On Tue, Mar 20, 2012 at 7:17 PM, R. Diez <[email protected]> wrote:
> Hi all:
>
> The attached patch makes targets rtl-tests and rtl-test-with-check in
> ORPSoCV2 print a clear error message and stop if a test case fails to
> compile or if the test results are not the ones expected.
>
> Without this patch, the makefile just carries on, and you have to visually
> inspect the log file to make sure all tests were successful. That's not
> ideal, especially in a daily build scenario.

Hi Ruben,

I think a better way of doing this is to try to run every test
irrespective of if the last one failed, and collate the failed tests
to report at the end. This helps when developing fixes for something,
which may fix what you're focusing on but then cause failures
elsewhere.

Maybe we ought to write out a little regression run report which
indicates the tests which were run and those which failed?

Julius
_______________________________________________
OpenRISC mailing list
[email protected]
http://lists.openrisc.net/listinfo/openrisc

Reply via email to