From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752022AbcIDGzQ (ORCPT ); Sun, 4 Sep 2016 02:55:16 -0400 Received: from mout.web.de ([212.227.15.14]:57897 "EHLO mout.web.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751015AbcIDGzO (ORCPT ); Sun, 4 Sep 2016 02:55:14 -0400 Subject: Re: sparc: bpf_jit: Move four assignments in bpf_jit_compile() To: Julian Calaby References: <566ABCD9.1060404@users.sourceforge.net> <2179bf7c-9878-adf7-da97-2746d5aa3d34@users.sourceforge.net> <1af7e987-0233-f972-6c00-6d5e00898188@users.sourceforge.net> Cc: sparclinux , Adam Buchbinder , Alexei Starovoitov , Daniel Borkmann , "David S. Miller" , Rabin Vincent , LKML , kernel-janitors , Julia Lawall , Paolo Bonzini From: SF Markus Elfring Message-ID: <193016ad-1b0d-50f1-44b8-ee34dbc2924b@users.sourceforge.net> Date: Sun, 4 Sep 2016 08:28:13 +0200 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:45.0) Gecko/20100101 Thunderbird/45.3.0 MIME-Version: 1.0 In-Reply-To: Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 8bit X-Provags-ID: V03:K0:6GOSP5hj9bnCUCNyZp0VT+Sk8dKrVL64TwuyTQe9QEU+FjANYm/ gqkmfENZNFG4Nz97PDgySx+Ao8kIbb4XKqRty5HBB7L1LBg+Ff4i/3ee6RxxfOKECMYB7xx DTqq6Tf7WKOlwymq88YE+tZT9XuKcMfgbLvPxHoUUM/YNXAp/TMs2ySlNnxAC5KCw1Ut5Cz mdpAW1pYH5yAg33FdGXZg== X-UI-Out-Filterresults: notjunk:1;V01:K0:OEofhBkFw68=:UpmSplsHIE+9jvCL0x4wxb FAVlaDIlrL9W4Wx7mLuGpiITYIRjbV2CXxkugaDhWUcpVHrU6mSv6BP6VFELq2qFiXcRvpYEH oDK7aESB3oTnMmfzCw7zWS2vYAgzTtkqK9CKsxmwBsFVc9VFmBAEu+7NJTGcmxfcD17s/GeBS PCngiiqOtd/PGMcoUW37YM/623vbmG8Ifya0CL4wrB8+9WL7R9bOYO7T2qZh5Rmzs4UAaFooO qH0joc3Hpu6j6mJ4hB2d6QrfBlwJFFIlhU1NqePhDNxIRyENgPe2MFltO2TYvxaMp265ZxOnf Mz1ZQRJwSfYPiQKl3Jwq8mO2UUlh7s1Yktf5m44cnXBmvw5ehayUIPCY1DIE++RtZBRIhHXbS b7NwS6BgkUPOtUylRboxnDHCUe0uObx4Cqwn71Rbk7CZuJFBj4nCNPz8nZMqQR7xJ0vPjzQas dFEC5Y2Au9huGbOoTZWGmz6jdvkpxlnkymA9LDsASXuGIa4RxejDyogZU3Vkhf/2kowxjfsTz s5PKWZniUGyoSNsjSKVe7fG/ykYvC6NGPbtIQjgf/J2N9PUQa6VY289Y1mUna2qLOxTuhOyA5 NY+cg+sWE71G6x5UvN7jXmxPG2W8W7M5pI8YykJIfGH15/K0IjU8iTYwpTc+FfwOzA54RPLlR 3V7jsIZaAkn0+pZRTl1DBkRzyrprkHhuRI6xBxUgaRRxNfesIW+7JKwjtA4vkPWsE63IjkxPW /JczPpUOjZYqNJU7w1wVQSiuRFg2Fjqf5Hv5Y2g3aqDAxTZVPQVvBsm8vQaQkjR9PGYLyH+xR C/LD3iC Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org >> Which test environments would you find acceptable for further clarification? > > Compiling it on GCC for Sparc, obviously. Are there any more configuration details to consider? >>> I must also point out that these sorts of optimisations are things the >>> compiler does automatically when compiling this code. >> >> Do you take this detail for granted? > > I trust that the GCC developers have done their work well. Will any more compiler implementations matter here? Do you like software which can run better by default also without application of special compilation parameters? > I'm looking for some glimmer of usefullness in this patch. I'm not seeing any. Thanks for your honest feedback. >> Should we avoid to compare software things similar to "apples" and "oranges" >> (while these fruits can make more fun)? ;-) > > Write a benchmark that exercises this function. Measure the time it > took without this change, measure the time it took with this change, > is there a difference. Is an accepted test system already available for the purpose that every commit would be checked in the way automatically you expect here? > You cannot expect people to take you seriously if you're proposing > performance changes without any actual ability or interest in > producing performance related data to go along with them. I suggested small changes which I found "logical". > You're essentially saying "I think doing things this way is better" Yes … > and providing _nothing_ else. You might be looking for more information than I can practically give you at the moment. > I think that things are perfectly fine the way they are. I have got an other impression for "perfection" in this software module. I found an implementation detail for further considerations. > Convince me with data or something else. I imagine that the "else" can become harder than you find reasonable. > Did someone do this to some other driver and you're applying the same fix elsewhere? Is a similar software development discussion still running for other modules? > You are the only person proposing changes like these ones as you are I am picking special software improvement opportunities up. > (as far as I know) the only person who thinks they have any value. I can accept that the value of specific changes will usually be categorised as lower than updates that you prefer so far. Regards, Markus