[PATCH] cselib_invalidate_regno performance

[Bernd Schmidt]

On Mon, 20 May 2002, David S. Miller wrote:

> Ok, here is another try at this, based upon Bernd's analysis.
> (although oddly he commented on this patch rather than the one
>  I actually sought input from him on, the approx_reg_cost stuff
>  since he is the one who added that :-)

Well, the other patch looked sensible so I didn't feel a great need
to comment on it :)


Bernd