ACPI boot fix for 2.6.15

ACPI boot fix for 2.6.15

[Andi Kleen]

Hallo,

I have the following patch in my x86-64 tree because it was needed
to boot without acpi=off on several of the machines here. 
Without it the ACPI device setup code would always create a loop
in the ACPI device tree and eventually some other code walking
it would hang the machine.

The patch has been around for a long time - it was originally 
done by Thomas Renninger - and has even been shipped in SUSE 10.0
so it was well tested
(which was 2.6.13 based, there the problem only hit occasionally,
in 2.6.14+ it seems to happen more frequently for some reason)

I think it was queued in the ACPI trees too and acked by the
ACPI poeople.

I know Len skipped the ACPI update for .15 for some reason,
but could we perhaps merge this safe patch. It would help
my test machines at least :)

----

From: trenn-l3A5Bk7waGM@public.gmane.org

Avoid endless loops in ACPI device tree.

Cc: Len.Brown-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org

Signed-off-by: Andi Kleen <ak-l3A5Bk7waGM@public.gmane.org>

Index: linux/drivers/acpi/scan.c
===================================================================
--- linux.orig/drivers/acpi/scan.c
+++ linux/drivers/acpi/scan.c
@@ -1110,7 +1110,7 @@ acpi_add_single_object(struct acpi_devic
 	 *
 	 * TBD: Assumes LDM provides driver hot-plug capability.
 	 */
-	result = acpi_bus_find_driver(device);
+	acpi_bus_find_driver(device);
 
       end:
 	if (!result)



-------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc. Do you grep through log files
for problems?  Stop!  Download the new AJAX search engine that makes
searching your log files as easy as surfing the  web.  DOWNLOAD SPLUNK!
http://ads.osdn.com/?ad_id=7637&alloc_id=16865&op=click

RE: ACPI boot fix for 2.6.15

[Brown, Len]

Andi,
this shipped already in 2.6.15-rc5, per
http://bugzilla.kernel.org/show_bug.cgi?id=5221

thanks,
-Len 

>-----Original Message-----
>From: Andi Kleen [mailto:ak-l3A5Bk7waGM@public.gmane.org] 
>Sent: Tuesday, December 13, 2005 12:29 AM
>To: torvalds-3NddpPZAyC0@public.gmane.org; akpm-3NddpPZAyC0@public.gmane.org; Brown, Len; 
>trenn-l3A5Bk7waGM@public.gmane.org; acpi-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org
>Subject: ACPI boot fix for 2.6.15
>
>
>Hallo,
>
>I have the following patch in my x86-64 tree because it was needed
>to boot without acpi=off on several of the machines here. 
>Without it the ACPI device setup code would always create a loop
>in the ACPI device tree and eventually some other code walking
>it would hang the machine.
>
>The patch has been around for a long time - it was originally 
>done by Thomas Renninger - and has even been shipped in SUSE 10.0
>so it was well tested
>(which was 2.6.13 based, there the problem only hit occasionally,
>in 2.6.14+ it seems to happen more frequently for some reason)
>
>I think it was queued in the ACPI trees too and acked by the
>ACPI poeople.
>
>I know Len skipped the ACPI update for .15 for some reason,
>but could we perhaps merge this safe patch. It would help
>my test machines at least :)
>
>----
>
>From: trenn-l3A5Bk7waGM@public.gmane.org
>
>Avoid endless loops in ACPI device tree.
>
>Cc: Len.Brown-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org
>
>Signed-off-by: Andi Kleen <ak-l3A5Bk7waGM@public.gmane.org>
>
>Index: linux/drivers/acpi/scan.c
>===================================================================
>--- linux.orig/drivers/acpi/scan.c
>+++ linux/drivers/acpi/scan.c
>@@ -1110,7 +1110,7 @@ acpi_add_single_object(struct acpi_devic
> 	 *
> 	 * TBD: Assumes LDM provides driver hot-plug capability.
> 	 */
>-	result = acpi_bus_find_driver(device);
>+	acpi_bus_find_driver(device);
> 
>       end:
> 	if (!result)
>
>


-------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc. Do you grep through log files
for problems?  Stop!  Download the new AJAX search engine that makes
searching your log files as easy as surfing the  web.  DOWNLOAD SPLUNK!
http://ads.osdn.com/?ad_idv37&alloc_id\x16865&op=click

Re: ACPI boot fix for 2.6.15

[Dave Jones]

On Tue, Dec 13, 2005 at 06:28:34AM +0100, Andi Kleen wrote:
 > 
 > Hallo,
 > 
 > I have the following patch in my x86-64 tree because it was needed
 > to boot without acpi=off on several of the machines here. 
 > Without it the ACPI device setup code would always create a loop
 > in the ACPI device tree and eventually some other code walking
 > it would hang the machine.
 > 
 > The patch has been around for a long time - it was originally 
 > done by Thomas Renninger - and has even been shipped in SUSE 10.0
 > so it was well tested
 > (which was 2.6.13 based, there the problem only hit occasionally,
 > in 2.6.14+ it seems to happen more frequently for some reason)
 > 
 > I think it was queued in the ACPI trees too and acked by the
 > ACPI poeople.
 > 
 > I know Len skipped the ACPI update for .15 for some reason,
 > but could we perhaps merge this safe patch. It would help
 > my test machines at least :)

It got merged on Dec 1st.

		Dave



-------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc. Do you grep through log files
for problems?  Stop!  Download the new AJAX search engine that makes
searching your log files as easy as surfing the  web.  DOWNLOAD SPLUNK!
http://ads.osdn.com/?ad_id=7637&alloc_id=16865&op=click

Re: ACPI boot fix for 2.6.15

[Andi Kleen]

On Tue, Dec 13, 2005 at 12:35:25AM -0500, Brown, Len wrote:
> Andi,
> this shipped already in 2.6.15-rc5, per
> http://bugzilla.kernel.org/show_bug.cgi?id=5221

Ok, my apologies Len. I must have looked in a rc4 kernel
by mistake.

-Andi


-------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc. Do you grep through log files
for problems?  Stop!  Download the new AJAX search engine that makes
searching your log files as easy as surfing the  web.  DOWNLOAD SPLUNK!
http://ads.osdn.com/?ad_id=7637&alloc_id=16865&op=click