From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1161088AbaJ3Rlo (ORCPT ); Thu, 30 Oct 2014 13:41:44 -0400 Received: from mail.skyhub.de ([78.46.96.112]:54495 "EHLO mail.skyhub.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1760674AbaJ3Rll (ORCPT ); Thu, 30 Oct 2014 13:41:41 -0400 Date: Thu, 30 Oct 2014 18:41:34 +0100 From: Borislav Petkov To: Henrique de Moraes Holschuh Cc: linux-kernel@vger.kernel.org, H Peter Anvin Subject: Re: [PATCH 4/8] x86, microcode, intel: add error logging to early update driver Message-ID: <20141030174134.GE11178@pd.tnic> References: <1410197875-19252-1-git-send-email-hmh@hmh.eng.br> <1410197875-19252-5-git-send-email-hmh@hmh.eng.br> <20141020150801.GE3524@pd.tnic> <20141021141015.GD22528@khazad-dum.debian.net> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Disposition: inline In-Reply-To: <20141021141015.GD22528@khazad-dum.debian.net> User-Agent: Mutt/1.5.23 (2014-03-12) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Tue, Oct 21, 2014 at 12:10:15PM -0200, Henrique de Moraes Holschuh wrote: > In fact, I have a patch somewhere that needs to add a new failure message: > we have several failure cases which we want to differentiate, at the very > least "processor didn't like it" and "it looks corrupt, so we didn't even > try to install it". Actually, I don't want to be too chatty with the loader: if the microcode blob passes checks but it is not for the current processor we're running, not saying anything is what I want to do. Why? Because I don't want to disturb people unnecessarily - if the microcode doesn't apply and everything else checks out, you simply don't need it. If one really wants to know, one can always check /proc/cpuinfo and read out the microcode revision from the blob. But that is for the 1% of the curious ones - everyone else should simply install microcode blob and boot machine. Fire and forget. Only the abnormal error cases should be vocal in saying what's wrong so that we can actually address those. -- Regards/Gruss, Boris. Sent from a fat crate under my desk. Formatting is fine. --