Re: (fyi) SEGV in jc1 for java.util.Date (from jikes bytecode only)

[Jon Olson <olson@mmsi.com>]

On Sun, 09 May 1999, Anthony Green wrote:
>Sandro wrote:
>> I'd send along the .class file in question, but I'm not sure what kind
>> of clean-room proceedures you might be using.
>
>You would be violating Sun's Binary Code License in doing so.  Even
>so, debugging this...
>
>> java/util/Date.java: In method `clone()':
>> java/util/Date.java:281: warning: Some bytecode operations (starting at pc 34) can never be executed
>
>...would require disassembling the class file, which is also in
>vviolation of the Binary Code License.
>
>It's likely that Jikes is doing something we do not expect.  A better
>solution would be to try to reproduce this problem with Jikes and some
>Free code.
>
>AG
>

I've found that jikes generates really terrible byte code.  The problem with
the error above is caused by jikes routinely generating code immediately
after an `athrow' instruction.  For byte code compilers, guavac generates
incorrect code, jikes generates terrible code.  The only good byte code
compiler I've found is Sun's `javac'.

 --
Jon Olson, Modular Mining Systems
	   3289 E. Hemisphere Loop
	   Tucson, AZ 85706
INTERNET:  olson@mmsi.com
PHONE:     (520)746-9127
FAX:       (520)889-5790