From mboxrd@z Thu Jan 1 00:00:00 1970 From: Mark Rutland Subject: Re: [PATCH] arm64/efi: remove spurious WARN_ON for !4K kernels Date: Thu, 23 Jun 2016 14:17:38 +0100 Message-ID: <20160623131738.GH8836@leverpostej> References: <1464189116-30898-1-git-send-email-mark.rutland@arm.com> <20160530211212.GG2984@codeblueprint.co.uk> <20160603100040.GJ9915@arm.com> <20160617211601.GN2658@codeblueprint.co.uk> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Content-Disposition: inline In-Reply-To: <20160617211601.GN2658-mF/unelCI9GS6iBeEJttW/XRex20P6io@public.gmane.org> Sender: linux-efi-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org To: Matt Fleming Cc: Will Deacon , linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r@public.gmane.org, Ard Biesheuvel , Catalin Marinas , Jeremy Linton , Leif Lindholm , linux-efi-u79uwXL29TY76Z2rM5mHXA@public.gmane.org List-Id: linux-efi@vger.kernel.org On Fri, Jun 17, 2016 at 10:16:01PM +0100, Matt Fleming wrote: > On Fri, 03 Jun, at 11:00:40AM, Will Deacon wrote: > > > > Mark, Jeremy? > > > > I've kept this patch marked as unread, but it's not clear to me (a) how > > urgent it is (b) what tree it should go through and (c) whether it's the > > right thing to do! > > Ping? Anyone got an update here? Sorry, I'd been meaning to come back to this. As Ard noted, there are serious cases that this detects which we cannot handle, and a WARN_ON for those is appropriate. For the alignment of regions as required for permissions, I believe we need clarification in the spec, and in the mean time I think we should keep the WARN_ON. To accurately report problems with differing attributes within a (kernel) page, we need more comprehensive checks on the EFI memory map, and potentially in some cases where we handle things dynamically at run time. I haven't had the chance to delve into those yet. So I guess for the timebeing this should stay as a WARN_ON, even if in some cases it is spurious. Thanks, Mark. From mboxrd@z Thu Jan 1 00:00:00 1970 From: mark.rutland@arm.com (Mark Rutland) Date: Thu, 23 Jun 2016 14:17:38 +0100 Subject: [PATCH] arm64/efi: remove spurious WARN_ON for !4K kernels In-Reply-To: <20160617211601.GN2658@codeblueprint.co.uk> References: <1464189116-30898-1-git-send-email-mark.rutland@arm.com> <20160530211212.GG2984@codeblueprint.co.uk> <20160603100040.GJ9915@arm.com> <20160617211601.GN2658@codeblueprint.co.uk> Message-ID: <20160623131738.GH8836@leverpostej> To: linux-arm-kernel@lists.infradead.org List-Id: linux-arm-kernel.lists.infradead.org On Fri, Jun 17, 2016 at 10:16:01PM +0100, Matt Fleming wrote: > On Fri, 03 Jun, at 11:00:40AM, Will Deacon wrote: > > > > Mark, Jeremy? > > > > I've kept this patch marked as unread, but it's not clear to me (a) how > > urgent it is (b) what tree it should go through and (c) whether it's the > > right thing to do! > > Ping? Anyone got an update here? Sorry, I'd been meaning to come back to this. As Ard noted, there are serious cases that this detects which we cannot handle, and a WARN_ON for those is appropriate. For the alignment of regions as required for permissions, I believe we need clarification in the spec, and in the mean time I think we should keep the WARN_ON. To accurately report problems with differing attributes within a (kernel) page, we need more comprehensive checks on the EFI memory map, and potentially in some cases where we handle things dynamically at run time. I haven't had the chance to delve into those yet. So I guess for the timebeing this should stay as a WARN_ON, even if in some cases it is spurious. Thanks, Mark.