Re: [PATCH 2/4] Bluetooth: hci_qca: Don't vote for specific voltage

[Balakrishna Godavarthi <bgodavar@codeaurora.org>]

Hi Matthias,

On 2019-10-22 22:45, Matthias Kaehlcke wrote:
> On Tue, Oct 22, 2019 at 11:35:43AM +0530, Balakrishna Godavarthi wrote:
>> Hi Matthias, Bjorn andresson,
>> 
>> On 2019-10-21 12:07, Harish Bandi wrote:
>> > + Bala
>> >
>> > On 2019-10-18 23:52, Matthias Kaehlcke wrote:
>> > > On Thu, Oct 17, 2019 at 10:24:02PM -0700, Bjorn Andersson wrote:
>> > > > Devices with specific voltage requirements should not request voltage
>> > > > from the driver, but instead rely on the system configuration to
>> > > > define
>> > > > appropriate voltages for each rail.
>> > > >
>> > > > This ensures that PMIC and board variations are accounted for,
>> > > > something
>> > > > that the 0.1V range in the hci_qca driver currently tries to address.
>> > > > But on the Lenovo Yoga C630 (with wcn3990) vddch0 is 3.1V, which
>> > > > means
>> > > > the driver will fail to set the voltage.
>> > > >
>> > > > Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
>> > > > ---
>> > > >  drivers/bluetooth/hci_qca.c | 26 ++++++++------------------
>> > > >  1 file changed, 8 insertions(+), 18 deletions(-)
>> > > >
>> > > > diff --git a/drivers/bluetooth/hci_qca.c
>> > > > b/drivers/bluetooth/hci_qca.c
>> > > > index c07c529b0d81..54aafcc69d06 100644
>> > > > --- a/drivers/bluetooth/hci_qca.c
>> > > > +++ b/drivers/bluetooth/hci_qca.c
>> > > > @@ -130,8 +130,6 @@ enum qca_speed_type {
>> > > >   */
>> > > >  struct qca_vreg {
>> > > >  	const char *name;
>> > > > -	unsigned int min_uV;
>> > > > -	unsigned int max_uV;
>> > > >  	unsigned int load_uA;
>> > > >  };
>> > > >
>> > > > @@ -1332,10 +1330,10 @@ static const struct hci_uart_proto
>> > > > qca_proto = {
>> > > >  static const struct qca_vreg_data qca_soc_data_wcn3990 = {
>> > > >  	.soc_type = QCA_WCN3990,
>> > > >  	.vregs = (struct qca_vreg []) {
>> > > > -		{ "vddio",   1800000, 1900000,  15000  },
>> > > > -		{ "vddxo",   1800000, 1900000,  80000  },
>> > > > -		{ "vddrf",   1300000, 1350000,  300000 },
>> > > > -		{ "vddch0",  3300000, 3400000,  450000 },
>> > > > +		{ "vddio", 15000  },
>> > > > +		{ "vddxo", 80000  },
>> > > > +		{ "vddrf", 300000 },
>> > > > +		{ "vddch0", 450000 },
>> > > >  	},
>> > > >  	.num_vregs = 4,
>> > > >  };
>> > > > @@ -1343,10 +1341,10 @@ static const struct qca_vreg_data
>> > > > qca_soc_data_wcn3990 = {
>> > > >  static const struct qca_vreg_data qca_soc_data_wcn3998 = {
>> > > >  	.soc_type = QCA_WCN3998,
>> > > >  	.vregs = (struct qca_vreg []) {
>> > > > -		{ "vddio",   1800000, 1900000,  10000  },
>> > > > -		{ "vddxo",   1800000, 1900000,  80000  },
>> > > > -		{ "vddrf",   1300000, 1352000,  300000 },
>> > > > -		{ "vddch0",  3300000, 3300000,  450000 },
>> > > > +		{ "vddio", 10000  },
>> > > > +		{ "vddxo", 80000  },
>> > > > +		{ "vddrf", 300000 },
>> > > > +		{ "vddch0", 450000 },
>> > > >  	},
>> > > >  	.num_vregs = 4,
>> > > >  };
>> > > > @@ -1386,13 +1384,6 @@ static int qca_power_off(struct hci_dev *hdev)
>> > > >  static int qca_enable_regulator(struct qca_vreg vregs,
>> > > >  				struct regulator *regulator)
>> > > >  {
>> > > > -	int ret;
>> > > > -
>> > > > -	ret = regulator_set_voltage(regulator, vregs.min_uV,
>> > > > -				    vregs.max_uV);
>> > > > -	if (ret)
>> > > > -		return ret;
>> > > > -
>> > > >  	return regulator_enable(regulator);
>> > > >
>> > > >  }
>> > > > @@ -1401,7 +1392,6 @@ static void qca_disable_regulator(struct
>> > > > qca_vreg vregs,
>> > > >  				  struct regulator *regulator)
>> > > >  {
>> > > >  	regulator_disable(regulator);
>> > > > -	regulator_set_voltage(regulator, 0, vregs.max_uV);
>> > > >
>> > > >  }
>> > >
>> > > This was brought up multiple times during the initial review, but
>> > > wasn't addressed.
>> > >
>> > > Reviewed-by: Matthias Kaehlcke <mka@chromium.org>
>> 
>> 
>> yes true PMIC dts regulator should do this.
>> But we have some real time issues observed.
>> 
>> Issue 1:
>> 
>> In PMIC dts node, ASAIK we have three levels of voltages.
>> 
>> 1. Default voltage.
>> 2. Minimum voltage. (mandatory entry)
>> 3. Maximum voltage. (mandatory entry)
>> 
>> Let us assume that the if PMIC regulator dts node supports  defaults 
>> voltage
>> to 3.2 Volts and Min  as 3.1 V and max as 3.3V
>> So default operating voltage is 3.1V  when we turn on BT (but 
>> according to
>> spec SoC requires min of 3.3V to operate,
>> Might have some functionality failures on end to end testing
> 
> The PMIC regulator shouldn't be configured with the entire range of 
> voltages
> it can generate, but with a range of voltages that is suitable for all 
> its
> consumers.
> 
> In other words if BT requires a minimum voltage of 3.3V the minimum 
> voltage
> of the regulator should be at least 3.3V.
> 
>> Issue 2:
>> 
>> WCN3990 RF is shared with WiFi, so it also try to turn on the 
>> regulators.
>> Wifi driver uses the same approach of restricting to min and max 
>> voltages in
>> driver.
>> Let us assume we turned ON BT and CH0 is set to 3.1v (as in your 
>> case), Wifi
>> is tuned on now, as its request the CH0 to operate at 3.3 Volts, 
>> regulator
>> will fail to set this voltage as BT is operating
>> at CH0 3.1v (assuming max voltage is 3.2v)
>> https://git.kernel.org/pub/scm/linux/kernel/git/bluetooth/bluetooth-next.git/tree/drivers/net/wireless/ath/ath10k/snoc.c#n39
> 
> see above
> 
>> Issue 3:
>> 
>> By mistake PMIC has low min or default voltage and high max voltages, 
>> that
>> is harm for WNC3990.
>> 
>> I would suggest to restrict the min and max voltages in driver, 
>> instead of
>> relaying on PMIC to do this.
>> BTW pmic will do this and doing it in our driver is safer.
> 
> What if another device switches the regulator on before BT?
> 
> Again, what you describe is a misconfiguration of the regulator and 
> should
> be fixed at its root, instead of implementing unreliable 'safeguards' 
> in each
> and every driver using regulators.

Thanks for detail analysis :)
-- 
Regards
Balakrishna.