Re: [PATCH v2 12/14] media: uapi: Add CAMSS ISP configuration definition

[Bryan O'Donoghue]

On 27/04/2026 13:43, Loic Poulain wrote:
> +/**
> + * struct camss_params_wb_gain - White Balance gains
> + *
> + * @header:   generic block header; @header.type = CAMSS_PARAMS_WB_GAIN
> + * @g_gain:   green channel gain (15uQ10)
> + * @b_gain:   blue channel gain (15uQ10)
> + * @r_gain:   red channel gain (15uQ10)
> + */
> +struct camss_params_wb_gain {
> +	struct v4l2_isp_params_block_header header;
> +	__u16 g_gain;
> +	__u16 b_gain;
> +	__u16 r_gain;
> +	__u16 _pad;
> +} __attribute__((aligned(8)));

So I published a comprehensive list of these structures a few days ago.

https://lore.kernel.org/linux-media/20260426000418.1158716-1-bryan.odonoghue@xxxxxxxxxx/T/#u

See:

struct camss_params_wb_gain - White balance gains and offsets

The ICP/HFI structures - at least for the stats I believe are how 
hardware writes data to memory.

OTOH on the way in, the params ought to be pretty uniform as again 
HFI/ICP needs to take that representation and either

- Hand it over to CDM or
- Pass it over HFI so that ICP can hand it over to CDM

The ordering and precision is a resolved thing. I'm ~ sure OPE must 
write and consume in the same format.

---
bod