strace 4.14 released

[Dmitry V. Levin]

On Wed, Oct 05, 2016 at 11:08:24AM +0200, Andreas Schwab wrote:
> I'm seeing this testsuite failure:
> 
> https://build.opensuse.org/package/live_build_log/home:Andreas_Schwab:Factory/strace/s/s390x
> 
> FAIL: count
> ===========
> 
> Pattern of expected output:  *[^ ]+ +(1\.[01]|0\.99)[^n]*nanosleep
> Actual output:
> % time     seconds  usecs/call     calls    errors syscall
> ------ ----------- ----------- --------- --------- ----------------
>  99.97    1.286494     1286494         1           nanosleep
>   0.01    0.000102         102         1           execve
>   0.01    0.000077          13         6           mmap
>   0.00    0.000040          13         3           mprotect
>   0.00    0.000026          13         2           open
>   0.00    0.000022          11         2           close
>   0.00    0.000022          11         2           fstat
>   0.00    0.000014          14         1           munmap
>   0.00    0.000013          13         1         1 access
>   0.00    0.000012          12         1           read
>   0.00    0.000012          12         1           brk
> ------ ----------- ----------- --------- --------- ----------------
> 100.00    1.286834                    21         1 total
> count.test: failed test: ../strace -cw ./sleep 1 output mismatch

I've seen exactly the same failure yesterday, perhaps that s390x build
environment just became temporarily slower than before.

The test allows up to 1.19* seconds for a 1-second nanosleep.
Allowing more time would decrease the chance to detect strace incapability
to measure the time spent in syscalls.

Any ideas how to workaround the slowness of test environment without
compromising the result of this test?


-- 
ldv
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 819 bytes
Desc: not available
URL: <http://lists.strace.io/pipermail/strace-devel/attachments/20161005/db042d09/attachment.bin>