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: zhanghailiang <zhang.zhanghailiang@huawei.com>,
	Jason Wang <jasowang@redhat.com>
Subject: [Qemu-devel] [PULL RESEND 08/19] colo-compare: kick compare thread to exit after some cleanup in finalization
Date: Mon,  6 Mar 2017 13:25:43 +0800	[thread overview]
Message-ID: <1488777954-4578-9-git-send-email-jasowang@redhat.com> (raw)
In-Reply-To: <1488777954-4578-1-git-send-email-jasowang@redhat.com>

From: zhanghailiang <zhang.zhanghailiang@huawei.com>

We should call g_main_loop_quit() to notify colo compare thread to
exit, Or it will run in g_main_loop_run() forever.

Besides, the finalizing process can't happen in context of colo thread,
it is reasonable to remove the 'if (qemu_thread_is_self(&s->thread))'
branch.

Before compare thead exits, some cleanup works need to be
done,  All unhandled packets need to be released and connection_track_table
needs to be freed, or there will be memory leak.

Signed-off-by: zhanghailiang <zhang.zhanghailiang@huawei.com>
Reviewed-by: Zhang Chen <zhangchen.fnst@cn.fujitsu.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
---
 net/colo-compare.c | 39 +++++++++++++++++++++++++++++----------
 1 file changed, 29 insertions(+), 10 deletions(-)

diff --git a/net/colo-compare.c b/net/colo-compare.c
index fdde788..37ce75c 100644
--- a/net/colo-compare.c
+++ b/net/colo-compare.c
@@ -83,6 +83,8 @@ typedef struct CompareState {
     GHashTable *connection_track_table;
     /* compare thread, a thread for each NIC */
     QemuThread thread;
+
+    GMainLoop *compare_loop;
 } CompareState;
 
 typedef struct CompareClass {
@@ -496,7 +498,6 @@ static gboolean check_old_packet_regular(void *opaque)
 static void *colo_compare_thread(void *opaque)
 {
     GMainContext *worker_context;
-    GMainLoop *compare_loop;
     CompareState *s = opaque;
     GSource *timeout_source;
 
@@ -507,7 +508,7 @@ static void *colo_compare_thread(void *opaque)
     qemu_chr_fe_set_handlers(&s->chr_sec_in, compare_chr_can_read,
                              compare_sec_chr_in, NULL, s, worker_context, true);
 
-    compare_loop = g_main_loop_new(worker_context, FALSE);
+    s->compare_loop = g_main_loop_new(worker_context, FALSE);
 
     /* To kick any packets that the secondary doesn't match */
     timeout_source = g_timeout_source_new(REGULAR_PACKET_CHECK_MS);
@@ -515,10 +516,10 @@ static void *colo_compare_thread(void *opaque)
                           (GSourceFunc)check_old_packet_regular, s, NULL);
     g_source_attach(timeout_source, worker_context);
 
-    g_main_loop_run(compare_loop);
+    g_main_loop_run(s->compare_loop);
 
     g_source_unref(timeout_source);
-    g_main_loop_unref(compare_loop);
+    g_main_loop_unref(s->compare_loop);
     g_main_context_unref(worker_context);
     return NULL;
 }
@@ -675,6 +676,23 @@ static void colo_compare_complete(UserCreatable *uc, Error **errp)
     return;
 }
 
+static void colo_flush_packets(void *opaque, void *user_data)
+{
+    CompareState *s = user_data;
+    Connection *conn = opaque;
+    Packet *pkt = NULL;
+
+    while (!g_queue_is_empty(&conn->primary_list)) {
+        pkt = g_queue_pop_head(&conn->primary_list);
+        compare_chr_send(&s->chr_out, pkt->data, pkt->size);
+        packet_destroy(pkt, NULL);
+    }
+    while (!g_queue_is_empty(&conn->secondary_list)) {
+        pkt = g_queue_pop_head(&conn->secondary_list);
+        packet_destroy(pkt, NULL);
+    }
+}
+
 static void colo_compare_class_init(ObjectClass *oc, void *data)
 {
     UserCreatableClass *ucc = USER_CREATABLE_CLASS(oc);
@@ -703,14 +721,15 @@ static void colo_compare_finalize(Object *obj)
     qemu_chr_fe_deinit(&s->chr_sec_in);
     qemu_chr_fe_deinit(&s->chr_out);
 
-    g_queue_free(&s->conn_list);
+    g_main_loop_quit(s->compare_loop);
+    qemu_thread_join(&s->thread);
 
-    if (qemu_thread_is_self(&s->thread)) {
-        /* compare connection */
-        g_queue_foreach(&s->conn_list, colo_compare_connection, s);
-        qemu_thread_join(&s->thread);
-    }
+    /* Release all unhandled packets after compare thead exited */
+    g_queue_foreach(&s->conn_list, colo_flush_packets, s);
+
+    g_queue_free(&s->conn_list);
 
+    g_hash_table_destroy(s->connection_track_table);
     g_free(s->pri_indev);
     g_free(s->sec_indev);
     g_free(s->outdev);
-- 
2.7.4

  parent reply	other threads:[~2017-03-06  5:26 UTC|newest]

Thread overview: 31+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-03-06  5:25 [Qemu-devel] [PULL RESEND 00/19] Net patches Jason Wang
2017-03-06  5:25 ` [Qemu-devel] [PULL RESEND 01/19] net: Remove useless local var pkt Jason Wang
2017-03-06  5:25 ` [Qemu-devel] [PULL RESEND 02/19] eth: Extend vlan stripping functions Jason Wang
2017-03-06  5:25 ` [Qemu-devel] [PULL RESEND 03/19] NetRxPkt: Fix memory corruption on VLAN header stripping Jason Wang
2017-03-06  5:25 ` [Qemu-devel] [PULL RESEND 04/19] NetRxPkt: Do not try to pull more data than present Jason Wang
2017-03-06  5:25 ` [Qemu-devel] [PULL RESEND 05/19] NetRxPkt: Account buffer with ETH header in IOV length Jason Wang
2017-03-06  5:25 ` [Qemu-devel] [PULL RESEND 06/19] NetRxPkt: Remove code duplication in net_rx_pkt_pull_data() Jason Wang
2017-03-06  5:25 ` [Qemu-devel] [PULL RESEND 07/19] colo-compare: use g_timeout_source_new() to process the stale packets Jason Wang
2017-03-06  5:25 ` Jason Wang [this message]
2017-03-06  5:25 ` [Qemu-devel] [PULL RESEND 09/19] char: remove the right fd been watched in qemu_chr_fe_set_handlers() Jason Wang
2017-03-06  5:25 ` [Qemu-devel] [PULL RESEND 10/19] colo-compare: Fix removing fds been watched incorrectly in finalization Jason Wang
2017-03-06  5:25 ` [Qemu-devel] [PULL RESEND 11/19] net/colo-compare: Fix memory free error Jason Wang
2017-03-06  5:25 ` [Qemu-devel] [PULL RESEND 12/19] vmxnet3: Convert ring values to uint32_t's Jason Wang
2017-03-13 20:20   ` Laurent Vivier
2017-03-14 11:16     ` Dr. David Alan Gilbert
2017-03-14 11:29       ` Laurent Vivier
2017-03-14 11:38         ` Dr. David Alan Gilbert
2017-03-14 11:47           ` Laurent Vivier
2017-03-14 12:32             ` Dr. David Alan Gilbert
2017-03-14 12:42               ` Laurent Vivier
2017-03-14 12:44             ` Dr. David Alan Gilbert
2017-03-14 13:31             ` Dr. David Alan Gilbert
2017-03-14 13:42               ` Laurent Vivier
2017-03-06  5:25 ` [Qemu-devel] [PULL RESEND 13/19] vmxnet3: VMStatify rx/tx q_descr and int_state Jason Wang
2017-03-06  5:25 ` [Qemu-devel] [PULL RESEND 14/19] net/colo: fix memory double free error Jason Wang
2017-03-06  5:25 ` [Qemu-devel] [PULL RESEND 15/19] filter-rewriter: skip net_checksum_calculate() while offset = 0 Jason Wang
2017-03-06  5:25 ` [Qemu-devel] [PULL RESEND 16/19] COLO-compare: Rename compare function and remove duplicate codes Jason Wang
2017-03-06  5:25 ` [Qemu-devel] [PULL RESEND 17/19] COLO-compare: Optimize compare_common and compare_tcp Jason Wang
2017-03-06  5:25 ` [Qemu-devel] [PULL RESEND 18/19] COLO-compare: Fix icmp and udp compare different packet always dump bug Jason Wang
2017-03-06  5:25 ` [Qemu-devel] [PULL RESEND 19/19] net/filter-mirror: Follow CODING_STYLE Jason Wang
2017-03-07  7:31 ` [Qemu-devel] [PULL RESEND 00/19] 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=1488777954-4578-9-git-send-email-jasowang@redhat.com \
    --to=jasowang@redhat.com \
    --cc=peter.maydell@linaro.org \
    --cc=qemu-devel@nongnu.org \
    --cc=zhang.zhanghailiang@huawei.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).