gdb/testsuite: fix default behavior of runto
The documented behavior of proc runto is to not emit a PASS when succeeding to to run to the specified location, but emit a FAIL when failing to do so. I suppose the intent is that it won't pollute the results normally passing tests (although I don't see why we would care), but make visible any problems. However, it seems like the implementation makes it default to never print anything. "no-message" is prependend to "args", so if "message" is not passed, we will always take the path that sets print_fail to 0, which will silence any failure. This unfortunately means that tests relying on runto_main won't emit a FAIL if failing to run to main. And since commit 4dfef5be ("gdb/testsuite: make runto_main not pass no-message to runto"), tests don't emit a FAIL themselves when failing to run to main. This means that tests failing to run to main just fail silently, and that's bad. This can be reproduced by hacking gdb.base/template.exp like so: diff --git a/gdb/testsuite/gdb.base/template.c b/gdb/testsuite/gdb.base/template.c index bcf39c377d92..052be5b79d73 100644 --- a/gdb/testsuite/gdb.base/template.c +++ b/gdb/testsuite/gdb.base/template.c @@ -15,6 +15,14 @@ You should have received a copy of the GNU General Public License along with this program. If not, see <http://www.gnu.org/licenses/>. */ +#include <stdlib.h> + +__attribute__((constructor)) +static void c (void) +{ + exit (1); +} + int main (void) { Running the modified gdb.base/template.exp shows that it exits without printing any result. Remove the line that prepends no-message to args, that should make runto's behavior match its documentation. This patch will appear to add many failures, but in reality they already existed, they were just silenced. Change-Id: I2a730d5bc72b6ef0698cd6aad962d9902aa7c3d6
Loading
Please register or sign in to comment