summaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorNick Ing-Simmons <nik@tiuk.ti.com>1999-08-14 09:20:58 +0000
committerNick Ing-Simmons <nik@tiuk.ti.com>1999-08-14 09:20:58 +0000
commita7c6d24429ab2b6db54575a3bdc62c7ed9f881cf (patch)
tree7d35f0d7ea024de4150a892601d0a84bcd7d1c59
parent7e291ef107a151a0b630a8ce85d6a562438da741 (diff)
downloadperl-a7c6d24429ab2b6db54575a3bdc62c7ed9f881cf.tar.gz
New lightweight Carp has a require. If Carp is used in a __DIE__ handler
this causes a POPSTACK panic. The problem seems to be that although die_where() has unwound the tail of perl_vdie() top_env setjmp has been set to resume execution there. Avoiding setting CATCH_SET(TRUE) in call_sv() avoids this. So invent a new G_NOCATCH flag to disable messing with CATCH_SET() in call_sv, use it in perl_vdie(). Add test to op/eval.t which will fail (panic) if bug comes back. >>> I AM NOT CONVINCED THIS IS CORRECT LONG TERM FIX <<< p4raw-id: //depot/perl@3988
-rw-r--r--cop.h1
-rw-r--r--perl.c10
-rwxr-xr-xt/op/eval.t7
-rw-r--r--util.c6
4 files changed, 20 insertions, 4 deletions
diff --git a/cop.h b/cop.h
index 84afd58967..f23251b543 100644
--- a/cop.h
+++ b/cop.h
@@ -297,6 +297,7 @@ struct context {
#define G_NOARGS 8 /* Don't construct a @_ array. */
#define G_KEEPERR 16 /* Append errors to $@, don't overwrite it */
#define G_NODEBUG 32 /* Disable debugging at toplevel. */
+#define G_NOCATCH 64 /* Don't do CATCH_SET() */
/* flag bits for PL_in_eval */
#define EVAL_NULL 0 /* not in an eval */
diff --git a/perl.c b/perl.c
index d81187989f..8c8b84c387 100644
--- a/perl.c
+++ b/perl.c
@@ -1225,10 +1225,16 @@ Perl_call_sv(pTHX_ SV *sv, I32 flags)
PL_op->op_private |= OPpENTERSUB_DB;
if (!(flags & G_EVAL)) {
- CATCH_SET(TRUE);
+ /* G_NOCATCH is a hack for perl_vdie using this path to call
+ a __DIE__ handler */
+ if (!(flags & G_NOCATCH)) {
+ CATCH_SET(TRUE);
+ }
call_xbody((OP*)&myop, FALSE);
retval = PL_stack_sp - (PL_stack_base + oldmark);
- CATCH_SET(FALSE);
+ if (!(flags & G_NOCATCH)) {
+ CATCH_SET(FALSE);
+ }
}
else {
cLOGOP->op_other = PL_op;
diff --git a/t/op/eval.t b/t/op/eval.t
index dc163e9e8f..abcb3794b7 100755
--- a/t/op/eval.t
+++ b/t/op/eval.t
@@ -1,6 +1,6 @@
#!./perl
-print "1..36\n";
+print "1..37\n";
eval 'print "ok 1\n";';
@@ -171,3 +171,8 @@ sub terminal { eval 'print $r' }
}
$x++;
+# Have we cured panic which occurred with require/eval in die handler ?
+$SIG{__DIE__} = sub { eval {1}; die shift };
+eval { die "ok ".$x++,"\n" };
+print $@;
+
diff --git a/util.c b/util.c
index 132ec5e127..715a5f1a15 100644
--- a/util.c
+++ b/util.c
@@ -1491,7 +1491,11 @@ Perl_vdie(pTHX_ const char* pat, va_list *args)
PUSHMARK(SP);
XPUSHs(msg);
PUTBACK;
- call_sv((SV*)cv, G_DISCARD);
+ /* HACK - REVISIT - avoid CATCH_SET(TRUE) in call_sv()
+ or we come back here due to a JMPENV_JMP() and do
+ a POPSTACK - but die_where() will have already done
+ one as it unwound - NI-S 1999/08/14 */
+ call_sv((SV*)cv, G_DISCARD|G_NOCATCH);
POPSTACK;
LEAVE;
}