ALSA: hda: Fix regression by strip mask fix

The commit e38e486 ("ALSA: hda: Modify stream stripe mask only
when needed") tried to address the regression by the unconditional
application of the stripe mask, but this caused yet another
regression for the previously working devices.  Namely, the patch
clears the azx_dev->stripe flag at snd_hdac_stream_clear(), but this
may be called multiple times before restarting the stream, so this
ended up with clearance of the flag for the whole time.

This patch fixes the regression by moving the azx_dev->stripe flag
clearance at the counter-part, the close callback of HDMI codec
driver instead.

Fixes: e38e486 ("ALSA: hda: Modify stream stripe mask only when needed")
BugLink: https://bugzilla.kernel.org/show_bug.cgi?id=205855
BugLink: https://bugzilla.kernel.org/show_bug.cgi?id=204477
Cc: <stable@vger.kernel.org>
Link: https://lore.kernel.org/r/20191214175217.31852-1-tiwai@suse.de
Signed-off-by: Takashi Iwai <tiwai@suse.de>

sound/hda/hdac_stream.c

@@ -120,10 +120,8 @@ void snd_hdac_stream_clear(struct hdac_stream *azx_dev)
 	snd_hdac_stream_updateb(azx_dev, SD_CTL,
 				SD_CTL_DMA_START | SD_INT_MASK, 0);
 	snd_hdac_stream_writeb(azx_dev, SD_STS, SD_INT_MASK); /* to be sure */
-	if (azx_dev->stripe) {
+	if (azx_dev->stripe)
 		snd_hdac_stream_updateb(azx_dev, SD_CTL_3B, SD_CTL_STRIPE_MASK, 0);
-		azx_dev->stripe = 0;
-	}
 	azx_dev->running = false;
 }
 EXPORT_SYMBOL_GPL(snd_hdac_stream_clear);

sound/pci/hda/patch_hdmi.c

@@ -2021,6 +2021,8 @@ static int hdmi_pcm_close(struct hda_pcm_stream *hinfo,
 		per_cvt->assigned = 0;
 		hinfo->nid = 0;
 
+		azx_stream(get_azx_dev(substream))->stripe = 0;
+
 		mutex_lock(&spec->pcm_lock);
 		snd_hda_spdif_ctls_unassign(codec, pcm_idx);
 		clear_bit(pcm_idx, &spec->pcm_in_use);