FIAT should not be involved. The timings are an average of running multiple times after JIT has been completed.
-- Anders mån 30 mars 2015 kl 13:53 skrev Andrew McRae <[email protected]>: > Is this with a cold cache or a warm cache? I noticed a FIAT slowdown > <https://bitbucket.org/fenics-project/fiat/issue/5/order-of-magnitude-slowdown-in-fiat> > last year, but this is run-once code. I really doubt this explains all the > difference, though. > > On 30 March 2015 at 12:34, Anders Logg <[email protected]> wrote: > >> See this question on the QA forum: >> >> >> http://fenicsproject.org/qa/6875/ubuntu-compile-from-source-which-provide-better-performance >> >> The Cahn-Hilliard demo takes 40 seconds with 1.3 Ubuntu packages and 52 >> seconds with 1.5+ built from source. Are these regressions in performance >> or is Johannes that much better at building Debian packages than I am >> building FEniCS (with HashDist). >> >> PS: Looking at the benchbot, there seem to have been some regressions >> in the timing facilities with the recent changes: >> >> http://fenicsproject.org/benchbot/ >> >> -- >> Anders >> >> >
_______________________________________________ fenics mailing list [email protected] http://fenicsproject.org/mailman/listinfo/fenics
