Received: by 10.42.150.2 with SMTP id y2mr1760599icv.7.1333224246698;
        Sat, 31 Mar 2012 13:04:06 -0700 (PDT)
X-BeenThere: v8-dev@googlegroups.com
Received: by 10.50.188.137 with SMTP id ga9ls4169751igc.0.gmail; Sat, 31 Mar
 2012 13:04:05 -0700 (PDT)
Received: by 10.50.45.193 with SMTP id p1mr1772055igm.1.1333224245207;
        Sat, 31 Mar 2012 13:04:05 -0700 (PDT)
Received: by 10.50.45.193 with SMTP id p1mr1772054igm.1.1333224245199;
        Sat, 31 Mar 2012 13:04:05 -0700 (PDT)
Return-Path: <3NGN3TwoJBlk5I9B.3FIIP7D19C.3FDMZ-45M7FF7C57IFLGJ....@m3kw2wvrgufz5godrsrytgd7.apphosting.bounces.google.com>
Received: from mail-ob0-f198.google.com (mail-ob0-f198.google.com [209.85.214.198])
        by gmr-mx.google.com with ESMTPS id us5si1601751igc.0.2012.03.31.13.04.05
        (version=TLSv1/SSLv3 cipher=OTHER);
        Sat, 31 Mar 2012 13:04:05 -0700 (PDT)
Received-SPF: pass (google.com: domain of 3NGN3TwoJBlk5I9B.3FIIP7D19C.3FDMZ-45M7FF7C57IFLGJ....@m3kw2wvrgufz5godrsrytgd7.apphosting.bounces.google.com designates 209.85.214.198 as permitted sender) client-ip=209.85.214.198;
Authentication-Results: gmr-mx.google.com; spf=pass (google.com: domain of 3NGN3TwoJBlk5I9B.3FIIP7D19C.3FDMZ-45M7FF7C57IFLGJ....@m3kw2wvrgufz5godrsrytgd7.apphosting.bounces.google.com designates 209.85.214.198 as permitted sender) smtp.mail=3NGN3TwoJBlk5I9B.3FIIP7D19C.3FDMZ-45M7FF7C57IFLGJ....@m3kw2wvrgufz5godrsrytgd7.apphosting.bounces.google.com
Received: by obbup16 with SMTP id up16so6210058obb.5
        for <v8-dev@googlegroups.com>; Sat, 31 Mar 2012 13:04:04 -0700 (PDT)
MIME-Version: 1.0
Received: by 10.50.154.231 with SMTP id vr7mr1916044igb.2.1333224244847; Sat,
 31 Mar 2012 13:04:04 -0700 (PDT)
Reply-To: erik.co...@gmail.com, u...@chromium.org, l...@google.com, 
	l...@chromium.org, v8-dev@googlegroups.com
Message-ID: <14dae93407c185277204bc8f75db@google.com>
Date: Sat, 31 Mar 2012 20:04:04 +0000
Subject: Re: RegExp: Add support for table-based character class (issue 9854020)
From: erik.co...@gmail.com
To: u...@chromium.org, l...@google.com, l...@chromium.org
Cc: v8-dev@googlegroups.com
Content-Type: text/plain; charset=ISO-8859-1; format=flowed; delsp=yes


https://chromiumcodereview.appspot.com/9854020/diff/11001/src/x64/regexp-macro-assembler-x64.cc
File src/x64/regexp-macro-assembler-x64.cc (right):

https://chromiumcodereview.appspot.com/9854020/diff/11001/src/x64/regexp-macro-assembler-x64.cc#newcode602
src/x64/regexp-macro-assembler-x64.cc:602: __ and_(rbx,
Immediate(kTableMask));
On 2012/03/30 07:46:28, Erik Corry wrote:
> On 2012/03/29 15:25:32, Lasse Reichstein wrote:
> > I don't think the register renamer will do that since the value of
> > current_character() needs to be retained as well. A move will use a
different
> > physical register.

> My understanding is that after renaming you have gone from 2-register
> instructions to 3-register instructions so the mov disappears.

But I tested it and your version is marginally faster.

https://chromiumcodereview.appspot.com/9854020/