Re: 29 failed tests in the Texinfo 7.1 testsuite on Debian with IBM POWER9

[Dennis Clarke]

On 6/1/24 05:55, Patrice Dumas wrote:
> On Sat, Jun 01, 2024 at 01:17:59AM -0400, Dennis Clarke via Bug reports for the 
> GNU Texinfo documentation system wrote:
> > FAIL: test_scripts/nested_formats_texi_nested_formats.sh
> >
> > Why fail? I have no idea.
> >
> > This is Debian on an IBM POWER9 server :
>
> Maybe something to do with specifics of POWER9, I guess it could be
> somewhat peculiar.  Also the gcc seems very recent to me, could also be
> that.

I was baffled. I also did the exact same tests on a bog standard off
the shelf boring Xeon machine and everything works like a charm with the
same GCC release :

oberon$ uname -a
Linux oberon 6.6.26-genunix #1 SMP Thu Apr 11 13:57:18 GMT 2024 x86_64 
GNU/Linux
oberon$ gcc --version
gcc (GENUNIX Wed May 29 23:34:00 UTC 2024) 14.1.0
Copyright (C) 2024 Free Software Foundation, Inc.
This is free software; see the source for copying conditions.  There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.

oberon$
oberon$
oberon$ as --version
GNU assembler (GNU Binutils) 2.42
Copyright (C) 2024 Free Software Foundation, Inc.
This program is free software; you may redistribute it under the terms of
the GNU General Public License version 3 or later.
This program has absolutely no warranty.
This assembler was configured for a target of `x86_64-pc-linux-gnu'.
oberon$

So .. that explains the "wow" email I just had to send in! :)

> Could you please send the log file of any of these failed tests, for
> example

Yes indeed !


> tp/tests/nested_formats/test_log/texi_nested_formats.log
>
> > dax$ gcc --version
> > gcc (GENUNIX Thu May 30 01:20:34 UTC 2024) 14.1.0

I could isolate just the failed tests but may as well tarball up
the entire pile?  For now .. just the failed tests. I have attached
a GNU tar file ( xz compressed ) of the logs and have also jsut run
at least "nested_formats_texi_nested_formats.sh" which spits out a
massive pile of diff lines :

dax$
dax$ uname -a
Linux dax 6.1.86-mrw1 #1 SMP Sat Apr 13 20:47:03 PDT 2024 ppc64le GNU/Linux
dax$
dax$ echo $SHELL
/bin/bash
dax$
dax$ env | grep '^L'
LOGNAME=dclarke
LESS_IS_MORE=1
LANG=en_US.UTF-8
LD=/opt/bw/bin/gld
LC_TIME=C
dax$
dax$ ./tp/tests/test_scripts/nested_formats_texi_nested_formats.sh ^C
dax$ cd ./tp/tests/
dax$
dax$ test_scripts/nested_formats_texi_nested_formats.sh
D: nested_formats//diffs/texi_nested_formats_html.diff (printed below)
diff -a -u -r 
./nested_formats//res_parser_html/texi_nested_formats/index.html 
nested_formats//out_parser_html/texi_nested_formats/index.html
--- ./nested_formats//res_parser_html/texi_nested_formats/index.html 
2023-10-17 12:13:33.000000000 +0000
+++ nested_formats//out_parser_html/texi_nested_formats/index.html 
2024-06-02 00:14:01.079764772 +0000
@@ -93,10 +93,10 @@
 </li></ol>

 <table class="multitable">
-<thead><tr><th width="60%">mu&ndash;ltitable headitem</th><th 
width="70%">another tab</th></tr></thead>
-<tbody><tr><td width="60%">mu&ndash;ltitable item</td><td 
width="70%">multitable tab</td></tr>
-<tr><td width="60%">mu&ndash;ltitable item 2</td><td 
width="70%">multitable tab 2</td></tr>
-<tr><td width="60%">lone mu&ndash;ltitable item</td></tr>
+<thead><tr><th width="59%">mu&ndash;ltitable headitem</th><th 
width="70%">another tab</th></tr></thead>
+<tbody><tr><td width="59%">mu&ndash;ltitable item</td><td 
width="70%">multitable tab</td></tr>
+<tr><td width="59%">mu&ndash;ltitable item 2</td><td 
width="70%">multitable tab 2</td></tr>
+<tr><td width="59%">lone mu&ndash;ltitable item</td></tr>
 </tbody>
 </table>

.
.
.
endless pile of similar diffs.


@@ -3667,10 +3667,10 @@
 </li></ol>

 <table class="multitable">
-<thead><tr><th width="60%">mu&ndash;ltitable headitem</th><th 
width="70%">another tab</th></tr></thead>
-<tbody><tr><td width="60%">mu&ndash;ltitable item</td><td 
width="70%">multitable tab</td></tr>
-<tr><td width="60%">mu&ndash;ltitable item 2</td><td 
width="70%">multitable tab 2</td></tr>
-<tr><td width="60%">lone mu&ndash;ltitable item</td></tr>
+<thead><tr><th width="59%">mu&ndash;ltitable headitem</th><th 
width="70%">another tab</th></tr></thead>
+<tbody><tr><td width="59%">mu&ndash;ltitable item</td><td 
width="70%">multitable tab</td></tr>
+<tr><td width="59%">mu&ndash;ltitable item 2</td><td 
width="70%">multitable tab 2</td></tr>
+<tr><td width="59%">lone mu&ndash;ltitable item</td></tr>
 </tbody>
 </table>

D: nested_formats//diffs/texi_nested_formats_html.diff (printed above)
testdir: nested_formats/
driving_file: ./nested_formats//list-of-tests
found special first line, commands now: : :_html
made result dir: ./nested_formats//res_parser/
made result dir: ./nested_formats//res_parser_html/

doing test texi_nested_formats, src_file 
./nested_formats//nested_formats.texi
format_option:
texi2any.pl texi_nested_formats -> 
nested_formats//out_parser/texi_nested_formats
 /opt/bw/bin/perl -w ./..//texi2any.pl  --force --conf-dir ./../t/init/ 
--conf-dir ./../init --conf-dir ./../ext -I ./nested_formats/ -I 
nested_formats// -I ./ -I . -I built_input --error-limit=1000 -c TEST=1 
--output nested_formats//out_parser/texi_nested_formats/ -c 
TEXINFO_OUTPUT_FORMAT=plaintexinfo ./nested_formats//nested_formats.texi 
> nested_formats//out_parser/texi_nested_formats/nested_formats.1 
2>nested_formats//out_parser/texi_nested_formats/nested_formats.2

doing test texi_nested_formats, src_file 
./nested_formats//nested_formats.texi
format_option: --html
texi2any.pl texi_nested_formats -> 
nested_formats//out_parser_html/texi_nested_formats
 /opt/bw/bin/perl -w ./..//texi2any.pl --html --force --conf-dir 
./../t/init/ --conf-dir ./../init --conf-dir ./../ext -I 
./nested_formats/ -I nested_formats// -I ./ -I . -I built_input 
--error-limit=1000 -c TEST=1  --output 
nested_formats//out_parser_html/texi_nested_formats/ -c 
TEXINFO_OUTPUT_FORMAT=plaintexinfo ./nested_formats//nested_formats.texi 
> nested_formats//out_parser_html/texi_nested_formats/nested_formats.1 
2>nested_formats//out_parser_html/texi_nested_formats/nested_formats.2

all done, exiting with status 1
dax$

Right. That makes no sense at all.

--
Dennis Clarke
RISC-V/SPARC/PPC/ARM/CISC
UNIX and Linux spoken

texinfo_fail_logs.tar.xz
Description: application/xz