2005-12-25 20:30:14

by Jaco Kroon

[permalink] [raw]
Subject: [PATCH] ati-agp suspend/resume support

Based on the patch at
http://unixhead.org/docs/thinkpad/ati-agp/ati-agp.diff, add support for
suspend/resume in the ati-agp module.

Signed-of-by: Jaco Kroon <[email protected]>

--- linux-2.6.15-rc6/drivers/char/agp/ati-agp.c.orig 2005-12-25
22:21:32.000000000 +0200
+++ linux-2.6.15-rc6/drivers/char/agp/ati-agp.c 2005-12-25
22:23:33.000000000 +0200
@@ -243,6 +243,15 @@
return 0;
}

+static int ati_resume(struct pci_dev *dev)
+{
+ return ati_configure();
+}
+
+static int ati_suspend(struct pci_dev *dev, pm_message_t state)
+{
+ return 0;
+}

/*
*Since we don't need contigious memory we just try
@@ -525,6 +534,8 @@
.id_table = agp_ati_pci_table,
.probe = agp_ati_probe,
.remove = agp_ati_remove,
+ .resume = ati_resume,
+ .suspend = ati_suspend,
};

static int __init agp_ati_init(void)

--
There are only 10 kinds of people in this world,
those that understand binary and those that don't.
http://www.kroon.co.za/


Attachments:
smime.p7s (3.09 kB)
S/MIME Cryptographic Signature

2005-12-25 21:14:39

by Pavel Machek

[permalink] [raw]
Subject: Re: [PATCH] ati-agp suspend/resume support

Hi!

> Based on the patch at
> http://unixhead.org/docs/thinkpad/ati-agp/ati-agp.diff, add support for
> suspend/resume in the ati-agp module.
>
> Signed-of-by: Jaco Kroon <[email protected]>
>
> --- linux-2.6.15-rc6/drivers/char/agp/ati-agp.c.orig 2005-12-25
> 22:21:32.000000000 +0200
> +++ linux-2.6.15-rc6/drivers/char/agp/ati-agp.c 2005-12-25
> 22:23:33.000000000 +0200
> @@ -243,6 +243,15 @@
> return 0;
> }
>
> +static int ati_resume(struct pci_dev *dev)
> +{
> + return ati_configure();
> +}
> +
> +static int ati_suspend(struct pci_dev *dev, pm_message_t state)
> +{
> + return 0;
> +}

I think you can just leave out empty function; that should work,
too. Otherwise it looks good, thanks... ...

> /*
> *Since we don't need contigious memory we just try
> @@ -525,6 +534,8 @@
> .id_table = agp_ati_pci_table,
> .probe = agp_ati_probe,
> .remove = agp_ati_remove,
> + .resume = ati_resume,
> + .suspend = ati_suspend,
> };

...aha, plus you may want to keep naming convention. It should
probably be agp_ati_resume.
Pavel
--
Thanks, Sharp!

2005-12-26 04:53:28

by Jaco Kroon

[permalink] [raw]
Subject: Re: [PATCH] ati-agp suspend/resume support

Pavel Machek wrote:
> Hi!
>
>
>>Based on the patch at
>>http://unixhead.org/docs/thinkpad/ati-agp/ati-agp.diff, add support for
>>suspend/resume in the ati-agp module.
>>
>>Signed-of-by: Jaco Kroon <[email protected]>
>>
>>--- linux-2.6.15-rc6/drivers/char/agp/ati-agp.c.orig 2005-12-25
>>22:21:32.000000000 +0200
>>+++ linux-2.6.15-rc6/drivers/char/agp/ati-agp.c 2005-12-25
>>22:23:33.000000000 +0200
>>@@ -243,6 +243,15 @@
>> return 0;
>> }
>>
>>+static int ati_resume(struct pci_dev *dev)
>>+{
>>+ return ati_configure();
>>+}
>>+
>>+static int ati_suspend(struct pci_dev *dev, pm_message_t state)
>>+{
>>+ return 0;
>>+}
>
>
> I think you can just leave out empty function; that should work,
> too. Otherwise it looks good, thanks... ...

That is indeed the case.

>> /*
>> *Since we don't need contigious memory we just try
>>@@ -525,6 +534,8 @@
>> .id_table = agp_ati_pci_table,
>> .probe = agp_ati_probe,
>> .remove = agp_ati_remove,
>>+ .resume = ati_resume,
>>+ .suspend = ati_suspend,
>> };
>
> ...aha, plus you may want to keep naming convention. It should
> probably be agp_ati_resume.

Done. Resubmitting in a second.
--
There are only 10 kinds of people in this world,
those that understand binary and those that don't.
http://www.kroon.co.za/


Attachments:
smime.p7s (3.09 kB)
S/MIME Cryptographic Signature