From mboxrd@z Thu Jan 1 00:00:00 1970 From: Keir Fraser Subject: Re: frame_table initialisation Date: Mon, 26 Feb 2007 16:02:12 +0000 Message-ID: References: <45E2F29A.2020601@cea.fr> Mime-Version: 1.0 Content-Type: text/plain; charset="US-ASCII" Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <45E2F29A.2020601@cea.fr> List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xensource.com Errors-To: xen-devel-bounces@lists.xensource.com To: PUCCETTI Armand , xen-devel@lists.xensource.com List-Id: xen-devel@lists.xenproject.org On 26/2/07 14:45, "PUCCETTI Armand" wrote: > 2. some lines before in __start_xen, this variable is referenced by > function map_pages_to_xen -> virt_to_xen_l2 -> alloc_xen_pagetable -> > mfn_to_page. > What should the value of frame_table be there? > > Do I miss some pre-initialisation(s)? The fact that frame_table is not initialised here (and hence will be NULL) doesn't actually matter, since the result of mfn_to_page() is only actually used as a value to page_to_mfn(). And page_to_mfn(mfn_to_page(m))==m for any (constant) value of frame_table. This code could be clearer I'll admit. It probably makes sense for alloc_xen_pagetable to return an MFN rather than a page_info pointer. -- Keir