Return-path: Received: from cdptpa-omtalb.mail.rr.com ([75.180.132.120]:41666 "EHLO cdptpa-omtalb.mail.rr.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754020Ab3BABys (ORCPT ); Thu, 31 Jan 2013 20:54:48 -0500 Date: Thu, 31 Jan 2013 20:54:42 -0500 From: Solomon Peachy To: "John W. Linville" Cc: Pontus Fuchs , linux-wireless@vger.kernel.org Subject: Re: RFCv3: ST-E CW1200 WLAN driver Message-ID: <20130201015442.GA2174@shaftnet.org> (sfid-20130201_025452_194025_EA559C03) References: <1357999575-12838-1-git-send-email-pizza@shaftnet.org> <50F3F743.1000802@gmail.com> <20130130190215.GA2167@tuxdriver.com> <20130130223232.GA7307@shaftnet.org> MIME-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="J2SCkAp4GZ/dPZZf" In-Reply-To: <20130130223232.GA7307@shaftnet.org> Sender: linux-wireless-owner@vger.kernel.org List-ID: --J2SCkAp4GZ/dPZZf Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Wed, Jan 30, 2013 at 05:32:32PM -0500, Solomon Peachy wrote: > This is my current to-do list: Was unexpectedly able to spend some time on this today! > * One final Kconfig option to eliminate Done. > * A possible crash (reported not three hours ago) that needs confirming Turns out to not be a crash, but it is a real problem. The driver can=20 get into an inconsistent internal state if a connection loss event=20 occurs. Proper fix triggers deadlock due to interdependencies. =20 Hackish workaround in place now, and I'm working on a proper fix. Or=20 will be once my head stops throbbing. The code is needlessly complex=20 and interdependent, and unravelling a bit at a time is difficult. This state problem actually explains a bunch of other odd quirks I've=20 seen, so fixing it properly is pretty important. > * Hot unplug is still iffy (but better!) More progress here, but more work to be done. (Only a problem now if=20 device isn't idle when yanked -- if the supplicant is attached, it=20 deadlocks still) > * Reworked SPI interrupt delivery completely untested > * IBSS is still busted (haven't had time to tackle it properly) > * Rebase against 3.8 > * Investigate adding OF bindings that can be used instead of platform_da= ta No progress here yet. I'll post another patch set this weekend. - Solomon --=20 Solomon Peachy pizza at shaftnet dot org =20 Melbourne, FL ^^ (mail/jabber/gtalk) ^^ Quidquid latine dictum sit, altum viditur. --J2SCkAp4GZ/dPZZf Content-Type: application/pgp-signature -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.13 (GNU/Linux) iD8DBQFRCyBiPuLgii2759ARAnjHAKCB+qRUxhJHHp1m7ZZLuTYGpNqhGgCfYP+T VS5wtB99UxU+puDZI5lTYfY= =TGNx -----END PGP SIGNATURE----- --J2SCkAp4GZ/dPZZf--