[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: RE : [Pnet-developers] [Bug #4014] [0.5.8] Hard limit for bytecode
From: |
Rhys Weatherley |
Subject: |
Re: RE : [Pnet-developers] [Bug #4014] [0.5.8] Hard limit for bytecode size is different in VM bytecode verifier and ilverify |
Date: |
Wed, 18 Jun 2003 21:18:33 +1000 |
User-agent: |
KMail/1.4.3 |
On Wednesday 18 June 2003 08:40 pm, Yannis BRES wrote:
> I was not talking about the stack size but the bytecode size : there
> does not seem to be any option in this respect...
Could you send me an example of a binary that has this problem by private
e-mail (or put it on the Web somewhere)? I will investigate.
The method cache is allocated in blocks of size 128 k, which does effectively
limit how big the CVM version of a method can be. Do you really have
individual methods that are that big?
An overly large method may appear to be a verfication failure because of the
way the CVM coder works. The verifier, ilverify, wouldn't exhibit this same
problem because it uses a null coder that doesn't generate CVM code.
Cheers,
Rhys.