Chromium Code Reviews
Help | Chromium Project | Sign in
(43)

Issue 5733004: Attempt to fix compare performance regression.

Can't Edit
Can't Publish+Mail
Start Review
Created:
4 years, 8 months ago by Vitaly Repeshko
Modified:
4 years, 8 months ago
Reviewers:
fschneider
CC:
v8-dev
Visibility:
Public.

Description

Attempt to fix compare performance regression.

Patch Set 1 #

Total comments: 1
Unified diffs Side-by-side diffs Delta from patch set Stats (+21 lines, -6 lines) Patch
M src/ia32/ic-ia32.cc View 1 chunk +12 lines, -1 line 0 comments Download
M src/ic.h View 1 chunk +2 lines, -1 line 0 comments Download
M src/ic.cc View 1 chunk +7 lines, -4 lines 1 comment Download
Trybot results: Sign in to try more bots
Project "None" does not have a commit queue.

Messages

Total messages: 2 (0 generated)
Vitaly Repeshko
4 years, 8 months ago (2010-12-11 02:20:28 UTC) #1
fschneider
4 years, 8 months ago (2010-12-11 15:18:46 UTC) #2
LGTM!

http://codereview.chromium.org/5733004/diff/1/src/ic.cc
File src/ic.cc (right):

http://codereview.chromium.org/5733004/diff/1/src/ic.cc#newcode2142
src/ic.cc:2142: x->IsNumber() && y->IsNumber()) return HEAP_NUMBERS;
Would the condition 
(state == UNINITIALIZED || state == SMIS) &&
x->IsNumber() &&
y->IsNumber())

be sufficient here? So that we always go from SMI to HEAP_NUMBER before going to
GENERIC.
Sign in to reply to this message.

Powered by Google App Engine
RSS Feeds Recent Issues | This issue
This is Rietveld c33a7a4