qemu-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Qemu-devel] [PATCH v3] vhost-user: disable the *broken* subprocess


From: Marc-André Lureau
Subject: Re: [Qemu-devel] [PATCH v3] vhost-user: disable the *broken* subprocess tests
Date: Tue, 05 Sep 2017 22:00:46 +0000

Hi

On Tue, Sep 5, 2017 at 9:41 PM Philippe Mathieu-Daudé <address@hidden>
wrote:

> not vhost-user-test, but this still failed:
>
> ERROR:tests/rcutorture.c:388:gtest_stress: assertion failed (n_mberror
> == 0): (1 == 0)
> GTester: last random seed: R02S8905ef6a0d8ed82670297d559c40e919
> make: *** [check-tests/rcutorture] Error 1
>
> CONFIG="--enable-debug --enable-debug-tcg --enable-trace-backends=log"
>
> https://travis-ci.org/philmd/qemu/jobs/272168998
>
> On 09/05/2017 03:06 PM, Philippe Mathieu-Daudé wrote:
> > tests/vhost-user-test keeps failing on build-system since Aug 15:
> >
> >    ERROR:tests/vhost-user-test.c:835:test_flags_mismatch: child process
> (/i386/vhost-user/flags-mismatch/subprocess [4836]) failed unexpectedly
> > ...
> >    ERROR:tests/vhost-user-test.c:807:test_connect_fail: child process
> (/x86_64/vhost-user/connect-fail/subprocess [58910]) failed unexpectedly
> >
> > Suggested-by: Peter Maydell <address@hidden>
> > Suggested-by: Daniel P. Berrange <address@hidden>
> > Signed-off-by: Philippe Mathieu-Daudé <address@hidden>
> > ---
> > v3: only disable tests using glib subprocess,
> >      migrate/multiqueue/guest-mem are still tested.
> >
> > v2: remove unuseful warning
> >
> >   tests/vhost-user-test.c | 17 ++++++++++-------
> >   1 file changed, 10 insertions(+), 7 deletions(-)
> >
> > diff --git a/tests/vhost-user-test.c b/tests/vhost-user-test.c
> > index d4da09f147..4b98018478 100644
> > --- a/tests/vhost-user-test.c
> > +++ b/tests/vhost-user-test.c
> > @@ -31,6 +31,8 @@
> >   #include <linux/virtio_net.h>
> >   #include <sys/vfs.h>
> >
> > +#define VHOST_USER_NET_TESTS_WORKING 0 /* broken as of 2.10.0 */
>

Not so, at least I can't reproduce locally yet.

If it's only reproducible on travis for now, shouldn't the failing tests be
disabled if getenv("TRAVIS") instead?

> +
> >   /* GLIB version compatibility flags */
> >   #if !GLIB_CHECK_VERSION(2, 26, 0)
> >   #define G_TIME_SPAN_SECOND              (G_GINT64_CONSTANT(1000000))
> > @@ -472,11 +474,6 @@ static void test_server_listen(TestServer *server)
> >       test_server_create_chr(server, ",server,nowait");
> >   }
> >
> > -static inline void test_server_connect(TestServer *server)
> > -{
> > -    test_server_create_chr(server, ",reconnect=1");
> > -}
> > -
> >   #define GET_QEMU_CMD(s)                                         \
> >       g_strdup_printf(QEMU_CMD, 512, 512, (root), (s)->chr_name,  \
> >                       (s)->socket_path, "", (s)->chr_name)
> > @@ -722,7 +719,12 @@ static void wait_for_rings_started(TestServer *s,
> size_t count)
> >       g_mutex_unlock(&s->data_mutex);
> >   }
> >
> > -#ifdef CONFIG_HAS_GLIB_SUBPROCESS_TESTS
> > +#if VHOST_USER_NET_TESTS_WORKING &&
> defined(CONFIG_HAS_GLIB_SUBPROCESS_TESTS)
> > +static inline void test_server_connect(TestServer *server)
> > +{
> > +    test_server_create_chr(server, ",reconnect=1");
> > +}
> > +
> >   static gboolean
> >   reconnect_cb(gpointer user_data)
> >   {
> > @@ -962,7 +964,8 @@ int main(int argc, char **argv)
> >       qtest_add_data_func("/vhost-user/read-guest-mem", server,
> read_guest_mem);
> >       qtest_add_func("/vhost-user/migrate", test_migrate);
> >       qtest_add_func("/vhost-user/multiqueue", test_multiqueue);
> > -#ifdef CONFIG_HAS_GLIB_SUBPROCESS_TESTS
> > +
> > +#if VHOST_USER_NET_TESTS_WORKING &&
> defined(CONFIG_HAS_GLIB_SUBPROCESS_TESTS)
> >       qtest_add_func("/vhost-user/reconnect/subprocess",
> >                      test_reconnect_subprocess);
> >       qtest_add_func("/vhost-user/reconnect", test_reconnect);
> >
>
> --
Marc-André Lureau


reply via email to

[Prev in Thread] Current Thread [Next in Thread]