Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Date:   Thu, 28 Oct 2021 11:18:45 -0700
From:   Bjorn Andersson <bjorn.andersson@linaro.org>
To:     Konrad Dybcio <konrad.dybcio@somainline.org>
Cc:     Yassine Oudjana <y.oudjana@protonmail.com>,
        Andy Gross <agross@kernel.org>,
        Michael Turquette <mturquette@baylibre.com>,
        Stephen Boyd <sboyd@kernel.org>,
        Rob Herring <robh+dt@kernel.org>,
        Ilia Lin <ilia.lin@kernel.org>,
        Viresh Kumar <vireshk@kernel.org>, Nishanth Menon <nm@ti.com>,
        "Rafael J. Wysocki" <rafael@kernel.org>,
        Loic Poulain <loic.poulain@linaro.org>,
        AngeloGioacchino Del Regno 
        <angelogioacchino.delregno@somainline.org>,
        linux-arm-msm@vger.kernel.org, linux-clk@vger.kernel.org,
        devicetree@vger.kernel.org, linux-kernel@vger.kernel.org,
        linux-pm@vger.kernel.org, ~postmarketos/upstreaming@lists.sr.ht,
        phone-devel@vger.kernel.org
Subject: Re: [PATCH 4/8] cpufreq: qcom_cpufreq_nvmem: Simplify reading kryo
 speedbin
Message-ID: <YXrphSSG9+zsZbJ9@ripper>
References: <20211014083016.137441-1-y.oudjana@protonmail.com>
 <20211014083016.137441-5-y.oudjana@protonmail.com>
 <f4af0e40-9e93-dba9-7401-cdfc76ba667e@somainline.org>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <f4af0e40-9e93-dba9-7401-cdfc76ba667e@somainline.org>
Precedence: bulk

On Fri 15 Oct 11:58 PDT 2021, Konrad Dybcio wrote:

> 
> On 14.10.2021 10:32, Yassine Oudjana wrote:
> > In preparation for adding a separate device tree for MSM8996 Pro, skip reading
> > msm-id from smem and just read the speedbin efuse.
> >
> While I'd really like for this to be merged, it's gonna totally wreck backwards
> 
> compatibility.. But then, since APCC was not defined properly before commit
> 
> 0a275a35ceab07 arm64: dts: qcom: msm8996: Make CPUCC actually probe (and work)
> 
> there's only 5.14/5.15 (both of which were non-LTS) which would *actually* break given
> 
> somebody decided that "ah yes, pulling in DTs from these specific mainline kernel releases
> 
> is a good idea"...
> 
> 
> If I were to judge, it would probably be fine to rid the old mechanism..
> 

Given that various people have reported instabilities on db820c in its
current form - and prior to that it was too slow - I think it's fine to
favour getting this sorted out properly over backwards compatibility.

Regards,
Bjorn