gcj/247: Lossage with BigInteger.java (probable compiler bug)

[Bryce McKinlay]

The following reply was made to PR gcj/247; it has been noted by GNATS.

From: Bryce McKinlay <bryce@albatross.co.nz>
To: zastai@hotmail.com
Cc: java-gnats@sourceware.cygnus.com
Subject: Re: gcj/247: Lossage with BigInteger.java (probable compiler bug)
Date: Fri, 26 May 2000 20:15:39 +1200

 Are you trying to build a cvs/snapshot libgcj with gcc 2.95.2? Did you
 apply my 2.95.2 java patches (see http://waitaki.otago.ac.nz/~bryce/ )
 before trying to do this? If you didn't, you will have a lot of
 difficulty building recent libgcjs.
 
 This is a known compiler bug that was fixed a long time ago: see PR
 gcj/155.
 
 regards
 
   [ bryce ]
 
 zastai@hotmail.com wrote:
 
 > BigInteger.java: In class `java.math.BigInteger':
 > BigInteger.java: In method `isNegative()':
 > BigInteger.java:283: internal error - SP mismatch