[RFC PATCH 2/2] lib: reset: thead: Correct the naming convention of dts

Conor Dooley conor.dooley at microchip.com
Tue Jun 13 23:56:14 PDT 2023


On Wed, Jun 14, 2023 at 09:11:01AM +0800, Guo Ren wrote:
> On Wed, Jun 14, 2023 at 3:36 AM Conor Dooley <conor at kernel.org> wrote:
> > On Tue, Jun 13, 2023 at 10:31:40AM +0800, Guo Ren wrote:
> > > On Mon, Jun 12, 2023 at 11:39 PM Conor Dooley <conor at kernel.org> wrote:
> > > > On Mon, Jun 12, 2023 at 08:57:36AM +0800, Guo Ren wrote:
> > > > > Jisheng Zhang would update the Linux yaml patch, here is the final dts
> > > > > format of the reset controller:

@Jisheng, iirc this was the only real outstanding thing for your Linux
patchset, if you re-submit without the reset controller this week it
should still be in time for v6.5.

> > > > >                reset-controller at ffff019050 {
> > > > >                        compatible = "thead,cpu-reset";
> >
> > Also, please just add the soc-specific compatibles even if the driver
> > will bind against the most common form of it.
> >
> > > > >                        reg = <0xff 0xff019050 0x0 0x4>, <0xff
> > > > > 0xff015004 0x0 0x0>;
> > > > >                        reset-ctrl = <0x1c>;
> > > > >                        clone-csrs = <0x7f3 0x7c0 0x7c1 0x7c2 0x7c3
> > > > > 0x7c5 0x7cc 0x7ce>;
> >
> > To be honest, I don't actually understand why "clone-csrs" is even part
> > of this "cpu-reset" node. It seems to serve a different purpose to
> > something that could take various parts of the SoC out of reset.
> > IMO, the "clone-csrs" stuff is something that should be done based on
> > the soc-level compatible, and not related to the "cpu-reset" node at
> > all, which (given the below) sounds more and more like a regular reset
> > controller.

> We need to put clone-csrs' execution entry into the reset entry (the
> first reg address).

So, take the first reg out too then, with the clone-csrs bits?

That'd leave you with a "regular" reset controller for bits that reset
various parts of the SoC & the clone-csrs stuff can be its own thing.

What am I missing?

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 228 bytes
Desc: not available
URL: <http://lists.infradead.org/pipermail/opensbi/attachments/20230614/257e6f15/attachment.sig>


More information about the opensbi mailing list