[Pm-utils] [PATCH 5/8] pm-utils 1.2.3 proposed patches

[Victor Lowther]

On Nov 29, 2008, at 6:28 PM, Robby Workman <rw at rlworkman.net> wrote:

> On Sat, 29 Nov 2008 08:51:29 -0800 (PST)
> Victor Lowther <victor.lowther at gmail.com> wrote:
>
>> 90clock takes over a second to run on suspend.
>>
>> Most systems do not need it.  Add NEED_CLOCK_SYNC, and only run
>> hwclock if it is set.  This should eventually turn into a quirk.
>
>
> Any idea which systems currently *do* need this?

Sorry, I am afraid I only have anecdata. None of my laptops have  
needed this code, so there are my two data points.

>  I'm assuming that the
> proper way to handle this in the future is patches to hal-info, but  
> I'm
> curious for the time being: can we expect a *lot* of bug reports if we
> remove this from our 1.2.x package? IOW, how much of a minority are
> the systems which actually need this hook?

I honestly have no idea. Going back through the history of the hook  
indicates that we have always done it this way, which probably means  
the code was copied without question from an earlier project (like  
acpi-support) and dates form an era when we actually needed that hack.
>
> -RW