Re: [PATCH v3 12/17] arm64: dts: nvidia: Add EL2 virtual timer interrupt

Re: [PATCH v3 12/17] arm64: dts: nvidia: Add EL2 virtual timer interrupt

[Jon Hunter]

Hi Marc,

On 23/05/2026 15:02, Marc Zyngier wrote:
> The ARMv8.2 based CPUs used in a number of nvidia SoCs are missing
> the EL2 virtual timer interrupt. Add it.
> 
> Signed-off-by: Marc Zyngier <maz@kernel.org>
> ---
>   arch/arm64/boot/dts/nvidia/tegra194.dtsi | 2 ++
>   arch/arm64/boot/dts/nvidia/tegra234.dtsi | 3 ++-
>   2 files changed, 4 insertions(+), 1 deletion(-)
> 
> diff --git a/arch/arm64/boot/dts/nvidia/tegra194.dtsi b/arch/arm64/boot/dts/nvidia/tegra194.dtsi
> index 849694f751d90..45cc180ac9973 100644
> --- a/arch/arm64/boot/dts/nvidia/tegra194.dtsi
> +++ b/arch/arm64/boot/dts/nvidia/tegra194.dtsi
> @@ -3163,6 +3163,8 @@ timer {
>   			     <GIC_PPI 11
>   				(GIC_CPU_MASK_SIMPLE(4) | IRQ_TYPE_LEVEL_LOW)>,
>   			     <GIC_PPI 10
> +				(GIC_CPU_MASK_SIMPLE(4) | IRQ_TYPE_LEVEL_LOW)>,
> +			     <GIC_PPI 12
>   				(GIC_CPU_MASK_SIMPLE(4) | IRQ_TYPE_LEVEL_LOW)>;
>   		interrupt-parent = <&gic>;
>   		always-on;
> diff --git a/arch/arm64/boot/dts/nvidia/tegra234.dtsi b/arch/arm64/boot/dts/nvidia/tegra234.dtsi
> index 04a95b6658caa..ab9813f9ba30c 100644
> --- a/arch/arm64/boot/dts/nvidia/tegra234.dtsi
> +++ b/arch/arm64/boot/dts/nvidia/tegra234.dtsi
> @@ -5872,7 +5872,8 @@ timer {
>   		interrupts = <GIC_PPI 13 (GIC_CPU_MASK_SIMPLE(4) | IRQ_TYPE_LEVEL_LOW)>,
>   			     <GIC_PPI 14 (GIC_CPU_MASK_SIMPLE(4) | IRQ_TYPE_LEVEL_LOW)>,
>   			     <GIC_PPI 11 (GIC_CPU_MASK_SIMPLE(4) | IRQ_TYPE_LEVEL_LOW)>,
> -			     <GIC_PPI 10 (GIC_CPU_MASK_SIMPLE(4) | IRQ_TYPE_LEVEL_LOW)>;
> +			     <GIC_PPI 10 (GIC_CPU_MASK_SIMPLE(4) | IRQ_TYPE_LEVEL_LOW)>,
> +			     <GIC_PPI 12 (GIC_CPU_MASK_SIMPLE(4) | IRQ_TYPE_LEVEL_LOW)>;
>   		interrupt-parent = <&gic>;
>   		always-on;
>   	};

Sorry for the delay. I gave this a test because I observed the warning 
that was added on the Tegra194 and Tegra234 platforms. This change fixes 
the warning for Tegra234, but on Tegra194 the platforms I tested hang on 
boot. It appears to be similar to the issue that Marek saw on his 
platforms and so I am wondering if Tegra194 also doesn't have this wired up?

Was there any resolution to the issue reported by Marek?

FYI, the Tegra194 SoC has the 'NVIDIA Carmel ARM v8.2' CPUs [0].

Jon

[0] 
https://www.nvidia.com/en-gb/autonomous-machines/embedded-systems/jetson-xavier-series/ 


-- 
nvpublic

Re: [PATCH v3 12/17] arm64: dts: nvidia: Add EL2 virtual timer interrupt

[Marc Zyngier]

Hi Jon,

On Tue, 30 Jun 2026 10:42:34 +0100,
Jon Hunter <jonathanh@nvidia.com> wrote:
> 
> Hi Marc,
> 
> On 23/05/2026 15:02, Marc Zyngier wrote:
> > The ARMv8.2 based CPUs used in a number of nvidia SoCs are missing
> > the EL2 virtual timer interrupt. Add it.
> > 
> > Signed-off-by: Marc Zyngier <maz@kernel.org>
> > ---
> >   arch/arm64/boot/dts/nvidia/tegra194.dtsi | 2 ++
> >   arch/arm64/boot/dts/nvidia/tegra234.dtsi | 3 ++-
> >   2 files changed, 4 insertions(+), 1 deletion(-)
> > 
> > diff --git a/arch/arm64/boot/dts/nvidia/tegra194.dtsi b/arch/arm64/boot/dts/nvidia/tegra194.dtsi
> > index 849694f751d90..45cc180ac9973 100644
> > --- a/arch/arm64/boot/dts/nvidia/tegra194.dtsi
> > +++ b/arch/arm64/boot/dts/nvidia/tegra194.dtsi
> > @@ -3163,6 +3163,8 @@ timer {
> >   			     <GIC_PPI 11
> >   				(GIC_CPU_MASK_SIMPLE(4) | IRQ_TYPE_LEVEL_LOW)>,
> >   			     <GIC_PPI 10
> > +				(GIC_CPU_MASK_SIMPLE(4) | IRQ_TYPE_LEVEL_LOW)>,
> > +			     <GIC_PPI 12
> >   				(GIC_CPU_MASK_SIMPLE(4) | IRQ_TYPE_LEVEL_LOW)>;
> >   		interrupt-parent = <&gic>;
> >   		always-on;
> > diff --git a/arch/arm64/boot/dts/nvidia/tegra234.dtsi b/arch/arm64/boot/dts/nvidia/tegra234.dtsi
> > index 04a95b6658caa..ab9813f9ba30c 100644
> > --- a/arch/arm64/boot/dts/nvidia/tegra234.dtsi
> > +++ b/arch/arm64/boot/dts/nvidia/tegra234.dtsi
> > @@ -5872,7 +5872,8 @@ timer {
> >   		interrupts = <GIC_PPI 13 (GIC_CPU_MASK_SIMPLE(4) | IRQ_TYPE_LEVEL_LOW)>,
> >   			     <GIC_PPI 14 (GIC_CPU_MASK_SIMPLE(4) | IRQ_TYPE_LEVEL_LOW)>,
> >   			     <GIC_PPI 11 (GIC_CPU_MASK_SIMPLE(4) | IRQ_TYPE_LEVEL_LOW)>,
> > -			     <GIC_PPI 10 (GIC_CPU_MASK_SIMPLE(4) | IRQ_TYPE_LEVEL_LOW)>;
> > +			     <GIC_PPI 10 (GIC_CPU_MASK_SIMPLE(4) | IRQ_TYPE_LEVEL_LOW)>,
> > +			     <GIC_PPI 12 (GIC_CPU_MASK_SIMPLE(4) | IRQ_TYPE_LEVEL_LOW)>;
> >   		interrupt-parent = <&gic>;
> >   		always-on;
> >   	};
> 
> Sorry for the delay. I gave this a test because I observed the warning
> that was added on the Tegra194 and Tegra234 platforms. This change
> fixes the warning for Tegra234, but on Tegra194 the platforms I tested
> hang on boot. It appears to be similar to the issue that Marek saw on
> his platforms and so I am wondering if Tegra194 also doesn't have this
> wired up?

I think you are in a better position than me to find out. It also
could be a firmware issue not making the PPI a Group-1 interrupt, and
therefore not allow Linux to configure the interrupt.

> 
> Was there any resolution to the issue reported by Marek?
> 
> FYI, the Tegra194 SoC has the 'NVIDIA Carmel ARM v8.2' CPUs [0].

There is no resolution so far. Florian was going to check what the
deal is with the Broadcom-related systems, but hasn't come back with
an answer yet.

The possibilities are as follows:

- remove the interrupt for the EL2 virtual timer and live with the
  warning

- add a patch such as [1], which should document the reason why this
  is now working (and fallback to the EL2 physical timer)

I'm happy either way, as long as we know exactly what we are dealing
with on each affected platform.

Thanks,

	M.

[1] https://lore.kernel.org/all/878q898ulx.wl-maz@kernel.org/

-- 
Without deviation from the norm, progress is not possible.

Re: [PATCH v3 12/17] arm64: dts: nvidia: Add EL2 virtual timer interrupt

[Jon Hunter]

Hi Marc,

On 30/06/2026 11:54, Marc Zyngier wrote:

...

>> Sorry for the delay. I gave this a test because I observed the warning
>> that was added on the Tegra194 and Tegra234 platforms. This change
>> fixes the warning for Tegra234, but on Tegra194 the platforms I tested
>> hang on boot. It appears to be similar to the issue that Marek saw on
>> his platforms and so I am wondering if Tegra194 also doesn't have this
>> wired up?
> 
> I think you are in a better position than me to find out. It also
> could be a firmware issue not making the PPI a Group-1 interrupt, and
> therefore not allow Linux to configure the interrupt.

Yes absolutely. I will see what I can find out.

>> Was there any resolution to the issue reported by Marek?
>>
>> FYI, the Tegra194 SoC has the 'NVIDIA Carmel ARM v8.2' CPUs [0].
> 
> There is no resolution so far. Florian was going to check what the
> deal is with the Broadcom-related systems, but hasn't come back with
> an answer yet.
> 
> The possibilities are as follows:
> 
> - remove the interrupt for the EL2 virtual timer and live with the
>    warning
> 
> - add a patch such as [1], which should document the reason why this
>    is now working (and fallback to the EL2 physical timer)
> 
> I'm happy either way, as long as we know exactly what we are dealing
> with on each affected platform.

I would like to get the warning fixed for Tegra234. Do you want to split 
that part out of your patch and then I can test and we can at least fix 
for that device while I see whats up with Tegra194?

Jon

-- 
nvpublic

Re: [PATCH v3 12/17] arm64: dts: nvidia: Add EL2 virtual timer interrupt

[Marc Zyngier]

On Tue, 30 Jun 2026 13:09:20 +0100,
Jon Hunter <jonathanh@nvidia.com> wrote:
> 
> Hi Marc,
> 
> On 30/06/2026 11:54, Marc Zyngier wrote:
> 
> ...
> 
> >> Sorry for the delay. I gave this a test because I observed the warning
> >> that was added on the Tegra194 and Tegra234 platforms. This change
> >> fixes the warning for Tegra234, but on Tegra194 the platforms I tested
> >> hang on boot. It appears to be similar to the issue that Marek saw on
> >> his platforms and so I am wondering if Tegra194 also doesn't have this
> >> wired up?
> > 
> > I think you are in a better position than me to find out. It also
> > could be a firmware issue not making the PPI a Group-1 interrupt, and
> > therefore not allow Linux to configure the interrupt.
> 
> Yes absolutely. I will see what I can find out.
> 
> >> Was there any resolution to the issue reported by Marek?
> >> 
> >> FYI, the Tegra194 SoC has the 'NVIDIA Carmel ARM v8.2' CPUs [0].
> > 
> > There is no resolution so far. Florian was going to check what the
> > deal is with the Broadcom-related systems, but hasn't come back with
> > an answer yet.
> > 
> > The possibilities are as follows:
> > 
> > - remove the interrupt for the EL2 virtual timer and live with the
> >    warning
> > 
> > - add a patch such as [1], which should document the reason why this
> >    is now working (and fallback to the EL2 physical timer)
> > 
> > I'm happy either way, as long as we know exactly what we are dealing
> > with on each affected platform.
> 
> I would like to get the warning fixed for Tegra234. Do you want to
> split that part out of your patch and then I can test and we can at
> least fix for that device while I see whats up with Tegra194?

Getting the warning fixed for Tegra234 is probably limited to fixing
the DT, Feel free to lift that from my original patch and slap your
name on it.

Also consider fixing most of the PPI interrupt specifiers while you're
at it, because aside from the PMU, they are all awfully wrong.

Thanks,

	M.

-- 
Without deviation from the norm, progress is not possible.