From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752218Ab1LLKBp (ORCPT ); Mon, 12 Dec 2011 05:01:45 -0500 Received: from arkanian.console-pimps.org ([212.110.184.194]:35964 "EHLO arkanian.console-pimps.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752028Ab1LLKBm (ORCPT ); Mon, 12 Dec 2011 05:01:42 -0500 Subject: Re: [PATCH] Revert "x86, efi: Calling __pa() with an ioremap()ed address is invalid" From: Matt Fleming To: Linus Torvalds Cc: Keith Packard , linux-kernel@vger.kernel.org, "H. Peter Anvin" , Matthew Garrett , Zhang Rui , Huang Ying , Andrew Morton , Ingo Molnar In-Reply-To: References: <1323648762-2148-1-git-send-email-keithp@keithp.com> Content-Type: text/plain; charset="UTF-8" Date: Mon, 12 Dec 2011 10:01:26 +0000 Message-ID: <1323684086.3669.16.camel@mfleming-mobl1.ger.corp.intel.com> Mime-Version: 1.0 X-Mailer: Evolution 2.32.3 (2.32.3-1.fc14) Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Sun, 2011-12-11 at 18:59 -0800, Linus Torvalds wrote: > That said, if you've verified that if fails for you, I guess we should > revert regardless. Matt, any suggestions for tests to do? A dmesg from Keith of a working kernel would be good since the most interesting things are the EFI memory map and the E820 memory map. At this late -rc stage I agree it makes sense to revert this. The CONFIG_X86_32 oops that the patch fixes has been around for years and no one seems to have encountered it (until now), whereas this new regression is likely to affect many more users. Looks like there are some macbook memory map funnies that still need to be addressed... -- Matt Fleming, Intel Open Source Technology Center