[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <6985874e-efae-746d-efbd-dc537b5f24c0@quicinc.com>
Date: Thu, 10 Jul 2025 15:56:57 +0530
From: Dikshita Agarwal <quic_dikshita@...cinc.com>
To: Bryan O'Donoghue <bryan.odonoghue@...aro.org>,
Jorge Ramirez-Ortiz
<jorge.ramirez@....qualcomm.com>,
<quic_vgarodia@...cinc.com>, <krzk+dt@...nel.org>,
<konradybcio@...nel.org>, <mchehab@...nel.org>, <robh@...nel.org>,
<conor+dt@...nel.org>, <andersson@...nel.org>,
<amit.kucheria@....qualcomm.com>
CC: <linux-media@...r.kernel.org>, <linux-arm-msm@...r.kernel.org>,
<devicetree@...r.kernel.org>, <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH v6 6/7] arm64: dts: qcom: qcm2290: Add Venus video node
On 7/10/2025 2:45 PM, Bryan O'Donoghue wrote:
> On 10/07/2025 09:57, Dikshita Agarwal wrote:
>>> + iommus = <&apps_smmu 0x860 0x0>,
>>> + <&apps_smmu 0x880 0x0>,
>>> + <&apps_smmu 0x861 0x04>,
>>> + <&apps_smmu 0x863 0x0>,
>>> + <&apps_smmu 0x804 0xe0>;
>> What’s the rationale behind having five entries here?
>> could you share the use-cases that justify this configuration?
>
> Already getting in trouble with non-pixel/secure # of iommus.
>
> Why not specify the maximum expected number hardware supports, specifically
> so we don't end up buried under incomplete schema again ?
Adding this in DT schema is fine, concern here is, adding it to the DT
without a present use-case may be premature. Surprising to see it doesn't
create any boot-up issues.
Thanks,
Dikshita
>
> ---
> bod
Powered by blists - more mailing lists