On Thu, Aug 1, 2019 at 2:43 PM Brendan Higgins <brendanhiggins@xxxxxxxxxx> wrote: > > On Thu, Aug 1, 2019 at 2:14 PM Stephen Boyd <sboyd@xxxxxxxxxx> wrote: > > > > Quoting Brendan Higgins (2019-08-01 11:59:57) > > > On Thu, Aug 1, 2019 at 11:55 AM Brendan Higgins > > > <brendanhiggins@xxxxxxxxxx> wrote: > > > > > > > > On Fri, Jul 26, 2019 at 1:31 AM Petr Mladek <pmladek@xxxxxxxx> wrote: > > > > > > > > > To be honest I do not fully understand KUnit design. I am not > > > > > completely sure how the tested code is isolated from the running > > > > > system. Namely, I do not know if the tested code shares > > > > > the same locks with the system running the test. > > > > > > > > No worries, I don't expect printk to be the hang up in those cases. It > > > > sounds like KUnit has a long way to evolve before printk is going to > > > > be a limitation. > > > > > > So Stephen, what do you think? > > > > > > Do you want me to go forward with the new kunit_assert API wrapping > > > the string_stream as I have it now? Would you prefer to punt this to a > > > later patch? Or would you prefer something else? > > > > > > > I like the struct based approach. If anything, it can be adjusted to > > make the code throw some records into a spinlock later on and delay the > > formatting of the assertion if need be. > > That's a fair point. > > > Can you resend with that > > approach? I don't think I'll have any more comments after that. I sent a new revision, v12, that incorporates the kunit_assert stuff. Let me know what you think!
