On Fri, 14 Feb 2025 at 17:53, Tamir Duberstein <tamird@xxxxxxxxx> wrote: > > On Fri, Feb 14, 2025 at 11:02 AM Andy Shevchenko > <andriy.shevchenko@xxxxxxxxxxxxxxx> wrote: > > > > On Fri, Feb 14, 2025 at 04:35:12PM +0100, Petr Mladek wrote: > > > I have just quickly tested this before leaving for a week. > > > And I am fine with the result. > > Thanks, Petr, for demonstrating how it looks in a failure case. > > Seems reasonable to me. But I want a consensus with Rasmus. > > I have a local v4 where I've added the same enhancement as the scanf > patches so that assertions log the line in the top-level test. > > I'll wait for Rasmus' reply before sending. I think all my concerns are addressed, with the lines printed in case of error telling what is wrong and not that memcmp() evaluating to 1 instead of 0, and with the final free-form comment including that "ran 448 tests". If you feel that word is confusing when there's "obviously" only 28 "test" being done, feel free to change that to "did 448 checks" or "did 448 individual checks" any other better wording. Rasmus
