qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Jason Wang <jasowang@redhat.com>
To: peter.maydell@linaro.org, qemu-devel@nongnu.org
Cc: Zhang Chen <zhangchen.fnst@cn.fujitsu.com>,
	Jason Wang <jasowang@redhat.com>
Subject: [Qemu-devel] [PULL 13/18] net/colo-compare.c: Optimize unpredictable tcp options comparison
Date: Fri,  8 Sep 2017 10:05:19 +0800	[thread overview]
Message-ID: <1504836324-4871-14-git-send-email-jasowang@redhat.com> (raw)
In-Reply-To: <1504836324-4871-1-git-send-email-jasowang@redhat.com>

From: Zhang Chen <zhangchen.fnst@cn.fujitsu.com>

When network is busy, some tcp options(like sack) will unpredictable
occur in primary side or secondary side. it will make packet size
not same, but the two packet's payload is identical. colo just
care about packet payload, so we skip the option field.

Signed-off-by: Zhang Chen <zhangchen.fnst@cn.fujitsu.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
---
 net/colo-compare.c | 39 +++++++++++++++++++++++++++------------
 1 file changed, 27 insertions(+), 12 deletions(-)

diff --git a/net/colo-compare.c b/net/colo-compare.c
index 5fe8e3f..612fbac 100644
--- a/net/colo-compare.c
+++ b/net/colo-compare.c
@@ -186,7 +186,10 @@ static int packet_enqueue(CompareState *s, int mode)
  * return:    0  means packet same
  *            > 0 || < 0 means packet different
  */
-static int colo_packet_compare_common(Packet *ppkt, Packet *spkt, int offset)
+static int colo_packet_compare_common(Packet *ppkt,
+                                      Packet *spkt,
+                                      int poffset,
+                                      int soffset)
 {
     if (trace_event_get_state_backends(TRACE_COLO_COMPARE_MISCOMPARE)) {
         char pri_ip_src[20], pri_ip_dst[20], sec_ip_src[20], sec_ip_dst[20];
@@ -201,12 +204,13 @@ static int colo_packet_compare_common(Packet *ppkt, Packet *spkt, int offset)
                                    sec_ip_src, sec_ip_dst);
     }
 
-    offset = ppkt->vnet_hdr_len + offset;
+    poffset = ppkt->vnet_hdr_len + poffset;
+    soffset = ppkt->vnet_hdr_len + soffset;
 
-    if (ppkt->size == spkt->size) {
-        return memcmp(ppkt->data + offset,
-                      spkt->data + offset,
-                      spkt->size - offset);
+    if (ppkt->size - poffset == spkt->size - soffset) {
+        return memcmp(ppkt->data + poffset,
+                      spkt->data + soffset,
+                      spkt->size - soffset);
     } else {
         trace_colo_compare_main("Net packet size are not the same");
         return -1;
@@ -263,13 +267,22 @@ static int colo_packet_compare_tcp(Packet *spkt, Packet *ppkt)
      * so we just need skip this field.
      */
     if (ptcp->th_off > 5) {
-        ptrdiff_t tcp_offset;
+        ptrdiff_t ptcp_offset, stcp_offset;
 
-        tcp_offset = ppkt->transport_header - (uint8_t *)ppkt->data
-                     + (ptcp->th_off * 4) - ppkt->vnet_hdr_len;
-        res = colo_packet_compare_common(ppkt, spkt, tcp_offset);
+        ptcp_offset = ppkt->transport_header - (uint8_t *)ppkt->data
+                      + (ptcp->th_off * 4) - ppkt->vnet_hdr_len;
+        stcp_offset = spkt->transport_header - (uint8_t *)spkt->data
+                      + (stcp->th_off * 4) - spkt->vnet_hdr_len;
+
+        /*
+         * When network is busy, some tcp options(like sack) will unpredictable
+         * occur in primary side or secondary side. it will make packet size
+         * not same, but the two packet's payload is identical. colo just
+         * care about packet payload, so we skip the option field.
+         */
+        res = colo_packet_compare_common(ppkt, spkt, ptcp_offset, stcp_offset);
     } else if (ptcp->th_sum == stcp->th_sum) {
-        res = colo_packet_compare_common(ppkt, spkt, ETH_HLEN);
+        res = colo_packet_compare_common(ppkt, spkt, ETH_HLEN, ETH_HLEN);
     } else {
         res = -1;
     }
@@ -330,6 +343,7 @@ static int colo_packet_compare_udp(Packet *spkt, Packet *ppkt)
      * the ip payload here.
      */
     ret = colo_packet_compare_common(ppkt, spkt,
+                                     network_header_length + ETH_HLEN,
                                      network_header_length + ETH_HLEN);
 
     if (ret) {
@@ -367,6 +381,7 @@ static int colo_packet_compare_icmp(Packet *spkt, Packet *ppkt)
      * the ip payload here.
      */
     if (colo_packet_compare_common(ppkt, spkt,
+                                   network_header_length + ETH_HLEN,
                                    network_header_length + ETH_HLEN)) {
         trace_colo_compare_icmp_miscompare("primary pkt size",
                                            ppkt->size);
@@ -404,7 +419,7 @@ static int colo_packet_compare_other(Packet *spkt, Packet *ppkt)
                                    sec_ip_src, sec_ip_dst);
     }
 
-    return colo_packet_compare_common(ppkt, spkt, 0);
+    return colo_packet_compare_common(ppkt, spkt, 0, 0);
 }
 
 static int colo_old_packet_check_one(Packet *pkt, int64_t *check_time)
-- 
2.7.4

  parent reply	other threads:[~2017-09-08  2:06 UTC|newest]

Thread overview: 20+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-09-08  2:05 [Qemu-devel] [PULL 00/18] Net patches Jason Wang
2017-09-08  2:05 ` [Qemu-devel] [PULL 01/18] net/filter-rewriter.c: Fix rewirter checksum bug when use virtio-net Jason Wang
2017-09-08  2:05 ` [Qemu-devel] [PULL 02/18] net/rocker: Remove the dead error handling Jason Wang
2017-09-08  2:05 ` [Qemu-devel] [PULL 03/18] net/rocker: Plug memory leak in pci_rocker_init() Jason Wang
2017-09-08  2:05 ` [Qemu-devel] [PULL 04/18] net/rocker: Convert to realize() Jason Wang
2017-09-08  2:05 ` [Qemu-devel] [PULL 05/18] net/rocker: Fix the unusual macro name Jason Wang
2017-09-08  2:05 ` [Qemu-devel] [PULL 06/18] net: rtl8139: do not use old_mmio accesses Jason Wang
2017-09-08  2:05 ` [Qemu-devel] [PULL 07/18] MAINTAINERS: Update mail address for COLO Proxy Jason Wang
2017-09-08  2:05 ` [Qemu-devel] [PULL 08/18] net/socket: Don't treat odd socket type as SOCK_STREAM Jason Wang
2017-09-08  2:05 ` [Qemu-devel] [PULL 09/18] net/socket: Convert several helper functions to Error Jason Wang
2017-09-08  2:05 ` [Qemu-devel] [PULL 10/18] net/net: Convert parse_host_port() " Jason Wang
2017-09-08  2:05 ` [Qemu-devel] [PULL 11/18] net/socket: Improve -net socket error reporting Jason Wang
2017-09-08  2:05 ` [Qemu-devel] [PULL 12/18] e1000: Rename the SEC symbol to SEQEC Jason Wang
2017-09-08  2:05 ` Jason Wang [this message]
2017-09-08  2:05 ` [Qemu-devel] [PULL 14/18] net/colo-compare.c: Adjust net queue pop order for performance Jason Wang
2017-09-08  2:05 ` [Qemu-devel] [PULL 15/18] net/colo-compare.c: Fix comments and scheme Jason Wang
2017-09-08  2:05 ` [Qemu-devel] [PULL 16/18] qemu-iothread: IOThread supports the GMainContext event loop Jason Wang
2017-09-08  2:05 ` [Qemu-devel] [PULL 17/18] colo-compare: Use IOThread to Check old packet regularly and Process pactkets of the primary Jason Wang
2017-09-08  2:05 ` [Qemu-devel] [PULL 18/18] colo-compare: Update the COLO document to add the IOThread configuration Jason Wang
2017-09-08 13:00 ` [Qemu-devel] [PULL 00/18] Net patches 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=1504836324-4871-14-git-send-email-jasowang@redhat.com \
    --to=jasowang@redhat.com \
    --cc=peter.maydell@linaro.org \
    --cc=qemu-devel@nongnu.org \
    --cc=zhangchen.fnst@cn.fujitsu.com \
    /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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).