On May 15, 2013, at 3:22 AM, Shakthi Kannan <skan...@redhat.com> wrote: > ----- Original Message ----- > | From: "Tom Tromey" <tro...@redhat.com> > | > | Does this cause test suite regressions? > \-- > > I built gcc-4.8.0 with and without the patch, and ran the test suite for both > instances using: > > $ make -k check > > There was no difference in the test result output.
The output of make -k, or the contents of all the .sum and .log files, or… For example, if you compared all the .sum files, and there were no differences (literally), then you did something wrong, at least the dates must differ. > Is this the same as the test suite regression? I like using ~/contrib/compare_tests gcc-before.sum gcc-after.sum to determine if there are regressions. You can also use that script to check for regressions between two build trees as well. The output is priority sorted with the important stuff first, and lessor things later. regression is defined as that script returning 1, and no regression is if the script returns 0. Anyway, life is more complicated than this, but, it is a good first order approximation.
