From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1161034AbVKXMYY (ORCPT ); Thu, 24 Nov 2005 07:24:24 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1030383AbVKXMYY (ORCPT ); Thu, 24 Nov 2005 07:24:24 -0500 Received: from viper.oldcity.dca.net ([216.158.38.4]:56989 "HELO viper.oldcity.dca.net") by vger.kernel.org with SMTP id S1030392AbVKXMYX (ORCPT ); Thu, 24 Nov 2005 07:24:23 -0500 Subject: stable review patch - ALSA ALC800 codec From: Lee Revell To: linux-kernel Cc: stable@kernel.org, Takashi Iwai Content-Type: text/plain Date: Thu, 24 Nov 2005 07:23:13 -0500 Message-Id: <1132834994.20390.4.camel@mindpipe> Mime-Version: 1.0 X-Mailer: Evolution 2.4.1 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org This fixes an apparent copy and paste bug that has really been causing problems for a lot of users. It closes ~10 different ALSA bug reports, see #1563 for example. Please apply ASAP for -stable. Lee This patch is to fix the problem of calculating the nid incorrectly when auto-probe for ALC880. The problem to be fixed often behaves with such words when using dmesg, 'num_steps = 0 for NID=0x8' when auto-probe for ALC880. The patch contains: - alsa-kernel/pci/hda/patch_realtek.c: replace 'alc880_dac_to_idx' with 'alc880_idx_to_dac' in function 'alc880_auto_fill_dac_nids()' Signed-off-by: Libin Yang --- alsa-driver-1.0.10/alsa-kernel/pci/hda/patch_realtek.c 2005-11-02 10:26:49.000000000 -0800 +++ my_alsa-driver-1.0.10/alsa-kernel/pci/hda/patch_realtek.c 2005-11-23 05:52:36.000000000 -0800 @@ -1809,7 +1809,7 @@ nid = cfg->line_out_pins[i]; if (alc880_is_fixed_pin(nid)) { int idx = alc880_fixed_pin_idx(nid); - spec->multiout.dac_nids[i] = alc880_dac_to_idx(idx); + spec->multiout.dac_nids[i] = alc880_idx_to_dac(idx); assigned[idx] = 1; } }