Created attachment 574610 [details] /var/log/portage/app-misc:beep-1.4.4:20190429-123938.log This package suffers from multiple issues that eventually cause it to fail to build. Namely: 1. It doesn't respect CC, and calls both 'gcc' AND 'clang' directly (no clue why it uses two compilers in the first place). 2. It forces -Werror which makes it fail due to invalid args to clang. 3. It calls 'git' while not in a git repo (that's minor). 4. The license should be GPL-2+. make: *** [GNUmakefile:198: beep-library.clang-o] Error 1 clang-9: error: argument unused during compilation: '--param l1-cache-size=64' [-Werror,-Wunused-command-line-argument] clang-9: error: argument unused during compilation: '--param l1-cache-line-size=64' [-Werror,-Wunused-command-line-argument] clang-9: error: argument unused during compilation: '--param l2-cache-size=512' [-Werror,-Wunused-command-line-argument] distcc[791] ERROR: compile beep-usage.c on localhost failed make: *** [GNUmakefile:198: beep-usage.clang-o] Error 1 rm beep.1 * ERROR: app-misc/beep-1.4.4::gentoo failed (compile phase): * emake failed
Created attachment 574614 [details] build.log What's exactly the point of this bug report? 1) The package compiles just fine in my sandbox. 2) I filed a PR to get the package tested by the Sec team. Whoever merged it should have picked up the issue first. 3) If the mere idea of having a package that apparently jeopardises the integrity of the Gentoo repository is so horrible to you, you can still revert the commit and/or mask the package. Then reopen the PR to let me know about the issue. Instead, you've decided to take the most cumbersome route and to file a bug whilst CCing QA in. Great use of everyone's time. Your witch hunt against me is becoming tedious, to say the least.
The purpose of this bug report is to get the package fixed *because it does not build on my system*, as presented in the attached bug report. Please kindly be mature enough to take responsibility for your mistakes rather than blaming 'Security team' and accusing me of witch-hunting.
you at least introduced a regression by having clang/gcc called directly, so this report is valid whoever reports this. (I do not have clang on my systems to state if this really fails building with it or not, though).
Correct me if I'm wrong but didn't I get a QA ban for not testing PRs before merging them? And now you're accusing me of "taking responsibility of my mistakes"? It's either one or the other but it can't be both at the time. Let's get this right for lurkers who happen to read this bug report: I didn't merge the PR hence didn't introduce the bug in the first place. This double standard is absolutely shocking for a so-called QA team.
(In reply to Patrice Clement from comment #4) > Correct me if I'm wrong but didn't I get a QA ban for not testing PRs before > merging them? And now you're accusing me of "taking responsibility of my > mistakes"? It's either one or the other but it can't be both at the time. > Let's get this right for lurkers who happen to read this bug report: I > didn't merge the PR hence didn't introduce the bug in the first place. > > This double standard is absolutely shocking for a so-called QA team. Your provocative behaviour is inapropriate. This is a bug about certian problem that has been found and should be fixed, not the ban discussion topic. If you try to provoke the QA team by only acusing the person who merged it, then this is wrong way. You still poses the developer status despite were imposed by the disciplinary action, that said it does not mean you can commit low quality code just because you do not temporary have access to the portage tree.
The bug has been referenced in the following commit(s): https://gitweb.gentoo.org/repo/gentoo.git/commit/?id=6569e6455dae2d9786dbb473550396486f83b5dc commit 6569e6455dae2d9786dbb473550396486f83b5dc Author: Aaron Bauman <bman@gentoo.org> AuthorDate: 2019-04-29 21:02:16 +0000 Commit: Aaron Bauman <bman@gentoo.org> CommitDate: 2019-04-29 21:02:16 +0000 Revert "app-misc/beep: version bump." This reverts commit 370f5643e13ef95e78e692752626e5c0391b10ef. * PR was merged from a fellow developer who I was not aware of being under a QA ban. * PR addressed an outstanding security issue with app-misc/beep hence the merge. * Reverted per the QA bug being opened. Bug: https://bugs.gentoo.org/684728 Bug: https://bugs.gentoo.org/659338 Signed-off-by: Aaron Bauman <bman@gentoo.org> app-misc/beep/Manifest | 1 - app-misc/beep/beep-1.4.4.ebuild | 38 -------------------------------------- 2 files changed, 39 deletions(-)
Thanks Aaron. @QA feel free come up with a fix for bug 659338. This bug report is no longer relevant since beep-1.4.4.ebuild has been discarded. Closing.
*** Bug 684780 has been marked as a duplicate of this bug. ***
For the record, ban doesn't mean you can't merge his stuff. It is exactly how he should contribute while under the ban. However, this doesn't mean he doesn't get to take responsibility for the problems or avoid having to submit fixes.
