Re: Your Jan 15 change broke the x86

[Richard Henderson <rth at cygnus dot com>]

On Sun, Jan 31, 1999 at 03:00:37PM -0700, Jeffrey A Law wrote:
> Should we reject it at expansion time or in reload?
> 
> I'd say expansion time, but I worry that we'd start balking at asms that have
> worked for years.  Then again, it may be the case that nobody uses the <>
> constraints in an asm without either an "o" or "m".

I dunno. 

The thing is, at the moment they'll fall prey to the strictness of
asm_operand_ok.  So if we accept them at expansion time, we have to
relax asm_operand_ok.  Which is ok, I guess.  It's intent is to get
us as close as possible to something reloadable.  If it isn't 
possible...


r~