On Jun 3, 2013, at 1:16 PM, Segher Boessenkool
wrote:
> define_subst is not flexible enough for our purpose by far
I had Kenny review define_subst when it went in, though it is a step in the
right direction, it misses out on a lot completeness that a port maintainer
might wish it had. For me
Well, my first comment is that this is more or less exactly what
we invented define_subst to attempt to handle. Could we have more
constructive feedback on that mechanism rather than inventing a
new mechanism private to the rs6000 backend?
Hi Richard,
I don't think this can be done using defin
On 06/03/2013 09:26 AM, Segher Boessenkool wrote:
> David alerted me to the proper way to ensure generated files
> will not be regenerated for end-users. I also cleaned up some
> whitespace/comment churn; and there was an embarrassing bug in
> the dot condition for sri, for -m32 -mpowerpc64. So,
David alerted me to the proper way to ensure generated files
will not be regenerated for end-users. I also cleaned up some
whitespace/comment churn; and there was an embarrassing bug in
the dot condition for sri, for -m32 -mpowerpc64. So, v2.
Bootstrapped and tested these six together, no regres
