[Ecls-list] Feedback needed: more detailed optimization settings

[Juan Jose Garcia-Ripoll]

On Fri, May 7, 2010 at 11:49 PM, Juan Jose Garcia-Ripoll <
juanjose.garciaripoll at googlemail.com> wrote:

> The problem I am having right now is how to specify the interaction of
> these flags (which could also appear in proclamations, DECLAIM, etc) with
> the standard optimization settings. One possibility would be defining the
> whole set of optimizations as a bitmap, B. Each of the standard optimization
> settings would then do two things: first compute a combined mask of
> optimizations that are activated, A, and a combined mask of flags that are
> deactivated, D, and compute B <- (LOGANDC2 (LOGIOR B A) D)
>

Not exactly in this form, but this idea has been already committed. This new
representation of "policy" flags is accompanied by changes in the compiler
which do the following:

- Improved optimization of AREF/ASET via compiler macros. As other
optimizations soon to come, they are very efficient but without losing type
safety.

- Fixed the generation of C variables for temporary results. Now the type of
the variable is more accurately computed, avoiding a lot of inefficient
calls to Lisp<->C conversion functions. Should really produce more efficient
code.

- Removed the type filter. Now ECL keeps more accurate information about the
type of arguments that a function expects, and this already helped us
uncover a couple of bugs in the library.

Juanjo

-- 
Instituto de Física Fundamental, CSIC
c/ Serrano, 113b, Madrid 28006 (Spain)
http://tream.dreamhosters.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.common-lisp.net/pipermail/ecl-devel/attachments/20100509/d772ce26/attachment.html>