On Wed, Oct 10, 2018 at 01:44:41PM +0200, SZEDER Gábor wrote: > > So that's really weird and counter-intuitive, since we should be doing > > strictly less work. I know that spatch tries to parallelize itself, > > though from my tests, 1.0.4 does not. I wonder if the version in Travis > > differs in that respect and starts too many threads, and the extra time > > is going to contention and context switches. > > I don't think it does any parallel work. > > Here is the timing again from my previous email: > > 960.50user 22.59system 16:23.74elapsed 99%CPU (0avgtext+0avgdata 1606156maxresident)k > > Notice that 16:23 is 983s, and that it matches the sum of the user and > system times. I usually saw this kind of timing with CPU-intensive > single-threaded programs, and if there were any parallelization, then I > would expect the elapsed time to be at least somewhat smaller than the > other two. Ah, right, I should have been able to figure that out myself. So scratch that theory. My "hypervisor stalling our memory reads" theory is still plausible, but I don't know how we would test it. I guess in some sense it doesn't matter. If it's slower, we're not likely to be able to fix that. So I guess we just need the fallback to the current behavior. -Peff
