[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <8c57ed65-002a-4ddc-9056-6b7908c0ff52@oss.qualcomm.com>
Date: Sat, 31 May 2025 13:11:04 +0200
From: Konrad Dybcio <konrad.dybcio@....qualcomm.com>
To: Renjiang Han <quic_renjiang@...cinc.com>,
Vikash Garodia <quic_vgarodia@...cinc.com>,
Dikshita Agarwal <quic_dikshita@...cinc.com>,
Bryan O'Donoghue <bryan.odonoghue@...aro.org>,
Mauro Carvalho Chehab <mchehab@...nel.org>,
Bjorn Andersson <andersson@...nel.org>,
Konrad Dybcio <konradybcio@...nel.org>, Rob Herring <robh@...nel.org>,
Krzysztof Kozlowski <krzk+dt@...nel.org>,
Conor Dooley <conor+dt@...nel.org>
Cc: linux-media@...r.kernel.org, linux-arm-msm@...r.kernel.org,
linux-kernel@...r.kernel.org, devicetree@...r.kernel.org
Subject: Re: [PATCH v8 1/3] media: venus: pm_helpers: use opp-table for the
frequency
On 5/30/25 6:02 AM, Renjiang Han wrote:
> The frequency value in the opp-table in the device tree and the freq_tbl
> in the driver are the same.
>
> Therefore, update pm_helpers.c to use the opp-table for frequency values
> for the v4 core.
> If getting data from the opp table fails, fall back to using the frequency
> table.
>
> Reviewed-by: Vikash Garodia <quic_vgarodia@...cinc.com>
> Signed-off-by: Renjiang Han <quic_renjiang@...cinc.com>
> ---
You can save yourself some error-checking pain in this patch if you
do something like
ret = devm_pm_opp_of_add_table(dev);
if (ret == -ENODEV) {
for (i = 0; i < freq_tbl_len; i++)
dev_pm_opp_add(dev, freq_tbl[i], 0);
}
to programmatically migrate everyone to OPP..
But - tangent - I'd say efforts to preserve compatibility with DTs that
don't even contain frequency data for a given target are rather futile..
Such descriptions where we only know the frequency (be it the tables that
we currently have, or the constructed OPP tables that only contain frequency
data and no voltage corners) are incomplete, and if the system manages to
not crash if the driver requests a TURBO/max freq, it's all because we got
lucky, as this consumer is not voting on (MM)CX.
That said, it's probably to keep the breakage to minimum, especially
given this is an old driver for old hardware.. I'll add the missing
OPPs across platforms with an intention to drop the hardcoding somewhere
in the future
Konrad
Powered by blists - more mailing lists