diff options
author | Nicholas Clark <nick@ccl4.org> | 2003-10-31 18:54:26 +0000 |
---|---|---|
committer | Nicholas Clark <nick@ccl4.org> | 2003-10-31 18:54:26 +0000 |
commit | 838efab482d30d6a82835e2296e673cfa68c80a7 (patch) | |
tree | 44feb052979f1037ba9b3f1bc2629f5d44048b92 /op.c | |
parent | f7218ed4fb64f81a57ca3333cb9e28f01f115558 (diff) | |
download | perl-838efab482d30d6a82835e2296e673cfa68c80a7.tar.gz |
Comment that this "optimisation" is actually a necessary fixup.
When the split occurs (to make the opti{miser,onal}) this isn't
(Richard Clamp reminded me that I forgot to do this)
p4raw-id: //depot/perl@21586
Diffstat (limited to 'op.c')
-rw-r--r-- | op.c | 4 |
1 files changed, 4 insertions, 0 deletions
@@ -6277,6 +6277,10 @@ Perl_peep(pTHX_ register OP *o) o->op_seq = PL_op_seqmax++; break; case OP_STUB: + /* XXX This makes sub {}; work as expected. + ie {return;} not {return @_;} + When optimiser is properly split into fixups and + optimisations, this needs to stay in the fixups. */ if(!oldop && o->op_next && o->op_next->op_type == OP_LEAVESUB) { |