dt-bindings: pwm: Allow decimal format in addition to hex format
Subsystems (2)
PATCH v3 1/2 dt-bindings: pwm: Allow decimal format in addition to hex format
Let's allow node numbering in decimal format too.
Simple human-readable increments/IDs are usually decimal, hex is only for addresses as noted by Krzysztof Kozlowski <[email protected]>.
Let's use an improved match suggested by Krzysztof Kozlowski <[email protected]> and improved a bit by Uwe Kleine-König <[email protected]>.
---
Changes since v2:
- Use pattern suggested by Krzysztof and Uwe
Changes since v1:
- New patch added to deal with pwm-omap-dmtimer binding
--- Documentation/devicetree/bindings/pwm/pwm.yaml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
Documentation/devicetree/bindings/pwm/pwm.yaml +1 -1
@@ -13,7 +13,7 @@ select: false properties: $nodename: - pattern: "^pwm(@.*|-[0-9a-f])*$" + pattern: "^pwm(@.+|-[0-9a-f]+)?$" "#pwm-cells": description:
Signature · 1 line
2.38.1
PATCH v3 2/2 dt-bindings: pwm: ti,pwm-omap-dmtimer: Update binding for yaml
Update for yaml and remove the old txt binding.
As we can replace most of the custom timer API with standard Linux frameworks such as clock framework, let's tag the properties for ti,prescaler and ti,clock-source as deprecated.
---
Changes since v2:
- Reviewed-by collected from Krzysztof, no changes
Changes since v1:
- Fix issues noticed by Krzysztof
--- .../bindings/pwm/pwm-omap-dmtimer.txt | 22 ------- .../bindings/pwm/ti,omap-dmtimer-pwm.yaml | 59 +++++++++++++++++++ 2 files changed, 59 insertions(+), 22 deletions(-) delete mode 100644 Documentation/devicetree/bindings/pwm/pwm-omap-dmtimer.txt create mode 100644 Documentation/devicetree/bindings/pwm/ti,omap-dmtimer-pwm.yaml
Documentation/devicetree/bindings/pwm/pwm-omap-dmtimer.txt deleted +0 -22
@@ -1,22 +0,0 @@ -* OMAP PWM for dual-mode timers - -Required properties: -- compatible: Shall contain "ti,omap-dmtimer-pwm". -- ti,timers: phandle to PWM capable OMAP timer. See timer/ti,timer-dm.yaml for info - about these timers. -- #pwm-cells: Should be 3. See pwm.yaml in this directory for a description of - the cells format. - -Optional properties: -- ti,prescaler: Should be a value between 0 and 7, see the timers datasheet -- ti,clock-source: Set dmtimer parent clock, values between 0 and 2: - - 0x00 - high-frequency system clock (timer_sys_ck) - - 0x01 - 32-kHz always-on clock (timer_32k_ck) - - 0x02 - external clock (timer_ext_ck, OMAP2 only) - -Example: - pwm9: dmtimer-pwm@9 { - compatible = "ti,omap-dmtimer-pwm"; - ti,timers = <&timer9>; - #pwm-cells = <3>; - };
Documentation/devicetree/bindings/pwm/ti,omap-dmtimer-pwm.yaml new file +59 -0
59 lines not on this page · raw message
Signature · 1 line
2.38.1
Quoted 9 lines · Tony Lindgren
Let's allow node numbering in decimal format too.
Simple human-readable increments/IDs are usually decimal, hex is only for addresses as noted by Krzysztof Kozlowski <[email protected]>.
Let's use an improved match suggested by Krzysztof Kozlowski <[email protected]> and improved a bit by Uwe Kleine-König <[email protected]>.
Best regards,
Krzysztof
Quoted 9 lines · Tony Lindgren
Let's allow node numbering in decimal format too.
Simple human-readable increments/IDs are usually decimal, hex is only for addresses as noted by Krzysztof Kozlowski <[email protected]>.
Let's use an improved match suggested by Krzysztof Kozlowski <[email protected]> and improved a bit by Uwe Kleine-König <[email protected]>.
Quoted 5 lines
Cc Thierry Reding <[email protected]>Suggested-by Krzysztof Kozlowski <[email protected]>Suggested-by Uwe Kleine-König <[email protected]>Signed-off-by Tony Lindgren <[email protected]>
Which tree should this merged through? The obvious candidates are pwm, dt and omap.
Best regards
Uwe
Signature · 2 lines
Pengutronix e.K. | Uwe Kleine-König | Industrial Linux Solutions | https://www.pengutronix.de/ |
Quoted 16 lines · Uwe Kleine-König
Quoted 9 lines · Tony Lindgren
Let's allow node numbering in decimal format too.
Simple human-readable increments/IDs are usually decimal, hex is only for addresses as noted by Krzysztof Kozlowski <[email protected]>.
Let's use an improved match suggested by Krzysztof Kozlowski <[email protected]> and improved a bit by Uwe Kleine-König <[email protected]>.
Quoted 5 lines
Cc Thierry Reding <[email protected]>Suggested-by Krzysztof Kozlowski <[email protected]>Suggested-by Uwe Kleine-König <[email protected]>Signed-off-by Tony Lindgren <[email protected]>
Quoted 4 lines
Acked-by Uwe Kleine-König <[email protected]>Which tree should this merged through? The obvious candidates are pwm, dt and omap.
The dt folks usually queue the binding patches separately.
Regards,
Tony
Hello Tony,
Quoted 17 lines · Tony Lindgren
Quoted 16 lines · Uwe Kleine-König
Quoted 14 lines · Tony Lindgren
Let's allow node numbering in decimal format too.
Simple human-readable increments/IDs are usually decimal, hex is only for addresses as noted by Krzysztof Kozlowski <[email protected]>.
Let's use an improved match suggested by Krzysztof Kozlowski <[email protected]> and improved a bit by Uwe Kleine-König <[email protected]>.
Cc Thierry Reding <[email protected]>Suggested-by Krzysztof Kozlowski <[email protected]>Suggested-by Uwe Kleine-König <[email protected]>Signed-off-by Tony Lindgren <[email protected]>
Quoted 6 lines
Quoted 4 lines
Acked-by Uwe Kleine-König <[email protected]>Which tree should this merged through? The obvious candidates are pwm, dt and omap.
The dt folks usually queue the binding patches separately.
fine, so I marked your series in the PWM patchwork as "handled-elsewhere".
Best regards
Uwe
Signature · 2 lines
Pengutronix e.K. | Uwe Kleine-König | Industrial Linux Solutions | https://www.pengutronix.de/ |
Rob Herring
maintainerQuoted 17 lines · Tony Lindgren
Quoted 16 lines · Uwe Kleine-König
Quoted 14 lines · Tony Lindgren
Let's allow node numbering in decimal format too.
Simple human-readable increments/IDs are usually decimal, hex is only for addresses as noted by Krzysztof Kozlowski <[email protected]>.
Let's use an improved match suggested by Krzysztof Kozlowski <[email protected]> and improved a bit by Uwe Kleine-König <[email protected]>.
Cc Thierry Reding <[email protected]>Suggested-by Krzysztof Kozlowski <[email protected]>Suggested-by Uwe Kleine-König <[email protected]>Signed-off-by Tony Lindgren <[email protected]>
Quoted 6 lines
Quoted 4 lines
Acked-by Uwe Kleine-König <[email protected]>Which tree should this merged through? The obvious candidates are pwm, dt and omap.
The dt folks usually queue the binding patches separately.
The default is they go thru the subsystem trees, and I mainly pickup what has not after some time (and is standalone binding change).
Rob
Rob Herring
maintainerQuoted 32 lines · Tony Lindgren
Let's allow node numbering in decimal format too.
Simple human-readable increments/IDs are usually decimal, hex is only for addresses as noted by Krzysztof Kozlowski <[email protected]>.
Let's use an improved match suggested by Krzysztof Kozlowski <[email protected]> and improved a bit by Uwe Kleine-König <[email protected]>.
Cc Thierry Reding <[email protected]>Suggested-by Krzysztof Kozlowski <[email protected]>Suggested-by Uwe Kleine-König <[email protected]>Signed-off-by Tony Lindgren <[email protected]>---
Changes since v2:
- Use pattern suggested by Krzysztof and Uwe
Changes since v1:
- New patch added to deal with pwm-omap-dmtimer binding
--- Documentation/devicetree/bindings/pwm/pwm.yaml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
Documentation/devicetree/bindings/pwm/pwm.yaml select: false · line 16
Quoted 5 lines
Documentation/devicetree/bindings/pwm/pwm.yaml +1
-1@@ -13,7 +13,7 @@ select: false properties: $nodename: - pattern: "^pwm(@.*|-[0-9a-f])*$" + pattern: "^pwm(@.+|-[0-9a-f]+)?$"
So now pwm-10 could be either?
I'm fine with decimal, but can we do that everywhere we do this -N naming?
Quoted 6 lines
Documentation/devicetree/bindings/pwm/pwm.yaml +0
-0"#pwm-cells": description:Signature · 1 line
2.38.1
Quoted 34 lines · Rob Herring
Quoted 33 lines · Tony Lindgren
Let's allow node numbering in decimal format too.
Simple human-readable increments/IDs are usually decimal, hex is only for addresses as noted by Krzysztof Kozlowski <[email protected]>.
Let's use an improved match suggested by Krzysztof Kozlowski <[email protected]> and improved a bit by Uwe Kleine-König <[email protected]>.
Cc Thierry Reding <[email protected]>Suggested-by Krzysztof Kozlowski <[email protected]>Suggested-by Uwe Kleine-König <[email protected]>Signed-off-by Tony Lindgren <[email protected]>---
Changes since v2:
- Use pattern suggested by Krzysztof and Uwe
Changes since v1:
- New patch added to deal with pwm-omap-dmtimer binding
--- Documentation/devicetree/bindings/pwm/pwm.yaml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)Documentation/devicetree/bindings/pwm/pwm.yaml +0
-0@@ -13,7 +13,7 @@ select: false
Quoted 6 lines
Documentation/devicetree/bindings/pwm/pwm.yaml select: false · line 16
Quoted 4 lines
Documentation/devicetree/bindings/pwm/pwm.yaml +1
-1properties: $nodename: - pattern: "^pwm(@.*|-[0-9a-f])*$" + pattern: "^pwm(@.+|-[0-9a-f]+)?$"So now pwm-10 could be either?
Yes.
Quoted 2 lines
I'm fine with decimal, but can we do that everywhere we do this -N naming?
Do you mean the '[0-9a-f]' users that don't use '[0-9af]+'?
These can be found with:
$ find Documentation/devicetree/bindings/ -name \*.yaml | \ xargs grep pattern: | grep '\[0-9a-f\]' | grep -v '\[0-9a-f\]+'
Not sure if some of these need to intentionally limit the node numbering to 15.
If you have some other criteria in mind, let me know :)
That sounds like a separate patch though.
Regards,
Tony
Rob Herring
maintainerQuoted 42 lines · Tony Lindgren
Quoted 34 lines · Rob Herring
Quoted 33 lines · Tony Lindgren
Let's allow node numbering in decimal format too.
Simple human-readable increments/IDs are usually decimal, hex is only for addresses as noted by Krzysztof Kozlowski <[email protected]>.
Let's use an improved match suggested by Krzysztof Kozlowski <[email protected]> and improved a bit by Uwe Kleine-König <[email protected]>.
Cc Thierry Reding <[email protected]>Suggested-by Krzysztof Kozlowski <[email protected]>Suggested-by Uwe Kleine-König <[email protected]>Signed-off-by Tony Lindgren <[email protected]>---
Changes since v2:
- Use pattern suggested by Krzysztof and Uwe
Changes since v1:
- New patch added to deal with pwm-omap-dmtimer binding
--- Documentation/devicetree/bindings/pwm/pwm.yaml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)Documentation/devicetree/bindings/pwm/pwm.yaml +0
-0@@ -13,7 +13,7 @@ select: falseQuoted 6 lines
Documentation/devicetree/bindings/pwm/pwm.yaml select: false · line 16
Quoted 4 lines
Documentation/devicetree/bindings/pwm/pwm.yaml +1
-1properties: $nodename: - pattern: "^pwm(@.*|-[0-9a-f])*$" + pattern: "^pwm(@.+|-[0-9a-f]+)?$"So now pwm-10 could be either?
Quoted 6 lines
Yes.
Quoted 2 lines
I'm fine with decimal, but can we do that everywhere we do this -N naming?
Do you mean the '[0-9a-f]' users that don't use '[0-9af]+'?
No, I mean for all cases of <nodename>-N, can be we consistent. Either we use hex or we use decimal.
Quoted 5 lines
These can be found with:
$ find Documentation/devicetree/bindings/ -name \*.yaml | \ xargs grep pattern: | grep '\[0-9a-f\]' | grep -v '\[0-9a-f\]+'
Not quite. It's just cases of '-N':
$ find Documentation/devicetree/bindings/ -name \*.yaml | xargs grep pattern: | grep '\-\[0-9a-f\]' | grep -v '\[0-9a-f\]+' Documentation/devicetree/bindings/phy/intel,combo-phy.yaml: pattern: "combophy(@.*|-[0-9a-f])*$" Documentation/devicetree/bindings/pwm/pwm.yaml: pattern: "^pwm(@.*|-[0-9a-f])*$" Documentation/devicetree/bindings/timestamp/hardware-timestamps-common.yaml: pattern: "^timestamp(@.*|-[0-9a-f])?$" Documentation/devicetree/bindings/watchdog/watchdog.yaml: pattern: "^watchdog(@.*|-[0-9a-f])?$" Documentation/devicetree/bindings/spi/spi-controller.yaml: pattern: "^spi(@.*|-[0-9a-f])*$" Documentation/devicetree/bindings/rtc/rtc.yaml: pattern: "^rtc(@.*|-[0-9a-f])*$"
And there's probably some more in dtschema.
Quoted 3 lines
Not sure if some of these need to intentionally limit the node numbering to 15.
No, I think it was all just cases of 15 should be enough for anyone.
Quoted 2 lines
If you have some other criteria in mind, let me know :)
If the numbering is hex, then it is as you have it. (Though we can't enforce that.
If it is decimal, then the regex should be ([0-9]|[1-9][0-9]+).
Rob
Hi Rob,
Quoted 41 lines · Rob Herring
Quoted 13 lines · Tony Lindgren
Quoted 11 lines · Rob Herring
Documentation/devicetree/bindings/pwm/pwm.yaml select: false · line 16
Quoted 8 lines · Tony Lindgren
Documentation/devicetree/bindings/pwm/pwm.yaml +1
-1@@ -13,7 +13,7 @@ select: false properties: $nodename: - pattern: "^pwm(@.*|-[0-9a-f])*$" + pattern: "^pwm(@.+|-[0-9a-f]+)?$"So now pwm-10 could be either?
Quoted 6 lines
Yes.
Quoted 2 lines
I'm fine with decimal, but can we do that everywhere we do this -N naming?
Do you mean the '[0-9a-f]' users that don't use '[0-9af]+'?
No, I mean for all cases of <nodename>-N, can be we consistent. Either we use hex or we use decimal.
Quoted 5 lines
These can be found with:
$ find Documentation/devicetree/bindings/ -name \*.yaml | \ xargs grep pattern: | grep '\[0-9a-f\]' | grep -v '\[0-9a-f\]+'Not quite. It's just cases of '-N':
$ find Documentation/devicetree/bindings/ -name \*.yaml | xargs grep pattern: | grep '\-\[0-9a-f\]' | grep -v '\[0-9a-f\]+' Documentation/devicetree/bindings/phy/intel,combo-phy.yaml: pattern: "combophy(@.*|-[0-9a-f])*$" Documentation/devicetree/bindings/pwm/pwm.yaml: pattern: "^pwm(@.*|-[0-9a-f])*$" Documentation/devicetree/bindings/timestamp/hardware-timestamps-common.yaml: pattern: "^timestamp(@.*|-[0-9a-f])?$" Documentation/devicetree/bindings/watchdog/watchdog.yaml: pattern: "^watchdog(@.*|-[0-9a-f])?$" Documentation/devicetree/bindings/spi/spi-controller.yaml: pattern: "^spi(@.*|-[0-9a-f])*$" Documentation/devicetree/bindings/rtc/rtc.yaml: pattern: "^rtc(@.*|-[0-9a-f])*$"
Quoted 1 line
And there's probably some more in dtschema.
Looking at this again, not exactly sure still what you want..
Can you please post some initial patch maybe, verbal patches are a bit tricky :)
Regards,
Tony
_______________________________________________
linux-arm-kernel mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel