linux-kernel - Re: [PATCH v8 5/7] arm64: dts: qcom: sdm845: Add DT nodes for the TBUs

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: <CAA8EJprGja_L6QXVmfSqoeMKCmSrf-qBeAWNToZDe++AsZ+Nyg@mail.gmail.com>
Date: Wed, 3 Jul 2024 12:54:48 +0300
From: Dmitry Baryshkov <dmitry.baryshkov@...aro.org>
To: Will Deacon <will@...nel.org>
Cc: Georgi Djakov <djakov@...nel.org>, Georgi Djakov <quic_c_gdjako@...cinc.com>, robin.murphy@....com, 
	joro@...tes.org, iommu@...ts.linux.dev, robh+dt@...nel.org, 
	krzysztof.kozlowski+dt@...aro.org, conor+dt@...nel.org, 
	devicetree@...r.kernel.org, andersson@...nel.org, konrad.dybcio@...aro.org, 
	robdclark@...il.com, linux-arm-kernel@...ts.infradead.org, 
	linux-kernel@...r.kernel.org, linux-arm-msm@...r.kernel.org, 
	quic_cgoldswo@...cinc.com, quic_sukadev@...cinc.com, quic_pdaly@...cinc.com, 
	quic_sudaraja@...cinc.com
Subject: Re: [PATCH v8 5/7] arm64: dts: qcom: sdm845: Add DT nodes for the TBUs

On Tue, 2 Jul 2024 at 19:39, Will Deacon <will@...nel.org> wrote:
>
> On Tue, Jun 25, 2024 at 03:59:27PM +0300, Dmitry Baryshkov wrote:
> > On Tue, 25 Jun 2024 at 15:57, Georgi Djakov <djakov@...nel.org> wrote:
> > >
> > > On 25.06.24 10:50, Dmitry Baryshkov wrote:
> > > > On Fri, 14 Jun 2024 at 21:05, Dmitry Baryshkov
> > > > <dmitry.baryshkov@...aro.org> wrote:
> > > >>
> > > >> On Wed, 17 Apr 2024 at 16:39, Georgi Djakov <quic_c_gdjako@...cinc.com> wrote:
> > > >>>
> > > >>> Add the device-tree nodes for the TBUs (translation buffer units) that
> > > >>> are present on the sdm845 platforms. The TBUs can be used debug the
> > > >>> kernel and provide additional information when a context faults occur.
> > > >>>
> > > >>> Describe the all registers, clocks, interconnects and power-domain
> > > >>> resources that are needed for each of the TBUs.
> > > >>>
> > > >>> Signed-off-by: Georgi Djakov <quic_c_gdjako@...cinc.com>
> > > >>
> > > >> This patch now prevents interconnect drivers from hitting the sync
> > > >> state on SDM845.
> > > >> The TBU driver is enabled only when the ARM_SMMU_QCOM_DEBUG is
> > > >> enabled, which is not a typical case on a normal system:
> > > >
> > > > Georgi, before I start acting like a bull in a china shop and sending
> > > > reverts, any update from your side?
> > >
> > > Hi Dmitry!
> > > Thanks for the report! We can easily add status = "disabled" to the DT
> > > nodes, but please give me some time to take a look what would be the best
> > > way to handle this, as i was out last week and now i am still catching up.
> >
> > I think the simplest thing would be to move the TBU driver to the
> > arm-qcom-smmu.c instead of having it in the -debug.c
>
> The TBUs aren't used for anything other than debugging, so I'd really
> rather they live with the debug code.

The problem is that not having any driver bound to the TBU devices
prevents interconnect drivers from hitting the sync state (and thus
lowering the bandwidth to the requested values).
Being that late in the development cycle, I think we should fix the
issue in the fastest way. So I'm close to sending a revert of the DT
changes. Setting status = "disabled" doesn't qualify as a logical
change as having TBU units enabled on a platform-to-platform basis
doesn't make sense.

-- 
With best wishes
Dmitry