Toshiba Satellite 1905S303 DSDT table

DISCUSSION Erico M Mendonca 5 participants 11 messages Nov 2, 2003 – Nov 5, 2003 7 messages not in the archive ACPI

Share & tools

Erico M Mendonca

author

2003-11-02 23:14 UTC #0 lore raw

I'm currently trying to debug the DSDT table of this aforementioned machine, and managed to clear all the errors, except for this one:
=================
Intel ACPI Component Architecture
ASL Optimizing Compiler / AML Disassembler version 20030918 [Sep 18 2003]
Copyright (C) 2000 - 2003 Intel Corporation
Supports ACPI Specification Revision 2.0b

s303.asl 468: If(\_OSI) {
Error 1028 - Too few arguments ^ (\_OSI requires 1)

s303.asl 469: "Windows 2001"
Error 1037 - ^ parse error, unexpected PARSEOP_STRING_LITERAL

ASL Input: s303.asl - 2871 lines, 112889 bytes, 2084 keywords Compilation complete. 2 Errors, 0 Warnings, 0 Remarks, 538 Optimizations
==============

I searched around the web and found only one message stating that acpidump didn't support the _OSI method yet (late 2002). Did this ever got solved?

-- Erico Mendonca
Techisa do Brasil

-------------------------------------------------------
This SF.net email is sponsored by: SF.net Giveback Program. Does SourceForge.net help you be more productive? Does it
help you create better code? SHARE THE LOVE, and help us help
YOU! Click Here: http://sourceforge.net/donate/

A message that is not in the archive (sent off-list or not yet received): 3FA58FE8.4000409-cz9lu//[email protected]

Erico M Mendonca

author

2003-11-02 23:40 UTC #2 lore raw

Quoted 13 lines · Erico M Mendonca

I'm currently trying to debug the DSDT table of this aforementioned machine, and managed to clear all the errors, except for this one:
=================
Intel ACPI Component Architecture
ASL Optimizing Compiler / AML Disassembler version 20030918 [Sep 18 2003]
Copyright (C) 2000 - 2003 Intel Corporation
Supports ACPI Specification Revision 2.0b
s303.asl 468: If(\_OSI) {
Error 1028 - Too few arguments ^ (\_OSI requires 1)
s303.asl 469: "Windows 2001"
Error 1037 - ^ parse error, unexpected PARSEOP_STRING_LITERAL

Quoted 10 lines

ASL Input: s303.asl - 2871 lines, 112889 bytes, 2084 keywords Compilation complete. 2 Errors, 0 Warnings, 0 Remarks, 538 Optimizations
==============
I searched around the web and found only one message stating that acpidump didn't support the _OSI method yet (late 2002). Did this ever got solved?
-- Erico Mendonca
Techisa do Brasil

Replying to myself...

The acpidump code I'm using is from the CVS, downloaded today:
6685 May 1 2003 acpidump.c

Ducrot Bruno

2003-11-03 09:55 UTC #3 lore raw

In reply to a message not in the archive

Quoted 15 lines · Erico M Mendonca

I'm currently trying to debug the DSDT table of this aforementioned machine, and managed to clear all the errors, except for this one:
=================
Intel ACPI Component Architecture
ASL Optimizing Compiler / AML Disassembler version 20030918 [Sep 18 2003] Copyright (C) 2000 - 2003 Intel Corporation
Supports ACPI Specification Revision 2.0b
s303.asl 468: If(\_OSI) {
Error 1028 - Too few arguments ^ (\_OSI requires 1)
s303.asl 469: "Windows 2001"
Error 1037 - ^ parse error,
unexpected PARSEOP_STRING_LITERAL

Quoted 8 lines

ASL Input: s303.asl - 2871 lines, 112889 bytes, 2084 keywords Compilation complete. 2 Errors, 0 Warnings, 0 Remarks, 538 Optimizations
==============
I searched around the web and found only one message stating that acpidump didn't support the _OSI method yet (late 2002). Did this ever got solved?

I remember that acpidmp is no more supported, in favor to iasl. acpidmp is now, under F-BSD, a wrapper for iasl, with some little goodies, like dumping more infos. Therefore, I doubt this will be corrected.

If you want more info, you should search FreeBSD mailing list (I don't remember the correct one, should be -current IIRC).

Signature · 11 lines

Ducrot Bruno

--  Which is worse:  ignorance or apathy?
--  Don't know.  Don't care.


-------------------------------------------------------
This SF.net email is sponsored by: SF.net Giveback Program.
Does SourceForge.net help you be more productive?  Does it
help you create better code?   SHARE THE LOVE, and help us help
YOU!  Click Here: http://sourceforge.net/donate/

A message that is not in the archive (sent off-list or not yet received): [email protected]

Nate Lawson

2003-11-03 21:25 UTC #5 lore raw

Quoted 24 lines · Ducrot Bruno

Quoted 15 lines · Erico M Mendonca
I'm currently trying to debug the DSDT table of this aforementioned machine, and managed to clear all the errors, except for this one:
=================
Intel ACPI Component Architecture
ASL Optimizing Compiler / AML Disassembler version 20030918 [Sep 18 2003] Copyright (C) 2000 - 2003 Intel Corporation
Supports ACPI Specification Revision 2.0b
s303.asl 468: If(\_OSI) {
Error 1028 - Too few arguments ^ (\_OSI requires 1)
s303.asl 469: "Windows 2001"
Error 1037 - ^ parse error,
unexpected PARSEOP_STRING_LITERAL
Quoted 7 lines
ASL Input: s303.asl - 2871 lines, 112889 bytes, 2084 keywords Compilation complete. 2 Errors, 0 Warnings, 0 Remarks, 538 Optimizations
==============
I searched around the web and found only one message stating that acpidump didn't support the _OSI method yet (late 2002). Did this ever got solved?

Quoted 7 lines

I remember that acpidmp is no more supported, in favor to iasl. acpidmp is now, under F-BSD, a wrapper for iasl, with some little goodies, like dumping more infos. Therefore, I doubt this will be corrected.
If you want more info, you should search FreeBSD mailing list (I don't remember the correct one, should be -current IIRC).

I'm the FreeBSD maintainer. acpidump(8) now just dumps the DSDT to a file and prints the fixed tables. It then calls iasl -d to disassemble the DSDT. I'm planning to add support for SSDT as well soon. If anyone from Linux is looking to port the new acpidump(8), let me know.

-Nate

Martin Michlmayr

2003-11-03 10:13 UTC #6 lore raw

In reply to a message not in the archive

Quoted 6 lines · Erico M Mendonca

machine, and managed to clear all the errors, except for this one:
s303.asl 468: If(\_OSI) {
Error 1028 - Too few arguments ^ (\_OSI requires 1)
s303.asl 469: "Windows 2001"
Error 1037 - ^ parse error,

I managed to fix this error with the following change:

dsdt.asl renamed from dsdt.asl.orig +1 -2

@@ -144,8 +144,7 @@
         }
         If(LOr(LEqual(C006, 0x0), LEqual(C006, 0x3))) {
             If(CondRefOf(\_OSI, Local0)) {
-                If(\_OSI) {
-                    "Windows 2001"
+                If(\_OSI ("Windows 2001")) {
                     Store(0x4, C006)
                 }
             }

This is basically in line with some other DSDTs I've seen on the net.

Signature · 9 lines

Martin Michlmayr
[email protected]


-------------------------------------------------------
This SF.net email is sponsored by: SF.net Giveback Program.
Does SourceForge.net help you be more productive?  Does it
help you create better code?   SHARE THE LOVE, and help us help
YOU!  Click Here: http://sourceforge.net/donate/

A message that is not in the archive (sent off-list or not yet received): 20031103101315.GA2100-u+sgIaa8TU6A7rR/[email protected]

Erico M Mendonca

author

2003-11-03 13:27 UTC #8 lore raw

Quoted 31 lines · Martin Michlmayr

Quoted 6 lines · Erico M Mendonca
machine, and managed to clear all the errors, except for this one:
Quoted 6 lines
s303.asl 468: If(\_OSI) {
Error 1028 - Too few arguments ^ (\_OSI requires 1)
s303.asl 469: "Windows 2001"
Error 1037 - ^ parse error,
I managed to fix this error with the following change:
dsdt.asl renamed from dsdt.asl.orig +1 -2
@@ -144,8 +144,7 @@
        }
        If(LOr(LEqual(C006, 0x0), LEqual(C006, 0x3))) {
            If(CondRefOf(\_OSI, Local0)) {
-                If(\_OSI) {
-                    "Windows 2001"
+                If(\_OSI ("Windows 2001")) {
                    Store(0x4, C006)
                }
            }

Quoted 3 lines

This is basically in line with some other DSDTs I've seen on the net.

Yesterday I did the same thing :)

Anyway, it did compile ok, I recompiled the kernel and added the fixed DSDT table. It worked quite nice, I now also have the power button generating the shutdown event.

I still have the fn-F5 panic though. I compiled the kernel with kdb and tried to redirect it to the serial console, but alas, this machine does not have a serial port, and as far as I have researched, support for USB serial adapters as console is available only on kernel 2.5+.

Signature · 12 lines

-- Erico Mendonca
Techisa do Brasil





-------------------------------------------------------
This SF.net email is sponsored by: SF.net Giveback Program.
Does SourceForge.net help you be more productive?  Does it
help you create better code?   SHARE THE LOVE, and help us help
YOU!  Click Here: http://sourceforge.net/donate/

Ducrot Bruno

2003-11-04 10:39 UTC #9 lore raw

In reply to a message not in the archive

Quoted 11 lines · Erico M Mendonca

I'm currently trying to debug the DSDT table of this aforementioned machine, and managed to clear all the errors, except for this one:
=================
Intel ACPI Component Architecture
ASL Optimizing Compiler / AML Disassembler version 20030918 [Sep 18 2003] Copyright (C) 2000 - 2003 Intel Corporation
Supports ACPI Specification Revision 2.0b
s303.asl 468: If(\_OSI) {
Error 1028 - Too few arguments ^ (\_OSI requires 1)

Quoted 8 lines

s303.asl 469: "Windows 2001"
Error 1037 - ^ parse error,
unexpected PARSEOP_STRING_LITERAL
ASL Input: s303.asl - 2871 lines, 112889 bytes, 2084 keywords Compilation complete. 2 Errors, 0 Warnings, 0 Remarks, 538 Optimizations
==============

FYI I have debugged a *lot* of similar models. What is the problem you encounter for that one?

Signature · 11 lines

Ducrot Bruno

--  Which is worse:  ignorance or apathy?
--  Don't know.  Don't care.


-------------------------------------------------------
This SF.net email is sponsored by: SF.net Giveback Program.
Does SourceForge.net help you be more productive?  Does it
help you create better code?   SHARE THE LOVE, and help us help
YOU!  Click Here: http://sourceforge.net/donate/

A message that is not in the archive (sent off-list or not yet received): [email protected]

Erico M Mendonca

author

2003-11-04 12:57 UTC #11 lore raw

Quoted 23 lines · Ducrot Bruno

Quoted 10 lines · Erico M Mendonca
I'm currently trying to debug the DSDT table of this aforementioned machine, and managed to clear all the errors, except for this one:
=================
Intel ACPI Component Architecture
ASL Optimizing Compiler / AML Disassembler version 20030918 [Sep 18 2003]
Copyright (C) 2000 - 2003 Intel Corporation
Supports ACPI Specification Revision 2.0b
s303.asl 468: If(\_OSI) {
Error 1028 - Too few arguments ^ (\_OSI requires 1)
Quoted 9 lines
s303.asl 469: "Windows 2001"
Error 1037 - ^ parse error, unexpected PARSEOP_STRING_LITERAL
ASL Input: s303.asl - 2871 lines, 112889 bytes, 2084 keywords Compilation complete. 2 Errors, 0 Warnings, 0 Remarks, 538 Optimizations
==============

Quoted 6 lines

FYI I have debugged a *lot* of similar models. What is the problem you encounter for that one?

I solved this by changing it to a simple

              If(\_OSI ("Windows 2001"))

as suggested by Martin earlier.

Anyway, I got the ACPI working fine for most functions, except:
- processor: no throttling (Pentium 4), but that may be out of the scope of ACPI(?)

processor id:            0
acpi id:                 0
bus mastering control:   no
power management:        yes
throttling control:      no
performance management:  no
limit interface:         no

- fan control/reporting
- thermal_zone and power_resource

The lid and power buttons are generating events after recompiling the DSDT table (they didn't before). Proper IRQ routing, battery and ac_adapter worked before.

As for the Fn-keys:

Fn-ESC (mute): doesn't work
Fn-F1 (screen lock): doesn't work
Fn-F2 (power toggle): doesn't work
Fn-F3 (suspend to RAM): doesn't work
Fn-F4 (hibernate): doesn't work
Fn-F5 (video switch): hangs the machine (read below) Fn-F6/F7 (brightness): works
Fn-F8 (wireless toggle): works
Fn-F9 (touchpad toggle): works Fn-F10/F11/F12: well, they should work anyway :)

I know some of these can be mapped to generate key events and call specific user programs. The one big problem I'm having now is with Fn-F5 (video switch). With an older kernel version (2.4.20) it worked flawlessly, but with 2.4.22 (from Mandrake 9.2) it hangs the machine solid. I tried setting a serial console to check out any oops/kernel panic messages, even tried KDB, but alas, this machine doesn't have a serial port.

I searched around and found out that there's an USB serial console patch, but for kernel 2.5+.

I'm just beginning with ACPI, so if you have any suggestions on this, I'd be glad to hear. I'm still reading the 500+ page tome of the ACPI spec... :)

Signature · 12 lines

-- Erico Mendonca
Techisa do Brasil





-------------------------------------------------------
This SF.net email is sponsored by: SF.net Giveback Program.
Does SourceForge.net help you be more productive?  Does it
help you create better code?   SHARE THE LOVE, and help us help
YOU!  Click Here: http://sourceforge.net/donate/

A message that is not in the archive (sent off-list or not yet received): 3FA7A224.1080907-cz9lu//[email protected]

Ducrot Bruno

2003-11-04 15:09 UTC #13 lore raw

Quoted 33 lines · Erico M Mendonca

Quoted 25 lines · Ducrot Bruno
Quoted 11 lines · Erico M Mendonca
I'm currently trying to debug the DSDT table of this aforementioned machine, and managed to clear all the errors, except for this one:
=================
Intel ACPI Component Architecture
ASL Optimizing Compiler / AML Disassembler version 20030918 [Sep 18 2003] Copyright (C) 2000 - 2003 Intel Corporation
Supports ACPI Specification Revision 2.0b
s303.asl 468: If(\_OSI) {
Error 1028 - Too few arguments ^ (\_OSI requires 1)
Quoted 10 lines
s303.asl 469: "Windows 2001"
Error 1037 - ^ parse error,
unexpected PARSEOP_STRING_LITERAL
ASL Input: s303.asl - 2871 lines, 112889 bytes, 2084 keywords Compilation complete. 2 Errors, 0 Warnings, 0 Remarks, 538 Optimizations
==============
Quoted 6 lines
FYI I have debugged a *lot* of similar models. What is the problem you encounter for that one?

Quoted 3 lines

I solved this by changing it to a simple
              If(\_OSI ("Windows 2001")) 

As I already tryed to say, unless you are on a BSD system, or unless someone port the FreeBSD acpidump, do not use acpidump. Or else, you are missing informations for debugging irq for example.

Quoted 3 lines

Anyway, I got the ACPI working fine for most functions, except:
- processor: no throttling (Pentium 4), but that may be out of the scope of ACPI(?)

It may be done via ACPI, but cpufreq can do that via p4-clockmod.

Quoted 7 lines

processor id:            0
acpi id:                 0
bus mastering control:   no
power management:        yes
throttling control:      no
performance management:  no
limit interface:         no

performance should work, strange. Anyway, cpufreq can do that via speedstep-ich.

Quoted 3 lines

- fan control/reporting
- thermal_zone and power_resource

A thermal_zone can be added to the AML for that model.

Quoted 7 lines

The lid and power buttons are generating events after recompiling the DSDT table (they didn't before). Proper IRQ routing, battery and ac_adapter worked before.
As for the Fn-keys:

Quoted 11 lines

Fn-ESC (mute): doesn't work
Fn-F1 (screen lock): doesn't work
Fn-F2 (power toggle): doesn't work
Fn-F3 (suspend to RAM): doesn't work
Fn-F4 (hibernate): doesn't work
Fn-F5 (video switch): hangs the machine (read below) Fn-F6/F7 (brightness): works
Fn-F8 (wireless toggle): works
Fn-F9 (touchpad toggle): works Fn-F10/F11/F12: well, they should work anyway :)

Some, if not all, of this Fn-key can be done via the omnibook kernel module (see http://sf.net/projects/omke). You can test at first the perl tool, in order to check if that is ok.

Cheers,

Signature · 11 lines

Ducrot Bruno

--  Which is worse:  ignorance or apathy?
--  Don't know.  Don't care.


-------------------------------------------------------
This SF.net email is sponsored by: SF.net Giveback Program.
Does SourceForge.net help you be more productive?  Does it
help you create better code?   SHARE THE LOVE, and help us help
YOU!  Click Here: http://sourceforge.net/donate/

A message that is not in the archive (sent off-list or not yet received): [email protected]

Len Brown

2003-11-05 01:07 UTC #15 lore raw

Quoted 6 lines · Ducrot Bruno

Quoted 1 line · Erico M Mendonca
Ducrot Bruno wrote:
As I already tryed to say, unless you are on a BSD system, or unless someone port the FreeBSD acpidump, do not use acpidump. Or else, you are missing informations for debugging irq for example.

huh?
Are you referring to acpidmp from Intel's pmtools package? What exactly is it missing?

thanks,
-Len

A message that is not in the archive (sent off-list or not yet received): [email protected]

Ducrot Bruno

2003-11-05 12:12 UTC #17 lore raw

Quoted 11 lines · Len Brown

Quoted 6 lines · Ducrot Bruno
Quoted 1 line · Erico M Mendonca
Ducrot Bruno wrote:
As I already tryed to say, unless you are on a BSD system, or unless someone port the FreeBSD acpidump, do not use acpidump. Or else, you are missing informations for debugging irq for example.
huh?
Are you referring to acpidmp from Intel's pmtools package? What exactly is it missing?

acpidump, with a 'u'. It has nothing to do with pmtools.

It was the only free, decent AML decompiler available, before iasl implemented the '-d' option, and is part of FreeBSD system.

A backport for other OS, like linux, were made, but is less complete than the FreeBSD version.

Signature · 11 lines

Ducrot Bruno

--  Which is worse:  ignorance or apathy?
--  Don't know.  Don't care.


-------------------------------------------------------
This SF.net email is sponsored by: SF.net Giveback Program.
Does SourceForge.net help you be more productive?  Does it
help you create better code?   SHARE THE LOVE, and help us help
YOU!  Click Here: http://sourceforge.net/donate/