Return-path: Received: from rv-out-0506.google.com ([209.85.198.232]:51917 "EHLO rv-out-0506.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750848AbYGICWK (ORCPT ); Tue, 8 Jul 2008 22:22:10 -0400 Received: by rv-out-0506.google.com with SMTP id k40so3387442rvb.1 for ; Tue, 08 Jul 2008 19:22:09 -0700 (PDT) Message-ID: (sfid-20080709_042229_864639_D7475156) Date: Tue, 8 Jul 2008 22:22:09 -0400 From: "Bob Copeland" To: "Helge Deller" Subject: Re: [ath5k-devel] [PATCH] ath5k: convert LED code to use mac80211 triggers Cc: ath5k-devel@lists.ath5k.org, linux-wireless@vger.kernel.org In-Reply-To: MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 References: <20080622181134.GA26571@hash.localnet> <20080626023528.GA16068@hash.localnet> Sender: linux-wireless-owner@vger.kernel.org List-ID: On Tue, Jul 8, 2008 at 7:08 PM, Helge Deller wrote: > I tested it against Linus' head version (it applied with minor > modifications) on my HP NC6000 laptop with this built-in Atheros card: [...] > > Basically it works, but I think the sc->led_on in: >> + sc->led_on = 0; /* active low */ > > needs to be: >> + sc->led_on = 1; /* active high */ > for the AR5212 in the HP laptop. Thanks for the test! > See also the changes to if_ath_pci.c in: > http://madwifi.org/attachment/ticket/1018/led.patch > and > http://madwifi.org/ticket/1018 > for an old discussion about madwifi for LEDs in the HP box. So current madwifi doesn't have this change unless I missed something? It appears to use active-low leds in the current version regardless of the vendor. > Without this change the LED stays on after a "rmmod ath5k" as well. Yeah, that would definitely happen if the logic was backwards. I guess the conservative approach is to just use sc->led_on = 1 in the VENDOR_ID_COMPAQ branch. -- Bob Copeland %% www.bobcopeland.com