On Oct 15, 2015, at 1:38 PM, Richard Sandiford <richard.sandif...@arm.com> wrote: > Mike Stump <mikest...@comcast.net> writes: >> On Oct 15, 2015, at 12:47 PM, Richard Sandiford >> <richard.sandif...@arm.com> wrote: >>> I can see that argument if people are only taking work items from >>> the PR database. But it's possible (likely even) that people will >>> independently find a problem like this and just fix it, if the missed >>> optimisation happens to be important to them. I don't think they >>> should then have to trawl the PR database to see which PRs their patch >>> fixes. >> >> There is no requirement that they do. > > But if they don't the original test stays #if 0d out.
Only until someone retests the PR. > I don't see why that's better than having an XFAIL become an XPASS, so that > it's > obvious that the XFAIL can be removed and we get the test back quicker. I have a slight preference to not split the test cases in this case. As I said, it isn’t a big deal, and if you feel it is better to split the test case, then your patch is ok. I think it is trivial as I don’t know of a reason why someone should reject it.