Thanks for the patch, but would it not make more sense not to build the
test at all (by modifying src/Makefile.am), rather than having a test
that always succeeds?
E g, like srbchannel-test is only builds if HAVE_SYS_EVENTFD_H.
On 2015-12-11 05:49, Kamil Rytarowski wrote:
> Original patch from pkgsrc by Ryo Ondera <ryoon at NetBSD.org>
> ---
> src/tests/rtpoll-test.c | 6 ++++++
> 1 file changed, 6 insertions(+)
>
> diff --git a/src/tests/rtpoll-test.c b/src/tests/rtpoll-test.c
> index 7db7177..d0398a0 100644
> --- a/src/tests/rtpoll-test.c
> +++ b/src/tests/rtpoll-test.c
> @@ -83,6 +83,7 @@ START_TEST (rtpoll_test) {
> END_TEST
>
> int main(int argc, char *argv[]) {
> +#ifdef SIGRTMIN
> int failed = 0;
> Suite *s;
> TCase *tc;
> @@ -103,4 +104,9 @@ int main(int argc, char *argv[]) {
> srunner_free(sr);
>
> return (failed == 0) ? EXIT_SUCCESS : EXIT_FAILURE;
> +#else
> + printf("Real-time signals are not supported on this platform\n");
> +
> + return EXIT_SUCCESS;
> +#endif
> }
>
--
David Henningsson, Canonical Ltd.
https://launchpad.net/~diwic
