Re: [Qemu-devel] [Bug 1179731] [NEW] is networking broken on windows hosts?

[Paolo Bonzini <pbonzini@redhat.com>]

Il 14/05/2013 13:39, TeLeMan ha scritto:
> On Tue, May 14, 2013 at 6:46 PM, Paolo Bonzini <pbonzini@redhat.com> wrote:
>> Il 14/05/2013 12:24, TeLeMan ha scritto:
>>> On Tue, May 14, 2013 at 3:51 PM, Stefan Hajnoczi <stefanha@gmail.com> wrote:
>>>> On Tue, May 14, 2013 at 12:02:24AM -0000, therock247uk wrote:
>>>>> just wondering as i just compiled the latest git and qemu goes into none
>>>>> responding mode when i try to do any networking stuff on guests (both
>>>>> linux and windows)
>>>>
>>>> Works for me on qemu.git/master on Linux:
>>>>
>>>>   $ git rev-parse HEAD
>>>>   b087143b4d010451208264b7c841436aafe1cbb1
>>>>   $ x86_64-softmmu/qemu-system-x86_64 -m 1024 -enable-kvm -cpu host \
>>>>           -drive if=virtio,cache=none,file=test.img
>>>>
>>>> Please include more information, like the QEMU command-line and commit
>>>> ID.
>>>>
>>>> Stefan
>>>>
>>>
>>> This regression occurs on the Windows host. SLIRP hangs in sorecvfrom().
>>
>> Can you bisect it?
>>
>> Paolo
>>
> The first break is the commit
> 5e3bc735d93dd23f074b5116fd11e1ad8cd4962f. But it need more packets
> than HEAD to reproduce this regression.

Please check if this partial revert of that commit fixes it:

diff --git a/main-loop.c b/main-loop.c
index f46aece..79c45b8 100644
--- a/main-loop.c
+++ b/main-loop.c
@@ -394,6 +394,20 @@ static int os_host_main_loop_wait(uint32_t timeout)
         return ret;
     }
 
+    FD_ZERO(&rfds);
+    FD_ZERO(&wfds);
+    FD_ZERO(&xfds);
+    nfds = pollfds_fill(gpollfds, &rfds, &wfds, &xfds);
+    if (nfds >= 0) {
+        select_ret = select(nfds + 1, &rfds, &wfds, &xfds, &tv0);
+        if (select_ret != 0) {
+            timeout = 0;
+        }
+        if (select_ret > 0) {
+            pollfds_poll(gpollfds, nfds, &rfds, &wfds, &xfds);
+        }
+    }
+
     g_main_context_prepare(context, &max_priority);
     n_poll_fds = g_main_context_query(context, max_priority, &poll_timeout,
                                       poll_fds, ARRAY_SIZE(poll_fds));
@@ -426,24 +440,11 @@ static int os_host_main_loop_wait(uint32_t timeout)
         g_main_context_dispatch(context);
     }
 
-    /* Call select after g_poll to avoid a useless iteration and therefore
-     * improve socket latency.
+    /* If an edge-triggered socket event occurred, select will return a
+     * positive result on the next iteration.  We do not need to do anything
+     * here.
      */
 
-    FD_ZERO(&rfds);
-    FD_ZERO(&wfds);
-    FD_ZERO(&xfds);
-    nfds = pollfds_fill(gpollfds, &rfds, &wfds, &xfds);
-    if (nfds >= 0) {
-        select_ret = select(nfds + 1, &rfds, &wfds, &xfds, &tv0);
-        if (select_ret != 0) {
-            timeout = 0;
-        }
-        if (select_ret > 0) {
-            pollfds_poll(gpollfds, nfds, &rfds, &wfds, &xfds);
-        }
-    }
-
     return select_ret || g_poll_ret;
 }
 #endif