HP Compaq nc6320 Suspend / Resume (changes in behaviour)

HP Compaq nc6320 Suspend / Resume (changes in behaviour)

[Peter Clifton]

Hi,

I just updated my bios to HP's latest, and have recently been upgraded
to Ubuntu Gutsy kernel version:

Linux version 2.6.22-1-generic (buildd@palmer) (gcc version 4.1.3
20070423 (prerelease) (Ubuntu 4.1.2-3ubuntu3)) #1 SMP Mon Apr 30
10:49:22 GMT 2007

For the first time ever, suspend to ram works!

Unfortunately, hibernation has broken. This happened somewhere between
Feisty's 2.6.20 kernel and the one I'm using now. (Although it could
have been an update released in Feisty which broke things).

I can hibernate, but after a short while being booted back up (and
perhaps accessing some ACPI related thing, e.g. cpu scaling, unplugging
AC etc.. it will crash (and due to being in X11 at the time, I've got no
useful debug output to go on).

Does anyone know specifically if something changed to fix suspend, (with
SATA drives no less!), or anything which may have broken hibernate.
(Which was broken before I updated from bios F.06 to the latest today)

After resume from suspend, I see:

dmesg  | grep ACPI | grep Error
[ 4763.428000] ACPI Error (psparse-0537): Method parse/execution failed [\_SB_.C241] (Node df94c9dc), AE_AML_PACKAGE_LIMIT
[ 4763.428000] ACPI Error (psparse-0537): Method parse/execution failed [\_SB_.C002.C0DC.C349._STM] (Node df94fba8), AE_AML_PACKAGE_LIMIT
[ 6417.528000] ACPI Error (psparse-0537): Method parse/execution failed [\_SB_.C241] (Node df94c9dc), AE_AML_PACKAGE_LIMIT
[ 6417.528000] ACPI Error (psparse-0537): Method parse/execution failed [\_SB_.C002.C0DC.C349._STM] (Node df94fba8), AE_AML_PACKAGE_LIMIT

Is this anything to worry about - should I post a dsdt for the new bios?

Kind regards,

-- 
Peter Clifton

Electrical Engineering Division,
Engineering Department,
University of Cambridge,
9, JJ Thomson Avenue,
Cambridge
CB3 0FA

Tel: +44 (0)7729 980173 - (No signal in the lab!)

Re: HP Compaq nc6320 Suspend / Resume (changes in behaviour)

[Rafael J. Wysocki]

Hi,

On Tuesday, 22 May 2007 20:49, Peter Clifton wrote:
> Hi,
> 
> I just updated my bios to HP's latest, and have recently been upgraded
> to Ubuntu Gutsy kernel version:
> 
> Linux version 2.6.22-1-generic (buildd@palmer) (gcc version 4.1.3
> 20070423 (prerelease) (Ubuntu 4.1.2-3ubuntu3)) #1 SMP Mon Apr 30
> 10:49:22 GMT 2007
> 
> For the first time ever, suspend to ram works!

That's a good news.

> Unfortunately, hibernation has broken. This happened somewhere between
> Feisty's 2.6.20 kernel and the one I'm using now. (Although it could
> have been an update released in Feisty which broke things).
> 
> I can hibernate, but after a short while being booted back up (and
> perhaps accessing some ACPI related thing, e.g. cpu scaling, unplugging
> AC etc.. it will crash (and due to being in X11 at the time, I've got no
> useful debug output to go on).
> 
> Does anyone know specifically if something changed to fix suspend, (with
> SATA drives no less!), or anything which may have broken hibernate.
> (Which was broken before I updated from bios F.06 to the latest today)

Well, threre's a couple of such things.  You may try the kernel.org kernel
2.6.22-rc2 with the hibernation and suspend patchset available from
http://www.sisk.pl/kernel/hibernation_and_suspend/2.6.22-rc2/patches .

> After resume from suspend, I see:
> 
> dmesg  | grep ACPI | grep Error
> [ 4763.428000] ACPI Error (psparse-0537): Method parse/execution failed [\_SB_.C241] (Node df94c9dc), AE_AML_PACKAGE_LIMIT
> [ 4763.428000] ACPI Error (psparse-0537): Method parse/execution failed [\_SB_.C002.C0DC.C349._STM] (Node df94fba8), AE_AML_PACKAGE_LIMIT
> [ 6417.528000] ACPI Error (psparse-0537): Method parse/execution failed [\_SB_.C241] (Node df94c9dc), AE_AML_PACKAGE_LIMIT
> [ 6417.528000] ACPI Error (psparse-0537): Method parse/execution failed [\_SB_.C002.C0DC.C349._STM] (Node df94fba8), AE_AML_PACKAGE_LIMIT
> 
> Is this anything to worry about - should I post a dsdt for the new bios?

This means we are unable to set the IDE controller transfer timings after the
resume.  I don't know how serious this is in practice, though.

Greetings,
Rafael

Re: HP Compaq nc6320 Suspend / Resume (changes in behaviour)

[Rafael J. Wysocki]

On Tuesday, 22 May 2007 22:57, Peter Clifton wrote:
> On Tue, 2007-05-22 at 21:27 +0200, Rafael J. Wysocki wrote:
> > Hi,
> > 
> > On Tuesday, 22 May 2007 20:49, Peter Clifton wrote:
> > > Hi,
> > > 
> > > I just updated my bios to HP's latest, and have recently been
> upgraded
> > > to Ubuntu Gutsy kernel version:
> > > 
> > > Linux version 2.6.22-1-generic (buildd@palmer) (gcc version 4.1.3
> > > 20070423 (prerelease) (Ubuntu 4.1.2-3ubuntu3)) #1 SMP Mon Apr 30
> > > 10:49:22 GMT 2007
> 
> [snip]
> 
> > Well, threre's a couple of such things.  You may try the kernel.org
> kernel
> > 2.6.22-rc2 with the hibernation and suspend patchset available from
> > http://www.sisk.pl/kernel/hibernation_and_suspend/2.6.22-rc2/patches .
> 
> I wonder if those are included in Ubuntu's kernel (I'm being dumb today,
> but I couldn't work out if the debian build was adding patches or not,
> nor which exact version they started from.
> 
> (The source filename is linux-source-2.6.22_2.6.22-5.11.tar.gz)
> 
> Just FYI, the link you posted has a file with funny permissions on it:
> 
> http://www.sisk.pl/kernel/hibernation_and_suspend/2.6.22-rc2/patches/16-usb-use-freezable-workqueue.patch
> 
> Gives me a 403 error.

Fixed now, sorry for the inconvenience.

Greetings,
Rafael

Re: HP Compaq nc6320 Suspend / Resume (changes in behaviour)

[Peter Clifton]

[-- Attachment #1: Type: text/plain, Size: 3045 bytes --]

On Tue, 2007-05-22 at 21:27 +0200, Rafael J. Wysocki wrote:
> Hi,
> 
> On Tuesday, 22 May 2007 20:49, Peter Clifton wrote:
> > Hi,
> > 
> > I just updated my bios to HP's latest, and have recently been
upgraded
> > to Ubuntu Gutsy kernel version:
> > 
> > Linux version 2.6.22-1-generic (buildd@palmer) (gcc version 4.1.3
> > 20070423 (prerelease) (Ubuntu 4.1.2-3ubuntu3)) #1 SMP Mon Apr 30
> > 10:49:22 GMT 2007

[snip]

> Well, threre's a couple of such things.  You may try the kernel.org
kernel
> 2.6.22-rc2 with the hibernation and suspend patchset available from
> http://www.sisk.pl/kernel/hibernation_and_suspend/2.6.22-rc2/patches .

I wonder if those are included in Ubuntu's kernel (I'm being dumb today,
but I couldn't work out if the debian build was adding patches or not,
nor which exact version they started from.

(The source filename is linux-source-2.6.22_2.6.22-5.11.tar.gz)

Just FYI, the link you posted has a file with funny permissions on it:

http://www.sisk.pl/kernel/hibernation_and_suspend/2.6.22-rc2/patches/16-usb-use-freezable-workqueue.patch

Gives me a 403 error.


> > After resume from suspend, I see:
> > 
> > dmesg  | grep ACPI | grep Error
> > [ 4763.428000] ACPI Error (psparse-0537): Method parse/execution
failed [\_SB_.C241] (Node df94c9dc), AE_AML_PACKAGE_LIMIT
> > [ 4763.428000] ACPI Error (psparse-0537): Method parse/execution
failed [\_SB_.C002.C0DC.C349._STM] (Node df94fba8), AE_AML_PACKAGE_LIMIT
> > [ 6417.528000] ACPI Error (psparse-0537): Method parse/execution
failed [\_SB_.C241] (Node df94c9dc), AE_AML_PACKAGE_LIMIT
> > [ 6417.528000] ACPI Error (psparse-0537): Method parse/execution
failed [\_SB_.C002.C0DC.C349._STM] (Node df94fba8), AE_AML_PACKAGE_LIMIT
> > 
> > Is this anything to worry about - should I post a dsdt for the new
bios?

Looking at the ACPI 3.0 spec:

_STM Optional control method that sets the IDE controller’s transfer
timing sttings. IDE-only

(IE. shouldn't be called for an SATA drive). That may mean its for my CD
writer (which shows up as hda). This is in the first SSDT, "HPQSAT"
which I've attached the dissasembly of.

> This means we are unable to set the IDE controller transfer timings
after the
> resume.  I don't know how serious this is in practice, though.

iasl decompiles and recompiles the dsdt and all ssdt ok (after minor
tweaking due to decompile discrepancies).

The remaining warnings only relate to _BCQ (A Microsoft induced typo for
_BQC?)


And, unreachable code in the _STA method of the "ACEL" device. (They
have blanked that function off with a Return (0x00). Thats a shame - if
it means what I think (that the acceleration sensor doesn't work).

Many places in the ACEL device code check If (_STA ()) around big blocks
of code, so presumably the AML code for it is not doing a great deal. (I
wounder how Windows uses it - a custom driver perhaps?)

Thanks for the great work on suspend / resume (and of course ACPI
support in general) everyone!

Peter

[-- Attachment #2: ssdt1.dsl --]
[-- Type: text/x-csrc, Size: 6261 bytes --]

/*
 * Intel ACPI Component Architecture
 * AML Disassembler version 20061109
 *
 * Disassembly of ssdt1.aml, Tue May 22 20:59:05 2007
 *
 *
 * Original Table Header:
 *     Signature        "SSDT"
 *     Length           0x00000326 (806)
 *     Revision         0x01
 *     OEM ID           "HP    "
 *     OEM Table ID     "HPQSAT"
 *     OEM Revision     0x00000001 (1)
 *     Creator ID       "MSFT"
 *     Creator Revision 0x0100000E (16777230)
 */
DefinitionBlock ("ssdt1.aml", "SSDT", 1, "HP    ", "HPQSAT", 0x00000001)
{
    External (C0EC, MethodObj)    // 6 Arguments
    External (C0E8, MethodObj)    // 5 Arguments
    External (C0E6, MethodObj)    // 2 Arguments
    External (C0E7, MethodObj)    // 1 Argument?
    External (C0E4, MethodObj)    // 2 Arguments
    External (C240)
    External (\_SB_.C14A)
    External (\_SB_.C248, MethodObj)
    External (\_SB_.C241, MethodObj)    // 4 Arguments
    External (\_SB_.C002, DeviceObj)
    External (\_SB_.C002.C0DC, DeviceObj)

    Scope (\_SB.C002)
    {
        Device (C351)
        {
            Name (_ADR, 0x001F0002)
            Device (C0F6)
            {
                Name (_ADR, 0xFFFF)
                Name (C352, Package (0x06)
                {
                    0x00, 
                    0x00, 
                    0x00, 
                    0x00, 
                    0x00, 
                    0x00
                })
                Method (_SDD, 1, Serialized)
                {
                    If (LEqual (SizeOf (Arg0), 0x0200))
                    {
                        Store (0x00, Local0)
                        Store (Buffer (Multiply (SizeOf (C240), 0x02)) {}, Local1)
                        While (LLess (Local0, SizeOf (C240)))
                        {
                            Store (DerefOf (Index (C240, Local0)), Local2)
                            Multiply (Local2, 0x02, Local2)
                            Multiply (Local0, 0x02, Local3)
                            Store (DerefOf (Index (Arg0, Local2)), Index (Local1, Local3))
                            Store (DerefOf (Index (Arg0, Add (Local2, 0x01))), Index (
                                Local1, Add (Local3, 0x01)))
                            Increment (Local0)
                        }

                        \_SB.C241 (0x00010000, 0x00, Local1, C352)
                    }
                }

                Method (_GTF, 0, NotSerialized)
                {
                    Return (\_SB.C248(C352, 0x00))
                }
            }
        }
    }

    Scope (\_SB.C002.C0DC)
    {
        Name (_ADR, 0x001F0001)
        Name (C353, 0x01)
        Device (C349)
        {
            Name (_ADR, 0x00)
            Name (C34A, Package (0x05)
            {
                0x00, 
                0x00, 
                0x00, 
                0x00, 
                0x00
            })
            Method (_GTM, 0, NotSerialized)
            {
                Store (Buffer (0x14)
                    {
                        /* 0000 */    0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 
                        /* 0008 */    0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 
                        /* 0010 */    0xFF, 0xFF, 0xFF, 0xFF
                    }, Local0)
                CreateDWordField (Local0, 0x00, C34B)
                CreateDWordField (Local0, 0x04, C34C)
                CreateDWordField (Local0, 0x08, C354)
                CreateDWordField (Local0, 0x0C, C355)
                CreateDWordField (Local0, 0x10, C34D)
                Store (C0E4 (0x00, 0x00), C34B)
                Store (C0E7 (0x00), C34D)
                If (And (C34D, 0x01))
                {
                    Store (C0E6 (0x00, 0x00), C34C)
                }
                Else
                {
                    Store (C34B, C34C)
                }

                If (LOr (LEqual (And (\_SB.C14A, 0x0F), 0x07), LEqual (And (
                    C353, 0x01), 0x00)))
                {
                    Store (0x78, C354)
                    Store (0x11, C355)
                    Or (C34D, 0x0C, C34D)
                }
                Else
                {
                    Store (C0E4 (0x00, 0x01), C354)
                    If (And (C34D, 0x04))
                    {
                        Store (C0E6 (0x00, 0x01), C355)
                    }
                    Else
                    {
                        Store (C354, C355)
                    }
                }

                Return (Local0)
            }

            Method (_STM, 3, NotSerialized)
            {
                CreateDWordField (Arg0, 0x00, C34B)
                CreateDWordField (Arg0, 0x04, C34C)
                CreateDWordField (Arg0, 0x08, C354)
                CreateDWordField (Arg0, 0x0C, C355)
                CreateDWordField (Arg0, 0x10, C34D)
                If (LEqual (And (\_SB.C14A, 0x0F), 0x01))
                {
                    Or (C353, 0x01, C353)
                    If (LEqual (SizeOf (Arg1), 0x0200))
                    {
                        Store (0x00, Local0)
                        Store (Buffer (Multiply (SizeOf (C240), 0x02)) {}, Local1)
                        While (LLess (Local0, SizeOf (C240)))
                        {
                            Store (DerefOf (Index (C240, Local0)), Local2)
                            Multiply (Local2, 0x02, Local2)
                            Multiply (Local0, 0x02, Local3)
                            Store (DerefOf (Index (Arg1, Local2)), Index (Local1, Local3))
                            Store (DerefOf (Index (Arg1, Add (Local2, 0x01))), Index (
                                Local1, Add (Local3, 0x01)))
                            Increment (Local0)
                        }

                        Store (C0E8 (0x00, 0x00, C34B, C34D, Local1), Local2)
                        Store (C0EC (0x00, 0x00, C34C, C34D, Local1, Local3), \_SB.C241 (Local2, Local3, Local1, C34A))
                    }
                }
            }

            Device (C34E)
            {
                Name (_ADR, 0x00)
                Method (_GTF, 0, NotSerialized)
                {
                    Return (\_SB.C248 (C34A, 0x00))
                }
            }
        }
    }
}

Re: HP Compaq nc6320 Suspend / Resume (changes in behaviour)

[Thomas Renninger]

On Tue, 2007-05-22 at 19:49 +0100, Peter Clifton wrote:
> Hi,
> 
> I just updated my bios to HP's latest, and have recently been upgraded
> to Ubuntu Gutsy kernel version:
> 
> Linux version 2.6.22-1-generic (buildd@palmer) (gcc version 4.1.3
> 20070423 (prerelease) (Ubuntu 4.1.2-3ubuntu3)) #1 SMP Mon Apr 30
> 10:49:22 GMT 2007
> 
> For the first time ever, suspend to ram works!
> 
> Unfortunately, hibernation has broken. This happened somewhere between
> Feisty's 2.6.20 kernel and the one I'm using now. (Although it could
> have been an update released in Feisty which broke things).
> 
> I can hibernate, but after a short while being booted back up (and
> perhaps accessing some ACPI related thing, e.g. cpu scaling, unplugging
> AC etc.. it will crash (and due to being in X11 at the time, I've got no
> useful debug output to go on).
> 
> Does anyone know specifically if something changed to fix suspend, (with
> SATA drives no less!), or anything which may have broken hibernate.
> (Which was broken before I updated from bios F.06 to the latest today)
> 
> After resume from suspend, I see:
> 
> dmesg  | grep ACPI | grep Error
> [ 4763.428000] ACPI Error (psparse-0537): Method parse/execution failed [\_SB_.C241] (Node df94c9dc), AE_AML_PACKAGE_LIMIT
> [ 4763.428000] ACPI Error (psparse-0537): Method parse/execution failed [\_SB_.C002.C0DC.C349._STM] (Node df94fba8), AE_AML_PACKAGE_LIMIT
> [ 6417.528000] ACPI Error (psparse-0537): Method parse/execution failed [\_SB_.C241] (Node df94c9dc), AE_AML_PACKAGE_LIMIT
> [ 6417.528000] ACPI Error (psparse-0537): Method parse/execution failed [\_SB_.C002.C0DC.C349._STM] (Node df94fba8), AE_AML_PACKAGE_LIMIT
> 
> Is this anything to worry about - should I post a dsdt for the new bios?

STR works (also the disk(s)?)?

This is strange, bug:
http://bugzilla.kernel.org/show_bug.cgi?id=8456
is about broken SATA disks after STR.

If you think this is related to bug #8456, you might want to add
yourself there or open a new bug if you think it's not.

You might want to add htejun@gmail.com (Tejun Heo), who did the latest
ACPI SATA changes. Best you also add acpidump output and also CC me,
pls.

   Thomas

Re: HP Compaq nc6320 Suspend / Resume (changes in behaviour)

[Peter Clifton]

On Thu, 2007-05-24 at 20:51 +0200, Thomas Renninger wrote:
>  
> > dmesg  | grep ACPI | grep Error
> > [ 4763.428000] ACPI Error (psparse-0537): Method parse/execution failed [\_SB_.C241] (Node df94c9dc), AE_AML_PACKAGE_LIMIT
> > [ 4763.428000] ACPI Error (psparse-0537): Method parse/execution failed [\_SB_.C002.C0DC.C349._STM] (Node df94fba8), AE_AML_PACKAGE_LIMIT
> > [ 6417.528000] ACPI Error (psparse-0537): Method parse/execution failed [\_SB_.C241] (Node df94c9dc), AE_AML_PACKAGE_LIMIT
> > [ 6417.528000] ACPI Error (psparse-0537): Method parse/execution failed [\_SB_.C002.C0DC.C349._STM] (Node df94fba8), AE_AML_PACKAGE_LIMIT
> > 
> > Is this anything to worry about - should I post a dsdt for the new bios?
> 
> STR works (also the disk(s)?)?

STR works, and resume - works fine afterwards.
STD works, resumes, but crashes some minutes afterwards.

> This is strange, bug:
> http://bugzilla.kernel.org/show_bug.cgi?id=8456
> is about broken SATA disks after STR.
> 
> If you think this is related to bug #8456, you might want to add
> yourself there or open a new bug if you think it's not.
>
> You might want to add htejun@gmail.com (Tejun Heo), who did the latest
> ACPI SATA changes. Best you also add acpidump output and also CC me,
> pls.

I'll open a new bug (or bugs, keeping the STD issue separate) some time
next week when I'll have a little more time to play with debugging
things.

Thanks for your reply,

Regards.

-- 
Peter Clifton

Electrical Engineering Division,
Engineering Department,
University of Cambridge,
9, JJ Thomson Avenue,
Cambridge
CB3 0FA

Tel: +44 (0)7729 980173 - (No signal in the lab!)