Your Jan 8th, 2001 rtlanal.c:note_stores change

[David S. Miller]

   From: kenner@vlsi1.ultra.nyu.edu (Richard Kenner)
   Date: Mon, 15 Apr 02 16:53:20 EDT
   
   I agree the CLOBBERs are wrong, but so is a SET.  It will cause
   problems to show each of those registers as being set to *the same
   thing*, whatever that thing is.
   
It sounds like you know of a current note_stores() callback that is
going to fail if they are given as SETs.  Where will these problems be
caused?  In particular, who inside a note_stores() callback interprets
sets of CALL returns in this weird way?  Is this some ADA specific
issue?

The only example I can possibly come up with are functions with
attribute const being optimized improperly, but as mentioned I failed
to come up with a test case demonstrating this failure with the
CLOBBER stuff reverted.