[1497] in java-interest
Re: lookupswitch byte code
daemon@ATHENA.MIT.EDU (Jeremy Fitzhardinge)
Wed Sep 6 04:33:15 1995
From: "Jeremy Fitzhardinge" <jeremy@suede.sw.oz.au>
Date: Wed, 6 Sep 1995 15:39:32 -0500
In-Reply-To: erik@kroete2.freinet.de (Erik Corry)
"lookupswitch byte code" (Sep 6, 1:35am)
To: ehcorry@inet.uni-c.dk (Erik Corry)
Cc: java-interest@java.sun.com
On Sep 6, 1:35am, Erik Corry wrote:
> Subject: lookupswitch byte code
> Looking at the description of the 'lookupswitch' byte code on
>
> http://java.sun.com/1.0alpha3/doc/vmspec/vmspec_28.html#HEADING177
>
> (which is presumably designed to implement sparse switch statements,)
> I notice that the match-offset pairs are not guaranteed to be sorted
> by match. This means that a binary chop search is not possible. Is
> this an omission in the docs, in the definition, or is there a good
> reason for it?
There's no reason a runtime couldn't sort the table in memory after loading
the instruction.
J
-
Note to Sun employees: this is an EXTERNAL mailing list!
Info: send 'help' to java-interest-request@java.sun.com
