Hi Alexandre, On Thu, Oct 17, 2024 at 12:03 AM Alexandre Belloni <alexandre.belloni@xxxxxxxxxxx> wrote: > On 11/10/2024 13:28:55+0300, claudiu beznea wrote: > > >> + compatible = "renesas,r9a08g045-rtca3", "renesas,rz-rtca3"; > > >> + reg = <0 0x1004ec00 0 0x400>; > > >> + interrupts = <GIC_SPI 315 IRQ_TYPE_LEVEL_HIGH>, > > >> + <GIC_SPI 316 IRQ_TYPE_LEVEL_HIGH>, > > >> + <GIC_SPI 317 IRQ_TYPE_LEVEL_HIGH>; > > >> + interrupt-names = "alarm", "period", "carry"; > > >> + clocks = <&cpg CPG_MOD R9A08G045_VBAT_BCLK>, <&vbattb VBATTB_VBATTCLK>; > > >> + clock-names = "bus", "counter"; > > >> + assigned-clocks = <&vbattb VBATTB_MUX>; > > >> + assigned-clock-parents = <&vbattb VBATTB_XC>; > > > > > > Don't the assigned-clock* properties belong in the board DTS? > > > > It makes sense to be in the board DTS, indeed. > > > > > In addition, I think they should be documented in the DT bindings, > > > and be made required, so board developers don't forget about them. > > > > It would be better, indeed. > > There were multiple recent emails from Rob saying that assigned-clocks > should not be part of the bindings. You mean "In general, assigned-clocks* do not need to be documented and should never be required."[1] and "Whatever clock setup needed is outside the scope of a binding"[2]? In this case, it's not generic, and not specific to "renesas,rz-rtca3", but related to the integration of RTC-A3 on the RZ/G3S SoC. Which is indeed not relevant for the RTC-A3 bindings, so I agree the assigned-clock* properties should not be part of this binding. At least they're present in the example in the bindings ;-) [1] https://lore.kernel.org/all/20241015211540.GA1968867-robh@xxxxxxxxxx [2] https://lore.kernel.org/all/20241015164609.GA1235530-robh@xxxxxxxxxx Gr{oetje,eeting}s, Geert -- Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@xxxxxxxxxxxxxx In personal conversations with technical people, I call myself a hacker. But when I'm talking to journalists I just say "programmer" or something like that. -- Linus Torvalds
