What I would like to be able to do, instead of -O1, -O2 -O3 flags to a compiler is for the "optimization level" to be measured in seconds. So -O1 would run the optimizer for 1 second and give me the best optimized code it could come up with in that time. -O3600 would let the machine think about it for an hour, using any and all heuristics and empirical tests and then giving me what it had at the end. Pre-release, I might want to run the optimizer for a week.
> Pre-release, I might want to run the optimizer for a week.
That's a dangerous model, as certain classes of bugs would only come out in the super-optimized version, but that version would presumably not get the same amount of testing as the regular builds.
Re:Will Parallel Code Ever Be Embraced?
by bunderbunder on Thursday July 19, 2012 @10:30AM
(#4266135)
It's amazing how much of that stuff isn't really the CPU's fault, though. For example, buying a solid-state drive and moving all my projects over to it has done things that are almost magical to my build times.
(Granted, my C++ days are behind me so particularly CPU-intensive builds just don't really happen to me anymore.)
That's just ... Weird. What would "maxing out" even mean for a CPU? Going "fast enough", somehow? Maybe because build cycles is exactly what I do a lot of my time in front of a computer, I really don't think CPU:s are ever going to be "fast enough". Even if just doing "ordinary computing", I often think e.g. browsers and office applications are rather slow.