libjava test suite keeps getting stuck

Mon Apr 23 12:44:00 GMT 2001

On Mon, Apr 23, 2001 at 03:51:44PM +0100, Andrew Haley wrote:
> I thought that gcc did not yet allow arbitrary insns to throw
> exceptions in the manner of the old "-fasynchronous-exceptions". If
> this is in fact supposed to work I'll investigate further.

This is supposed to work. If flag_non_call_exceptions is set
(as it is in java_init_options) then any may_trap_p instruction
that appears within an exception region at the beginning of
optimization will be tagged and optimized appropriately.
The bits I was talking to you about earlier should reduce the
number of false positives so that you get better code generation.
Moreover, on two platforms (x86 and alpha linux) you should be
able to throw from the signal handler with no special tricks.
r~