Hi On Wed, Apr 27, 2022 at 1:41 PM Thomas Huth <th...@redhat.com> wrote: > > On 26/04/2022 11.27, marcandre.lur...@redhat.com wrote: > > From: Marc-André Lureau <marcandre.lur...@redhat.com> > > > > The call is POSIX-specific. Use the dedicated GLib API. > > g_unix_set_fd_nonblocking() is also available on Unix-like systems according > to its name, I suppose? So what's the advantage of this change? >
This is a preliminary patch before the last patch (https://patchew.org/QEMU/20220426092715.3931705-1-marcandre.lur...@redhat.com/20220426092715.3931705-27-marcandre.lur...@redhat.com/) We should reserve qemu_socket_set_*block() usage to socket-like fd. thanks > Thomas > > > > Signed-off-by: Marc-André Lureau <marcandre.lur...@redhat.com> > > --- > > tests/qtest/vhost-user-test.c | 4 +++- > > 1 file changed, 3 insertions(+), 1 deletion(-) > > > > diff --git a/tests/qtest/vhost-user-test.c b/tests/qtest/vhost-user-test.c > > index ee30f5479648..a2cec8768462 100644 > > --- a/tests/qtest/vhost-user-test.c > > +++ b/tests/qtest/vhost-user-test.c > > @@ -302,6 +302,7 @@ static int chr_can_read(void *opaque) > > > > static void chr_read(void *opaque, const uint8_t *buf, int size) > > { > > + g_autoptr(GError) err = NULL; > > TestServer *s = opaque; > > CharBackend *chr = &s->chr; > > VhostUserMsg msg; > > @@ -394,7 +395,8 @@ static void chr_read(void *opaque, const uint8_t *buf, > > int size) > > * The receive function forces it to be blocking, > > * so revert it back to non-blocking. > > */ > > - qemu_set_nonblock(fd); > > + g_unix_set_fd_nonblocking(fd, true, &err); > > + g_assert_no_error(err); > > break; > > > > case VHOST_USER_SET_LOG_BASE: >
