| Message ID | 1428515100-1496-1-git-send-email-harald@ccbib.org (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
Hi Harald, [ add devicetree mailinglist ] > Harald Geyer <harald@ccbib.org> hat am 8. April 2015 um 19:45 geschrieben: > > > The imx23-olinuxino board has an i2c interface exposed on UEXT connector. AFAIK the iMX233-OLinuXino-MICRO don't have a UEXT connector. Maybe we reached the point to make dts files for every Olinuxino board. In that case we could also handle the USB Host / Peripherial issue. > This patch provides the necessary devicetree code. > Tested with MOD-LCD1x9 from Olimex. > > This patch is based on work by Fadil Berisha with his permission. However > all bugs are mine. > > Signed-off-by: Harald Geyer <harald@ccbib.org> > --- > arch/arm/boot/dts/imx23-olinuxino.dts | 6 +++++ > arch/arm/boot/dts/imx23.dtsi | 40 ++++++++++++++++++++++++++++++++- > 2 files changed, 45 insertions(+), 1 deletion(-) > > [...] > --- a/arch/arm/boot/dts/imx23.dtsi > +++ b/arch/arm/boot/dts/imx23.dtsi > @@ -308,6 +308,39 @@ > fsl,voltage = <MXS_VOLTAGE_HIGH>; > fsl,pull-up = <MXS_PULL_ENABLE>; > }; > + > + i2c_pins_a: i2c@0 { > + reg = <0>; > + fsl,pinmux-ids = < > + MX23_PAD_I2C_SCL__I2C_SCL > + MX23_PAD_I2C_SDA__I2C_SDA > + >; > + fsl,drive-strength = <MXS_DRIVE_8mA>; > + fsl,voltage = <MXS_VOLTAGE_HIGH>; > + fsl,pull-up = <MXS_PULL_ENABLE>; > + }; > + > + i2c_pins_b: i2c@1 { > + reg = <1>; > + fsl,pinmux-ids = < > + MX23_PAD_LCD_ENABLE__I2C_SCL > + MX23_PAD_LCD_HSYNC__I2C_SDA > + >; > + fsl,drive-strength = <MXS_DRIVE_8mA>; > + fsl,voltage = <MXS_VOLTAGE_HIGH>; > + fsl,pull-up = <MXS_PULL_ENABLE>; > + }; > + > + i2c_pins_c: i2c@2 { > + reg = <2>; > + fsl,pinmux-ids = < > + MX23_PAD_SSP1_DATA1__I2C_SCL > + MX23_PAD_SSP1_DATA2__I2C_SDA > + >; > + fsl,drive-strength = <MXS_DRIVE_8mA>; > + fsl,voltage = <MXS_VOLTAGE_HIGH>; > + fsl,pull-up = <MXS_PULL_ENABLE>; > + }; Please add only i2c_pins_b because this pin muxing is the only used one. Thanks Stefan
Hi Stefan! On Thu, 9 Apr 2015 00:12:55 +0200 (CEST), Stefan Wahren <stefan.wahren@i2se.com> wrote: > Hi Harald, > > [ add devicetree mailinglist ] > >> Harald Geyer <harald@ccbib.org> hat am 8. April 2015 um 19:45 >> geschrieben: >> >> >> The imx23-olinuxino board has an i2c interface exposed on UEXT connector. > > AFAIK the iMX233-OLinuXino-MICRO don't have a UEXT connector. Yes, but I don't think having default i2c pins there would be a bad thing. > Maybe we reached the point to make dts files for every Olinuxino board. > In that case we could also handle the USB Host / Peripherial issue. Yes, I'm thinking in this direction. But even if peripherial mode works on micro/nano (which I haven't been able to confirm yet) that doesn't mean that it should be set in the default device tree. After all Olimex themself advertise the port as host. >> This patch provides the necessary devicetree code. >> Tested with MOD-LCD1x9 from Olimex. >> >> This patch is based on work by Fadil Berisha with his permission. However >> all bugs are mine. >> >> Signed-off-by: Harald Geyer <harald@ccbib.org> >> --- >> arch/arm/boot/dts/imx23-olinuxino.dts | 6 +++++ >> arch/arm/boot/dts/imx23.dtsi | 40 ++++++++++++++++++++++++++++++++- >> 2 files changed, 45 insertions(+), 1 deletion(-) >> >> [...] >> --- a/arch/arm/boot/dts/imx23.dtsi >> +++ b/arch/arm/boot/dts/imx23.dtsi >> @@ -308,6 +308,39 @@ >> fsl,voltage = <MXS_VOLTAGE_HIGH>; >> fsl,pull-up = <MXS_PULL_ENABLE>; >> }; >> + >> + i2c_pins_a: i2c@0 { >> + reg = <0>; >> + fsl,pinmux-ids = < >> + MX23_PAD_I2C_SCL__I2C_SCL >> + MX23_PAD_I2C_SDA__I2C_SDA >> + >; >> + fsl,drive-strength = <MXS_DRIVE_8mA>; >> + fsl,voltage = <MXS_VOLTAGE_HIGH>; >> + fsl,pull-up = <MXS_PULL_ENABLE>; >> + }; >> + >> + i2c_pins_b: i2c@1 { >> + reg = <1>; >> + fsl,pinmux-ids = < >> + MX23_PAD_LCD_ENABLE__I2C_SCL >> + MX23_PAD_LCD_HSYNC__I2C_SDA >> + >; >> + fsl,drive-strength = <MXS_DRIVE_8mA>; >> + fsl,voltage = <MXS_VOLTAGE_HIGH>; >> + fsl,pull-up = <MXS_PULL_ENABLE>; >> + }; >> + >> + i2c_pins_c: i2c@2 { >> + reg = <2>; >> + fsl,pinmux-ids = < >> + MX23_PAD_SSP1_DATA1__I2C_SCL >> + MX23_PAD_SSP1_DATA2__I2C_SDA >> + >; >> + fsl,drive-strength = <MXS_DRIVE_8mA>; >> + fsl,voltage = <MXS_VOLTAGE_HIGH>; >> + fsl,pull-up = <MXS_PULL_ENABLE>; >> + }; > > Please add only i2c_pins_b because this pin muxing is the only used one. I disagree: Having all possible i2c pin groups documented in devicetree is a good thing. Also imx23.dtsi is more general then olinuxino - somebody might produce a imx23 based board where one of the other pin groups makes more sense to use. Well, even on olinuxino somebody might use i2c_pins_a if they don't need an UART, but use an LCD. Thanks, Harald
Hi Stefan, On Wed, Apr 8, 2015 at 7:12 PM, Stefan Wahren <stefan.wahren@i2se.com> wrote: > AFAIK the iMX233-OLinuXino-MICRO don't have a UEXT connector. Maybe we reached > the point > to make dts files for every Olinuxino board. In that case we could also handle > the USB Host / Peripherial issue. Yes, I agree. it would be better to split the dts files for the olinuxino variants. Regards, Fabio Estevam
Hi Harald, >>> This patch provides the necessary devicetree code. >>> Tested with MOD-LCD1x9 from Olimex. >>> >>> This patch is based on work by Fadil Berisha with his permission. > However >>> all bugs are mine. >>> >>> Signed-off-by: Harald Geyer <harald@ccbib.org> >>> --- >>> arch/arm/boot/dts/imx23-olinuxino.dts | 6 +++++ >>> arch/arm/boot/dts/imx23.dtsi | 40 ++++++++++++++++++++++++++++++++- >>> 2 files changed, 45 insertions(+), 1 deletion(-) >>> >>> [...] >>> --- a/arch/arm/boot/dts/imx23.dtsi >>> +++ b/arch/arm/boot/dts/imx23.dtsi >>> @@ -308,6 +308,39 @@ >>> fsl,voltage = <MXS_VOLTAGE_HIGH>; >>> fsl,pull-up = <MXS_PULL_ENABLE>; >>> }; >>> + >>> + i2c_pins_a: i2c@0 { >>> + reg = <0>; >>> + fsl,pinmux-ids = < >>> + MX23_PAD_I2C_SCL__I2C_SCL >>> + MX23_PAD_I2C_SDA__I2C_SDA >>> + >; >>> + fsl,drive-strength = <MXS_DRIVE_8mA>; >>> + fsl,voltage = <MXS_VOLTAGE_HIGH>; >>> + fsl,pull-up = <MXS_PULL_ENABLE>; >>> + }; >>> + >>> + i2c_pins_b: i2c@1 { >>> + reg = <1>; >>> + fsl,pinmux-ids = < >>> + MX23_PAD_LCD_ENABLE__I2C_SCL >>> + MX23_PAD_LCD_HSYNC__I2C_SDA >>> + >; >>> + fsl,drive-strength = <MXS_DRIVE_8mA>; >>> + fsl,voltage = <MXS_VOLTAGE_HIGH>; >>> + fsl,pull-up = <MXS_PULL_ENABLE>; >>> + }; >>> + >>> + i2c_pins_c: i2c@2 { >>> + reg = <2>; >>> + fsl,pinmux-ids = < >>> + MX23_PAD_SSP1_DATA1__I2C_SCL >>> + MX23_PAD_SSP1_DATA2__I2C_SDA >>> + >; >>> + fsl,drive-strength = <MXS_DRIVE_8mA>; >>> + fsl,voltage = <MXS_VOLTAGE_HIGH>; >>> + fsl,pull-up = <MXS_PULL_ENABLE>; >>> + }; >> Please add only i2c_pins_b because this pin muxing is the only used one. > I disagree: Having all possible i2c pin groups documented in devicetree is > a good thing. Also imx23.dtsi is more general then olinuxino - somebody > might produce a imx23 based board where one of the other pin groups makes > more sense to use. Well, even on olinuxino somebody might use i2c_pins_a > if they don't need an UART, but use an LCD. Sure. But this 2 additional muxes have nothing to do with adding i2c support to olinuxino and should be a separate patch. In case of a revert of this patch the other i.MX23 boards are also affected. So i think it should be a separate patch. Btw i read that Shawn only want used muxes [1]. [1] - http://lists.infradead.org/pipermail/linux-arm-kernel/2012-August/115779.html > Thanks, > Harald
Hi Stefan! On Thu, 09 Apr 2015 11:20:10 +0200, Stefan Wahren <stefan.wahren@i2se.com> wrote: > Hi Harald, > >>>> This patch provides the necessary devicetree code. >>>> Tested with MOD-LCD1x9 from Olimex. >>>> >>>> This patch is based on work by Fadil Berisha with his permission. >> However >>>> all bugs are mine. >>>> >>>> Signed-off-by: Harald Geyer <harald@ccbib.org> >>>> --- >>>> arch/arm/boot/dts/imx23-olinuxino.dts | 6 +++++ >>>> arch/arm/boot/dts/imx23.dtsi | 40 ++++++++++++++++++++++++++++++++- >>>> 2 files changed, 45 insertions(+), 1 deletion(-) >>>> >>>> [...] >>>> --- a/arch/arm/boot/dts/imx23.dtsi >>>> +++ b/arch/arm/boot/dts/imx23.dtsi >>>> @@ -308,6 +308,39 @@ >>>> fsl,voltage = <MXS_VOLTAGE_HIGH>; >>>> fsl,pull-up = <MXS_PULL_ENABLE>; >>>> }; >>>> + >>>> + i2c_pins_a: i2c@0 { >>>> + reg = <0>; >>>> + fsl,pinmux-ids = < >>>> + MX23_PAD_I2C_SCL__I2C_SCL >>>> + MX23_PAD_I2C_SDA__I2C_SDA >>>> + >; >>>> + fsl,drive-strength = <MXS_DRIVE_8mA>; >>>> + fsl,voltage = <MXS_VOLTAGE_HIGH>; >>>> + fsl,pull-up = <MXS_PULL_ENABLE>; >>>> + }; >>>> + >>>> + i2c_pins_b: i2c@1 { >>>> + reg = <1>; >>>> + fsl,pinmux-ids = < >>>> + MX23_PAD_LCD_ENABLE__I2C_SCL >>>> + MX23_PAD_LCD_HSYNC__I2C_SDA >>>> + >; >>>> + fsl,drive-strength = <MXS_DRIVE_8mA>; >>>> + fsl,voltage = <MXS_VOLTAGE_HIGH>; >>>> + fsl,pull-up = <MXS_PULL_ENABLE>; >>>> + }; >>>> + >>>> + i2c_pins_c: i2c@2 { >>>> + reg = <2>; >>>> + fsl,pinmux-ids = < >>>> + MX23_PAD_SSP1_DATA1__I2C_SCL >>>> + MX23_PAD_SSP1_DATA2__I2C_SDA >>>> + >; >>>> + fsl,drive-strength = <MXS_DRIVE_8mA>; >>>> + fsl,voltage = <MXS_VOLTAGE_HIGH>; >>>> + fsl,pull-up = <MXS_PULL_ENABLE>; >>>> + }; >>> Please add only i2c_pins_b because this pin muxing is the only used one. >> I disagree: Having all possible i2c pin groups documented in devicetree >> is >> a good thing. Also imx23.dtsi is more general then olinuxino - somebody >> might produce a imx23 based board where one of the other pin groups makes >> more sense to use. Well, even on olinuxino somebody might use i2c_pins_a >> if they don't need an UART, but use an LCD. > > Sure. But this 2 additional muxes have nothing to do with adding i2c > support to olinuxino and should be a separate patch. In case of a revert > of this patch the other i.MX23 boards are also affected. So i think it > should be a separate patch. If it is prefered, I will split the changes to imx23.dtsi and imx23-olinuxino.dts into two patches. > Btw i read that Shawn only want used muxes [1]. > > [1] - > http://lists.infradead.org/pipermail/linux-arm-kernel/2012-August/115779.html If that is an issue, then I'll just test all three pin groups on my olinuxino board. After all it would be really awkward to have only one pin group and that one is labeled "i2c_pins_b" ... Thanks, Harald
Hi Harald, > harald@ccbib.org hat am 9. April 2015 um 15:03 geschrieben: > > > Hi Stefan! > > On Thu, 09 Apr 2015 11:20:10 +0200, Stefan Wahren <stefan.wahren@i2se.com> > wrote: > > > Btw i read that Shawn only want used muxes [1]. > > > > [1] - > > > http://lists.infradead.org/pipermail/linux-arm-kernel/2012-August/115779.html > > If that is an issue, then I'll just test all three pin groups on my > olinuxino > board. After all it would be really awkward to have only one pin group and > that > one is labeled "i2c_pins_b" ... from my understanding you are free about the labels. So you could rename it to "i2c_pins_a" and change the other pin groups. > > Thanks, > Harald Stefan
Stefan Wahren writes: > > On Thu, 09 Apr 2015 11:20:10 +0200, Stefan Wahren <stefan.wahren@i2se.com> > > wrote: > > > > > Btw i read that Shawn only want used muxes [1]. > > > > > > [1] - > > > > > http://lists.infradead.org/pipermail/linux-arm-kernel/2012-August/115779.html > > > > If that is an issue, then I'll just test all three pin groups on my > > olinuxino > > board. After all it would be really awkward to have only one pin group and > > that > > one is labeled "i2c_pins_b" ... > > from my understanding you are free about the labels. So you could > rename it to "i2c_pins_a" and change the other pin groups. Theoretically yes, but there are two reasons not to do that: 1) When Fadil Berisha chose the labels he didn't do so arbitrarily. The labels make sense the way the pinmux table is organised in the datasheet. (otherwise he sure would have chosen "i2c_pins_a" for the default pin group on olinuxino). 2) The labels are in use for about two years now by a lot of people. (Yes it really takes that long to have this merged.) I would hate to break their setups for no strong reason. Harald
diff --git a/arch/arm/boot/dts/imx23-olinuxino.dts b/arch/arm/boot/dts/imx23-olinuxino.dts index 7e6eef2..85718df 100644 --- a/arch/arm/boot/dts/imx23-olinuxino.dts +++ b/arch/arm/boot/dts/imx23-olinuxino.dts @@ -73,6 +73,12 @@ status = "okay"; }; + i2c: i2c@80058000 { + pinctrl-names = "default"; + pinctrl-0 = <&i2c_pins_b>; + status = "okay"; + }; + duart: serial@80070000 { pinctrl-names = "default"; pinctrl-0 = <&duart_pins_a>; diff --git a/arch/arm/boot/dts/imx23.dtsi b/arch/arm/boot/dts/imx23.dtsi index bbcfb5a..c892d58 100644 --- a/arch/arm/boot/dts/imx23.dtsi +++ b/arch/arm/boot/dts/imx23.dtsi @@ -308,6 +308,39 @@ fsl,voltage = <MXS_VOLTAGE_HIGH>; fsl,pull-up = <MXS_PULL_ENABLE>; }; + + i2c_pins_a: i2c@0 { + reg = <0>; + fsl,pinmux-ids = < + MX23_PAD_I2C_SCL__I2C_SCL + MX23_PAD_I2C_SDA__I2C_SDA + >; + fsl,drive-strength = <MXS_DRIVE_8mA>; + fsl,voltage = <MXS_VOLTAGE_HIGH>; + fsl,pull-up = <MXS_PULL_ENABLE>; + }; + + i2c_pins_b: i2c@1 { + reg = <1>; + fsl,pinmux-ids = < + MX23_PAD_LCD_ENABLE__I2C_SCL + MX23_PAD_LCD_HSYNC__I2C_SDA + >; + fsl,drive-strength = <MXS_DRIVE_8mA>; + fsl,voltage = <MXS_VOLTAGE_HIGH>; + fsl,pull-up = <MXS_PULL_ENABLE>; + }; + + i2c_pins_c: i2c@2 { + reg = <2>; + fsl,pinmux-ids = < + MX23_PAD_SSP1_DATA1__I2C_SCL + MX23_PAD_SSP1_DATA2__I2C_SDA + >; + fsl,drive-strength = <MXS_DRIVE_8mA>; + fsl,voltage = <MXS_VOLTAGE_HIGH>; + fsl,pull-up = <MXS_PULL_ENABLE>; + }; }; digctl@8001c000 { @@ -444,8 +477,13 @@ status = "disabled"; }; - i2c@80058000 { + i2c: i2c@80058000 { + #address-cells = <1>; + #size-cells = <0>; + compatible = "fsl,imx23-i2c"; reg = <0x80058000 0x2000>; + interrupts = <27>; + clock-frequency = <100000>; dmas = <&dma_apbx 3>; dma-names = "rx-tx"; status = "disabled";
The imx23-olinuxino board has an i2c interface exposed on UEXT connector. This patch provides the necessary devicetree code. Tested with MOD-LCD1x9 from Olimex. This patch is based on work by Fadil Berisha with his permission. However all bugs are mine. Signed-off-by: Harald Geyer <harald@ccbib.org> --- arch/arm/boot/dts/imx23-olinuxino.dts | 6 +++++ arch/arm/boot/dts/imx23.dtsi | 40 ++++++++++++++++++++++++++++++++- 2 files changed, 45 insertions(+), 1 deletion(-)