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 autolearn=ham autolearn_force=no version=3.4.4 X-Google-Thread: 103376,7767a311e01e1cd X-Google-Attributes: gid103376,public X-Google-Language: ENGLISH,ASCII-7-bit Path: g2news2.google.com!news4.google.com!border1.nntp.dca.giganews.com!nntp.giganews.com!local01.nntp.dca.giganews.com!nntp.comcast.com!news.comcast.com.POSTED!not-for-mail NNTP-Posting-Date: Mon, 23 Oct 2006 07:03:54 -0500 Date: Mon, 23 Oct 2006 07:55:49 -0400 From: Jeffrey Creem User-Agent: Mozilla Thunderbird 1.0.7 (Windows/20050923) X-Accept-Language: en-us, en MIME-Version: 1.0 Newsgroups: comp.lang.ada Subject: Re: GNAT compiler switches and optimization References: <1161341264.471057.252750@h48g2000cwc.googlegroups.com> <9Qb_g.111857$aJ.65708@attbi_s21> <434o04-7g7.ln1@newserver.thecreems.com> <4539ce34$1_2@news.bluewin.ch> <453A532F.2070709@obry.net> <9kfq04-sgm.ln1@newserver.thecreems.com> <5vgs04-64f.ln1@newserver.thecreems.com> <453bc74e$0$19614$426a74cc@news.free.fr> <4jit04-0gq.ln1@newserver.thecreems.com> In-Reply-To: Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Message-ID: NNTP-Posting-Host: 24.147.74.171 X-Trace: sv3-u45ZhNyNYf7lQn6hFmN6jB1Ho6IhbNq2hmjxlZudVLp7VQkSAQDJ+92uVnX7TeAfmyR+SisweglIzVq!humdyIyDQPWtIjqC2dwCQAoUyk5LSttvroQ0jOmq07YTDAA7lThObq6kqZZS68rNl41knFUsI2O8!rhk= X-Complaints-To: abuse@comcast.net X-DMCA-Complaints-To: dmca@comcast.net X-Abuse-and-DMCA-Info: Please be sure to forward a copy of ALL headers X-Abuse-and-DMCA-Info: Otherwise we will be unable to process your complaint properly X-Postfilter: 1.3.32 Xref: g2news2.google.com comp.lang.ada:7158 Date: 2006-10-23T07:55:49-04:00 List-Id: Jeffrey R. Carter wrote: > Jeffrey Creem wrote: > >> >> Having said that, I did update the wiki site with some timing numbers >> people asked for from a dual processor machine. >> >> http://gnuada.sourceforge.net/pmwiki.php/Main/Oct2006CLAGFORTRANComparison >> >> >> Scroll to the bottom and start reading at "Other Versions" if you are >> interested. This is getting too messy for plain text conversation. > > > Wow, a greater than 50% speedup from using 2 tasks instead of 1 (from > ada/tst_array to ada/tst_array_task_2). Something is odd there. > > This shows that, for those with dual-processor machines, easy-to-create > Ada is faster than easy-to-create FORTRAN. > > This doesn't help those with single processors, of course. > > I doubt if anything will beat matmul (short of 640_000 processors). But > Ada with explicit loops and FORTRAN with matmul are hardly equivalent. > You need to convention-FORTRAN the Ada array type, import matmul, and > call it to get a fair comparison. There shouldn't be much difference. > I looked at the gcc FORTRAN matmul. Unless there some additional trickery going on behind the scenes, it is not anything magical. It looks like a matmul implemented in C with manual array subscripting logic (i.e. uses a single dimensional array overlay).. In any case, it is not so much matmul I am trying to make faster here but rather just the nature of 2d array traversals in native language structures. I just included the FORTRAN matmul to be "more than fair" to FORTRAN as I am in no way trying to bash FORTRAN. The speedup for the tasks is quite odd though. I'll need to disassemble it tonight. I also just finished a 4.0.2 install last night so I'll get those numbers to see if all of this mess is simply a regression someplace in the compiler.