linux-kernel - Re: [PATCH v8 3/4] arm64: dts: qcom: sm6115: Add EUD dt node and dwc3 connector

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:   Mon, 17 Jul 2023 20:26:37 +0200
From:   Stephan Gerhold <stephan@...hold.net>
To:     Bhupesh Sharma <bhupesh.sharma@...aro.org>
Cc:     linux-arm-msm@...r.kernel.org, devicetree@...r.kernel.org,
        linux-usb@...r.kernel.org, agross@...nel.org, andersson@...nel.org,
        konrad.dybcio@...aro.org, linux-kernel@...r.kernel.org,
        bhupesh.linux@...il.com, robh+dt@...nel.org,
        krzysztof.kozlowski+dt@...aro.org, krzysztof.kozlowski@...aro.org,
        quic_schowdhu@...cinc.com, gregkh@...uxfoundation.org
Subject: Re: [PATCH v8 3/4] arm64: dts: qcom: sm6115: Add EUD dt node and
 dwc3 connector

On Mon, Jul 17, 2023 at 11:33:40PM +0530, Bhupesh Sharma wrote:
> On Mon, 17 Jul 2023 at 16:15, Stephan Gerhold <stephan@...hold.net> wrote:
> >
> > On Mon, Jul 17, 2023 at 04:02:35PM +0530, Bhupesh Sharma wrote:
> > > Add the Embedded USB Debugger(EUD) device tree node for
> > > SM6115 / SM4250 SoC.
> > >
> > > The node contains EUD base register region, EUD mode manager
> > > register region and TCSR Base register region along with the
> > > interrupt entry.
> > >
> > > [...]
> > >
> > > Reviewed-by: Konrad Dybcio <konrad.dybcio@...aro.org>
> > > Signed-off-by: Bhupesh Sharma <bhupesh.sharma@...aro.org>
> > > ---
> > >  arch/arm64/boot/dts/qcom/sm6115.dtsi | 50 ++++++++++++++++++++++++++++
> > >  1 file changed, 50 insertions(+)
> > >
> > > diff --git a/arch/arm64/boot/dts/qcom/sm6115.dtsi b/arch/arm64/boot/dts/qcom/sm6115.dtsi
> > > index 839c603512403..db45337c1082c 100644
> > > --- a/arch/arm64/boot/dts/qcom/sm6115.dtsi
> > > +++ b/arch/arm64/boot/dts/qcom/sm6115.dtsi
> > > [...]
> > > @@ -789,6 +801,37 @@ gcc: clock-controller@...0000 {
> > >                       #power-domain-cells = <1>;
> > >               };
> > >
> > > +             eud: eud@...0000 {
> > > +                     compatible = "qcom,sm6115-eud", "qcom,eud";
> > > +                     reg = <0x0 0x01610000 0x0 0x2000>,
> > > +                           <0x0 0x01612000 0x0 0x1000>,
> > > +                           <0x0 0x003c0000 0x0 0x40000>;
> > > +                     reg-names = "eud-base", "eud-mode-mgr", "tcsr-base";
> >
> > TCSR is a separate hardware block unrelated to the EUD. IMHO it
> > shouldn't be listed as "reg" here.
> >
> > Typically we describe it as syscon and then reference it from other
> > nodes. See e.g. sm8450.dtsi "tcsr: syscon@...0000" referenced in &scm
> > "qcom,dload-mode = <&tcsr 0x13000>". This is pretty much exactly the
> > same use case as you have. It also uses this to write something with
> > qcom_scm_io_writel() at the end.
> 
> That was discussed a bit during v1 patchset review. Basically, if we
> use a tcsr syscon approach here, we will need to define a 'qcom,xx'
> vendor specific dt-property and use something like this in the eud
> node:
> 
> qcom,eud-sec-reg = <&tcsr_reg yyyy>
> 
> which would be then used by the eud driver (via
> syscon_regmap_lookup_by_phandle()).
> 
> But for sm6115 / qcm2290 this would be an over complicated solution as
> normally the eud driver (say sc7280) doesn't need tcsr based secure
> mode manager access. So defining a new soc / vendor specific
> dt-property might be an overkill.
> 

IMO a vendor-specific DT property is still better than messing up the
device separation in the device tree. The same "tcsr-base" reg would
also appear on the actual tcsr syscon device tree node. Having two
device tree nodes with the same reg region is generally not valid.

Something like qcom,eud-sec-reg = <&tcsr_reg yyyy> would at least make
clear that this points into a region that is shared between multiple
different devices, while adding it as reg suggests that TCSR belongs
exclusively to EUD.

Thanks,
Stephan