SIGSEGV in jc1

Jeff Sturm jsturm@one-point.com
Tue May 8 22:38:00 GMT 2001


On 8 May 2001, Per Bothner wrote:
> Could some other people try Hans's testcase using a branch or trunk
> tree, and report the result?

Sure. With a fresh branch on i686-pc-linux-gnu:

$ make rpn_calc
/opt/java/bin/javac -d . UnaryCRFunction.java CR.java
PrecisionOverflowError.java AbortedError.java TestCR.java
gcj -O -o rpn_calc --main=rpn_calc rpn_calc.java com/sgi/math/*.class
$ ./rpn_calc
1 2 +
3.00

Building from source seems to work too:

$ gcj -O -o rpn_calc --main=rpn_calc *.java
$ ./rpn_calc
3 4 *
12.00

except when building TestCR:

$ gcj -O -o TestCR --main=com.sgi.math.TestCR *.java
$ ./TestCR
Segmentation fault (core dumped)

Strange, that one.

Jeff




More information about the Java mailing list