On Wednesday, November 26, 2014 01:54:31 PM Eduardo Valentin wrote:
--UugvWAfsgieZRqgk Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable
Hello Viresh,
Thanks for providing a proposal.
On Wed, Nov 26, 2014 at 11:22:55AM +0530, Viresh Kumar wrote:
Hi Rafael/Eduardo, =20 Currently there is no callback for cpufreq drivers which is called once t=
he
policy is ready to be used. There are some requirements where such a call=
back is
required. =20 One of them is registering a cooling device with the help of of_cpufreq_cooling_register(). This routine tries to get 'struct cpufreq_=
policy'
for CPUs which isn't yet initialed at the time ->init() is called and so =
we face
issues while registering the cooling device. =20 Because we can't register cooling device from ->init(), we need a callbac=
k that
is called after the policy is ready to be used and hence ->usable() callb=
ack.
=20 The first patch fixes few formatting issues, so that the third patch does=
n't
throw any checkpatch warnings. Second one fixes a potential bug in cpufre=
q-dt
driver. Third one introduces ->usable() callback which will be used in the fourth patch. =20 Last three are fixes for cooling core, which may be applied separately by Eduardo if he wants. Sent them in this series as they were sort of connec=
ted
with cpufreq in general. =20 Let me know if it still doesn't work properly.
For the series, the last three patches somehow breaks things. I didn't not investigate the reason now, because, well, I think we should take one thing at a time.
Agreed!