| Summary: | dev-util/rr-4.5.0 : fails tests, 64bit_child (Failed) and more | ||
|---|---|---|---|
| Product: | Gentoo Linux | Reporter: | Toralf Förster <toralf> |
| Component: | Current packages | Assignee: | Gentoo Toolchain Maintainers <toolchain> |
| Status: | RESOLVED FIXED | ||
| Severity: | normal | CC: | lu_zero, mudler, slyfox |
| Priority: | Normal | Keywords: | TESTFAILURE |
| Version: | unspecified | ||
| Hardware: | All | ||
| OS: | Linux | ||
| Whiteboard: | |||
| Package list: | Runtime testing required: | --- | |
| Attachments: |
emerge-info.txt
dev-util:rr-4.5.0:20170929-085219.log.bz2 emerge-history.txt environment etc.portage.tbz2 LastTest.log.bz2 logs.tbz2 tests.tbz2 |
||
Created attachment 497058 [details]
emerge-info.txt
Created attachment 497060 [details]
dev-util:rr-4.5.0:20170929-085219.log.bz2
Created attachment 497062 [details]
emerge-history.txt
Created attachment 497064 [details]
environment
Created attachment 497066 [details]
etc.portage.tbz2
Created attachment 497068 [details]
LastTest.log.bz2
Created attachment 497070 [details]
logs.tbz2
Created attachment 497072 [details]
tests.tbz2
Does it fail on modern rr? (In reply to Sergei Trofimovich from comment #9) > Does it fail on modern rr? no idea, bug 699886 prevent a current re-test ;) (In reply to Toralf Förster from comment #10) > (In reply to Sergei Trofimovich from comment #9) > > Does it fail on modern rr? > > no idea, bug 699886 prevent a current re-test ;) Gah! Thank you! I'll have a look. Tried tests locally and it's quite a disaster :) """ 2% tests passed, 2144 tests failed out of 2193 Total Test time (real) = 2698.98 sec """ Looking at the failure: """ [FATAL /tmp/portage/dev-util/rr-5.2.0-r1/work/rr-5.2.0/src/Task.cc:2239:write_bytes_helper() errno: EIO] (task 967940 (rec:967602) at time 18) -> Assertion `nwritten == buf_size' failed to hold. Should have written 144 bytes to 0x7fff573a98a0, but only wrote -1 warning: No executable has been specified and target does not support determining executable automatically. Try using the "file" command. [FATAL /tmp/portage/dev-util/rr-5.2.0-r1/work/rr-5.2.0/src/log.cc:371:emergency_debug() errno: EIO] Can't resume execution from invalid state """ write_bytes_helper() does memory write over ptrace interface. Unfortunately sandbox already runs everything under ptrace. Let's skip test when attempting those under sandbox. The bug has been referenced in the following commit(s): https://gitweb.gentoo.org/repo/gentoo.git/commit/?id=899e40469ea6be6d4b4202a6771ecdffd9264e5c commit 899e40469ea6be6d4b4202a6771ecdffd9264e5c Author: Sergei Trofimovich <slyfox@gentoo.org> AuthorDate: 2019-11-12 22:43:56 +0000 Commit: Sergei Trofimovich <slyfox@gentoo.org> CommitDate: 2019-11-12 22:45:09 +0000 dev-util/rr: run tests only under FEATURE=-usersandbox rr extensively uses ptrace() to set memory up for replays. sandbox is also using ptrace() extensively. As a result 'rr' can't use ptrace() at all: ``` [FATAL rr-5.2.0-r1/work/rr-5.2.0/src/Task.cc:2239:write_bytes_helper() errno: EIO] (task 967940 (rec:967602) at time 18) -> Assertion `nwritten == buf_size' failed to hold. Should have written 144 bytes to 0x7fff573a98a0, but only wrote -1 ``` Similar to strace test suite allow running tests only under FEATURES=-usersandbox. This makes most tests to start passing for me. Reported-by: Toralf Förster Bug: https://bugs.gentoo.org/632394 Package-Manager: Portage-2.3.79, Repoman-2.3.18 Signed-off-by: Sergei Trofimovich <slyfox@gentoo.org> dev-util/rr/rr-5.2.0-r1.ebuild | 9 +++++++++ 1 file changed, 9 insertions(+) The bug has been closed via the following commit(s): https://gitweb.gentoo.org/repo/gentoo.git/commit/?id=e015ba7525117156c1672ed8af667e242b4dec02 commit e015ba7525117156c1672ed8af667e242b4dec02 Author: Sergei Trofimovich <slyfox@gentoo.org> AuthorDate: 2019-11-12 22:52:06 +0000 Commit: Sergei Trofimovich <slyfox@gentoo.org> CommitDate: 2019-11-12 22:52:12 +0000 dev-util/rr: RESTRICT=test, bug #632394 Too many tests fail due to specific kernel or toolchain version. Tests are still nice to run to see if anything works at all but are not very useful for regression testing. Closes: https://bugs.gentoo.org/632394 Package-Manager: Portage-2.3.79, Repoman-2.3.18 Signed-off-by: Sergei Trofimovich <slyfox@gentoo.org> dev-util/rr/rr-5.2.0-r1.ebuild | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) |
The following tests FAILED: 1 - 64bit_child (Failed) 2 - 64bit_child-no-syscallbuf (Failed) 3 - _llseek (Failed) 4 - _llseek-no-syscallbuf (Failed) ------------------------------------------------------------------- This is an unstable amd64 chroot image at a tinderbox (==build bot) name: 13.0-desktop_libressl_20170918-162903 ------------------------------------------------------------------- gcc-config -l: [1] x86_64-pc-linux-gnu-6.4.0 * llvm-config: 3.9.1 Available Python interpreters, in order of preference: [1] python3.4 [2] python3.6 (fallback) [3] python2.7 (fallback) java-config: The following VMs are available for generation-2: 1) IcedTea JDK 7.2.6.11 [icedtea-bin-7] *) IcedTea JDK 3.5.1 [icedtea-bin-8] Available Java Virtual Machines: [1] icedtea-bin-7 [2] icedtea-bin-8 system-vm emerge -qpv dev-util/rr [ebuild N ] dev-util/rr-4.5.0 USE="{-test}" PYTHON_TARGETS="python2_7"