oggdemux: use an adaptive chunksize for performance reasons

Ogg data is read chunk by chunk, and the chunk size used was
originally taken from libvorbisfile. However, this value leads
to poor performance when used on an Ogg file with large pages
(Ogg pages can be close to 64 KB).

We can't just use a larger chunk size, since this will decrease
performance on small page streams, so we use an adaptive scheme
where the chunk size is twice the largest page size we've seen
so far in the stream. For "typical" Ogg/Vorbis, this gives us
almost the same chunk size (a bit lower), and this lets us get
better performance on streams with large pages.

1 files changed, 36 insertions, 14 deletions

diff --git a/ext/ogg/gstoggdemux.c b/ext/ogg/gstoggdemux.c
index 9d27942b4..65ac060b9 100644
--- a/ext/ogg/gstoggdemux.c
+++ b/ext/ogg/gstoggdemux.c
@@ -2070,6 +2070,8 @@ gst_ogg_demux_init (GstOggDemux * ogg)
   ogg->stats_bisection_max_steps[1] = 0;
 
   ogg->newsegment = NULL;
+
+  ogg->chunk_size = CHUNKSIZE;
 }
 
 static void
@@ -2110,6 +2112,8 @@ gst_ogg_demux_reset_streams (GstOggDemux * ogg)
   GST_DEBUG_OBJECT (ogg, "Resetting current chain");
   ogg->current_chain = NULL;
   ogg->resync = TRUE;
+
+  ogg->chunk_size = CHUNKSIZE;
 }
 
 static gboolean
@@ -2285,15 +2289,17 @@ gst_ogg_demux_get_data (GstOggDemux * ogg, gint64 end_offset)
   if (ogg->read_offset == ogg->length)
     goto eos;
 
-  oggbuffer = ogg_sync_buffer (&ogg->sync, CHUNKSIZE);
+  oggbuffer = ogg_sync_buffer (&ogg->sync, ogg->chunk_size);
   if (G_UNLIKELY (oggbuffer == NULL))
     goto no_buffer;
 
   buffer =
-      gst_buffer_new_wrapped_full (0, oggbuffer, CHUNKSIZE, 0, CHUNKSIZE, NULL,
-      NULL);
+      gst_buffer_new_wrapped_full (0, oggbuffer, ogg->chunk_size, 0,
+      ogg->chunk_size, NULL, NULL);
 
-  ret = gst_pad_pull_range (ogg->sinkpad, ogg->read_offset, CHUNKSIZE, &buffer);
+  ret =
+      gst_pad_pull_range (ogg->sinkpad, ogg->read_offset, ogg->chunk_size,
+      &buffer);
   if (ret != GST_FLOW_OK)
     goto error;
 
@@ -2440,11 +2446,12 @@ gst_ogg_demux_get_prev_page (GstOggDemux * ogg, ogg_page * og, gint64 * offset)
   GST_LOG_OBJECT (ogg, "getting page before %" G_GINT64_FORMAT, begin);
 
   while (cur_offset == -1) {
-    begin -= CHUNKSIZE;
+    begin -= ogg->chunk_size;
     if (begin < 0)
       begin = 0;
 
-    /* seek CHUNKSIZE back */
+    /* seek ogg->chunk_size back */
+    GST_LOG_OBJECT (ogg, "seeking back to %" G_GINT64_FORMAT, begin);
     gst_ogg_demux_seek (ogg, begin);
 
     /* now continue reading until we run out of data, if we find a page
@@ -2769,13 +2776,14 @@ do_binary_search (GstOggDemux * ogg, GstOggChain * chain, gint64 begin,
   while (begin < end) {
     gint64 bisect;
 
-    if ((end - begin < CHUNKSIZE) || (endtime == begintime)) {
+    if ((end - begin < ogg->chunk_size) || (endtime == begintime)) {
       bisect = begin;
     } else {
       /* take a (pretty decent) guess, avoiding overflow */
       gint64 rate = (end - begin) * GST_MSECOND / (endtime - begintime);
 
-      bisect = (target - begintime) / GST_MSECOND * rate + begin - CHUNKSIZE;
+      bisect =
+          (target - begintime) / GST_MSECOND * rate + begin - ogg->chunk_size;
 
       if (bisect <= begin)
         bisect = begin;
@@ -2802,7 +2810,7 @@ do_binary_search (GstOggDemux * ogg, GstOggChain * chain, gint64 begin,
           if (bisect == 0)
             goto seek_error;
 
-          bisect -= CHUNKSIZE;
+          bisect -= ogg->chunk_size;
           if (bisect <= begin)
             bisect = begin + 1;
 
@@ -2861,7 +2869,7 @@ do_binary_search (GstOggDemux * ogg, GstOggChain * chain, gint64 begin,
           } else {
             if (end == ogg->offset) {   /* we're pretty close - we'd be stuck in */
               end = result;
-              bisect -= CHUNKSIZE;      /* an endless loop otherwise. */
+              bisect -= ogg->chunk_size;        /* an endless loop otherwise. */
               if (bisect <= begin)
                 bisect = begin + 1;
               gst_ogg_demux_seek (ogg, bisect);
@@ -3517,7 +3525,7 @@ gst_ogg_demux_perform_seek_push (GstOggDemux * ogg, GstEvent * event)
 
   /* offset by typical page length, and ensure our best guess is within
      reasonable bounds */
-  best -= CHUNKSIZE;
+  best -= ogg->chunk_size;
   if (best < 0)
     best = 0;
   if (ogg->push_byte_length > 0 && best >= ogg->push_byte_length)
@@ -3618,7 +3626,7 @@ gst_ogg_demux_bisect_forward_serialno (GstOggDemux * ogg,
   while (searched < endsearched) {
     gint64 bisect;
 
-    if (endsearched - searched < CHUNKSIZE) {
+    if (endsearched - searched < ogg->chunk_size) {
       bisect = searched;
     } else {
       bisect = (searched + endsearched) / 2;
@@ -3843,7 +3851,7 @@ gst_ogg_demux_read_end_chain (GstOggDemux * ogg, GstOggChain * chain)
   gint i;
 
   while (!done) {
-    begin -= CHUNKSIZE;
+    begin -= ogg->chunk_size;
     if (begin < 0)
       begin = 0;
 
@@ -4079,6 +4087,17 @@ no_last_page:
   }
 }
 
+static void
+gst_ogg_demux_update_chunk_size (GstOggDemux * ogg, ogg_page * page)
+{
+  long size = page->header_len + page->body_len;
+  long chunk_size = size * 2;
+  if (chunk_size > ogg->chunk_size) {
+    GST_LOG_OBJECT (ogg, "Updating chunk size to %ld", chunk_size);
+    ogg->chunk_size = chunk_size;
+  }
+}
+
 static GstFlowReturn
 gst_ogg_demux_handle_page (GstOggDemux * ogg, ogg_page * page)
 {
@@ -4090,6 +4109,8 @@ gst_ogg_demux_handle_page (GstOggDemux * ogg, ogg_page * page)
   serialno = ogg_page_serialno (page);
   granule = ogg_page_granulepos (page);
 
+  gst_ogg_demux_update_chunk_size (ogg, page);
+
   GST_LOG_OBJECT (ogg,
       "processing ogg page (serial %08x, "
       "pageno %ld, granulepos %" G_GINT64_FORMAT ", bos %d)", serialno,
@@ -4340,7 +4361,8 @@ gst_ogg_demux_loop_forward (GstOggDemux * ogg)
   }
 
   GST_LOG_OBJECT (ogg, "pull data %" G_GINT64_FORMAT, ogg->offset);
-  ret = gst_pad_pull_range (ogg->sinkpad, ogg->offset, CHUNKSIZE, &buffer);
+  ret =
+      gst_pad_pull_range (ogg->sinkpad, ogg->offset, ogg->chunk_size, &buffer);
   if (ret != GST_FLOW_OK) {
     GST_LOG_OBJECT (ogg, "Failed pull_range");
     goto done;