I set 10 seconds for the timeout of the i915 audio component binding with a hope that recent machines are fast enough to handle all probe tasks in that period, but I was too optimistic. The binding may take longer than that, and this caused a problem on the machine with both audio and graphics driver modules loaded in parallel, as Paul Menzel experienced. This problem haven't hit so often just because the KMS driver is loaded in initrd on most machines.
As a simple workaround, extend the timeout to 60 seconds.
Fixes: f9b54e1961c7 ("ALSA: hda/i915: Allow delayed i915 audio component binding") Reported-by: Paul Menzel pmenzel+alsa-devel@molgen.mpg.de Cc: stable@vger.kernel.org Signed-off-by: Takashi Iwai tiwai@suse.de --- sound/hda/hdac_i915.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/sound/hda/hdac_i915.c b/sound/hda/hdac_i915.c index 617ff1aa818f..27eb0270a711 100644 --- a/sound/hda/hdac_i915.c +++ b/sound/hda/hdac_i915.c @@ -144,9 +144,9 @@ int snd_hdac_i915_init(struct hdac_bus *bus) return -ENODEV; if (!acomp->ops) { request_module("i915"); - /* 10s timeout */ + /* 60s timeout */ wait_for_completion_timeout(&bind_complete, - msecs_to_jiffies(10 * 1000)); + msecs_to_jiffies(60 * 1000)); } if (!acomp->ops) { dev_info(bus->dev, "couldn't bind with audio component\n");