httpd-apreq-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Joe Schaefer <joe+gm...@sunstarsys.com>
Subject Re: 2.03-dev-rc1 available
Date Thu, 10 Jun 2004 19:39:08 GMT
"Edward J. Sabol" <sabol@alderaan.gsfc.nasa.gov> writes:

> Martin Nilsson wrote:
> > I have changed the following:
> > env/t/TEST.PL added these functions.
> >
> > sub cmodules_makefile_vars {
> >      return <<EOF;
> > MAKE = gmake
> > EOF
> > }
> [...]
> > This seems to be a problem with Apache::Test not libapreq!
> 
> I'm not a fan of Apache::Test's over-reliance on %Config either. (See
> previous e-mails from me regarding $Config{perlpath}.)
> 
> Would the following work?
> 
> sub cmodules_makefile_vars {
>      my $make = $ENV{MAKE} || $Config{make};
>      return <<EOF;
> MAKE = $make
> EOF
> }
> 
> Most all makes, including gmake, should set the MAKE environment variable...
> Just wondering if we can come up with a generic solution that will work for
> everyone...

+1- I see we had the same thought :-).

On a related note, I get the same segfaults with 5.6.1 (no ithreads)
on Debian woody.  Everything compiles and all the non-perl tests
pass, but all the perl tests segfault.  The backtraces are wierd, 
though:

% bt
#0  apr_palloc (pool=0x403a2600, size=16) at apr_pools.c:632
632         if (size < (apr_size_t)(active->endp - active->first_avail)) {
(gdb) bt
#0  apr_palloc (pool=0x403a2600, size=16) at apr_pools.c:632
#1  0x401eeb26 in apr_pmemdup (a=0x403a2600, m=0x81dce38, n=11)
    at apr_strings.c:156
#2  0x40612df5 in XS_Apache__RequestRec_content_type ()
   from /usr/local/lib/perl/5.6.1/Apache2/auto/Apache/RequestRec/RequestRec.so
#3  0x405787b1 in Perl_pp_entersub () from /usr/lib/libperl.so.5.6
#4  0x40572621 in Perl_runops_standard () from /usr/lib/libperl.so.5.6
#5  0x4052abac in perl_call_sv () from /usr/lib/libperl.so.5.6
#6  0x4052a91e in perl_call_sv () from /usr/lib/libperl.so.5.6
#7  0x404f4966 in modperl_callback () from /opt/apache2/modules/mod_perl.so
#8  0x404f4cfd in modperl_callback_run_handlers ()
   from /opt/apache2/modules/mod_perl.so
#9  0x404f4d92 in modperl_callback_per_dir ()
   from /opt/apache2/modules/mod_perl.so
#10 0x404f224a in modperl_response_handler_run ()
   from /opt/apache2/modules/mod_perl.so
#11 0x404f22bf in modperl_response_handler ()
   from /opt/apache2/modules/mod_perl.so
#12 0x08066f39 in ap_run_handler (r=0x81c20c0) at config.c:194
#13 0x08067483 in ap_invoke_handler (r=0x81c20c0) at config.c:401
#14 0x08064796 in ap_process_request (r=0x81c20c0) at http_request.c:288
#15 0x0806080a in ap_process_http_connection (c=0x81bc180) at http_core.c:293
#16 0x0806f188 in ap_run_process_connection (c=0x81bc180) at connection.c:85
#17 0x0806f42c in ap_process_connection (c=0x81bc180, csd=0x81bc0a8)
    at connection.c:211
#18 0x08065c10 in child_main (child_num_arg=0) at prefork.c:694
#19 0x08065d6a in make_child (s=0x809be68, slot=0) at prefork.c:788
#20 0x08065f11 in perform_idle_server_maintenance (p=0x809a0c8)
    at prefork.c:923


Since mp2's test suite runs fine for me, I suspect some of our 
THX-related code may need fixing for 5.6 (IIRC modern MM's 
generate a helper header file, maybe we need to use that)?  
I'm out of my depth here, hopefully Stas might be able to help us...

-- 
Joe Schaefer


Mime
View raw message