Re: [PATCH 04/16] arm64: dts: qcom: sm8350: Specify clock-frequency for arch timer

From: Bjorn Andersson <bjorn.andersson@linaro.org>
To: Konrad Dybcio <konrad.dybcio@somainline.org>
Cc: Stephen Boyd <sboyd@kernel.org>,
	~postmarketos/upstreaming@lists.sr.ht,
	martin.botka@somainline.org,
	angelogioacchino.delregno@somainline.org,
	marijn.suijten@somainline.org, jamipkettunen@somainline.org,
	Andy Gross <agross@kernel.org>, Rob Herring <robh+dt@kernel.org>,
	linux-arm-msm@vger.kernel.org, devicetree@vger.kernel.org,
	linux-kernel@vger.kernel.org
Subject: Re: [PATCH 04/16] arm64: dts: qcom: sm8350: Specify clock-frequency for arch timer
Date: Wed, 1 Dec 2021 16:17:29 -0800	[thread overview]
Message-ID: <YagQmQzk7GorCUVv@ripper> (raw)
In-Reply-To: <0b171f2e-4bbc-a54a-7615-87fd4559fee9@somainline.org>

On Wed 01 Dec 16:00 PST 2021, Konrad Dybcio wrote:

> 
> On 01.12.2021 21:45, Stephen Boyd wrote:
> > Quoting Konrad Dybcio (2021-11-30 11:59:03)
> >> On 30/11/2021 03:05, Stephen Boyd wrote:
> >>> Quoting Konrad Dybcio (2021-11-13 17:27:43)
> >>>> Arch timer runs at 19.2 MHz. Specify the rate in the timer node.
> >>>>
> >>>> Signed-off-by: Konrad Dybcio <konrad.dybcio@somainline.org>
> >>>> ---
> >>>>   arch/arm64/boot/dts/qcom/sm8350.dtsi | 1 +
> >>>>   1 file changed, 1 insertion(+)
> >>>>
> >>>> diff --git a/arch/arm64/boot/dts/qcom/sm8350.dtsi b/arch/arm64/boot/dts/qcom/sm8350.dtsi
> >>>> index a30ba3193d84..60866a20a55c 100644
> >>>> --- a/arch/arm64/boot/dts/qcom/sm8350.dtsi
> >>>> +++ b/arch/arm64/boot/dts/qcom/sm8350.dtsi
> >>>> @@ -2484,5 +2484,6 @@ timer {
> >>>>                               <GIC_PPI 14 (GIC_CPU_MASK_SIMPLE(8) | IRQ_TYPE_LEVEL_LOW)>,
> >>>>                               <GIC_PPI 11 (GIC_CPU_MASK_SIMPLE(8) | IRQ_TYPE_LEVEL_LOW)>,
> >>>>                               <GIC_PPI 10 (GIC_CPU_MASK_SIMPLE(8) | IRQ_TYPE_LEVEL_LOW)>;
> >>>> +               clock-frequency = <19200000>;
> >>> Does the firmware not set the frequency properly?
> >> It does on my device on the current firmware version (it wouldn't really 
> >> boot if it didn't, no?),
> >>
> >> but who knows if it always will, or if it always has been..
> >>
> >>
> >> It's present in downstream too, so I reckon it does not hurt to have it 
> >> here too, even
> >>
> >> for completeness-of-describing-the-machine-properly sake.
> >>
> > No. We don't want dts files to have this. The property is only there to
> > workaround bad firmware that doesn't set the frequency. Please drop this
> > patch.
> 
> After looking at it again, I see I was indeed wrong, and so was this patch.
> 
> Sorry, and green light for dropping..
> 

Can you please send me a patch that reverts the change as I merged it
into my -next branch already? Both to simplify for me and to document
why it shouldn't be here for others to refer to in the future.

Thanks,
Bjorn