Re: abort in verify_wide_reg_1(), in 2.96 20000502, simpletest case

[Mark Mitchell <mark at codesourcery dot com>]

>>>>> "Richard" == Richard Henderson <rth@cygnus.com> writes:

    Richard> On Thu, May 04, 2000 at 04:56:45PM -0700, Mark Mitchell
    Richard> wrote: Fixed thus.
    >>  And tested where? :-) Thanks!

    Richard> Erum... actually, nowhere.

    Richard> I do have builds running, but I didn't go through a
    Richard> complete build cycle specifically for this change.  This
    Richard> one just seemed obvious.

Fair enough.  The obvious fix rule is still a good one.

In that case saying "obvious fix; spinning tests now" seems like a
good idea.

The point of all this is to inspire our mutual confidence in
one-another, to force us to remember to test, and to help understand
what might be the problem when something does go wrong.  (Hmm, I know
this worked on Alpha, so I wonder what's different on SPARC?  Oh, I
know, it's that SPARC has register windows, and that means that ...)

Thanks,

--
Mark Mitchell                   mark@codesourcery.com
CodeSourcery, LLC               http://www.codesourcery.com