[PATCH] iommu/arm: fix ARM_SMMU_QCOM compilation

[PATCH] iommu/arm: fix ARM_SMMU_QCOM compilation

[Arnd Bergmann]

From: Arnd Bergmann <arnd@arndb.de>

My previous bugfix ended up making things worse for the QCOM IOMMU
driver when it forgot to add the Kconfig symbol that is getting used to
control the compilation of the SMMU implementation specific code
for Qualcomm.

Fixes: 424953cf3c66 ("qcom_scm: hide Kconfig symbol")
Reported-by: Daniel Lezcano <daniel.lezcano@linaro.org>
Reported-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reported-by: John Stultz <john.stultz@linaro.org>
Link: https://lore.kernel.org/lkml/20211010023350.978638-1-dmitry.baryshkov@linaro.org/
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
---
In case we want fix it this way after all, here is the patch
I made. Either this one or Dmitry patch from the link above
is required for v5.15
---
 drivers/iommu/Kconfig | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/drivers/iommu/Kconfig b/drivers/iommu/Kconfig
index c5c71b7ab7e8..3eb68fa1b8cc 100644
--- a/drivers/iommu/Kconfig
+++ b/drivers/iommu/Kconfig
@@ -355,6 +355,14 @@ config ARM_SMMU_DISABLE_BYPASS_BY_DEFAULT
 	  'arm-smmu.disable_bypass' will continue to override this
 	  config.
 
+config ARM_SMMU_QCOM
+	def_tristate y
+	depends on ARM_SMMU && ARCH_QCOM
+	select QCOM_SCM
+	help
+	  When running on a Qualcomm platform that has the custom variant
+	  of the ARM SMMU, this needs to be built into the SMMU driver.
+
 config ARM_SMMU_V3
 	tristate "ARM Ltd. System MMU Version 3 (SMMUv3) Support"
 	depends on ARM64
-- 
2.29.2

Re: [PATCH] iommu/arm: fix ARM_SMMU_QCOM compilation

[Will Deacon]

On Tue, Oct 12, 2021 at 05:18:00PM +0200, Arnd Bergmann wrote:
> From: Arnd Bergmann <arnd@arndb.de>
> 
> My previous bugfix ended up making things worse for the QCOM IOMMU
> driver when it forgot to add the Kconfig symbol that is getting used to
> control the compilation of the SMMU implementation specific code
> for Qualcomm.
> 
> Fixes: 424953cf3c66 ("qcom_scm: hide Kconfig symbol")
> Reported-by: Daniel Lezcano <daniel.lezcano@linaro.org>
> Reported-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
> Reported-by: John Stultz <john.stultz@linaro.org>
> Link: https://lore.kernel.org/lkml/20211010023350.978638-1-dmitry.baryshkov@linaro.org/
> Signed-off-by: Arnd Bergmann <arnd@arndb.de>
> ---
> In case we want fix it this way after all, here is the patch
> I made. Either this one or Dmitry patch from the link above
> is required for v5.15
> ---
>  drivers/iommu/Kconfig | 8 ++++++++
>  1 file changed, 8 insertions(+)
> 
> diff --git a/drivers/iommu/Kconfig b/drivers/iommu/Kconfig
> index c5c71b7ab7e8..3eb68fa1b8cc 100644
> --- a/drivers/iommu/Kconfig
> +++ b/drivers/iommu/Kconfig
> @@ -355,6 +355,14 @@ config ARM_SMMU_DISABLE_BYPASS_BY_DEFAULT
>  	  'arm-smmu.disable_bypass' will continue to override this
>  	  config.
>  
> +config ARM_SMMU_QCOM
> +	def_tristate y
> +	depends on ARM_SMMU && ARCH_QCOM
> +	select QCOM_SCM
> +	help
> +	  When running on a Qualcomm platform that has the custom variant
> +	  of the ARM SMMU, this needs to be built into the SMMU driver.
> +

FWIW, I prefer this solution over changing the driver code, so:

Acked-by: Will Deacon <will@kernel.org>

I assume you'll be getting this fixed for 5.15?

Cheers,

Will

Re: [PATCH] iommu/arm: fix ARM_SMMU_QCOM compilation

[Arnd Bergmann]

On Wed, Oct 13, 2021 at 9:58 AM Will Deacon <will@kernel.org> wrote:
> On Tue, Oct 12, 2021 at 05:18:00PM +0200, Arnd Bergmann wrote:
> > From: Arnd Bergmann <arnd@arndb.de>
> >
> > My previous bugfix ended up making things worse for the QCOM IOMMU
> > driver when it forgot to add the Kconfig symbol that is getting used to
> > control the compilation of the SMMU implementation specific code
> > for Qualcomm.
> >
> > Fixes: 424953cf3c66 ("qcom_scm: hide Kconfig symbol")
> > Reported-by: Daniel Lezcano <daniel.lezcano@linaro.org>
> > Reported-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
> > Reported-by: John Stultz <john.stultz@linaro.org>
> > Link: https://lore.kernel.org/lkml/20211010023350.978638-1-dmitry.baryshkov@linaro.org/
> > Signed-off-by: Arnd Bergmann <arnd@arndb.de>
> > ---
> > In case we want fix it this way after all, here is the patch
> > I made. Either this one or Dmitry patch from the link above
> > is required for v5.15
> > ---
> >  drivers/iommu/Kconfig | 8 ++++++++
> >  1 file changed, 8 insertions(+)
> >
> > diff --git a/drivers/iommu/Kconfig b/drivers/iommu/Kconfig
> > index c5c71b7ab7e8..3eb68fa1b8cc 100644
> > --- a/drivers/iommu/Kconfig
> > +++ b/drivers/iommu/Kconfig
> > @@ -355,6 +355,14 @@ config ARM_SMMU_DISABLE_BYPASS_BY_DEFAULT
> >         'arm-smmu.disable_bypass' will continue to override this
> >         config.
> >
> > +config ARM_SMMU_QCOM
> > +     def_tristate y
> > +     depends on ARM_SMMU && ARCH_QCOM
> > +     select QCOM_SCM
> > +     help
> > +       When running on a Qualcomm platform that has the custom variant
> > +       of the ARM SMMU, this needs to be built into the SMMU driver.
> > +
>
> FWIW, I prefer this solution over changing the driver code, so:
>
> Acked-by: Will Deacon <will@kernel.org>
>
> I assume you'll be getting this fixed for 5.15?

I was hoping you and Joerg could just pick your preferred patch
into the iommu fixes tree for v5.15.

I currently have nothing else pending for my asm-generic tree that
introduced the regression, but I can take it through there if that helps
you.

         Arnd

Re: [PATCH] iommu/arm: fix ARM_SMMU_QCOM compilation

[Will Deacon]

Hi Arnd,

On Wed, Oct 13, 2021 at 10:33:55AM +0200, Arnd Bergmann wrote:
> On Wed, Oct 13, 2021 at 9:58 AM Will Deacon <will@kernel.org> wrote:
> > On Tue, Oct 12, 2021 at 05:18:00PM +0200, Arnd Bergmann wrote:
> > > From: Arnd Bergmann <arnd@arndb.de>
> > >
> > > My previous bugfix ended up making things worse for the QCOM IOMMU
> > > driver when it forgot to add the Kconfig symbol that is getting used to
> > > control the compilation of the SMMU implementation specific code
> > > for Qualcomm.
> > >
> > > Fixes: 424953cf3c66 ("qcom_scm: hide Kconfig symbol")
> > > Reported-by: Daniel Lezcano <daniel.lezcano@linaro.org>
> > > Reported-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
> > > Reported-by: John Stultz <john.stultz@linaro.org>
> > > Link: https://lore.kernel.org/lkml/20211010023350.978638-1-dmitry.baryshkov@linaro.org/
> > > Signed-off-by: Arnd Bergmann <arnd@arndb.de>
> > > ---
> > > In case we want fix it this way after all, here is the patch
> > > I made. Either this one or Dmitry patch from the link above
> > > is required for v5.15
> > > ---
> > >  drivers/iommu/Kconfig | 8 ++++++++
> > >  1 file changed, 8 insertions(+)
> > >
> > > diff --git a/drivers/iommu/Kconfig b/drivers/iommu/Kconfig
> > > index c5c71b7ab7e8..3eb68fa1b8cc 100644
> > > --- a/drivers/iommu/Kconfig
> > > +++ b/drivers/iommu/Kconfig
> > > @@ -355,6 +355,14 @@ config ARM_SMMU_DISABLE_BYPASS_BY_DEFAULT
> > >         'arm-smmu.disable_bypass' will continue to override this
> > >         config.
> > >
> > > +config ARM_SMMU_QCOM
> > > +     def_tristate y
> > > +     depends on ARM_SMMU && ARCH_QCOM
> > > +     select QCOM_SCM
> > > +     help
> > > +       When running on a Qualcomm platform that has the custom variant
> > > +       of the ARM SMMU, this needs to be built into the SMMU driver.
> > > +
> >
> > FWIW, I prefer this solution over changing the driver code, so:
> >
> > Acked-by: Will Deacon <will@kernel.org>
> >
> > I assume you'll be getting this fixed for 5.15?
> 
> I was hoping you and Joerg could just pick your preferred patch
> into the iommu fixes tree for v5.15.
> 
> I currently have nothing else pending for my asm-generic tree that
> introduced the regression, but I can take it through there if that helps
> you.

I also don't have any fixes pending, and I don't see any in Joerg's tree so
it's probably quickest if you send it on yourself. Is that ok?

Cheers,

Will

Re: [PATCH] iommu/arm: fix ARM_SMMU_QCOM compilation

[Arnd Bergmann]

On Wed, Oct 13, 2021 at 6:20 PM Will Deacon <will@kernel.org> wrote:
> On Wed, Oct 13, 2021 at 10:33:55AM +0200, Arnd Bergmann wrote:
> > On Wed, Oct 13, 2021 at 9:58 AM Will Deacon <will@kernel.org> wrote:
> > > On Tue, Oct 12, 2021 at 05:18:00PM +0200, Arnd Bergmann wrote:

> > I was hoping you and Joerg could just pick your preferred patch
> > into the iommu fixes tree for v5.15.
> >
> > I currently have nothing else pending for my asm-generic tree that
> > introduced the regression, but I can take it through there if that helps
> > you.
>
> I also don't have any fixes pending, and I don't see any in Joerg's tree so
> it's probably quickest if you send it on yourself. Is that ok?

Sure, no problem. I ended up adding it to the arm/fixes branch of the
soc tree, as I just merged some other fixes there, and it seems as good
as any of the other trees.

      Arnd

Re: [PATCH] iommu/arm: fix ARM_SMMU_QCOM compilation

[Will Deacon]

On Wed, Oct 13, 2021 at 09:31:40PM +0200, Arnd Bergmann wrote:
> On Wed, Oct 13, 2021 at 6:20 PM Will Deacon <will@kernel.org> wrote:
> > On Wed, Oct 13, 2021 at 10:33:55AM +0200, Arnd Bergmann wrote:
> > > On Wed, Oct 13, 2021 at 9:58 AM Will Deacon <will@kernel.org> wrote:
> > > > On Tue, Oct 12, 2021 at 05:18:00PM +0200, Arnd Bergmann wrote:
> 
> > > I was hoping you and Joerg could just pick your preferred patch
> > > into the iommu fixes tree for v5.15.
> > >
> > > I currently have nothing else pending for my asm-generic tree that
> > > introduced the regression, but I can take it through there if that helps
> > > you.
> >
> > I also don't have any fixes pending, and I don't see any in Joerg's tree so
> > it's probably quickest if you send it on yourself. Is that ok?
> 
> Sure, no problem. I ended up adding it to the arm/fixes branch of the
> soc tree, as I just merged some other fixes there, and it seems as good
> as any of the other trees.

Thanks, Arnd!

Will