On 12/03/12 15:27 , Mike Stump wrote:

So, I guess the question is, is there a down side to doing that?  I can imagine 
a

        if [ -r "$srcdir/contrib/testsuite-management/$target.xfail ]; then
                $srcdir/contrib/testsuite-management/validate_failures.py bla 
bla
        fi

Yeah. I had something like this in mind originally, but never followed through (we use the validator from within our own build harness).

Ideally, though, we would not even need this hack. 'make check' should return 0 when every test is nominal. Period.

Our own guidelines are the culprit here: '... , this means comparing post-patch test results to pre-patch results by testing twice or comparing with recent posts to the gcc-testresults list.' (http://gcc.gnu.org/contribute.html#testing).

I have argued before that we should change this, but I am yet to do anything concrete about it.


Diego.

Reply via email to