From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-qk0-f199.google.com (mail-qk0-f199.google.com [209.85.220.199]) by kanga.kvack.org (Postfix) with ESMTP id 2D5476B0007 for ; Fri, 27 Apr 2018 06:20:24 -0400 (EDT) Received: by mail-qk0-f199.google.com with SMTP id a124so989805qkb.19 for ; Fri, 27 Apr 2018 03:20:24 -0700 (PDT) Received: from mx1.redhat.com (mx3-rdu2.redhat.com. [66.187.233.73]) by mx.google.com with ESMTPS id p12-v6si332685qvl.241.2018.04.27.03.20.23 for (version=TLS1_2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128/128); Fri, 27 Apr 2018 03:20:23 -0700 (PDT) Date: Fri, 27 Apr 2018 06:20:16 -0400 (EDT) From: Mikulas Patocka Subject: Re: [dm-devel] [PATCH v5] fault-injection: introduce kvmalloc fallback options In-Reply-To: <20180427082555.GC17484@dhcp22.suse.cz> Message-ID: References: <1524694663.4100.21.camel@HansenPartnership.com> <1524697697.4100.23.camel@HansenPartnership.com> <23266.8532.619051.784274@quad.stoffel.home> <20180427005213-mutt-send-email-mst@kernel.org> <20180427082555.GC17484@dhcp22.suse.cz> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-linux-mm@kvack.org List-ID: To: Michal Hocko Cc: "Michael S. Tsirkin" , John Stoffel , James Bottomley , Michal@stoffel.org, eric.dumazet@gmail.com, netdev@vger.kernel.org, jasowang@redhat.com, Randy Dunlap , linux-kernel@vger.kernel.org, Matthew Wilcox , linux-mm@kvack.org, dm-devel@redhat.com, Vlastimil Babka , Andrew@stoffel.org, David Rientjes , Morton , virtualization@lists.linux-foundation.org, David Miller , edumazet@google.com On Fri, 27 Apr 2018, Michal Hocko wrote: > On Thu 26-04-18 18:52:05, Mikulas Patocka wrote: > > > > > > On Fri, 27 Apr 2018, Michael S. Tsirkin wrote: > [...] > > > But assuming it's important to control this kind of > > > fault injection to be controlled from > > > a dedicated menuconfig option, why not the rest of > > > faults? > > > > The injected faults cause damage to the user, so there's no point to > > enable them by default. vmalloc fallback should not cause any damage > > (assuming that the code is correctly written). > > But you want to find those bugs which would BUG_ON easier, so there is a > risk of harm IIUC Yes, I want to harm them, but I only want to harm the users using the debugging kernel. Testers should be "harmed" by crashes - so that the users of production kernels are harmed less. If someone hits this, he should report it, use the kernel parameter to turn it off and continue with the testing. > and this is not much different than other fault injecting paths. Fault injections causes misbehavior even on completely bug-free code (for example, syscalls randomly returning -ENOMEM). This won't cause misbehavior on bug-free code. Mikulas