Handle accessors on the prototype chain in StoreICs. (issue 10735003) - v8-dev

The old Google Groups will be going away soon, but your browser is incompatible with the new version.
Message from discussion Handle accessors on the prototype chain in StoreICs. (issue 10735003)
Received-SPF: pass (google.com: domain of rossb...@google.com designates 209.85.220.181 as permitted sender) client-ip=209.85.220.181;
MIME-Version: 1.0
In-Reply-To: <CAPD5VRttX6fNX=_MXuuGupmTtfZtoz=FA2jqxTZNCr9oRVa...@mail.gmail.com>
References: <bcaec5014a37eca90404c4016...@google.com> <CAPD5VRvacGUDU12WmySXNt9ibWCFvJLX1DDgogfT7nPafmx...@mail.gmail.com>
 <CAPD5VRttX6fNX=_MXuuGupmTtfZtoz=FA2jqxTZNCr9oRVa...@mail.gmail.com>
From: Andreas Rossberg <rossb...@google.com>
Date: Wed, 4 Jul 2012 16:14:49 +0200
Message-ID: <CADjR9NpUwpN-q+s+0KtqDme4NceHE0=3j38O2OpB4OCPL3O...@mail.gmail.com>
Subject: Re: Handle accessors on the prototype chain in StoreICs. (issue 10735003)
To: Vyacheslav Egorov <vego...@google.com>
Cc: svenpa...@chromium.org, rossb...@chromium.org, vego...@chromium.org, 
	v8-dev@googlegroups.com
Content-Type: multipart/alternative; boundary=bcaec501c5c097c5b204c401a85d

--bcaec501c5c097c5b204c401a85d
Content-Type: text/plain; charset=ISO-8859-1

On 4 July 2012 15:59, Vyacheslav Egorov <vego...@google.com> wrote:

> argh, gibberish should read:
>
> but should not the fixed semantics...
> --
> Vyacheslav Egorov
>
>
> On Wed, Jul 4, 2012 at 3:58 PM, Vyacheslav Egorov <vego...@google.com>
> wrote:
> > But it's should fixed semantics affect only cases when setter
> > _appears_ on the prototype and store site has a stub compiled for
> > assignment into holder?
>

Yes, as I said, it _probably_ doesn't matter. But when I recently fixed
most of the setters-appearing-on-prototype behaviour along with
readonliness, 370 tests broke. Which is why I introduced the flag as a
temporary measure. I didn't look at all of the failures individually, so I
don't know if setters were among the causes, or if it only was readonliness.

Either way, I intend to find out soon. ;)

--bcaec501c5c097c5b204c401a85d
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

<div class=3D"gmail_extra"><div class=3D"gmail_quote">On 4 July 2012 15:59,=
 Vyacheslav Egorov <span dir=3D"ltr">&lt;<a href=3D"mailto:vego...@google.c=
om" target=3D"_blank">vego...@google.com</a>&gt;</span> wrote:<br><blockquo=
te class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1px #ccc so=
lid;padding-left:1ex">

argh, gibberish should read:<br>
<br>
but should not the fixed semantics...<br>
--<br>
Vyacheslav Egorov<br>
<div class=3D""><div class=3D"h5"><br>
<br>
On Wed, Jul 4, 2012 at 3:58 PM, Vyacheslav Egorov &lt;<a href=3D"mailto:veg=
o...@google.com">vego...@google.com</a>&gt; wrote:<br>
&gt; But it&#39;s should fixed semantics affect only cases when setter<br>
&gt; _appears_ on the prototype and store site has a stub compiled for<br>
&gt; assignment into holder?<br></div></div></blockquote><div><br></div><di=
v>Yes, as I said, it _probably_ doesn&#39;t matter. But when I recently fix=
ed most of the setters-appearing-on-prototype behaviour along with readonli=
ness, 370 tests broke.=A0Which is why I introduced the flag as a temporary =
measure.=A0I didn&#39;t look at all of the failures individually, so I don&=
#39;t know if setters were among the causes, or if it only was readonliness=
.</div>

<div><br></div><div>Either way, I intend to find out soon. ;)</div></div></=
div>

--bcaec501c5c097c5b204c401a85d--
Create a group - Google Groups - Google Home - Terms of Service - Privacy Policy
Account Options
