From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Original-To: caml-list@sympa.inria.fr Delivered-To: caml-list@sympa.inria.fr Received: from mail3-relais-sop.national.inria.fr (mail3-relais-sop.national.inria.fr [192.134.164.104]) by sympa.inria.fr (Postfix) with ESMTPS id 7CDF680211 for ; Mon, 25 Sep 2017 10:33:04 +0200 (CEST) Authentication-Results: mail3-smtp-sop.national.inria.fr; spf=None smtp.pra=mshinwell@janestreet.com; spf=Pass smtp.mailfrom=mshinwell@janestreet.com; spf=None smtp.helo=postmaster@mxout1.mail.janestreet.com Received-SPF: None (mail3-smtp-sop.national.inria.fr: no sender authenticity information available from domain of mshinwell@janestreet.com) identity=pra; client-ip=38.105.200.78; receiver=mail3-smtp-sop.national.inria.fr; envelope-from="mshinwell@janestreet.com"; x-sender="mshinwell@janestreet.com"; x-conformance=sidf_compatible Received-SPF: Pass (mail3-smtp-sop.national.inria.fr: domain of mshinwell@janestreet.com designates 38.105.200.78 as permitted sender) identity=mailfrom; client-ip=38.105.200.78; receiver=mail3-smtp-sop.national.inria.fr; envelope-from="mshinwell@janestreet.com"; x-sender="mshinwell@janestreet.com"; x-conformance=sidf_compatible; x-record-type="v=spf1" Received-SPF: None (mail3-smtp-sop.national.inria.fr: no sender authenticity information available from domain of postmaster@mxout1.mail.janestreet.com) identity=helo; client-ip=38.105.200.78; receiver=mail3-smtp-sop.national.inria.fr; envelope-from="mshinwell@janestreet.com"; x-sender="postmaster@mxout1.mail.janestreet.com"; x-conformance=sidf_compatible IronPort-PHdr: =?us-ascii?q?9a23=3AQlnmdB0CNhsKr69IsmDT+DRfVm0co7zxezQtwd8Z?= =?us-ascii?q?sesfLfad9pjvdHbS+e9qxAeQG96Eu7QZ06L/iOPJZy8p2d65qncMcZhBBVcuqP?= =?us-ascii?q?49uEgeOvODElDxN/XwbiY3T4xoXV5h+GynYwAOQJ6tL3WbmHC57CYTFxPjLkI1?= =?us-ascii?q?Y72tQs+Bx/iwgsK14J2bSQhSiDu7bPsmLhKtrS3bu9MagI1vbKA81k2ajGFPfr?= =?us-ascii?q?FzzHlrbWmTnhDj68O0tMpm6CRRve0s88poU6z8cr8kV7FVEHItNGVjt56jjgXK?= =?us-ascii?q?UQbavihUaW4RiBcdRlGdtBw=3D?= X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: =?us-ascii?q?A0CSAQD8vshZfU7IaSZbGwEBAQMBAQEJA?= =?us-ascii?q?QEBFQEBAQECAQEBAQgBAQEBhBNuJweDb4E2mGOBdpBsh1AKJYUWAoQmB0MUAQE?= =?us-ascii?q?BAQEBAQEBAQESAQEJFghXgjMFAR4BBYI7AQEBAQIBIx0BATcBBAsLBAcNKgICI?= =?us-ascii?q?hIBBQEcBhOKKwgDDZpbQIsga4IngwkBAQWIAgEBAQEBAQEBAgEBAQEBAQEBAQE?= =?us-ascii?q?WAwUSgxmCAoFRhQ+FMIJmgmCIMZhzhmN6jH+LXIcqlTIUBR+BFTaBL1MlXhoGh?= =?us-ascii?q?EEqghN1BYZagVQBAQE?= X-IPAS-Result: =?us-ascii?q?A0CSAQD8vshZfU7IaSZbGwEBAQMBAQEJAQEBFQEBAQECAQE?= =?us-ascii?q?BAQgBAQEBhBNuJweDb4E2mGOBdpBsh1AKJYUWAoQmB0MUAQEBAQEBAQEBAQESA?= =?us-ascii?q?QEJFghXgjMFAR4BBYI7AQEBAQIBIx0BATcBBAsLBAcNKgICIhIBBQEcBhOKKwg?= =?us-ascii?q?DDZpbQIsga4IngwkBAQWIAgEBAQEBAQEBAgEBAQEBAQEBAQEWAwUSgxmCAoFRh?= =?us-ascii?q?Q+FMIJmgmCIMZhzhmN6jH+LXIcqlTIUBR+BFTaBL1MlXhoGhEEqghN1BYZagVQ?= =?us-ascii?q?BAQE?= X-IronPort-AV: E=Sophos;i="5.42,435,1500933600"; d="scan'208,217";a="238585155" Received: from mxout1.mail.janestreet.com ([38.105.200.78]) by mail3-smtp-sop.national.inria.fr with ESMTP/TLS/DHE-RSA-AES256-GCM-SHA384; 25 Sep 2017 10:32:30 +0200 Received: from [172.27.56.106] (helo=tot-qpr-mailcore2) by mxout1.mail.janestreet.com with esmtps (TLSv1.2:ECDHE-RSA-AES256-GCM-SHA384:256) (Exim 4.89) (envelope-from ) id 1dwOoW-00049B-So for caml-list@inria.fr; Mon, 25 Sep 2017 04:32:28 -0400 X-JS-Flow: external X-JS-Scanner-attachment: (ok) No attachments Received: by tot-qpr-mailcore2 with ocaml/mailcore/mailcore 1.0+114 (d8ed010257c7) (envelope-from ) id BZyL8c-8iBUIA-bd; 2017-09-25 04:32:28.883598-04:00 Received: from mail-pg0-f72.google.com ([74.125.83.72]) by mxgoog1.mail.janestreet.com with esmtps (TLSv1.2:ECDHE-RSA-AES128-GCM-SHA256:128) (Exim 4.89) (envelope-from ) id 1dwOoW-0001Ov-RO for caml-list@inria.fr; Mon, 25 Sep 2017 04:32:28 -0400 Received: by mail-pg0-f72.google.com with SMTP id i130so15041103pgc.5 for ; Mon, 25 Sep 2017 01:32:28 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=janestreet.com; s=google; h=mime-version:in-reply-to:references:from:date:message-id:subject:to :cc; bh=zVo7BdBJ1fR9AWQlooceoVDeFeBnprj1VCb0KXX1mgo=; b=xMxP006vAqr0djR3+TM+APLtzBt3b9yniS30Yplgui83LpBgq9mKuy2i/vhGHVac2B w+7aAZ4PMEh7vT7mKXVulVMK4bKo22vxsIOSfj9953SLWYxBYEf2Pn3AhKPghjYUbUuL /ZqmfElSWrUyEYOTnHoSehyZLwJH7nLlAIowM= X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:mime-version:in-reply-to:references:from:date :message-id:subject:to:cc; bh=zVo7BdBJ1fR9AWQlooceoVDeFeBnprj1VCb0KXX1mgo=; b=Fx+IVJFOWZvlAlMuv5AeZ7iKPAjYXmOoTzSw8Ti4ZIFNXGND+ci9tCG02HJX+aMihe ReZiXkeTBa24FWwRYsSO3dcC7jnmTvs0WQcAgkAEOQZ0QxdQ0xM5PGcst6wsOtB2ObmA XQZF2bHGius3GrWdkI1M3NDi/FLIzaNoyVy1U3cpGYUbDlXWa58CgvXtL1shnilRGMJ6 Al7CyLJZpD7zUquQuctOP5TbsoNiwIL3rwLxIyC4oe7ixpAERo1fUCVI8SEtTSls939/ 9/EIDPrsET5vmq+C2thF+rwOofYQ2DtTxLsk4qGcWwGLl61Mka58PtKI8PyVGXbavZB8 JrGA== X-Gm-Message-State: AHPjjUg58WgErPKfgu6cOU55EATGyek19/WtyEUmdv04vj2xh822CeTB YCPlhtN/yHf17LUS/+QIuVaL+O1Ul3cJl6nI8s6bwIHGLGBfmrb/p8+F0Il6ozxWS4JsJAbThZR U5CtzZYfc12qoxXa53s/zUorrlw== X-Received: by 10.98.93.85 with SMTP id r82mr6898398pfb.327.1506328348075; Mon, 25 Sep 2017 01:32:28 -0700 (PDT) X-Google-Smtp-Source: AOwi7QDgE2vhjOKwlOe6i3CGEKWxn8yqsp03ej5Fd/akTXNC7trIYfoPNw/FOmQbIaWlgeGo5o8t8DRyu7+tpa/50Ys= X-Received: by 10.98.93.85 with SMTP id r82mr6898396pfb.327.1506328347902; Mon, 25 Sep 2017 01:32:27 -0700 (PDT) MIME-Version: 1.0 Received: by 10.100.179.1 with HTTP; Mon, 25 Sep 2017 01:32:27 -0700 (PDT) In-Reply-To: <33303E31-1874-4BD1-90D9-E9E8F623A6C6@berdine.net> References: <41FF6520-51D5-4EB7-A8AF-6BB9C495D1E8@gmx.net> <690A132C-892E-45DC-A884-415E92054AC4@gmx.net> <03ED974A-ACAE-4EEB-AD46-FC4B863F228E@gmx.net> <33303E31-1874-4BD1-90D9-E9E8F623A6C6@berdine.net> From: Mark Shinwell Date: Mon, 25 Sep 2017 09:32:27 +0100 Message-ID: To: Josh Berdine Cc: caml users Content-Type: multipart/alternative; boundary="001a113de7de17a2d90559ff6609" X-JS-Exim-Data-Received: 2017-09-25 04:32:28-0400 X-JS-Processed-by: mailcore X-Validation-by: mshinwell@janestreet.com Subject: Re: [Caml-list] Performance penalty for using monad --001a113de7de17a2d90559ff6609 Content-Type: text/plain; charset="UTF-8" I would also point out that Facebook's experience with compilation speed under Flambda is actually rather worse than what we've found at Jane Street. We are typically seeing closer to 2.5x rather than 5x at -O3. However we are working to reduce this further. Mark On 24 September 2017 at 21:04, Josh Berdine wrote: > On Sep 24, 2017, at 12:28 PM, Yaron Minsky wrote: > > For what it's worth, the main blocker for us turning Flambda on by default > in classic mode is getting build artifact size and compilation speed down > basically to the same level as closure-compilation. We're getting pretty > close to that goal, though it will take a bit more time to get the > improvements in question upstreamed. > > So getting Flambda enabled by default isn't that far away (though most of > the real benefits will require -O3, which will still lengthen compilation > by quite a bit.) > > > Another data point on this: at facebook we recently switched the > production infer binaries over to flambda -O3 ( > https://github.com/facebook/infer/commit/f8d7c810452ce3a4d2e7027e38f5d0 > 0426a2a917). For local builds during development, we usually build > without flambda, or actually even just bytecode. But for infer, flambda -O3 > is worth 15-20% elapsed (~25% cpu) time, so it does not take an abnormal > analysis run before that pays off the ~5x compile time deficit. (Given that > we have to distribute a custom clang with the analyzer, build artifact size > is basically in the noise.) > > Cheers, Josh > > --001a113de7de17a2d90559ff6609 Content-Type: text/html; charset="UTF-8" Content-Transfer-Encoding: quoted-printable
I would also point out that Facebook's experience with= compilation speed under Flambda is actually rather worse than what we'= ve found at Jane Street.=C2=A0 We are typically seeing closer to 2.5x rathe= r than 5x at -O3.=C2=A0 However we are working to reduce this further.
<= div>
Mark

On 24 September 2017 at 21:04, Josh Berdine <josh@berdine= .net> wrote:
On Sep 24, 2017, at 12:28 PM, Yaron = Minsky <ymin= sky@janestreet.com> wrote:
F= or what it's worth, the main blocker for us turning Flambda on by defau= lt in classic mode is getting build artifact size and compilation speed dow= n basically to the same level as closure-compilation. We're getting pre= tty close to that goal, though it will take a bit more time to get the impr= ovements in question upstreamed.

So getting Flambd= a enabled by default isn't that far away (though most of the real benef= its will require -O3, which will still lengthen compilation by quite a bit.= )

Another d= ata point on this: at facebook we recently switched the production infer bi= naries over to flambda -O3 (https://gi= thub.com/facebook/infer/commit/f8d7c810452ce3a4d2e7027e38f5d00426a2a917). For local builds during development, we usually build wit= hout flambda, or actually even just bytecode. But for infer, flambda -O3 is= worth 15-20% elapsed (~25% cpu) time, so it does not take an abnormal anal= ysis run before that pays off the ~5x compile time deficit. (Given that we = have to distribute a custom clang with the analyzer, build artifact size is= basically in the noise.)

Cheers, Josh
<= br>

--001a113de7de17a2d90559ff6609--