Analyzing Optimization: What's New in 3.4?

[Scott Robert Ladd]

Nathanael Nerode wrote:
> Be sure to mess around with -fschedule-insns and -fschedule-insns2.
> It seems that on many machines -fschedule-insns creates substantially 
> *worse* code (so use -fno-schedule-insns) and -fschedule-insns2 creates 
> slightly better code, but all four combinations should be tried.  ;-)

One of my running experiments has rejected -fschedule-insns from the 
outset, while showing a strong preference for -fschedule-insns2. This 
is, of course, on Intel ia32 hardware; other processors will undergo 
testing "soon."

-- 
Scott Robert Ladd
Coyote Gulch Productions (http://www.coyotegulch.com)
Software Invention for High-Performance Computing