Re: [PATCH v2] usb: r8a66597-hcd: fix buffer overflow on odd-length FIFO reads

[John Paul Adrian Glaubitz]

Hi Karl,

On Thu, 2026-10-08 at 08:18 +0200, Karl Mehltretter wrote:
> On Wed, Oct 07, 2026 at 11:12:46AM +0100, John Paul Adrian Glaubitz wrote:
> > > Fixes: 5d3043586db4 ("USB: r8a66597-hcd: host controller driver for R8A66597")
> > > Cc: stable@xxxxxxxxxxxxxxx
> > > Reported-by: John Paul Adrian Glaubitz <glaubitz@xxxxxxxxxxxxxxxxxxx>
> > > Closes: https://lore.kernel.org/linux-sh/3bd32eaf159db61ed1d423e1d52a869b3689c682.camel@xxxxxxxxxxxxxxxxxxx/
> > > Tested-by: John Paul Adrian Glaubitz <glaubitz@xxxxxxxxxxxxxxxxxxx>
> > > Assisted-by: LLM
> > 
> > I was wondering: Would it maybe be better if you disclosed what LLM you used?
> > I have seen that many others have done so and I'm not sure if it's actually
> > prepared to have the model disclosed.
> > 
> 
> The generic Assisted-by: LLM format is fairly new. Christian Brauner
> proposed simplifying the attribution in July [1], and it is now the
> documented format [2]. I expect we may see more patches using it as
> people adopt the change.
> 
> I prefer the generic Assisted-by: LLM tag. It acknowledges the
> assistance without advertising a particular vendor. 
> 
> A model name alone also leaves out the harness and settings, and I
> sometimes use different models for implementation and review. Tracking
> and listing all of that would make the attribution rather cumbersome.
> 
> [1] https://lore.kernel.org/r/20260701-work-coding-assistants-v1-1-a20a94d1d606@xxxxxxxxxx/
> [2] https://docs.kernel.org/process/coding-assistants.html#attribution

Thanks a lot for the pointer! I will read this thread and unless I find
anything speaking against the proposal, I'll adopt it and continue with
the reviews.

Adrian

-- 
 .''`.  John Paul Adrian Glaubitz
: :' :  Debian Developer
`. `'   Physicist
  `-    GPG: 62FF 8A75 84E0 2956 9546  0006 7426 3B37 F5B5 F913