[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <CAK8P3a1_97b14=wbYrFodRLBq_2o2gOOjk2zORp3DLG72O229w@mail.gmail.com>
Date: Wed, 13 Apr 2022 19:42:48 +0200
From: Arnd Bergmann <arnd@...db.de>
To: "Hawkins, Nick" <nick.hawkins@....com>
Cc: Arnd Bergmann <arnd@...db.de>,
"Verdun, Jean-Marie" <verdun@....com>,
Olof Johansson <olof@...om.net>,
"soc@...nel.org" <soc@...nel.org>,
Rob Herring <robh+dt@...nel.org>,
"linux-arm-kernel@...ts.infradead.org"
<linux-arm-kernel@...ts.infradead.org>,
"devicetree@...r.kernel.org" <devicetree@...r.kernel.org>,
"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH v3 09/10] arch: arm: boot: dts: Introduce HPE GXP Device tree
On Wed, Apr 13, 2022 at 6:48 PM Hawkins, Nick <nick.hawkins@....com> wrote:
> > I would count passing a register address to the child device as a private interface.
> > It's a minimalistic one, but that is not a bad thing here.
>
> Thank you. Now that the parent/child issue with timer/watchdog is resolved I am now
> having an issue trying to access "iopclk" from "gxp-timer". I have tried use of
> of_clk_get_by_name and of_clk_get which both fail to find the clock.
> Is it because clocks is outside of the axi -> ahb hierarchy?
No, that should work, but you have to pass the right device node, which would
correspond to the parent device. You should also be able to pass the parent
device pointer (i.e. linux device, not device_node) and use clk_get().
Arnd
Powered by blists - more mailing lists