Re: ACPI error messages

Re: ACPI error messages

[Zhang Rui]

On Wed, 2010-06-02 at 09:32 +0800, Alan Stern wrote:
> On Wed, 2 Jun 2010, Zhang Rui wrote:
> 
> > On Tue, 2010-06-01 at 21:11 +0800, Alan Stern wrote:
> > > My kernel log is full of hundreds of messages like these:
> > >
> > > ACPI Exception: AE_AML_BUFFER_LIMIT, Index (000000046) is beyond end of object (20090903/exoparg2-445)
> > > ACPI Error (psparse-0537): Method parse/execution failed [\_SB_.PCI0.SBRG.EC0_.STBR] (Node f701c918), AE_AML_BUFFER_LIMIT
> > > ACPI Error (psparse-0537): Method parse/execution failed [\_SB_.PCI0.VGA_.LCDD._BCM] (Node f7017b58), AE_AML_BUFFER_LIMIT
> > > ACPI Error: Evaluating _BCM failed (20090903/video-528)
> > >
> > This sounds like a win7 compatibility issue that I've seen before.
> > please try boot option acpi_osi="!Windows 2009" and see if it helps.
> 
> Yes, indeed it does!  It results in a few other differences too; I
> can't tell whether they are good, bad, or neutral.
> 
well, I don't see any other difference except some code execution
ordering issue.

I read the acpidump of this laptop, and it also confirms that this (_BCM
stuff) is the only place that acpi_osi="!Windows 2009" changes.

thanks,
rui

> > BTW, it would be great if you can attach the acpidump of this laptop and
> > the dmesg output after boot.
> 
> Attached (the dmesg log is from a boot without the acpi_osi option).
> Below is a diff between that boot and one with acpi_osi="!Windows
> 2009".  (I believe all those "name_count maxed" lines come from
> mounting an NTFS filesystem; you can ignore them.)
> 
> Alan Stern
> 
> 
> --- dmesg1.txt  2010-06-01 21:23:34.569204151 -0400
> +++ dmesg2.txt  2010-06-01 21:18:21.329732003 -0400

Re: ACPI error messages

[Alan Stern]

On Wed, 2 Jun 2010, Zhang Rui wrote:

> On Wed, 2010-06-02 at 09:32 +0800, Alan Stern wrote:
> > On Wed, 2 Jun 2010, Zhang Rui wrote:
> > 
> > > On Tue, 2010-06-01 at 21:11 +0800, Alan Stern wrote:
> > > > My kernel log is full of hundreds of messages like these:
> > > >
> > > > ACPI Exception: AE_AML_BUFFER_LIMIT, Index (000000046) is beyond end of object (20090903/exoparg2-445)
> > > > ACPI Error (psparse-0537): Method parse/execution failed [\_SB_.PCI0.SBRG.EC0_.STBR] (Node f701c918), AE_AML_BUFFER_LIMIT
> > > > ACPI Error (psparse-0537): Method parse/execution failed [\_SB_.PCI0.VGA_.LCDD._BCM] (Node f7017b58), AE_AML_BUFFER_LIMIT
> > > > ACPI Error: Evaluating _BCM failed (20090903/video-528)
> > > >
> > > This sounds like a win7 compatibility issue that I've seen before.
> > > please try boot option acpi_osi="!Windows 2009" and see if it helps.
> > 
> > Yes, indeed it does!  It results in a few other differences too; I
> > can't tell whether they are good, bad, or neutral.
> > 
> well, I don't see any other difference except some code execution
> ordering issue.

You are right.  Well, there is one difference: the lines about 
"cfg80211: World regulatory domain updated:".  I'm not concerned about 
that.

> I read the acpidump of this laptop, and it also confirms that this (_BCM
> stuff) is the only place that acpi_osi="!Windows 2009" changes.

All right; I'll keep that boot config option.  Is there a plan to fix 
this compatibility issue?

Alan Stern

Re: ACPI error messages

[Zhang Rui]

On Wed, 2010-06-02 at 23:10 +0800, Alan Stern wrote:
> On Wed, 2 Jun 2010, Zhang Rui wrote:
> 
> > On Wed, 2010-06-02 at 09:32 +0800, Alan Stern wrote:
> > > On Wed, 2 Jun 2010, Zhang Rui wrote:
> > > 
> > > > On Tue, 2010-06-01 at 21:11 +0800, Alan Stern wrote:
> > > > > My kernel log is full of hundreds of messages like these:
> > > > >
> > > > > ACPI Exception: AE_AML_BUFFER_LIMIT, Index (000000046) is beyond end of object (20090903/exoparg2-445)
> > > > > ACPI Error (psparse-0537): Method parse/execution failed [\_SB_.PCI0.SBRG.EC0_.STBR] (Node f701c918), AE_AML_BUFFER_LIMIT
> > > > > ACPI Error (psparse-0537): Method parse/execution failed [\_SB_.PCI0.VGA_.LCDD._BCM] (Node f7017b58), AE_AML_BUFFER_LIMIT
> > > > > ACPI Error: Evaluating _BCM failed (20090903/video-528)
> > > > >
> > > > This sounds like a win7 compatibility issue that I've seen before.
> > > > please try boot option acpi_osi="!Windows 2009" and see if it helps.
> > > 
> > > Yes, indeed it does!  It results in a few other differences too; I
> > > can't tell whether they are good, bad, or neutral.
> > > 
> > well, I don't see any other difference except some code execution
> > ordering issue.
> 
> You are right.  Well, there is one difference: the lines about 
> "cfg80211: World regulatory domain updated:".  I'm not concerned about 
> that.
> 
> > I read the acpidump of this laptop, and it also confirms that this (_BCM
> > stuff) is the only place that acpi_osi="!Windows 2009" changes.
> 
> All right; I'll keep that boot config option.  Is there a plan to fix 
> this compatibility issue?
> 
This is a BIOS bug to me. The code runs in this way:

if (not Win7)
  good BIOS code
else
  buggy BIOS code

This is a regression because ACPI declares it's win7 compatible in
2.6.31. I'm afraid we can do nothing in Linux kernel to fix/workaround
it. IMO, we should ping Asus to fix it in their new BIOS release.

thanks,
rui






 We can not fix it in kernel.

> Alan Stern
> 

Re: ACPI error messages

[Alan Stern]

On Thu, 3 Jun 2010, Zhang Rui wrote:

> This is a BIOS bug to me. The code runs in this way:
> 
> if (not Win7)
>   good BIOS code
> else
>   buggy BIOS code
> 
> This is a regression because ACPI declares it's win7 compatible in
> 2.6.31. I'm afraid we can do nothing in Linux kernel to fix/workaround
> it. IMO, we should ping Asus to fix it in their new BIOS release.

Okay.  Thanks for the help.

Alan Stern

Re: ACPI error messages

[Matthew Garrett]

On Thu, Jun 03, 2010 at 09:10:06AM +0800, Zhang Rui wrote:

> This is a BIOS bug to me. The code runs in this way:
> 
> if (not Win7)
>   good BIOS code
> else
>   buggy BIOS code
> 
> This is a regression because ACPI declares it's win7 compatible in
> 2.6.31. I'm afraid we can do nothing in Linux kernel to fix/workaround
> it. IMO, we should ping Asus to fix it in their new BIOS release.

So if you install Windows 7 on these machines, backlight control doesn't 
work?

-- 
Matthew Garrett | mjg59@srcf.ucam.org

Re: ACPI error messages

[Zhang Rui]

On Thu, 2010-06-03 at 22:16 +0800, Matthew Garrett wrote:
> On Thu, Jun 03, 2010 at 09:10:06AM +0800, Zhang Rui wrote:
> 
> > This is a BIOS bug to me. The code runs in this way:
> > 
> > if (not Win7)
> >   good BIOS code
> > else
> >   buggy BIOS code
> > 
> > This is a regression because ACPI declares it's win7 compatible in
> > 2.6.31. I'm afraid we can do nothing in Linux kernel to fix/workaround
> > it. IMO, we should ping Asus to fix it in their new BIOS release.
> 
> So if you install Windows 7 on these machines, backlight control doesn't 
> work?
> 
I've heard that win7 prefers the ACPI wmi driver instead of ACPI video
driver for backlight control. But not sure if win7 prefers the platform
driver as well.

Alan,
can you help me verify if the backlight control works well when you boot
w/ acpi_backlight=vendor and w/o acpi_osi="!Windwos 2009"?

thanks,
rui

Re: ACPI error messages

[Alan Stern]

On Fri, 4 Jun 2010, Zhang Rui wrote:

> I've heard that win7 prefers the ACPI wmi driver instead of ACPI video
> driver for backlight control. But not sure if win7 prefers the platform
> driver as well.
> 
> Alan,
> can you help me verify if the backlight control works well when you boot
> w/ acpi_backlight=vendor and w/o acpi_osi="!Windwos 2009"?

I don't have the laptop with me here, so this will have to wait a 
little.

Also, I've never used the backlight control and I don't know how it 
works.  What should I do in order to test it?

Alan Stern

Re: ACPI error messages

[Zhang Rui]

On Fri, 2010-06-04 at 22:52 +0800, Alan Stern wrote:
> On Fri, 4 Jun 2010, Zhang Rui wrote:
> 
> > I've heard that win7 prefers the ACPI wmi driver instead of ACPI video
> > driver for backlight control. But not sure if win7 prefers the platform
> > driver as well.
> > 
> > Alan,
> > can you help me verify if the backlight control works well when you boot
> > w/ acpi_backlight=vendor and w/o acpi_osi="!Windwos 2009"?
> 
> I don't have the laptop with me here, so this will have to wait a 
> little.
> 
okay.

> Also, I've never used the backlight control and I don't know how it 
> works.  What should I do in order to test it?
> 
You can change the backlight by poking /sys/class/backlight/*/brightness

thanks,
rui

Re: ACPI error messages

[Alan Stern]

On Mon, 7 Jun 2010, Zhang Rui wrote:

> On Fri, 2010-06-04 at 22:52 +0800, Alan Stern wrote:
> > On Fri, 4 Jun 2010, Zhang Rui wrote:
> > 
> > > I've heard that win7 prefers the ACPI wmi driver instead of ACPI video
> > > driver for backlight control. But not sure if win7 prefers the platform
> > > driver as well.
> > > 
> > > Alan,
> > > can you help me verify if the backlight control works well when you boot
> > > w/ acpi_backlight=vendor and w/o acpi_osi="!Windwos 2009"?
> > 
> > I don't have the laptop with me here, so this will have to wait a 
> > little.
> > 
> okay.
> 
> > Also, I've never used the backlight control and I don't know how it 
> > works.  What should I do in order to test it?
> > 
> You can change the backlight by poking /sys/class/backlight/*/brightness

With acpi_backlight=vendor instead of acpi_osi="!Windows 2009", the 
backlight control still works.  And those error messages do not appear 
in the system log.

Which option should I use?

Alan Stern

ACPI error messages

[Alan Stern]

My kernel log is full of hundreds of messages like these:

ACPI Exception: AE_AML_BUFFER_LIMIT, Index (000000046) is beyond end of object (20090903/exoparg2-445)
ACPI Error (psparse-0537): Method parse/execution failed [\_SB_.PCI0.SBRG.EC0_.STBR] (Node f701c918), AE_AML_BUFFER_LIMIT
ACPI Error (psparse-0537): Method parse/execution failed [\_SB_.PCI0.VGA_.LCDD._BCM] (Node f7017b58), AE_AML_BUFFER_LIMIT
ACPI Error: Evaluating _BCM failed (20090903/video-528)

This is on an ASUS UL20A laptop (ICH9 chipset) with Fedora-12 (kernel
2.6.32.12).  What is the reason for these error messages?  Can anything 
be done to fix them?

Thanks,

Alan Stern

Re: ACPI error messages

[Zhang Rui]

On Tue, 2010-06-01 at 21:11 +0800, Alan Stern wrote:
> My kernel log is full of hundreds of messages like these:
> 
> ACPI Exception: AE_AML_BUFFER_LIMIT, Index (000000046) is beyond end of object (20090903/exoparg2-445)
> ACPI Error (psparse-0537): Method parse/execution failed [\_SB_.PCI0.SBRG.EC0_.STBR] (Node f701c918), AE_AML_BUFFER_LIMIT
> ACPI Error (psparse-0537): Method parse/execution failed [\_SB_.PCI0.VGA_.LCDD._BCM] (Node f7017b58), AE_AML_BUFFER_LIMIT
> ACPI Error: Evaluating _BCM failed (20090903/video-528)
> 
This sounds like a win7 compatibility issue that I've seen before.
please try boot option acpi_osi="!Windows 2009" and see if it helps.

BTW, it would be great if you can attach the acpidump of this laptop and
the dmesg output after boot.

thanks,
rui

> This is on an ASUS UL20A laptop (ICH9 chipset) with Fedora-12 (kernel
> 2.6.32.12).  What is the reason for these error messages?  Can anything 
> be done to fix them?
> 
> Thanks,
> 
> Alan Stern
> 
> --
> To unsubscribe from this list: send the line "unsubscribe linux-acpi" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html