linux-kernel - Re: [PATCH v2 2/3] dt-bindings: arm: mediatek: mmsys: Add OF graph support for board path

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: Fri, 19 Apr 2024 09:54:02 +0200
From: AngeloGioacchino Del Regno <angelogioacchino.delregno@...labora.com>
To: Rob Herring <robh@...nel.org>
Cc: chunkuang.hu@...nel.org, krzysztof.kozlowski+dt@...aro.org,
 conor+dt@...nel.org, p.zabel@...gutronix.de, airlied@...il.com,
 daniel@...ll.ch, maarten.lankhorst@...ux.intel.com, mripard@...nel.org,
 tzimmermann@...e.de, matthias.bgg@...il.com, shawn.sung@...iatek.com,
 yu-chang.lee@...iatek.com, ck.hu@...iatek.com, jitao.shi@...iatek.com,
 devicetree@...r.kernel.org, linux-kernel@...r.kernel.org,
 dri-devel@...ts.freedesktop.org, linux-mediatek@...ts.infradead.org,
 linux-arm-kernel@...ts.infradead.org, wenst@...omium.org,
 kernel@...labora.com
Subject: Re: [PATCH v2 2/3] dt-bindings: arm: mediatek: mmsys: Add OF graph
 support for board path

Il 10/04/24 21:15, Rob Herring ha scritto:
> On Tue, Apr 09, 2024 at 02:02:10PM +0200, AngeloGioacchino Del Regno wrote:
>> Document OF graph on MMSYS/VDOSYS: this supports up to three DDP paths
>> per HW instance (so potentially up to six displays for multi-vdo SoCs).
>>
>> The MMSYS or VDOSYS is always the first component in the DDP pipeline,
>> so it only supports an output port with multiple endpoints - where each
>> endpoint defines the starting point for one of the (currently three)
>> possible hardware paths.
>>
>> Signed-off-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@...labora.com>
>> ---
>>   .../bindings/arm/mediatek/mediatek,mmsys.yaml | 23 +++++++++++++++++++
>>   1 file changed, 23 insertions(+)
>>
>> diff --git a/Documentation/devicetree/bindings/arm/mediatek/mediatek,mmsys.yaml b/Documentation/devicetree/bindings/arm/mediatek/mediatek,mmsys.yaml
>> index b3c6888c1457..4e9acd966aa5 100644
>> --- a/Documentation/devicetree/bindings/arm/mediatek/mediatek,mmsys.yaml
>> +++ b/Documentation/devicetree/bindings/arm/mediatek/mediatek,mmsys.yaml
>> @@ -93,6 +93,29 @@ properties:
>>     '#reset-cells':
>>       const: 1
>>   
>> +  port:
>> +    $ref: /schemas/graph.yaml#/properties/port
>> +    description:
>> +      Output port node. This port connects the MMSYS/VDOSYS output to
>> +      the first component of one display pipeline, for example one of
>> +      the available OVL or RDMA blocks.
>> +      Some MediaTek SoCs support up to three display outputs per MMSYS.
> 
> I'm have a hard time understanding this, but is it 3 outputs
> simultaneously or connect mmsys to 1 of 3. Generally it's multiple ports
> for the former and endpoints for the latter.
> 

Yes I feel you, MediaTek SoCs are a bit strange, but I do have a reason to
use one port and multiple endpoints, instead of multiple ports and one endpoint.

On MediaTek SoCs, there are multiple ports: those multiple ports are represented
by multiple MMSYS or multiple VDOSYS (depending on the SoC), which do then have
multiple endpoints.

However, the multiple ports, at least for now, are represented by multiple MMSYS
and/or multiple VDOSYS nodes instead of one MM/VDO node with multiple iostart for
the multiple blocks in `reg`.

The multiple iostart "thing" was the initial design by MediaTek, but there was no
way to get them really connected the right way unless adding an iostart restriction
in the driver itself (so that the mmsys driver would check an iostart to probe the
mediatek-drm components for the right IP number), so, after quite many reviews and
many series versions, they had to resort to use multiple nodes for each VDO.

I think that, after this series, we could also clean that mess up (sorry for the
strong words) and make it right - assigning the MMIO for all VDOSYS blocks to one
node, and adding the multiple ports - however, that will require a bit of work that
is simply too much for this series alone.

Summarizing, so that you don't have to carefully proof-read all this wall of text:

- MediaTek SoCs have got multiple `port` for MMSYS and VDOSYS
   - Currently the driver implementation doesn't allow that
     - MediaTek had to work around no OF graph support!
   - Multiple ports are the multiple MMSYS/VDOSYS
- One MMSYS / One VDOSYS have multiple `endpoints`

That's how the HW is.

Hope that's clear now?

Cheers,
Angelo


>> +    properties:
>> +      endpoint@0:
>> +        $ref: /schemas/graph.yaml#/properties/endpoint
>> +        description: Output to the primary display pipeline
>> +
>> +      endpoint@1:
>> +        $ref: /schemas/graph.yaml#/properties/endpoint
>> +        description: Output to the secondary display pipeline
>> +
>> +      endpoint@2:
>> +        $ref: /schemas/graph.yaml#/properties/endpoint
>> +        description: Output to the tertiary display pipeline
>> +
>> +    required:
>> +      - endpoint@0
>> +
>>   required:
>>     - compatible
>>     - reg
>> -- 
>> 2.44.0
>>