Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1761454AbbBIVmZ (ORCPT ); Mon, 9 Feb 2015 16:42:25 -0500 Received: from mail-la0-f46.google.com ([209.85.215.46]:32968 "EHLO mail-la0-f46.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1760442AbbBIVmW (ORCPT ); Mon, 9 Feb 2015 16:42:22 -0500 From: Rasmus Villemoes To: "Wang\, Yalin" Cc: "'Andrew Morton'" , "'Kirill A. Shutemov'" , "'arnd\@arndb.de'" , "'linux-arch\@vger.kernel.org'" , "'linux-kernel\@vger.kernel.org'" , "'linux\@arm.linux.org.uk'" , "'linux-arm-kernel\@lists.infradead.org'" Subject: Re: [RFC] change non-atomic bitops method Organization: D03 References: <35FD53F367049845BC99AC72306C23D1044A02027E0A@CNBJMBX05.corpusers.net> <20150202152909.13bfd11f192fb0268b2ab4bf@linux-foundation.org> <20150203011730.GA15653@node.dhcp.inet.fi> <35FD53F367049845BC99AC72306C23D1044A02027E0B@CNBJMBX05.corpusers.net> <35FD53F367049845BC99AC72306C23D1044A02027E0C@CNBJMBX05.corpusers.net> <20150202223851.f30768d0.akpm@linux-foundation.org> <35FD53F367049845BC99AC72306C23D1044A02027E0D@CNBJMBX05.corpusers.net> <35FD53F367049845BC99AC72306C23D1044A02027E0E@CNBJMBX05.corpusers.net> <20150203025925.d1c95fb8.akpm@linux-foundation.org> <35FD53F367049845BC99AC72306C23D1044A02027E12@CNBJMBX05.corpusers.net> X-Hashcash: 1:20:150209:linux-kernel@vger.kernel.org::ow+knP0eMYRMaI20:00000000000000000000000000000000004Gp X-Hashcash: 1:20:150209:linux@arm.linux.org.uk::mBXyVTHa101e/vuV:0000000000000000000000000000000000000000eBG X-Hashcash: 1:20:150209:linux-arm-kernel@lists.infradead.org::YpvGnUu7CP+gi+7p:00000000000000000000000000hYn X-Hashcash: 1:20:150209:yalin.wang@sonymobile.com::f7jQ6hGBllgI2pZX:0000000000000000000000000000000000001rqx X-Hashcash: 1:20:150209:arnd@arndb.de::R6pH/vOX0VNiZdrF:000030tI X-Hashcash: 1:20:150209:linux-arch@vger.kernel.org::6JYY8ZBlYi8wFLSm:000000000000000000000000000000000002uCk X-Hashcash: 1:20:150209:akpm@linux-foundation.org::WFYBf4Ziwbd9NsWW:0000000000000000000000000000000000003f9H X-Hashcash: 1:20:150209:kirill@shutemov.name::7MAHxCgPDUB9+Mo7:0000000000000000000000000000000000000000054md Date: Mon, 09 Feb 2015 22:42:18 +0100 In-Reply-To: <35FD53F367049845BC99AC72306C23D1044A02027E12@CNBJMBX05.corpusers.net> (Yalin Wang's message of "Mon, 9 Feb 2015 16:18:10 +0800") Message-ID: <87386evk1x.fsf@rasmusvillemoes.dk> User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/24.3 (gnu/linux) MIME-Version: 1.0 Content-Type: text/plain Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 2102 Lines: 50 On Mon, Feb 09 2015, "Wang, Yalin" wrote: > I te-test the patch on 3.10 kernel. > The result like this: > > VmallocChunk: 251498164 kB > __set_bit_miss_count:11730 __set_bit_success_count:1036316 > __clear_bit_miss_count:209640 __clear_bit_success_count:4806556 > __test_and_set_bit_miss_count:0 __test_and_set_bit_success_count:121 > __test_and_clear_bit_miss_count:0 __test_and_clear_bit_success_count:445 > > __clear_bit miss rate is a little high, > I check the log, and most miss coming from this code: > > <6>[ 442.701798] [] warn_slowpath_fmt+0x4c/0x58 > <6>[ 442.701805] [] __clear_bit+0x98/0xa4 > <6>[ 442.701813] [] __alloc_fd+0xc8/0x124 > <6>[ 442.701821] [] get_unused_fd_flags+0x28/0x34 > <6>[ 442.701828] [] do_sys_open+0x10c/0x1c0 > <6>[ 442.701835] [] SyS_openat+0xc/0x18 > In __clear_close_on_exec(fd, fdt); > > > > <6>[ 442.695354] [] warn_slowpath_fmt+0x4c/0x58 > <6>[ 442.695359] [] __clear_bit+0x98/0xa4 > <6>[ 442.695367] [] dup_fd+0x1d4/0x280 > <6>[ 442.695375] [] copy_process.part.56+0x42c/0xe38 > <6>[ 442.695382] [] do_fork+0xe0/0x360 > <6>[ 442.695389] [] SyS_clone+0x10/0x1c > In __clear_open_fd(open_files - i, new_fdt); > > Do we need test_bit() before clear_bit()at these 2 place? > In the second case, new_fdt->open_fds has just been filled by a memcpy, and no-one can possibly have written to that cache line in the meantime. In the first case, testing is also likely wasteful if fdt->max_fds is less than half the number of bits in a cacheline (fdt->close_on_exec and fdt->open_fds are always contiguous, and the latter is unconditionally written to). Rasmus -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/