From: Eric Wong <e@80x24.org>
To: spew@80x24.org
Subject: [PATCH] fix optimization for hash aset/aref with fstring
Date: Thu, 22 Dec 2016 00:47:52 +0000 [thread overview]
Message-ID: <20161222004752.4086-1-e@80x24.org> (raw)
I don't like the idea of making insns.def any bigger to support
a corner case, and "test_hash_aref_fstring_identity" shows
how contrived this is.
[ruby-core:78783] [Bug #12855]
---
hash.c | 4 +---
insns.def | 8 ++++++--
internal.h | 1 +
test/ruby/test_hash.rb | 13 +++++++++++++
4 files changed, 21 insertions(+), 5 deletions(-)
diff --git a/hash.c b/hash.c
index 1867c09..b128cb3 100644
--- a/hash.c
+++ b/hash.c
@@ -2758,8 +2758,6 @@ rb_hash_compact_bang(VALUE hash)
return Qnil;
}
-static VALUE rb_hash_compare_by_id_p(VALUE hash);
-
/*
* call-seq:
* hsh.compare_by_identity -> hsh
@@ -2795,7 +2793,7 @@ rb_hash_compare_by_id(VALUE hash)
*
*/
-static VALUE
+VALUE
rb_hash_compare_by_id_p(VALUE hash)
{
if (!RHASH(hash)->ntbl)
diff --git a/insns.def b/insns.def
index b64a168..41e4498 100644
--- a/insns.def
+++ b/insns.def
@@ -1881,7 +1881,9 @@ opt_aset_with
(VALUE recv, VALUE val)
(VALUE val)
{
- if (!SPECIAL_CONST_P(recv) && RBASIC_CLASS(recv) == rb_cHash && BASIC_OP_UNREDEFINED_P(BOP_ASET, HASH_REDEFINED_OP_FLAG)) {
+ if (!SPECIAL_CONST_P(recv) && RBASIC_CLASS(recv) == rb_cHash &&
+ BASIC_OP_UNREDEFINED_P(BOP_ASET, HASH_REDEFINED_OP_FLAG) &&
+ rb_hash_compare_by_id_p(recv) == Qfalse) {
rb_hash_aset(recv, key, val);
}
else {
@@ -1903,7 +1905,9 @@ opt_aref_with
(VALUE recv)
(VALUE val)
{
- if (!SPECIAL_CONST_P(recv) && RBASIC_CLASS(recv) == rb_cHash && BASIC_OP_UNREDEFINED_P(BOP_AREF, HASH_REDEFINED_OP_FLAG)) {
+ if (!SPECIAL_CONST_P(recv) && RBASIC_CLASS(recv) == rb_cHash &&
+ BASIC_OP_UNREDEFINED_P(BOP_AREF, HASH_REDEFINED_OP_FLAG) &&
+ rb_hash_compare_by_id_p(recv) == Qfalse) {
val = rb_hash_aref(recv, key);
}
else {
diff --git a/internal.h b/internal.h
index 7aefcda..63f5d36 100644
--- a/internal.h
+++ b/internal.h
@@ -1088,6 +1088,7 @@ long rb_objid_hash(st_index_t index);
long rb_dbl_long_hash(double d);
st_table *rb_init_identtable(void);
st_table *rb_init_identtable_with_size(st_index_t size);
+VALUE rb_hash_compare_by_id_p(VALUE hash);
#define RHASH_TBL_RAW(h) rb_hash_tbl_raw(h)
VALUE rb_hash_keys(VALUE hash);
diff --git a/test/ruby/test_hash.rb b/test/ruby/test_hash.rb
index b79f9aa..040d25e 100644
--- a/test/ruby/test_hash.rb
+++ b/test/ruby/test_hash.rb
@@ -237,6 +237,19 @@ def test_ASET_fstring_key
assert_same a.keys[0], b.keys[0]
end
+ def test_hash_aset_fstring_identity
+ h = {}.compare_by_identity
+ h['abc'] = 1
+ h['abc'] = 2
+ assert_equal 2, h.size, '[ruby-core:78783] [Bug #12855]'
+ end
+
+ def test_hash_aref_fstring_identity
+ h = {}.compare_by_identity
+ h['abc'] = 1
+ assert_nil h['abc'], '[ruby-core:78783] [Bug #12855]'
+ end
+
def test_NEWHASH_fstring_key
a = {"ABC" => :t}
b = {"ABC" => :t}
--
EW
reply other threads:[~2016-12-22 0:47 UTC|newest]
Thread overview: [no followups] expand[flat|nested] mbox.gz Atom feed
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20161222004752.4086-1-e@80x24.org \
--to=e@80x24.org \
--cc=spew@80x24.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).