Re: [PATCH V4 3/4] block: queue work on unbound wq

[Amit Kucheria]

On Thu, Apr 4, 2013 at 3:24 AM, Tejun Heo <tj@xxxxxxxxxx> wrote:
> Hello, Viresh.
>
> Sorry about the delay.  Lost this one somehow.
>
> On Mon, Apr 01, 2013 at 12:01:05PM +0530, Viresh Kumar wrote:
>> Just wanted to make this clear before writing it:
>>
>> You want me to do something like (With better names):
>>
>> int wq_unbound_for_power_save_enabled = 0;
>>
>> #ifdef CONFIG_WQ_UNBOUND_FOR_POWER_SAVE
>> #define WQ_UNBOUND_FOR_POWER_SAVE   wq_unbound_for_power_save_enabled
>> #else
>> #define WQ_UNBOUND_FOR_POWER_SAVE 0
>>
>> And provide a call to enable/disable wq_unbound_for_power_save_enabled ??
>
> Not a call, probably a module_param() so that it can be switched
> on/off during boot.  You can make the param writable so that it can be
> flipped run-time but updating existing workqueues would be non-trivial
> and I don't think it's gonna be worthwhile.
>
> Thanks!

Does it make sense to collect this sort of power optimisation under a
CONFIG_PM sub-item, say, CONFIG_PM_OPTIMISATIONS?

For people tuning for power, it could be single place to go enable
stuff and for people looking for performance regressions it would be a
single place to make sure nothing is enabled.

/Amit
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/