Date: Sat, 12 Nov 2005 10:47:33 -0700 (MST) From: "M. Warner Losh" <imp@bsdimp.com> To: dirkx@webweaving.org Cc: freebsd-hackers@FreeBSD.org Subject: Re: Sierra Wireless / AC775 / Loading an external cisfile as a 'quirck' Message-ID: <20051112.104733.13599970.imp@bsdimp.com> In-Reply-To: <20051112072658.Q14448@skutsje.san.webweaving.org> References: <20051112072658.Q14448@skutsje.san.webweaving.org>
next in thread | previous in thread | raw e-mail | index | archive | help
In message: <20051112072658.Q14448@skutsje.san.webweaving.org>
Dirk-Willem van Gulik <dirkx@webweaving.org> writes:
:
: I've got the (correct) CIS files below for a Sierra Wireeless Airnet 775
: (AC775) card. The actual data on the card (see second cut-and-paste) is
: wrong (i.e. multi function instead of a modem).
:
: So I am looking for a way to either import a CIS file dynamically; at run
: time - or to translate a CIS file in what I assume should be a quirk entry
: in the file /sys/dev/pccard/pccard_cis_quirks.c (is that corect) ?
OpenBSD solved this problem by listing this card in
pcmica_cis_quirk.c. Maybe you can try that as an iterrum solution?
That's the short term solution. I really hate the quirks in the
present system and plan on doing something about it.
: I.e. something akin to:
:
: #cat SW_775_SER.dat > /dev/pccard0.cis
:
: Looking for a solution on either 5, 6 or CURRENT.
:
: Any hints - or guidance how to best map this into the quirk section - as
: the fields are not quite clear to me.
We just got the ability to dump the CIS out of a card. It is unclear
how we should go about replacing the CIS in the kernel. Do we do this
by loading the CIS in before the card is inserted? Do we keep a
database of bad cards and automatically subsitute the right one? Do
we detach an reprobe the card when a new CIS is loaded in (like your
example above)?
Being able to load CIS in would fix a bunch of multi-function cards
that aren't listed as such (and vice versa).
Warner
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20051112.104733.13599970.imp>