From mboxrd@z Thu Jan 1 00:00:00 1970 X-Spam-Checker-Version: SpamAssassin 3.4.4 (2020-01-24) on polar.synack.me X-Spam-Level: X-Spam-Status: No, score=-1.9 required=5.0 tests=BAYES_00,FREEMAIL_FROM autolearn=unavailable autolearn_force=no version=3.4.4 X-Received: by 10.129.69.21 with SMTP id s21mr8827378ywa.66.1497214090406; Sun, 11 Jun 2017 13:48:10 -0700 (PDT) X-Received: by 10.157.26.93 with SMTP id u29mr899759otu.2.1497214090367; Sun, 11 Jun 2017 13:48:10 -0700 (PDT) Path: eternal-september.org!reader01.eternal-september.org!reader02.eternal-september.org!news.eternal-september.org!news.eternal-september.org!feeder.eternal-september.org!news.glorb.com!s24no1050050qte.0!news-out.google.com!m134ni201itb.0!nntp.google.com!67no1688517itx.0!postnews.google.com!glegroupsg2000goo.googlegroups.com!not-for-mail Newsgroups: comp.lang.ada Date: Sun, 11 Jun 2017 13:48:10 -0700 (PDT) In-Reply-To: <65c5271c-33a1-4f36-977e-372af449d9c1@googlegroups.com> Complaints-To: groups-abuse@google.com Injection-Info: glegroupsg2000goo.googlegroups.com; posting-host=2601:191:8303:2100:5985:2c17:9409:aa9c; posting-account=fdRd8woAAADTIlxCu9FgvDrUK4wPzvy3 NNTP-Posting-Host: 2601:191:8303:2100:5985:2c17:9409:aa9c References: <65c5271c-33a1-4f36-977e-372af449d9c1@googlegroups.com> User-Agent: G2/1.0 MIME-Version: 1.0 Message-ID: <84e92c0f-8877-496d-b295-cf590e047ed4@googlegroups.com> Subject: Re: Poor performance with GNAT-GPL-2017 From: Robert Eachus Injection-Date: Sun, 11 Jun 2017 20:48:10 +0000 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable Xref: news.eternal-september.org comp.lang.ada:46934 Date: 2017-06-11T13:48:10-07:00 List-Id: On Saturday, June 10, 2017 at 7:23:34 AM UTC-4, Charly wrote: =20 > Any ideas?? =20 Three ideas really. First there may be some change in the tasking and rend= ezvous parts of GNAT-GPL that results in twice as many scheduling actions (= or more) and thus the slowdown. Most probable. Second there is no guarantee that the tasking actions will be identical bet= ween the two instances. Worst case, you have one thread which is the high = pole in the tent starting later. The "easy" way to check this is to take o= ut the tasking and make the program single-threaded with identical program = behavior. You could run some tasking tests which I hope have been around forever (ACE= S), to see what changed in tasking timings. Finally, and highly unlikely given the manitude of the timing change, the c= hange in versions will change the order (to some extent) in which alternati= ves are tested. These changes can result in the program run time being dif= ferent because more instances are checked before determining that there is = no valid way to continue from that point. You could check this by putting = in an instance counter that is incremented each time a tile is placed. (Ke= eping separate counters for each task is an obvious optimization.) It would be very interesting if the number of instances checked varied, eit= her from run-to-run or compiler version to compiler version.