From dev-return-30422-apmail-harmony-dev-archive=harmony.apache.org@harmony.apache.org Wed Nov 07 09:03:44 2007 Return-Path: Delivered-To: apmail-harmony-dev-archive@www.apache.org Received: (qmail 22505 invoked from network); 7 Nov 2007 09:03:44 -0000 Received: from hermes.apache.org (HELO mail.apache.org) (140.211.11.2) by minotaur.apache.org with SMTP; 7 Nov 2007 09:03:44 -0000 Received: (qmail 43733 invoked by uid 500); 7 Nov 2007 09:03:31 -0000 Delivered-To: apmail-harmony-dev-archive@harmony.apache.org Received: (qmail 43705 invoked by uid 500); 7 Nov 2007 09:03:31 -0000 Mailing-List: contact dev-help@harmony.apache.org; run by ezmlm Precedence: bulk List-Help: List-Unsubscribe: List-Post: List-Id: Reply-To: dev@harmony.apache.org Delivered-To: mailing list dev@harmony.apache.org Received: (qmail 43695 invoked by uid 99); 7 Nov 2007 09:03:31 -0000 Received: from athena.apache.org (HELO athena.apache.org) (140.211.11.136) by apache.org (qpsmtpd/0.29) with ESMTP; Wed, 07 Nov 2007 01:03:31 -0800 X-ASF-Spam-Status: No, hits=-0.0 required=10.0 tests=SPF_PASS X-Spam-Check-By: apache.org Received-SPF: pass (athena.apache.org: domain of maksim.ananjev@gmail.com designates 209.85.146.181 as permitted sender) Received: from [209.85.146.181] (HELO wa-out-1112.google.com) (209.85.146.181) by apache.org (qpsmtpd/0.29) with ESMTP; Wed, 07 Nov 2007 09:03:34 +0000 Received: by wa-out-1112.google.com with SMTP id k22so3903837waf for ; Wed, 07 Nov 2007 01:03:12 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=beta; h=domainkey-signature:received:received:message-id:date:from:to:subject:in-reply-to:mime-version:content-type:content-transfer-encoding:content-disposition:references; bh=b4XclZs/mHA59qKfo8OMaITAhLVprWPrdTkf5qMmyBY=; b=n1ZqzCqyCFYtBirDTUrAMrOvagM2FeJ/wY8BUuZt1Vn5b8JrT815UCpWNlqeWyl+vLqEgzTfrJIs623y5gCZyW/C3eic9P2PL/UJmlvSs7xTXOdxYAk/3VqTUflQ4EFQEVN20fecfE0DoRhrqeU9yR6ZWiz4tOuNKnK4cPzzXg4= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=beta; h=received:message-id:date:from:to:subject:in-reply-to:mime-version:content-type:content-transfer-encoding:content-disposition:references; b=nCjfK3FKPcIpZfZHMM/XWbnwj7mOKSeyP3EmGckUwG7Jdi1JyDcqYVHhf5DaoQFqw93pdvZogoVYxBUyAshMr8QOTmJPgfB657iHRKpBkxX90u307Pj1A3FisLZzytMWYOuvGNRhXEs8nSpio1L+9IReefSKYfXJv59Pu8MLQt4= Received: by 10.114.73.1 with SMTP id v1mr954525waa.1194426192682; Wed, 07 Nov 2007 01:03:12 -0800 (PST) Received: by 10.114.74.20 with HTTP; Wed, 7 Nov 2007 01:03:12 -0800 (PST) Message-ID: Date: Wed, 7 Nov 2007 15:03:12 +0600 From: "Maksim Ananjev" To: dev@harmony.apache.org Subject: Re: [perf] Comparative benchmarking In-Reply-To: MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline References: <472C548F.6030609@gmail.com> <47304FB3.7020306@gmail.com> <4bebff790711060438o422abd60ke22ac2aea811b50e@mail.gmail.com> <473068D4.8030901@gmail.com> <4bebff790711060603h62efddf0p85ccebccffae6e92@mail.gmail.com> <0vqejf3gzke.fsf@gmail.com> <0vqabprgj42.fsf@gmail.com> <4bebff790711061402v78140d33h21f4123a6eb7147f@mail.gmail.com> X-Virus-Checked: Checked by ClamAV on apache.org Btw, why not to implement api magic that would substitute sqrt(x) by hardware sse sqrt instruction? "sqrtpd", afair It might be faster then JNI call. Just an idea. I am not sure if such JIT hack looks pretty from the point of view of design. 2007/11/7, Egor Pasko : > On the 0x387 day of Apache Harmony Aleksey Shipilev wrote: > > On 07 Nov 2007 00:27:25 +0300, Egor Pasko wrote: > > > Vladimir, guess what? :) I actually mixed several things > > > altogether. > > (sigh) As usual, miracle does not happen :) I dreamed to see > > software sqrt() implementation that could be faster than hardware one. > > > > > > > So, we are left with SSE asm that can be inlined by JIT and AFAI can > > > see it is not as fast as HotSpot? Weird :) > > No, for now we have just the intrinsic in native code, so we also have > > overheads for JNI transition (and it is heavy!), parameter passing, > > chains of calls, etc. I believe that sqrt() magic will lead that NBody > > performance very close to RI. > > Ha! I thought, you made it without JNI. Then we have a chance! If we > do not get it through magics we still have another chance to implement > 1/sqrt() :) > > > BTW, in my thought this benchmark is like the top of the iceberg > > called "FP performance problems". > > Yes, I know, DRLVM has not been actively optimized for FP calc. > > -- > Egor Pasko > > > -- Maksim