All of lore.kernel.org
 help / color / mirror / Atom feed
From: "Daniel P. Berrange" <berrange@redhat.com>
To: qemu-devel@nongnu.org
Cc: Peter Maydell <peter.maydell@linaro.org>,
	"Daniel P. Berrange" <berrange@redhat.com>
Subject: [Qemu-devel] [PULL v1 05/11] io: simplify websocket ping reply handling
Date: Mon, 16 Oct 2017 21:16:44 +0100	[thread overview]
Message-ID: <20171016201650.18399-6-berrange@redhat.com> (raw)
In-Reply-To: <20171016201650.18399-1-berrange@redhat.com>

We must ensure we don't get flooded with ping replies if the outbound
channel is slow. Currently we do this by keeping the ping reply in a
separate temporary buffer and only writing it if the encoutput buffer
is completely empty. This is overly pessimistic, as it is reasonable
to add a ping reply to the encoutput buffer even if it has previous
data in it, as long as that previous data doesn't include a ping
reply.

To track this better, put the ping reply directly into the encoutput
buffer, and then record the size of encoutput at this time in
pong_remain. As we write encoutput to the underlying channel, we
can decrement the pong_remain counter. Once it hits zero, we can
accept further ping replies for transmission.

Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
---
 include/io/channel-websock.h |  2 +-
 io/channel-websock.c         | 28 +++++++++++++++-------------
 2 files changed, 16 insertions(+), 14 deletions(-)

diff --git a/include/io/channel-websock.h b/include/io/channel-websock.h
index ff32d8651b..3762707b9c 100644
--- a/include/io/channel-websock.h
+++ b/include/io/channel-websock.h
@@ -60,8 +60,8 @@ struct QIOChannelWebsock {
     Buffer encoutput;
     Buffer rawinput;
     Buffer rawoutput;
-    Buffer ping_reply;
     size_t payload_remain;
+    size_t pong_remain;
     QIOChannelWebsockMask mask;
     guint io_tag;
     Error *io_err;
diff --git a/io/channel-websock.c b/io/channel-websock.c
index 04bcc059cd..6083f74c9b 100644
--- a/io/channel-websock.c
+++ b/io/channel-websock.c
@@ -825,11 +825,14 @@ static int qio_channel_websock_decode_payload(QIOChannelWebsock *ioc,
         }
         return -1;
     } else if (ioc->opcode == QIO_CHANNEL_WEBSOCK_OPCODE_PING) {
-        /* ping frames produce an immediate reply */
-        buffer_reset(&ioc->ping_reply);
-        qio_channel_websock_encode_buffer(
-            ioc, &ioc->ping_reply, QIO_CHANNEL_WEBSOCK_OPCODE_PONG,
-            &ioc->encinput);
+        /* ping frames produce an immediate reply, as long as we've not still
+         * got a previous pong queued, in which case we drop the new pong */
+        if (ioc->pong_remain == 0) {
+            qio_channel_websock_encode_buffer(
+                ioc, &ioc->encoutput, QIO_CHANNEL_WEBSOCK_OPCODE_PONG,
+                &ioc->encinput);
+            ioc->pong_remain = ioc->encoutput.offset;
+        }
     }   /* pong frames are ignored */
 
     if (payload_len) {
@@ -888,7 +891,6 @@ static void qio_channel_websock_finalize(Object *obj)
     buffer_free(&ioc->encoutput);
     buffer_free(&ioc->rawinput);
     buffer_free(&ioc->rawoutput);
-    buffer_free(&ioc->ping_reply);
     object_unref(OBJECT(ioc->master));
     if (ioc->io_tag) {
         g_source_remove(ioc->io_tag);
@@ -946,12 +948,7 @@ static ssize_t qio_channel_websock_write_wire(QIOChannelWebsock *ioc,
     ssize_t ret;
     ssize_t done = 0;
 
-    /* ping replies take priority over binary data */
-    if (!ioc->ping_reply.offset) {
-        qio_channel_websock_encode(ioc);
-    } else if (!ioc->encoutput.offset) {
-        buffer_move_empty(&ioc->encoutput, &ioc->ping_reply);
-    }
+    qio_channel_websock_encode(ioc);
 
     while (ioc->encoutput.offset > 0) {
         ret = qio_channel_write(ioc->master,
@@ -968,6 +965,11 @@ static ssize_t qio_channel_websock_write_wire(QIOChannelWebsock *ioc,
         }
         buffer_advance(&ioc->encoutput, ret);
         done += ret;
+        if (ioc->pong_remain < ret) {
+            ioc->pong_remain = 0;
+        } else {
+            ioc->pong_remain -= ret;
+        }
     }
     return done;
 }
@@ -1026,7 +1028,7 @@ static void qio_channel_websock_set_watch(QIOChannelWebsock *ioc)
         return;
     }
 
-    if (ioc->encoutput.offset || ioc->ping_reply.offset) {
+    if (ioc->encoutput.offset) {
         cond |= G_IO_OUT;
     }
     if (ioc->encinput.offset < QIO_CHANNEL_WEBSOCK_MAX_BUFFER &&
-- 
2.13.5

  parent reply	other threads:[~2017-10-16 20:17 UTC|newest]

Thread overview: 15+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-10-16 20:16 [Qemu-devel] [PULL v1 00/11] Merge QIO 2017-10-16 Daniel P. Berrange
2017-10-16 20:16 ` [Qemu-devel] [PULL v1 01/11] sockets: factor out a new try_bind() function Daniel P. Berrange
2017-10-16 20:16 ` [Qemu-devel] [PULL v1 02/11] sockets: factor out create_fast_reuse_socket Daniel P. Berrange
2017-10-16 20:16 ` [Qemu-devel] [PULL v1 03/11] sockets: Handle race condition between binds to the same port Daniel P. Berrange
2017-11-03 18:54   ` Peter Maydell
2017-11-06 10:40     ` Daniel P. Berrange
2017-10-16 20:16 ` [Qemu-devel] [PULL v1 04/11] io: monitor encoutput buffer size from websocket GSource Daniel P. Berrange
2017-10-16 20:16 ` Daniel P. Berrange [this message]
2017-10-16 20:16 ` [Qemu-devel] [PULL v1 06/11] io: get rid of qio_channel_websock_encode helper method Daniel P. Berrange
2017-10-16 20:16 ` [Qemu-devel] [PULL v1 07/11] io: pass a struct iovec into qio_channel_websock_encode Daniel P. Berrange
2017-10-16 20:16 ` [Qemu-devel] [PULL v1 08/11] io: get rid of bounce buffering in websock write path Daniel P. Berrange
2017-10-16 20:16 ` [Qemu-devel] [PULL v1 09/11] io: cope with websock 'Connection' header having multiple values Daniel P. Berrange
2017-10-16 20:16 ` [Qemu-devel] [PULL v1 10/11] io: add trace points for websocket HTTP protocol headers Daniel P. Berrange
2017-10-16 20:16 ` [Qemu-devel] [PULL v1 11/11] io: fix mem leak in websock error path Daniel P. Berrange
2017-10-17 12:12 ` [Qemu-devel] [PULL v1 00/11] Merge QIO 2017-10-16 Peter Maydell

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20171016201650.18399-6-berrange@redhat.com \
    --to=berrange@redhat.com \
    --cc=peter.maydell@linaro.org \
    --cc=qemu-devel@nongnu.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.