From mboxrd@z Thu Jan 1 00:00:00 1970 Date: Mon, 20 Aug 2001 11:47:42 -0400 From: Ethan Blanton To: Derrik Pates Cc: linuxppc-dev@lists.linuxppc.org Subject: Re: Airport driver sleep problem and solution Message-ID: <20010820114742.J22957@localhost.localdomain> References: Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="wHh0aNzodMFDTGdO" In-Reply-To: ; from dpates@dsdk12.net on Mon, Aug 20, 2001 at 09:38:15AM -0600 Sender: owner-linuxppc-dev@lists.linuxppc.org List-Id: --wHh0aNzodMFDTGdO Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable Derrik Pates spake unto us the following wisdom: > I recently found that, if the Airport card in an iBook I've been running > Linux on was not brought "up" at the time the system was put to sleep (by > putting the lid down), after bringing it back out of sleep, it would no > longer correctly operate until the driver was removed and reloaded - > ifconfig would receive an ENODEV error, and iwconfig would report that the > device had no wireless extensions. I found that the problem is on line 179 > of linux/drivers/net/wireless/airport.c (v0.06f, in 2.4.9-benh0), where > the driver only calls "netif_device_attach" if card->open is TRUE - but it > unconditionally calls "netif_device_detach" in the PBOOK_SLEEP_NOW case. > Line 179 should be changed from "else if (card->open)" to just "else", so > that the network device is "reattached" to the networking subsystem. Spectacular! I've had this trouble as well, and I solved it my just leaving my Airport card configured all the time, regardless of whether I was on a network or not. I'm glad to see a real fix. :-) Ethan --=20 If I've told you once, I've told you once And once is all that you needed. -- The Refreshments, "Carefree" --wHh0aNzodMFDTGdO Content-Type: application/pgp-signature Content-Disposition: inline --wHh0aNzodMFDTGdO-- ** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/