From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from resqmta-h1p-028593.sys.comcast.net (resqmta-h1p-028593.sys.comcast.net [IPv6:2001:558:fd02:2446::7]) by sourceware.org (Postfix) with ESMTPS id CD5953858C39 for ; Sat, 14 Jan 2023 02:13:06 +0000 (GMT) DMARC-Filter: OpenDMARC Filter v1.4.2 sourceware.org CD5953858C39 Authentication-Results: sourceware.org; dmarc=pass (p=none dis=none) header.from=comcast.net Authentication-Results: sourceware.org; spf=pass smtp.mailfrom=comcast.net Received: from resomta-h1p-028517.sys.comcast.net ([96.102.179.206]) by resqmta-h1p-028593.sys.comcast.net with ESMTP id GVp6p2yGMDzFEGW2Zp48HC; Sat, 14 Jan 2023 02:13:03 +0000 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=comcast.net; s=20190202a; t=1673662383; bh=6np+tzTm859wF79PccsVfz+re3jLAJ/P5d0UCdQttWo=; h=Received:Received:Content-Type:Mime-Version:Subject:From:Date: Message-Id:To:Xfinity-Spam-Result; b=Nj1cU4uZFO8squ1SCC/w8UCXRgVfYg9PHl7JoBeYqlLHCoAJmjnT+mo5/FdMQQis6 quKhff9EbM4Ry0jllWURVHpAW4uIm78tjNwptT9GpyhjZhZoHLHMQ/P2Gh3pO6JT+J ztZfUYqkmhUqgxRdZITzL/VFNed9k2aavAs+TBDaXawuRd+KhPZgLoRCqvNesJfkUn YvcGmfMckw/nEwEMrHL+aj309h1wUQRu9iyqOXTIb6ItoqPJ7SEXw3MgYDqnSbPDmT Bh6fSXSHE2vwp7fz3QVqabr0PKrYiN8K/VOcNo9EtVFY+ctQEgeFoGs5eTy10GNoWv WbLIDXWAdwTKA== Received: from smtpclient.apple ([73.60.223.101]) by resomta-h1p-028517.sys.comcast.net with ESMTPSA id GW2XpUDPjqfrKGW2YpY8n5; Sat, 14 Jan 2023 02:13:03 +0000 X-Xfinity-VAAS: gggruggvucftvghtrhhoucdtuddrgedvhedrleelgdeghecutefuodetggdotefrodftvfcurfhrohhfihhlvgemucevohhmtggrshhtqdftvghsihdpqfgfvfdppffquffrtefokffrnecuuegrihhlohhuthemuceftddunecusecvtfgvtghiphhivghnthhsucdlqddutddtmdenucfjughrpegtggfuhfgjffevgffkfhfvofesthhqmhdthhdtvdenucfhrhhomheprfgruhhlucfmohhnihhnghcuoehprghulhhkohhnihhnghestghomhgtrghsthdrnhgvtheqnecuggftrfgrthhtvghrnhepveekveelffeliefgiedufeehgeejtdfhgedujeehueekiedtgfetffevgffggfdvnecukfhppeejfedriedtrddvvdefrddutddunecuvehluhhsthgvrhfuihiivgeptdenucfrrghrrghmpehhvghlohepshhmthhptghlihgvnhhtrdgrphhplhgvpdhinhgvthepjeefrdeitddrvddvfedruddtuddpmhgrihhlfhhrohhmpehprghulhhkohhnihhnghestghomhgtrghsthdrnhgvthdpnhgspghrtghpthhtohepfedprhgtphhtthhopeholhhivhgrsegruggrtghorhgvrdgtohhmpdhrtghpthhtoheprhhitghhrghrugdrghhuvghnthhhvghrsehgmhgrihhlrdgtohhmpdhrtghpthhtohepghgttgdqphgrthgthhgvshesghgttgdrghhnuhdrohhrgh X-Xfinity-VMeta: sc=-100.00;st=legit Content-Type: text/plain; charset=us-ascii Mime-Version: 1.0 (Mac OS X Mail 16.0 \(3696.120.41.1.1\)) Subject: Re: [RFC] Introduce -finline-memset-loops From: Paul Koning In-Reply-To: Date: Fri, 13 Jan 2023 21:12:59 -0500 Cc: Richard Biener , gcc-patches@gcc.gnu.org Content-Transfer-Encoding: quoted-printable Message-Id: <5268E9F0-32BC-4BE3-8ADE-F7A430DBBCCA@comcast.net> References: To: Alexandre Oliva X-Mailer: Apple Mail (2.3696.120.41.1.1) X-Spam-Status: No, score=-2.8 required=5.0 tests=BAYES_00,DKIM_SIGNED,DKIM_VALID,DKIM_VALID_AU,DKIM_VALID_EF,FREEMAIL_FROM,SPF_HELO_PASS,SPF_PASS,TXREP autolearn=ham autolearn_force=no version=3.4.6 X-Spam-Checker-Version: SpamAssassin 3.4.6 (2021-04-09) on server2.sourceware.org List-Id: > On Jan 13, 2023, at 8:54 PM, Alexandre Oliva via Gcc-patches = wrote: >=20 > Hello, Richard, >=20 > Thank you for the feedback. >=20 > On Jan 12, 2023, Richard Biener wrote: >=20 >> On Tue, Dec 27, 2022 at 5:12 AM Alexandre Oliva via Gcc-patches >> wrote: >=20 >>> This patch extends the memset expansion to start with a loop, so as = to >>> still take advantage of known alignment even with long lengths, but >>> without necessarily adding store blocks for every power of two. >=20 >> I wonder if that isn't better handled by targets via the setmem = pattern, >=20 > That was indeed where I started, but then I found myself duplicating = the > logic in try_store_by_multiple_pieces on a per-target basis. >=20 > Target-specific code is great for tight optimizations, but the main > purpose of this feature is not an optimization. AFAICT it actually > slows things down in general (due to code growth, and to conservative > assumptions about alignment),=20 I thought machinery like the memcpy patterns have as one of their = benefits the ability to find the alignment of their operands and from = that optimize things. So I don't understand why you'd say = "conservative". paul