-----Original Message-----
From: Levente Farkas <[EMAIL PROTECTED]>
To: Mo DeJong <[EMAIL PROTECTED]>; Java-Linux
<[EMAIL PROTECTED]>; [EMAIL PROTECTED] <[EMAIL PROTECTED]>;
[EMAIL PROTECTED] <[EMAIL PROTECTED]>
Date: Tuesday, December 19, 2000 4:42 PM
Subject: Re: What are the disadvantages of using jikes instead of javac?
>Mo DeJong wrote:
>>
>> On Mon, 18 Dec 2000, Joseph Shraibman wrote:
>>
>> > In at least a few instances that I know of, jikes produces slower code.
>> > But I don't think javac is a really great optimizing compiler either.
>>
>> That does not matter. Java code should be optimized by
>> the JIT not the Java compiler. On a pure interpreted JVM
>> it might matter a bit, but who uses those these days?
>> Besides, you could just switch to javac when you compile
>> the final executable, if it mattered.
>>
>> > There is at least one bug that causes jikes to crash where javac
>> > compiles the code. Don't remember exactly what it was, I think it was
>> > something like private inner classes. I just made them non-private and
>> > they compiled fine.
>>
>> That is a known bug, it is covered by this Jacks regression test:
>>
>> 8.8.5.1-accessible-explicit-super-invocation-args-3
>>
>> class T8851aesia3_super {
>> private T8851aesia3_super(int i) {}
>>
>> static class T8851aesia3 extends T8851aesia3_super {
>> T8851aesia3() { super(1); }
>> }
>> }
>>
>> It should be fixed in the 1.13 release.
>>
>> There are also cases where javac compiles your
>> code even though it is not legal. For example:
>>
>> class T8851mti2 {
>> T8851mti2() { this(0); }
>> T8851mti2(int i) { this(); }
>> }
>>
>> Jikes correctly prints an error in this
>> case, since the function would loop forever.
>>
>> Javac does not.
>
>so if it's time to send another bug report of jikes (or javac or there
>is no standard way?)
>the static inicializer:
>------------
>static
>{
>//...
>}
>------------
>with jikes compiles as final (if you look into the bytecode):
>static final {}
>while with sun's javac
>static {}
>one can think it's not realy a matter, BUT when you generate the
>serialVersionUID with jdk 1.3 (jdk 1.1 it ok) gives different number !!!
>so if you compile the same code sometimes with jikes and on another site
>with javac and try to use them (eg RMI which serialize the object)
>than the same object seems different (since the serialversions are
>different)!
>and it's realy hard to find;-(((
>
>
> -- Levente http://petition.eurolinux.org/index_html
> "The only thing worse than not knowing the truth is
> ruining the bliss of ignorance."
>
>
>----------------------------------------------------------------------
>To UNSUBSCRIBE, email to [EMAIL PROTECTED]
>with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
>
Jesus Christ Never Said That And Who The Fuck Is Alice.
YOURS SINCERELY
SHIVAKANTH
----------------------------------------------------------------------
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
