From mboxrd@z Thu Jan 1 00:00:00 1970 From: Borislav Petkov Subject: Re: [PATCH v6 1/2] acpi: apei: Rename ghes_severity() to ghes_cper_severity() Date: Tue, 22 May 2018 16:54:26 +0200 Message-ID: <20180522145426.GG5512@pd.tnic> References: <20180521135003.32459-1-mr.nuke.me@gmail.com> <20180521135003.32459-2-mr.nuke.me@gmail.com> <53d0ba88-6929-a7cf-6c3e-4ca389f7249a@gmail.com> <20180522135015.GF5512@pd.tnic> <0b758a1c-90e3-6f76-4f83-1e22c8fc9cd6@gmail.com> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Return-path: Content-Disposition: inline In-Reply-To: <0b758a1c-90e3-6f76-4f83-1e22c8fc9cd6@gmail.com> Sender: linux-kernel-owner@vger.kernel.org To: "Alex G." Cc: "Rafael J. Wysocki" , alex_gagniuc@dellteam.com, austin_bolen@dell.com, shyam_iyer@dell.com, "Rafael J. Wysocki" , Len Brown , Tony Luck , Tyler Baicar , Will Deacon , James Morse , Shiju Jose , "Jonathan (Zhixiong) Zhang" , Dongjiu Geng , ACPI Devel Maling List , Linux Kernel Mailing List List-Id: linux-acpi@vger.kernel.org On Tue, May 22, 2018 at 09:39:15AM -0500, Alex G. wrote: > No, the problem is with the current approach, not with mine. The problem > is trying to handle the error outside of the existing handler. That's a > no-no, IMO. Let me save you some time: until you come up with a proper solution for *all* PCIe errors so that the kernel can correctly decide what to do for each error based on its actual severity, consider this NAKed. I don't care about outside or inside of the handler - this thing needs to be done properly and not just to serve your particular use case of abrupt removal of devices causing PCIe errors, and punish the rest. I especially don't want to have the case where a PCIe error is *really* fatal and then we noodle in some handlers debating about the severity because it got marked as recoverable intermittently and end up causing data corruption on the storage device. Here's a real no-no for ya. -- Regards/Gruss, Boris. Good mailing practices for 400: avoid top-posting and trim the reply.