On Fri, 7 Nov 2014 22:08:43 +0200 Giulio Camuffo <[email protected]> wrote:
> Looks good, they fail as expected. > > Reviewed-by: Giulio Camuffo <[email protected]> > > 2014-09-24 15:37 GMT+03:00 Marek Chalupa <[email protected]>: > > sanity test if timeouts work. > > > > Signed-off-by: Marek Chalupa <[email protected]> > > --- > > tests/sanity-test.c | 25 +++++++++++++++++++++++++ > > 1 file changed, 25 insertions(+) > > > > diff --git a/tests/sanity-test.c b/tests/sanity-test.c > > index 46f4f85..5a03e7c 100644 > > --- a/tests/sanity-test.c > > +++ b/tests/sanity-test.c > > @@ -29,6 +29,9 @@ > > #include "test-runner.h" > > #include "wayland-util.h" > > > > +#define WL_HIDE_DEPRECATED > > +#include "test-compositor.h" > > + > > extern int leak_check_enabled; > > > > TEST(empty) > > @@ -125,3 +128,25 @@ TEST(sanity_fd_exec) > > > > exec_fd_leak_check(nr_fds + 2); > > } > > + > > +FAIL_TEST(timeout_tst) > > +{ > > + test_set_timeout(1); > > + sleep(2); > > +} > > + > > +static void > > +client_timeout(void) > > +{ > > + test_set_timeout(1); > > + sleep(2); > > +} > > + > > +/* test timeout with test-compositor */ > > +FAIL_TEST(tc_timeout_tst) > > +{ > > + struct display *d = display_create(); > > + client_create(d, client_timeout); > > + display_run(d); > > + display_destroy(d); > > +} Unfortunately, this is not guaranteed. 'man alarm' says: "sleep(3) may be implemented using SIGALRM; mixing calls to alarm() and sleep(3) is a bad idea." I suppose we should use nanosleep()? Like we replaced usleep() with nanosleep() too. Thanks, pq _______________________________________________ wayland-devel mailing list [email protected] http://lists.freedesktop.org/mailman/listinfo/wayland-devel
