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 mail2-relais-roc.national.inria.fr (mail2-relais-roc.national.inria.fr [192.134.164.83]) by sympa.inria.fr (Postfix) with ESMTPS id A15D08021B for ; Sun, 24 Sep 2017 22:05:00 +0200 (CEST) Authentication-Results: mail2-smtp-roc.national.inria.fr; spf=None smtp.pra=josh@berdine.net; spf=Pass smtp.mailfrom=josh@berdine.net; spf=None smtp.helo=postmaster@out1-smtp.messagingengine.com Received-SPF: None (mail2-smtp-roc.national.inria.fr: no sender authenticity information available from domain of josh@berdine.net) identity=pra; client-ip=66.111.4.25; receiver=mail2-smtp-roc.national.inria.fr; envelope-from="josh@berdine.net"; x-sender="josh@berdine.net"; x-conformance=sidf_compatible Received-SPF: Pass (mail2-smtp-roc.national.inria.fr: domain of josh@berdine.net designates 66.111.4.25 as permitted sender) identity=mailfrom; client-ip=66.111.4.25; receiver=mail2-smtp-roc.national.inria.fr; envelope-from="josh@berdine.net"; x-sender="josh@berdine.net"; x-conformance=sidf_compatible; x-record-type="v=spf1" Received-SPF: None (mail2-smtp-roc.national.inria.fr: no sender authenticity information available from domain of postmaster@out1-smtp.messagingengine.com) identity=helo; client-ip=66.111.4.25; receiver=mail2-smtp-roc.national.inria.fr; envelope-from="josh@berdine.net"; x-sender="postmaster@out1-smtp.messagingengine.com"; x-conformance=sidf_compatible IronPort-PHdr: =?us-ascii?q?9a23=3AucV2gx35l5V33PzXsmDT+DRfVm0co7zxezQtwd8Z?= =?us-ascii?q?sesQLPad9pjvdHbS+e9qxAeQG96Eu7QZ06L/iOPJZy8p2d65qncMcZhBBVcuqP?= =?us-ascii?q?49uEgeOvODElDxN/XwbiY3T4xoXV5h+GynYwAOQJ6tLw6annrn5jcXHl3iLgdv?= =?us-ascii?q?PayhEYfXi4Gz1vuu05zVeQRBwjSnN+BcNhKz+CfWrcleq4Z4Jqc3x1OdoHxTe8?= =?us-ascii?q?xcyHlkJFSe2RDx+pHjr9ZY7y1Mtqd5pIZ7WqLgcvFgQA=3D=3D?= X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: =?us-ascii?q?A0DuAQCJDshZhhkEb0JcGwEBAQMBAQEJA?= =?us-ascii?q?QEBFwEBBAEBCgEBhBMDayePCo8FgXaYPAolhRYChCRDFAEBAQEBAQEBAQEBEgE?= =?us-ascii?q?BAQgLCwgoL4IzIoJEAQIDQAEBOA8LBBQuVwYTijMQpkyDEoMJAQEFiAIBAQEBB?= =?us-ascii?q?gEBAQEBARoDBYMrggKDYwuCcoUwgikMYIIxiDGYc4Zjeo1xilcThyqSD4M3gTk?= =?us-ascii?q?2gS8yIQgoCkkSAYJ0ggQfgXR1BYgIAQEB?= X-IPAS-Result: =?us-ascii?q?A0DuAQCJDshZhhkEb0JcGwEBAQMBAQEJAQEBFwEBBAEBCgE?= =?us-ascii?q?BhBMDayePCo8FgXaYPAolhRYChCRDFAEBAQEBAQEBAQEBEgEBAQgLCwgoL4IzI?= =?us-ascii?q?oJEAQIDQAEBOA8LBBQuVwYTijMQpkyDEoMJAQEFiAIBAQEBBgEBAQEBARoDBYM?= =?us-ascii?q?rggKDYwuCcoUwgikMYIIxiDGYc4Zjeo1xilcThyqSD4M3gTk2gS8yIQgoCkkSA?= =?us-ascii?q?YJ0ggQfgXR1BYgIAQEB?= X-IronPort-AV: E=Sophos;i="5.42,433,1500933600"; d="scan'208,217";a="292493127" Received: from out1-smtp.messagingengine.com ([66.111.4.25]) by mail2-smtp-roc.national.inria.fr with ESMTP/TLS/DHE-RSA-AES256-GCM-SHA384; 24 Sep 2017 22:04:59 +0200 Received: from compute1.internal (compute1.nyi.internal [10.202.2.41]) by mailout.nyi.internal (Postfix) with ESMTP id C53A4209B1; Sun, 24 Sep 2017 16:04:58 -0400 (EDT) Received: from frontend2 ([10.202.2.161]) by compute1.internal (MEProxy); Sun, 24 Sep 2017 16:04:58 -0400 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=berdine.net; h= content-type:date:from:in-reply-to:message-id:mime-version :references:subject:to:x-me-sender:x-me-sender:x-sasl-enc :x-sasl-enc; s=fm1; bh=bVnCP3u9tSTvCjc1QiX9E/MZfsG3o6Ev4TfzSh2sD qs=; b=mwG4FX4MiiEOFJKdsSu8JszzPyNk/mVCdx/tgHlkfH+WDS1Va+dx/tuIC zsyM+mSHJRaBZtvo+VgP/jFGKWQNFrkwyFqVBD0zGoRragriMa9nOWFOGJqB4LsG cJ43p3aiqol8b+UOQWZ1HI3M7flvFChbq6UEhkRDBmKsVv4KiukVGHO3wNHOIZT7 Yr7cTeILX6iMjxMtyqV6y812PgTDZ87kC9Hu248aQiHNTCOBAYXImm6bxzCAx0O8 mYYtoV+6IvcBqRMZuweGQe9hWt9rQZbNTA8Br/iXLa77zCQeHHA0oUVL1O4/+VqX hS3Yhw0MqHyLy1MjeatzsIG4BtFCg== DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d= messagingengine.com; h=content-type:date:from:in-reply-to :message-id:mime-version:references:subject:to:x-me-sender :x-me-sender:x-sasl-enc:x-sasl-enc; s=fm1; bh=bVnCP3u9tSTvCjc1Qi X9E/MZfsG3o6Ev4TfzSh2sDqs=; b=GJK+kw8SXJP7XLHQPOwDLjIuG/fESokFw0 RJ3Ow+QpN3IFLSI2g4rn7B4W/HiPnd5fmb/kf8D7VtFmnDgpt9i0zyoV5U53Mirs MuYn8Wk3cDRwe8ZSo5k+ish4H8eljnIqwuUMET/CWVzA/ayJim49/0zmsWeeimZM fGETpkV2+8dsUod5BURUglXOhrH0T1TXiggnuf13cnTmcgiegxTZOJE+KYvL+hn6 hFgEmzrrmPQHxdfxsFp8wXVm7YMxj4zQliEXXGGF4FToBtBlj5Djw7K42TYLk7ca Zo1JAKZW1THj1S8WVjNS4yP1ErGIZC+kiZZ1+tCb+PwH2G5D5eTA== X-ME-Sender: X-Sasl-enc: IvlkDw1AicVLTtrQGJYX1c4MhdZTWnRqpWiulorVzfTm 1506283498 Received: from [172.16.3.210] (unknown [199.201.66.3]) by mail.messagingengine.com (Postfix) with ESMTPA id 5D87C2489A for ; Sun, 24 Sep 2017 16:04:58 -0400 (EDT) From: Josh Berdine Content-Type: multipart/alternative; boundary="Apple-Mail=_2ACA5685-8453-45F8-AA62-D236CFEDE2E2" Mime-Version: 1.0 (Mac OS X Mail 10.3 \(3273\)) Date: Sun, 24 Sep 2017 21:04:56 +0100 References: <41FF6520-51D5-4EB7-A8AF-6BB9C495D1E8@gmx.net> <690A132C-892E-45DC-A884-415E92054AC4@gmx.net> <03ED974A-ACAE-4EEB-AD46-FC4B863F228E@gmx.net> To: caml users In-Reply-To: Message-Id: <33303E31-1874-4BD1-90D9-E9E8F623A6C6@berdine.net> X-Mailer: Apple Mail (2.3273) Subject: Re: [Caml-list] Performance penalty for using monad --Apple-Mail=_2ACA5685-8453-45F8-AA62-D236CFEDE2E2 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset=us-ascii 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 defaul= t in classic mode is getting build artifact size and compilation speed down= basically to the same level as closure-compilation. We're getting pretty c= lose to that goal, though it will take a bit more time to get the improveme= nts in question upstreamed. >=20 > 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/comm= it/f8d7c810452ce3a4d2e7027e38f5d00426a2a917 ). For local builds dur= ing development, we usually build without flambda, or actually even just by= tecode. 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 --Apple-Mail=_2ACA5685-8453-45F8-AA62-D236CFEDE2E2 Content-Transfer-Encoding: quoted-printable Content-Type: text/html; charset=us-ascii On Sep 24, 2017= , at 12:28 PM, Yaron Minsky <yminsky@janestreet.com> wrote:
For wha= t it's worth, the main blocker for us turning Flambda on by default in clas= sic mode is getting build artifact size and compilation speed down basicall= y to the same level as closure-compilation. We're getting pretty close to t= hat goal, though it will take a bit more time to get the improvements in qu= estion 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.)

<= /div>
Another data point on this: at facebook we recently switched the = production infer binaries over to flambda -O3 (https://github.com/facebook/infer/commit/f8d7c810452ce3a4d2e7027e38f5d004= 26a2a917). For local builds during development, we usually build withou= t flambda, or actually even just bytecode. But for infer, flambda -O3 is wo= rth 15-20% elapsed (~25% cpu) time, so it does not take an abnormal analysi= s run before that pays off the ~5x compile time deficit. (Given that we hav= e to distribute a custom clang with the analyzer, build artifact size is ba= sically in the noise.)

Cheers, Josh

= --Apple-Mail=_2ACA5685-8453-45F8-AA62-D236CFEDE2E2--