new optimizations for lunar - part1

Zbigniew Łuszpiński zbiggy at o2.pl
Wed May 4 11:54:20 UTC 2005


> On Wed, May 04, 2005 at 09:32:14AM +0200, Auke Kok wrote:
> ..
>
> > - the '-freorder-blocks' looks interesting, however I don't think there
> > are many people who use -Os or -O1 optimization right now, so the
> > usability is low again. However for the ISO's it might be of use, or
> > anyone who's experimenting with smaller but still optimized code.
>
> I use -Os ;-) Never noticed much of a difference though. In case anyone was
> interested in that information...

I also use -Os, people says we are in minority:

"- the '-freorder-blocks' looks interesting, however I don't think there 
are many people who use -Os or -O1 optimization right now, so the 
usability is low again."

Time to organize and fight to protect our little laws :-) :-P :-D

But to be serious:
Sofar has right. Messing menu with useless options is bad. But the additional 
function -freorder-blocks does not take much space in configuration menu 
window. So why throwing it away? This function is optional - anyone can turn 
it on/off. There is no so much options worth of adding to menu, I only know 2 
more , they are not gcc options but passed to code being compilated.

> I still think it would be much better to have only optimization level, arch
> and linker options and have a freeform entry for additional options some
> (crazy) people would like.

I think that optimization wizard should be still available but at the end it 
should display freeform with chosen options allowing to add/remove/modify 
before saving finally. Newbies will simply save as it is, gcc optimization 
maniacs will put all their crazyness into the freeform and then save.

> 	Jasper

thanks, Jasper, for additional ideas :-) They look interesting and worth 
implementing.

zbiggy
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
Url : http://foo-projects.org/pipermail/lunar/attachments/20050504/8f87dc48/attachment.bin


More information about the Lunar mailing list