-- Antoniu Miclăuş > -----Original Message----- > From: Jonathan Cameron <Jonathan.Cameron@xxxxxxxxxx> > Sent: Friday, November 5, 2021 7:10 PM > To: Miclaus, Antoniu <Antoniu.Miclaus@xxxxxxxxxx> > Cc: Jonathan Cameron <jic23@xxxxxxxxxx>; Rob Herring <robh@xxxxxxxxxx>; > linux-iio@xxxxxxxxxxxxxxx; devicetree@xxxxxxxxxxxxxxx; linux- > kernel@xxxxxxxxxxxxxxx; Sa, Nuno <Nuno.Sa@xxxxxxxxxx> > Subject: Re: [PATCH v3 2/2] dt-bindings: iio: frequency: add admv1013 doc > > [External] > > On Fri, 5 Nov 2021 08:38:42 +0000 > "Miclaus, Antoniu" <Antoniu.Miclaus@xxxxxxxxxx> wrote: > > > > -----Original Message----- > > > From: Jonathan Cameron <jic23@xxxxxxxxxx> > > > Sent: Thursday, November 4, 2021 8:27 PM > > > To: Miclaus, Antoniu <Antoniu.Miclaus@xxxxxxxxxx> > > > Cc: Rob Herring <robh@xxxxxxxxxx>; linux-iio@xxxxxxxxxxxxxxx; > > > devicetree@xxxxxxxxxxxxxxx; linux-kernel@xxxxxxxxxxxxxxx; Sa, Nuno > > > <Nuno.Sa@xxxxxxxxxx> > > > Subject: Re: [PATCH v3 2/2] dt-bindings: iio: frequency: add admv1013 > doc > > > > > > [External] > > > > > > On Wed, 3 Nov 2021 14:30:56 +0000 > > > "Miclaus, Antoniu" <Antoniu.Miclaus@xxxxxxxxxx> wrote: > > > > > > > Example: > > > > In the setup that we tested the driver, we had a clock chip that was > > > hardware-routed only to the positive side of the local oscillator input (LOP > > > pin) from admv1013. > > > > > > > > Therefore, I think keeping the property in the DT might be useful. > > > > > > I think Rob's question was more general than that one property... See > below. > > > > > > > > Regards, > > > > -- > > > > Antoniu Miclăuş > > > > > > > > > -----Original Message----- > > > > > From: Miclaus, Antoniu > > > > > Sent: Wednesday, November 3, 2021 10:09 AM > > > > > To: Rob Herring <robh@xxxxxxxxxx> > > > > > Cc: jic23@xxxxxxxxxx; linux-iio@xxxxxxxxxxxxxxx; > > > devicetree@xxxxxxxxxxxxxxx; > > > > > linux-kernel@xxxxxxxxxxxxxxx; Sa, Nuno <Nuno.Sa@xxxxxxxxxx> > > > > > Subject: RE: [PATCH v3 2/2] dt-bindings: iio: frequency: add admv1013 > > > doc > > > > > > > > > > Hello Rob, > > > > > > > > > > These properties are fixed and available in the datasheet (binary > format): > > > > > https://www.analog.com/media/en/technical-documentation/data- > > > > > sheets/ADMV1013.pdf > > > > > > > > > > Please see Page 37 of 39, Table 15, QUAD_SE_MODE. > > > > > > > > > > Regards, > > > > > -- > > > > > Antoniu Miclăuş > > > > > > > > > > > -----Original Message----- > > > > > > From: Rob Herring <robh@xxxxxxxxxx> > > > > > > Sent: Tuesday, November 2, 2021 7:51 PM > > > > > > To: Miclaus, Antoniu <Antoniu.Miclaus@xxxxxxxxxx> > > > > > > Cc: jic23@xxxxxxxxxx; linux-iio@xxxxxxxxxxxxxxx; > > > > > devicetree@xxxxxxxxxxxxxxx; > > > > > > linux-kernel@xxxxxxxxxxxxxxx; Sa, Nuno <Nuno.Sa@xxxxxxxxxx> > > > > > > Subject: Re: [PATCH v3 2/2] dt-bindings: iio: frequency: add > admv1013 > > > doc > > > > > > > > > > > > [External] > > > > > > > > > > > > On Mon, Nov 01, 2021 at 12:04:20PM +0200, Antoniu Miclaus wrote: > > > > > > > Add device tree bindings for the ADMV1013 Upconverter. > > > > > > > > > > > > > > Signed-off-by: Antoniu Miclaus <antoniu.miclaus@xxxxxxxxxx> > > > > > > > --- > > > > > > > .../bindings/iio/frequency/adi,admv1013.yaml | 119 > > > > > > ++++++++++++++++++ > > > > > > > 1 file changed, 119 insertions(+) > > > > > > > create mode 100644 > > > > > > > Documentation/devicetree/bindings/iio/frequency/adi,admv1013.yaml > > > > > > > > > > > > > > diff --git > > > > > > > > > > a/Documentation/devicetree/bindings/iio/frequency/adi,admv1013.yaml > > > > > > > > > > b/Documentation/devicetree/bindings/iio/frequency/adi,admv1013.yaml > > > > > > > new file mode 100644 > > > > > > > index 000000000000..47993253a586 > > > > > > > --- /dev/null > > > > > > > +++ > > > > > > > > > > b/Documentation/devicetree/bindings/iio/frequency/adi,admv1013.yaml > > > > > > > @@ -0,0 +1,119 @@ > > > > > > > +# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause) > > > > > > > +%YAML 1.2 > > > > > > > +--- > > > > > > > +$id: > > > > > > > > > > > > > > > https://urldefense.com/v3/__http://devicetree.org/schemas/iio/frequency > > > > > > > > > > > > > > > /adi,admv1013.yaml*__;Iw!!A3Ni8CS0y2Y!uTDPalOgj6YS_vZ6bsDSbA_Qna6Q > > > > > > OwMpoRxzo6nn06i5TNuGWZEk9PvtbC6SKQGXrugy$ > > > > > > > +$schema: > > > https://urldefense.com/v3/__http://devicetree.org/meta- > > > > > > > > > > > > > > > schemas/core.yaml*__;Iw!!A3Ni8CS0y2Y!uTDPalOgj6YS_vZ6bsDSbA_Qna6Q > > > > > > OwMpoRxzo6nn06i5TNuGWZEk9PvtbC6SKYugV1fM$ > > > > > > > + > > > > > > > +title: ADMV1013 Microwave Upconverter > > > > > > > + > > > > > > > +maintainers: > > > > > > > + - Antoniu Miclaus <antoniu.miclaus@xxxxxxxxxx> > > > > > > > + > > > > > > > +description: | > > > > > > > + Wideband, microwave upconverter optimized for point to point > > > > > > microwave > > > > > > > + radio designs operating in the 24 GHz to 44 GHz frequency > range. > > > > > > > + > > > > > > > + https://www.analog.com/en/products/admv1013.html > > > > > > > + > > > > > > > +properties: > > > > > > > + compatible: > > > > > > > + enum: > > > > > > > + - adi,admv1013 > > > > > > > + > > > > > > > + reg: > > > > > > > + maxItems: 1 > > > > > > > + > > > > > > > + spi-max-frequency: > > > > > > > + maximum: 1000000 > > > > > > > + > > > > > > > + clocks: > > > > > > > + description: > > > > > > > + Definition of the external clock. > > > > > > > + minItems: 1 > > > > > > > + > > > > > > > + clock-names: > > > > > > > + items: > > > > > > > + - const: lo_in > > > > > > > + > > > > > > > + clock-output-names: > > > > > > > + maxItems: 1 > > > > > > > + > > > > > > > + vcm-supply: > > > > > > > + description: > > > > > > > + Analog voltage regulator. > > > > > > > + > > > > > > > + adi,vga-powerdown: > > > > > > > + description: > > > > > > > + Power Down the Voltage Gain Amplifier Circuit available at > > > > > > > + BG_RBIAS2 pin. > > > > > > > + type: boolean > > > > > > What wiring would make it sensible to always have this powered down? > > > If we can describe that rather than vga-powerdown then that is what > should > > > be in the binding. If there isn't any wiring based justification and this > > > is just turning off part of the device, then it should not be in the binding. > > > > > This part is similar to ADRF6780 that has also an Enable Register that can > power down circuit blocks within the part: > > > https://urldefense.com/v3/__https://git.kernel.org/pub/scm/linux/kernel/g > it/jic23/iio.git/tree/drivers/iio/frequency/adrf6780.c?h=togreg__;!!A3Ni8CS0 > y2Y!sZfM9jaILdXhyFOXGx3KQHQOxNJaKopL7sEw2F- > xzEWT0xHdoIW5sUEYE9ezdwHxPPWy$ > > > > Hmm. Guess we weren't super awake with that one :( > > > I preferred to expose these properties in the device tree, rather than > creating custom device attributes in the driver. > > > > But I guess these properties can be managed also only from debug fs. > > The ideal situation is to control these based on what the device is being used > for > and not provide explicit control to userspace or via dt. > > So if it doesn't make sense to power up a unit because it isn't wired to > anything > then express the wiring in DT. All these powerdown properties have individual supply pins assigned (VCC_MIXER, VCC_BG, VCC_QUAD... datasheet - page 8 of 39). So I guess we can consider to enable/disable these blocks based on the wiring. Looking forward to your feedback. > > If it makes sense to power down a unit because it isn't being used currently > due to what features are in use, then power it down automatically. Runtime > pm > handles a lot of these cases by letting you do autopowerdown if not used > after > a certain period. That way, we don't end up powering up and down rapidly > but > only do it if we have go reason to believe it is worth powering down. > > Could we use either of these approaches here? > > Jonathan > > > > I can adapt the code based on your decision. Looking forward to your > feedback. > > > > > > > + > > > > > > > + adi,mixer-powerdown: > > > > > > > + description: > > > > > > > + Power Down the Mixer Circuit. Enable to put the block in > > > > > > > + a power down state. > > > > > > Same for all these other power downs. > > > > > > > > > > + type: boolean > > > > > > > + > > > > > > > + adi,quad-powerdown: > > > > > > > + description: > > > > > > > + Power Down the Quadrupler. Enable to put the block in > > > > > > > + a power down state. > > > > > > > + type: boolean > > > > > > > + > > > > > > > + adi,bg-powerdown: > > > > > > > + description: > > > > > > > + Power Down the Transmitter Band Gap. Enable to put the > part in > > > > > > > + a power down state. > > > > > > > + type: boolean > > > > > > > + > > > > > > > + adi,mixer-if-enable: > > > > > > > + description: > > > > > > > + Enable the Intermediate Frequency Mode. Either IF Mode or > I/Q > > > > > Mode > > > > > > > + can be enabled at a time. > > > > > > > + type: boolean > > > > > > > + > > > > > > > + adi,detector-enable: > > > > > > > + description: > > > > > > > + Enable the Envelope Detector available at output pins > VENV_P > > > and > > > > > > > + VENV_N. Disable to reduce power consumption. > > > > > > > + type: boolean > > > > > > > + > > > > > > > + adi,quad-se-mode: > > > > > > > + description: > > > > > > > + Switch the LO path from differential to single-ended > operation. > > > > > > > + Set value 6 for Single-Ended Mode, Negative Side Disabled. > > > > > > > + Set value 9 for Single-Ended Mode, Positive Side Disabled. > > > > > > > + Set value 12 for Differential Mode. > > > > > > > + $ref: /schemas/types.yaml#/definitions/uint32 > > > > > > > + enum: [6, 9, 12] > > > > > > > > > > > > All these vendor properties are fixed based on the board design or > > > > > > something a user may want to change? The latter does not belong > in > > > DT. > > > > > > > > > > > > Rob > >
