I don't recall that the FreeBSD version of egcs is built with Haifa
turned on, which is supposed to improve optimizations as the level
is increased (more aggressive instruction scheduling.)
> With egcs, the '-O' flag doesn't specify the optimization level like it
> does in GCC. It specifies
Garrett Wollman wrote:
> Just for completeness, I did one final run of HINT with just `-O'
> specified (our usual default).? `-O' results in significantly better
> integer performance than `-O4'.? (Floating-point performance is just
> the opposite.)
>
> This suggests that compiling the world with
Garrett Wollman wrote:
> Just for completeness, I did one final run of HINT with just `-O'
> specified (our usual default).? `-O' results in significantly better
> integer performance than `-O4'.? (Floating-point performance is just
> the opposite.)
>
> This suggests that compiling the world with
With egcs, the '-O' flag doesn't specify the optimization level like it
does in GCC. It specifies the desired stability of the generated code. Lower
numbers (0,1,2) request higher stability. ;)
DS
> Dan Nelson wrote:
> > -O4 doesn't exist in egcs (or it didn't a month or so ago).
Dan Nelson wrote:
> -O4 doesn't exist in egcs (or it didn't a month or so ago). According
> to the source, -O2 enables all optimizations except -funroll-all-loops,
> and all -O3 does is enable -funroll-all-loops.
I think I recall reading somewhere that EGCS uses -O numbers > 3 to test
experimenta
< said:
> I'd like to see separate runs, one with each -march= option (i386,
> i486, i586, i686), so see if those many any difference at all.
As you will have seen from my post yesterday, there is no measurable
difference between -march=pentiumpro and no -march= option. You're
certainly welcome
In the last episode (May 25), Garrett Wollman said:
> Just for completeness, I did one final run of HINT with just `-O'
> specified (our usual default). `-O' results in significantly better
> integer performance than `-O4'. (Floating-point performance is just
> the opposite.)
>
> This suggests t
Just for completeness, I did one final run of HINT with just `-O'
specified (our usual default). `-O' results in significantly better
integer performance than `-O4'. (Floating-point performance is just
the opposite.)
This suggests that compiling the world with `-O' levels higher than
one is prob
