220 17233 <3be1928f-2116-4712-be20-7368682abf26@isocpp.org> article Path: news.gmane.org!not-for-mail From: tef@tentity.com Newsgroups: gmane.comp.lang.c++.isocpp.proposals Subject: Generic Parallel Computing (GPC) API Date: Fri, 3 Apr 2015 05:15:13 -0700 (PDT) Lines: 229 Approved: news@gmane.org Message-ID: <3be1928f-2116-4712-be20-7368682abf26@isocpp.org> Reply-To: std-proposals@isocpp.org NNTP-Posting-Host: plane.gmane.org Mime-Version: 1.0 Content-Type: multipart/mixed; boundary="----=_Part_297_924641526.1428063313015" X-Trace: ger.gmane.org 1428063322 7844 80.91.229.3 (3 Apr 2015 12:15:22 GMT) X-Complaints-To: usenet@ger.gmane.org NNTP-Posting-Date: Fri, 3 Apr 2015 12:15:22 +0000 (UTC) To: std-proposals@isocpp.org Original-X-From: std-proposals+bncBDL4BSXDYUBBBUMI7KUAKGQEG7JD7CQ@isocpp.org Fri Apr 03 14:15:16 2015 Return-path: Envelope-to: gclcip-std-proposals@m.gmane.org Original-Received: from mail-pa0-f71.google.com ([209.85.220.71]) by plane.gmane.org with esmtp (Exim 4.69) (envelope-from ) id 1Ye0VH-0000GE-RF for gclcip-std-proposals@m.gmane.org; Fri, 03 Apr 2015 14:15:16 +0200 Original-Received: by pactp5 with SMTP id tp5sf179632419pac.0 for ; Fri, 03 Apr 2015 05:15:14 -0700 (PDT) X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:date:from:to:message-id:subject:mime-version :content-type:x-original-sender:reply-to:precedence:mailing-list :list-id:list-post:list-help:list-archive:list-subscribe :list-unsubscribe; bh=mD4wkg9cY/tmwdnnYenTOiQDb2cVliRH6JGZ9O8vk4I=; b=XFm7sFP6swmBWiHpmZQdGT6ijOKsn1iJqzn0wK2EzVpXlKuniL+Cf7YaHic0svwERH 3cEVb6IO8sXy5dBZC2cVNRS/YXVYTBA0Pgecj/P0Epw57+U7bHO5jnAEgXM3/Ynh1qRF K6V7sFOuxPgELnn6FmBeF4IHiiSCzjEPJUTK4ybU5kUpcIpzvovOxVTRtauXPsnQfpBM 1qagAp2fvsjEgjGFr+/VAa6H3TAbDFBI00spjzqa/aly97tI0+GtpepRGLw/Nwn74CBl YD6Ysu3BoZfdJyl+Jz0J36TQ4RYmzfeIsnJ9u/hZByFk0FCd70uSamfsBNltJPsPLlUV g6Lg== X-Gm-Message-State: ALoCoQmGK/soRvVK9rMAEgVIeb1nZDq5iqRVEStsHeItPiaOUQX1nHjUCTj/aBs+gNqmUx2cS3nU X-Received: by 10.66.236.74 with SMTP id us10mr2625050pac.28.1428063314669; Fri, 03 Apr 2015 05:15:14 -0700 (PDT) X-BeenThere: std-proposals@isocpp.org Original-Received: by 10.140.90.239 with SMTP id x102ls1232931qgd.86.gmail; Fri, 03 Apr 2015 05:15:13 -0700 (PDT) X-Received: by 10.140.39.148 with SMTP id v20mr18539qgv.24.1428063313753; Fri, 03 Apr 2015 05:15:13 -0700 (PDT) X-Original-Sender: tef@tentity.com Precedence: list Mailing-list: list std-proposals@isocpp.org; contact std-proposals+owners@isocpp.org List-ID: X-Google-Group-Id: 399137483710 List-Post: , List-Help: , List-Archive: List-Subscribe: , List-Unsubscribe: , Xref: news.gmane.org gmane.comp.lang.c++.isocpp.proposals:17233 Archived-At: ------=_Part_297_924641526.1428063313015 Content-Type: multipart/alternative; boundary="----=_Part_298_1258457790.1428063313015" ------=_Part_298_1258457790.1428063313015 Content-Type: text/plain; charset=UTF-8 Now, C++ includes built-in support for threads, mutual exclusion, condition variables, and futures. But some essential paradigm for parallel computing is missing. I am talking about well known *fork/join* but enhanced for task parallelism. So, in GPC API: - the *fork* API does not clone a process or thread but submit a function/lambda to a thread pool. - the *join* API submits a function/lambda to a quetex. Quetex is Mutual Exclusion Queue. It eliminates blocking/waiting and thread switching in comparison to mutexes and futures. gpc::quetex qtx; while( /* condition */ ) { gpc::fork([qtx]() { // do something, then qtx.join([](gpc::mutex& state) { // exclusive access to state.setData() or state.getData() }); }); } Here is an example of the usage: Reduction I assume that if C++ includes threads it can include thread pools also. And if there is mutexes it could be also quetexes there. I have an implementation of GPC API that I use for my development. You can try it at http://parallect.codeplex.com If you have a use case for task parallelization, please share. It would be interesting to see how the *fork/join* would work or not. Thanks Andrew -- --- You received this message because you are subscribed to the Google Groups "ISO C++ Standard - Future Proposals" group. To unsubscribe from this group and stop receiving emails from it, send an email to std-proposals+unsubscribe@isocpp.org. To post to this group, send email to std-proposals@isocpp.org. Visit this group at http://groups.google.com/a/isocpp.org/group/std-proposals/. ------=_Part_298_1258457790.1428063313015 Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: quoted-printable

Now, C++ includes built-in support for t= hreads, mutual exclusion, condition variables, and futures. But some essent= ial paradigm for parallel computing is missing.

I= am talking about well known fork/join but enhanced for ta= sk parallelism. 

So, in GPC API:

  • the fork= API does not clone = a process or thread but submit a function/lambda to a thread pool. 
  • the join API submits a function/lambda to a= quetex.

Quetex is Mutual Exclusion Qu= eue. It eliminates blocking/waiting and thread switching in comparis= on to mutexes and futures.

<= span style=3D"color: #000;" class=3D"styled-by-prettify">
 gpc::quetex<int> qtx;

 while( /* condition */ ) {

  &= nbsp; gpc:= :fork([qtx]() {

         // do something, th= en

=          qtx.join([](gpc:= :mutex<int>&<= span style=3D"color: #000;" class=3D"styled-by-prettify"> state) {

           =    // exclusive access to <int>

          &nbs= p;   state.setD= ata() or state.getData()

        });
    });

 }


Here is an e= xample of the usage: Reduction

I assume= that if C++ includes threads it can include thread pools also. And if ther= e is mutexes it could be also quetexes there.

I have an = implementation of GPC API that I use for my development. You can try it at&= nbsp;http://parallect.codeplex.co= m

If you have a use case for task parallelization, please sha= re. It would be interesting to see how the fork/join would= work or not.

Thanks

Andrew



--

---
You received this message because you are subscribed to the Google Groups &= quot;ISO C++ Standard - Future Proposals" group.
To unsubscribe from this group and stop receiving emails from it, send an e= mail to std-proposa= ls+unsubscribe@isocpp.org.
To post to this group, send email to std-proposals@isocpp.org.
Visit this group at http://groups.google.com/a/isocpp.org/group/std-proposals/<= /a>.
------=_Part_298_1258457790.1428063313015-- ------=_Part_297_924641526.1428063313015-- .