Re: [PATCH 1/3] drm/bridge: ite-it66121: improve reading IDs and log on errors

[sashiko-bot@kernel.org]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Low] drm: Sparse warnings due to __u16 usage for little-endian data
--

commit 8f50576e49682d9068991d3e7de12fc063b0685f
Author: Luca Ceresoli <luca.ceresoli@bootlin.com>
drm/bridge: ite-it66121: improve reading IDs and log on errors

This commit switches from individual regmap_read() calls to a single
regmap_bulk_read() when fetching vendor and device IDs. It introduces a
union to explicitly handle the data bytes and adds proper error checking
to prevent silently continuing with invalid values.

> diff --git a/drivers/gpu/drm/bridge/ite-it66121.c b/drivers/gpu/drm/bridge/ite-it66121.c
> --- a/drivers/gpu/drm/bridge/ite-it66121.c
> +++ b/drivers/gpu/drm/bridge/ite-it66121.c
> @@ -1579,7 +1579,16 @@ static const struct it66121_chip_info it66xx_chip_info[] = {
>  
>  static int it66121_probe(struct i2c_client *client)
>  {
> -	u32 revision_id, vendor_ids[2] = { 0 }, device_ids[2] = { 0 };
> +	union {
> +		__u8 byte[4];
> +		struct {
> +			__u16 vendor;
> +			__u16 device;
> +		};

[Severity: Low]
Should these fields be declared as __le16 instead of __u16?

Since they hold little-endian data populated directly by regmap_bulk_read()
and are later passed to le16_to_cpu(), using natively-typed __u16 here
violates the kernel's endianness typing rules and triggers Sparse warnings
for incorrect types.

> +	} chip_id_regs;
> +	BUILD_BUG_ON(sizeof(chip_id_regs) != 4);
> +	u16 vendor_id, device_id;

[ ... ]

> @@ -1643,19 +1652,22 @@ static int it66121_probe(struct i2c_client *client)

[ ... ]

> +	vendor_id = le16_to_cpu(chip_id_regs.vendor);
> +	device_id = le16_to_cpu(chip_id_regs.device);

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20261015-drm-ite-it66121-log-probe-errors-v1-0-7fd6994e0be0@bootlin.com?part=1