Merge branch 'bisection-example-20220227' into blead

For: #19468

Porting/bisect-runner.pl

@@ -927,6 +927,66 @@ =head2 Point in time where code started to segfault is unknown
 
 =back
 
+=head2 Interaction of debug flags caused crash on C<-DDEBUGGING> builds
+
+=over 4
+
+=item * Problem
+
+In C<-DDEBUGGING> builds, the debug flags C<Xvt> would crash a program when
+F<strict.pm> was loaded via C<require> or C<use>.
+
+=item * Solution
+
+Two-stage solution.  In each stage, to shorten debugging time investigator
+made use of existing set of production releases of F<perl> built with
+C<-DDEBUGGING>.
+
+=over 4
+
+=item * Stage 1
+
+Investigator used existing C<-DDEBUGGING> builds to determine the production
+cycle in which crash first appeared.  Then:
+
+    .../perl/Porting/bisect.pl \
+        --start v5.20.0 \
+        --end v5.22.1 \
+        -DDEBUGGING \
+        --target miniperl \
+        --crash \
+        -- ./miniperl -Ilib -DXvt -Mstrict -e 1
+
+First bad commit was identified as
+L<ed958fa315|https://github.com/Perl/perl5/commit/ed958fa315>.
+
+=item * Stage 2
+
+A second investigator was able to create a reduction of the code needed to
+trigger a crash, then used this reduced case and the commit reported at the
+end of Stage 1 to further bisect.
+
+    .../perl/Porting/bisect.pl \
+        --start v5.18.4 \
+        --end ed958fa315 \
+        -DDEBUGGING \
+        --target miniperl \
+        --crash \
+        -- ./miniperl -Ilib -DXv -e '{ my $n=1; *foo= sub () { $n }; }'
+
+=back
+
+The first bisect determined the point at which code was introduced to
+F<strict.pm> that triggered the problem. With an understanding of the trigger,
+the second bisect then determined the point at which such a trigger started
+causing a crash.
+
+* Reference
+
+L<GH issue 193463|https://github.com/Perl/perl5/issues/19463>
+
+=back
+
 =head2 When did perl start failing to build on a certain platform using C<g++> as the C-compiler?
 
 =over 4

t/porting/known_pod_issues.dat

@@ -411,6 +411,7 @@ pod/perlsolaris.pod	Verbatim line length including indents exceeds 78 by	13
 pod/perltie.pod	Verbatim line length including indents exceeds 78 by	3
 pod/perltru64.pod	Verbatim line length including indents exceeds 78 by	1
 pod/perlwin32.pod	Verbatim line length including indents exceeds 78 by	1
+porting/bisect-runner.pl	Verbatim line length including indents exceeds 78 by	1
 porting/epigraphs.pod	Verbatim line length including indents exceeds 78 by	28
 porting/release_managers_guide.pod	Verbatim line length including indents exceeds 78 by	2
 porting/todo.pod	? Should you be using F<...> or maybe L<...> instead of	1