Mailing-List: contact dev-help@harmony.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@harmony.apache.org
Received-SPF: pass (herse.apache.org: domain of pmcfirst@gmail.com designates
 209.85.132.240 as permitted sender)
DomainKey-Signature: a=rsa-sha1; c=nofws;
        d=gmail.com; s=beta;
        h=received:message-id:date:from:to:subject:in-reply-to:mime-version:content-type:references;
        b=o7jjZ5cOCpuuU7kt3S9lB66y8jrmRYNDGdW38SUOGmRVzboNoX/tHvF0j/5YefT494wwcwTbABFie0eqfhcPz8WCKCpp0XPwiOmCccitP7f4GtVdAhkzf7rdI5DRHzNEQ+IKepQFWppOk+MeJgT9AeRm3D3v0bKtbpt5cW1Qagw=
Message-ID: <e0f125db0705030117q5b31105bo425620a39506a1c3@mail.gmail.com>
Date: Thu, 3 May 2007 12:17:35 +0400
From: "Pavel Pervov" <pmcfirst@gmail.com>
To: dev@harmony.apache.org
Subject: Re: [drlvm][jvmti] RedefineClasses.
In-Reply-To: <f1c5aj$sic$1@sea.gmane.org>
MIME-Version: 1.0
Content-Type: multipart/alternative;
	boundary="----=_Part_24955_5463625.1178180255270"
References: <e0f125db0705010956r195a6c26s35afc811b27fe13e@mail.gmail.com>
	 <0vqslagsa75.fsf@gmail.com>
	 <e0f125db0705012148s460c745di9d566b406957d382@mail.gmail.com>
	 <46382A1D.8000009@gmail.com>
	 <e0f125db0705022230w227842f0mf6a88fc4c5efd0f9@mail.gmail.com>
	 <f1c5aj$sic$1@sea.gmane.org>

------=_Part_24955_5463625.1178180255270
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
Content-Disposition: inline

George,

Constant pool indexes are used in class' bytecodes. We should either:
1) parse methods bytecodes in class file parser and provide per-instruction
interface to all other users of bytecode.
2) "Patch" bytecode with 32-bit constant pool indexes.

Both will affect EEs and verifier. I, personally, would prefer the first
one, cause it will allow to remove four other bytecode parsers we have now.

Pavel.


On 5/3/07, George Timoshenko <george.timoshenko@gmail.com> wrote:
>
> I see now. Thanks for clarification.
>
> As we are using "merged cp" inside DRLVM, can we just not take care of
> exceeding 16-bit boundary and keep cp indexes as 32-bit values?
>
> Or we export this information somewhere...?
>
>
> Pavel Pervov wrote:
> > In current codebase "merged" constant pool will cause many problems for
> > interpreter and one problem for JIT - if call to
> "throw_linking_exception"
> > stub is compiled into some method. As you may remember it contains
> > immediate
> > value of constant pool index of a referenced class.
> >
> > Also, Mikhail noted new functionality currently being implemented - lazy
> > resolution. It will also be heavily affected with constant pool merging.
> >
> > Regards,
> >    Pavel.
>
>


-- 
Pavel Pervov,
Intel Enterprise Solutions Software Division

------=_Part_24955_5463625.1178180255270--