From owner-svn-src-all@FreeBSD.ORG Tue Aug 21 08:40:01 2012 Return-Path: Delivered-To: svn-src-all@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [69.147.83.52]) by hub.freebsd.org (Postfix) with ESMTP id 25096106564A; Tue, 21 Aug 2012 08:40:01 +0000 (UTC) (envelope-from dim@FreeBSD.org) Received: from tensor.andric.com (cl-327.ede-01.nl.sixxs.net [IPv6:2001:7b8:2ff:146::2]) by mx1.freebsd.org (Postfix) with ESMTP id CE0258FC0A; Tue, 21 Aug 2012 08:40:00 +0000 (UTC) Received: from [IPv6:2001:7b8:3a7:0:dd99:997b:50d0:d751] (unknown [IPv6:2001:7b8:3a7:0:dd99:997b:50d0:d751]) (using TLSv1 with cipher DHE-RSA-CAMELLIA256-SHA (256/256 bits)) (No client certificate requested) by tensor.andric.com (Postfix) with ESMTPSA id BCB885C59; Tue, 21 Aug 2012 10:39:51 +0200 (CEST) Message-ID: <50334957.6020201@FreeBSD.org> Date: Tue, 21 Aug 2012 10:39:51 +0200 From: Dimitry Andric Organization: The FreeBSD Project User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:15.0) Gecko/20120815 Thunderbird/15.0 MIME-Version: 1.0 To: Jan Beich References: <201208201833.q7KIX58p075876@svn.freebsd.org> <1T3c9K-000Jlb-7g@internal.tormail.org> In-Reply-To: <1T3c9K-000Jlb-7g@internal.tormail.org> Content-Type: multipart/mixed; boundary="------------010001080807070805040402" Cc: svn-src-head@freebsd.org, svn-src-all@freebsd.org, src-committers@freebsd.org, Jason Evans Subject: jemalloc and clang (was: Re: svn commit: r239462 - in head) X-BeenThere: svn-src-all@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: "SVN commit messages for the entire src tree \(except for " user" and " projects" \)" List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 21 Aug 2012 08:40:01 -0000 This is a multi-part message in MIME format. --------------010001080807070805040402 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit On 2012-08-21 02:17, Jan Beich wrote: ... > Time to revert r228540? > > Index: contrib/jemalloc/include/jemalloc/jemalloc_FreeBSD.h > =================================================================== > --- contrib/jemalloc/include/jemalloc/jemalloc_FreeBSD.h (revision 239467) > +++ contrib/jemalloc/include/jemalloc/jemalloc_FreeBSD.h (working copy) > @@ -56,10 +56,6 @@ > #ifndef JEMALLOC_TLS_MODEL > # define JEMALLOC_TLS_MODEL /* Default. */ > #endif > -#ifdef __clang__ > -# undef JEMALLOC_TLS_MODEL > -# define JEMALLOC_TLS_MODEL /* clang does not support tls_model yet. */ > -#endif > > #define STATIC_PAGE_SHIFT PAGE_SHIFT > #define LG_SIZEOF_INT 2 Well, if Jason would like to support upstream jemalloc for different versions of clang, it is probably better to to use __has_feature() instead. Jason, what do you think of the attached patch? Or could we just remove the whole #ifdef __clang__ part? --------------010001080807070805040402 Content-Type: text/x-diff; name="jemalloc-clang-tlsmodel-1.diff" Content-Transfer-Encoding: 7bit Content-Disposition: attachment; filename="jemalloc-clang-tlsmodel-1.diff" Index: contrib/jemalloc/include/jemalloc/jemalloc_FreeBSD.h =================================================================== --- contrib/jemalloc/include/jemalloc/jemalloc_FreeBSD.h (revision 239463) +++ contrib/jemalloc/include/jemalloc/jemalloc_FreeBSD.h (working copy) @@ -56,9 +56,9 @@ #ifndef JEMALLOC_TLS_MODEL # define JEMALLOC_TLS_MODEL /* Default. */ #endif -#ifdef __clang__ +#if defined(__clang__) && !__has_feature(tls_model) # undef JEMALLOC_TLS_MODEL -# define JEMALLOC_TLS_MODEL /* clang does not support tls_model yet. */ +# define JEMALLOC_TLS_MODEL /* This clang does not support tls_model. */ #endif #define STATIC_PAGE_SHIFT PAGE_SHIFT --------------010001080807070805040402--