From mboxrd@z Thu Jan 1 00:00:00 1970 From: Roy Franz Subject: Re: [PATCH V2 05/12] replace split_value() with truncate_string() Date: Mon, 18 Aug 2014 16:38:03 -0700 Message-ID: References: <1405989815-25236-1-git-send-email-roy.franz@linaro.org> <1405989815-25236-6-git-send-email-roy.franz@linaro.org> <53D0CFAD02000078000255F2@mail.emea.novell.com> <53E337A30200007800029FF2@mail.emea.novell.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <53E337A30200007800029FF2@mail.emea.novell.com> List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xen.org Errors-To: xen-devel-bounces@lists.xen.org To: Jan Beulich Cc: keir , Ian Campbell , tim , xen-devel , Stefano Stabellini , linaro-uefi , Fu Wei List-Id: xen-devel@lists.xenproject.org On Wed, Aug 6, 2014 at 11:24 PM, Jan Beulich wrote: >>>> On 07.08.14 at 00:37, wrote: >> On Thu, Jul 24, 2014 at 12:19 AM, Jan Beulich wrote: >>> Furthermore splitting out the place_string() doesn't seem very >>> efficient, as imo the goal ought to be for efi_start() to become >>> common code (or at least the module loading part of fit), i.e. >>> there's no win at all from the change you're doing here. >> >> I don't think that combining the x86 and arm efi_start() will work out >> that cleanly. Arm is using device tree from getting information from >> GRUB and/or the firmware, so I think you'd end up with a lot of conditional >> code. > > But that is precisely what you'd add (arch-specific) calls out of the > function for, in the extreme case doing nothing on x86. (And that is > also specifically why I'd favor the approach outlined in the earlier > reply to patch 1.) > > Jan > Hi Jan, I've spent a little time prototyping what you are suggesting, and I think I should be able to have a patchset in reasonable shape in time for the 4.5 freeze. This will require more extensive changes to the x86 code than my previous patchset, but most of it appears to be benign re-ordering of various bits of code to bring blocks of arch specific code together. I don't have an EFI based x86 system to test on, which makes me a bit nervous about this. I will be at LinuxCon this week (arriving Tuesday evening, through Sat.), and it would be great if we could find some time to chat about this in person. Thanks, Roy