[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: 'make check' errors for m4-1.4.5 on various platforms
From: |
Eric Blake |
Subject: |
Re: 'make check' errors for m4-1.4.5 on various platforms |
Date: |
Wed, 26 Jul 2006 20:17:27 +0000 (UTC) |
User-agent: |
Loom/3.14 (http://gmane.org/) |
Bert Deknuydt <Bert.Deknuydt <at> esat.kuleuven.be> writes:
>
>
> Hello,
>
> Thanks for the latest version of m4! However I encountered some bugs while
> running the testsuite.
Thanks for the report. Failures in test 79 and 82 are already well-reported
and fixed in CVS. I'm focusing on the other thing you turned up:
> Platform: alpha-dec-osf5.1
> Compiler: gcc (GCC) 3.4.6
> Configure options: ./configure --prefix=/freeware/gnu-tools --exec-
prefix=/freeware/gnu-tools/alpha
> Build: Builds correctly
> Check: Error; relevant output below
>
> [...]
> Checking ./096.sysval
> ./check-them: 36246 Hangup
> <at> ../doc/m4.texinfo:3788: Origin of test
> ./096.sysval: stdout mismatch
> 1,4d0
> <
> < 256
> <
> < 2304
This was a testsuite bug, for assuming that kill -1 reliably kills shells with
signal 1 (as a counterexample, ksh traps SIGHUP and exits normally with status
127), but I've already tried to improve things in CVS. So far, I have not
found any counter-case for signal 13 reliably killing a shell. What does this
output for you, with m4 1.4.5?
$ echo 'syscmd(kill -13 $$)sysval' | m4
--
Eric Blake