2017-03-09 12:34 GMT+01:00 'bertschi' via Clojure <clojure@googlegroups.com>: > Thanks for your comments. As suggested I ran a small benchmark of both > versions. Turns out that the difference between (ref-set ref @ref) and > ensure is huge ... > I'm running clojure 1.8.0 by the way. According to VisualVM the (ref-set ref > @ref) version spends most of its time in > clojure.lang.LockingTransaction$RetryEx.<init> () while ensure gets basically > stuck (> 99%) at clojure.lang.LockingTransaction.tryWriteLock ().
On my machine, results don't look as bad, but ensure still has a significantly worse bad case (the upper quantile), the best case (lower quantile) is about on par with ref-set. Disabling tiered compilation helps close the gap (possibly due to lock straightening?) but ref-set still comes out ahead. The tests run significantly faster on server compilation than on tiered compilation and also run faster on clojure-1.9.0-alpha14, than on -1.8.0 I'm on % java -version openjdk version "1.8.0_121" OpenJDK Runtime Environment (build 1.8.0_121-13) OpenJDK 64-Bit Server VM (build 25.121-b13, mixed mode) Here is a gist with my test program + output listings: https://gist.github.com/bendlas/6e3b24bb639c248fdc47d192f5117c39 cc clojure-dev: is this ready for a ticket? do we need any more information to show what's going on? -- You received this message because you are subscribed to the Google Groups "Clojure" group. To post to this group, send email to clojure@googlegroups.com Note that posts from new members are moderated - please be patient with your first post. To unsubscribe from this group, send email to clojure+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/clojure?hl=en --- You received this message because you are subscribed to the Google Groups "Clojure" group. To unsubscribe from this group and stop receiving emails from it, send an email to clojure+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
