Account Options

  1. Sign in
The old Google Groups will be going away soon, but your browser is incompatible with the new version.
Google Groups Home
« Groups Home
v8-dev
Home
+ new post
About this group
Join this group
Add rotate-right instruction to hydrogen and use it instead of bitwise operations (issue 11033005)
Options
There are currently too many topics in this group that display first. To make this topic appear first, remove this option from another topic.
There was an error processing your request. Please try again.
flag
   4 messages - Collapse all  -  Translate all to Translated (View all originals)
The group you are posting to is a Usenet group. Messages posted to this group will make your email address visible to anyone on the Internet.
Your reply message has not been sent.
Your post was successful
 
From:
To:
Cc:
Followup To:
Add Cc | Add Followup-to | Edit Subject
Subject:
Validation:
For verification purposes please type the characters you see in the picture below or the numbers you hear by clicking the accessibility icon. Listen and type the numbers you hear
 
u...@chromium.org  
View profile  
 More options Oct 1 2012, 2:15 pm
From: u...@chromium.org
Date: Mon, 01 Oct 2012 18:15:19 +0000
Local: Mon, Oct 1 2012 2:15 pm
Subject: Add rotate-right instruction to hydrogen and use it instead of bitwise operations (issue 11033005)
Print | Individual message | Show original | Report this message | Find messages by this author
Reviewers: danno, Michael Starzinger, Jay Conrod,

Message:
Please take a look.

https://chromiumcodereview.appspot.com/11033005/diff/1/src/hydrogen-i...
File src/hydrogen-instructions.h (right):

https://chromiumcodereview.appspot.com/11033005/diff/1/src/hydrogen-i...
src/hydrogen-instructions.h:902: bool deleteable_operation =
IsBitwiseBinaryOperation();
danno, michael: any suggestions for other instruction types to add here?

https://chromiumcodereview.appspot.com/11033005/diff/1/src/hydrogen.cc
File src/hydrogen.cc (right):

https://chromiumcodereview.appspot.com/11033005/diff/1/src/hydrogen.c...
src/hydrogen.cc:3414: HPhase phase("H_EliminateUnusedInstructions",
this);
Maybe I should just do one pass instead of this complex worklist
algorithm?

Description:
Add rotate-right instruction to hydrogen and use it instead of bitwise
operations
of the form ((x >>> i) | (x << (32 - i))).

This CL is based on https://chromiumcodereview.appspot.com/10984057/
by Jay Conrod <dcon...@codeaurora.org>.

R=da...@chromium.org,mstarzin...@chromium.org,dcon...@codeaurora.org

Please review this at https://chromiumcodereview.appspot.com/11033005/

SVN Base: https://v8.googlecode.com/svn/branches/bleeding_edge

Affected files:
   M src/arm/lithium-arm.cc
   M src/arm/lithium-codegen-arm.cc
   M src/arm/simulator-arm.cc
   M src/hydrogen-instructions.h
   M src/hydrogen.h
   M src/hydrogen.cc
   M src/ia32/assembler-ia32.h
   M src/ia32/assembler-ia32.cc
   M src/ia32/lithium-codegen-ia32.cc
   M src/ia32/lithium-ia32.cc
   M src/token.h
   M src/x64/assembler-x64.h
   M src/x64/lithium-codegen-x64.cc
   M src/x64/lithium-x64.cc
   A test/mjsunit/compiler/rotate.js


 
You must Sign in before you can post messages.
To post a message you must first join this group.
Please update your nickname on the subscription settings page before posting.
You do not have the permission required to post.
dcon...@codeaurora.org  
View profile  
 More options Oct 1 2012, 3:40 pm
From: dcon...@codeaurora.org
Date: Mon, 01 Oct 2012 19:40:40 +0000
Local: Mon, Oct 1 2012 3:40 pm
Subject: Re: Add rotate-right instruction to hydrogen and use it instead of bitwise operations (issue 11033005)
Print | Individual message | Show original | Report this message | Find messages by this author
LGTM. I was worried having pattern matching inside the graph building phase
would overcomplicate things, but it actually seems simpler since you don't  
have
to worry about representation or HChanges. I can't really comment on the
ia32/x64 parts. Does MIPS need support, too?

http://codereview.chromium.org/11033005/


 
You must Sign in before you can post messages.
To post a message you must first join this group.
Please update your nickname on the subscription settings page before posting.
You do not have the permission required to post.
dcon...@codeaurora.org  
View profile  
 More options Oct 1 2012, 3:41 pm
From: dcon...@codeaurora.org
Date: Mon, 01 Oct 2012 19:41:17 +0000
Local: Mon, Oct 1 2012 3:41 pm
Subject: Re: Add rotate-right instruction to hydrogen and use it instead of bitwise operations (issue 11033005)
Print | Individual message | Show original | Report this message | Find messages by this author
lgtm

http://codereview.chromium.org/11033005/diff/2001/src/arm/simulator-a...
File src/arm/simulator-arm.cc (right):

http://codereview.chromium.org/11033005/diff/2001/src/arm/simulator-a...
src/arm/simulator-arm.cc:1503: } else {
Erik Corry expressed some suspicion in my first rev about the shifts,
and I agree. "result" is signed (int32_t), so >> will sign extend. Also
*carry_out should contain the last bit that was shifted, so it can be
(static_cast<uint32_t>(result) >> 31) != 0, right?

http://codereview.chromium.org/11033005/


 
You must Sign in before you can post messages.
To post a message you must first join this group.
Please update your nickname on the subscription settings page before posting.
You do not have the permission required to post.
u...@chromium.org  
View profile  
 More options Oct 2 2012, 8:20 am
From: u...@chromium.org
Date: Tue, 02 Oct 2012 12:20:26 +0000
Local: Tues, Oct 2 2012 8:20 am
Subject: Re: Add rotate-right instruction to hydrogen and use it instead of bitwise operations (issue 11033005)
Print | Individual message | Show original | Report this message | Find messages by this author
I uploaded a new patch set that addresses the comments and fixes a
representation bug.

http://codereview.chromium.org/11033005/diff/2001/src/arm/simulator-a...
File src/arm/simulator-arm.cc (right):

http://codereview.chromium.org/11033005/diff/2001/src/arm/simulator-a...
src/arm/simulator-arm.cc:1503: } else {
On 2012/10/01 19:41:17, Jay Conrod wrote:

> Erik Corry expressed some suspicion in my first rev about the shifts,
and I
> agree. "result" is signed (int32_t), so >> will sign extend. Also
*carry_out
> should contain the last bit that was shifted, so it can be
> (static_cast<uint32_t>(result) >> 31) != 0, right?

Yes, you're right. I am taking the updated implementation that you
uploaded in the other CL.

http://codereview.chromium.org/11033005/


 
You must Sign in before you can post messages.
To post a message you must first join this group.
Please update your nickname on the subscription settings page before posting.
You do not have the permission required to post.
End of messages
« Back to Discussions « Newer topic     Older topic »

Create a group - Google Groups - Google Home - Terms of Service - Privacy Policy
©2013 Google