Lots of "make check" regressions

[Bryce McKinlay <bryce at albatross dot co dot nz>]

Okay folks, gcj now builds once again (at least, the mainline does,
we'll have to move the libjava changes over to the branch.)

Unfortunatly, I am seeing a large number of testsuite failures:


                === libjava Summary ===

# of expected passes            1522
# of unexpected failures        46
# of unexpected successes       34
# of expected failures          122

I'm pretty sure we were looking much better than this a few weeks ago.

One of the more disturbing regressions (see
InterfaceDispatch.java) is:

class CC {}

public class Test
{
  public static void main(String args[])
  {
    System.out.println (new CC[1] instanceof Object[]);
  }
}

$ ./test
false

Upon inspection, gcj is never even generating a call to
_Jv_IsInstanceOf() (arguably it shouldn't anyway because the type of
the array is provable, but I was not aware that gcj had gained this
optimzation)

I am absolutely sure that this worked last time I ran the testsuite.
Does this ring bells with anyone?

regards

  [ bryce ]