[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [bug-gawk] Beta release of gawk 4.2.0 now available
From: |
arnold |
Subject: |
Re: [bug-gawk] Beta release of gawk 4.2.0 now available |
Date: |
Wed, 04 Oct 2017 04:22:17 -0600 |
User-agent: |
Heirloom mailx 12.4 7/29/08 |
Hi Assaf.
> 1.
> On FreeBSD-11.0p1, the following fails:
>
> ======== Starting shared library tests ========
> [...]
> inplace1
> ./inplace1.ok _inplace1 differ: char 47, line 2
> inplace2
> ./inplace2.ok _inplace2 differ: char 47, line 2
> inplace3
> ./inplace3.ok _inplace3 differ: char 47, line 2
> [...]
Please send results from make diffout.
> ============== _backsmalls1 =============
> *** backsmalls1.ok Sun Apr 5 08:13:50 2015
> --- _backsmalls1 Wed Oct 4 07:15:17 2017
This appears to be a locale issue.
> 2.
> On older Alpine-Linux 3.5.2 (which uses musl-libc instead of glibc, and
> busybox instead of coreutils) - the following fails:
>
> ls: unrecognized option: f
> BusyBox v1.25.1 (2016-10-26 16:15:20 GMT) multi-call binary.
> [...]
> ===================================
>
> So I assume one test uses "ls -f" which isn't always in busybox.
Busybox is (or was) non-standard here. Not my problem. :-)
> 3.
> On OpenSolaris 5.11, (x86 and sparc), compilation fails with:
> ===
> /bin/sh ./libtool --tag=CC --mode=compile gcc -DHAVE_CONFIG_H -I.
> -I./.. -g -O2 -DNDEBUG -MT inplace.lo -MD -MP -MF .deps/inplace.Tpo -c
> -o inplace.lo inplace.c
> libtool: compile: gcc -DHAVE_CONFIG_H -I. -I./.. -g -O2 -DNDEBUG -MT
> inplace.lo -MD -MP -MF .deps/inplace.Tpo -c inplace.c -fPIC -DPIC -o
> .libs/inplace.o
> In file included from /usr/include/stdio.h:15:0,
> from inplace.c:37:
> /opt/csw/lib/gcc/i386-pc-solaris2.10/5.2.0/include-fixed/sys/feature_tests.h:346:2:
> error: #error "Compiler or options invalid for pre-UNIX 03 X/Open
> applications and pre-2001 POSIX applications"
> #error "Compiler or options invalid for pre-UNIX 03 X/Open applications \
> ^
> make[4]: *** [inplace.lo] Error 1
> make[4]: Leaving directory
Please try to figure out what additional flags are needed here and we
can update a README file. You may need -std=gnu99 or some such.
> 4.
> On OpenSolaris 5.10, compilation fails with:
According to your log, it's the same issue as above:
/opt/csw/lib/gcc/sparc-sun-solaris2.10/5.2.0/include-fixed/sys/feature_tests.h:3
46:2: error: #error "Compiler or options invalid for pre-UNIX 03 X/Open applicat
ions and pre-2001 POSIX applications"
#error "Compiler or options invalid for pre-UNIX 03 X/Open applications \
^
> 5.
> On AIX (from the GCC compile farm), the following fail:
> ====
> forcenum
> ./forcenum.ok _forcenum differ: char 61, line 3
> make: 1254-004 The error code from the last command is 1.
> make: 1254-005 Ignored error code 1 from last command.
> ...
> mbstr1
> ./mbstr1.ok _mbstr1 differ: char 1, line 1
> make: 1254-004 The error code from the last command is 1.
> make: 1254-005 Ignored error code 1 from last command.
> mbstr2
> ./mbstr2.ok _mbstr2 differ: char 6, line 2
> make: 1254-004 The error code from the last command is 1.
> make: 1254-005 Ignored error code 1 from last command.
> muldimposix
> ...
> ============== _forcenum =============
> *** forcenum.ok Wed Aug 16 23:58:57 2017
> --- _forcenum Wed Oct 4 02:24:23 2017
> ***************
> *** 1,6 ****
> [] -> 0 (type string)
> [5apple] -> 5 (type string)
> ! [+NaN] -> nan (type strnum)
> [ 6] -> 6 (type strnum)
> [0x1az] -> 26 (type string)
> [011Q] -> 9 (type string)
> --- 1,6 ----
> [] -> 0 (type string)
> [5apple] -> 5 (type string)
> ! [+NaN] -> NaNQ (type strnum)
> [ 6] -> 6 (type strnum)
> [0x1az] -> 26 (type string)
> [011Q] -> 9 (type string)
> make: 1254-004 The error code from the last command is 1.
This is harmless - a libc difference.
> 6.
> This might be a false positive, but for some reason that I haven't
> pin-pointed yet, on some of my (automated) tests running 'make check'
> results only in "make 'CFLAGS=-g -O2 -DNDEBUG' 'LDFLAGS=' check-local"
> which does nothing - and so the tests are not executed.
>
> Perhaps related to not having tty ? or some other weirdness (these tests
> do not run on a terminal).
No clue. Sorry.
Thanks for the reports,
Arnold