From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from relay1.mentorg.com ([192.94.38.131]) by linuxtogo.org with esmtp (Exim 4.72) (envelope-from ) id 1Qmp3t-0004kk-6q for openembedded-core@lists.openembedded.org; Fri, 29 Jul 2011 17:33:17 +0200 Received: from svr-orw-fem-01.mgc.mentorg.com ([147.34.98.93]) by relay1.mentorg.com with esmtp id 1Qmozl-00077f-Cs from Tom_Rini@mentor.com for openembedded-core@lists.openembedded.org; Fri, 29 Jul 2011 08:29:01 -0700 Received: from SVR-ORW-FEM-02.mgc.mentorg.com ([147.34.96.206]) by svr-orw-fem-01.mgc.mentorg.com over TLS secured channel with Microsoft SMTPSVC(6.0.3790.4675); Fri, 29 Jul 2011 08:29:01 -0700 Received: from [172.30.80.87] (147.34.91.1) by svr-orw-fem-02.mgc.mentorg.com (147.34.96.168) with Microsoft SMTP Server id 14.1.289.1; Fri, 29 Jul 2011 08:29:00 -0700 Message-ID: <4E32D1B3.4010707@mentor.com> Date: Fri, 29 Jul 2011 08:28:51 -0700 From: Tom Rini Organization: Mentor Graphics Corporation User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.9.2.18) Gecko/20110617 Lightning/1.0b2 Thunderbird/3.1.11 MIME-Version: 1.0 To: References: <2241650b6cb4e409a5463d38cfa0274eec3a3ee2.1311836896.git.lei.mei@intel.com> <4E317B82.1020907@mentor.com> <749B9D3DBF0F054390025D9EAFF47F2212D0F6D3D5@shsmsx501.ccr.corp.intel.com> <4E321053.1050305@mentor.com> <749B9D3DBF0F054390025D9EAFF47F2212D0FCC70E@shsmsx501.ccr.corp.intel.com> In-Reply-To: <749B9D3DBF0F054390025D9EAFF47F2212D0FCC70E@shsmsx501.ccr.corp.intel.com> X-Enigmail-Version: 1.1.1 X-OriginalArrivalTime: 29 Jul 2011 15:29:01.0217 (UTC) FILETIME=[3DE53910:01CC4E04] Subject: Re: [PATCH 1/1] python-native: Fix a compiler finding issue X-BeenThere: openembedded-core@lists.openembedded.org X-Mailman-Version: 2.1.11 Precedence: list Reply-To: Patches and discussions about the oe-core layer List-Id: Patches and discussions about the oe-core layer List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 29 Jul 2011 15:33:17 -0000 Content-Type: text/plain; charset="ISO-8859-1" Content-Transfer-Encoding: 7bit On 07/28/2011 11:58 PM, Mei, Lei wrote: > > >> -----Original Message----- >> From: openembedded-core-bounces@lists.openembedded.org >> [mailto:openembedded-core-bounces@lists.openembedded.org] On Behalf Of >> Tom Rini >> Sent: Friday, July 29, 2011 9:44 AM >> To: openembedded-core@lists.openembedded.org >> Subject: Re: [OE-core] [PATCH 1/1] python-native: Fix a compiler finding issue >> >> On 07/28/2011 06:15 PM, Mei, Lei wrote: >>> >>> >>>> -----Original Message----- >>>> From: openembedded-core-bounces@lists.openembedded.org >>>> [mailto:openembedded-core-bounces@lists.openembedded.org] On Behalf >> Of >>>> Tom Rini >>>> Sent: Thursday, July 28, 2011 11:09 PM >>>> To: openembedded-core@lists.openembedded.org >>>> Subject: Re: [OE-core] [PATCH 1/1] python-native: Fix a compiler finding >> issue >>>> >>>> On 07/28/2011 12:20 AM, Mei Lei wrote: >>>>> The CC variable sometimes add option information after compiler name, >> but >>>> python can't get the real compiler name if those information added. >>>>> Fix this issue by dropping the option information when finding compiler >> name. >>>>> >>>>> Signed-off-by: Mei Lei >>>> >>>> I think this is going to cause problems when you must pass flags to gcc >>>> to have it work, eg 'gcc -m64'. >>> >>> This patch fixed your worried issue. >>> The CC variable, sometimes like: "x86_64-poky-linux-gcc -m64 >> --sysroot=/${TMPDIR}/sysroots/qemux86-64", contains flags information. >>> This will lead to wrong compiler name "qemux86-64" rather than >> "x86_64-poky-linux-gcc" when python finding the compiler name, so add this >> patch to find the real gcc name. >> >> No, what I'm saying is I have a compiler that must be invoked as 'gcc >> -m64' (which is what BUILD_CC is). So, I think after saying that, the >> right answer is to modify python to read the OE-specific BUILD_CC variable. > > > I think I didn't describe this patch exactly before. > > This patch is only for function runtime_library_dir_option, this function is to detect which platform we are running and what compiler we used, then decide what option information should be passed to compiler. > > By default, our cross-compiler's name be recognized as "qemux86" rather than " x86_64-poky-linux-gcc" in this function, this is wrong, this will induce a wrong option information passed to x86_64-poky-linux-gcc and block the compile process. > > And function runtime_library_dir_option only return the option information, so didn't influence compiler name in global. > > By the way, I think BUILD_CC is host compiler name, not for target. You're patching python-native, not python, which means the host python and not the target python. -- Tom Rini Mentor Graphics Corporation