smoke signs suppression

The iperlsys.h reverts an earlier tweak of mine since it
doesn't seem
to help, others pretty standard smoke-induced moves.





--- iperlsys.h.dist	2006-12-31 15:34:06.000000000 +0200
+++ iperlsys.h	2006-12-31 15:34:39.000000000 +0200
 -339,7
+339,7 
 #define PerlSIO_set_cnt(f,c)		PerlIOProc_abort()
 #endif
 #if defined(USE_STDIO_PTR) &&
defined(STDIO_PTR_LVALUE)
-#define PerlSIO_set_ptr(f,p)		FILE_ptr(f) = ((STDCHAR*)(p))
+#define PerlSIO_set_ptr(f,p)		(FILE_ptr(f) = (p))
 #else
 #define PerlSIO_set_ptr(f,p)		PerlIOProc_abort()
 #endif
--- perly.c.dist	2006-12-31 15:36:30.000000000 +0200
+++ perly.c	2006-12-31 15:36:36.000000000 +0200
 -319,7
+319,7 
     int yyresult;
 
     /* Lookahead token as an internal (translated) token
number.  */
-    int yytoken;
+    int yytoken = 0;
 
     SV *parser_sv;		    /* SV whose PVX holds the parser
object */
     register yy_parser *parser;	    /* the parser object */
--- regcomp.c.dist	2006-12-31 15:33:07.000000000 +0200
+++ regcomp.c	2006-12-31 15:37:37.000000000 +0200
 -5065,7
+5065,7 
 		/*notreached*/
             { /* named and numeric backreferences */
                 I32 num;
-                char * parse_start;
+                char * parse_start = NULL;
             case '&':            /* (?&NAME) */
                 parse_start = RExC_parse - 1;
               named_recursion:
 -8144,7
+8144,8 
             I32 n;
             if (name) {
                 for ( n=0; n<SvIVX(sv_dat); n++ ) {
-                    Perl_sv_catpvf(aTHX_ sv,
"%s%"IVdf, ( n ? "," : "" ),
nums[n]);
+                    Perl_sv_catpvf(aTHX_ sv,
"%s%"IVdf,
+				   (n ? "," : ""), (IV)nums[n]);
                 }
                 Perl_sv_catpvf(aTHX_ sv, "
'%"SVf"'", *name );
             }

On Sun, 31 Dec 2006 08:51:52 -0500, Jarkko Hietaniemi
<jhiiki.fi> wrote:

> The iperlsys.h reverts an earlier tweak of mine since
it doesn't seem
> to help, others pretty standard smoke-induced moves.

Thanks, applied as change #29645

-- 
H.Merijn Brand         Amsterdam Perl Mongers (http://amsterdam.pm.org/
)
using & porting perl 5.6.2, 5.8.x, 5.9.x   on HP-UX
10.20, 11.00, 11.11,
& 11.23, SuSE 10.0 & 10.1, AIX 4.3 & 5.2, and
Cygwin. http://qa.perl.org
http://mirrors.de
velooper.com/hpux/            http://www.test-smoke.org
                        http
://www.goldmark.org/jeff/stupid-disclaimers/

H.Merijn Brand wrote:
> On Sun, 31 Dec 2006 08:51:52 -0500, Jarkko Hietaniemi
<jhiiki.fi> wrote:
> 
>> The iperlsys.h reverts an earlier tweak of mine
since it doesn't seem
>> to help, others pretty standard smoke-induced
moves.
> 
> Thanks, applied as change #29645
> 

Thanks.

Could you perhaps also apply Ilya's suggestion

http://www.xray.mpe.mpg.de/mailing-l
ists/perl5-porters/2006-12/msg00646.html

for the Self* failures, you seem to have one of the failing
systems:

http://www.nntp.perl.org/group/perl.daily-build.rep
orts/43589

On Sun, Dec 31, 2006 at 10:18:59AM -0500, Jarkko Hietaniemi
wrote:
> H.Merijn Brand wrote:
> > On Sun, 31 Dec 2006 08:51:52 -0500, Jarkko
Hietaniemi <jhiiki.fi> wrote:
> > 
> >> The iperlsys.h reverts an earlier tweak of
mine since it doesn't seem
> >> to help, others pretty standard smoke-induced
moves.
> > 
> > Thanks, applied as change #29645
> > 
> 
> Thanks.
> 
> Could you perhaps also apply Ilya's suggestion
> 
> http://www.xray.mpe.mpg.de/mailing-l
ists/perl5-porters/2006-12/msg00646.html
> 
> for the Self* failures, you seem to have one of the
failing systems:
> 
> http://www.nntp.perl.org/group/perl.daily-build.rep
orts/43589

A slightly modified version of the suggestion got the tests
passing for
me.

Steve Peters
stevefisharerojo.org

Change 29646 by stevepstevep-kirk on 2006/12/31 16:37:54

	Fix to test failures caused by change #29606 when
PERLIO=stdio.
	Based on suggestions from Ilya Zakharevich in 
	http://www.xray.mpe.mpg.de/mailing-l
ists/perl5-porters/2006-12/msg00646.html

Affected files ...

... //depot/perl/lib/SelfLoader.pm#22 edit

Differences ...

==== //depot/perl/lib/SelfLoader.pm#22 (text) ====

 -61,10
+61,12 
     croak("$callpack doesn't contain an __DATA__
token")
         unless defined fileno($fh);
     # Protect: fork() shares the file pointer between the
parent and the kid
-    open my $nfh, '<&', $fh or croak "reopen:
$!";# dup() the fd
-    close $fh or die "close: $1";                
# autocloses, but be paranoid
-    open $fh, '<&', $nfh or croak "reopen2:
$!";  # dup() the fd "back"
-    close $nfh or die "close after reopen: $1";  
# autocloses, but be paranoid
+    if(sysseek($fh, tell($fh), 0)) {
+      open my $nfh, '<&', $fh or croak "reopen:
$!";# dup() the fd
+      close $fh or die "close: $1";              
  # autocloses, but be paranoid
+      open $fh, '<&', $nfh or croak "reopen2:
$!";  # dup() the fd "back"
+      close $nfh or die "close after reopen: $1";
  # autocloses, but be paranoid
+    }
     $Cache{"$::<DATA"} = 1;   #
indicate package is cached
 
     local($/) = "n";

Steve Peters wrote:
> On Sun, Dec 31, 2006 at 10:18:59AM -0500, Jarkko
Hietaniemi wrote:
>> H.Merijn Brand wrote:
>>> On Sun, 31 Dec 2006 08:51:52 -0500, Jarkko
Hietaniemi <jhiiki.fi> wrote:
>>>
>>>> The iperlsys.h reverts an earlier tweak of
mine since it doesn't seem
>>>> to help, others pretty standard
smoke-induced moves.
>>> Thanks, applied as change #29645
>>>
>> Thanks.
>>
>> Could you perhaps also apply Ilya's suggestion
>>
>> http://www.xray.mpe.mpg.de/mailing-l
ists/perl5-porters/2006-12/msg00646.html
>>
>> for the Self* failures, you seem to have one of the
failing systems:
>>
>> http://www.nntp.perl.org/group/perl.daily-build.rep
orts/43589
> 
> A slightly modified version of the suggestion got the
tests passing for
> me.
> 
> Steve Peters
> stevefisharerojo.org
> 
> Change 29646 by stevepstevep-kirk on 2006/12/31
16:37:54
> 
> 	Fix to test failures caused by change #29606 when
PERLIO=stdio.
> 	Based on suggestions from Ilya Zakharevich in 
> 	http://www.xray.mpe.mpg.de/mailing-l
ists/perl5-porters/2006-12/msg00646.html

Sweet.

It seems that we have some persistent smoke failures:

t/op/utftaint.t in Darwin and powerpc debian and Solaris:

http://www.nntp.perl.org/group/perl.daily-build.rep
orts/43586
http://www.nntp.perl.org/group/perl.daily-build.rep
orts/43583
http://www.nntp.perl.org/group/perl.daily-build.rep
orts/43581

an x86 Debian that's very unhappy in various areas:

http://www.nntp.perl.org/group/perl.daily-build.rep
orts/43585

a ppc Debian also very unhappy:

http://www.nntp.perl.org/group/perl.daily-build.rep
orts/43583

and many more...

On Sun, 31 Dec 2006 14:35:16 -0500, Jarkko Hietaniemi
<jhiiki.fi> wrote:

> Steve Peters wrote:
> > On Sun, Dec 31, 2006 at 10:18:59AM -0500, Jarkko
Hietaniemi wrote:
> >> H.Merijn Brand wrote:
> >>> On Sun, 31 Dec 2006 08:51:52 -0500, Jarkko
Hietaniemi <jhiiki.fi> wrote:
> >>>
> >>>> The iperlsys.h reverts an earlier
tweak of mine since it doesn't seem
> >>>> to help, others pretty standard
smoke-induced moves.
> >>> Thanks, applied as change #29645
> >>>
> >> Thanks.
> >>
> >> Could you perhaps also apply Ilya's suggestion
> >>
> >> http://www.xray.mpe.mpg.de/mailing-l
ists/perl5-porters/2006-12/msg00646.html
> >>
> >> for the Self* failures, you seem to have one
of the failing systems:
> >>
> >> http://www.nntp.perl.org/group/perl.daily-build.rep
orts/43589
> > 
> > A slightly modified version of the suggestion got
the tests passing for
> > me.
> > 
> > Steve Peters
> > stevefisharerojo.org
> > 
> > Change 29646 by stevepstevep-kirk on 2006/12/31
16:37:54
> > 
> > 	Fix to test failures caused by change #29606 when
PERLIO=stdio.
> > 	Based on suggestions from Ilya Zakharevich in 
> > 	http://www.xray.mpe.mpg.de/mailing-l
ists/perl5-porters/2006-12/msg00646.html
> 
> Sweet.
> 
> It seems that we have some persistent smoke failures:

Not here:

Checking status for configuration 'p59_config' (5.9.x)
  Change number 29646 started on Sun Dec 31 17:49:24 2006.
    5 out of 16 configurations finished in 2 hours 48
minutes.
    2 configurations showed failures (F).
    32 failures in the running configuration.
    11 configurations to finish, estimated completion in 5
hours 52 minutes
    Average smoke duration: 33 minutes 47 seconds.
  Matrix, using cc version 4.1.0 (SUSE Linux):
     29646     Configuration (common) none
  -----------
---------------------------------------------------------
  O X O O O O
  X O O O O O -Duse64bitint
  O O O F ? - -Duselongdouble
  | | | | | +- LC_ALL = en_US.utf8 -DDEBUGGING
  | | | | +--- PERLIO = perlio -DDEBUGGING
  | | | +----- PERLIO = stdio  -DDEBUGGING
  | | +------- LC_ALL = en_US.utf8
  | +--------- PERLIO = perlio
  +----------- PERLIO = stdio

> t/op/utftaint.t in Darwin and powerpc debian and
Solaris:
> 
> http://www.nntp.perl.org/group/perl.daily-build.rep
orts/43586
> http://www.nntp.perl.org/group/perl.daily-build.rep
orts/43583
> http://www.nntp.perl.org/group/perl.daily-build.rep
orts/43581
> 
> an x86 Debian that's very unhappy in various areas:
> 
> http://www.nntp.perl.org/group/perl.daily-build.rep
orts/43585
> 
> a ppc Debian also very unhappy:
> 
> http://www.nntp.perl.org/group/perl.daily-build.rep
orts/43583
> 
> and many more...
> 
> 
> 


-- 
H.Merijn Brand         Amsterdam Perl Mongers (http://amsterdam.pm.org/
)
using & porting perl 5.6.2, 5.8.x, 5.9.x   on HP-UX
10.20, 11.00, 11.11,
& 11.23, SuSE 10.0 & 10.1, AIX 4.3 & 5.2, and
Cygwin. http://qa.perl.org
http://mirrors.de
velooper.com/hpux/            http://www.test-smoke.org
                        http
://www.goldmark.org/jeff/stupid-disclaimers/