Re: [PATCH v2] staging: rtl8723bs: refactor BSS Coexistence channel report logic

[Greg Kroah-Hartman <gregkh@linuxfoundation.org>]

On Thu, Jan 29, 2026 at 01:54:37AM -0800, Te-Hsiu Huang wrote:
> On Thu, Jan 29, 2026 at 12:54 AM Greg Kroah-Hartman
> <gregkh@linuxfoundation.org> wrote:
> 
> > That's a lot of different things all in one single patch.  SHouldn't
> > this be split up into different ones?
> I kept these changes in one patch because they all work together to
> replace the old ICS logic and remove the magic numbers. Splitting them
> might break the build or make the code inconsistent halfway through.
> I was suggested renaming and using a bool array structure in the
> earlier patch here:
> https://lore.kernel.org/all/e87acd109ea5eb4e3f8fc233788cd0dbd128407d.camel@perches.com/
> I kept the 2D structure to maintain parity with the IEEE 802.11 spec
> as original implementation. It makes the code semantically correct and
> easier to extend for other Operating Classes in the future.
> 
> > And did you use AI to generate this patch?
> I used AI tools to polish the language of my commit message and to
> perform a parity check, ensuring the refactored boolean logic is
> functionally identical to the original code.

Then as per our development rules, you must document this properly.
Please read and follow them for when you resubmit this.

> 
> > Also, how was this tested?  What prompted you to want to make this
> > change in the first place?
> The patch was tested by:
> 1. Cross-compile the module for the rtl8723bs driver (make
> M=drivers/staging/rtl8723bs) to ensure no build regressions.
> 2. Ran it through scripts/checkpatch.pl to confirm it meets kernel
> coding standards.
> 3. Since I do not have the physical hardware, I performed a manual
> check (by AI verification) to ensure the logic remains identical.

AI can not "verify" anything, you must always check that, sorry.

greg k-h