Hello, I installed today the testing version of portage 2.6.1.8. It seems that this version ignores the -stricter feature I have set in make.conf and thus I can not emerge packages that fail QA checks. Unfortunately I can confirm it only for media-tv/xbmc, a hard masked ebuild, as I don't know any other package that fails QA checks in order to try it too. With portage 2.6.1.7 I get the expected behavior, the package gets installed despite failing the stricter tests. Reproducible: Always Steps to Reproduce: 1. emerge portage-2.6.1.8 2. Set feature -stricter at make.conf 3. emerge an ebuild that fails stricter tests (I tried only media-tv/xbmc) Actual Results: * Function `completion_matches' implicitly converted to pointer at /var/tmp/portage/media-tv/xbmc-9999/work/xbmc-9999/xbmc/lib/libPython/Python/Modules/readline.c:683 * * Please file a bug about this at http://bugs.gentoo.org/ * with the maintaining herd of the package. * * * ERROR: media-tv/xbmc-9999 failed. * Call stack: * misc-functions.sh, line 729: Called install_qa_check * misc-functions.sh, line 421: Called die * The specific snippet of code: * die "poor code kills airplanes" * The die message: * poor code kills airplanes * * If you need support, post the topmost build error, and the call stack if relevant. * A complete build log is located at '/var/tmp/portage/media-tv/xbmc-9999/temp/build.log'. * The ebuild environment file is located at '/var/tmp/portage/media-tv/xbmc-9999/temp/environment'. * Expected Results: Install the package despite the failed tests.
It is intended behavior: http://sources.gentoo.org/viewcvs.py/portage?rev=12793&view=rev These warnings indicate very probable segmentation faults on 64-bit architectures, so you should report bugs for all affected packages.
Thanks, I hadn't notice this change. Sorry for the useless bug report.
