On Tue, Mar 13, 2012 at 01:30:29PM -0700, Mike Stump wrote:
>On Mar 13, 2012, at 9:38 AM, Bernhard Reutner-Fischer wrote:
>> Could some of the testsuite maintainers please eyeball?
>
>I've eyed it, the only thing that stood out was:
>
>-foreach testcase [lsort [glob -nocomplain $srcdir/$subdir/*.F]] {
>- if ![runtest_file_p $runtests $testcase] then {
>- continue
>- }
>- fortran-torture $testcase
>-}
>-
>-foreach testcase [lsort [glob -nocomplain $srcdir/$subdir/*.f90]] {
>- if ![runtest_file_p $runtests $testcase] then {
>- continue
>- }
>- fortran-torture $testcase
>-}
>
>which, I hope does what you want.
It replaces a manually unrolled loop with a loop over the
testcase-extensions and works as expected. Or did you mean something
else?
>
>Assuming you like those parts, Ok.
committed as r185430.
>
>Since it touches non-fortran, please watch for and respond to any problems it
>might cause.
Of course, as always.
PS: I saw that the cleanup-tree-dump is also a bit redundant.
When looking at e.g. gcc/testsuite/gcc.dg/pr52578.c one would think that
scan-tree-dump-times might be able to automagically collect which
dumpfiles to cleanup as last step in dg-final.
One could go one step further and even implicitly add the corresponding
dg-options according to the expected scan-tree-dump files although this
might not be worth it iff there are passes that may produce dump-files
different to their pass name (did not look if this is possible). One
would have to strip the pass-number off the dump-file for passes that
are run more than once, too. Just a thought..
Thanks and cheers,
