FAQ
1. Encode causes AIX 4.2 't's, see below

Automated smoke report for patch 16109 cc gcc
HP-UX 11.00 B.11.11.04 3.0.4 32-bit
O = OK | 3.0.4-64 64-bit +GNUld
F = Failure(s), extended report at the bottom | HP-UX 10.20 A.10.32.30 3.0.4
? = still running or test results not (yet) available | AIX 4.3.3.0 vac 5.0.2.2 3.0.4
Build failures during: - = unknown, = skipped | AIX 4.2.1.0 xlc 3.1.4.10 3.0.4
c = Configure, m = make, t = make test-prep | Cygwin 1.3.10 2.95.3-5

HP-UX HP-UX HP-UX HP-UX AIX AIX
11.00 11.00 10.20 10.20 4.3.3 4.2.1
HPc gcc HPc gcc vac xlc
16109 16109 16109 16109 16109 16109 Configuration
------- ------- ------- ------- ------- ------- --------------------------------------------------------------------
F F F F F F ? ? F F F F F F F F F F F F t t t t -Uuseperlio
F F F F ? ? ? ? F F F F F F F F F F F F t t t t -Duseperlio
F F F F ? ? ? ? F F F F -Duseperlio -Duse64bitint
F F F F ? ? ? ? -Duseperlio -Duse64bitall
F F F F ? ? ? ? F F F F F F F F F F F F t t t t -Duseperlio -Duselongdouble
F F F F ? ? ? ? F F F F -Duseperlio -Dusemorebits
F F F F ? ? ? ? -Duseperlio -Duse64bitall -Duselongdouble
F F F F ? ? ? ? m m F F F ? ? ? F F ? ? t t t t -Uuseperlio -Dusethreads -Duseithreads
F F F F ? ? ? ? F F F F ? ? ? ? ? ? ? ? t t t t -Duseperlio -Dusethreads -Duseithreads
F F F F ? ? ? ? ? ? ? ? ? ? ? ? -Duseperlio -Duse64bitint -Dusethreads -Duseithreads
F F F F ? ? ? ? ? ? ? ? ? ? ? ? -Duseperlio -Duse64bitall -Dusethreads -Duseithreads
F F F F ? ? ? ? F F F F ? ? ? ? ? ? ? ? t t t t -Duseperlio -Duselongdouble -Dusethreads -Duseithreads
F F F F ? ? ? ? ? ? ? ? ? ? ? ? -Duseperlio -Dusemorebits -Dusethreads -Duseithreads
F F F F ? ? ? ? ? ? ? ? ? ? ? ? -Duseperlio -Duse64bitall -Duselongdouble -Dusethreads -Duseithreads
+----- PERLIO = perlio | +- PERLIO = perlio -DDEBUGGING
+------- PERLIO = stdio +--- PERLIO = stdio -DDEBUGGING

Failures:

l1 MANIFEST errors:
did not declare ext/Encode/t/11156.utf8
did not declare ext/Encode/t/11159.pio
did not declare ext/Encode/t/11159.sio
did not declare ext/Encode/t/14430.utf8
did not declare ext/Encode/t/14433.pio
did not declare ext/Encode/t/14433.sio
did not declare ext/Encode/t/15987.utf8
did not declare ext/Encode/t/15990.pio
did not declare ext/Encode/t/15990.sio
did not declare ext/Encode/t/17595.utf8
did not declare ext/Encode/t/17598.pio
did not declare ext/Encode/t/17598.sio
did not declare ext/Encode/t/1827.utf8
did not declare ext/Encode/t/1830.pio
did not declare ext/Encode/t/1830.sio
did not declare ext/Encode/t/1832.utf8
did not declare ext/Encode/t/1835.pio
did not declare ext/Encode/t/1835.sio
did not declare ext/Encode/t/19116.utf8
did not declare ext/Encode/t/19119.pio
did not declare ext/Encode/t/19119.sio
did not declare ext/Encode/t/19535.utf8
did not declare ext/Encode/t/19538.pio
did not declare ext/Encode/t/19538.sio
did not declare ext/Encode/t/21818.utf8
did not declare ext/Encode/t/21821.pio
did not declare ext/Encode/t/21821.sio
did not declare ext/Encode/t/22691.utf8
did not declare ext/Encode/t/22694.pio
did not declare ext/Encode/t/22694.sio
did not declare ext/Encode/t/24939.utf8
did not declare ext/Encode/t/24942.pio
did not declare ext/Encode/t/24942.sio
did not declare ext/Encode/t/2844.utf8
did not declare ext/Encode/t/2847.pio
did not declare ext/Encode/t/2847.sio
did not declare ext/Encode/t/4916.utf8
did not declare ext/Encode/t/4919.pio
did not declare ext/Encode/t/4919.sio
did not declare ext/Encode/t/4940.utf8
did not declare ext/Encode/t/4943.pio
did not declare ext/Encode/t/4943.sio
did not declare ext/Encode/t/6048.utf8
did not declare ext/Encode/t/6051.pio
did not declare ext/Encode/t/6051.sio
did not declare ext/Encode/t/8030.utf8
did not declare ext/Encode/t/8033.pio
did not declare ext/Encode/t/8033.sio
did not declare lib/ExtTest.pm
did not declare lib/ExtUtils/t/Big-Fat-Dummy/Liar/Makefile
did not declare lib/ExtUtils/t/Big-Fat-Dummy/Makefile
did not declare opcode.h-old
did not declare opnames.h-old
did not declare pp.sym-old
did not declare pp_proto.h-old

HP-UX 11.00 stdio/perlio -Uuseperlio
HP-UX 11.00 stdio/perlio -DDEBUGGING -Uuseperlio
HP-UX 11.00 stdio/perlio -Duseperlio
HP-UX 11.00 stdio/perlio -DDEBUGGING -Duseperlio
HP-UX 11.00 stdio/perlio -Duseperlio -Duse64bitint
HP-UX 11.00 stdio/perlio -DDEBUGGING -Duseperlio -Duse64bitint
HP-UX 11.00 stdio/perlio -Duseperlio -Duse64bitall
HP-UX 11.00 stdio/perlio -DDEBUGGING -Duseperlio -Duse64bitall
HP-UX 11.00 stdio/perlio -Duseperlio -Duselongdouble
HP-UX 11.00 stdio/perlio -DDEBUGGING -Duseperlio -Duselongdouble
HP-UX 11.00 stdio/perlio -Duseperlio -Dusemorebits
HP-UX 11.00 stdio/perlio -DDEBUGGING -Duseperlio -Dusemorebits
HP-UX 11.00 stdio/perlio -Duseperlio -Duse64bitall -Duselongdouble
HP-UX 11.00 stdio/perlio -DDEBUGGING -Duseperlio -Duse64bitall -Duselongdouble
HP-UX 11.00 stdio/perlio -Uuseperlio -Dusethreads -Duseithreads
HP-UX 11.00 stdio/perlio -DDEBUGGING -Uuseperlio -Dusethreads -Duseithreads
HP-UX 11.00 stdio/perlio -Duseperlio -Dusethreads -Duseithreads
HP-UX 11.00 stdio/perlio -DDEBUGGING -Duseperlio -Dusethreads -Duseithreads
HP-UX 11.00 stdio/perlio -Duseperlio -Duse64bitint -Dusethreads -Duseithreads
HP-UX 11.00 stdio/perlio -DDEBUGGING -Duseperlio -Duse64bitint -Dusethreads -Duseithreads
HP-UX 11.00 stdio/perlio -Duseperlio -Duse64bitall -Dusethreads -Duseithreads
HP-UX 11.00 stdio/perlio -DDEBUGGING -Duseperlio -Duse64bitall -Dusethreads -Duseithreads
HP-UX 11.00 stdio/perlio -Duseperlio -Duselongdouble -Dusethreads -Duseithreads
HP-UX 11.00 stdio/perlio -DDEBUGGING -Duseperlio -Duselongdouble -Dusethreads -Duseithreads
HP-UX 11.00 stdio/perlio -Duseperlio -Dusemorebits -Dusethreads -Duseithreads
HP-UX 11.00 stdio/perlio -DDEBUGGING -Duseperlio -Dusemorebits -Dusethreads -Duseithreads
HP-UX 11.00 stdio/perlio -Duseperlio -Duse64bitall -Duselongdouble -Dusethreads -Duseithreads
HP-UX 11.00 stdio/perlio -DDEBUGGING -Duseperlio -Duse64bitall -Duselongdouble -Dusethreads -Duseithreads
lib/ExtUtils/t/INST_PREFIX..........FAILED at test 20

HPUX 11 gcc stdio/perlio -Uuseperlio
lib/ExtUtils/t/INST_PREFIX..........FAILED at test 20

HP-UX 10.20 stdio/perlio -Uuseperlio
HP-UX 10.20 stdio/perlio -DDEBUGGING -Uuseperlio
HP-UX 10.20 stdio/perlio -Duseperlio
HP-UX 10.20 stdio/perlio -DDEBUGGING -Duseperlio
HP-UX 10.20 stdio/perlio -Duseperlio -Duselongdouble
HP-UX 10.20 stdio/perlio -DDEBUGGING -Duseperlio -Duselongdouble
HP-UX 10.20 stdio/perlio -DDEBUGGING -Uuseperlio -Dusethreads -Duseithreads
HP-UX 10.20 stdio/perlio -Duseperlio -Dusethreads -Duseithreads
HP-UX 10.20 stdio/perlio -DDEBUGGING -Duseperlio -Dusethreads -Duseithreads
HP-UX 10.20 stdio/perlio -Duseperlio -Duselongdouble -Dusethreads -Duseithreads
HP-UX 10.20 stdio/perlio -DDEBUGGING -Duseperlio -Duselongdouble -Dusethreads -Duseithreads
lib/ExtUtils/t/INST_PREFIX..........FAILED at test 20

HPUX 10 gcc stdio/perlio -Uuseperlio
HPUX 10 gcc stdio/perlio -DDEBUGGING -Uuseperlio
HPUX 10 gcc stdio/perlio -Duseperlio
HPUX 10 gcc stdio/perlio -DDEBUGGING -Duseperlio
HPUX 10 gcc stdio/perlio -Duseperlio -Duselongdouble
HPUX 10 gcc stdio/perlio -DDEBUGGING -Duseperlio -Duselongdouble
lib/ExtUtils/t/INST_PREFIX..........FAILED at test 20

HPUX 10 gcc stdio -Uuseperlio -Dusethreads -Duseithreads
lib/ExtUtils/t/INST_PREFIX..........FAILED at test 20

AIX 4.3 stdio/perlio -Uuseperlio
AIX 4.3 stdio/perlio -DDEBUGGING -Uuseperlio
AIX 4.3 stdio/perlio -Duseperlio
AIX 4.3 stdio/perlio -DDEBUGGING -Duseperlio
lib/ExtUtils/t/INST_PREFIX..........FAILED at test 20

AIX 4.3 stdio/perlio -Duseperlio -Duse64bitint
AIX 4.3 stdio/perlio -DDEBUGGING -Duseperlio -Duse64bitint
ext/Encode/t/fallback...............FAILED at test 14
lib/ExtUtils/t/INST_PREFIX..........FAILED at test 20

AIX 4.3 stdio/perlio -Duseperlio -Duselongdouble
AIX 4.3 stdio/perlio -DDEBUGGING -Duseperlio -Duselongdouble
lib/ExtUtils/t/INST_PREFIX..........FAILED at test 20

AIX 4.3 stdio/perlio -Duseperlio -Dusemorebits
AIX 4.3 stdio/perlio -DDEBUGGING -Duseperlio -Dusemorebits
ext/Encode/t/fallback...............FAILED at test 14
lib/ExtUtils/t/INST_PREFIX..........FAILED at test 20

AIX 4.3 stdio/perlio -Uuseperlio -Dusethreads -Duseithreads
lib/ExtUtils/t/INST_PREFIX..........FAILED at test 20


AIX 4.2 ---

Reading big5-eten (big5-eten)
Reading big5-hkscs (big5-hkscs)
Reading cp950 (cp950)
Reading MacChineseTrad (MacChineseTrad)
Writing compiled form
102322 bytes in string tables
410276 bytes (20%) saved spotting duplicates
../../../miniperl "-I../../../lib" "-I../../../lib" ../../../lib/ExtUtils/xsubpp -nolinenumbers -typemap ../../../lib/ExtUtils/typemap TW.xs > TW.xsc && mv TW.xsc TW.c
xlc -c -I../Encode -DDEBUGGING -D_ALL_SOURCE -D_ANSI_C_SOURCE -D_POSIX_SOURCE -qmaxmem=16384 -qnoansialias -I/pro/local/include -O2 -DVERSION=\"1.25\" -DXS_VERSION=\"1.25\" "-I../../.." TW.c
xlc -c -I../Encode -DDEBUGGING -D_ALL_SOURCE -D_ANSI_C_SOURCE -D_POSIX_SOURCE -qmaxmem=16384 -qnoansialias -I/pro/local/include -O2 -DVERSION=\"1.25\" -DXS_VERSION=\"1.25\" "-I../../.." big5_t.c
xlc: 1501-229 Compilation ended due to lack of space
make[2]: *** [big5_t.o] Error 68
make[2]: Leaving directory `/pro/3gl/CPAN/perl-current/ext/Encode/TW'
make[1]: *** [subdirs] Error 2
make[1]: Leaving directory `/pro/3gl/CPAN/perl-current/ext/Encode'
make: *** [lib/auto/Encode/Encode.so] Error 2
ibm:/pro/3gl/CPAN/perl-current 103 >

--
H.Merijn Brand Amsterdam Perl Mongers (http://amsterdam.pm.org/)
using perl-5.6.1, 5.7.3 & 631 on HP-UX 10.20 & 11.00, AIX 4.2, AIX 4.3,
WinNT 4, Win2K pro & WinCE 2.11. Smoking perl CORE: smokers@perl.org
http://archives.develooper.com/daily-build@perl.org/ perl-qa@perl.org
send smoke reports to: smokers-reports@perl.org, QA: http://qa.perl.org

Search Discussions

  • Dan Kogai at Apr 24, 2002 at 11:56 am

    On Wednesday, April 24, 2002, at 08:44 , H.Merijn Brand wrote:
    Reading big5-eten (big5-eten)
    Reading big5-hkscs (big5-hkscs)
    Reading cp950 (cp950)
    Reading MacChineseTrad (MacChineseTrad)
    Writing compiled form
    102322 bytes in string tables
    410276 bytes (20%) saved spotting duplicates
    ../../../miniperl "-I../../../lib" "-I../../../lib"
    ../../../lib/ExtUtils/xsubpp -nolinenumbers -typemap
    ../../../lib/ExtUtils/typemap TW.xs > TW.xsc && mv TW.xsc TW.c
    xlc -c -I../Encode -DDEBUGGING -D_ALL_SOURCE -D_ANSI_C_SOURCE
    -D_POSIX_SOURCE -qmaxmem=16384 -qnoansialias -I/pro/local/include -O2
    -DVERSION=\"1.25\" -DXS_VERSION=\"1.25\" "-I../../.." TW.c
    xlc -c -I../Encode -DDEBUGGING -D_ALL_SOURCE -D_ANSI_C_SOURCE
    -D_POSIX_SOURCE -qmaxmem=16384 -qnoansialias -I/pro/local/include -O2
    -DVERSION=\"1.25\" -DXS_VERSION=\"1.25\" "-I../../.." big5_t.c
    xlc: 1501-229 Compilation ended due to lack of space
    make[2]: *** [big5_t.o] Error 68
    make[2]: Leaving directory `/pro/3gl/CPAN/perl-current/ext/Encode/TW'
    make[1]: *** [subdirs] Error 2
    make[1]: Leaving directory `/pro/3gl/CPAN/perl-current/ext/Encode'
    make: *** [lib/auto/Encode/Encode.so] Error 2
    ibm:/pro/3gl/CPAN/perl-current 103 >
    It occurs to me that your computer is simply out of memory. "due to
    lack of space" ?

    Dan the Encode Maintainer
  • H.Merijn Brand at Apr 24, 2002 at 12:33 pm

    On Wed 24 Apr 2002 13:56, Dan Kogai wrote:
    On Wednesday, April 24, 2002, at 08:44 , H.Merijn Brand wrote:
    Reading big5-eten (big5-eten)
    Reading big5-hkscs (big5-hkscs)
    Reading cp950 (cp950)
    Reading MacChineseTrad (MacChineseTrad)
    Writing compiled form
    102322 bytes in string tables
    410276 bytes (20%) saved spotting duplicates
    ../../../miniperl "-I../../../lib" "-I../../../lib"
    ../../../lib/ExtUtils/xsubpp -nolinenumbers -typemap
    ../../../lib/ExtUtils/typemap TW.xs > TW.xsc && mv TW.xsc TW.c
    xlc -c -I../Encode -DDEBUGGING -D_ALL_SOURCE -D_ANSI_C_SOURCE
    -D_POSIX_SOURCE -qmaxmem=16384 -qnoansialias -I/pro/local/include -O2
    -DVERSION=\"1.25\" -DXS_VERSION=\"1.25\" "-I../../.." TW.c
    xlc -c -I../Encode -DDEBUGGING -D_ALL_SOURCE -D_ANSI_C_SOURCE
    -D_POSIX_SOURCE -qmaxmem=16384 -qnoansialias -I/pro/local/include -O2
    -DVERSION=\"1.25\" -DXS_VERSION=\"1.25\" "-I../../.." big5_t.c
    xlc: 1501-229 Compilation ended due to lack of space
    make[2]: *** [big5_t.o] Error 68
    make[2]: Leaving directory `/pro/3gl/CPAN/perl-current/ext/Encode/TW'
    make[1]: *** [subdirs] Error 2
    make[1]: Leaving directory `/pro/3gl/CPAN/perl-current/ext/Encode'
    make: *** [lib/auto/Encode/Encode.so] Error 2
    ibm:/pro/3gl/CPAN/perl-current 103 >
    It occurs to me that your computer is simply out of memory. "due to
    lack of space" ?
    Optimizer issue, but still a show stopper

    ibm:/pro/3gl/CPAN/perl-current/ext/Encode/TW 116 > xlc -c -I../Encode -DDEBUGGING -D_ALL_SOURCE -D_ANSI_C_SOURCE -D_POSIX_SOURCE -qmaxmem=16384 -qnoansialias -I/pro/local/include -O2 -DVERSION=\"1.25\" -DXS_VERSION=\"1.25\" "-I../../.." big5_t.c
    xlc: 1501-229 Compilation ended due to lack of space
    ibm:/pro/3gl/CPAN/perl-current/ext/Encode/TW 117 > xlc -c -I../Encode -DDEBUGGING -D_ALL_SOURCE -D_ANSI_C_SOURCE -D_POSIX_SOURCE -qmaxmem=16384000 -qnoansialias -I/pro/local/include -O2 -DVERSION=\"1.25\" -DXS_VERSION=\"1.25\" "-I../../.." big5_t.c
    xlc: 1501-229 Compilation ended due to lack of space
    ibm:/pro/3gl/CPAN/perl-current/ext/Encode/TW 118 > xlc -c -I../Encode -DDEBUGGING -D_ALL_SOURCE -D_ANSI_C_SOURCE -D_POSIX_SOURCE -qmaxmem=16384 -qnoansialias -I/pro/local/include -DVERSION=\"1.25\" -DXS_VERSION=\"1.25\" "-I../../.." big5_t.c
    ibm:/pro/3gl/CPAN/perl-current/ext/Encode/TW 119 >

    If you can't come up with a generic change, we must hint this file to not
    optimize on AIX-4.2/xlc (preferably in the Encode section)
    Dan the Encode Maintainer
    --
    H.Merijn Brand Amsterdam Perl Mongers (http://amsterdam.pm.org/)
    using perl-5.6.1, 5.7.3 & 631 on HP-UX 10.20 & 11.00, AIX 4.2, AIX 4.3,
    WinNT 4, Win2K pro & WinCE 2.11. Smoking perl CORE: smokers@perl.org
    http://archives.develooper.com/daily-build@perl.org/ perl-qa@perl.org
    send smoke reports to: smokers-reports@perl.org, QA: http://qa.perl.org
  • Jarkko Hietaniemi at Apr 24, 2002 at 1:47 pm
    I think also Laszlo reported optimizer problems with Encode. Would it
    be possible somehow to split the large tables?

    Optimizers tend to get overly "excited" about large pieces of code,
    and splitting files by a factor of N might not only make the
    optimization succeed but actually speed up the total time spent in
    compilation by more than the factor of N.

    --
    $jhi++; # http://www.iki.fi/jhi/
    # There is this special biologist word we use for 'stable'.
    # It is 'dead'. -- Jack Cohen
  • Dan Kogai at Apr 24, 2002 at 2:24 pm

    On Wednesday, April 24, 2002, at 10:47 , Jarkko Hietaniemi wrote:
    I think also Laszlo reported optimizer problems with Encode. Would it
    be possible somehow to split the large tables?
    That can be done by 'deaggregating' similar encodings, like SHIFT_JIS
    and MacJapanese. However, that results in larger loadable object.

    I can add a hook to Makefile.PL so that when it is invoked it splits.
    environment variable or $Config{} would be ok. Then you can add the
    entry to the hints file. I'll experiment....

    Dan
  • Jarkko Hietaniemi at Apr 24, 2002 at 2:28 pm

    On Wed, Apr 24, 2002 at 11:24:45PM +0900, Dan Kogai wrote:
    On Wednesday, April 24, 2002, at 10:47 , Jarkko Hietaniemi wrote:
    I think also Laszlo reported optimizer problems with Encode. Would it
    be possible somehow to split the large tables?
    That can be done by 'deaggregating' similar encodings, like SHIFT_JIS
    and MacJapanese. However, that results in larger loadable object.
    Huh, how does splitting the source code files result in larger
    loadable objects?
    I can add a hook to Makefile.PL so that when it is invoked it splits.
    environment variable or $Config{} would be ok. Then you can add the
    entry to the hints file. I'll experiment....
    Uhh, no. Too messy. Split always.
    Dan
    --
    $jhi++; # http://www.iki.fi/jhi/
    # There is this special biologist word we use for 'stable'.
    # It is 'dead'. -- Jack Cohen
  • Dan Kogai at Apr 24, 2002 at 2:39 pm

    On Wednesday, April 24, 2002, at 11:28 , Jarkko Hietaniemi wrote:
    On Wed, Apr 24, 2002 at 11:24:45PM +0900, Dan Kogai wrote:
    On Wednesday, April 24, 2002, at 10:47 , Jarkko Hietaniemi wrote:
    I think also Laszlo reported optimizer problems with Encode. Would it
    be possible somehow to split the large tables?
    That can be done by 'deaggregating' similar encodings, like SHIFT_JIS
    and MacJapanese. However, that results in larger loadable object.
    Huh, how does splitting the source code files result in larger
    loadable objects?
    Here is how it works.

    JP/Makefile.PL
    my %tables = (
    euc_jp_t => ['euc-jp.ucm'],
    sjis_t => ['shiftjis.ucm',
    'macJapanese.ucm',
    'cp932.ucm'],
    raw_t => [
    qw(jis0201.ucm jis0208.ucm jis0212.ucm)
    ],
    );

    When 'perl Makefile.PL' is invoked, it creates *_t.fnm files that looks
    like this

    .../ucm/shiftjis.ucm
    .../ucm/macJapanese.ucm
    .../ucm/cp932.ucm

    then during the 'make' phase, enc2xs tries to creat *_t.c out of ucm
    files listed here. in doing so, it tries to "optimize" the resulting
    *.c by eliminating duplicate entries therein. the idea of splitting is
    to let these *.fnm have only one ucm file entry. So the resulting *.c
    is smaller but the resulting *.so (or *.dylib in MacOS X or whatever)
    may get bloated.

    Dan
  • Jarkko Hietaniemi at Apr 24, 2002 at 2:47 pm

    Here is how it works.

    JP/Makefile.PL
    my %tables = (
    euc_jp_t => ['euc-jp.ucm'],
    sjis_t => ['shiftjis.ucm',
    'macJapanese.ucm',
    'cp932.ucm'],
    raw_t => [
    qw(jis0201.ucm jis0208.ucm jis0212.ucm)
    ],
    );

    When 'perl Makefile.PL' is invoked, it creates *_t.fnm files that looks
    like this

    ../ucm/shiftjis.ucm
    ../ucm/macJapanese.ucm
    ../ucm/cp932.ucm

    then during the 'make' phase, enc2xs tries to creat *_t.c out of ucm
    files listed here. in doing so, it tries to "optimize" the resulting
    *.c by eliminating duplicate entries therein. the idea of splitting is
    to let these *.fnm have only one ucm file entry. So the resulting *.c
    is smaller but the resulting *.so (or *.dylib in MacOS X or whatever)
    may get bloated.
    Okay, now I understand. So the total size of the *.so might get
    bigger. However...

    (1) with the smaller *.c we might actually get as far the *.so
    (2) with the smaller individual *.so, well, the individual *.so
    are smaller... that is, not only in their disk but memory
    footprint (instead of loading all of shiftjis, macJapanese,
    and cp932), one can separately load what one really needs.
    Dan
    --
    $jhi++; # http://www.iki.fi/jhi/
    # There is this special biologist word we use for 'stable'.
    # It is 'dead'. -- Jack Cohen
  • Dan Kogai at Apr 24, 2002 at 2:59 pm

    On Wednesday, April 24, 2002, at 11:46 , Jarkko Hietaniemi wrote:
    Okay, now I understand. So the total size of the *.so might get
    bigger. However...

    (1) with the smaller *.c we might actually get as far the *.so
    (2) with the smaller individual *.so, well, the individual *.so
    are smaller... that is, not only in their disk but memory
    footprint (instead of loading all of shiftjis, macJapanese,
    and cp932), one can separately load what one really needs.
    For part 2, I once tought of giving EACH encoding an .so but gave up for
    the time being because makes the build process too complicated with a
    given infrastructure of current Encode. Please make it 5.8.1 todo (I
    also have even more radical idea on that but that's little bit too late
    for 5.8.0).

    But part 1 is much easier to achieve; All I have to do is revise
    XX/Makefile.PL. I'll benchmark them first then if it goes well, I'll go
    for it.

    I think Bytes/EBCDIC/Symbol should all remain as is (they are ALL
    aggregated within the module). It's amasing that with all Bytes
    combined it's only 1/5 of shiftjis alone!

    Dan
  • Jarkko Hietaniemi at Apr 24, 2002 at 3:05 pm

    On Wed, Apr 24, 2002 at 11:59:17PM +0900, Dan Kogai wrote:
    On Wednesday, April 24, 2002, at 11:46 , Jarkko Hietaniemi wrote:
    Okay, now I understand. So the total size of the *.so might get
    bigger. However...

    (1) with the smaller *.c we might actually get as far the *.so
    (2) with the smaller individual *.so, well, the individual *.so
    are smaller... that is, not only in their disk but memory
    footprint (instead of loading all of shiftjis, macJapanese,
    and cp932), one can separately load what one really needs.
    For part 2, I once tought of giving EACH encoding an .so but gave up for
    the time being because makes the build process too complicated with a
    given infrastructure of current Encode. Please make it 5.8.1 todo (I
    also have even more radical idea on that but that's little bit too late
    for 5.8.0).

    But part 1 is much easier to achieve; All I have to do is revise
    XX/Makefile.PL. I'll benchmark them first then if it goes well, I'll go
    for it.

    I think Bytes/EBCDIC/Symbol should all remain as is (they are ALL
    aggregated within the module). It's amasing that with all Bytes
    combined it's only 1/5 of shiftjis alone!
    Here's wc of the *.c files. If I were an optimizer I would start
    choking at a 2.7 MiB file...

    326688 757100 14292826 total
    60577 136014 2684619 TW/big5_t.c
    41023 81978 1815019 KR/euc_kr_t.c
    31715 62617 1551838 CN/euc_cn_t.c
    30442 62913 1379481 JP/sjis_t.c
    30404 82087 1232811 JP/euc_jp_t.c
    29475 75370 1294474 JP/raw_t.c
    21260 50560 882848 KR/johab_t.c
    20289 53276 865635 KR/5601_t.c
    17229 45572 744629 CN/12345_t.c
    15548 40066 666179 CN/ir_165_t.c
    15052 39042 641383 CN/2312_t.c
    9632 18298 400201 Byte/byte_t.c
    1376 2573 56002 Symbol/symbol_t.c
    941 1985 37573 EBCDIC/ebcdic_t.c
    775 2175 18146 Encode.c
    305 905 7234 Unicode/Unicode.c
    153 667 3905 encengine.c
    79 148 3468 def_t.c
    63 130 1139 CN/CN.c
    61 126 1093 KR/KR.c
    61 126 1089 JP/JP.c
    57 118 1023 Symbol/Symbol.c
    57 118 1023 EBCDIC/EBCDIC.c
    57 118 1011 Byte/Byte.c
    57 118 1003 TW/TW.c
    Dan
    --
    $jhi++; # http://www.iki.fi/jhi/
    # There is this special biologist word we use for 'stable'.
    # It is 'dead'. -- Jack Cohen
  • Nicholas Clark at Apr 24, 2002 at 3:16 pm

    On Wed, Apr 24, 2002 at 06:05:12PM +0300, Jarkko Hietaniemi wrote:
    On Wed, Apr 24, 2002 at 11:59:17PM +0900, Dan Kogai wrote:

    I think Bytes/EBCDIC/Symbol should all remain as is (they are ALL
    aggregated within the module). It's amasing that with all Bytes
    combined it's only 1/5 of shiftjis alone!
    Here's wc of the *.c files. If I were an optimizer I would start
    choking at a 2.7 MiB file...

    326688 757100 14292826 total
    60577 136014 2684619 TW/big5_t.c
    41023 81978 1815019 KR/euc_kr_t.c
    31715 62617 1551838 CN/euc_cn_t.c
    30442 62913 1379481 JP/sjis_t.c
    30404 82087 1232811 JP/euc_jp_t.c
    29475 75370 1294474 JP/raw_t.c
    21260 50560 882848 KR/johab_t.c
    20289 53276 865635 KR/5601_t.c
    17229 45572 744629 CN/12345_t.c
    15548 40066 666179 CN/ir_165_t.c
    15052 39042 641383 CN/2312_t.c
    9632 18298 400201 Byte/byte_t.c
    1376 2573 56002 Symbol/symbol_t.c
    941 1985 37573 EBCDIC/ebcdic_t.c
    775 2175 18146 Encode.c
    305 905 7234 Unicode/Unicode.c
    153 667 3905 encengine.c
    79 148 3468 def_t.c
    63 130 1139 CN/CN.c
    61 126 1093 KR/KR.c
    61 126 1089 JP/JP.c
    57 118 1023 Symbol/Symbol.c
    57 118 1023 EBCDIC/EBCDIC.c
    57 118 1011 Byte/Byte.c
    57 118 1003 TW/TW.c
    It looks like it would be good enough just to split things somewhat in the
    {CN,JP,KR,TW} directories.

    I've been thinking about another optimiser idea for a bit - rather than the
    current optimiser techniques of (1) looking for exact duplicate strings
    (2) looking for substrings I was wondering what happens if the encode
    compiler simply aggregated all the strings into a single huge common
    character array, and made every other pointer into a reference within it.

    Effectively it's the (2) optimiser looking for substrings, and if you don't
    find your string as a substring you tack it onto the end and then "find" it
    there.

    This may well reduce the size of the .c file, and the .so file, and give the
    optimiser less ways of getting its knickers in a twist.

    Nicholas Clark
  • Dan Kogai at Apr 24, 2002 at 4:19 pm

    On Thursday, April 25, 2002, at 12:16 , Nicholas Clark wrote:
    It looks like it would be good enough just to split things somewhat in
    the
    {CN,JP,KR,TW} directories.
    I just found that splitting *.c for each *.ucm for CJKT has worked
    well. Here is before and after the split. On FreeBSD 4.5 with
    breadperl compiled with '-g -O2' option. So in practice, it is even
    smaller (it is not likely that ports version in future would use "-g".
    It also strips every binary on install). And it seems that it compiles
    slightly faster!

    Before
    344 blib/arch/auto/Encode/Byte/Byte.so*
    2784 blib/arch/auto/Encode/CN/CN.so*
    84 blib/arch/auto/Encode/EBCDIC/EBCDIC.so*
    90 blib/arch/auto/Encode/Encode.so*
    3208 blib/arch/auto/Encode/JP/JP.so*
    2752 blib/arch/auto/Encode/KR/KR.so*
    96 blib/arch/auto/Encode/Symbol/Symbol.so*
    2120 blib/arch/auto/Encode/TW/TW.so*
    63 blib/arch/auto/Encode/Unicode/Unicode.so* And After
    344 blib/arch/auto/Encode/Byte/Byte.so*
    3152 blib/arch/auto/Encode/CN/CN.so*
    84 blib/arch/auto/Encode/EBCDIC/EBCDIC.so*
    90 blib/arch/auto/Encode/Encode.so*
    4096 blib/arch/auto/Encode/JP/JP.so*
    3544 blib/arch/auto/Encode/KR/KR.so*
    96 blib/arch/auto/Encode/Symbol/Symbol.so*
    3048 blib/arch/auto/Encode/TW/TW.so*
    63 blib/arch/auto/Encode/Unicode/Unicode.so*
    New version will be released in a few hours.

    Dan the Encode Maintainer
  • Nick Ing-Simmons at Apr 24, 2002 at 7:14 pm

    Jarkko Hietaniemi writes:
    Here's wc of the *.c files. If I were an optimizer I would start
    choking at a 2.7 MiB file...

    326688 757100 14292826 total
    60577 136014 2684619 TW/big5_t.c
    Look at that file. There is no code in it all it is all data.
    An optimizer has no business messing with it ;-)

    --
    Nick Ing-Simmons
    http://www.ni-s.u-net.com/
  • Jarkko Hietaniemi at Apr 24, 2002 at 7:21 pm

    On Wed, Apr 24, 2002 at 08:10:03PM +0100, Nick Ing-Simmons wrote:
    Jarkko Hietaniemi <jhi@iki.fi> writes:
    Here's wc of the *.c files. If I were an optimizer I would start
    choking at a 2.7 MiB file...

    326688 757100 14292826 total
    60577 136014 2684619 TW/big5_t.c
    Look at that file. There is no code in it all it is all data.
    An optimizer has no business messing with it ;-)
    I don't really care if it's the optimizer or the black cat of Bill
    Gates that is messing with it. If splitting them up helps Merijn's
    and Laszlo's compilers, it's good.
    --
    Nick Ing-Simmons
    http://www.ni-s.u-net.com/
    --
    $jhi++; # http://www.iki.fi/jhi/
    # There is this special biologist word we use for 'stable'.
    # It is 'dead'. -- Jack Cohen
  • Jarkko Hietaniemi at Apr 24, 2002 at 7:50 pm

    On Wed, Apr 24, 2002 at 10:20:54PM +0300, Jarkko Hietaniemi wrote:
    On Wed, Apr 24, 2002 at 08:10:03PM +0100, Nick Ing-Simmons wrote:
    Jarkko Hietaniemi <jhi@iki.fi> writes:
    Here's wc of the *.c files. If I were an optimizer I would start
    choking at a 2.7 MiB file...

    326688 757100 14292826 total
    60577 136014 2684619 TW/big5_t.c
    Look at that file. There is no code in it all it is all data.
    An optimizer has no business messing with it ;-)
    I don't really care if it's the optimizer or the black cat of Bill
    Gates that is messing with it. If splitting them up helps Merijn's
    and Laszlo's compilers, it's good.
    That being said, we don't yet have positive proof on that the
    splitting will help. My guess is that it will, but then again,
    I guessed that 5.8 would have been released aeons ago...

    --
    $jhi++; # http://www.iki.fi/jhi/
    # There is this special biologist word we use for 'stable'.
    # It is 'dead'. -- Jack Cohen
  • Dan Kogai at Apr 24, 2002 at 8:56 pm

    On Thursday, April 25, 2002, at 04:50 , Jarkko Hietaniemi wrote:
    That being said, we don't yet have positive proof on that the
    splitting will help.
    I am not sure of the compiler but enc2xs needs significantly more memory
    to generate *.c file when it attempts to aggregate *.ucm. That may have
    hurt low-memory machines as well. That reminds me; How much memory do
    we say is enough to build a perl?

    Among the machines at my fingertip with compilers, ThinkPad i1620 (X21
    done wrong) has the leaset with 128MB. Never encountered
    memory-deficiency kind of smoke with Cygwin, not to mention FreeBSD. My
    main toolbox machine, with the same CPU as X-BOX, has 512 MB installed
    (I usually "make -j8 buildworld" to get the most out of it) and so does
    my, um, our iMac (available only when my kids are asleep; I give 128MB
    for Virtual PC). Boy, I must have been spoiled recently....
    My guess is that it will, but then again,
    I guessed that 5.8 would have been released aeons ago...
    One good side is that 5.8 is more like what we have hoped Perl5 to be --
    Storable that stores utf8 keys and restricted hashes, Test::More without
    resorting to CPAN, and *working* Encode with supported encodings and
    options magnitudes more than any other utilities....

    Dan the Perl5 Porter
  • Nicholas Clark at Apr 25, 2002 at 1:09 pm

    On Thu, Apr 25, 2002 at 05:56:24AM +0900, Dan Kogai wrote:
    On Thursday, April 25, 2002, at 04:50 , Jarkko Hietaniemi wrote:
    That being said, we don't yet have positive proof on that the
    splitting will help.
    I am not sure of the compiler but enc2xs needs significantly more memory
    to generate *.c file when it attempts to aggregate *.ucm. That may have
    hurt low-memory machines as well. That reminds me; How much memory do
    we say is enough to build a perl?
    My FreeBSD box does it with 16M.
    It has got about 50M swap, but I'm not sure how much of that it needs.
    -O3 will work, but the perl that produces actually seems to be slower.

    Big switch statements are bad. POSIX.xs no longer has one. The most recent
    big switch statement bust up was in parrot's core ops - perl hasn't taken
    out any machine I've used for a while.

    (http://bugs.perl.org/_perlbug.cgi?req=bug_id&bug_id=20001126.002&format=H&trim=25&range=953
    being a bug report for constant in POSIX.xs for 5.7.0 taking out gcc on x86
    Linux box with 64M RAM)

    Nicholas Clark
  • Nick Ing-Simmons at Apr 24, 2002 at 7:06 pm

    Jarkko Hietaniemi writes:
    On Wed, Apr 24, 2002 at 11:24:45PM +0900, Dan Kogai wrote:
    On Wednesday, April 24, 2002, at 10:47 , Jarkko Hietaniemi wrote:
    I think also Laszlo reported optimizer problems with Encode. Would it
    be possible somehow to split the large tables?
    That can be done by 'deaggregating' similar encodings, like SHIFT_JIS
    and MacJapanese. However, that results in larger loadable object.
    Huh, how does splitting the source code files result in larger
    loadable objects?
    As an example consider western encodings ASCII and iso-8859-*
    All the entries for 0x20..0x7e att least are the same so the data structures
    only occur once and are shared between all the encodings.
    Other clumps occur too.

    The same is true for related Asian encodings, the regularity may not be
    as high, but the data structures "saved" are bigger
    Nick C spent some time optimizing our "data optimizer".

    I can add a hook to Makefile.PL so that when it is invoked it splits.
    environment variable or $Config{} would be ok. Then you can add the
    entry to the hints file. I'll experiment....
    Uhh, no. Too messy. Split always.
    Dan
    --
    Nick Ing-Simmons
    http://www.ni-s.u-net.com/
  • Jarkko Hietaniemi at Apr 24, 2002 at 7:18 pm

    As an example consider western encodings ASCII and iso-8859-*
    All the entries for 0x20..0x7e att least are the same so the data structures
    only occur once and are shared between all the encodings.
    Other clumps occur too.

    The same is true for related Asian encodings, the regularity may not be
    as high, but the data structures "saved" are bigger
    Nick C spent some time optimizing our "data optimizer".
    Data optimizer good. (Ab)using it to create mega(lomaniac) source
    code files that make compilers choke bad.

    I claim that splitting up the CJK a bit in fact makes to memory use
    *smaller* since we will be loading only, say, big5-hsckcs into memory,
    instead of all four of big5-eten, big5-hskcs, MacChinTrad, and cp950.
    In normal use people are more likely to use only one or maybe two of
    them, not all of them at the same time.

    --
    $jhi++; # http://www.iki.fi/jhi/
    # There is this special biologist word we use for 'stable'.
    # It is 'dead'. -- Jack Cohen
  • Dan Kogai at Apr 24, 2002 at 8:29 pm
    I am determined to extinguish the smoke when I released Encode 1.60,
    available as follows;

    Whole:
    http://www.dan.co.jp/~dankogai/Encode-1.60.tar.gz
    and CPAN
    Diff against current (932 lines):
    http://www.dan.co.jp/~dankogai/current-1.60.diff.gz

    Here is the summery of Changes. Detailed Changes right after the sig.

    * For CJKT encodings, now *.c is created for each *.ucm files. Though
    it makes the resulting shared object 30-40% larger, it should reduce the
    memory required to build so hopefully "not enough memory" type of smokes
    are gone with this
    * needs_lines() and perlio_ok() are explicitly defined in all encoding
    objects that come with Encode. Encode::Encodings are there so you can
    "use base" that but I wanted them explicitly defined for all that comes
    standard.
    * Two more fallback modes for Encode::XS
    * 'Guess' decoding
    * MIME header transcoding.
    The above two were the only missing feature that Jcode had.

    Dan the Encode Maintainer

    $Revision: 1.60 $ $Date: 2002/04/24 20:06:52 $
    ! Encode.xs
    "Thou shalt not assume %x works." -- jhi
    Message-Id: <20020424210618.E24347@alpha.hut.fi>
    ! CN/Makefile.PL JP/Makefile.PL KR/Makefile.PL TW/Makefile.PL To make
    low-memory build machines happy, now *.c is created for each *.ucm
    (no table aggregation). You can still override this by setting
    $ENV{AGGREGATE_TABLES}.
    Message-Id: <00B1B3E4-579F-11D6-A441-00039301D480@dan.co.jp>
    + lib/Encode/Guess.pm
    + lib/Encode/JP/JIS7.pm
    Encoding-autodetect (mainly for Japanese encoding) added. In a
    course of development, JIS7.pm was improved.
    + lib/Encode/HTML/Header.pm
    + lib/Encode/Config.pm
    MIME B/Q Header Encoding Added!
    ! Encode.pm Encode.xs t/fallback.t
    new fallbacks; XMLCREF and HTMLCREF upon Bart's request.
    Message-Id: <20020424130709.GA14211@tanglefoot>

    1.59 $ 2002/04/22 23:54:22
    ! Encode.pm Encode.xs
    needs_lines() and perlio_ok() are added to Internal encodings such
    as utf8 so XML::SAX is happy. FB_* stub xsubs are now prototyped.
  • Nick Ing-Simmons at Apr 26, 2002 at 6:37 pm

    Dan Kogai writes:
    * needs_lines() and perlio_ok() are explicitly defined in all encoding
    objects that come with Encode. Encode::Encodings are there so you can
    "use base" that but I wanted them explicitly defined for all that comes
    standard.
    So long as the 'use base' AS WELL I don't mind if they have their own.
    But if they don't have a base class then every time we tweak the
    Encoding API we have to go and add methods to every encoding class :-(


    --
    Nick Ing-Simmons
    http://www.ni-s.u-net.com/
  • Jarkko Hietaniemi at Apr 24, 2002 at 2:27 pm

    On Wed, Apr 24, 2002 at 04:47:21PM +0300, Jarkko Hietaniemi wrote:
    I think also Laszlo reported optimizer problems with Encode. Would it
    be possible somehow to split the large tables?

    Optimizers tend to get overly "excited" about large pieces of code,
    and splitting files by a factor of N might not only make the
    optimization succeed but actually speed up the total time spent in
    compilation by more than the factor of N.
    As a very simple "N" we could have compile big5-eten, big5-hskcs,
    MacChinTrad, and cp950 as separate files. Ditto for the other CJK.

    --
    $jhi++; # http://www.iki.fi/jhi/
    # There is this special biologist word we use for 'stable'.
    # It is 'dead'. -- Jack Cohen
  • Nick Ing-Simmons at Apr 24, 2002 at 7:01 pm

    Jarkko Hietaniemi writes:
    I think also Laszlo reported optimizer problems with Encode. Would it
    be possible somehow to split the large tables?

    Optimizers tend to get overly "excited" about large pieces of code,
    and splitting files by a factor of N might not only make the
    optimization succeed but actually speed up the total time spent in
    compilation by more than the factor of N.
    A. They are no code they are data.
    B. If you split 'em up they use more memory at run time

    --
    Nick Ing-Simmons
    http://www.ni-s.u-net.com/
  • Jarkko Hietaniemi at Apr 24, 2002 at 7:04 pm

    On Wed, Apr 24, 2002 at 08:00:17PM +0100, Nick Ing-Simmons wrote:
    Jarkko Hietaniemi <jhi@iki.fi> writes:
    I think also Laszlo reported optimizer problems with Encode. Would it
    be possible somehow to split the large tables?

    Optimizers tend to get overly "excited" about large pieces of code,
    and splitting files by a factor of N might not only make the
    optimization succeed but actually speed up the total time spent in
    compilation by more than the factor of N.
    A. They are no code they are data.
    A. Tell that to the compilers, not to me.
    B. If you split 'em up they use more memory at run time
    B. If we don't split 'em up there will be no run time.
    --
    Nick Ing-Simmons
    http://www.ni-s.u-net.com/
    --
    $jhi++; # http://www.iki.fi/jhi/
    # There is this special biologist word we use for 'stable'.
    # It is 'dead'. -- Jack Cohen
  • H.Merijn Brand at Apr 24, 2002 at 12:47 pm

    On Wed 24 Apr 2002 13:44, "H.Merijn Brand" wrote:
    1. Encode causes AIX 4.2 't's, see below

    Automated smoke report for patch 16109 cc gcc
    Cygwin@16125 :

    /lib/Env/t/env.....................ok
    /lib/Exporter......................ok
    /lib/ExtUtils/t/basic..............# Failed test (../lib/ExtUtils/t/basic.t
    at line 32)
    /lib/ExtUtils/t/basic..............NOK 1# chdir failed: No such file or director
    y
    Can't open Liar/t/sanity.t: No such file or directory at ../lib/ExtUtils/t/basic
    .t line 49.
    # Looks like you planned 17 tests but only ran 1.
    # Looks like your test died just after 1.
    /lib/ExtUtils/t/basic..............dubious


    Failed Test Stat Wstat Total Fail Failed List of Failed
    -------------------------------------------------------------------------------
    ../lib/ExtUtils/t/Command.t 2 512 ?? ?? % ??
    ../lib/ExtUtils/t/Constant.t 2 512 ?? ?? % ??
    ../lib/ExtUtils/t/Embed.t 2 512 ?? ?? % ??
    ../lib/ExtUtils/t/INST.t 2 512 ?? ?? % ??
    ../lib/ExtUtils/t/INST_PREFIX.t 2 512 ?? ?? % ??
    ../lib/ExtUtils/t/Installed.t 2 512 ?? ?? % ??
    ../lib/ExtUtils/t/MM_BeOS.t 2 512 ?? ?? % ??
    ../lib/ExtUtils/t/MM_Cygwin.t 2 512 ?? ?? % ??
    ../lib/ExtUtils/t/MM_OS2.t 2 512 ?? ?? % ??
    ../lib/ExtUtils/t/MM_Unix.t 2 512 ?? ?? % ??
    ../lib/ExtUtils/t/MM_VMS.t 2 512 ?? ?? % ??
    ../lib/ExtUtils/t/MM_Win32.t 2 512 ?? ?? % ??
    ../lib/ExtUtils/t/Manifest.t 4 1024 33 4 12.12% 15 26-28
    ../lib/ExtUtils/t/Mkbootstrap.t 2 512 ?? ?? % ??
    ../lib/ExtUtils/t/Packlist.t 2 512 ?? ?? % ??
    ../lib/ExtUtils/t/basic.t 255 65280 17 17 100.00% 1-17
    ../lib/ExtUtils/t/hints.t 2 512 ?? ?? % ??
    ../lib/ExtUtils/t/prefixify.t 2 512 ?? ?? % ??
    ../lib/ExtUtils/t/problems.t 2 512 ?? ?? % ??
    ../lib/ExtUtils/t/testlib.t 2 512 ?? ?? % ??
    ../lib/Fatal.t 2 512 15 15 100.00% 1-15
    ../lib/File/Basename.t 2 512 ?? ?? % ??
    ../lib/File/CheckTree.t 2 512 ?? ?? % ??
    ../lib/File/Copy.t 2 512 ?? ?? % ??
    ../lib/File/DosGlob.t 2 512 ?? ?? % ??
    ../lib/File/Path.t 2 512 ?? ?? % ??
    ../lib/File/Spec/t/Functions.t 2 512 ?? ?? % ??
    ../lib/File/Spec/t/Spec.t 2 512 ?? ?? % ??
    ../lib/File/Spec/t/rel2abs2rel.t 2 512 ?? ?? % ??
    ../lib/File/Temp/t/mktemp.t 2 512 ?? ?? % ??
    ../lib/File/Temp/t/posix.t 2 512 ?? ?? % ??
    ../lib/File/Temp/t/security.t 2 512 ?? ?? % ??
    ../lib/File/Temp/t/tempfile.t 2 512 ?? ?? % ??
    ../lib/File/stat.t 2 512 ?? ?? % ??
    ../lib/FileCache.t 2 512 ?? ?? % ??
    ../lib/FileHandle.t 2 512 ?? ?? % ??
    ../lib/Filter/Simple/t/data.t 2 512 ?? ?? % ??
    ../lib/Filter/Simple/t/export.t 2 512 1 1 100.00% 1
    ../lib/Filter/Simple/t/filter.t 2 512 ?? ?? % ??
    ../lib/Filter/Simple/t/filter_onl 2 512 ?? ?? % ??
    ../lib/Filter/Simple/t/import.t 2 512 4 4 100.00% 1-4
    ../lib/FindBin.t 2 512 ?? ?? % ??
    ../lib/Getopt/Long/t/gol-basic.t 2 512 ?? ?? % ??
    ../lib/Getopt/Long/t/gol-compat.t 2 512 ?? ?? % ??
    ../lib/Getopt/Long/t/gol-linkage. 2 512 ?? ?? % ??
    ../lib/Getopt/Long/t/gol-oo.t 2 512 ?? ?? % ??
    ../lib/Getopt/Std.t 2 512 ?? ?? % ??
    ../lib/Hash/Util.t 2 512 ?? ?? % ??
    ../lib/I18N/Collate.t 2 512 ?? ?? % ??
    ../lib/IPC/Open2.t 2 512 ?? ?? % ??
    ../lib/IPC/Open3.t 2 512 ?? ?? % ??
    ../lib/IPC/SysV.t 2 512 ?? ?? % ??
    ../lib/Internals.t 2 512 ?? ?? % ??
    ../lib/Locale/Codes/t/all.t 2 512 ?? ?? % ??
    ../lib/Locale/Codes/t/constants.t 2 512 ?? ?? % ??
    ../lib/Locale/Codes/t/country.t 2 512 ?? ?? % ??
    ../lib/Locale/Codes/t/currency.t 2 512 ?? ?? % ??
    ../lib/Locale/Codes/t/languages.t 2 512 ?? ?? % ??
    ?Failed Test Stat Wstat Total Fail Failed List of Failed

    -------------------------------------------------------------------------------
    ../lib/Locale/Codes/t/uk.t 2 512 ?? ?? % ??
    ../lib/Math/BigInt/t/bare_mbf.t 2 512 ?? ?? % ??
    ../lib/Math/BigInt/t/bare_mbi.t 2 512 ?? ?? % ??
    ../lib/Math/BigInt/t/bigfltpm.t 2 512 ?? ?? % ??
    ../lib/Math/BigInt/t/bigintpm.t 2 512 2237 2237 100.00% 1-2237
    ../lib/Math/BigInt/t/calling.t 2 512 ?? ?? % ??
    ../lib/Math/BigInt/t/isa.t 2 512 ?? ?? % ??
    ../lib/Math/BigInt/t/mbimbf.t 2 512 ?? ?? % ??
    ../lib/Math/BigInt/t/require.t 2 512 ?? ?? % ??
    ../lib/Math/BigInt/t/sub_mbf.t 2 512 ?? ?? % ??
    ../lib/Math/BigInt/t/sub_mbi.t 2 512 ?? ?? % ??
    ../lib/Math/BigInt/t/sub_mif.t 2 512 ?? ?? % ??
    ../lib/Math/BigInt/t/upgrade.t 2 512 ?? ?? % ??
    ../lib/Math/BigInt/t/use.t 2 512 ?? ?? % ??
    ../lib/Math/BigInt/t/use_lib1.t 2 512 ?? ?? % ??
    ../lib/Math/BigInt/t/use_lib2.t 2 512 ?? ?? % ??
    ../lib/Math/BigInt/t/use_lib3.t 2 512 ?? ?? % ??
    ../lib/Math/BigInt/t/use_lib4.t 2 512 ?? ?? % ??
    ../lib/Math/BigInt/t/with_sub.t 2 512 ?? ?? % ??
    ../lib/Math/BigRat/t/bigfltrt.t 2 512 ?? ?? % ??
    ../lib/Math/BigRat/t/bigratpm.t 2 512 ?? ?? % ??
    ../lib/Math/Complex.t 2 512 ?? ?? % ??
    ../lib/Math/Trig.t 2 512 ?? ?? % ??
    ../lib/Memoize/t/array.t 2 512 ?? ?? % ??
    ../lib/Memoize/t/array_confusion. 2 512 ?? ?? % ??
    ../lib/Memoize/t/correctness.t 2 512 ?? ?? % ??
    ../lib/Memoize/t/errors.t 2 512 ?? ?? % ??
    ../lib/Memoize/t/expire.t 2 512 ?? ?? % ??
    ../lib/Memoize/t/flush.t 2 512 ?? ?? % ??
    ../lib/Memoize/t/normalize.t 2 512 ?? ?? % ??
    ../lib/Memoize/t/prototype.t 2 512 ?? ?? % ??
    ../lib/Memoize/t/speed.t 2 512 ?? ?? % ??
    ../lib/Memoize/t/tie.t 2 512 ?? ?? % ??
    ../lib/Memoize/t/tie_gdbm.t 2 512 ?? ?? % ??
    ../lib/Memoize/t/tie_ndbm.t 2 512 ?? ?? % ??
    ../lib/Memoize/t/tie_sdbm.t 2 512 ?? ?? % ??
    ../lib/Memoize/t/tie_storable.t 2 512 ?? ?? % ??
    ../lib/Memoize/t/tiefeatures.t 2 512 ?? ?? % ??
    ../lib/Memoize/t/unmemoize.t 2 512 ?? ?? % ??
    ../lib/NEXT/t/actual.t 2 512 9 9 100.00% 1-9
    ../lib/NEXT/t/actuns.t 2 512 5 5 100.00% 1-5
    ../lib/NEXT/t/next.t 2 512 25 25 100.00% 1-25
    ../lib/NEXT/t/unseen.t 2 512 4 4 100.00% 1-4
    ../lib/Net/hostent.t 2 512 ?? ?? % ??
    ../lib/Net/netent.t 2 512 ?? ?? % ??
    ../lib/Net/protoent.t 2 512 ?? ?? % ??
    ../lib/Net/servent.t 2 512 ?? ?? % ??
    ../lib/Pod/t/Functions.t 2 512 ?? ?? % ??
    ../lib/Pod/t/InputObjects.t 2 512 ?? ?? % ??
    ../lib/Pod/t/Select.t 2 512 ?? ?? % ??
    ../lib/Pod/t/Usage.t 2 512 ?? ?? % ??
    ../lib/Pod/t/basic.t 2 512 11 11 100.00% 1-11
    ../lib/Pod/t/htmlescp.t 2 512 ?? ?? % ??
    ../lib/Pod/t/htmlview.t 2 512 ?? ?? % ??
    ../lib/Pod/t/latex.t 2 512 ?? ?? % ??
    ../lib/Pod/t/man.t 2 512 2 2 100.00% 1-2
    ../lib/Pod/t/text-errors.t 2 512 5 5 100.00% 1-5
    ../lib/Pod/t/utils.t 2 512 ?? ?? % ??
    ?Failed Test Stat Wstat Total Fail Failed List of Failed

    -------------------------------------------------------------------------------
    ../lib/Search/Dict.t 2 512 ?? ?? % ??
    ../lib/SelectSaver.t 2 512 ?? ?? % ??
    ../lib/SelfLoader.t 2 512 19 19 100.00% 1-19
    ../lib/Shell.t 2 512 ?? ?? % ??
    ../lib/Switch/t/given.t 2 512 ?? ?? % ??
    ../lib/Switch/t/nested.t 2 512 ?? ?? % ??
    ../lib/Switch/t/switch.t 2 512 ?? ?? % ??
    ../lib/Symbol.t 2 512 ?? ?? % ??
    ../lib/Term/Cap.t 2 512 ?? ?? % ??
    ../lib/Term/Complete.t 2 512 ?? ?? % ??
    ../lib/Term/ReadLine.t 2 512 ?? ?? % ??
    ../lib/Test/Harness/t/00compile.t 2 512 ?? ?? % ??
    ../lib/Test/Harness/t/assert.t 2 512 ?? ?? % ??
    ../lib/Test/Harness/t/base.t 1 1 100.00% 1
    ../lib/Test/Harness/t/callback.t 2 512 ?? ?? % ??
    ../lib/Test/Harness/t/strap-analy 2 512 ?? ?? % ??
    ../lib/Test/Harness/t/strap.t 2 512 ?? ?? % ??
    ../lib/Test/Harness/t/test-harnes 2 512 ?? ?? % ??
    ../lib/Test/Simple/t/Builder.t 2 512 ?? ?? % ??
    ../lib/Test/Simple/t/More.t 2 512 ?? ?? % ??
    ../lib/Test/Simple/t/buffer.t 2 512 ?? ?? % ??
    ../lib/Test/Simple/t/diag.t 2 512 ?? ?? % ??
    ../lib/Test/Simple/t/extra.t 2 2 100.00% 1-2
    ../lib/Test/Simple/t/fail-like.t 2 512 ?? ?? % ??
    ../lib/Test/Simple/t/fail-more.t 2 512 ?? ?? % ??
    ../lib/Test/Simple/t/fail.t 2 512 ?? ?? % ??
    ../lib/Test/Simple/t/filehandles. 2 512 ?? ?? % ??
    ../lib/Test/Simple/t/import.t 2 512 ?? ?? % ??
    ../lib/Test/Simple/t/is_deeply.t 2 512 ?? ?? % ??
    ../lib/Test/Simple/t/missing.t 2 2 100.00% 1-2
    ../lib/Test/Simple/t/no_ending.t 2 512 ?? ?? % ??
    ../lib/Test/Simple/t/no_header.t 2 512 ?? ?? % ??
    ../lib/Test/Simple/t/no_plan.t 12 10 83.33% 3-12
    ../lib/Test/Simple/t/output.t 2 512 ?? ?? % ??
    ../lib/Test/Simple/t/plan.t 2 512 ?? ?? % ??
    ../lib/Test/Simple/t/plan_is_nopl 2 512 ?? ?? % ??
    ../lib/Test/Simple/t/plan_no_plan 2 512 ?? ?? % ??
    ../lib/Test/Simple/t/plan_skip_al 2 512 ?? ?? % ??
    ../lib/Test/Simple/t/simple.t 2 512 ?? ?? % ??
    ../lib/Test/Simple/t/skip.t 2 512 ?? ?? % ??
    ../lib/Test/Simple/t/skipall.t 2 512 ?? ?? % ??
    ../lib/Test/Simple/t/todo.t 2 512 ?? ?? % ??
    ../lib/Test/Simple/t/undef.t 2 512 ?? ?? % ??
    ../lib/Test/Simple/t/use_ok.t 2 512 ?? ?? % ??
    ../lib/Test/Simple/t/useing.t 2 512 ?? ?? % ??
    ../lib/Text/Abbrev.t 2 512 ?? ?? % ??
    ../lib/Text/Balanced/t/extbrk.t 2 512 19 19 100.00% 1-19
    ../lib/Text/Balanced/t/extcbk.t 2 512 41 41 100.00% 1-41
    ../lib/Text/Balanced/t/extdel.t 2 512 45 45 100.00% 1-45
    ../lib/Text/Balanced/t/extmul.t 2 512 85 85 100.00% 1-85
    ../lib/Text/Balanced/t/extqlk.t 2 512 89 89 100.00% 1-89
    ../lib/Text/Balanced/t/exttag.t 2 512 53 53 100.00% 1-53
    ../lib/Text/Balanced/t/extvar.t 2 512 181 181 100.00% 1-181
    ../lib/Text/Balanced/t/gentag.t 2 512 35 35 100.00% 1-35
    ../lib/Text/ParseWords.t 2 512 ?? ?? % ??
    ../lib/Text/Soundex.t 2 512 ?? ?? % ??
    ../lib/Text/TabsWrap/t/fill.t 2 512 ?? ?? % ??
    ../lib/Text/TabsWrap/t/tabs.t 2 512 ?? ?? % ??
    ?Failed Test Stat Wstat Total Fail Failed List of Failed

    -------------------------------------------------------------------------------
    ../lib/Text/TabsWrap/t/wrap.t 2 512 ?? ?? % ??
    ../lib/Tie/Array/push.t 2 512 ?? ?? % ??
    ../lib/Tie/Array/splice.t 2 512 ?? ?? % ??
    ../lib/Tie/Array/std.t 2 512 ?? ?? % ??
    ../lib/Tie/Array/stdpush.t 2 512 ?? ?? % ??
    ../lib/Tie/Handle/stdhandle.t 2 512 ?? ?? % ??
    ../lib/Tie/Memoize.t 2 512 ?? ?? % ??
    ../lib/Tie/RefHash.t 2 512 ?? ?? % ??
    ../lib/Tie/Scalar.t 2 512 ?? ?? % ??
    ../lib/Tie/SubstrHash.t 2 512 ?? ?? % ??
    ../lib/Time/Local.t 2 512 ?? ?? % ??
    ../lib/Time/gmtime.t 2 512 ?? ?? % ??
    ../lib/Time/localtime.t 2 512 ?? ?? % ??
    ../lib/Unicode/UCD.t 2 512 ?? ?? % ??
    ../lib/User/grent.t 2 512 ?? ?? % ??
    ../lib/User/pwent.t 2 512 ?? ?? % ??
    ../lib/fields.t 2 512 ?? ?? % ??
    ../lib/filetest.t 2 512 ?? ?? % ??
    ../lib/h2xs.t 2 512 ?? ?? % ??
    ../lib/if.t 2 512 ?? ?? % ??
    ../lib/integer.t 2 512 ?? ?? % ??
    ../lib/lib.t 2 512 ?? ?? % ??
    ../lib/locale.t 2 512 ?? ?? % ??
    ../lib/open.t 2 512 ?? ?? % ??
    ../lib/overload.t 2 512 ?? ?? % ??
    ../lib/sigtrap.t 2 512 ?? ?? % ??
    ../lib/sort.t 2 512 ?? ?? % ??
    ../lib/strict.t 2 512 ?? ?? % ??
    ../lib/subs.t 9 9 100.00% 1-9
    ../lib/utf8.t 2 512 ?? ?? % ??
    ../lib/vars.t 2 512 ?? ?? % ??
    ../lib/vmsish.t 2 512 ?? ?? % ??
    ../lib/warnings.t 2 512 ?? ?? % ??
    pod/emptycmd.t 2 512 ?? ?? % ??
    pod/find.t 2 512 ?? ?? % ??
    pod/for.t 2 512 ?? ?? % ??
    pod/headings.t 2 512 ?? ?? % ??
    pod/include.t 2 512 ?? ?? % ??
    pod/included.t 2 512 ?? ?? % ??
    pod/lref.t 2 512 ?? ?? % ??
    pod/multiline_items.t 2 512 ?? ?? % ??
    pod/nested_items.t 2 512 ?? ?? % ??
    pod/nested_seqs.t 2 512 ?? ?? % ??
    pod/oneline_cmds.t 2 512 ?? ?? % ??
    pod/plainer.t 2 512 ?? ?? % ??
    pod/pod2usage.t 2 512 ?? ?? % ??
    pod/poderrs.t 2 512 ?? ?? % ??
    pod/podselect.t 2 512 ?? ?? % ??
    pod/special_seqs.t 2 512 ?? ?? % ??
    42 tests and 154 subtests skipped.
    Failed 223/677 test scripts, 67.06% okay. 2930/43118 subtests failed, 93.20% okay.
    --
    H.Merijn Brand Amsterdam Perl Mongers (http://amsterdam.pm.org/)
    using perl-5.6.1, 5.7.3 & 631 on HP-UX 10.20 & 11.00, AIX 4.2, AIX 4.3,
    WinNT 4, Win2K pro & WinCE 2.11. Smoking perl CORE: smokers@perl.org
    http://archives.develooper.com/daily-build@perl.org/ perl-qa@perl.org
    send smoke reports to: smokers-reports@perl.org, QA: http://qa.perl.org
  • Michael G Schwern at Apr 24, 2002 at 2:43 pm
    On Wed, Apr 24, 2002 at 02:47:00PM +0200, H.Merijn Brand wrote:
    in@16125 :
    /lib/Env/t/env.....................ok
    /lib/Exporter......................ok
    /lib/ExtUtils/t/basic..............# Failed test (../lib/ExtUtils/t/basic.t
    at line 32)
    /lib/ExtUtils/t/basic..............NOK 1# chdir failed: No such file or director
    y
    Can't open Liar/t/sanity.t: No such file or directory at ../lib/ExtUtils/t/basic
    .t line 49.
    # Looks like you planned 17 tests but only ran 1.
    # Looks like your test died just after 1.
    /lib/ExtUtils/t/basic..............dubious
    The MakeMaker tests are relying that the tests will run in alphabetical
    order, numbers first. I betcha lib/ExtUtils/t/00setup_dummy.t didn't run
    before basic.t?


    --

    Michael G. Schwern <schwern@pobox.com> http://www.pobox.com/~schwern/
    Perl Quality Assurance <perl-qa@perl.org> Kwalitee Is Job One
    you're a bleeding-heart
    liberal. LET ME PLUG MY
    ASS WITH PASTE RIGHT NOW!
    -- japhy
  • H.Merijn Brand at Apr 24, 2002 at 3:03 pm

    On Wed 24 Apr 2002 15:45, Michael G Schwern wrote:
    On Wed, Apr 24, 2002 at 02:47:00PM +0200, H.Merijn Brand wrote:
    in@16125 :
    /lib/Env/t/env.....................ok
    /lib/Exporter......................ok
    /lib/ExtUtils/t/basic..............# Failed test (../lib/ExtUtils/t/basic.t
    at line 32)
    /lib/ExtUtils/t/basic..............NOK 1# chdir failed: No such file or director
    y
    Can't open Liar/t/sanity.t: No such file or directory at ../lib/ExtUtils/t/basic
    .t line 49.
    # Looks like you planned 17 tests but only ran 1.
    # Looks like your test died just after 1.
    /lib/ExtUtils/t/basic..............dubious
    The MakeMaker tests are relying that the tests will run in alphabetical
    order, numbers first. I betcha lib/ExtUtils/t/00setup_dummy.t didn't run
    before basic.t?
    <dummy mode>

    # rsync ...
    # Configure -Dusedevel -Duseperlio -Duse64bitint -des
    # make
    # make test_prep
    # ./perl -Ilib t/harness

    <shivver mode>

    --
    H.Merijn Brand Amsterdam Perl Mongers (http://amsterdam.pm.org/)
    using perl-5.6.1, 5.7.3 & 631 on HP-UX 10.20 & 11.00, AIX 4.2, AIX 4.3,
    WinNT 4, Win2K pro & WinCE 2.11. Smoking perl CORE: smokers@perl.org
    http://archives.develooper.com/daily-build@perl.org/ perl-qa@perl.org
    send smoke reports to: smokers-reports@perl.org, QA: http://qa.perl.org

Related Discussions

Discussion Navigation
viewthread | post
Discussion Overview
groupperl5-porters @
categoriesperl
postedApr 24, '02 at 11:45a
activeApr 26, '02 at 6:37p
posts27
users6
websiteperl.org

People

Translate

site design / logo © 2022 Grokbase