state of 3.2.1-pre: how far from release?

[David Edelsohn]

>>>>> Mark Mitchell writes:

Mark> My preference would be to see the load multiple/store multiple
Mark> instructions that are causing the problem disabled.  Better a slow
Mark> compiler than a broken one, since David says the problem can come up
Mark> relatively often.

Mark> If the RS6000 maintainers do not want to do that, then David's
Mark> work-around for abi_check.cc is OK.

	The failure is not complicated, but does not appear to occur
often, so disabling the pattern does not seem necessary.  We can consider
this more for GCC 3.3.

David