From mboxrd@z Thu Jan 1 00:00:00 1970 From: Matthew Garrett Subject: Re: [PATCH] Fix efifb initialisation when the only GOP device implements ConOut. Date: Mon, 7 Jan 2013 18:43:20 +0000 Message-ID: <20130107184320.GA1400@srcf.ucam.org> References: <1357431222.9270.73.camel@shinybook.infradead.org> <1357524250.9270.86.camel@shinybook.infradead.org> <1357578923.8203.72.camel@mfleming-mobl1.ger.corp.intel.com> <1357581080.2844.17.camel@shinybook.infradead.org> <20130107175846.GA385@srcf.ucam.org> <1357583795.2844.18.camel@shinybook.infradead.org> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: Content-Disposition: inline In-Reply-To: <1357583795.2844.18.camel-Fexsq3y4057IgHVZqg5X0TlWvGAXklZc@public.gmane.org> Sender: linux-efi-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org To: David Woodhouse Cc: Matt Fleming , x86-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org, "H. Peter Anvin" , linux-efi List-Id: linux-efi@vger.kernel.org On Mon, Jan 07, 2013 at 06:36:35PM +0000, David Woodhouse wrote: > On Mon, 2013-01-07 at 17:58 +0000, Matthew Garrett wrote: > > Ignore this case? You can't run 32-bit kernels on 64-bit EFI, or=20 > > vice-versa. You'll die on the first call to UEFI services. >=20 > The kernel has code to avoid calling UEFI services in that case. Even= if > we don't want to support that (any more?), we should at least fail > gracefully rather than just jumping to the wrong entry point and havi= ng > RandomShit=E2=84=A2 happening. The kernel does, the EFI boot stub doesn't. There's probably an argumen= t=20 for having it fail in that case, but it wouldn't even be able to print=20 an error message without some gymnastics. --=20 Matthew Garrett | mjg59-1xO5oi07KQx4cg9Nei1l7Q@public.gmane.org