Gateway 200x - trouble supplying ECDT table

Gateway 200x - trouble supplying ECDT table

[Greg Sarjeant]

Hello,

   I know that this has been covered here before, but I have gotten stuck, so I was wondering if someone could give me a nudge in the right direction.

   I have a Gateway 200x. Apparently, as discussed here previously (http://sourceforge.net/mailarchive/message.php?msg_id=6514510), these machines do not provide an ECDT. I am getting some EmbeddedController errors in dmesg, and do not have battery, button, thermal or ac adapter support. Obviously, it would be nice to resolve this :)

   I followed the thread above, and noticed that there is a patch for the 2.4.23 kernel that allows you to supply an ECDT as a boot parameter. Sounds like just the thing. I have applied the patch and am sending what I believe to be the correct parameters to the kernel in my grub.conf, but it will not use the faked ECDT. I imagine that I am just doing something stupid, but I can't figure out what it is right now. I have tried a number of different ways of specifying these parameters, but none has worked so far. I've posted some files (DSDT, dmesg, grub.conf and the version of the patch that I applied) at http://www.morningdave.org/acpi/. If someone could look them over and let me know if anything jumps out as obviously wrong, I would really appreciate it. Right now, I'm hopeful that it is just a subtle difference in the syntax for lilo.conf and grub.conf that I am not catching. If you nee
 d any other information, please let me know.

   Also, if it wouldn't be too much trouble, could someone let me know how one determines the command and data ports that are used by the EmbeddedController? I am using the values that folks seemed to agree on in the previous thread for this machine (0x66 and 0x62, respectively). But my only real justification is that they were mentioned in that thread, and I see those values in the DSDT. I don't really understand how I could have identified them without someone who knows what he is doing providing them first.

   Thanks,
   Greg


-------------------------------------------------------
This SF.net email is sponsored by: IBM Linux Tutorials.
Become an expert in LINUX or just sharpen your skills.  Sign up for IBM's
Free Linux Tutorials.  Learn everything from the bash shell to sys admin.
Click now! http://ads.osdn.com/?ad_id=1278&alloc_id=3371&op=click

Re: Gateway 200x - trouble supplying ECDT table

[Greg Sarjeant]

Hello again,

    I've made a bit of progress on this, but am still a bit puzzled.
 
    It turns out that grub.conf does not require the double-slash when specifying the path to the Embedded Controller device. My grub.conf looked like this before:

	title=Gentoo Linux (2.4.23 fake-ecdt)
	root (hd0,5)
	kernel (hd0,5)/boot/linux-2.4.23 root=/dev/hda8 ecdt_fake=0x66:0x62:0x1c:0x01:\\_SB_.PCI0.LPCB.H_EC

dmesg would print the following messages when I got to the fake ECDT stuff:

	ACPI: Faking ECDT
	ACPI: Could not use ECDT

Now, my grub.conf is:

	title=Gentoo Linux (2.4.23 fake-ecdt)
	root (hd0,5)
	kernel (hd0,5)/boot/linux-2.4.23 root=/dev/hda8 ecdt_fake=0x66:0x62:0x1c:0x01:\_SB_.PCI0.LPCB.H_EC

And dmesg gives me:

ACPI: Faking ECDT
evregion-0345: *** Error: Handler for [EmbeddedControl] returned AE_BAD_PARAMETER
 dswexec-0422 [12] ds_exec_end_op        : [Store]: Could not resolve operands, AE_BAD_PARAMETER
 psparse-1120: *** Error: Method execution failed [\_SB_.PCI0.LPCB.H_EC._REG] (Node df5e4408), AE_BAD_PARAMETER
ACPI: Could not use ECDT

So, instead of giving up on the fake ECDT immediately, it throws a couple of errors and then gives up. I think that that's progress, as it at least seems to be recognizing my attempt. I'm still not sure why I continue to get AE_BAD_PARAMETER errors, though. Perhaps there is also something wrong with the DSDT? I did notice that /proc/acpi/embeddedcontroller/H_EC/info does not seem to agree with my dsdt, as it appeared that other people's did in the previous email thread about this. Here is the output of cat /proc/acpi/embeddedcontroller/H_EC/info:

	gpe bit:                 0x00
	ports:                   0x08, 0x05
	use global lock:         yes


I have updated the grub.conf and dmesg files and added /proc/acpi/embeddedcontroller/H_EC/info to http://www.morningdave.org/acpi/ to reflect these stunning new developments, in case anyone is really bored and would like to look them over.

Thanks again,
 Greg
  
On Mon, 22 Dec 2003 10:10:55 -0500
Greg Sarjeant <greg-QNIYhHqVzB9kr2E5YSwMOQ@public.gmane.org> wrote:

> Hello,
> 
>    I know that this has been covered here before, but I have gotten stuck, so I was wondering if someone could give me a nudge in the right direction.
> 
>    I have a Gateway 200x. Apparently, as discussed here previously (http://sourceforge.net/mailarchive/message.php?msg_id=6514510), these machines do not provide an ECDT. I am getting some EmbeddedController errors in dmesg, and do not have battery, button, thermal or ac adapter support. Obviously, it would be nice to resolve this :)
> 
>    I followed the thread above, and noticed that there is a patch for the 2.4.23 kernel that allows you to supply an ECDT as a boot parameter. Sounds like just the thing. I have applied the patch and am sending what I believe to be the correct parameters to the kernel in my grub.conf, but it will not use the faked ECDT. I imagine that I am just doing something stupid, but I can't figure out what it is right now. I have tried a number of different ways of specifying these parameters, but none has worked so far. I've posted some files (DSDT, dmesg, grub.conf and the version of the patch that I applied) at http://www.morningdave.org/acpi/. If someone could look them over and let me know if anything jumps out as obviously wrong, I would really appreciate it. Right now, I'm hopeful that it is just a subtle difference in the syntax for lilo.conf and grub.conf that I am not catching. If you n
 eed any other information, please let me know.
> 
>    Also, if it wouldn't be too much trouble, could someone let me know how one determines the command and data ports that are used by the EmbeddedController? I am using the values that folks seemed to agree on in the previous thread for this machine (0x66 and 0x62, respectively). But my only real justification is that they were mentioned in that thread, and I see those values in the DSDT. I don't really understand how I could have identified them without someone who knows what he is doing providing them first.
> 
>    Thanks,
>    Greg
> 
> 
> -------------------------------------------------------
> This SF.net email is sponsored by: IBM Linux Tutorials.
> Become an expert in LINUX or just sharpen your skills.  Sign up for IBM's
> Free Linux Tutorials.  Learn everything from the bash shell to sys admin.
> Click now! http://ads.osdn.com/?ad_id=1278&alloc_id=3371&op=click
> _______________________________________________
> Acpi-devel mailing list
> Acpi-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org
> https://lists.sourceforge.net/lists/listinfo/acpi-devel
> 




-------------------------------------------------------
This SF.net email is sponsored by: IBM Linux Tutorials.
Become an expert in LINUX or just sharpen your skills.  Sign up for IBM's
Free Linux Tutorials.  Learn everything from the bash shell to sys admin.
Click now! http://ads.osdn.com/?ad_id=1278&alloc_id=3371&op=click

RE: Gateway 200x - trouble supplying ECDT table

[Yu, Luming]

There is a patch at http://bugzilla.kernel.org/show_bug.cgi?id=1690
trying to solve ec._INI failure due to lacking ECDT.      
But it's not a generic solution.  Since you are faking ECDT using data
of EC device, why not do it automatically?

--Luming                                                            


-------------------------------------------------------
This SF.net email is sponsored by: IBM Linux Tutorials.
Become an expert in LINUX or just sharpen your skills.  Sign up for IBM's
Free Linux Tutorials.  Learn everything from the bash shell to sys admin.
Click now! http://ads.osdn.com/?ad_id\x1278&alloc_id371&op=click

Re: Gateway 200x - trouble supplying ECDT table

[Greg Sarjeant]

Luming,

    I am sorry that you will receive this twice. I did not send my original reply back to the mailinglist.

    Regards,
    Greg

############################################################################

Thanks, Luming.

    I applied this patch to my 2.4.23 kernel (started back from the fresh sources and applied the latest ACPI patch first). Unfortunately, I'm still getting the same sorts of messages on boot. The _REG method of the Embedded Control fails as follows:

ACPI: Embedded Controller [H_EC] (gpe 28)
evregion-0348: *** Error: Handler for [EmbeddedControl] returned AE_BAD_PARAMETER
 dswexec-0435 [30] ds_exec_end_op        : [Store]: Could not resolve operands, AE_BAD_PARAMETER
 psparse-1120: *** Error: Method execution failed [\_SB_.PCI0.LPCB.H_EC._REG] (Node df5e7ae8), AE_BAD_PARAMETER


    I then get similar errors from the battery, adapter and thermal modules, but the root always appears to be the Embedded Controller:

	ACPI: Processor [CPU0] (supports C1 C2 C3, 5 performance states, 8 throttling states)
	ACPI: Power Button (FF) [PWRF]
	ACPI: Lid Switch [LID0]
	ACPI: Sleep Button (CM) [SLPB]
	evregion-0348: *** Error: Handler for [EmbeddedControl] returned AE_BAD_PARAMETER
	 dswexec-0435 [45] ds_exec_end_op        : [Store]: Could not resolve operands, AE_BAD_PARAMETER
	 psparse-1120: *** Error: Method execution failed [\_SB_.ADP1._PSR] (Node df5f40c8), AE_BAD_PARAMETER
	 acpi_ac-0083 [36] acpi_ac_get_state     : Error reading AC Adapter state
	evregion-0348: *** Error: Handler for [EmbeddedControl] returned AE_BAD_PARAMETER
	dswexec-0435 [46] ds_exec_end_op        : [LEqual]: Could not resolve operands, AE_BAD_PARAMETER
	dswstate-0273 [48] ds_result_pop_from_bot: No result objects! State=df5fbc28
	dsutils-0525 [48] ds_create_operand     : Missing or null operand, AE_AML_NO_RETURN_VALUE
	psparse-1120: *** Error: Method execution failed [\_SB_.BAT1._STA] (Node df5f4848), AE_AML_NO_RETURN_VALUE
	evregion-0348: *** Error: Handler for [EmbeddedControl] returned AE_BAD_PARAMETER
	dswexec-0435 [46] ds_exec_end_op        : [Store]: Could not resolve operands, AE_BAD_PARAMETER
 	psparse-1120: *** Error: Method execution failed [\_TZ_.THRM._TMP] (Node df5f6f68), AE_BAD_PARAMETER
	ACPI: Fan [FAN0] (on)


    My assumption at this point is that the battery, adapter and thermal modules rely on the EC, so if I can provide an ECDT, the other errors should also go away.

    I had previously found this bug at bugzilla that deals with the same problem on the same machine:
	http://bugme.osdl.org/show_bug.cgi?id=1515

    Unfortunately, the only suggestion is to provide the ECDT manually, which is, of course, where I am stuck now. Neither of the methods that I have tried so far has worked, and I'm beginning to wonder if the DSDT is buggy enough that even providing a correct ECDT just causes errors on initialization. Is that even possible?

    I'm a little reluctant to open a new bug, since I know both the problem and the solution, Besides, it isn't really an ACPI bug, it's Gateway's issue. My only problem at this point is implementing the workaround.

    Have I missed something in applying your patch? You said that it isn't a generic solution, but I'm not quite sure what you meant by that. Do I have to modify it for my system, or do you just mean that it should allow me to circumvent the EC initialization error, but it won't help later functions that require an initialized EC?

    I have provided the entire dmesg output here in case you would find it helpful:

http://www.morningdave.org/acpi/dmesg-luming-patch

    I have also posted my DSDT here, in hope that it may shed some light on the problem:

http://www.morningdave.org/acpi/dsdt-disasm

     Thank you for your help. I'll keep digging around bugzilla to see if I can find some more pointers on supplying the ECDT manually. If you think it would be helpful for me to open a bug and post all of this information on bugzilla, I'll be happy to do it. I just don't want to give you extra bug reports for problems with known solutions (though I guess I'm cluttering your inbox with such reports now as it is).

    Thanks,
     Greg



On Tue, 23 Dec 2003 14:07:22 +0800
"Yu, Luming" <luming.yu-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org> wrote:

> There is a patch at http://bugzilla.kernel.org/show_bug.cgi?id=1690
> trying to solve ec._INI failure due to lacking ECDT.      
> But it's not a generic solution.  Since you are faking ECDT using data
> of EC device, why not do it automatically?
> 
> --Luming                                                            
> 




-------------------------------------------------------
This SF.net email is sponsored by: IBM Linux Tutorials.
Become an expert in LINUX or just sharpen your skills.  Sign up for IBM's
Free Linux Tutorials.  Learn everything from the bash shell to sys admin.
Click now! http://ads.osdn.com/?ad_id=1278&alloc_id=3371&op=click

RE: Gateway 200x - trouble supplying ECDT table

[Yu, Luming]

>I am sorry that you will receive this twice.
>I did not send my original reply back to the mailinglist.
>

Never mind!

>  I applied this patch to my 2.4.23 kernel 
>(started back from the fresh sources and applied the latest ACPI patch
first)

After a few investigation on your problem, that patch doesn't apply.
>From your DSDT, 
ECDT seems to be  unnecessary to your box.

>Unfortunately, I'm still getting the same sorts of >>messages on boot. 
>The _REG method of the Embedded Control fails as follows:

Obviously, this issue is relative to acpi_ec_space_handler.
 
>ACPI: Embedded Controller [H_EC] (gpe 28)
>evregion-0348: *** Error: Handler for [EmbeddedControl] returned
AE_BAD_PARAMETER
> dswexec-0435 [30] ds_exec_end_op        : [Store]: Could not resolve
operands, AE_BAD_PARAMETER
> psparse-1120: *** Error: Method execution failed
[\_SB_.PCI0.LPCB.H_EC._REG] (Node df5e7ae8), AE_BAD_PARAMETER

I just filed a patch for you testing, just try it.
http://bugzilla.kernel.org/show_bug.cgi?id=1744

Maybe we need to support other access bit width.  I found AccessType:
DWordAcc in your EC region definition.

Thanks,
Luming



-------------------------------------------------------
This SF.net email is sponsored by: IBM Linux Tutorials.
Become an expert in LINUX or just sharpen your skills.  Sign up for IBM's
Free Linux Tutorials.  Learn everything from the bash shell to sys admin.
Click now! http://ads.osdn.com/?ad_id\x1278&alloc_id371&op=click

Re: Gateway 200x - trouble supplying ECDT table

[Greg Sarjeant]

Thanks, Luming. That patch helped, though it has not completely fixed the problem. I have attached the details to the bug you created. I will submit any future updates to that bug as well, instead of this mailing list.

  Thanks again,
  Greg



On Thu, 25 Dec 2003 12:08:41 +0800
"Yu, Luming" <luming.yu-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org> wrote:

> >I am sorry that you will receive this twice.
> >I did not send my original reply back to the mailinglist.
> >
> 
> Never mind!
> 
> >  I applied this patch to my 2.4.23 kernel 
> >(started back from the fresh sources and applied the latest ACPI patch
> first)
> 
> After a few investigation on your problem, that patch doesn't apply.
> >From your DSDT, 
> ECDT seems to be  unnecessary to your box.
> 
> >Unfortunately, I'm still getting the same sorts of >>messages on boot. 
> >The _REG method of the Embedded Control fails as follows:
> 
> Obviously, this issue is relative to acpi_ec_space_handler.
>  
> >ACPI: Embedded Controller [H_EC] (gpe 28)
> >evregion-0348: *** Error: Handler for [EmbeddedControl] returned
> AE_BAD_PARAMETER
> > dswexec-0435 [30] ds_exec_end_op        : [Store]: Could not resolve
> operands, AE_BAD_PARAMETER
> > psparse-1120: *** Error: Method execution failed
> [\_SB_.PCI0.LPCB.H_EC._REG] (Node df5e7ae8), AE_BAD_PARAMETER
> 
> I just filed a patch for you testing, just try it.
> http://bugzilla.kernel.org/show_bug.cgi?id=1744
> 
> Maybe we need to support other access bit width.  I found AccessType:
> DWordAcc in your EC region definition.
> 
> Thanks,
> Luming
> 
> 
> 
> -------------------------------------------------------
> This SF.net email is sponsored by: IBM Linux Tutorials.
> Become an expert in LINUX or just sharpen your skills.  Sign up for IBM's
> Free Linux Tutorials.  Learn everything from the bash shell to sys admin.
> Click now! http://ads.osdn.com/?ad_id\x1278&alloc_id371&op_k
> _______________________________________________
> Acpi-devel mailing list
> Acpi-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org
> https://lists.sourceforge.net/lists/listinfo/acpi-devel
> 




-------------------------------------------------------
This SF.net email is sponsored by: IBM Linux Tutorials.
Become an expert in LINUX or just sharpen your skills.  Sign up for IBM's
Free Linux Tutorials.  Learn everything from the bash shell to sys admin.
Click now! http://ads.osdn.com/?ad_id\x1278&alloc_id371&op=click