Re: [patch/hppa-linux] MD_FALLBACK_FRAME_STATE_FOR definition

[Andrew Haley <aph at redhat dot com>]

Randolph Chung writes:
 > >  > ok, let me turn that on and retest.
 > > 
 > > testsuite/libjava.lang/Throw_2.java is the key test case.
 > 
 > well, it passed that test before with my patch. it still passes now, 
 > after i turned on can_unwind_signal.

Without can_unwind_signal, we insert null pointer checks in the code
that gcj generates, so except in a few corner cases like Throw_2,
MD_FALLBACK_FRAME_STATE_FOR won't be used.

If after rebuilding with can_unwind_signal Throw_2 still works, then
you're fine.

 > # of expected passes            3324
 > # of unexpected failures        2
 > # of expected failures          14
 > # of untested testcases         16
 > 
 > The two failed cases are:
 > FAIL: negzero output - gij test
 > FAIL: negzero output - gij test
 > 
 > these were there before my patch too...
 > 
 > so, are there other things you want me to test? or just resend the patch
 > with Dave's comments + can_unwind_signal=yes ?

Do that.

Andrew.