[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <3b325b3b-df64-4f16-9284-cddfaaa40a21@altera.com>
Date: Thu, 20 Mar 2025 16:24:02 -0700
From: "Gerlach, Matthew" <matthew.gerlach@...era.com>
To: Rob Herring <robh@...nel.org>
Cc: krzk+dt@...nel.org, conor+dt@...nel.org, dinguyen@...nel.org,
bp@...en8.de, tony.luck@...el.com, james.morse@....com, mchehab@...nel.org,
rric@...nel.org, devicetree@...r.kernel.org, linux-kernel@...r.kernel.org,
linux-edac@...r.kernel.org
Subject: Re: [PATCH 3/4] arm64: dts: agilex: Update eccmgr in DTSI to reflect
hw/yaml
On 3/20/2025 12:34 PM, Rob Herring wrote:
> On Thu, Mar 20, 2025 at 09:46:21AM -0700, Matthew Gerlach wrote:
> > Update socfpga_agilex.dtsi to track the actual hardware description
> > provided in altr,socfpga-s10-ecc-manager.yaml.
> >
> > Signed-off-by: Matthew Gerlach <matthew.gerlach@...era.com>
> > ---
> > arch/arm64/boot/dts/intel/socfpga_agilex.dtsi | 18 ++++++------------
> > 1 file changed, 6 insertions(+), 12 deletions(-)
> >
> > diff --git a/arch/arm64/boot/dts/intel/socfpga_agilex.dtsi b/arch/arm64/boot/dts/intel/socfpga_agilex.dtsi
> > index 1235ba5a9865..708cb8e762b6 100644
> > --- a/arch/arm64/boot/dts/intel/socfpga_agilex.dtsi
> > +++ b/arch/arm64/boot/dts/intel/socfpga_agilex.dtsi
> > @@ -602,8 +602,7 @@ sdr: sdr@...11100 {
> > };
> >
> > eccmgr {
> > - compatible = "altr,socfpga-s10-ecc-manager",
> > - "altr,socfpga-a10-ecc-manager";
> > + compatible = "altr,socfpga-s10-ecc-manager";
>
> You are breaking the ABI here. Before this series, the driver required
> altr,socfpga-a10-ecc-manager.
Yes, breaking the ABI required the change in PATCH 2/4, which is
suboptimal.
>
> > altr,sysmgr-syscon = <&sysmgr>;
> > #address-cells = <1>;
> > #size-cells = <1>;
> > @@ -619,40 +618,35 @@ sdramedac {
> > };
> >
> > ocram-ecc@...cc000 {
> > - compatible = "altr,socfpga-s10-ocram-ecc",
> > - "altr,socfpga-a10-ocram-ecc";
> > + compatible = "altr,socfpga-a10-ocram-ecc";
>
> AIUI, nothing used altr,socfpga-s10-ocram-ecc, so this change is okay I
> guess. Normally, we'd require both because there might be some
> difference you find later on, but here you could just look at the parent
> node compatible.
>
> If it were me, I'd just add the compatible string in the schema and
> avoid the .dts change. That would have been less work...
Less work sounds better. I will look at just creating the yaml, with no
dtsi/dts and no driver changes.
>
> Rob
Thanks for the feedback,
Matthew Gerlach
Powered by blists - more mailing lists