HTTP: improve performance by reducing forward seeks

This commit optimizes HTTP performance by reducing forward seeks, instead
favoring a read-ahead and discard on the current connection (referred to
as a short seek) for seeks that are within a TCP window's worth of data.
This improves performance because with TCP flow control, a window's worth
of data will be in the local socket buffer already or in-flight from the
sender once congestion control on the sender is fully utilizing the window.

Note: this approach doesn't attempt to differentiate from a newly opened
connection which may not be fully utilizing the window due to congestion
control vs one that is. The receiver can't get at this information, so we
assume worst case; that full window is in use (we did advertise it after all)
and that data could be in-flight

The previous behavior of closing the connection, then opening a new
with a new HTTP range value results in a massive amounts of discarded
and re-sent data when large TCP windows are used.  This has been observed
on MacOS/iOS which starts with an initial window of 256KB and grows up to
1MB depending on the bandwidth-product delay.

When seeking within a window's worth of data and we close the connection,
then open a new one within the same window's worth of data, we discard
from the current offset till the end of the window.  Then on the new
connection the server ends up re-sending the previous data from new
offset till the end of old window.

Example (assumes full window utilization):

TCP window size: 64KB
Position: 32KB
Forward seek position: 40KB

      *                      (Next window)
32KB |--------------| 96KB |---------------| 160KB
        *
  40KB |---------------| 104KB

Re-sent amount: 96KB - 40KB = 56KB

For a real world test example, I have MP4 file of ~25MB, which ffplay
only reads ~16MB and performs 177 seeks. With current ffmpeg, this results
in 177 HTTP GETs and ~73MB worth of TCP data communication.  With this
patch, ffmpeg issues 4 HTTP GETs and 3 seeks for a total of ~22MB of TCP data
communication.

To support this feature, the short seek logic in avio_seek() has been
extended to call a function to get the short seek threshold value.  This
callback has been plumbed to the URLProtocol structure, which now has
infrastructure in HTTP and TCP to get the underlying receiver window size
via SO_RCVBUF.  If the underlying URL and protocol don't support returning
a short seek threshold, the default s->short_seek_threshold is used

This feature has been tested on Windows 7 and MacOS/iOS.  Windows support
is slightly complicated by the fact that when TCP window auto-tuning is
enabled, SO_RCVBUF doesn't report the real window size, but it does if
SO_RCVBUF was manually set (disabling auto-tuning). So we can only use
this optimization on Windows in the later case

Signed-off-by: Joel Cunningham <joel.cunningham@me.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

6 files changed, 68 insertions, 1 deletions

diff --git a/libavformat/avio.c b/libavformat/avio.c
index 3606eb0fda..62233a633c 100644
--- a/libavformat/avio.c
+++ b/libavformat/avio.c
@@ -645,6 +645,13 @@ int ffurl_get_multi_file_handle(URLContext *h, int **handles, int *numhandles)
     return h->prot->url_get_multi_file_handle(h, handles, numhandles);
 }
 
+int ffurl_get_short_seek(URLContext *h)
+{
+    if (!h->prot->url_get_short_seek)
+        return AVERROR(ENOSYS);
+    return h->prot->url_get_short_seek(h);
+}
+
 int ffurl_shutdown(URLContext *h, int flags)
 {
     if (!h->prot->url_shutdown)
diff --git a/libavformat/avio.h b/libavformat/avio.h
index e2cb4af7a2..8040094423 100644
--- a/libavformat/avio.h
+++ b/libavformat/avio.h
@@ -313,6 +313,12 @@ typedef struct AVIOContext {
      */
     enum AVIODataMarkerType current_type;
     int64_t last_time;
+
+    /**
+     * A callback that is used instead of short_seek_threshold.
+     * This is current internal only, do not use from outside.
+     */
+    int (*short_seek_get)(void *opaque);
 } AVIOContext;
 
 /**
diff --git a/libavformat/aviobuf.c b/libavformat/aviobuf.c
index bf7e5f85a0..4ade4d0d7e 100644
--- a/libavformat/aviobuf.c
+++ b/libavformat/aviobuf.c
@@ -119,6 +119,7 @@ int ffio_init_context(AVIOContext *s,
     s->ignore_boundary_point = 0;
     s->current_type          = AVIO_DATA_MARKER_UNKNOWN;
     s->last_time             = AV_NOPTS_VALUE;
+    s->short_seek_get        = NULL;
 
     return 0;
 }
@@ -233,6 +234,7 @@ int64_t avio_seek(AVIOContext *s, int64_t offset, int whence)
     int64_t pos;
     int force = whence & AVSEEK_FORCE;
     int buffer_size;
+    int short_seek;
     whence &= ~AVSEEK_FORCE;
 
     if(!s)
@@ -254,13 +256,21 @@ int64_t avio_seek(AVIOContext *s, int64_t offset, int whence)
     if (offset < 0)
         return AVERROR(EINVAL);
 
+    if (s->short_seek_get) {
+        short_seek = s->short_seek_get(s->opaque);
+        /* fallback to default short seek */
+        if (short_seek <= 0)
+            short_seek = s->short_seek_threshold;
+    } else
+        short_seek = s->short_seek_threshold;
+
     offset1 = offset - pos; // "offset1" is the relative offset from the beginning of s->buffer
     if (!s->must_flush && (!s->direct || !s->seek) &&
         offset1 >= 0 && offset1 <= buffer_size - s->write_flag) {
         /* can do the seek inside the buffer */
         s->buf_ptr = s->buffer + offset1;
     } else if ((!s->seekable ||
-               offset1 <= buffer_size + s->short_seek_threshold) &&
+               offset1 <= buffer_size + short_seek) &&
                !s->write_flag && offset1 >= 0 &&
                (!s->direct || !s->seek) &&
               (whence != SEEK_END || force)) {
@@ -858,6 +868,12 @@ static int64_t io_seek(void *opaque, int64_t offset, int whence)
     return ffurl_seek(internal->h, offset, whence);
 }
 
+static int io_short_seek(void *opaque)
+{
+    AVIOInternal *internal = opaque;
+    return ffurl_get_short_seek(internal->h);
+}
+
 static int io_read_pause(void *opaque, int pause)
 {
     AVIOInternal *internal = opaque;
@@ -919,6 +935,7 @@ int ffio_fdopen(AVIOContext **s, URLContext *h)
         (*s)->read_pause = io_read_pause;
         (*s)->read_seek  = io_read_seek;
     }
+    (*s)->short_seek_get = io_short_seek;
     (*s)->av_class = &ff_avio_class;
     return 0;
 fail:
diff --git a/libavformat/http.c b/libavformat/http.c
index bd1be3f7bb..293a8a7204 100644
--- a/libavformat/http.c
+++ b/libavformat/http.c
@@ -1533,6 +1533,12 @@ static int http_get_file_handle(URLContext *h)
     return ffurl_get_file_handle(s->hd);
 }
 
+static int http_get_short_seek(URLContext *h)
+{
+    HTTPContext *s = h->priv_data;
+    return ffurl_get_short_seek(s->hd);
+}
+
 #define HTTP_CLASS(flavor)                          \
 static const AVClass flavor ## _context_class = {   \
     .class_name = # flavor,                         \
@@ -1554,6 +1560,7 @@ const URLProtocol ff_http_protocol = {
     .url_seek            = http_seek,
     .url_close           = http_close,
     .url_get_file_handle = http_get_file_handle,
+    .url_get_short_seek  = http_get_short_seek,
     .url_shutdown        = http_shutdown,
     .priv_data_size      = sizeof(HTTPContext),
     .priv_data_class     = &http_context_class,
@@ -1573,6 +1580,7 @@ const URLProtocol ff_https_protocol = {
     .url_seek            = http_seek,
     .url_close           = http_close,
     .url_get_file_handle = http_get_file_handle,
+    .url_get_short_seek  = http_get_short_seek,
     .url_shutdown        = http_shutdown,
     .priv_data_size      = sizeof(HTTPContext),
     .priv_data_class     = &https_context_class,
diff --git a/libavformat/tcp.c b/libavformat/tcp.c
index 5f00ba7588..3055e48015 100644
--- a/libavformat/tcp.c
+++ b/libavformat/tcp.c
@@ -266,6 +266,26 @@ static int tcp_get_file_handle(URLContext *h)
     return s->fd;
 }
 
+static int tcp_get_window_size(URLContext *h)
+{
+    TCPContext *s = h->priv_data;
+    int avail;
+    int avail_len = sizeof(avail);
+
+#if HAVE_WINSOCK2_H
+    /* SO_RCVBUF with winsock only reports the actual TCP window size when
+    auto-tuning has been disabled via setting SO_RCVBUF */
+    if (s->recv_buffer_size < 0) {
+        return AVERROR(ENOSYS);
+    }
+#endif
+
+    if (getsockopt(s->fd, SOL_SOCKET, SO_RCVBUF, &avail, &avail_len)) {
+        return ff_neterrno();
+    }
+    return avail;
+}
+
 const URLProtocol ff_tcp_protocol = {
     .name                = "tcp",
     .url_open            = tcp_open,
@@ -274,6 +294,7 @@ const URLProtocol ff_tcp_protocol = {
     .url_write           = tcp_write,
     .url_close           = tcp_close,
     .url_get_file_handle = tcp_get_file_handle,
+    .url_get_short_seek  = tcp_get_window_size,
     .url_shutdown        = tcp_shutdown,
     .priv_data_size      = sizeof(TCPContext),
     .flags               = URL_PROTOCOL_FLAG_NETWORK,
diff --git a/libavformat/url.h b/libavformat/url.h
index 5c5024553e..910f1e00b3 100644
--- a/libavformat/url.h
+++ b/libavformat/url.h
@@ -84,6 +84,7 @@ typedef struct URLProtocol {
     int (*url_get_file_handle)(URLContext *h);
     int (*url_get_multi_file_handle)(URLContext *h, int **handles,
                                      int *numhandles);
+    int (*url_get_short_seek)(URLContext *h);
     int (*url_shutdown)(URLContext *h, int flags);
     int priv_data_size;
     const AVClass *priv_data_class;
@@ -249,6 +250,13 @@ int ffurl_get_file_handle(URLContext *h);
 int ffurl_get_multi_file_handle(URLContext *h, int **handles, int *numhandles);
 
 /**
+ * Return the current short seek threshold value for this URL.
+ *
+ * @return threshold (>0) on success or <=0 on error.
+ */
+int ffurl_get_short_seek(URLContext *h);
+
+/**
  * Signal the URLContext that we are done reading or writing the stream.
  *
  * @param h pointer to the resource