linux-kernel - Re: [PATCH v2 1/3] soc: qcom: ice: Add OPP-based clock scaling support for ICE

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <aTpbZHRUj6QF3KSE@hu-arakshit-hyd.qualcomm.com>
Date: Thu, 11 Dec 2025 11:19:24 +0530
From: Abhinaba Rakshit <abhinaba.rakshit@....qualcomm.com>
To: Konrad Dybcio <konrad.dybcio@....qualcomm.com>
Cc: Bjorn Andersson <andersson@...nel.org>,
        Konrad Dybcio <konradybcio@...nel.org>,
        Manivannan Sadhasivam <mani@...nel.org>,
        "James E.J. Bottomley" <James.Bottomley@...senpartnership.com>,
        "Martin K. Petersen" <martin.petersen@...cle.com>,
        Neeraj Soni <neeraj.soni@....qualcomm.com>,
        linux-arm-msm@...r.kernel.org, linux-kernel@...r.kernel.org,
        linux-scsi@...r.kernel.org
Subject: Re: [PATCH v2 1/3] soc: qcom: ice: Add OPP-based clock scaling
 support for ICE

On Fri, Nov 21, 2025 at 02:46:52PM +0100, Konrad Dybcio wrote:
> On 11/21/25 11:36 AM, Abhinaba Rakshit wrote:
> > Register optional operation-points-v2 table for ICE device
> > and aquire its minimum and maximum frequency during ICE
> > device probe.
> > 
> > Introduce clock scaling API qcom_ice_scale_clk which scale ICE
> > core clock if valid (non-zero) frequencies are obtained from
> > OPP-table. Zero min and max (default values) frequencies depicts
> > clock scaling is disabled.
> > 
> > When an ICE-device specific OPP table is available, use the PM OPP
> > framework to manage frequency scaling and maintain proper power-domain
> > constraints. For legacy targets without an ICE-device specific OPP table,
> > fall back to the standard clock framework APIs to set the frequency.
> 
> You can still set a frequency through OPP APIs if the table is empty
> (and one is always created even if devm_pm_opp_of_add_table() fails)
> 
> [...]
> 
> >  	/*
> >  	 * Legacy DT binding uses different clk names for each consumer,
> > -	 * so lets try those first. If none of those are a match, it means
> > -	 * the we only have one clock and it is part of the dedicated DT node.
> > -	 * Also, enable the clock before we check what HW version the driver
> > -	 * supports.
> > +	 * so lets try those first. Also get its corresponding clock index.
> > +	 */
> 
> I would argue *not* setting the rate on targets utilizing a binding without
> an OPP table for the ICE is probably a smart thing to do, because we may
> brownout the SoC this way

Also, adding to the previous reply, if we completly remove clock scaling for any target
we see a major degradation in IO throughput. On targets like Talos (target with legacy binding),
we observed around ~50%-70% degradation of IO performance if we don't keep the ICE clock
on par with storage clock and leaving it operate at default frequency.

> 
> Konrad