GCJ/minGW produced executables and linux/wine

Tue Mar 4 19:20:00 GMT 2003

On Tue, 4 Mar 2003, Andrew Haley wrote:
> > I think this test isn't quite right:
> >
> > if test $can_unwind_signal = no && test $enable_sjlj_exceptions = no; then
> > CHECKREFSPEC=-fcheck-references
> > ...
> >
> > since it gives no assurances that signal handling a.k.a. exception filters
> > work at all.
>> The test is correct, but it may need to be overriden on some targets.
> If we're using sjlj, we shouldn't need to use -fcheck-references.

Shouldn't, no. My problem is that I don't have a viable debugger. I did
try one test on a bonafide Windows machine with the same results though.
However it seems to me that targets supporting a SIGSEGV handler or
equivalent are the exception, not the rule, in which case
-fcheck-references would be more appropriate as the configuration default
except on targets for which we know better.
> Please let me know which tests still fail.

FAIL: SyncTest execution - bytecode->native test
FAIL: SyncTest -O output - bytecode->native test
FAIL: TestProxy execution - source compiled test
FAIL: TestProxy execution - bytecode->native test
FAIL: TestProxy -O execution - source compiled test
FAIL: TestProxy -O execution - bytecode->native test
FAIL: Thread_Interrupt execution - source compiled test
FAIL: Thread_Interrupt execution - bytecode->native test
FAIL: Thread_Interrupt -O execution - source compiled test
FAIL: Thread_Interrupt -O execution - bytecode->native test
FAIL: Thread_Wait_2 execution - source compiled test
FAIL: Thread_Wait_2 execution - bytecode->native test
FAIL: Thread_Wait_2 -O execution - source compiled test
FAIL: Thread_Wait_2 -O execution - bytecode->native test
FAIL: Thread_Wait_Interrupt output - source compiled test
FAIL: Thread_Wait_Interrupt output - bytecode->native test
FAIL: Thread_Wait_Interrupt -O output - source compiled test
FAIL: Thread_Wait_Interrupt -O output - bytecode->native test
FAIL:
/export/home/jsturm/gcc-3.3/build-mingw/i686-pc-mingw32/libjava/testsuite/TestEarlyGC.exe
execution -
/export/home/jsturm/gcc-3.3/build-mingw/i686-pc-mingw32/libjava/testsuite/TestEarlyGC.exe
FAIL:
/export/home/jsturm/gcc-3.3/build-mingw/i686-pc-mingw32/libjava/testsuite/TestLeak.exe
execution -
/export/home/jsturm/gcc-3.3/build-mingw/i686-pc-mingw32/libjava/testsuite/TestLeak.exe
Thread.interrupt() tests are expected to fail. SyncTest looks
suspicious, since it didn't timeout. A broken GC might be responsible for
some of these.
Jeff