[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

upstream test case



Similar problem (jar to native compilation) reported in
http://gcc.gnu.org/PR18931
(closed)

the _exact_ bug is reported in
http://gcc.gnu.org/PR19505
(open)

test case PR19505.class still fails with gcj-snapshot (20050227)

GNU Java version 4.1.0 20050227 (experimental) (i486-linux-gnu)
        compiled by GNU C version 4.1.0 20050227 (experimental).
GGC heuristics: --param ggc-min-expand=47 --param ggc-min-heapsize=32066
...

Note comment #2 on http://gcc.gnu.org/PR19505 : gcj fails to compile
bytecode produced by Sun's javac, but not when compiling bytecode
produced by gcj (with -C) or ecj.

comment #2 is posted in 2005-01-25 (same as the failing gcj I am
using), but I am compiling a .jar produced by gcj (I guess bug was
introduced some time between 2004-12-28 and 2005-01-25)

I am testing again:

gcj-snapshot -O2 -Wl,-Bsymbolic -shared -fPIC -fjni -findirect-dispatch -o /tmp/jdt.jar.so /usr/share/eclipse/plugins/org.eclipse.jdt.ui_3.0.1/jdt.jar

now. It take long time, but problem with gcj 4.0 20050125 happen soon,
I still do not see problem, by now. The jar is compiled with gcj 4.0
20050125



Reply to: