[0/3] cpufreq: dt: Allow platforms to provide suspend/resume hooks

Message ID cover.1524549644.git.viresh.kumar@linaro.org
Headers show
Series
  • cpufreq: dt: Allow platforms to provide suspend/resume hooks
Related show

Message

Viresh Kumar April 24, 2018, 6:07 a.m.
Hi Miquel,

This is in response to the patch[1] you sent.

I have updated the cpufreq-dt driver to allow platform specific
suspend/resume hooks and done some minor cleanup in the driver. I have
updated your patch and applied that on top of all this. I haven't tested
any of this and would need your help to get that done.

Thanks.

--
viresh

[1] https://lkml.kernel.org/r/20180421141943.25705-1-miquel.raynal@bootlin.com

Miquel Raynal (1):
  cpufreq: add suspend/resume support in Armada 37xx DVFS driver

Viresh Kumar (2):
  cpufreq: dt: Allow platform specific suspend/resume callbacks
  cpufreq: armada: Free resources on error paths

 drivers/cpufreq/armada-37xx-cpufreq.c | 107 +++++++++++++++++++++++++++++-----
 drivers/cpufreq/cpufreq-dt.c          |  10 +++-
 drivers/cpufreq/cpufreq-dt.h          |   5 ++
 3 files changed, 104 insertions(+), 18 deletions(-)

-- 
2.15.0.194.g9af6a3dea062

Comments

Miquel Raynal April 24, 2018, 9:08 a.m. | #1
Hi Viresh,

On Tue, 24 Apr 2018 11:37:33 +0530, Viresh Kumar
<viresh.kumar@linaro.org> wrote:

> Hi Miquel,

> 

> This is in response to the patch[1] you sent.


Thank you very much for taking the time to look into it.

> 

> I have updated the cpufreq-dt driver to allow platform specific

> suspend/resume hooks and done some minor cleanup in the driver. I have

> updated your patch and applied that on top of all this. I haven't tested

> any of this and would need your help to get that done.


I tested on an EspressoBin (Armada 3720), it works.

However, I first had a panic due to the clock being 'put' twice. I
suggest you to merge this [1] into the second patch. With this (or
something similar):

Tested-by: Miquel Raynal <miquel.raynal@bootlin.com>


And the path error was reached because I did not select the cpufreq-dt
driver in menuconfig, which I think should be selected by default when
enabling the armada37xx-cpufreq driver. I wrote a small patch for
that too, see [2]. Tell me if you want me to send it of if you want to
include it in the next version ?

[1] http://code.bulix.org/auzdqa-323646
[2] http://code.bulix.org/e1yk5h-323647

Thank you,
Miquèl


> 

> Thanks.

> 

> --

> viresh

> 

> [1] https://lkml.kernel.org/r/20180421141943.25705-1-miquel.raynal@bootlin.com

> 

> Miquel Raynal (1):

>   cpufreq: add suspend/resume support in Armada 37xx DVFS driver

> 

> Viresh Kumar (2):

>   cpufreq: dt: Allow platform specific suspend/resume callbacks

>   cpufreq: armada: Free resources on error paths

> 

>  drivers/cpufreq/armada-37xx-cpufreq.c | 107 +++++++++++++++++++++++++++++-----

>  drivers/cpufreq/cpufreq-dt.c          |  10 +++-

>  drivers/cpufreq/cpufreq-dt.h          |   5 ++

>  3 files changed, 104 insertions(+), 18 deletions(-)

> 




-- 
Miquel Raynal, Bootlin (formerly Free Electrons)
Embedded Linux and Kernel engineering
https://bootlin.com
Viresh Kumar April 24, 2018, 9:17 a.m. | #2
On 24-04-18, 11:08, Miquel Raynal wrote:
> I tested on an EspressoBin (Armada 3720), it works.


Great.

> However, I first had a panic due to the clock being 'put' twice. I

> suggest you to merge this [1] into the second patch. With this (or

> something similar):


NULL is a valid clock. I have fixed it now, a bit differently though.

> Tested-by: Miquel Raynal <miquel.raynal@bootlin.com>


Thanks.

> And the path error was reached because I did not select the cpufreq-dt

> driver in menuconfig, which I think should be selected by default when

> enabling the armada37xx-cpufreq driver. I wrote a small patch for

> that too, see [2]. Tell me if you want me to send it of if you want to

> include it in the next version ?


Maybe just send that separately (and you don't need to resend any of
the patches which are part of this series).

-- 
viresh