Hi,
On 7/21/2017 10:54 PM, Stephen Boyd wrote:
> On 07/21/2017 04:01 AM, Manu Gautam wrote:
>> Driver can turn off clocks during runtime suspend.
>> Also, runtime suspend is not as a result of host mode
>> selective suspend then PHY can be powered off as well.
>>
>> Signed-off-by: Manu Gautam <mgau...@codeaurora.org>
>>
>> diff --git a/drivers/phy/qualcomm/phy-qcom-qusb2.c
>> b/drivers/phy/qualcomm/phy-qcom-qusb2.c
>> index fa60a99..b505681 100644
>> --- a/drivers/phy/qualcomm/phy-qcom-qusb2.c
>> +++ b/drivers/phy/qualcomm/phy-qcom-qusb2.c
>> @@ -132,6 +132,9 @@ struct qusb2_phy {
>>
>> const struct qusb2_phy_cfg *cfg;
>> bool has_se_clk_scheme;
>> + bool phy_initialized;
>> + bool powered_on;
> Is the powered_on flag here because the controller driver has unbalanced
> power on calls to the phy? Same comment applies for the phy_initialized
> flag. Both of these look like workarounds for some odd behavior in the
> controller driver.
phy_initialized flag is to make runtime_suspend/resume no-ops
until PHY gets initialized.
powered_on flag is not related to any issue issue in core (as of now).
I just added that to bail out early from phy power_on/off which is now
called from phy_init/exit and runtime_suspend/resume as well.
