Skip site navigation (1)Skip section navigation (2)
Date:      Fri, 8 Apr 2005 15:58:28 -0400
From:      "Will Saxon" <WillS@housing.ufl.edu>
To:        "Warner Losh" <imp@bsdimp.com>
Cc:        usb@FreeBSD.org
Subject:   RE: Please commit fix outlined in PR/63837
Message-ID:  <0E972CEE334BFE4291CD07E056C76ED805A2C657@bragi.housing.ufl.edu>

next in thread | raw e-mail | index | archive | help
> -----Original Message-----
> From: Warner Losh [mailto:imp@bsdimp.com]
> Sent: Friday, April 08, 2005 3:40 PM
> To: Will Saxon
> Cc: usb@FreeBSD.org
> Subject: Re: Please commit fix outlined in PR/63837
>=20
>=20
> From: "Will Saxon" <WillS@housing.ufl.edu>
> Subject: Please commit fix outlined in PR/63837
> Date: Fri, 8 Apr 2005 15:34:14 -0400
>=20
> > Is it possible to get the 1 line fix posted in PR/63837 added=20
> > before 5.4? This has been a problem since before 5.3, the fix is=20
> > trivial and it works fine.
> >=20
> > It is kind of annoying to have to edit hid.c every time I cvsup.
> >=20
> > This has been asked about before by myself and others... Is there
> > a stylistic problem or a planned cleanup that has held this back?
> >=20
> > --- /usr/src/sys/dev/usb/hid.c.orig  Fri Apr  8 15:26:46 2005
> > +++ /usr/src/sys/dev/usb/hid.c       Fri Apr  8 14:55:01 2005
> > @@ -375,7 +375,7 @@
> >=20
> > 	id =3D 0;
> > 	for (d =3D hid_start_parse(buf, len, 1<<k); hid_get_item(d, &h); )
> > -		if (h.report_ID !=3D 0)
> > + 		if (h.report_ID !=3D 0 && !id)
> > 			id =3D h.report_ID;
> > 	hid_end_parse(d);
> > 	size =3D h.loc.pos;
>=20
> So this patch just makes report_ID the FIRST id in the item list,
> right?

Looks like it, although really I haven't the foggiest. I do know that=20
without the patch, the mouse included with my MX Duo doesn't work.=20
With it, the mouse works.=20

Reading through the PR notes, it appears that a) the answer to=20
your question is yes and b) the below patch is more of a hack=20
that makes things work and that 'this seems to be a horrible=20
interface.'

That's fine, but the last message on the PR that was NOT a 'hey this=20
works for me too' was mid November. The assertion was that we should=20
try to adopt the solution(s) NetBSD has to fix this and other things,=20
however I don't think that has happened yet.

It would just be nice to get an update, it would make editing
hid.c less aggravating if I knew for sure it wasn't always going
to be this way. I guess I just forget that I have to do it and then
I get mad when I've built a new kernel and my mouse stops working :).

-Will



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?0E972CEE334BFE4291CD07E056C76ED805A2C657>