From mboxrd@z Thu Jan 1 00:00:00 1970 From: Russ Anderson Date: Thu, 06 Jan 2005 20:50:12 +0000 Subject: Re: [patch] per cpu MCA/INIT fixes. Message-Id: <200501062050.j06KoCL0218437@ben.americas.sgi.com> List-Id: References: <200501060308.j0638M7s194286@ben.americas.sgi.com> In-Reply-To: <200501060308.j0638M7s194286@ben.americas.sgi.com> MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: linux-ia64@vger.kernel.org Tony Luck wrote: > > I just applied and pushed Jack's patch to the test-2.6.11 tree, Russ, can > you fix up your patch to apply cleanly on top of Jack's. Yup. > >> +#ifdef CONFIG_SMP > >> +#include > >> +#endif > > > >Just a nit, header files should always be safe to include > >unconditionally. If they're not, the headers need fixing (barring out of tree > >stuff like kdb of course). > > Agreed ... this looks really ugly ... surely we want to support MCA > on uni-processors as well as SMP! What's the issue that provoked > adding the #ifdef here? The only reason was that the changes are all in per_cpu_init() and that routine is inside #ifdef CONFIG_SMP. I built the patch with all configs in arch/ia64/configs, but now see that they all have CONFIG_SMP=y. (sigh) Given all the infrastructure setup done under #ifdef CONFIG_SMP, would it work with uni-processor? I'll start digging through the code... -- Russ Anderson, OS RAS/Partitioning Project Lead SGI - Silicon Graphics Inc rja@sgi.com