Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Test suite fails on NetBSD #837

Open
gson1703 opened this issue May 16, 2017 · 3 comments
Open

Test suite fails on NetBSD #837

gson1703 opened this issue May 16, 2017 · 3 comments

Comments

@gson1703
Copy link

Running "make check" in jemalloc 4.5.0 on a NetBSD/amd64 7.0.2 system
fails in the thread_arena test:

=== test/integration/thread_arena ===
thd_start:test/integration/thread_arena.c:15: Failed assertion: (p) != (NULL) --> 0x0 == 0x0: Error in malloc()
[1] Segmentation fault (core dumped) export MALLOC_CO...
Test harness error
Makefile:410: recipe for target 'check_integration_decay' failed

Using "git bisect", I have determined that the tests started failing
with git revision c2f970c. I have also tested every revision between
c2f970c and the 4.5.0 tag, and they all fail, though sometimes it
takes multiple runs of the test suite for the failure to occur.

Looking at the "dev" branch, things are more complicated. There, the
tests also broke with commit c2f970c, but were at least mostly fixed
by 4e910fc, and then later broken, fixed, and broken once again, and
are now (as of 18a8368) failing with the following messages:

=== test/integration/thread_arena ===
test_thread_arena (non-reentrant): pass
test_thread_arena (libc-reentrant): pass
[1] Segmentation fault (core dumped) ${JEMALLOC_TEST_...
Test harness error: test/integration/thread_arena w/ MALLOC_CONF="dirty_decay_time:-1,muzzy_decay_time:-1"
Use prefix to debug, e.g. JEMALLOC_TEST_PREFIX="gdb --args" sh test/test.sh test/integration/thread_arena
Makefile:478: recipe for target 'check_integration_decay' failed

Let me know if there is some additional information I can provide to help track this down.

@jasone
Copy link
Member

jasone commented May 16, 2017

We don't have continuous integration coverage for NetBSD, nor do any of the primary developers use NetBSD in any capacity. If you're able to diagnose and fix the regressions, we'lll be happy to take patches, but we don't have the time/compute resources to do the work directly ourselves.

jsonn pushed a commit to jsonn/pkgsrc that referenced this issue May 16, 2017
jsonn pushed a commit to jsonn/pkgsrc that referenced this issue May 17, 2017
This also has the benefit of not triggering the issue reported in
jemalloc/jemalloc#837 .
@gson1703
Copy link
Author

It looks like the root cause of this is an issue with NetBSD's mmap(); see http://gnats.netbsd.org/52239 for my analysis. If os_overcommits is set to true, which seems like correct value in any case given http://gnats.netbsd.org/12451 , the tests pass.

@jasone
Copy link
Member

jasone commented May 18, 2017

Wow, I bet that was interesting to diagnose! Thanks for digging into the issue.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

No branches or pull requests

2 participants