From mboxrd@z Thu Jan 1 00:00:00 1970 From: Matt Fleming Subject: Re: [PATCH] x86, boot: Allow 64bit EFI kernel to be loaded above 4G Date: Thu, 12 Feb 2015 14:59:42 +0000 Message-ID: <20150212145942.GE4665@codeblueprint.co.uk> References: <1423015400-12629-1-git-send-email-yinghai@kernel.org> <20150209182742.GQ6461@codeblueprint.co.uk> <20150211155524.GC4665@codeblueprint.co.uk> <20150211162958.GA18062@fenchurch.internal.datastacks.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Content-Disposition: inline In-Reply-To: <20150211162958.GA18062@fenchurch.internal.datastacks.com> Sender: linux-doc-owner@vger.kernel.org To: Peter Jones Cc: Yinghai Lu , Thomas Gleixner , "H. Peter Anvin" , Ingo Molnar , Jonathan Corbet , Kees Cook , Junjie Mao , "linux-doc@vger.kernel.org" , "linux-efi@vger.kernel.org" , Linux Kernel Mailing List , Matthew Garrett List-Id: linux-efi@vger.kernel.org On Wed, 11 Feb, at 11:29:58AM, Peter Jones wrote: > > From grub's point of view I'm not sure why we'd care - the pages kernel > and initramfs land in are both from the Boot Services allocator, so if the > machine doesn't support high addresses, they won't be there. It's not that some implementations don't "support" higher addresses, it's that the EFI_FILE_PROTOCOL is buggy and it corrupts the memory when reading into it; you can allocate it just fine. At least, that's what I remember from the limited investigation I performed. But since grub doesn't use EFI_FILE_PROTOCOL, we should be cool. -- Matt Fleming, Intel Open Source Technology Center