Re: [PATCH v7 3/3] HID: nintendo: Add unified report format support

[sashiko-bot@kernel.org]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [High] The global HID report size check was incorrectly tightened to 17 bytes, silently dropping valid input reports for non-unified controllers.
--

commit 2414bf53574e3d2b6aee6b978aeb58f941052e21
Author: Vicki Pfau <vi@endrift.com>

HID: nintendo: Add unified report format support

This commit adds support for the unified report format for various Nintendo
controllers. It maps their respective buttons and axes to the new unified
formats.

> diff --git a/drivers/hid/hid-nintendo.c b/drivers/hid/hid-nintendo.c
> index a53c554a54ae3..4920c9691b08a 100644
> --- a/drivers/hid/hid-nintendo.c
> +++ b/drivers/hid/hid-nintendo.c

[ ... ]

> @@ -3765,7 +3866,7 @@ static int switch2_event(struct hid_device *hdev, struct hid_report *report, uin
>  	if (report->type != HID_INPUT_REPORT)
>  		return 0;
>  
> -	if (size < 15)
> +	if (size < 17)
>  		return -EINVAL;

[Severity: High]
Does tightening this global size check to 17 bytes silently drop valid input
reports for non-unified controllers?

Looking at the other cases in switch2_event(), such as NS2_REPORT_GC, the
driver previously validated against a 15-byte minimum because it only
accesses up to raw_data[14]. By unconditionally checking if the size is
less than 17 before evaluating the report ID, any 15-byte or 16-byte report
will hit this early return and be rejected with -EINVAL.

Should this size requirement be localized strictly to the NS2_REPORT_UNIFIED
branch to avoid breaking functionality for legacy controllers?

>  	guard(rcu)();
> @@ -3776,11 +3877,51 @@ static int switch2_event(struct hid_device *hdev, struct hid_report *report, uin
>  
>  	switch (report->id) {
>  	case NS2_REPORT_UNIFIED:

[ ... ]

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260630084838.2313379-1-vi@endrift.com?part=3