From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from alt-proxy28.mail.unifiedlayer.com (alt-proxy28.mail.unifiedlayer.com [74.220.216.123]) by sourceware.org (Postfix) with ESMTPS id 677B63857BBD for ; Fri, 15 Jul 2022 19:06:05 +0000 (GMT) DMARC-Filter: OpenDMARC Filter v1.4.1 sourceware.org 677B63857BBD Authentication-Results: sourceware.org; dmarc=none (p=none dis=none) header.from=tromey.com Authentication-Results: sourceware.org; spf=pass smtp.mailfrom=tromey.com Received: from cmgw13.mail.unifiedlayer.com (unknown [10.0.90.128]) by progateway1.mail.pro1.eigbox.com (Postfix) with ESMTP id B80A91003A622 for ; Fri, 15 Jul 2022 19:05:49 +0000 (UTC) Received: from box5379.bluehost.com ([162.241.216.53]) by cmsmtp with ESMTP id CQdJodQICWBOVCQdJoHvpG; Fri, 15 Jul 2022 19:05:49 +0000 X-Authority-Reason: nr=8 X-Authority-Analysis: v=2.4 cv=cpReL30i c=1 sm=1 tr=0 ts=62d1ba8d a=ApxJNpeYhEAb1aAlGBBbmA==:117 a=ApxJNpeYhEAb1aAlGBBbmA==:17 a=dLZJa+xiwSxG16/P+YVxDGlgEgI=:19 a=RgO8CyIxsXoA:10:nop_rcvd_month_year a=Qbun_eYptAEA:10:endurance_base64_authed_username_1 a=CCpqsmhAAAAA:8 a=SnvJyErxX2ha6y56pnMA:9 a=ul9cdbp4aOFLsgKbc677:22 DKIM-Signature: v=1; a=rsa-sha256; q=dns/txt; c=relaxed/relaxed; d=tromey.com; s=default; h=Content-Type:MIME-Version:Message-ID:In-Reply-To:Date:References :Subject:Cc:To:From:Sender:Reply-To:Content-Transfer-Encoding:Content-ID: Content-Description:Resent-Date:Resent-From:Resent-Sender:Resent-To:Resent-Cc :Resent-Message-ID:List-Id:List-Help:List-Unsubscribe:List-Subscribe: List-Post:List-Owner:List-Archive; bh=mAQ7lvd4WG3L+XxMcgMkNmCmpvJeu5D7CajjK9OvXqY=; b=J9lR6b1yU049+LSVg1snFq5G7A hzSZ90AIttb2a2DqaKZKOzXaAtuxTP4KYXqbrvsWxrfHJkQd6c9DEoO+FBw1MCY4j7IdKG9fOT5BM sQ0u63b+b8mxMvib3tKWulCEa; Received: from 71-211-185-228.hlrn.qwest.net ([71.211.185.228]:56128 helo=murgatroyd) by box5379.bluehost.com with esmtpsa (TLS1.2) tls TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384 (Exim 4.95) (envelope-from ) id 1oCQdJ-004G6u-4r; Fri, 15 Jul 2022 13:05:49 -0600 From: Tom Tromey To: Tom de Vries via Gdb-patches Subject: Re: [RFC][gdbsupport] Improve thread scheduling in parallel_for_each References: <20220715094034.GA10751@delia.home> X-Attribution: Tom Date: Fri, 15 Jul 2022 13:05:48 -0600 In-Reply-To: <20220715094034.GA10751@delia.home> (Tom de Vries via Gdb-patches's message of "Fri, 15 Jul 2022 11:40:35 +0200") Message-ID: <87y1wumc1f.fsf@tromey.com> User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/27.2 (gnu/linux) MIME-Version: 1.0 Content-Type: text/plain X-AntiAbuse: This header was added to track abuse, please include it with any abuse report X-AntiAbuse: Primary Hostname - box5379.bluehost.com X-AntiAbuse: Original Domain - sourceware.org X-AntiAbuse: Originator/Caller UID/GID - [47 12] / [47 12] X-AntiAbuse: Sender Address Domain - tromey.com X-BWhitelist: no X-Source-IP: 71.211.185.228 X-Source-L: No X-Exim-ID: 1oCQdJ-004G6u-4r X-Source: X-Source-Args: X-Source-Dir: X-Source-Sender: 71-211-185-228.hlrn.qwest.net (murgatroyd) [71.211.185.228]:56128 X-Source-Auth: tom+tromey.com X-Email-Count: 3 X-Source-Cap: ZWx5bnJvYmk7ZWx5bnJvYmk7Ym94NTM3OS5ibHVlaG9zdC5jb20= X-Local-Domain: yes X-Spam-Status: No, score=-3022.8 required=5.0 tests=BAYES_00, DKIM_SIGNED, DKIM_VALID, JMQ_SPF_NEUTRAL, RCVD_IN_DNSWL_NONE, RCVD_IN_MSPIKE_H2, SPF_HELO_NONE, SPF_PASS, TXREP autolearn=no autolearn_force=no version=3.4.6 X-Spam-Checker-Version: SpamAssassin 3.4.6 (2021-04-09) on server2.sourceware.org X-BeenThere: gdb-patches@sourceware.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: Gdb-patches mailing list List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 15 Jul 2022 19:06:06 -0000 >>>>> "Tom" == Tom de Vries via Gdb-patches writes: Tom> This introduces a performance regression on a particular test-case I happened Tom> to use: Tom> ... Tom> $ for n in $(seq 1 10); do \ Tom> time gdb -q -batch libxul.so.debug 2>&1 | grep real:; \ Tom> done Tom> ... Tom> so revert to the original schedule by reducing the worker threads: Tom> ... This seems like making a change and then undoing it somewhere else? Tom> Still, the performance experiment yields a slight performance loss. Sounds bad. Tom> if (n_threads < 0) Tom> - n_threads = std::thread::hardware_concurrency (); Tom> + { Tom> + n_threads = std::thread::hardware_concurrency (); Tom> + if (n_threads > 0) Tom> + /* Account for main thread. */ Tom> + n_threads--; Tom> + } I think it's better if the setting just directly controls how many threads there are. Then elsewhere we can decide what that means -- like if it performs better with the defaults to not do any work in the main thread, then parallel_for_each can be modified to just send tasks to the workers and do nothing in the main thread except wait for results. Tom> size_t elts_per_thread = 0; [...] Tom> + elts_per_thread = n_elements / n_threads; The initial declaration can be removed and then this latter line can declare the variable as well. Tom> for (int i = 0; i < count; ++i) Tom> { Tom> RandomIt end = first + elts_per_thread; Tom> + if (i < left_over) Tom> + end++; It may be nice to mention the distribution of leftovers in a comment somewhere. Tom