Re: [PATCH v2 2/2] riscv: dts: anlogic: drop unused DR1V90 external clocks

From: Conor Dooley

Date: Thu Oct 08 2026 - 10:13:13 EST


On Fri, Oct 02, 2026 at 01:57:45PM -0400, Brian Masney wrote:
> On Fri, Oct 02, 2026 at 05:33:24PM +0100, Conor Dooley wrote:
> > On Sat, Oct 03, 2026 at 12:28:40AM +0800, Junhui Liu wrote:
> > > The external CAN and watchdog clock inputs are not routed on the MLKPAI
> > > FS01, the only board currently supported. Drop the two fixed-clock
> > > placeholders from the common dtsi, as fixed clocks require a valid
> > > clock-frequency property.
> > >
> > > Both clock muxes select their internal parent clocks after reset, so
> > > the CRU operates correctly without the external inputs. Keep only the
> > > required oscillator input to the CRU.
> > >
> > > A board using these external clocks should define them in its own
> > > board-level dts and override the clocks and clock-names properties of
> > > the CRU node.
> > >
> > > Reported-by: Yixun Lan <dlan@xxxxxxxxxx>
> > > Fixes: 7849521aaac2 ("riscv: dts: anlogic: add clocks and CRU for DR1V90")
> > > Reviewed-by: Conor Dooley <conor.dooley@xxxxxxxxxxxxx>
> > > Acked-by: Conor Dooley <conor.dooley@xxxxxxxxxxxxx>
> > > Signed-off-by: Junhui Liu <junhui.liu@xxxxxxxxxxxxx>
> >
> > Per the v1, Brian said he would rebase the clock tree, so I'll just
> > squash this into the original patch I think.
>
> It's already rebased, and dropped from the clk-next history.

Cool. Note to self: 7,5 material, cos I don't have the headers to do
anything with "riscv: dts: anlogic: add clocks and CRU for DR1V90"

Cheers,
Conor.

Attachment: signature.asc
Description: PGP signature