In perl.git, the branch blead has been updated
<http://perl5.git.perl.org/perl.git/commitdiff/b33a1a96a953a17844057156988e6402f194e0ba?hp=818355acb63e1db6f7c5d48165cd5afe6e8e0bc4>
- Log -----------------------------------------------------------------
commit b33a1a96a953a17844057156988e6402f194e0ba
Author: Father Chrysostomos <spr...@cpan.org>
Date: Mon Oct 20 17:52:36 2014 -0700
Increase $B::Deparse::VERSION to 1.30
M lib/B/Deparse.pm
commit 8538e2d4d47f3b235052cbaabfe1aa035ae18916
Author: Father Chrysostomos <spr...@cpan.org>
Date: Mon Oct 20 17:16:01 2014 -0700
Rmv restrictions on op tree structure for padrange
The padrange optimisation was only happening if the op tree was laid
out in a way that would allow B::Deparse easily to restore the origi-
nal op_next pointers (via $B::overlay) for the sake of deparsing.
It turns out that B::Deparse doesnât even use the op_next pointers
(except for deparsing âuseâ statements, which are not affected here),
so there is no need for these restrictions.
In the case of my($a,$b)=@_, @_ was expected to be the sibling of the
preceding pushmark-cum-padrange. Removing that restriction doesnât
allow the padrange optimisation to happen in any new cases as far as I
can tell. It just means there is less checking to do.
In the more general case of a list of lexicals with no =@_, this
allows padrange to be used for array and hash slices in conjunction
with a separate list-in-list-context optimisation, that formerly con-
flicted with padrange.
When we compile @a[$lex1,$lex2], we end up with a redundant list-in-
list-context:
7 aslice
1 pushmark
5 list
2 pushmark
3 padsv
4 padsv
6 padav (assuming the array is lexical
Pushmark pushes a mark on to the markstack, and list pops one, so
pushmark and list cancel each other out in list context.
In 5.18, when padrange was introduced, this was optimised like this:
5 aslice
1 pushmark
3 list
2 padrange
- padsv
- padsv
4 padav
Then a separate optimisation was added in 5.20 that allows a list
with a pushmark kid to turn into null ops, and this cancelled out
the padrange:
5 aslice
1 pushmark
- ex-list
- ex-pushmark
2 padsv
3 padsv
4 padav
Bunchmarks show that this is slightly faster than padrange with two
lexicals, but more lexicals would make padrange faster. This commit
brings us the goodness of both optimisations at once:
3 aslice
1 padrange
- ex-list
- ex-pushmark
- padsv
- padsv
2 padav
And it is faster than just one optimisation.
M lib/B/Deparse.pm
M op.c
commit 40ba772c3f3fa292e5635bb62c0a68609c27fcdb
Author: Father Chrysostomos <spr...@cpan.org>
Date: Mon Oct 20 16:51:55 2014 -0700
op.c: No need to check the name of *_
Checking against PL_defgv is sufficient.
M op.c
-----------------------------------------------------------------------
Summary of changes:
lib/B/Deparse.pm | 5 +----
op.c | 13 +------------
2 files changed, 2 insertions(+), 16 deletions(-)
diff --git a/lib/B/Deparse.pm b/lib/B/Deparse.pm
index 3866f32..53f4e6b 100644
--- a/lib/B/Deparse.pm
+++ b/lib/B/Deparse.pm
@@ -21,7 +21,7 @@ use B qw(class main_root main_start main_cv svref_2object
opnumber perlstring
CVf_METHOD CVf_LVALUE
PMf_KEEP PMf_GLOBAL PMf_CONTINUE PMf_EVAL PMf_ONCE
PMf_MULTILINE PMf_SINGLELINE PMf_FOLD PMf_EXTENDED PMf_EXTENDED_MORE);
-$VERSION = '1.29';
+$VERSION = '1.30';
use strict;
use vars qw/$AUTOLOAD/;
use warnings ();
@@ -358,9 +358,6 @@ sub _pessimise_walk {
type => OP_PUSHMARK,
name => 'pushmark',
private => ($op->private & OPpLVAL_INTRO),
- next => ($op->flags & OPf_SPECIAL)
- ? $op->sibling->first
- : $op->sibling,
};
}
diff --git a/op.c b/op.c
index c1672ed..6ac8f3f 100644
--- a/op.c
+++ b/op.c
@@ -11937,21 +11937,15 @@ Perl_rpeep(pTHX_ OP *o)
/* look for a pushmark -> gv[_] -> rv2av */
{
- GV *gv;
OP *rv2av, *q;
p = o->op_next;
if ( p->op_type == OP_GV
- && (gv = cGVOPx_gv(p)) && isGV(gv)
- && GvNAMELEN_get(gv) == 1
- && *GvNAME_get(gv) == '_'
- && GvSTASH(gv) == PL_defstash
+ && cGVOPx_gv(p) == PL_defgv
&& (rv2av = p->op_next)
&& rv2av->op_type == OP_RV2AV
&& !(rv2av->op_flags & OPf_REF)
&& !(rv2av->op_private & (OPpLVAL_INTRO|OPpMAYBE_LVSUB))
&& ((rv2av->op_flags & OPf_WANT) == OPf_WANT_LIST)
- && OP_SIBLING(o) == rv2av /* these two for Deparse */
- && cUNOPx(rv2av)->op_first == p
) {
q = rv2av->op_next;
if (q->op_type == OP_NULL)
@@ -11963,11 +11957,6 @@ Perl_rpeep(pTHX_ OP *o)
}
}
if (!defav) {
- /* To allow Deparse to pessimise this, it needs to be able
- * to restore the pushmark's original op_next, which it
- * will assume to be the same as OP_SIBLING. */
- if (o->op_next != OP_SIBLING(o))
- break;
p = o;
}
--
Perl5 Master Repository
