From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jes Sorensen Date: Tue, 31 Jan 2006 10:57:05 +0000 Subject: Re: [patch 5/5] fix sn rw_mmr.h to use intrinsic Message-Id: List-Id: References: <200601270137.k0R1big21236@unix-os.sc.intel.com> In-Reply-To: <200601270137.k0R1big21236@unix-os.sc.intel.com> MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: linux-ia64@vger.kernel.org >>>>> "Ken" = Chen, Kenneth W writes: Ken> Jes Sorensen wrote on Friday, January 27, 2006 1:57 AM >> I'll have to test this. However it really doesn't matter much as >> we won't be building the SN kernel with ICC. Ken> When I build a generic kernel, all sn code are pulled in and need Ken> to be compiled. It would be nice to have sn portion of kernel Ken> code free of asm statement in the source. I heard that gcc is Ken> also going in the direction of implementing intrinsic in the long Ken> term. Or do we have an option of CONFIG_NON_SGI? I can cook up Ken> something if this is really what people Want (I suspect not). Ken, Speaking solely on my own behalf, and not that of my employer here. I have to say that I have always found it very discouraging how the kernel had this mess added which is badly obfuscating the code solely to support one vendor's broken compiler. The real solution to this problem would be for Intel's compiler people to fix the brokenness of ICC instead of adding any more of this obfuscation to the kernel ;-( Regards, Jes