Message ID | CANqRtoRcsAhvFB=DNoJHzJt4dejKX+Z8ar04D8O1cFq1XkxWkw@mail.gmail.com (mailing list archive) |
---|---|
State | Not Applicable |
Delegated to: | Simon Horman |
Headers | show |
On Tue, Aug 08, 2017 at 07:29:28PM +0900, Magnus Damm wrote: > Hi Simon, > > On Tue, Aug 8, 2017 at 5:39 PM, Simon Horman <horms+renesas@verge.net.au> wrote: > > Use newly added R-Car GPIO Gen 1, 2 and 3 fallback compat strings in place > > of now deprecated non-generation specific R-Car GPIO fallback compat string > > in the DT of Renesas ARM and arm64 based SoCs. > > > > This should have no run-time effect as the driver matches against the > > per-SoC compat string before considering the fallback compat string. > > Thanks for your efforts.I have no issue with your series (apart from > the GPIO and SATA mistake), but at the same time I believe the GPIO > hardware itself is backwards compatible between various generations. > > In the nitpick department I would like to point out that the level of > hardware difference between say R-Car Gen1 GPIO and R-Car Gen2 GPIO is > similar to say good old uarts like 8250 and 16450 hardware. Basically > a couple of registers were added to the hardware in a > backwards-compatible way if I recall correctly. > > So if we are going to use "compatible" to point out if hardware is > compatible or not then I would do this instead: Thanks for your feedback. When the generation specific compat strings were recently added the renesas,gpio-rcar compat string was marked as deprecated as I was under the understanding that it was only compatibile with gen 1 SoCs. It now seems that was not the best thing to do and renesas,gpio-rcar should be re-instated as being a generic fallback for all R-Car versions supported in upstream. Do you concur? > > --- a/arch/arm64/boot/dts/renesas/r8a7796.dtsi > +++ b/arch/arm64/boot/dts/renesas/r8a7796.dtsi > @@ -214,7 +214,7 @@ > > gpio0: gpio@e6050000 { > compatible = "renesas,gpio-r8a7796", > + "renesas,rcar-gen3-gpio"; > "renesas,gpio-rcar"; > reg = <0 0xe6050000 0 0x50>; > interrupts = <GIC_SPI 4 IRQ_TYPE_LEVEL_HIGH>; > #gpio-cells = <2>; > > At the same time I'm not sure if I care _that_ much. =) > > Thanks, > > / magnus >
On Wed, Aug 09, 2017 at 09:55:26AM +0200, Simon Horman wrote: > On Tue, Aug 08, 2017 at 07:29:28PM +0900, Magnus Damm wrote: > > Hi Simon, > > > > On Tue, Aug 8, 2017 at 5:39 PM, Simon Horman <horms+renesas@verge.net.au> wrote: > > > Use newly added R-Car GPIO Gen 1, 2 and 3 fallback compat strings in place > > > of now deprecated non-generation specific R-Car GPIO fallback compat string > > > in the DT of Renesas ARM and arm64 based SoCs. > > > > > > This should have no run-time effect as the driver matches against the > > > per-SoC compat string before considering the fallback compat string. > > > > Thanks for your efforts.I have no issue with your series (apart from > > the GPIO and SATA mistake), but at the same time I believe the GPIO > > hardware itself is backwards compatible between various generations. > > > > In the nitpick department I would like to point out that the level of > > hardware difference between say R-Car Gen1 GPIO and R-Car Gen2 GPIO is > > similar to say good old uarts like 8250 and 16450 hardware. Basically > > a couple of registers were added to the hardware in a > > backwards-compatible way if I recall correctly. > > > > So if we are going to use "compatible" to point out if hardware is > > compatible or not then I would do this instead: > > Thanks for your feedback. > > When the generation specific compat strings were recently > added the renesas,gpio-rcar compat string was marked as deprecated > as I was under the understanding that it was only compatibile with gen 1 SoCs. > > It now seems that was not the best thing to do and renesas,gpio-rcar should > be re-instated as being a generic fallback for all R-Car versions supported > in upstream. > > Do you concur? I posted a patch to un-deprecate the renesas,gpio-rcar compat string in the bindings documentation. [PATCH] gpio: rcar: reinstate generic compat string
Hi Simon, On Wed, Aug 9, 2017 at 5:22 PM, Simon Horman <horms@verge.net.au> wrote: > On Wed, Aug 09, 2017 at 09:55:26AM +0200, Simon Horman wrote: >> On Tue, Aug 08, 2017 at 07:29:28PM +0900, Magnus Damm wrote: >> > Hi Simon, >> > >> > On Tue, Aug 8, 2017 at 5:39 PM, Simon Horman <horms+renesas@verge.net.au> wrote: >> > > Use newly added R-Car GPIO Gen 1, 2 and 3 fallback compat strings in place >> > > of now deprecated non-generation specific R-Car GPIO fallback compat string >> > > in the DT of Renesas ARM and arm64 based SoCs. >> > > >> > > This should have no run-time effect as the driver matches against the >> > > per-SoC compat string before considering the fallback compat string. >> > >> > Thanks for your efforts.I have no issue with your series (apart from >> > the GPIO and SATA mistake), but at the same time I believe the GPIO >> > hardware itself is backwards compatible between various generations. >> > >> > In the nitpick department I would like to point out that the level of >> > hardware difference between say R-Car Gen1 GPIO and R-Car Gen2 GPIO is >> > similar to say good old uarts like 8250 and 16450 hardware. Basically >> > a couple of registers were added to the hardware in a >> > backwards-compatible way if I recall correctly. >> > >> > So if we are going to use "compatible" to point out if hardware is >> > compatible or not then I would do this instead: >> >> Thanks for your feedback. >> >> When the generation specific compat strings were recently >> added the renesas,gpio-rcar compat string was marked as deprecated >> as I was under the understanding that it was only compatibile with gen 1 SoCs. >> >> It now seems that was not the best thing to do and renesas,gpio-rcar should >> be re-instated as being a generic fallback for all R-Car versions supported >> in upstream. >> >> Do you concur? > > I posted a patch to un-deprecate the renesas,gpio-rcar compat string > in the bindings documentation. > > [PATCH] gpio: rcar: reinstate generic compat string Thanks, this looks good to me. I would like to hear what Geert thinks as well! Cheers, / magnus
On Wed, Aug 09, 2017 at 08:11:24PM +0900, Magnus Damm wrote: > Hi Simon, > > On Wed, Aug 9, 2017 at 5:22 PM, Simon Horman <horms@verge.net.au> wrote: > > On Wed, Aug 09, 2017 at 09:55:26AM +0200, Simon Horman wrote: > >> On Tue, Aug 08, 2017 at 07:29:28PM +0900, Magnus Damm wrote: > >> > Hi Simon, > >> > > >> > On Tue, Aug 8, 2017 at 5:39 PM, Simon Horman <horms+renesas@verge.net.au> wrote: > >> > > Use newly added R-Car GPIO Gen 1, 2 and 3 fallback compat strings in place > >> > > of now deprecated non-generation specific R-Car GPIO fallback compat string > >> > > in the DT of Renesas ARM and arm64 based SoCs. > >> > > > >> > > This should have no run-time effect as the driver matches against the > >> > > per-SoC compat string before considering the fallback compat string. > >> > > >> > Thanks for your efforts.I have no issue with your series (apart from > >> > the GPIO and SATA mistake), but at the same time I believe the GPIO > >> > hardware itself is backwards compatible between various generations. > >> > > >> > In the nitpick department I would like to point out that the level of > >> > hardware difference between say R-Car Gen1 GPIO and R-Car Gen2 GPIO is > >> > similar to say good old uarts like 8250 and 16450 hardware. Basically > >> > a couple of registers were added to the hardware in a > >> > backwards-compatible way if I recall correctly. > >> > > >> > So if we are going to use "compatible" to point out if hardware is > >> > compatible or not then I would do this instead: > >> > >> Thanks for your feedback. > >> > >> When the generation specific compat strings were recently > >> added the renesas,gpio-rcar compat string was marked as deprecated > >> as I was under the understanding that it was only compatibile with gen 1 SoCs. > >> > >> It now seems that was not the best thing to do and renesas,gpio-rcar should > >> be re-instated as being a generic fallback for all R-Car versions supported > >> in upstream. > >> > >> Do you concur? > > > > I posted a patch to un-deprecate the renesas,gpio-rcar compat string > > in the bindings documentation. > > > > [PATCH] gpio: rcar: reinstate generic compat string > > Thanks, this looks good to me. I would like to hear what Geert thinks as well! Thanks, good plan.
--- a/arch/arm64/boot/dts/renesas/r8a7796.dtsi +++ b/arch/arm64/boot/dts/renesas/r8a7796.dtsi @@ -214,7 +214,7 @@ gpio0: gpio@e6050000 { compatible = "renesas,gpio-r8a7796", + "renesas,rcar-gen3-gpio"; "renesas,gpio-rcar"; reg = <0 0xe6050000 0 0x50>; interrupts = <GIC_SPI 4 IRQ_TYPE_LEVEL_HIGH>; #gpio-cells = <2>;