Re: [PATCH -mm 3/3] PM: Improve handling of ACPI system state indicator (rev. 3)

["Rafael J. Wysocki" <rjw@sisk.pl>]

On Wednesday, 29 August 2007 18:54, Moore, Robert wrote:
> No, it's not safe to run the AML interpreter with interrupts disabled.

OK

> I don't have any problem with introducing finer granularity enter/exit
> sleep interfaces if they are required.
> 
> I would suggest that we rename things a bit however.
> 
> Currently:
> acpi_enter_sleep_state_prep
> acpi_enter_sleep_state_prep_late
> acpi_enter_sleep_state
> 
> acpi_leave_sleep_state_prep
> acpi_leave_sleep_state
> 
> I think we can truncate and clarify:
> 
> acpi_sleep_setup1
> acpi_sleep_setup2
> acpi_sleep
> 
> acpi_wake_setup1
> acpi_wake

That's perfectly fine by me.  I'll update the 2/3 patch to use these names.

Also, I think we can remove acpi_enter_sleep_state_s4bios() entirely (in a
separate patch).

> acpi_set_sleep_state_indicator:
> 
> I'm not sure if we have any external interfaces that simply execute a
> control method, seems like overkill.
> 
> Please give me more information as to why _SSI needs to be moved (other
> than executing it after _BFS)

The _SST after _BFS is okay, but the invocation of _SST in acpi_wake()
(currently acpi_leave_sleep_state) is problematic, since it causes the
indicator to be set to "working" during hibernation, before the image is
saved.  Thus, during hibernation _SST shouldn't be called from acpi_wake().

For this reason, I thought it would be a good idea to call _SST from a separate
routine that might be invoked by higher level functions as desired.

Greetings,
Rafael