On Wed, 6 Sep 2000, Peter Prymmer wrote:

> >     $ MMS TEST
> > will produce the LF's (same ones that Peter sees).
> > 
> > But:
> >     $ @[.vms]test ".exe" 1
> > doesn't have extra LF's.  Both were run as batch jobs so SYS$ERROR
> > output would go to the log file.
> > 
> > To compare results so far:
> >  Who    system              compiler            make            result
> > -----------------------------------------------------------------------
> > Prymmer ?                   ?                   ?               LF's
> 
> make that line read:
> 
>  Prymmer VMS/AXP 7.1          DECC V6.0-001       MMS V3.3-4      LF's

By the way that listed result was with the perl-5.7.0 kit.
In addition to that I now see:

  Prymmer VMS/VAX 6.2          DECC V5.3-006       MMS V3.0-05     LF's
(vax mess is discussed in detail below)

  Prymmer VMS/AXP 7.2-1        DECC V6.0-001       MMS V3.3-4      LF's

  Prymmer VMS/AXP 7.1          DECC V5.6-003       MMS V3.2-01     LF's
  Prymmer VMS/AXP 7.1          DECC V5.6-003       MMK V           almost ok ('cept 
for test failures)

These latter two tests with MMS and MMK were both done on the same machine
all in the same source kit build path.  I do note that I see far fewer
LF's with MMK *but* I do see these extra things (that I really should 
not be seeing at all):

 [.io]open...............ok
 Can't locate warnings/register.pm in @INC (@INC contains: ../lib/VMS_AXP../lib
 perl_root:[lib.VMS_AXP.5_7_0] perl_root:[lib]
 perl_root:[lib.site_perl.VMS_AXP]
 perl_root:[lib.site_perl] /perl_root/lib/site_perl .) at ../lib/vars.pm line 12.
 BEGIN failed--compilation aborted at ../lib/vars.pm line 12.
 Compilation failed in require at ../lib/File/Spec.pm line 4.
 BEGIN failed--compilation aborted at ../lib/File/Spec.pm line 4.
 Compilation failed in require at ../lib/IO/File.pm line 113.
 BEGIN failed--compilation aborted at ../lib/IO/File.pm line 113.
 Compilation failed in require at ../lib/FileHandle.pm line 9.
 Compilation failed in require at -e line 1.
 BEGIN failed--compilation aborted at -e line 1.
 [.io]openpid............FAILED on test 7

<snip>

 [.lib]st-recurse........ok
 Corrupted storable file (binary v2.2) at ../../lib/storable.pm (autosplit
 into [--.lib.auto.Storable]retrieve.al) line 187, at [.lib]st-retrieve.t line 51
 [.lib]st-retrieve.......FAILED on test 6
 Corrupted storable file (binary v2.2) at ../../lib/storable.pm (autosplit
 into [--.lib.auto.Storable]retrieve.al) line 187, at [.lib]st-store.t line 45
 [.lib]st-store..........FAILED on test 3

<snip>

 [.op]subst_wamp.........ok
 [.op]sysio..............ok
 # Note: logical name 'PATH' may have been deleted
 [.op]taint..............ok
 [.op]tie................ok

<snip>

BTW that failure of openpid.t is in addition to the known doubles store
test failures.

I have also stumbled across a config/build failure on a VAX with:

  Prymmer VMS/VAX 6.2          DECC V5.3-006       MMS V3.0-05     LF's

Here the build warns at the compilation of pp.c with:

CC/DECC/Include=[]/Standard=Relaxed_ANSI/Prefix=All/Obj=PP.obj/NoList/Define=PERL_CORE 
PP.C
              if (Perl_isnan(left) || Perl_isnan(right)) {
        ..........^
%CC-I-IMPLICITFUNC, In this statement, the identifier "isnan" is
implicitly
 declared as a function.
                At line number 1222 in 219333$DIA1:[PERL-5_7_0]PP.C;1.

%VCG-I-SUMMARY, Completed with 0 error(s), 0 warning(s), and
                1 informational messages.
                At line number 5296 in 219333$DIA1:[PERL-5_7_0]PP.C;1.


Then later the build dies at the link of miniperl with:

Link /NoTrace/NoMap/Trace/Exe=MINIPERL.EXE
miniperlmain.obj,libperl.olb/Library/
Include=globals  ,[]crtl.opt/Options
%LINK-W-NUDFSYMS, 1 undefined symbol:
%LINK-I-UDFSYM,         ISNAN
%LINK-W-USEUNDEF, undefined symbol ISNAN referenced
        in psect $CODE offset %X00004119
        in module PP file 219333$DIA1:[PERL-5_7_0]LIBPERL.OLB;1
%LINK-W-USEUNDEF, undefined symbol ISNAN referenced
        in psect $CODE offset %X00004127
        in module PP file 219333$DIA1:[PERL-5_7_0]LIBPERL.OLB;1
%MMS-F-ABORT, For target MINIPERL.EXE, CLI returned abort status:
%X10648268.

And it looks like I've uncovered yet another bug in configure.com 
since I see this in my config.h:

#define HAS_ISNAN               /**/
#undef HAS_ISNANL               /**/
                                         
Here is a VMS style diff (sorry, not yet a GNU uni diff patch):

$ diff configure.com
************
File 219333$DIA1:[PERL-5_7_0]CONFIGURE.COM;2
 4750   $ WC "d_isnan='" + d_isnan + "'"
 4751   $ WC "d_isnanl='" + d_isnanl + "'"
******
File 219333$DIA1:[PERL-5_7_0]CONFIGURE.COM;1
 4750   $ WC "d_isnan='define'"
 4751   $ WC "d_isnanl='" + d_isnanl + "'"

Oops.  Do note that this is currently misconfigured on AXPen that are not
building with 64bitints.

Warning seen building VMS::Stdio:

Copy/NoConfirm STDIO.OPT [---.LIB.AUTO.VMS.STDIO]STDIO.OPT
MCR 219333$dia1:[perl-5_7_0]miniperl.exe "-I[---.lib]" "-I[---.lib]"
[---.lib.Ex
tUtils]xsubpp  -typemap [---.lib.ExtUtils]typemap STDIO.xs >STDIO.C
CC/DECC/Include=[]/Standard=Relaxed_ANSI/Prefix=All/Obj=.obj/NoList/Define=("VE
RSION=""2.2""","XS_VERSION=""2.2""")/Include=([---])/Optimize  STDIO.c
                    if ((retsts = sys$setddir(&dirdsc,0,0)) & 1) ST(0) = &PL_sv_yes;
        ..........................^
%CC-I-IMPLICITFUNC, In this statement, the identifier "sys$setddir" is
 implicitly declared as a function.
                At line number 259 in stdio.xs.

%VCG-I-SUMMARY, Completed with 0 error(s), 0 warning(s), and
                1 informational messages.
                At line number 572 in stdio.c.



Peter Prymmer

Reply via email to