On Sun, Jul 17, 2011 at 11:47:52AM -0700, Michael G Schwern wrote: > Moving out of the realm of what you can do today, it may be worth adding a key > to the meta file which indicates the tests are not safe for parallel testing. > > I prefer this over the inverse (a flag stating that the tests ARE safe for > parallel testing) because moving forward it would be most efficient for large > scale testing if we encourage authors to make their tests safe for parallel > processing.
I agree that this is an important area to look at. My initial thoughts were also that some indication in the meta file would be useful. I'm not sure it really matters which way around the flag is - ideally it would be neutral. The important thing is what to do when the flag is not present, which would be the normal case initially. For safety, in the absence of any other information, the installer would need to assume that the tests were not safe for parallel testing. If they wanted to, some smokers could default to testing in parallel and then send feedback to authors - and perhaps even a patch to the meta file on success. But care needs to be taken here. One parallel success doesn't mean the testsuite is parallel safe in general. -- Paul Johnson - p...@pjcj.net http://www.pjcj.net