Re: [PATCH v4 12/12] arm64: dts: qcom: Add EL2 overlay for Lemans

[Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>]

On 10/8/25 9:32 AM, Mukesh Ojha wrote:
> On Tue, Oct 07, 2025 at 02:55:04PM -0700, Manivannan Sadhasivam wrote:
>> On Tue, Oct 07, 2025 at 10:18:57PM +0530, Mukesh Ojha wrote:
>>> All the Lemans IOT variants boards are using Gunyah hypervisor which
>>> means that, so far, Linux-based OS could only boot in EL1 on those
>>> devices.  However, it is possible for us to boot Linux at EL2 on these
>>> devices [1].
>>>
>>> When running under Gunyah, remote processor firmware IOMMU streams is
>>> controlled by the Gunyah however when Linux take ownership of it in EL2,
>>> It need to configure it properly to use remote processor.
>>>
>>> Add a EL2-specific DT overlay and apply it to Lemans IOT variant
>>> devices to create -el2.dtb for each of them alongside "normal" dtb.
>>>
>>> [1]
>>> https://docs.qualcomm.com/bundle/publicresource/topics/80-70020-4/boot-developer-touchpoints.html#uefi
>>>
>>> Signed-off-by: Mukesh Ojha <mukesh.ojha@oss.qualcomm.com>
>>> ---

[...]

>>> +&iris {
>>> +	/* TODO: Add video-firmware iommus to start IRIS from EL2 */
>>
>> So the missing 'iommus' property is the only blocker to enable IRIS?
> 
> Yes, but this would require some changes in the driver as well to create
> a platform device for the firmware and IOMMU domain, and to attach the
> firmware device to the domain.

"yes, but actually no"

> 
> There was some discussion around this in v2[1], where I added support
> for video. However, the change involved hooking into the video-firmware
> property, which was used for Venus.  There were concerns[2] about following
> the same approach for Iris, and due to that, we have dropped video
> support for now.  However, an RFC has been posted from our side to
> handle such scenarios[3], as of now unclear about the direction for
> Iris.
> 
> 
> [1] https://lore.kernel.org/lkml/aKooCFoV3ZYwOMRx@linaro.org/
> 
> [2] https://lore.kernel.org/lkml/20250627-video_cb-v3-0-51e18c0ffbce@quicinc.com/
> 
> [3] https://lore.kernel.org/all/20250928171718.436440-1-charan.kalla@oss.qualcomm.com/
> 
> [4] https://lore.kernel.org/lkml/4a32bbec-2baf-4210-a7c1-1ddcd45d30c8@oss.qualcomm.com/

Reword the comment to "/* More driver work is needed */" since there
is nothing more you can do at the moment and you most likely don't
want to block your series on resolving the above entanglement

Konrad