* Andrew Dunstan (andrew@dunslane.net) wrote:
> On 09/08/2015 02:58 PM, Stephen Frost wrote:
> >Would be great to get feedback on it as I'm no Makefile expert (it took
> >me far too long to run down what was happening and work out what seemed
> >like the right fix..).
>
>
> Seems OK to me from a quick look - I guess the proof of the pudding
> is in the eating. Does every "make check" target that was working
> still work? I assume so. It's interesting that nobody has noticed
> this for about 6 weeks.
This only impacted individuals running the TAP tests on REL9_4_STABLE
using 'make check', which I'm guessing is a pretty small set (possibly
a set of 1).
make check-world works for me with this patch, I'm guessing that covers
all of the 'make check' cases which use the TAP tests, but even if it
doesn't, it's probably more than enough to be representative.
I'm doing one final run with the latest changes to REL9_4_STABLE and
once those complete, assuming all still looks good, I'll go ahead and
push this.
Thanks!
Stephen