[PATCH 0/5] ALSA: usb-audio: Delay account cleanup and latency reduction

[PATCH 0/5] ALSA: usb-audio: Delay account cleanup and latency reduction

[Takashi Iwai]

Hi,

this is a patch set for USB-audio driver for the code refactoring of
PCM delay account as well as the reduction of the playback latency.
The last change is a bit intrusive (despite of the smallness), and
hopefully it won't break things.  My initial tests through various
backends were OK, but if anyone finds a problem, please let me know.


Takashi

===

Takashi Iwai (5):
  ALSA: usb-audio: Make snd_usb_pcm_delay() static
  ALSA: usb-audio: Pre-calculate buffer byte size
  ALSA: usb-audio: Refactoring delay account code
  ALSA: usb-audio: Factor out DSD bitrev copy function
  ALSA: usb-audio: Reduce latency at playback start

 sound/usb/card.h     |   8 +-
 sound/usb/endpoint.c |   1 +
 sound/usb/pcm.c      | 202 ++++++++++++++++++++-----------------------
 sound/usb/pcm.h      |   3 -
 4 files changed, 101 insertions(+), 113 deletions(-)

-- 
2.26.2

[PATCH 1/5] ALSA: usb-audio: Make snd_usb_pcm_delay() static

[Takashi Iwai]

It's a local function, let's make it static.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
---
 sound/usb/pcm.c | 4 ++--
 sound/usb/pcm.h | 3 ---
 2 files changed, 2 insertions(+), 5 deletions(-)

diff --git a/sound/usb/pcm.c b/sound/usb/pcm.c
index e5311b6bb3f6..359c759a7023 100644
--- a/sound/usb/pcm.c
+++ b/sound/usb/pcm.c
@@ -29,8 +29,8 @@
 #define SUBSTREAM_FLAG_SYNC_EP_STARTED	1
 
 /* return the estimated delay based on USB frame counters */
-snd_pcm_uframes_t snd_usb_pcm_delay(struct snd_usb_substream *subs,
-				    unsigned int rate)
+static snd_pcm_uframes_t snd_usb_pcm_delay(struct snd_usb_substream *subs,
+					   unsigned int rate)
 {
 	int current_frame_number;
 	int frame_diff;
diff --git a/sound/usb/pcm.h b/sound/usb/pcm.h
index 06c586467d3f..493a4e34d78d 100644
--- a/sound/usb/pcm.h
+++ b/sound/usb/pcm.h
@@ -2,9 +2,6 @@
 #ifndef __USBAUDIO_PCM_H
 #define __USBAUDIO_PCM_H
 
-snd_pcm_uframes_t snd_usb_pcm_delay(struct snd_usb_substream *subs,
-				    unsigned int rate);
-
 void snd_usb_set_pcm_ops(struct snd_pcm *pcm, int stream);
 int snd_usb_pcm_suspend(struct snd_usb_stream *as);
 int snd_usb_pcm_resume(struct snd_usb_stream *as);
-- 
2.26.2

[PATCH 2/5] ALSA: usb-audio: Pre-calculate buffer byte size

[Takashi Iwai]

There are a bunch of lines calculating the buffer size in bytes at
each time.  Keep the value in subs->buffer_bytes and use it
consistently for the code simplicity.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
---
 sound/usb/card.h |  1 +
 sound/usb/pcm.c  | 48 ++++++++++++++++++++++++------------------------
 2 files changed, 25 insertions(+), 24 deletions(-)

diff --git a/sound/usb/card.h b/sound/usb/card.h
index a741e7da83a2..b346653d4b76 100644
--- a/sound/usb/card.h
+++ b/sound/usb/card.h
@@ -158,6 +158,7 @@ struct snd_usb_substream {
 
 	unsigned int running: 1;	/* running status */
 
+	unsigned int buffer_bytes;	/* buffer size in bytes */
 	unsigned int hwptr_done;	/* processed byte position in the buffer */
 	unsigned int transfer_done;		/* processed frames since last period update */
 	unsigned int frame_limit;	/* limits number of packets in URB */
diff --git a/sound/usb/pcm.c b/sound/usb/pcm.c
index 359c759a7023..e8121af8e1d5 100644
--- a/sound/usb/pcm.c
+++ b/sound/usb/pcm.c
@@ -600,6 +600,7 @@ static int snd_usb_pcm_prepare(struct snd_pcm_substream *substream)
 		goto unlock;
 
 	/* reset the pointer */
+	subs->buffer_bytes = frames_to_bytes(runtime, runtime->buffer_size);
 	subs->hwptr_done = 0;
 	subs->transfer_done = 0;
 	subs->last_delay = 0;
@@ -1147,8 +1148,8 @@ static void retire_capture_urb(struct snd_usb_substream *subs,
 		spin_lock_irqsave(&subs->lock, flags);
 		oldptr = subs->hwptr_done;
 		subs->hwptr_done += bytes;
-		if (subs->hwptr_done >= runtime->buffer_size * stride)
-			subs->hwptr_done -= runtime->buffer_size * stride;
+		if (subs->hwptr_done >= subs->buffer_bytes)
+			subs->hwptr_done -= subs->buffer_bytes;
 		frames = (bytes + (oldptr % stride)) / stride;
 		subs->transfer_done += frames;
 		if (subs->transfer_done >= runtime->period_size) {
@@ -1166,9 +1167,9 @@ static void retire_capture_urb(struct snd_usb_substream *subs,
 
 		spin_unlock_irqrestore(&subs->lock, flags);
 		/* copy a data chunk */
-		if (oldptr + bytes > runtime->buffer_size * stride) {
-			unsigned int bytes1 =
-					runtime->buffer_size * stride - oldptr;
+		if (oldptr + bytes > subs->buffer_bytes) {
+			unsigned int bytes1 = subs->buffer_bytes - oldptr;
+
 			memcpy(runtime->dma_area + oldptr, cp, bytes1);
 			memcpy(runtime->dma_area, cp + bytes1, bytes - bytes1);
 		} else {
@@ -1184,10 +1185,9 @@ static inline void fill_playback_urb_dsd_dop(struct snd_usb_substream *subs,
 					     struct urb *urb, unsigned int bytes)
 {
 	struct snd_pcm_runtime *runtime = subs->pcm_substream->runtime;
-	unsigned int stride = runtime->frame_bits >> 3;
 	unsigned int dst_idx = 0;
 	unsigned int src_idx = subs->hwptr_done;
-	unsigned int wrap = runtime->buffer_size * stride;
+	unsigned int wrap = subs->buffer_bytes;
 	u8 *dst = urb->transfer_buffer;
 	u8 *src = runtime->dma_area;
 	u8 marker[] = { 0x05, 0xfa };
@@ -1233,8 +1233,8 @@ static inline void fill_playback_urb_dsd_dop(struct snd_usb_substream *subs,
 			subs->hwptr_done++;
 		}
 	}
-	if (subs->hwptr_done >= runtime->buffer_size * stride)
-		subs->hwptr_done -= runtime->buffer_size * stride;
+	if (subs->hwptr_done >= subs->buffer_bytes)
+		subs->hwptr_done -= subs->buffer_bytes;
 }
 
 static void copy_to_urb(struct snd_usb_substream *subs, struct urb *urb,
@@ -1242,10 +1242,10 @@ static void copy_to_urb(struct snd_usb_substream *subs, struct urb *urb,
 {
 	struct snd_pcm_runtime *runtime = subs->pcm_substream->runtime;
 
-	if (subs->hwptr_done + bytes > runtime->buffer_size * stride) {
+	if (subs->hwptr_done + bytes > subs->buffer_bytes) {
 		/* err, the transferred area goes over buffer boundary. */
-		unsigned int bytes1 =
-			runtime->buffer_size * stride - subs->hwptr_done;
+		unsigned int bytes1 = subs->buffer_bytes - subs->hwptr_done;
+
 		memcpy(urb->transfer_buffer + offset,
 		       runtime->dma_area + subs->hwptr_done, bytes1);
 		memcpy(urb->transfer_buffer + offset + bytes1,
@@ -1255,8 +1255,8 @@ static void copy_to_urb(struct snd_usb_substream *subs, struct urb *urb,
 		       runtime->dma_area + subs->hwptr_done, bytes);
 	}
 	subs->hwptr_done += bytes;
-	if (subs->hwptr_done >= runtime->buffer_size * stride)
-		subs->hwptr_done -= runtime->buffer_size * stride;
+	if (subs->hwptr_done >= subs->buffer_bytes)
+		subs->hwptr_done -= subs->buffer_bytes;
 }
 
 static unsigned int copy_to_urb_quirk(struct snd_usb_substream *subs,
@@ -1295,7 +1295,7 @@ static void prepare_playback_urb(struct snd_usb_substream *subs,
 	int i, stride, period_elapsed = 0;
 	unsigned long flags;
 
-	stride = runtime->frame_bits >> 3;
+	stride = ep->stride;
 
 	frames = 0;
 	urb->number_of_packets = 0;
@@ -1304,8 +1304,8 @@ static void prepare_playback_urb(struct snd_usb_substream *subs,
 	for (i = 0; i < ctx->packets; i++) {
 		counts = snd_usb_endpoint_next_packet_size(ep, ctx, i);
 		/* set up descriptor */
-		urb->iso_frame_desc[i].offset = frames * ep->stride;
-		urb->iso_frame_desc[i].length = counts * ep->stride;
+		urb->iso_frame_desc[i].offset = frames * stride;
+		urb->iso_frame_desc[i].length = counts * stride;
 		frames += counts;
 		urb->number_of_packets++;
 		subs->transfer_done += counts;
@@ -1320,14 +1320,14 @@ static void prepare_playback_urb(struct snd_usb_substream *subs,
 					frames -= subs->transfer_done;
 					counts -= subs->transfer_done;
 					urb->iso_frame_desc[i].length =
-						counts * ep->stride;
+						counts * stride;
 					subs->transfer_done = 0;
 				}
 				i++;
 				if (i < ctx->packets) {
 					/* add a transfer delimiter */
 					urb->iso_frame_desc[i].offset =
-						frames * ep->stride;
+						frames * stride;
 					urb->iso_frame_desc[i].length = 0;
 					urb->number_of_packets++;
 				}
@@ -1340,7 +1340,7 @@ static void prepare_playback_urb(struct snd_usb_substream *subs,
 		    !snd_usb_endpoint_implicit_feedback_sink(ep))
 			break;
 	}
-	bytes = frames * ep->stride;
+	bytes = frames * stride;
 
 	if (unlikely(ep->cur_format == SNDRV_PCM_FORMAT_DSD_U16_LE &&
 		     subs->cur_audiofmt->dsd_dop)) {
@@ -1350,14 +1350,14 @@ static void prepare_playback_urb(struct snd_usb_substream *subs,
 		/* bit-reverse the bytes */
 		u8 *buf = urb->transfer_buffer;
 		for (i = 0; i < bytes; i++) {
-			int idx = (subs->hwptr_done + i)
-				% (runtime->buffer_size * stride);
+			int idx = (subs->hwptr_done + i) % subs->buffer_bytes;
+
 			buf[i] = bitrev8(runtime->dma_area[idx]);
 		}
 
 		subs->hwptr_done += bytes;
-		if (subs->hwptr_done >= runtime->buffer_size * stride)
-			subs->hwptr_done -= runtime->buffer_size * stride;
+		if (subs->hwptr_done >= subs->buffer_bytes)
+			subs->hwptr_done -= subs->buffer_bytes;
 	} else {
 		/* usual PCM */
 		if (!subs->tx_length_quirk)
-- 
2.26.2

[PATCH 3/5] ALSA: usb-audio: Refactoring delay account code

[Takashi Iwai]

The PCM delay accounting in USB-audio driver is a bit complex to
follow, and this is an attempt to improve the readability and provide
some potential fix.

Basically, the PCM position delay is calculated from two factors: the
in-flight data on URBs and the USB frame counter.  For the playback
stream, we advance the hwptr already at submitting URBs.  Those
"in-flight" data amount is now tracked, and this is used as the base
value for the PCM delay correction.  The in-flight data is decreased
again at URB completion in return.  For the capture stream, OTOH,
there is no in-flight data, hence the delay base is zero.

The USB frame counter is used in addition for correcting the current
position.  The reference frame counter is updated at each submission
and receiving time, and the difference from the current counter value
is taken into account.

In this patch, each in-flight data bytes is recorded in the new
snd_usb_ctx.queued field, and the total in-flight amount is tracked in
snd_usb_substream.inflight_bytes field, as the replacement of
last_delay field.

Note that updating the hwptr after URB completion doesn't work for
PulseAudio who tries to scratch the buffer on the fly; USB-audio is
basically a double-buffer implementation, hence the scratching the
buffer can't work for the already submitted data.  So we always update
hwptr beforehand.  It's not ideal, but the delay account should give
enough correctness.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
---
 sound/usb/card.h     |   7 ++-
 sound/usb/endpoint.c |   1 +
 sound/usb/pcm.c      | 128 +++++++++++++++++--------------------------
 3 files changed, 56 insertions(+), 80 deletions(-)

diff --git a/sound/usb/card.h b/sound/usb/card.h
index b346653d4b76..5577a776561b 100644
--- a/sound/usb/card.h
+++ b/sound/usb/card.h
@@ -54,6 +54,7 @@ struct snd_urb_ctx {
 	struct snd_usb_endpoint *ep;
 	int index;	/* index for urb array */
 	int packets;	/* number of packets per urb */
+	int queued;	/* queued data bytes by this urb */
 	int packet_size[MAX_PACKS_HS]; /* size of packets for next submission */
 	struct list_head ready_list;
 };
@@ -159,8 +160,9 @@ struct snd_usb_substream {
 	unsigned int running: 1;	/* running status */
 
 	unsigned int buffer_bytes;	/* buffer size in bytes */
+	unsigned int inflight_bytes;	/* in-flight data bytes on buffer (for playback) */
 	unsigned int hwptr_done;	/* processed byte position in the buffer */
-	unsigned int transfer_done;		/* processed frames since last period update */
+	unsigned int transfer_done;	/* processed frames since last period update */
 	unsigned int frame_limit;	/* limits number of packets in URB */
 
 	/* data and sync endpoints for this stream */
@@ -175,8 +177,7 @@ struct snd_usb_substream {
 	struct list_head fmt_list;	/* format list */
 	spinlock_t lock;
 
-	int last_frame_number;          /* stored frame number */
-	int last_delay;                 /* stored delay */
+	unsigned int last_frame_number;	/* stored frame number */
 
 	struct {
 		int marker;
diff --git a/sound/usb/endpoint.c b/sound/usb/endpoint.c
index 014c43862826..da649211bff3 100644
--- a/sound/usb/endpoint.c
+++ b/sound/usb/endpoint.c
@@ -275,6 +275,7 @@ static void prepare_silent_urb(struct snd_usb_endpoint *ep,
 
 	urb->number_of_packets = ctx->packets;
 	urb->transfer_buffer_length = offs * ep->stride + ctx->packets * extra;
+	ctx->queued = 0;
 }
 
 /*
diff --git a/sound/usb/pcm.c b/sound/usb/pcm.c
index e8121af8e1d5..8ee45f2e8dce 100644
--- a/sound/usb/pcm.c
+++ b/sound/usb/pcm.c
@@ -30,14 +30,20 @@
 
 /* return the estimated delay based on USB frame counters */
 static snd_pcm_uframes_t snd_usb_pcm_delay(struct snd_usb_substream *subs,
-					   unsigned int rate)
+					   struct snd_pcm_runtime *runtime)
 {
-	int current_frame_number;
-	int frame_diff;
+	unsigned int current_frame_number;
+	unsigned int frame_diff;
 	int est_delay;
+	int queued;
 
-	if (!subs->last_delay)
-		return 0; /* short path */
+	if (subs->direction == SNDRV_PCM_STREAM_PLAYBACK) {
+		queued = bytes_to_frames(runtime, subs->inflight_bytes);
+		if (!queued)
+			return 0;
+	} else if (!subs->running) {
+		return 0;
+	}
 
 	current_frame_number = usb_get_current_frame_number(subs->dev);
 	/*
@@ -49,14 +55,14 @@ static snd_pcm_uframes_t snd_usb_pcm_delay(struct snd_usb_substream *subs,
 
 	/* Approximation based on number of samples per USB frame (ms),
 	   some truncation for 44.1 but the estimate is good enough */
-	est_delay =  frame_diff * rate / 1000;
-	if (subs->direction == SNDRV_PCM_STREAM_PLAYBACK)
-		est_delay = subs->last_delay - est_delay;
-	else
-		est_delay = subs->last_delay + est_delay;
+	est_delay = frame_diff * runtime->rate / 1000;
+
+	if (subs->direction == SNDRV_PCM_STREAM_PLAYBACK) {
+		est_delay = queued - est_delay;
+		if (est_delay < 0)
+			est_delay = 0;
+	}
 
-	if (est_delay < 0)
-		est_delay = 0;
 	return est_delay;
 }
 
@@ -65,17 +71,17 @@ static snd_pcm_uframes_t snd_usb_pcm_delay(struct snd_usb_substream *subs,
  */
 static snd_pcm_uframes_t snd_usb_pcm_pointer(struct snd_pcm_substream *substream)
 {
-	struct snd_usb_substream *subs = substream->runtime->private_data;
+	struct snd_pcm_runtime *runtime = substream->runtime;
+	struct snd_usb_substream *subs = runtime->private_data;
 	unsigned int hwptr_done;
 
 	if (atomic_read(&subs->stream->chip->shutdown))
 		return SNDRV_PCM_POS_XRUN;
 	spin_lock(&subs->lock);
 	hwptr_done = subs->hwptr_done;
-	substream->runtime->delay = snd_usb_pcm_delay(subs,
-						substream->runtime->rate);
+	runtime->delay = snd_usb_pcm_delay(subs, runtime);
 	spin_unlock(&subs->lock);
-	return hwptr_done / (substream->runtime->frame_bits >> 3);
+	return bytes_to_frames(runtime, hwptr_done);
 }
 
 /*
@@ -601,9 +607,9 @@ static int snd_usb_pcm_prepare(struct snd_pcm_substream *substream)
 
 	/* reset the pointer */
 	subs->buffer_bytes = frames_to_bytes(runtime, runtime->buffer_size);
+	subs->inflight_bytes = 0;
 	subs->hwptr_done = 0;
 	subs->transfer_done = 0;
-	subs->last_delay = 0;
 	subs->last_frame_number = 0;
 	runtime->delay = 0;
 
@@ -1156,14 +1162,9 @@ static void retire_capture_urb(struct snd_usb_substream *subs,
 			subs->transfer_done -= runtime->period_size;
 			period_elapsed = 1;
 		}
-		/* capture delay is by construction limited to one URB,
-		 * reset delays here
-		 */
-		runtime->delay = subs->last_delay = 0;
 
 		/* realign last_frame_number */
 		subs->last_frame_number = current_frame_number;
-		subs->last_frame_number &= 0xFF; /* keep 8 LSBs */
 
 		spin_unlock_irqrestore(&subs->lock, flags);
 		/* copy a data chunk */
@@ -1181,6 +1182,18 @@ static void retire_capture_urb(struct snd_usb_substream *subs,
 		snd_pcm_period_elapsed(subs->pcm_substream);
 }
 
+static void urb_ctx_queue_advance(struct snd_usb_substream *subs,
+				  struct urb *urb, unsigned int bytes)
+{
+	struct snd_urb_ctx *ctx = urb->context;
+
+	ctx->queued += bytes;
+	subs->inflight_bytes += bytes;
+	subs->hwptr_done += bytes;
+	if (subs->hwptr_done >= subs->buffer_bytes)
+		subs->hwptr_done -= subs->buffer_bytes;
+}
+
 static inline void fill_playback_urb_dsd_dop(struct snd_usb_substream *subs,
 					     struct urb *urb, unsigned int bytes)
 {
@@ -1191,6 +1204,7 @@ static inline void fill_playback_urb_dsd_dop(struct snd_usb_substream *subs,
 	u8 *dst = urb->transfer_buffer;
 	u8 *src = runtime->dma_area;
 	u8 marker[] = { 0x05, 0xfa };
+	unsigned int queued = 0;
 
 	/*
 	 * The DSP DOP format defines a way to transport DSD samples over
@@ -1229,12 +1243,11 @@ static inline void fill_playback_urb_dsd_dop(struct snd_usb_substream *subs,
 				dst[dst_idx++] = bitrev8(src[idx]);
 			else
 				dst[dst_idx++] = src[idx];
-
-			subs->hwptr_done++;
+			queued++;
 		}
 	}
-	if (subs->hwptr_done >= subs->buffer_bytes)
-		subs->hwptr_done -= subs->buffer_bytes;
+
+	urb_ctx_queue_advance(subs, urb, queued);
 }
 
 static void copy_to_urb(struct snd_usb_substream *subs, struct urb *urb,
@@ -1254,9 +1267,8 @@ static void copy_to_urb(struct snd_usb_substream *subs, struct urb *urb,
 		memcpy(urb->transfer_buffer + offset,
 		       runtime->dma_area + subs->hwptr_done, bytes);
 	}
-	subs->hwptr_done += bytes;
-	if (subs->hwptr_done >= subs->buffer_bytes)
-		subs->hwptr_done -= subs->buffer_bytes;
+
+	urb_ctx_queue_advance(subs, urb, bytes);
 }
 
 static unsigned int copy_to_urb_quirk(struct snd_usb_substream *subs,
@@ -1298,6 +1310,7 @@ static void prepare_playback_urb(struct snd_usb_substream *subs,
 	stride = ep->stride;
 
 	frames = 0;
+	ctx->queued = 0;
 	urb->number_of_packets = 0;
 	spin_lock_irqsave(&subs->lock, flags);
 	subs->frame_limit += ep->max_urb_frames;
@@ -1355,9 +1368,7 @@ static void prepare_playback_urb(struct snd_usb_substream *subs,
 			buf[i] = bitrev8(runtime->dma_area[idx]);
 		}
 
-		subs->hwptr_done += bytes;
-		if (subs->hwptr_done >= subs->buffer_bytes)
-			subs->hwptr_done -= subs->buffer_bytes;
+		urb_ctx_queue_advance(subs, urb, bytes);
 	} else {
 		/* usual PCM */
 		if (!subs->tx_length_quirk)
@@ -1367,14 +1378,7 @@ static void prepare_playback_urb(struct snd_usb_substream *subs,
 			/* bytes is now amount of outgoing data */
 	}
 
-	/* update delay with exact number of samples queued */
-	runtime->delay = subs->last_delay;
-	runtime->delay += frames;
-	subs->last_delay = runtime->delay;
-
-	/* realign last_frame_number */
 	subs->last_frame_number = usb_get_current_frame_number(subs->dev);
-	subs->last_frame_number &= 0xFF; /* keep 8 LSBs */
 
 	if (subs->trigger_tstamp_pending_update) {
 		/* this is the first actual URB submitted,
@@ -1398,48 +1402,17 @@ static void retire_playback_urb(struct snd_usb_substream *subs,
 			       struct urb *urb)
 {
 	unsigned long flags;
-	struct snd_pcm_runtime *runtime = subs->pcm_substream->runtime;
-	struct snd_usb_endpoint *ep = subs->data_endpoint;
-	int processed = urb->transfer_buffer_length / ep->stride;
-	int est_delay;
-
-	/* ignore the delay accounting when processed=0 is given, i.e.
-	 * silent payloads are processed before handling the actual data
-	 */
-	if (!processed)
-		return;
+	struct snd_urb_ctx *ctx = urb->context;
 
 	spin_lock_irqsave(&subs->lock, flags);
-	if (!subs->last_delay)
-		goto out; /* short path */
-
-	est_delay = snd_usb_pcm_delay(subs, runtime->rate);
-	/* update delay with exact number of samples played */
-	if (processed > subs->last_delay)
-		subs->last_delay = 0;
-	else
-		subs->last_delay -= processed;
-	runtime->delay = subs->last_delay;
-
-	/*
-	 * Report when delay estimate is off by more than 2ms.
-	 * The error should be lower than 2ms since the estimate relies
-	 * on two reads of a counter updated every ms.
-	 */
-	if (abs(est_delay - subs->last_delay) * 1000 > runtime->rate * 2)
-		dev_dbg_ratelimited(&subs->dev->dev,
-			"delay: estimated %d, actual %d\n",
-			est_delay, subs->last_delay);
-
-	if (!subs->running) {
-		/* update last_frame_number for delay counting here since
-		 * prepare_playback_urb won't be called during pause
-		 */
-		subs->last_frame_number =
-			usb_get_current_frame_number(subs->dev) & 0xff;
+	if (ctx->queued) {
+		if (subs->inflight_bytes >= ctx->queued)
+			subs->inflight_bytes -= ctx->queued;
+		else
+			subs->inflight_bytes = 0;
 	}
 
- out:
+	subs->last_frame_number = usb_get_current_frame_number(subs->dev);
 	spin_unlock_irqrestore(&subs->lock, flags);
 }
 
@@ -1504,6 +1477,7 @@ static int snd_usb_substream_capture_trigger(struct snd_pcm_substream *substream
 		snd_usb_endpoint_set_callback(subs->data_endpoint,
 					      NULL, retire_capture_urb,
 					      subs);
+		subs->last_frame_number = usb_get_current_frame_number(subs->dev);
 		subs->running = 1;
 		dev_dbg(&subs->dev->dev, "%d:%d Start Capture PCM\n",
 			subs->cur_audiofmt->iface,
-- 
2.26.2

[PATCH 4/5] ALSA: usb-audio: Factor out DSD bitrev copy function

[Takashi Iwai]

Just minor code refactoring.  Like DOP DSD code, it can be better in a
separate function for code readability.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
---
 sound/usb/pcm.c | 28 +++++++++++++++++++---------
 1 file changed, 19 insertions(+), 9 deletions(-)

diff --git a/sound/usb/pcm.c b/sound/usb/pcm.c
index 8ee45f2e8dce..e26d37365f02 100644
--- a/sound/usb/pcm.c
+++ b/sound/usb/pcm.c
@@ -1250,6 +1250,24 @@ static inline void fill_playback_urb_dsd_dop(struct snd_usb_substream *subs,
 	urb_ctx_queue_advance(subs, urb, queued);
 }
 
+/* copy bit-reversed bytes onto transfer buffer */
+static void fill_playback_urb_dsd_bitrev(struct snd_usb_substream *subs,
+					 struct urb *urb, unsigned int bytes)
+{
+	struct snd_pcm_runtime *runtime = subs->pcm_substream->runtime;
+	const u8 *src = runtime->dma_area;
+	u8 *buf = urb->transfer_buffer;
+	int i, ofs = subs->hwptr_done;
+
+	for (i = 0; i < bytes; i++) {
+		*buf++ = bitrev8(src[ofs]);
+		if (++ofs >= subs->buffer_bytes)
+			ofs = 0;
+	}
+
+	urb_ctx_queue_advance(subs, urb, bytes);
+}
+
 static void copy_to_urb(struct snd_usb_substream *subs, struct urb *urb,
 			int offset, int stride, unsigned int bytes)
 {
@@ -1360,15 +1378,7 @@ static void prepare_playback_urb(struct snd_usb_substream *subs,
 		fill_playback_urb_dsd_dop(subs, urb, bytes);
 	} else if (unlikely(ep->cur_format == SNDRV_PCM_FORMAT_DSD_U8 &&
 			   subs->cur_audiofmt->dsd_bitrev)) {
-		/* bit-reverse the bytes */
-		u8 *buf = urb->transfer_buffer;
-		for (i = 0; i < bytes; i++) {
-			int idx = (subs->hwptr_done + i) % subs->buffer_bytes;
-
-			buf[i] = bitrev8(runtime->dma_area[idx]);
-		}
-
-		urb_ctx_queue_advance(subs, urb, bytes);
+		fill_playback_urb_dsd_bitrev(subs, urb, bytes);
 	} else {
 		/* usual PCM */
 		if (!subs->tx_length_quirk)
-- 
2.26.2

[PATCH 5/5] ALSA: usb-audio: Reduce latency at playback start

[Takashi Iwai]

USB-audio driver behaves a bit strangely for the playback stream --
namely, it starts sending silent packets at PCM prepare state while
the actual data is submitted at first when the trigger START is kicked
off.  This is a workaround for the behavior where URBs are processed
too quickly at the beginning.  That is, if we start submitting URBs at
trigger START, the first few URBs will be immediately completed, and
this would result in the immediate period-elapsed calls right after
the start, which may confuse applications.

OTOH, submitting the data after silent URBs would, of course, result
in a certain delay of the actual data processing, and this is rather
more serious problem on modern systems, in practice.

This patch tries to revert the workaround and lets the URB submission
starting at PCM trigger for the playback again.  As far as I've tested
with various backends (native ALSA, PA, JACK, PW), I haven't seen any
problems (famous last words :)

Note that the capture stream handling needs no such workaround, since
the capture is driven per received URB.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
---
 sound/usb/pcm.c | 14 +++++++++-----
 1 file changed, 9 insertions(+), 5 deletions(-)

diff --git a/sound/usb/pcm.c b/sound/usb/pcm.c
index e26d37365f02..c66831ee15f9 100644
--- a/sound/usb/pcm.c
+++ b/sound/usb/pcm.c
@@ -613,11 +613,6 @@ static int snd_usb_pcm_prepare(struct snd_pcm_substream *substream)
 	subs->last_frame_number = 0;
 	runtime->delay = 0;
 
-	/* for playback, submit the URBs now; otherwise, the first hwptr_done
-	 * updates for all URBs would happen at the same time when starting */
-	if (subs->direction == SNDRV_PCM_STREAM_PLAYBACK)
-		ret = start_endpoints(subs);
-
  unlock:
 	snd_usb_unlock_shutdown(chip);
 	return ret;
@@ -1430,6 +1425,7 @@ static int snd_usb_substream_playback_trigger(struct snd_pcm_substream *substrea
 					      int cmd)
 {
 	struct snd_usb_substream *subs = substream->runtime->private_data;
+	int err;
 
 	switch (cmd) {
 	case SNDRV_PCM_TRIGGER_START:
@@ -1440,6 +1436,14 @@ static int snd_usb_substream_playback_trigger(struct snd_pcm_substream *substrea
 					      prepare_playback_urb,
 					      retire_playback_urb,
 					      subs);
+		if (cmd == SNDRV_PCM_TRIGGER_START) {
+			err = start_endpoints(subs);
+			if (err < 0) {
+				snd_usb_endpoint_set_callback(subs->data_endpoint,
+							      NULL, NULL, NULL);
+				return err;
+			}
+		}
 		subs->running = 1;
 		dev_dbg(&subs->dev->dev, "%d:%d Start Playback PCM\n",
 			subs->cur_audiofmt->iface,
-- 
2.26.2