Re: [PATCH 2/2] media: qcom: camss: Enable the streams API on SC7280
From: Hitesh Patel
Date: Wed Sep 16 2026 - 04:34:45 EST
On 16/09/2026 13:38, Bryan O'Donoghue wrote:
> On 15/09/2026 13:15, Hitesh Patel wrote:
>> @@ -2280,6 +2280,7 @@ static const struct camss_subdev_resources csiphy_res_7280[] = {
>> .reg = { "csiphy0" },
>> .interrupt = { "csiphy0" },
>> .csiphy = {
>> + .streams_enable = true,
>
> Set it @ the SoC not for each csiphy and csid.
Agreed. This patch only mirrors what 8/8 of the series does for SM8250
because that is where the flag lives in v1; Gjorgji already agreed in the
1/8 review to move it into struct camss_resources for v2, at which point
this becomes a single .streams_enable = true in camss_resources_7280.
Rather than respin against a shape that is about to change, I would
prefer Gjorgji to fold the SC7280 enablement into his v2 as a one-liner
(and 1/2 as well, if the CSID->VFE pairing rework discussed on 5/8 does
not already make it unnecessary on SC7280: 5 CSIDs and 5 VFEs pair 1:1
there). Gjorgji, is that OK with you? I will retest v2 on RB3 Gen2 as
soon as it is out.
Thanks,
Hitesh