summaryrefslogtreecommitdiffstats
diff options
context:
space:
mode:
authorJaroslav Kysela <perex@perex.cz>2009-07-10 12:27:31 +0200
committerTakashi Iwai <tiwai@suse.de>2009-07-10 12:55:49 +0200
commit9d30937accf2c01e8b0bd59787409a7348cbbcb7 (patch)
tree6ee1c5a8e9487ca8f56d015e64dbb9cfd3ca0001
parentALSA: hda - targa and targa-2ch fix (diff)
downloadlinux-9d30937accf2c01e8b0bd59787409a7348cbbcb7.tar.xz
linux-9d30937accf2c01e8b0bd59787409a7348cbbcb7.zip
ALSA: hda_intel: more strict alc880_parse_auto_config dig_nid checking
On some IbexPeak systems with ALC889A errors like "azx_get_response timeout, switching to polling mode: last cmd=0xaf9f000b" are produced, because non-existent codec #10 is wrongly accessed. The problem is that snd_hda_get_connections() returns out-of-range result for NID 0x1c (something like 0xf8f9 or 0xffff). This patch adds a check to alc880_parse_auto_config() to avoid using of this out-of-range NIDs. A better fix maybe to improve snd_hda_get_connections() routine to check for valid NID ranges if NIDs are expected as result. Signed-off-by: Jaroslav Kysela <perex@perex.cz> Signed-off-by: Takashi Iwai <tiwai@suse.de>
-rw-r--r--sound/pci/hda/patch_realtek.c6
1 files changed, 6 insertions, 0 deletions
diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c
index bbb9b42e2604..7e99763ca527 100644
--- a/sound/pci/hda/patch_realtek.c
+++ b/sound/pci/hda/patch_realtek.c
@@ -4505,6 +4505,12 @@ static int alc880_parse_auto_config(struct hda_codec *codec)
&dig_nid, 1);
if (err < 0)
continue;
+ if (dig_nid > 0x7f) {
+ printk(KERN_ERR "alc880_auto: invalid dig_nid "
+ "connection 0x%x for NID 0x%x\n", dig_nid,
+ spec->autocfg.dig_out_pins[i]);
+ continue;
+ }
if (!i)
spec->multiout.dig_out_nid = dig_nid;
else {