Re: [PATCH] External vectorization library (target) support

[Gabriel Dos Reis <gdr at cs dot tamu dot edu>]

On Sun, 10 Dec 2006, Richard Guenther wrote:

| On Sun, 10 Dec 2006, Gabriel Dos Reis wrote:
|
| > Richard Guenther <rguenther@suse.de> writes:
| >
| > | This is the alternative way to tackle vectorization of intrinsics
| > | (the other way is libgcc-math which lacks reviewers, approvers,
| > | political willingness?).
| >
| > As I have expressed interest in libgcc-math (for libstdc++), I'm a bit
| > unclear about what you mean by the above.  Please could you expand on
| > it? (You have done that before, but I've lost track of parts of GCC
| > lists traffic).
|
| Well, I really don't know either (this is why there is a question mark
| after the sentence).  I expected some feedback on the notes from the BOF
| (http://gcc.gnu.org/ml/libstdc++/2006-06/msg00108.html) or on the
| re-submission of libgcc-math.

Thanks for the pointer.  I believe the conclusion in your report is
positive for libstdc++.

|  There was a discussion on the SC-list
| about libgcc-math early this year which ended in a way that libgcc-math
| was removed from 4.2.  There was more in this discussion and RMS agreed
| to something but there seems to be uncertainty about what exactly was
| agreed on.

:-(

| So, basically the SC (or RMS) needs to decide this issue (again).
|
| Or we could do a techincal decision - but as there isn't even a
| technical discussion about the matter I don't see things going
| forward (neither from my peekings on the SC through my private
| channels).

I think we can proceed with the technical aspects. And leave the
political aspect, if there really is one to others.


How do we determine when we link libgc-math into programs?
Could GCC assume that any standard C/C++ mathematical function is
"defined" by GCc (e.g. libgcc-math) and therefore apply more knowledge
in optimizations?

-- Gaby