Re: Add support to cprop to propagate from parallels

[Jan Hubicka <hubicka at atrey dot karlin dot mff dot cuni dot cz>]

> On Mon, Nov 01, 1999 at 04:55:34AM -0700, Jeffrey A Law wrote:
> > This should be fixed in cse or by some post-cse pass to strip off useless
> > clobbers.  Useless clobbers are going to do nothing but hinder potential
> > optimizations.
> 
> Rather, the clobbers shouldn't be added until later in the compilation. 
> They are never useless, because their existance affects the dependancies
> created during life_analysis (for combine) and for scheduling.
The kind of clobbers in moves we was talking about was ones "forgotten"
in folded expression. When expression (with correct clobber) is folded to constant
or register, move pattern with extra clobber is created....
These really ought to be stripped.
> 
> This goes back to the idea of a `mid-level rtl' with no libcalls, not
> decomposing multiplies etc, and letting the first few passes operate on that.
This sounds like interesting idea...

Honza
> 
> 
> 
> r~

-- 
                       OK. Lets make a signature file.
+-------------------------------------------------------------------------+
|        Jan Hubicka (Jan Hubi\v{c}ka in TeX) hubicka@freesoft.cz         |
|         Czech free software foundation: http://www.freesoft.cz          |
|AA project - the new way for computer graphics - http://www.ta.jcu.cz/aa |
|  homepage: http://www.paru.cas.cz/~hubicka/, games koules, Xonix, fast  |
|  fractal zoomer XaoS, index of Czech GNU/Linux/UN*X documentation etc.  | 
+-------------------------------------------------------------------------+