From owner-freebsd-questions@freebsd.org Tue May 12 12:32:18 2020 Return-Path: Delivered-To: freebsd-questions@mailman.nyi.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2610:1c1:1:606c::19:1]) by mailman.nyi.freebsd.org (Postfix) with ESMTP id 2D0C12EDD1D for ; Tue, 12 May 2020 12:32:18 +0000 (UTC) (envelope-from 4250.82.1d4c1000580a213.84c5845b14bad6b9a6edf2dbcf6cfbb7@email-od.com) Received: from s1-b0c6.socketlabs.email-od.com (s1-b0c6.socketlabs.email-od.com [142.0.176.198]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (Client did not present a certificate) by mx1.freebsd.org (Postfix) with ESMTPS id 49LxyS4h91z4CR3 for ; Tue, 12 May 2020 12:32:16 +0000 (UTC) (envelope-from 4250.82.1d4c1000580a213.84c5845b14bad6b9a6edf2dbcf6cfbb7@email-od.com) DKIM-Signature: v=1; a=rsa-sha256; d=email-od.com;i=@email-od.com;s=dkim; c=relaxed/relaxed; q=dns/txt; t=1589286737; x=1591878737; h=content-transfer-encoding:content-type:mime-version:references:in-reply-to:message-id:subject:cc:to:from:date:x-thread-info; bh=GUH7zaue+aeCh946E/ucjr1fF69DHojtF1oFp4VlOOY=; b=baQxo4WMPruEsl760VKuA0rHehLOGkZ1gbULmJrb5/P5h2HR9s68tVwJqDYIAiW8J/FO3h0Ft+2pJJiyQdEJAeQ21uufjv5v5SDxmUwfY5UudGxv0iqgX6Dzh86yNPANz5Df+xfV5Jr6nx2uaDjZiUETyviQAxuZlPQ9TYl4uUs= X-Thread-Info: NDI1MC45Mi4xZDRjMTAwMDU4MGEyMTMuZnJlZWJzZC1xdWVzdGlvbnM9ZnJlZWJzZC5vcmc= Received: from r2.h.in.socketlabs.com (r2.h.in.socketlabs.com [142.0.180.12]) by mxsg2.email-od.com with ESMTP(version=Tls12 cipher=Aes256 bits=256); Tue, 12 May 2020 08:32:13 -0400 Received: from smtp.lan.sohara.org (EMTPY [185.202.17.215]) by r2.h.in.socketlabs.com with ESMTP(version=Tls12 cipher=Aes256 bits=256); Tue, 12 May 2020 08:32:12 -0400 Received: from [192.168.63.1] (helo=steve.lan.sohara.org) by smtp.lan.sohara.org with smtp (Exim 4.92.3 (FreeBSD)) (envelope-from ) id 1jYU4w-0005ND-PI; Tue, 12 May 2020 13:32:10 +0100 Date: Tue, 12 May 2020 13:32:10 +0100 From: Steve O'Hara-Smith To: Nikita Stepanov Cc: freebsd-questions@freebsd.org Subject: Re: Recompiling the kernel increases performance? Message-Id: <20200512133210.3e5ed9aac65903ebdee3ea56@sohara.org> In-Reply-To: <825101589284883@sas1-ffdbcd5f1d77.qloud-c.yandex.net> References: <825101589284883@sas1-ffdbcd5f1d77.qloud-c.yandex.net> X-Mailer: Sylpheed 3.7.0 (GTK+ 2.24.32; amd64-portbld-freebsd12.0) X-Clacks-Overhead: "GNU Terry Pratchett" Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Rspamd-Queue-Id: 49LxyS4h91z4CR3 X-Spamd-Bar: / Authentication-Results: mx1.freebsd.org; dkim=pass header.d=email-od.com header.s=dkim header.b=baQxo4WM; dmarc=none; spf=pass (mx1.freebsd.org: domain of 4250.82.1d4c1000580a213.84c5845b14bad6b9a6edf2dbcf6cfbb7@email-od.com designates 142.0.176.198 as permitted sender) smtp.mailfrom=4250.82.1d4c1000580a213.84c5845b14bad6b9a6edf2dbcf6cfbb7@email-od.com X-Spamd-Result: default: False [-0.37 / 15.00]; ARC_NA(0.00)[]; NEURAL_HAM_MEDIUM(-0.74)[-0.743,0]; R_DKIM_ALLOW(-0.20)[email-od.com:s=dkim]; FROM_HAS_DN(0.00)[]; TO_DN_SOME(0.00)[]; R_SPF_ALLOW(-0.20)[+ip4:142.0.176.0/20]; MV_CASE(0.50)[]; MIME_GOOD(-0.10)[text/plain]; RCVD_TLS_LAST(0.00)[]; DMARC_NA(0.00)[sohara.org]; NEURAL_HAM_LONG(-0.95)[-0.950,0]; RCVD_COUNT_THREE(0.00)[4]; TO_MATCH_ENVRCPT_SOME(0.00)[]; DKIM_TRACE(0.00)[email-od.com:+]; RCPT_COUNT_TWO(0.00)[2]; RCVD_IN_DNSWL_NONE(0.00)[198.176.0.142.list.dnswl.org : 127.0.15.0]; IP_SCORE(0.02)[ip: (-0.19), ipnet: 142.0.176.0/22(0.27), asn: 7381(0.08), country: US(-0.05)]; FORGED_SENDER(0.30)[steve@sohara.org,4250.82.1d4c1000580a213.84c5845b14bad6b9a6edf2dbcf6cfbb7@email-od.com]; SUBJECT_ENDS_QUESTION(1.00)[]; MIME_TRACE(0.00)[0:+]; ASN(0.00)[asn:7381, ipnet:142.0.176.0/22, country:US]; FROM_NEQ_ENVFROM(0.00)[steve@sohara.org,4250.82.1d4c1000580a213.84c5845b14bad6b9a6edf2dbcf6cfbb7@email-od.com]; MID_RHS_MATCH_FROM(0.00)[] X-BeenThere: freebsd-questions@freebsd.org X-Mailman-Version: 2.1.33 Precedence: list List-Id: User questions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 12 May 2020 12:32:18 -0000 On Tue, 12 May 2020 18:01:23 +0600 Nikita Stepanov wrote: Recompiling the kernel may improve the performance of some CPU bound operations if you adjust optimisation flags and compile to a specific CPU architecture. However unless your system load is such that it is CPU bound in kernel space it is unlikely to make a difference to the overall performance of the system at the cost of increasing the risk of hitting optimisation bugs. Very high performance routing is one of the few applications I can think of where I might expect a noticeable difference from optimising kernel code. -- Steve O'Hara-Smith