ACPICA release 20101013 linuxized patches

ACPICA release 20101013 linuxized patches

[Lin Ming]

[-- Attachment #1: Type: text/plain, Size: 848 bytes --]

Hi, Len

ACPICA release 20101013 linuxized patches attached.
There is a small conflict for PATCH 5 between test and acpica branches,
so I made 2 mbox files for each branch to apply, you choose it.

acpica-v20101013-on_top_of_test_branch.mbox applied cleanly on top of
your linux-acpi-2.6/test branch.

acpica-v20101013-on_top_of_acpica_branch.mbox applied cleanly on top of
your linux-acpi-2.6/acpica branch.

Build/Tested on my 32/64 bit machines, both work OK.

[PATCH 1/6] ACPICA: Comment update; no functional change
[PATCH 2/6] ACPICA: Change type of _TZ from ThermalZone to Device
[PATCH 3/6] ACPICA: Eliminate duplicate code in acpi_ut_execute_* functions
[PATCH 4/6] ACPICA: Add Vista SP2 to supported _OSI strings
[PATCH 5/6] ACPICA: Clear PCIEXP_WAKE_STS when clearing ACPI events
[PATCH 6/6] ACPICA: Update version to 20101013

Lin Ming

[-- Attachment #2: acpica-v20101013-on_top_of_acpica_branch.mbox --]
[-- Type: application/mbox, Size: 10477 bytes --]

[-- Attachment #3: acpica-v20101013-on_top_of_test_branch.mbox --]
[-- Type: application/mbox, Size: 10487 bytes --]

Re: ACPICA release 20101013 linuxized patches

[Len Brown]

patch 5 doesn't look quite right,
because it leaves both

#define ACPI_BITMASK_PCIEXP_WAKE_STATUS         0x4000  /* ACPI 3.0 */
and
#define ACPI_BITMASK_PCIEXP_WAKE_DISABLE        0x4000  /* ACPI 3.0 */

in aclocal.h

If we're going to re-name what the original patch did,
then we should finsh the job and delete the original name, yes?

thanks,
Len Brown, Intel Open Source Technology Center

Re: ACPICA release 20101013 linuxized patches

[Lin Ming]

On Mon, 2010-10-18 at 13:31 +0800, Len Brown wrote:
> patch 5 doesn't look quite right,
> because it leaves both
> 
> #define ACPI_BITMASK_PCIEXP_WAKE_STATUS         0x4000  /* ACPI 3.0 */
> and
> #define ACPI_BITMASK_PCIEXP_WAKE_DISABLE        0x4000  /* ACPI 3.0 */
> 
> in aclocal.h
> 
> If we're going to re-name what the original patch did,
> then we should finsh the job and delete the original name, yes?

No, we don't re-name anything.

#define ACPI_BITMASK_PCIEXP_WAKE_STATUS         0x4000  /* ACPI 3.0 */
#define ACPI_BITMASK_PCIEXP_WAKE_DISABLE        0x4000  /* ACPI 3.0 */

These are the bit positions for the status and enable registers.

Colin's original patch mis-used ACPI_BITMASK_PCIEXP_WAKE_DISABLE which
should be ACPI_BITMASK_PCIEXP_WAKE_STATUS, but it didn't cause any
problem because they have the same value 0x4000.

Bob has corrected it, see comment #1 at
http://www.acpica.org/bugzilla/show_bug.cgi?id=880

Lin Ming

> 
> thanks,
> Len Brown, Intel Open Source Technology Center
>