Add kernel parameters for debugfs disable_*_ba values

Add kernel parameters to set default values for debugfs disable_*_ba values

This provides a method for setting default values for the debugfs parameters for all phyNs at boot time, while continuing to allow runtime per-phy modifications.  This is needed because debugfs parameters cannot easily be statically pushed onto a phy that may or may not yet be loaded, since an init script would have to mount the debugfs then monitor the directory until the phyN parameter appears.

BUG=none
TEST=rebooted with various config options

[Sam Leffler, 2010-07-02 21:48:11]

LGTM

please update the issue to explain that this is being done mostly because of the
problem where mount'ing debugfs happens asynchronously so you cannot set the
parameters w/o polling for the debugfs files to appear in the namespace

[Olof Johansson, 2010-07-08 02:11:46]

Hi,

Has this been posted for upstream, or will it be lost whenever we move to a
newer compat-wireless or base kernel?

If it's local to chromeos and has to be carried forward manually, it should
ideally be prefixed with CHROMIUM:, to not miss these cases.

Also, adding a Signed-off-by is a good habit.

[Sam Leffler, 2010-07-08 02:15:43]

It's likely to be a local patch for a while.  We've got a handful of
debugfs knobs we use for driver workarounds and for testing that
upstream is unlikely to care about.

-Sam

On Wed, Jul 7, 2010 at 7:11 PM,  <olofj@chromium.org> wrote:
> Hi,
>
> Has this been posted for upstream, or will it be lost whenever we move to a
> newer compat-wireless or base kernel?
>
> If it's local to chromeos and has to be carried forward manually, it should
> ideally be prefixed with CHROMIUM:, to not miss these cases.
>
> Also, adding a Signed-off-by is a good habit.
>
> http://codereview.chromium.org/2887009/show
>

[Paul Stewart, 2010-07-08 20:56:11]

Could you clarify what should happen in order to create this
"CHROMIUM:" prefix?  I'm a relative newbie here.

--
Paul

On Wed, Jul 7, 2010 at 7:11 PM,  <olofj@chromium.org> wrote:
> Hi,
>
> Has this been posted for upstream, or will it be lost whenever we move to a
> newer compat-wireless or base kernel?
>
> If it's local to chromeos and has to be carried forward manually, it should
> ideally be prefixed with CHROMIUM:, to not miss these cases.
>
> Also, adding a Signed-off-by is a good habit.
>
> http://codereview.chromium.org/2887009/show
>

[Mandeep Singh Baines, 2010-07-08 23:16:28]

Paul Stewart (pstew@chromium.org) wrote:
> Could you clarify what should happen in order to create this
> "CHROMIUM:" prefix?  I'm a relative newbie here.
> 

http://dev.chromium.org/chromium-os/chromiumos-design-docs/chromium-os-kernel

> --
> Paul
> 
> On Wed, Jul 7, 2010 at 7:11 PM,  <olofj@chromium.org> wrote:
> > Hi,
> >
> > Has this been posted for upstream, or will it be lost whenever we move to a
> > newer compat-wireless or base kernel?
> >
> > If it's local to chromeos and has to be carried forward manually, it should
> > ideally be prefixed with CHROMIUM:, to not miss these cases.
> >
> > Also, adding a Signed-off-by is a good habit.
> >
> > http://codereview.chromium.org/2887009/show
> >

[pstew1, 2010-07-09 00:00:53]

Thanks!

On Thu, Jul 8, 2010 at 4:16 PM, Mandeep Singh Baines <msb@chromium.org> wrote:
> Paul Stewart (pstew@chromium.org) wrote:
>> Could you clarify what should happen in order to create this
>> "CHROMIUM:" prefix?  I'm a relative newbie here.
>>
>
> http://dev.chromium.org/chromium-os/chromiumos-design-docs/chromium-os-kernel
>
>> --
>> Paul
>>
>> On Wed, Jul 7, 2010 at 7:11 PM,  <olofj@chromium.org> wrote:
>> > Hi,
>> >
>> > Has this been posted for upstream, or will it be lost whenever we move to a
>> > newer compat-wireless or base kernel?
>> >
>> > If it's local to chromeos and has to be carried forward manually, it should
>> > ideally be prefixed with CHROMIUM:, to not miss these cases.
>> >
>> > Also, adding a Signed-off-by is a good habit.
>> >
>> > http://codereview.chromium.org/2887009/show
>> >
>