On 08/02/12 06:04 +0000, Dietmar Maurer wrote:
Just compiled the latest 0.10.0 release and got:
=======================================
libqb 0.10.0: tests/test-suite.log
========================================
1 of 8 tests failed.
.. contents:: :depth: 2
FAIL: loop.test (exit: 1)
=========================
Running suite(s): loop_job
loop_timers
[error] can't add a timer with either (l == NULL || timer_fn == NULL)
[error] can't add a timer with either (l == NULL || timer_fn == NULL)
average error for 1500000 ns timer is 121723 (ns) (8.114895)
average error for 10500000 ns timer is 101712 (ns) (0.968686)
average error for 19500000 ns timer is 101931 (ns) (0.522726)
average error for 28500000 ns timer is 111700 (ns) (0.391933)
average error for 37500000 ns timer is 103482 (ns) (0.275953)
average error for 46500000 ns timer is 124853 (ns) (0.268501)
average error for 55500000 ns timer is 123825 (ns) (0.223110)
average error for 64500000 ns timer is 125102 (ns) (0.193957)
average error for 73500000 ns timer is 136308 (ns) (0.185453)
average error for 82500000 ns timer is 146708 (ns) (0.177828)
average error for 100000000 ns timer is 146534 (ns) (0.146535)
loop_signal_suite
90%: Checks: 10, Failures: 1, Errors: 0
check_loop.c:118:P:limits:test_loop_job_input:0: Passed
check_loop.c:135:P:run_one:test_loop_job_1:0: Passed
check_loop.c:152:P:run_recursive:test_loop_job_4:0: Passed
check_loop.c:167:P:run_500:test_loop_job_nuts:0: Passed
check_loop.c:213:F:rate_limit:test_job_rate_limit:0: Assertion
'job_1_run_count<(RATE_LIMIT_RUNTIME_SEC * (QB_TIME_MS_IN_SEC/50)) + 1' failed:
job_1_run_count==61, (RATE_LIMIT_RUNTIME_SEC * (QB_TIME_MS_IN_SEC/50)) + 1==61
check_loop.c:264:P:limits:test_loop_timer_input:0: Passed
check_loop.c:309:P:basic:test_loop_timer_basic:0: Passed
check_loop.c:338:P:precision:test_loop_timer_precision:0: Passed
check_loop.c:458:P:expire_leak:test_loop_timer_expire_leak:0: Passed
check_loop.c:492:P:signals:test_loop_sig_handling:0: Passed
The test passes if I use <= in the assertion:
_ck_assert_int(job_1_run_count, <=, (RATE_LIMIT_RUNTIME_SEC * (QB_TIME_MS_IN_SEC/50))
+ 1);
I'll add 10 instead of 1, really not the issue. I just want to make sure
it is not a million more. This might also be sensitive to machine
performance too. So I'll patch now.
Thanks for the report.
-Angus
- Dietmar
_______________________________________________
quarterback-devel mailing list
quarterback-devel(a)lists.fedorahosted.org
https://fedorahosted.org/mailman/listinfo/quarterback-devel