From mboxrd@z Thu Jan 1 00:00:00 1970 From: Gerd Knorr Subject: Re: PAE support revisited - questions Date: 30 Mar 2005 12:20:51 +0200 Message-ID: <87acolbfjw.fsf@bytesex.org> References: <1111533714.19319.28.camel@kwdevel.austin.ibm.com> <87oed2bpv5.fsf@bytesex.org> <1112142537.21025.60.camel@kwdevel.austin.ibm.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: In-Reply-To: <1112142537.21025.60.camel@kwdevel.austin.ibm.com> List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xensource.com Errors-To: xen-devel-bounces@lists.xensource.com To: Karen White Cc: Ian Pratt , keir.fraser@cl.cam.ac.uk, xen-devel@lists.xensource.com, Nuutti Kotivuori List-Id: xen-devel@lists.xenproject.org Karen White writes: > > Thats my intention. Maybe build some CONFIG_PAGING_LEVELS_2/3/4 from > > this and i386 vs. x86_64 to make the #ifdefs in the code a bit more > > readable. > > I submitted a CONFIG_PAE patch, but Keir says PAE support will be > determined at boot time, not compile time. I think that was just because it was just the config option without it being actually used anywhere. > > I'd create include/asm-x86/x86_32/page-2l.h and page-3l.h and let > > page.h include the correct one depending on CONFIG_X86_PAE (and also > > have the common stuff there). > > Are you working on this? I have modeled a page.h after x86_32/page.h > but for level-3 page tables. Not finished with it. I have a old patch for that, it will not apply any more due to the page table macro cleanups though. Once the page table macro stuff is finished I want to look at this again. Gerd -- #define printk(args...) fprintf(stderr, ## args)