| Message ID | 20170525172911.11467-5-stillcompiling@gmail.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
Hi, Am 25.05.2017 um 19:29 schrieb Joshua Clayton: > Add support for Altera V FPGA connected to an spi port Did you mean "Altera Cyclone V"? > to the evi devicetree file > > Signed-off-by: Joshua Clayton <stillcompiling@gmail.com> > --- > arch/arm/boot/dts/imx6q-evi.dts | 16 ++++++++++++++++ > 1 file changed, 16 insertions(+) > > diff --git a/arch/arm/boot/dts/imx6q-evi.dts b/arch/arm/boot/dts/imx6q-evi.dts > index 24fe093a66db..a0cbb2d84803 100644 > --- a/arch/arm/boot/dts/imx6q-evi.dts > +++ b/arch/arm/boot/dts/imx6q-evi.dts > @@ -82,6 +82,15 @@ > pinctrl-names = "default"; > pinctrl-0 = <&pinctrl_ecspi1 &pinctrl_ecspi1cs>; > status = "okay"; > + > + fpga_spi: cyclonespi@0 { "cyclonespi" does not strike me as the best node name. I am guessing this is a sub-node of a SPI controller node, so no need to repeat "spi", and Cyclone seems more or less implied by "altr,fpga-". Note that the example in the bindings doc uses "evi-fpga-spi". Nodes don't need to be (shouldn't be?) prefixed with the board. Note that bindings examples tend to get copied a lot. Any reason not to just use "fpga@0" in both places for simplicity? > + compatible = "altr,fpga-passive-serial"; > + spi-max-frequency = <20000000>; > + reg = <0>; > + pinctrl-0 = <&pinctrl_fpgaspi>; > + nconfig-gpios = <&gpio4 9 GPIO_ACTIVE_LOW>; > + nstat-gpios = <&gpio4 11 GPIO_ACTIVE_LOW>; > + }; > }; > > &ecspi3 { [snip] Regards, Andreas
On Friday, June 2, 2017 6:30:12 PM PDT Andreas Färber wrote: > Hi, > > Am 25.05.2017 um 19:29 schrieb Joshua Clayton: > > Add support for Altera V FPGA connected to an spi port > > Did you mean "Altera Cyclone V"? I meant to shorten it from Altera Cyclone V to Altera FPGA. Didn't quite make it. Will fix and resubmit today. > > > to the evi devicetree file > > > > Signed-off-by: Joshua Clayton <stillcompiling@gmail.com> > > --- > > > > arch/arm/boot/dts/imx6q-evi.dts | 16 ++++++++++++++++ > > 1 file changed, 16 insertions(+) > > > > diff --git a/arch/arm/boot/dts/imx6q-evi.dts > > b/arch/arm/boot/dts/imx6q-evi.dts index 24fe093a66db..a0cbb2d84803 100644 > > --- a/arch/arm/boot/dts/imx6q-evi.dts > > +++ b/arch/arm/boot/dts/imx6q-evi.dts > > @@ -82,6 +82,15 @@ > > > > pinctrl-names = "default"; > > pinctrl-0 = <&pinctrl_ecspi1 &pinctrl_ecspi1cs>; > > status = "okay"; > > > > + > > + fpga_spi: cyclonespi@0 { > > "cyclonespi" does not strike me as the best node name. > > I am guessing this is a sub-node of a SPI controller node, so no need to > repeat "spi", and Cyclone seems more or less implied by "altr,fpga-". True. > > Note that the example in the bindings doc uses "evi-fpga-spi". Nodes > don't need to be (shouldn't be?) prefixed with the board. Note that > bindings examples tend to get copied a lot. > > Any reason not to just use "fpga@0" in both places for simplicity? Sure. fpga: fpga@0 is probably better. I'll change it in both the dts and the binding doc. > > > + compatible = "altr,fpga-passive-serial"; > > + spi-max-frequency = <20000000>; > > + reg = <0>; > > + pinctrl-0 = <&pinctrl_fpgaspi>; > > + nconfig-gpios = <&gpio4 9 GPIO_ACTIVE_LOW>; > > + nstat-gpios = <&gpio4 11 GPIO_ACTIVE_LOW>; > > + }; > > > > }; > > > > &ecspi3 { > > [snip] > > Regards, > Andreas
Am 02.06.2017 um 21:39 schrieb stillcompiling@gmail.com: > On Friday, June 2, 2017 6:30:12 PM PDT Andreas Färber wrote: >> Am 25.05.2017 um 19:29 schrieb Joshua Clayton: >>> diff --git a/arch/arm/boot/dts/imx6q-evi.dts >>> b/arch/arm/boot/dts/imx6q-evi.dts index 24fe093a66db..a0cbb2d84803 100644 >>> --- a/arch/arm/boot/dts/imx6q-evi.dts >>> +++ b/arch/arm/boot/dts/imx6q-evi.dts >>> @@ -82,6 +82,15 @@ >>> >>> pinctrl-names = "default"; >>> pinctrl-0 = <&pinctrl_ecspi1 &pinctrl_ecspi1cs>; >>> status = "okay"; >>> >>> + >>> + fpga_spi: cyclonespi@0 { >> >> "cyclonespi" does not strike me as the best node name. >> >> I am guessing this is a sub-node of a SPI controller node, so no need to >> repeat "spi", and Cyclone seems more or less implied by "altr,fpga-". > True. >> >> Note that the example in the bindings doc uses "evi-fpga-spi". Nodes >> don't need to be (shouldn't be?) prefixed with the board. Note that >> bindings examples tend to get copied a lot. >> >> Any reason not to just use "fpga@0" in both places for simplicity? > Sure. fpga: fpga@0 is probably better. Note that I was only commenting on the node name, the latter part. I'm not aware of any rules for the label, so that could remain unchanged or adopt cyclone_spi from the old node name or whatever is unique and syntactically valid. > I'll change it in both the dts and the binding doc. Thanks. Maybe double-check if there's any conventions Xilinx/Lattice DTs are using. Cheers, Andreas
On Friday, June 2, 2017 9:54:22 PM PDT Andreas Färber wrote: > Am 02.06.2017 um 21:39 schrieb stillcompiling@gmail.com: > > On Friday, June 2, 2017 6:30:12 PM PDT Andreas Färber wrote: > >> Am 25.05.2017 um 19:29 schrieb Joshua Clayton: > >>> diff --git a/arch/arm/boot/dts/imx6q-evi.dts > >>> b/arch/arm/boot/dts/imx6q-evi.dts index 24fe093a66db..a0cbb2d84803 > >>> 100644 > >>> --- a/arch/arm/boot/dts/imx6q-evi.dts > >>> +++ b/arch/arm/boot/dts/imx6q-evi.dts > >>> @@ -82,6 +82,15 @@ > >>> > >>> pinctrl-names = "default"; > >>> pinctrl-0 = <&pinctrl_ecspi1 &pinctrl_ecspi1cs>; > >>> status = "okay"; > >>> > >>> + > >>> + fpga_spi: cyclonespi@0 { > >> > >> "cyclonespi" does not strike me as the best node name. > >> > >> I am guessing this is a sub-node of a SPI controller node, so no need to > >> repeat "spi", and Cyclone seems more or less implied by "altr,fpga-". > > > > True. > > > >> Note that the example in the bindings doc uses "evi-fpga-spi". Nodes > >> don't need to be (shouldn't be?) prefixed with the board. Note that > >> bindings examples tend to get copied a lot. > >> > >> Any reason not to just use "fpga@0" in both places for simplicity? > > > > Sure. fpga: fpga@0 is probably better. > > Note that I was only commenting on the node name, the latter part. > > I'm not aware of any rules for the label, so that could remain unchanged > or adopt cyclone_spi from the old node name or whatever is unique and > syntactically valid. Too late! Patches posted. Oh, well, I'm not changing it back. > > > I'll change it in both the dts and the binding doc. > > Thanks. Maybe double-check if there's any conventions Xilinx/Lattice DTs > are using. > Of the conventions I found, fpga seemed the most "hardware descriptive" for a plain FPGA. The other one several binding doc examples are using is "fpga-mgr". > Cheers, > Andreas
On Fri, Jun 2, 2017 at 4:10 PM, <stillcompiling@gmail.com> wrote: > On Friday, June 2, 2017 9:54:22 PM PDT Andreas Färber wrote: >> Am 02.06.2017 um 21:39 schrieb stillcompiling@gmail.com: >> > On Friday, June 2, 2017 6:30:12 PM PDT Andreas Färber wrote: >> >> Am 25.05.2017 um 19:29 schrieb Joshua Clayton: >> >>> diff --git a/arch/arm/boot/dts/imx6q-evi.dts >> >>> b/arch/arm/boot/dts/imx6q-evi.dts index 24fe093a66db..a0cbb2d84803 >> >>> 100644 >> >>> --- a/arch/arm/boot/dts/imx6q-evi.dts >> >>> +++ b/arch/arm/boot/dts/imx6q-evi.dts >> >>> @@ -82,6 +82,15 @@ >> >>> >> >>> pinctrl-names = "default"; >> >>> pinctrl-0 = <&pinctrl_ecspi1 &pinctrl_ecspi1cs>; >> >>> status = "okay"; >> >>> >> >>> + >> >>> + fpga_spi: cyclonespi@0 { >> >> >> >> "cyclonespi" does not strike me as the best node name. >> >> >> >> I am guessing this is a sub-node of a SPI controller node, so no need to >> >> repeat "spi", and Cyclone seems more or less implied by "altr,fpga-". >> > >> > True. >> > >> >> Note that the example in the bindings doc uses "evi-fpga-spi". Nodes >> >> don't need to be (shouldn't be?) prefixed with the board. Note that >> >> bindings examples tend to get copied a lot. >> >> >> >> Any reason not to just use "fpga@0" in both places for simplicity? >> > >> > Sure. fpga: fpga@0 is probably better. >> >> Note that I was only commenting on the node name, the latter part. >> >> I'm not aware of any rules for the label, so that could remain unchanged >> or adopt cyclone_spi from the old node name or whatever is unique and >> syntactically valid. > Too late! Patches posted. > Oh, well, I'm not changing it back. It's fine as it is in v12. fpga_mgr: fpga-mgr@0 is what I've been using most recently. It distinguishes the one block that is used to program the fpga from the fpga and hardware in the fpga. But no need to respin this. Alan >> >> > I'll change it in both the dts and the binding doc. >> >> Thanks. Maybe double-check if there's any conventions Xilinx/Lattice DTs >> are using. >> > Of the conventions I found, fpga seemed the most "hardware descriptive" > for a plain FPGA. > The other one several binding doc examples are using is "fpga-mgr". > >> Cheers, >> Andreas > > > -- > ~Joshua A Clayton
diff --git a/arch/arm/boot/dts/imx6q-evi.dts b/arch/arm/boot/dts/imx6q-evi.dts index 24fe093a66db..a0cbb2d84803 100644 --- a/arch/arm/boot/dts/imx6q-evi.dts +++ b/arch/arm/boot/dts/imx6q-evi.dts @@ -82,6 +82,15 @@ pinctrl-names = "default"; pinctrl-0 = <&pinctrl_ecspi1 &pinctrl_ecspi1cs>; status = "okay"; + + fpga_spi: cyclonespi@0 { + compatible = "altr,fpga-passive-serial"; + spi-max-frequency = <20000000>; + reg = <0>; + pinctrl-0 = <&pinctrl_fpgaspi>; + nconfig-gpios = <&gpio4 9 GPIO_ACTIVE_LOW>; + nstat-gpios = <&gpio4 11 GPIO_ACTIVE_LOW>; + }; }; &ecspi3 { @@ -313,6 +322,13 @@ >; }; + pinctrl_fpgaspi: fpgaspigrp { + fsl,pins = < + MX6QDL_PAD_KEY_ROW1__GPIO4_IO09 0x1b0b0 + MX6QDL_PAD_KEY_ROW2__GPIO4_IO11 0x1b0b0 + >; + }; + pinctrl_gpminand: gpminandgrp { fsl,pins = < MX6QDL_PAD_NANDF_CLE__NAND_CLE 0xb0b1
Add support for Altera V FPGA connected to an spi port to the evi devicetree file Signed-off-by: Joshua Clayton <stillcompiling@gmail.com> --- arch/arm/boot/dts/imx6q-evi.dts | 16 ++++++++++++++++ 1 file changed, 16 insertions(+)