Re: [PATCH 1/3] soc: mediatek: mtk-socinfo: Add driver for getting chip information

[AngeloGioacchino Del Regno]

Il 18/07/23 13:21, William-tw Lin ha scritto:
> Add driver for socinfo retrieval. This patch includes the following:
> 1. mtk-socinfo driver for chip info retrieval
> 2. Related changes to Makefile and Kconfig
>
> Signed-off-by: William-tw Lin <william-tw.lin@xxxxxxxxxxxx>
> ---
>   drivers/soc/mediatek/Kconfig       |  18 +++
>   drivers/soc/mediatek/Makefile      |   1 +
>   drivers/soc/mediatek/mtk-socinfo.c | 203 +++++++++++++++++++++++++++
>   drivers/soc/mediatek/mtk-socinfo.h | 213 +++++++++++++++++++++++++++++
>   4 files changed, 435 insertions(+)
>   create mode 100644 drivers/soc/mediatek/mtk-socinfo.c
>   create mode 100644 drivers/soc/mediatek/mtk-socinfo.h

..snip..

> +
> +/* begin 8186 info */
> +#define mtk_mt8186_EFUSE_DATA_COUNT 1
> +static struct efuse_data mtk_mt8186_efuse_data_info[][mtk_mt8186_EFUSE_DATA_COUNT] = {
> +	{{.nvmem_cell_name = "socinfo-data1", .efuse_data = 0x81861001}},
> +	{{.nvmem_cell_name = "socinfo-data1", .efuse_data = 0x81862001}},
> +};

I'm *sure* that there's a way to avoid specifying the soc-specific compatible, as
the efuse_data is used only for validation, while what you do here is to simply
read the SoC ID from eFuse array.

In theory, this driver could simply have

compatible = "mediatek,socinfo";

as compatible string, without any SoC-specific string in the devicetree, at all,
so that the SoC ID would get recognized completely dynamically - and adding the
name and other strings would be a consequence.

> +
> +static struct name_data mtk_mt8186_name_data_info[] = {
> +	{.soc_name = "MT8186",
> +	 .soc_segment_name = "MT8186GV/AZA",
> +	 .marketing_name = "Kompanio 520"},
> +	{.soc_name = "MT8186T",
> +	 .soc_segment_name = "MT8186TV/AZA",
> +	 .marketing_name = "Kompanio 528"},
> +};
> +/* end 8186 info */
> +

Regards,
Angelo