userprefs.cgi?tab=email
> ------- You are receiving this mail because: -------
> You reported the bug.
 		 	   		  
_________________________________________________________________
Hotmail: Trusted email with powerful SPAM protection.
https://signup.live.com/signup.aspx?id=60969
Comment 5 Anonymous Emailer 2010-03-24 19:10:56 UTC
Reply-To: bp@amd64.org

From: Masoud Nahali <masoudnahali@live.com>
Date: Wed, Mar 24, 2010 at 05:26:06PM +0000

Hi,

> Many thanks for your prompt answer and attention. I think that the
> type of the RAM is non-ECC. the Part number is : " CM3X2G1333C9DHX".
> If the Ram is non-ECC , can I solve the problem or not?!

well, strictly speaking, this is not a problem. Please read the message
carefully - it simply tells you that the driver won't load since your
DIMMs do not support ECC.

If the message annoys you, you can do two things:

1) buy DIMMs which support ECC

2) disable CONFIG_EDAC_AMD64 in your kernel config so that the driver
doesn't exist on your machine, rebuild and install your kernel.

Hope that helps.
Comment 6 masoud 2010-03-25 22:38:16 UTC
Dear kernel-bugs users

The message in dmesg does not annoy me, the main problem is occurred when I execute "Xcrysden" software !
when I write ./xcrysden I see the graphic page of Xcrysden only for 1 second and then disappears.

koa@linux-vs0r:~/XCrysden/usr/bin> ./xcrysden
Running on platform : unix
Operating system : Linux
Package ImageMagick's convert: /usr/bin/convert
Package ImageMagick's import: /usr/bin/import
Package Gifsicle: /usr/bin/gifsicle
Package Xwd: /usr/bin/xwd
Executing: /home/koa/XCrysden/usr/lib/xcrysden/bin/ftnunit
drmRadeonCmdBuffer: -22. Kernel failed to parse or rejected command stream. See dmesg for more info.


I cannot buy another DIMMs which support ECC. How can I disable CONFIG_EDAC_AMD64 and rebuild the kernel ?
 Can I use "kernelcfg" for the purpose ? will I lose my configuration of Linux and installed software's by 
installing the kernel? excuse me because of the long story and many thanks for your helps.  
    
Sincerely Yours
Masoud Nahali


> Date: Wed, 24 Mar 2010 19:11:03 +0000
> From: bugzilla-daemon@bugzilla.kernel.org
> To: masoudnahali@live.com
> Subject: [Bug 15485] amd_adac error
> 
> https://bugzilla.kernel.org/show_bug.cgi?id=15485
> 
> 
> 
> 
> 
> --- Comment #5 from Anonymous Emailer <anonymous@kernel-bugs.osdl.org> 
> 2010-03-24 19:10:56 ---
> Reply-To: bp@amd64.org
> 
> From: Masoud Nahali <masoudnahali@live.com>
> Date: Wed, Mar 24, 2010 at 05:26:06PM +0000
> 
> Hi,
> 
> > Many thanks for your prompt answer and attention. I think that the
> > type of the RAM is non-ECC. the Part number is : " CM3X2G1333C9DHX".
> > If the Ram is non-ECC , can I solve the problem or not?!
> 
> well, strictly speaking, this is not a problem. Please read the message
> carefully - it simply tells you that the driver won't load since your
> DIMMs do not support ECC.
> 
> If the message annoys you, you can do two things:
> 
> 1) buy DIMMs which support ECC
> 
> 2) disable CONFIG_EDAC_AMD64 in your kernel config so that the driver
> doesn't exist on your machine, rebuild and install your kernel.
> 
> Hope that helps.
> 
> -- 
> Configure bugmail: https://bugzilla.kernel.org/userprefs.cgi?tab=email
> ------- You are receiving this mail because: -------
> You reported the bug.
 		 	   		  
_________________________________________________________________
Your E-mail and More On-the-Go. Get Windows Live Hotmail Free.
https://signup.live.com/signup.aspx?id=60969
Comment 7 Borislav Petkov 2010-03-26 07:58:38 UTC
The actual problem was:

drmRadeonCmdBuffer: -22. Kernel failed to parse or rejected command >
stream. See dmesg for more info

which is not EDAC-related. 

Closing.
Comment 8 Anonymous Emailer 2010-03-26 08:33:29 UTC
Reply-To: bp@alien8.de

From: bugzilla-daemon@bugzilla.kernel.org
Date: Thu, Mar 25, 2010 at 10:38:21PM +0000

> --- Comment #6 from masoud <masoudnahali@live.com>  2010-03-25 22:38:16 ---
> Dear kernel-bugs users
> 
> The message in dmesg does not annoy me, the main problem is occurred when I
> execute "Xcrysden" software !
> when I write ./xcrysden I see the graphic page of Xcrysden only for 1 second
> and then disappears.
> 
> koa@linux-vs0r:~/XCrysden/usr/bin> ./xcrysden
> Running on platform : unix
> Operating system : Linux
> Package ImageMagick's convert: /usr/bin/convert
> Package ImageMagick's import: /usr/bin/import
> Package Gifsicle: /usr/bin/gifsicle
> Package Xwd: /usr/bin/xwd
> Executing: /home/koa/XCrysden/usr/lib/xcrysden/bin/ftnunit
> drmRadeonCmdBuffer: -22. Kernel failed to parse or rejected command stream.
> See
> dmesg for more info.

ok, now I get it. You tried to do this and mixed the radeon driver line
with the edac driver error from dmesg.

So,

if you, like me, would simply had googled

"drmRadeonCmdBuffer: -22. Kernel failed to parse or rejected command stream. See dmesg for more info."

for example, you would've found the following:

http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=563257#10

according to which, this bug should be fixed in kernel 2.6.34-rc1.

Closing bugzulla entry.

Note You need to log in before you can comment on or make changes to this bug.