[PATCH 2/2] phy: qcom-mipi-csi2: Add a CSI2 MIPI D-PHY driver

[neil.armstrong at linaro.org]

On 10/07/2025 18:16, Bryan O'Donoghue wrote:
> Add a new MIPI CSI2 driver in D-PHY mode initially. The entire set of
> existing CAMSS CSI PHY init sequences are imported in order to save time
> and effort in later patches.
> 
> In-line with other PHY drivers the process node name is omitted from the
> compat string while the soc name is included.
> 
> At the moment we follow the assignment of lane positions - the bitmap of
> physical input lanes to logical lane numbers as a linear list per the
> existing DPHY @lanes data-member.
> 
> This is fine for us in upstream since we also map the lanes contiguously
> but, our hardware can support different lane mappings so we should in the
> future extend out the DPHY structure to capture the mapping.
> 
> The Qualcomm 3PH class of PHYs can do both D-PHY and C-PHY mode. For now only
> D-PHY is supported.
> 
> In porting some of the logic over from camss-csiphy*.c to here its also
> possible to rationalise some of the code.
> 
> In particular use of regulator_bulk and clk_bulk as well as dropping the
> seemingly useless and unused interrupt handler.
> 
> The PHY sequences and a lot of the logic that goes with them are well proven
> in CAMSS and mature so the main thing to watch out for here is how to get
> the right sequencing of regulators, clocks and register-writes.
> 
> Signed-off-by: Bryan O'Donoghue <bryan.odonoghue at linaro.org>
> ---
>   MAINTAINERS                                        |  11 +
>   drivers/phy/qualcomm/Kconfig                       |  11 +
>   drivers/phy/qualcomm/Makefile                      |   6 +
>   drivers/phy/qualcomm/phy-qcom-mipi-csi2-3ph-dphy.c | 491 +++++++++++++++++++++
>   drivers/phy/qualcomm/phy-qcom-mipi-csi2-core.c     | 281 ++++++++++++
>   drivers/phy/qualcomm/phy-qcom-mipi-csi2.h          | 101 +++++
>   6 files changed, 901 insertions(+)
> 
>
<snip>

> +const struct mipi_csi2phy_clk_freq zero = { 0 };
> +
> +const struct mipi_csi2phy_clk_freq dphy_4nm_x1e_csiphy = {
> +	.freq = {
> +		300000000, 400000000, 480000000
> +	},
> +	.num_freq = 3,
> +};
> +
> +const struct mipi_csi2phy_clk_freq dphy_4nm_x1e_csiphy_timer = {
> +	.freq = {
> +		266666667, 400000000
> +	},
> +	.num_freq = 2,
> +};
> +
> +const struct mipi_csi2phy_soc_cfg mipi_csi2_dphy_4nm_x1e = {
> +	.ops = &phy_qcom_mipi_csi2_ops_3ph_1_0,
> +	.reg_info = {
> +		.init_seq = lane_regs_x1e80100,
> +		.lane_array_size = ARRAY_SIZE(lane_regs_x1e80100),
> +		.offset = 0x1000,
> +		.generation = GEN2,
> +	},
> +	.supply_names = (const char *[]){
> +		"vdda-0p8",
> +		"vdda-1p2"
> +	},
> +	.num_supplies = 2,
> +	.clk_names = (const char *[]) {
> +		"camnoc_axi",
> +		"cpas_ahb",
> +		"csiphy",
> +		"csiphy_timer"
> +	},
> +	.num_clk = 4,
> +	.clk_freq = {
> +		zero,

It seems clang doesn't like this at all:
drivers/phy/qualcomm/phy-qcom-mipi-csi2-3ph-dphy.c:486:3: error: initializer element is not a compile-time constant
                 zero,
                 ^~~~
1 error generated.

> +		zero,
> +		dphy_4nm_x1e_csiphy,
> +		dphy_4nm_x1e_csiphy_timer,
> +	},
> +};
<snip>