headers
Luca Lesinigo | 19 Jun 2008 09:38

m4-1.4.11 test fails on HPPA

Using m4-1.4.11,

test-strtod.c:667: assertion failed
test-strtod.c:668: assertion failed
test-strtod.c:688: assertion failed
test-strtod.c:717: assertion failed
test-strtod.c:718: assertion failed
FAIL: test-strtod

I found the problem with m4 from Gentoo Portage and confirmed it  
with ./configure - make - make check in the sources from the original  
m4-1.4.11 tarball, compiling -O2 or -O0 doesn't change anything.

Confirmed on an HPPA 2.0 with:
- gentoo's gcc-4.1.2 and glibc-2.7-r2
- gentoo's gcc-4.2.4 and glibc-2.7-r2
but all checks are successful on HPPA 2.0 with:
- gentoo's gcc-4.3.1 and glibc-2.8_p20080602

I tried also on an x86_64 box with gcc-3.4.6 and glibc-2.6.1 and all  
checks are successful.

I'm not 100% sure if this is a problem in m4 or somewhere else. Feel  
free to mail me if you need any other test or if you have suggestions.

--
Luca Lesinigo
Permalink | Reply | 1 comment |
headers
Eric Blake | 19 Jun 2008 14:56
Gravatar

Re: m4-1.4.11 test fails on HPPA


According to Luca Lesinigo on 6/19/2008 1:38 AM:
| Using m4-1.4.11,
|
| test-strtod.c:667: assertion failed
|
| Confirmed on an HPPA 2.0 with:
| - gentoo's gcc-4.1.2 and glibc-2.7-r2
| - gentoo's gcc-4.2.4 and glibc-2.7-r2
| but all checks are successful on HPPA 2.0 with:
| - gentoo's gcc-4.3.1 and glibc-2.8_p20080602

Thanks for the report.  This has previously been reported; it is a
combination of two bugs: older glibc's strtod(3) is not C99 compliant, and
m4's configure file was only detecting a subset, rather than all, known
glibc strtod bugs.  Both problems have been fixed in the respective
version control repositories.

You can work around it by using 'gl_cv_func_strtod_works=no ./configure'
instead of plain './configure'.  Or you can try a snapshot of what will
become the future m4 1.6:

http://home.comcast.net/~ericblake/m4-1.5.89a.31-d9538d.tar.gz
http://home.comcast.net/~ericblake/m4-1.5.89a.31-d9538d.tar.gz.asc
http://home.comcast.net/~ericblake/m4-1.5.89a.31-d9538d.tar.bz2
http://home.comcast.net/~ericblake/m4-1.5.89a.31-d9538d.tar.bz2.asc
http://home.comcast.net/~ericblake/m4-1.5.89a.31-d9538d.tar.lzma
http://home.comcast.net/~ericblake/m4-1.5.89a.31-d9538d.tar.lzma.asc

|
(Continue reading)

Permalink | Reply | 0 comments |
headers
Diab Jerius | 24 Jun 2008 20:05
Picon

m4-1.4.11: make check fails on Fedora Core 8

m4 : 1.4.11
OS : Fedora Core 8
CPU : Intel 32bit Xeon
gcc : 4.1.2 20070925 (Red Hat 4.1.2-33)
libc : 2.7 with extensions:
        The C stubs add-on version 2.1.2.
        crypt add-on version 2.1 by Michael Glad and others
        GNU Libidn by Simon Josefsson
        Native POSIX Threads Library by Ulrich Drepper et al
        BIND-8.2.3-T5B
        RT using linux kernel aio

./configure is invoked with no arguments.

make check fails with:

test-strtod.c:667: assertion failed
test-strtod.c:668: assertion failed
test-strtod.c:688: assertion failed
test-strtod.c:717: assertion failed
test-strtod.c:718: assertion failed

Thanks,
Diab

--

-- 
Diab Jerius <djerius <at> cfa.harvard.edu>
Smithsonian Astrophysical Observatory
(Continue reading)

Permalink | Reply | 1 comment |
headers
Diab Jerius | 24 Jun 2008 20:34
Picon

Re: m4-1.4.11: make check fails on Fedora Core 8

Whoops.  Shoulda searched the bug list archives first.
No need to reply.

My apologies for the line noise.

Diab

On Tue, 2008-06-24 at 14:05 -0400, Diab Jerius wrote:
> m4 : 1.4.11
> OS : Fedora Core 8
> CPU : Intel 32bit Xeon
> gcc : 4.1.2 20070925 (Red Hat 4.1.2-33)
> libc : 2.7 with extensions:
>         The C stubs add-on version 2.1.2.
>         crypt add-on version 2.1 by Michael Glad and others
>         GNU Libidn by Simon Josefsson
>         Native POSIX Threads Library by Ulrich Drepper et al
>         BIND-8.2.3-T5B
>         RT using linux kernel aio
> 
> ./configure is invoked with no arguments.
> 
> make check fails with:
> 
> test-strtod.c:667: assertion failed
> test-strtod.c:668: assertion failed
> test-strtod.c:688: assertion failed
> test-strtod.c:717: assertion failed
> test-strtod.c:718: assertion failed
>
(Continue reading)

Permalink | Reply | 0 comments |
headers
Ludovic Courtès | 1 Jul 2008 15:37
Picon

m4-1.4.11: Failure of `126.diversion'

Hi,

On GNU/Linux (NixOS), I observe the following failure:

  Checking ./126.diversions
   <at>  ../doc/m4.texinfo:4711: Origin of test
  ./126.diversions: stderr mismatch
  --- m4-tmp.17114/m4-xerr        2008-07-01 13:21:45.000000000 +0000
  +++ m4-tmp.17114/m4-err 2008-07-01 13:21:45.000000000 +0000
   <at>  <at>  -0,0 +1,2  <at>  <at> 
  +m4: ERROR: copying inserted file: Broken pipe
  +m4: write error

The thing is run by a build daemon in a chroot with no `/tmp'.

Any idea what's going on?

Thanks in advance,
Ludovic.
Permalink | Reply | 0 comments |
headers
Eric Blake | 1 Jul 2008 19:16
Picon

Re: m4-1.4.11: Failure of `126.diversion'

>   ./126.diversions: stderr mismatch
>   --- m4-tmp.17114/m4-xerr        2008-07-01 13:21:45.000000000 +0000
>   +++ m4-tmp.17114/m4-err 2008-07-01 13:21:45.000000000 +0000
>    <at>  <at>  -0,0 +1,2  <at>  <at> 
>   +m4: ERROR: copying inserted file: Broken pipe
>   +m4: write error
> 
> The thing is run by a build daemon in a chroot with no `/tmp'.
> 
> Any idea what's going on?

Yes.  Your buildbot is ignoring SIGPIPE.  This has been previously
reported, and the testsuite in the latest git version already works
around such enviornments (even though I personally don't like
running with SIGPIPE ignored, because of the extra error messages
that it tends to produce).

--

-- 
Eric Blake
Permalink | Reply | 1 comment |
headers
Ludovic Courtès | 1 Jul 2008 19:56
Picon

Re: m4-1.4.11: Failure of `126.diversion'

Hi,

ericblake <at> comcast.net (Eric Blake) writes:

> Yes.  Your buildbot is ignoring SIGPIPE.  This has been previously
> reported, and the testsuite in the latest git version already works
> around such enviornments (even though I personally don't like
> running with SIGPIPE ignored, because of the extra error messages
> that it tends to produce).

Hmm, I found commit 546109144ee2014257166698df2a0a9d14526a5a, which
removes SIGPIPE from the `kill' example, but that dates back to before
1.4.11 AFAICS.  Maybe that's another problem here?

Thanks,
Ludovic.
Permalink | Reply | 0 comments |
headers
Eric Blake | 1 Jul 2008 20:38
Picon

Re: m4-1.4.11: Failure of `126.diversion'

> Hmm, I found commit 546109144ee2014257166698df2a0a9d14526a5a, which
> removes SIGPIPE from the `kill' example, but that dates back to before
> 1.4.11 AFAICS.

Unrelated to the problem at hand.

>  Maybe that's another problem here?

Much more recent:

http://lists.gnu.org/archive/html/m4-discuss/2008-06/msg00005.html

http://git.savannah.gnu.org/gitweb/?p=m4.git;a=commitdiff;h=2b570fc

--

-- 
Eric Blake
Permalink | Reply | 501 comments |
headers
Ludovic Courtès | 1 Jul 2008 21:44
Picon

Re: m4-1.4.11: Failure of `126.diversion'

ericblake <at> comcast.net (Eric Blake) writes:

> Much more recent:
>
> http://lists.gnu.org/archive/html/m4-discuss/2008-06/msg00005.html
>
> http://git.savannah.gnu.org/gitweb/?p=m4.git;a=commitdiff;h=2b570fc

OK, works like a charm.

Thanks!

Ludovic.
Permalink | Reply | 501 comments |

Gmane