On Wednesday 09 May 2007 10:42, Fred Moyer wrote:
> > to have these 2 patches applied to trunk or to get a statement why they
> > cannot be applied?
> >
> > http://www.gossamer-threads.com/lists/modperl/dev/93415
> > http://www.gossamer-threads.com/lists/modperl/dev/93466
>
> I have been following this somewhat, but not grokking all of it.
>
> Can you write a test that shows the change and its effect?
So, this patch adds an extra test to show the effect of the first patch. With
trunk the handling thread is locked forever after the 2nd test. If the number
of interpreters is limited to 1 it is locked after the 1st request.
With the patch applied all works smoothly.
Torsten
--- /dev/null 2006-11-25 13:17:37.000000000 +0100
+++ t/perl/ithreads3.t 2007-05-10 10:06:11.000000000 +0200
@@ -0,0 +1,39 @@
+use strict;
+use warnings FATAL => 'all';
+
+use Apache::Test;
+use Apache::TestUtil;
+use Apache::TestRequest 'GET_BODY';
+
+plan tests => 6, need_apache_mpm('worker') && need_perl('ithreads');
+
+my $module = 'TestPerl::ithreads3';
+
+sub u {Apache::TestRequest::module2url($module, {path=>$_[0]})}
+sub t {
+ my $rc;
+ eval {
+ local $SIG{ALRM}=sub {die "Timeout\n"};
+ alarm 2;
+ eval {
+ $rc=GET_BODY u(shift);
+ };
+ alarm 0;
+ };
+ alarm 0;
+ return $rc;
+}
+
+t_debug("connecting to ".u(''));
+ok t_cmp t('/perl-script?1'), 2, 'perl-script 1';
+ok t_cmp t('/modperl?1'), 2, 'modperl 1';
+
+ok t_cmp t('/perl-script?2'), 5, 'perl-script 2';
+ok t_cmp t('/modperl?2'), 5, 'modperl 2';
+
+ok t_cmp t('/perl-script?3'), 3, 'perl-script 3';
+ok t_cmp t('/modperl?3'), 3, 'modperl 3';
+
+# Local Variables: #
+# mode: cperl #
+# End: #
--- /dev/null 2006-11-25 13:17:37.000000000 +0100
+++ t/response/TestPerl/ithreads3.pm 2007-05-10 10:58:00.000000000 +0200
@@ -0,0 +1,108 @@
+package TestPerl::ithreads3;
+
+use strict;
+use warnings FATAL => 'all';
+
+use Apache2::RequestRec;
+use Apache2::RequestIO;
+use Apache2::RequestUtil;
+use APR::Pool;
+use Apache2::Const -compile => 'OK', 'DECLINED';
+
+# XXX: These tests rely on the assumption that the virtual host is not
+# otherwise accessed. In this case the same interpreter is chosen
+# for each phase. The $counter counts them.
+# Of course if only 1 interp is configured it must be hit each time.
+
+my $counter=0;
+
+sub response {
+ my $r=shift;
+ $r->content_type('text/plain');
+ $r->print($counter);
+ return Apache2::Const::OK;
+}
+
+sub count { $counter++; return Apache2::Const::DECLINED; }
+
+sub clear_pool {
+ delete $_[0]->pnotes->{my_pool};
+ return Apache2::Const::DECLINED;
+}
+
+sub trans {
+ my $r=shift;
+ my $test=$r->args;
+ $counter=0;
+ if( $test eq '1' ) {
+ # this is to check for a bug in modperl_response_handler versus
+ # modperl_response_handler_cgi. The former used to allocate an
+ # extra interpreter for its work. In both cases $counter should be
+ # 2 in the response phase
+ $r->push_handlers( PerlMapToStorageHandler=>__PACKAGE__.'::count' );
+ $r->push_handlers( PerlFixupHandler=>__PACKAGE__.'::count' );
+ }
+ elsif( $test eq '2' ) {
+ # now add an extra PerlCleanupHandler. It is run each time the
+ # interp is released. So it is run after Trans, MapToStorage and
+ # Fixup. In the response phase $counter should be 5. After Response
+ # it is run again but that is after.
+ # This used to eat up all interpreters because modperl_interp_unselect
+ # calls modperl_config_request_cleanup that allocates a new interp
+ # to handle the cleanup. When this interp is then unselected
+ # modperl_interp_unselect gets called again but the cleanup handler is
+ # still installed. So the cycle starts again until all interpreters
+ # are in use or the stack runs out. Then the thread is locked infinitely
+ # or a segfault appears.
+ $r->push_handlers( PerlMapToStorageHandler=>__PACKAGE__.'::count' );
+ $r->push_handlers( PerlFixupHandler=>__PACKAGE__.'::count' );
+ $r->push_handlers( PerlCleanupHandler=>__PACKAGE__.'::count' );
+ }
+ elsif( $test eq '3' ) {
+ # a subpool adds an extra reference to the interp. So it is preserved
+ # and bound to the request until the pool is destroyed. So the cleanup
+ # handler is run only once after Fixup. Hence the counter is 3.
+ $r->push_handlers( PerlMapToStorageHandler=>__PACKAGE__.'::count' );
+ $r->push_handlers( PerlFixupHandler=>__PACKAGE__.'::count' );
+ $r->push_handlers( PerlCleanupHandler=>__PACKAGE__.'::count' );
+ $r->pnotes->{my_pool}=$r->pool->new;
+ $r->push_handlers( PerlFixupHandler=>__PACKAGE__.'::clear_pool' );
+ }
+ return Apache2::Const::DECLINED;
+}
+
+1;
+
+__END__
+# APACHE_TEST_CONFIG_ORDER 942
+
+<VirtualHost TestPerl::ithreads3>
+
+ <IfDefine PERL_USEITHREADS>
+ # a new interpreter pool
+ PerlOptions +Parent
+ PerlInterpStart 3
+ PerlInterpMax 3
+ PerlInterpMinSpare 1
+ PerlInterpMaxSpare 3
+ PerlInterpScope handler
+ </IfDefine>
+
+ # use test system's @INC
+ PerlSwitches [EMAIL PROTECTED]@
+ PerlRequire "conf/modperl_inc.pl"
+ PerlModule TestPerl::ithreads3
+
+ <Location /modperl>
+ SetHandler modperl
+ PerlResponseHandler TestPerl::ithreads3::response
+ </Location>
+
+ <Location /perl-script>
+ SetHandler perl-script
+ PerlResponseHandler TestPerl::ithreads3::response
+ </Location>
+
+ PerlTransHandler TestPerl::ithreads3::trans
+
+</VirtualHost>
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
