Resuming IDE devices with ACPI

Resuming IDE devices with ACPI

[Matthew Garrett]

http://godmode.ca/bookNref/win/XpFastBootUp.pdf (a copy of a Microsoft
document for system integrators) has this to say:

Actions for BIOS Wake Improvements 

...

Do not configure IDE drives in BIOS; use the _GTF method in your ASL
code.

which makes it unsurprising that hardware without working IDE on resume
is continuing to appear. In order to resume it, we need to be able to
make the appropriate taskfile calls with the output of the _GTF method.
For this to be possible, we need to call each IDE controller's _GTF
method on resume. How can we associate each IDE controller with a given
piece of ACPI namespace?

-- 
Matthew Garrett | mjg59-1xO5oi07KQx4cg9Nei1l7Q@public.gmane.org



-------------------------------------------------------
This SF.Net email is sponsored by: NEC IT Guy Games.
Get your fingers limbered up and give it your best shot. 4 great events, 4
opportunities to win big! Highest score wins.NEC IT Guy Games. Play to
win an NEC 61 plasma display. Visit http://www.necitguy.com/?r=20

Re: Resuming IDE devices with ACPI

[Alan Cox]

On Llu, 2005-05-09 at 14:51, Matthew Garrett wrote:
> which makes it unsurprising that hardware without working IDE on resume
> is continuing to appear. In order to resume it, we need to be able to
> make the appropriate taskfile calls with the output of the _GTF method.
> For this to be possible, we need to call each IDE controller's _GTF
> method on resume. How can we associate each IDE controller with a given
> piece of ACPI namespace?

We do know the resource allocations of each controller.

You also then have to retune the drives etc and resume operations having
checked various things like drive model/id to spot cold plugs (otherwise
if you get the BIOS numbering wrong or something moves you'll do serious
damage to the
users data)

Alan



-------------------------------------------------------
This SF.Net email is sponsored by: NEC IT Guy Games.
Get your fingers limbered up and give it your best shot. 4 great events, 4
opportunities to win big! Highest score wins.NEC IT Guy Games. Play to
win an NEC 61 plasma display. Visit http://www.necitguy.com/?r=20

RE: Resuming IDE devices with ACPI

[Li, Shaohua]

>http://godmode.ca/bookNref/win/XpFastBootUp.pdf (a copy of a Microsoft
>document for system integrators) has this to say:
>
>Actions for BIOS Wake Improvements
>
>...
>
>Do not configure IDE drives in BIOS; use the _GTF method in your ASL
>code.
>
>which makes it unsurprising that hardware without working IDE on resume
>is continuing to appear. In order to resume it, we need to be able to
>make the appropriate taskfile calls with the output of the _GTF method.
>For this to be possible, we need to call each IDE controller's _GTF
>method on resume. How can we associate each IDE controller with a given
>piece of ACPI namespace?
Hi Matthew,
I have an internal patch which can get IDE's ACPI device (currently only
support PCI IDE controller) and execute _GTF, _GTM and _STM at
suspend/resume. Unfortunately, I'm not familiar with IDE, so I don't
know how to invoke the IDE commands returned from _GTF at resume.
I sent an email to IDE list for help, but got no response :(.
I believe it's a quite generic suspend/resume bug. Did anybody know
Linux IDE in the list? If you need more info such as the IDE commands
from one of my system, please let me know.

Thanks,
Shaohua


-------------------------------------------------------
This SF.Net email is sponsored by Oracle Space Sweepstakes
Want to be the first software developer in space?
Enter now for the Oracle Space Sweepstakes!
http://ads.osdn.com/?ad_ids93&alloc_id\x16281&op=click

RE: Resuming IDE devices with ACPI

[Matthew Garrett]

On Tue, 2005-05-10 at 08:55 +0800, Li, Shaohua wrote:

> I have an internal patch which can get IDE's ACPI device (currently only
> support PCI IDE controller) and execute _GTF, _GTM and _STM at
> suspend/resume. Unfortunately, I'm not familiar with IDE, so I don't
> know how to invoke the IDE commands returned from _GTF at resume.
> I sent an email to IDE list for help, but got no response :(.
> I believe it's a quite generic suspend/resume bug. Did anybody know
> Linux IDE in the list? If you need more info such as the IDE commands
> from one of my system, please let me know.

I /believe/ that it the buffer returned from _GTF just needs to be split
into 7-byte chunks and then fed to taskfile_output_data. This seems to
be what happens when taskfile commands are triggered by IDE ioctls (see
the hdparm or smartmontools source code, for example). Is your patch
available anywhere for testing?

Thanks,
-- 
Matthew Garrett | mjg59-1xO5oi07KQx4cg9Nei1l7Q@public.gmane.org



-------------------------------------------------------
This SF.Net email is sponsored by Oracle Space Sweepstakes
Want to be the first software developer in space?
Enter now for the Oracle Space Sweepstakes!
http://ads.osdn.com/?ad_id=7393&alloc_id=16281&op=click