11706 vs ([lno] Canonical iv creation)

[Gabriel Dos Reis]

Paolo Carlini <pcarlini@suse.de> writes:

| Gabriel Dos Reis wrote:
| 
| >It would be misleading to let people believe that is only about
| >"cleaness".  The raison d'être of pow(T, int) is precesily that common
| >implementations of pow(T, floating-point) try to recover an
| >information they don't have access to but which where available at the
| >call site, namely whether the exponent is integer or not and apply the
| >appropriate algorithm.  The pow() we are discussing was provided as a
| > way to communicate that information to the library.  Forwarding the
| > call back again to pow(T, floating-point) defeats the
| >whole purpose of pow(T, int).  It is pessimization -- note calls to
| >pow() are not just with constant integers.  I don't see that
| >as improvement.
| 
| Well, Gaby, I don't want to discuss this issue more. After all, you
| gonna take the responsability for the performance of pow(T, int) in
| 3.4, not me.

Paolo --

   You have been unnecessarily annoying for a long time now.

Each time I want to take the time to have a sound discussion with
you, you excell in sidetracking the main issue and concentrating on
me. 

If you don't want to discuss the issue, why do you persist then?

   Assume you don't have the monopole of "pragmatism" and my
disagreeing with you does not mean being less pragmatic than you.

   As I said very early, this issue should not be attacked in rush.
There are lies, dead lies and benchmarks.  Pushing this issue as
you're doing by just comparing against pow(, -1), pow(, 0), pow(, 1)
or pow(, 2) is a myopic view, which despite your efforts, misses
the whole point of this problem.  Have you been carefully listening to
people, especially to what Richard G. has been saying/suggesting (look
carefully at his "fixes") and what I'm saying, you'll notice that the
difference in our respective points of view is not just about
forwarding to __builtin_pow().

   Did you do your homework by looking at the archive, you'll notice
that I've been following Roger's work on built-ins (even if I can't
recite by heart all the patches he sent).  So I assume that your
asking is not an expression of an irresistible desire of being silly,
but simply the result of a profond ignorance of what went on the
built-ins.  I can't blame you for that.

   This ends this conversation with you.  I'm willing to have further
technical discussions with you on pow() issue, provided:
   (1) you do further investigations, looking at the core issues;
   (2) you stop fixing on me.  I don't want to be your fix-point.

-- Gaby