Blob Blame History Raw
From fa1f875c120fa44572c561d86022af2f6b0774c7 Mon Sep 17 00:00:00 2001
From: Peter Ujfalusi <peter.ujfalusi@ti.com>
Date: Mon, 10 Feb 2020 17:14:02 +0200
Subject: [PATCH] ALSA: dmaengine_pcm: Consider DMA cache caused delay in pointer callback
Git-commit: fa1f875c120fa44572c561d86022af2f6b0774c7
Patch-mainline: v5.7-rc1
References: jsc#SLE-16518

Some DMA engines can have big FIFOs which adds to the latency.
The DMAengine framework can report the FIFO utilization in bytes. Use this
information for the delay reporting.

Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Reviewed-by: Takashi Iwai <tiwai@suse.de>
Link: https://lore.kernel.org/r/20200210151402.29634-1-peter.ujfalusi@ti.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Acked-by: Takashi Iwai <tiwai@suse.de>

---
 sound/core/pcm_dmaengine.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/sound/core/pcm_dmaengine.c b/sound/core/pcm_dmaengine.c
index 5749a8a49784..d8be7b488162 100644
--- a/sound/core/pcm_dmaengine.c
+++ b/sound/core/pcm_dmaengine.c
@@ -247,9 +247,14 @@ snd_pcm_uframes_t snd_dmaengine_pcm_pointer(struct snd_pcm_substream *substream)
 
 	status = dmaengine_tx_status(prtd->dma_chan, prtd->cookie, &state);
 	if (status == DMA_IN_PROGRESS || status == DMA_PAUSED) {
+		struct snd_pcm_runtime *runtime = substream->runtime;
+
 		buf_size = snd_pcm_lib_buffer_bytes(substream);
 		if (state.residue > 0 && state.residue <= buf_size)
 			pos = buf_size - state.residue;
+
+		runtime->delay = bytes_to_frames(runtime,
+						 state.in_flight_bytes);
 	}
 
 	return bytes_to_frames(substream->runtime, pos);
-- 
2.16.4