PASS: status-request-revoked PASS: set_x509_ocsp_multi_cli FAIL: dtls-with-seccomp PASS: tls-with-seccomp FAIL: dtls-client-with-seccomp PASS: tls-client-with-seccomp ------------------------------------------------------------------- This is an unstable amd64 chroot image at a tinderbox (==build bot) name: 17.1_desktop_plasma-test-20200228-231731 ------------------------------------------------------------------- gcc-config -l: [1] x86_64-pc-linux-gnu-9.2.0 * llvm: 9.0.1 Available Python interpreters, in order of preference: [1] python3.6 [2] python3.8 (fallback) [3] python3.7 (fallback) [4] python2.7 (fallback) repository: ==> /var/db/repos/gentoo/metadata/timestamp.chk <== Sat, 29 Feb 2020 11:07:24 +0000 emerge -qpvO net-libs/gnutls [ebuild U ] net-libs/gnutls-3.6.12 [3.6.7-r1] USE="cxx idn nls openssl seccomp test* tls-heartbeat -dane -doc -examples -guile -pkcs11 -sslv2 -sslv3 -static-libs -test-full -tools -valgrind" ABI_X86="(64) -32 (-x32)"
Created attachment 616632 [details] emerge-info.txt
Created attachment 616634 [details] emerge-history.txt
Created attachment 616636 [details] environment
Created attachment 616638 [details] etc.portage.tbz2
Created attachment 616640 [details] logs.tbz2
Created attachment 616642 [details] net-libs:gnutls-3.6.12:20200229-170828.log.bz2
Created attachment 616644 [details] temp.tbz2
This is similar to bug 649396, but for me all the 4 seccomp ones fail in an arm64 systemd-nspawn; here 2 pass, suggesting a completely different issue
Created attachment 638482 [details] strace of dtls-client-with-seccomp
Created attachment 638484 [details] strace of dtls-with-seccomp
Same issue here, I setup an amd64 chroot on an amd64 host for helping out testing failures with clang. I have attached the straces of the failing tests (complete of follow-forks, verbose, relative timestamps). It's a chroot that was setup 5 days ago from a stage3, and since then I recompiled world with GCC, then switched on tests. No configuration changes were made to the chroot except for /etc/portage/* and /etc/locale.gen, and I am not aware that I ever touched anything seccomp-related on the host.
The bug has been referenced in the following commit(s): https://gitweb.gentoo.org/repo/gentoo.git/commit/?id=807088725f04adee3a1e0ed9a8b41d8d647262b3 commit 807088725f04adee3a1e0ed9a8b41d8d647262b3 Author: Thomas Deutschmann <whissi@gentoo.org> AuthorDate: 2020-09-04 15:28:14 +0000 Commit: Thomas Deutschmann <whissi@gentoo.org> CommitDate: 2020-09-04 16:13:47 +0000 net-libs/gnutls: bump to v3.6.15 Bug: https://bugs.gentoo.org/649396 Bug: https://bugs.gentoo.org/711104 Bug: https://bugs.gentoo.org/740390 Package-Manager: Portage-3.0.4, Repoman-3.0.1 Signed-off-by: Thomas Deutschmann <whissi@gentoo.org> net-libs/gnutls/Manifest | 1 + .../gnutls-3.6.15-skip-dtls-seccomp-tests.patch | 26 ++++ net-libs/gnutls/gnutls-3.6.15.ebuild | 134 +++++++++++++++++++++ 3 files changed, 161 insertions(+)
Update summary. [something] is used when something refers to an overlay
Created attachment 767539 [details] build.log.xz (ppc, gnutls-3.7.3-r1) These -with-seccomp tests still fail on ppc.
Created attachment 767540 [details] test-suite.log (ppc, gnutls-3.7.3-r1)
tinderbox_musl has reproduced this issue with version 3.7.7 - Updating summary.
