On Mon, 2019-09-23 at 07:48 +0200, Dagobert Michelsen wrote: > Hi Paul, > > Am 22.09.2019 um 23:42 schrieb Paul Smith <psm...@gnu.org>: > > On Tue, 2019-09-17 at 09:40 +0200, Dagobert Michelsen wrote: > > > Additionally, I have two failed tests on Solaris 10 borg Sparc > > > x86: > > > > The wildcard issue is due to the old glob.c implementation in GNU > > make. I don't know if I can fix that before the next release. > > It's not a regression, however: this is just a new test that we > > didn't used to run. The behavior of GNU make is the same as it was > > in previous releases. > > Maybe it should be then marked as XFAIL? Now the build process stops > after the testsuite because of the failures.
It won't fail on systems that provide GNU libc. It only fails on systems which need to use GNU make's built-in glob. Unfortunately there's no such thing in the GNU make test suite as XFAIL. I could skip these tests if we are using the built-in glob, that's all I can do. I'm not sure what you mean by "the build process stops"... there's no build process that involves running the tests as part a pipeline, at least none provided by make itself. Maybe this is an extra build process that you are using locally? _______________________________________________ Bug-make mailing list Bug-make@gnu.org https://lists.gnu.org/mailman/listinfo/bug-make
