Subject: Re: kern/34219: ubtbcmfw.c
To: None <plunky@NetBSD.org, gnats-admin@netbsd.org,>
From: Iain Hibbert <plunky@rya-online.net>
List: netbsd-bugs
Date: 08/19/2006 15:30:03
The following reply was made to PR kern/34219; it has been noted by GNATS.
From: Iain Hibbert <plunky@rya-online.net>
To: gnats-bugs@netbsd.org, Andreas Wrede <andreas@planix.com>
Cc:
Subject: Re: kern/34219: ubtbcmfw.c
Date: Sat, 19 Aug 2006 16:24:07 +0100 (BST)
The reason we can't use firmload(9) really is to do with timing - there is
no realy way for the kernel to know when its safe to do so, and because
the device detaches after you load the firmware (and comes back as
something else) it would get complicated if we waited until the device was
being activated before we loaded it.
I have a bcmfw package (pending in pkgsrc-wip at the moment) that will
push the firmware into a ugen device, so in a bit I will remove the
ubtbcmfw(4) device and add something to ubt_match to make it ignore the
vendor/product ID pair of the firmware-less version.
iain