From: Kirill A. Korinsky <kirill@korins.ky>
Subject: Re: sys/ihidev: respect RESET_RESPONSE and adjsut POWER ON timeout
To: Marcus Glocker <marcus@nazgul.ch>
Cc: tech@openbsd.org, Mark Kettenis <mark.kettenis@xs4all.nl>, mvs@openbsd.org
Date: Sat, 04 Jan 2025 18:04:00 +0100

On Sat, 04 Jan 2025 17:33:42 +0100,
Marcus Glocker <marcus@nazgul.ch> wrote:
> 
> > +		for (i = 0; i < 50; i++) {
> > +			ihidev_sleep(sc, 100);
> > +			res = iic_exec(sc->sc_tag, I2C_OP_READ_WITH_STOP, sc->sc_addr,
> > +			    NULL, 0, buf, sizeof(buf), 0);
> 
> You could have fixed this line break if you anyway had to create a new
> patch ;-)
>

I really thought I had fixed it by moving sc->sc_addr into the next line
before NULL. But my diff juggling sems to framed me :)

> > +			DPRINTF(("%s: read attempt %d: 0x%x, 0x%x, res: %d\n",
> > +			    sc->sc_dev.dv_xname, i, buf[0], buf[1], res));
> > +			if (!res)
> > +				res = (buf[0] != 0x00 || buf[1] != 0x00);
> > +			if (!res)
> > +				break;
> > +		}
> > +
> > +		break;
> > +	}
> >  	default:
> >  		printf("%s: unknown command %d\n", sc->sc_dev.dv_xname,
> >  		    hidcmd);
> > @@ -557,7 +584,11 @@ ihidev_reset(struct ihidev_softc *sc)
> 
> Have you intentionally removed the DPRINTF "resetting" from your last
> patch?  I think it would be better placed in the reset function itself,
> as in your last patch.
>

Yep, to keep this changes as small as possible which just adds support of
RESET RESPONSE and nothing else.

-- 
wbr, Kirill