linux-kernel - Re: [PATCH v5 09/18] arm64: dts: qcom: qrb5165-rb5: model the PMU of the QCA6391

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

Date: Tue, 20 Feb 2024 14:38:33 +0100
From: Bartosz Golaszewski <brgl@...ev.pl>
To: Mark Brown <broonie@...nel.org>
Cc: Marcel Holtmann <marcel@...tmann.org>, Luiz Augusto von Dentz <luiz.dentz@...il.com>, 
	"David S . Miller" <davem@...emloft.net>, Eric Dumazet <edumazet@...gle.com>, 
	Jakub Kicinski <kuba@...nel.org>, Paolo Abeni <pabeni@...hat.com>, Rob Herring <robh@...nel.org>, 
	Krzysztof Kozlowski <krzysztof.kozlowski+dt@...aro.org>, Conor Dooley <conor+dt@...nel.org>, 
	Kalle Valo <kvalo@...nel.org>, Bjorn Andersson <andersson@...nel.org>, 
	Konrad Dybcio <konrad.dybcio@...aro.org>, Liam Girdwood <lgirdwood@...il.com>, 
	Catalin Marinas <catalin.marinas@....com>, Will Deacon <will@...nel.org>, 
	Bjorn Helgaas <bhelgaas@...gle.com>, Saravana Kannan <saravanak@...gle.com>, 
	Geert Uytterhoeven <geert+renesas@...der.be>, Arnd Bergmann <arnd@...db.de>, 
	Neil Armstrong <neil.armstrong@...aro.org>, Marek Szyprowski <m.szyprowski@...sung.com>, 
	Alex Elder <elder@...aro.org>, Srini Kandagatla <srinivas.kandagatla@...aro.org>, 
	Greg Kroah-Hartman <gregkh@...uxfoundation.org>, Abel Vesa <abel.vesa@...aro.org>, 
	Manivannan Sadhasivam <mani@...nel.org>, Lukas Wunner <lukas@...ner.de>, 
	Dmitry Baryshkov <dmitry.baryshkov@...aro.org>, linux-bluetooth@...r.kernel.org, 
	netdev@...r.kernel.org, devicetree@...r.kernel.org, 
	linux-kernel@...r.kernel.org, linux-wireless@...r.kernel.org, 
	linux-arm-msm@...r.kernel.org, linux-arm-kernel@...ts.infradead.org, 
	linux-pci@...r.kernel.org, linux-pm@...r.kernel.org, 
	Bartosz Golaszewski <bartosz.golaszewski@...aro.org>
Subject: Re: [PATCH v5 09/18] arm64: dts: qcom: qrb5165-rb5: model the PMU of
 the QCA6391

On Tue, Feb 20, 2024 at 2:31 PM Mark Brown <broonie@...nel.org> wrote:
>
> On Tue, Feb 20, 2024 at 12:16:10PM +0100, Bartosz Golaszewski wrote:
> > On Mon, Feb 19, 2024 at 8:59 PM Mark Brown <broonie@...nel.org> wrote:
> > > On Mon, Feb 19, 2024 at 07:48:20PM +0100, Bartosz Golaszewski wrote:
>
> > > > No, the users don't request any regulators (or rather: software
> > > > representations thereof) because - as per the cover letter - no
> > > > regulators are created by the PMU driver. This is what is physically
> > > > on the board - as the schematics and the datasheet define it. I took
>
> > > The above makes no sense.  How can constraints be "what is physically on
> > > the board", particularly variable constrants when there isn't even a
> > > consumer?  What values are you taking from which documentation?
>
> > The operating conditions for PMU outputs. I took them from a
> > confidential datasheet. There's a table for input constraints and
> > possible output values.
>
> That sounds like you're just putting the maximum range of voltages that
> the PMU can output in there.  This is a fundamental misunderstanding of
> what the constraints are for, the constraints exist to specify what is
> safe on a specific board which will in essentially all cases be much
> more restricted.  The regulator driver should describe whatever the PMU
> can support by itself, the constraints whatever is actually safe and
> functional on the specific board.
>

Ok, got it. Yeah I misunderstood that, but I think it's maybe the
second or third time I'm adding a regulators node myself to DT. I'll
change that.

> > And what do you mean by there not being any consumers? The WLAN and BT
> > *are* the consumers.
>
> There are no drivers that bind to the regulators and vary the voltages
> at runtime.
>

Even with the above misunderstanding clarified: so what? DT is the
representation of hardware. There's nothing that obligates us to model
DT sources in drivers 1:1.

> > > > the values from the docs verbatim. In C, we create a power sequencing
> > > > provider which doesn't use the regulator framework at all.
>
> > > For something that doesn't use the regulator framework at all what
> > > appears to be a provider in patch 16 ("power: pwrseq: add a driver for
> > > the QCA6390 PMU module") seems to have a lot of regualtor API calls?
>
> > This driver is a power sequencing *provider* but also a regulator
> > *consumer*. It gets regulators from the host and exposes a power
> > sequencer to *its* consumers (WLAN and BT). On DT it exposes
> > regulators (LDO outputs of the PMU) but we don't instantiate them in
> > C.
>
> Right, which sounds a lot like being a user of the regualtor framework.

Ok, I meant "user" as a regulator provider but maybe I wasn't clear enough.

Bart