Working on a battery charger and fuel gauge driver it happened to
have the need to read from VADC_LR_MUX2_BAT_ID channel. Here is
the declaration from downstream msm8916.dtsi:
chan@31 {
label = "batt_id";
reg = <0x31>;
qcom,decimation = <0>;
qcom,pre-div-channel-scaling = <0>;
qcom,calibration-type = "ratiometric";
qcom,scale-function = <0>;
qcom,hw-settle-time = <0xb>;
qcom,fast-avg-setup = <0>;
};
Those two patches set channel scaling accordingly and add it to
pm8916 device tree.
Jonathan Albrieux (2):
iio:adc:qcom-spmi-vadc: add default scale to LR_MUX2_BAT_ID channel
arm64: dts: qcom: pm8916: Add batt_id channel node
arch/arm64/boot/dts/qcom/pm8916.dtsi | 5 +++++
drivers/iio/adc/qcom-spmi-vadc.c | 2 +-
2 files changed, 6 insertions(+), 1 deletion(-)
--
2.17.1
Checking at both msm8909-pm8916.dtsi and msm8916.dtsi from downstream
it is indicated that "batt_id" channel has to be scaled with the default
function:
chan@31 {
label = "batt_id";
reg = <0x31>;
qcom,decimation = <0>;
qcom,pre-div-channel-scaling = <0>;
qcom,calibration-type = "ratiometric";
qcom,scale-function = <0>;
qcom,hw-settle-time = <0xb>;
qcom,fast-avg-setup = <0>;
};
Change LR_MUX2_BAT_ID scaling accordingly.
Signed-off-by: Jonathan Albrieux <[email protected]>
---
drivers/iio/adc/qcom-spmi-vadc.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/iio/adc/qcom-spmi-vadc.c b/drivers/iio/adc/qcom-spmi-vadc.c
index b0388f8a69f4..7e7d408452ec 100644
--- a/drivers/iio/adc/qcom-spmi-vadc.c
+++ b/drivers/iio/adc/qcom-spmi-vadc.c
@@ -598,7 +598,7 @@ static const struct vadc_channels vadc_chans[] = {
VADC_CHAN_NO_SCALE(P_MUX16_1_3, 1)
VADC_CHAN_NO_SCALE(LR_MUX1_BAT_THERM, 0)
- VADC_CHAN_NO_SCALE(LR_MUX2_BAT_ID, 0)
+ VADC_CHAN_VOLT(LR_MUX2_BAT_ID, 0, SCALE_DEFAULT)
VADC_CHAN_NO_SCALE(LR_MUX3_XO_THERM, 0)
VADC_CHAN_NO_SCALE(LR_MUX4_AMUX_THM1, 0)
VADC_CHAN_NO_SCALE(LR_MUX5_AMUX_THM2, 0)
--
2.17.1
On Wed 13 Jan 09:18 CST 2021, Jonathan Albrieux wrote:
> Checking at both msm8909-pm8916.dtsi and msm8916.dtsi from downstream
> it is indicated that "batt_id" channel has to be scaled with the default
> function:
>
> chan@31 {
> label = "batt_id";
> reg = <0x31>;
> qcom,decimation = <0>;
> qcom,pre-div-channel-scaling = <0>;
> qcom,calibration-type = "ratiometric";
> qcom,scale-function = <0>;
> qcom,hw-settle-time = <0xb>;
> qcom,fast-avg-setup = <0>;
> };
>
> Change LR_MUX2_BAT_ID scaling accordingly.
>
Acked-by: Bjorn Andersson <[email protected]>
Not entirely sure, but looking at the history I think this used to work
- but it's obvious that no one has read this channel for a while...
But I think below is a regression and should be mentioned:
Fixes: 7c271eea7b8a ("iio: adc: spmi-vadc: Changes to support different scaling")
> Signed-off-by: Jonathan Albrieux <[email protected]>
Jonathan Cameron, if you merge this through your tree I can take the dts
addition through the Qualcomm tree.
Regards,
Bjorn
> ---
> drivers/iio/adc/qcom-spmi-vadc.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/drivers/iio/adc/qcom-spmi-vadc.c b/drivers/iio/adc/qcom-spmi-vadc.c
> index b0388f8a69f4..7e7d408452ec 100644
> --- a/drivers/iio/adc/qcom-spmi-vadc.c
> +++ b/drivers/iio/adc/qcom-spmi-vadc.c
> @@ -598,7 +598,7 @@ static const struct vadc_channels vadc_chans[] = {
> VADC_CHAN_NO_SCALE(P_MUX16_1_3, 1)
>
> VADC_CHAN_NO_SCALE(LR_MUX1_BAT_THERM, 0)
> - VADC_CHAN_NO_SCALE(LR_MUX2_BAT_ID, 0)
> + VADC_CHAN_VOLT(LR_MUX2_BAT_ID, 0, SCALE_DEFAULT)
> VADC_CHAN_NO_SCALE(LR_MUX3_XO_THERM, 0)
> VADC_CHAN_NO_SCALE(LR_MUX4_AMUX_THM1, 0)
> VADC_CHAN_NO_SCALE(LR_MUX5_AMUX_THM2, 0)
> --
> 2.17.1
>
On Fri, 15 Jan 2021 10:30:49 -0600
Bjorn Andersson <[email protected]> wrote:
> On Wed 13 Jan 09:18 CST 2021, Jonathan Albrieux wrote:
>
> > Checking at both msm8909-pm8916.dtsi and msm8916.dtsi from downstream
> > it is indicated that "batt_id" channel has to be scaled with the default
> > function:
> >
> > chan@31 {
> > label = "batt_id";
> > reg = <0x31>;
> > qcom,decimation = <0>;
> > qcom,pre-div-channel-scaling = <0>;
> > qcom,calibration-type = "ratiometric";
> > qcom,scale-function = <0>;
> > qcom,hw-settle-time = <0xb>;
> > qcom,fast-avg-setup = <0>;
> > };
> >
> > Change LR_MUX2_BAT_ID scaling accordingly.
> >
>
> Acked-by: Bjorn Andersson <[email protected]>
>
> Not entirely sure, but looking at the history I think this used to work
> - but it's obvious that no one has read this channel for a while...
>
> But I think below is a regression and should be mentioned:
>
> Fixes: 7c271eea7b8a ("iio: adc: spmi-vadc: Changes to support different scaling")
>
Yikes that was a while ago :)
> > Signed-off-by: Jonathan Albrieux <[email protected]>
>
> Jonathan Cameron, if you merge this through your tree I can take the dts
> addition through the Qualcomm tree.
Applied to the fixes-togreg branch of iio.git and marked for stable.
I'm not going to rush this one given age of the bug, but if I happen to
have anything else going it'll make it before the end of this cycle.
Thanks,
Jonathan
>
> Regards,
> Bjorn
>
> > ---
> > drivers/iio/adc/qcom-spmi-vadc.c | 2 +-
> > 1 file changed, 1 insertion(+), 1 deletion(-)
> >
> > diff --git a/drivers/iio/adc/qcom-spmi-vadc.c b/drivers/iio/adc/qcom-spmi-vadc.c
> > index b0388f8a69f4..7e7d408452ec 100644
> > --- a/drivers/iio/adc/qcom-spmi-vadc.c
> > +++ b/drivers/iio/adc/qcom-spmi-vadc.c
> > @@ -598,7 +598,7 @@ static const struct vadc_channels vadc_chans[] = {
> > VADC_CHAN_NO_SCALE(P_MUX16_1_3, 1)
> >
> > VADC_CHAN_NO_SCALE(LR_MUX1_BAT_THERM, 0)
> > - VADC_CHAN_NO_SCALE(LR_MUX2_BAT_ID, 0)
> > + VADC_CHAN_VOLT(LR_MUX2_BAT_ID, 0, SCALE_DEFAULT)
> > VADC_CHAN_NO_SCALE(LR_MUX3_XO_THERM, 0)
> > VADC_CHAN_NO_SCALE(LR_MUX4_AMUX_THM1, 0)
> > VADC_CHAN_NO_SCALE(LR_MUX5_AMUX_THM2, 0)
> > --
> > 2.17.1
> >
On Sat, Jan 16, 2021 at 05:47:35PM +0000, Jonathan Cameron wrote:
> On Fri, 15 Jan 2021 10:30:49 -0600
> Bjorn Andersson <[email protected]> wrote:
>
> > On Wed 13 Jan 09:18 CST 2021, Jonathan Albrieux wrote:
> >
> > > Checking at both msm8909-pm8916.dtsi and msm8916.dtsi from downstream
> > > it is indicated that "batt_id" channel has to be scaled with the default
> > > function:
> > >
> > > chan@31 {
> > > label = "batt_id";
> > > reg = <0x31>;
> > > qcom,decimation = <0>;
> > > qcom,pre-div-channel-scaling = <0>;
> > > qcom,calibration-type = "ratiometric";
> > > qcom,scale-function = <0>;
> > > qcom,hw-settle-time = <0xb>;
> > > qcom,fast-avg-setup = <0>;
> > > };
> > >
> > > Change LR_MUX2_BAT_ID scaling accordingly.
> > >
> >
> > Acked-by: Bjorn Andersson <[email protected]>
> >
> > Not entirely sure, but looking at the history I think this used to work
> > - but it's obvious that no one has read this channel for a while...
> >
> > But I think below is a regression and should be mentioned:
> >
> > Fixes: 7c271eea7b8a ("iio: adc: spmi-vadc: Changes to support different scaling")
> >
>
> Yikes that was a while ago :)
>
> > > Signed-off-by: Jonathan Albrieux <[email protected]>
> >
> > Jonathan Cameron, if you merge this through your tree I can take the dts
> > addition through the Qualcomm tree.
>
> Applied to the fixes-togreg branch of iio.git and marked for stable.
>
> I'm not going to rush this one given age of the bug, but if I happen to
> have anything else going it'll make it before the end of this cycle.
>
> Thanks,
>
> Jonathan
>
> >
> > Regards,
> > Bjorn
> >
> > > ---
> > > drivers/iio/adc/qcom-spmi-vadc.c | 2 +-
> > > 1 file changed, 1 insertion(+), 1 deletion(-)
> > >
> > > diff --git a/drivers/iio/adc/qcom-spmi-vadc.c b/drivers/iio/adc/qcom-spmi-vadc.c
> > > index b0388f8a69f4..7e7d408452ec 100644
> > > --- a/drivers/iio/adc/qcom-spmi-vadc.c
> > > +++ b/drivers/iio/adc/qcom-spmi-vadc.c
> > > @@ -598,7 +598,7 @@ static const struct vadc_channels vadc_chans[] = {
> > > VADC_CHAN_NO_SCALE(P_MUX16_1_3, 1)
> > >
> > > VADC_CHAN_NO_SCALE(LR_MUX1_BAT_THERM, 0)
> > > - VADC_CHAN_NO_SCALE(LR_MUX2_BAT_ID, 0)
> > > + VADC_CHAN_VOLT(LR_MUX2_BAT_ID, 0, SCALE_DEFAULT)
> > > VADC_CHAN_NO_SCALE(LR_MUX3_XO_THERM, 0)
> > > VADC_CHAN_NO_SCALE(LR_MUX4_AMUX_THM1, 0)
> > > VADC_CHAN_NO_SCALE(LR_MUX5_AMUX_THM2, 0)
> > > --
> > > 2.17.1
> > >
>
Thank you Bjorn, thank you Jonathan,
Regards,
Jonathan