From mboxrd@z Thu Jan 1 00:00:00 1970 From: Thomas Meyer Subject: Re: [PATCH 1/2] efi/x86: Move efi stub globals from .bss to .data Date: Sat, 11 Apr 2020 10:50:01 +0200 Message-ID: <20200411085001.GA28742@localhost.localdomain> References: <20200406180614.429454-1-nivedita@alum.mit.edu> <20200410112605.GA3344@localhost.localdomain> <20200410143821.GB936997@rani.riverdale.lan> Mime-Version: 1.0 Return-path: Content-Disposition: inline In-Reply-To: <20200410143821.GB936997@rani.riverdale.lan> Sender: linux-kernel-owner@vger.kernel.org List-ID: Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: Arvind Sankar Cc: Ard Biesheuvel , Borislav Petkov , Sergey Shatunov , hpa@zytor.com, Linux Kernel Mailing List , mingo@redhat.com, Thomas Gleixner , x86@kernel.org, linux-efi , initramfs@vger.kernel.org, Donovan Tremura , Harald Hoyer On Fri, Apr 10, 2020 at 10:38:21AM -0400, Arvind Sankar wrote: > On Fri, Apr 10, 2020 at 01:26:05PM +0200, Thomas Meyer wrote: > > On Mon, Apr 06, 2020 at 02:06:13PM -0400, Arvind Sankar wrote: > > > > Hi, > > > > I did write an email to x86@kernel.org, which sadly seems to have no > > mailing list archive, I wonder if my problem has anything to do with the > > patches you are discussing here: > > > > I found this reply, which contains my original email in my inbox: > > > > Subject: Kernel v5.5 doesn't boot on my x86 laptop > > > > Hi, > > > > I'm using an old MacBookPro1,1 to run Fedora 30 (the last one to support > > x86) and a upstream up-to-date kernel, currently 5.4.16. > > > > I'm using sd-boot to boot into an EFI-enabled kernel which contains > > an embedded initram cpio image (because loading the image from kernel's > > EFI stub doesn't seem to work for some unknown reason, I tried to debug > > this but my early debugging foo is too weak). > > > > Kernel 5.4.x works correctly with this setup (but resuming from disk > > seems to have broken in 5.4.x or maybe even earlier - when resuming from > > disk I get all kind of funky OOPSes/errors, but that's another story, hopefully > > 5.5.x was fixed in this regards). > > > > So I did have a look at the commits under arch/x86/boot and "x86/boot: > > Introduce setup_indirect" (b3c72fc9a78e74161f9d05ef7191706060628f8c) did > > talk about "bump setup_header version in arch/x86/boot/header.S", so I > > did revert above commit and I was finally able to boot into v5.5 kernel! > > > > So either sd-boot also needs an upgrade or this commit does break > > something. > > Any help is welcome, don't hesitate to get in contact with me if you > > have any questions. > > > > mfg > > thomas > > > > If it is a problem with 5.5, it would be unrelated to this thread, which > is about problems introduced by patches for the upcoming 5.7. okay, ping me if I should test something on real hardware. > > Thanks.