Re: RFC: merging MEM_ATTRS

[Richard Henderson <rth at redhat dot com>]

On Tue, Feb 10, 2004 at 06:22:54PM +0000, Joern Rennecke wrote:
> If we zero the MEM_EXPR if one or more of the alternatives accesses
> a variable that never has its address taken and that we reasoned thus
> can only be accessed with explicit accesses, we would incorrectly assume
> the MEM can't affect/use the variable any more.

This is false.  If MEM_EXPR is null, then we assume we don't know
what variable it references, rather than assuming it references 
no variables.



r~