From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753220AbeDPLQV (ORCPT ); Mon, 16 Apr 2018 07:16:21 -0400 Received: from mail-pf0-f172.google.com ([209.85.192.172]:36831 "EHLO mail-pf0-f172.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751099AbeDPLQT (ORCPT ); Mon, 16 Apr 2018 07:16:19 -0400 X-Google-Smtp-Source: AIpwx4+yuWQe6SlfUfpUlUv3ymqmtRRnzSHDFFNWPNvv3vljZqZvoOcazy29S9mP2EpKX0H8MJFwoZV85kTuK3sOv20= MIME-Version: 1.0 In-Reply-To: References: <20180323201634.GA16350@gmail.com> From: Dmitry Vyukov Date: Mon, 16 Apr 2018 13:15:58 +0200 Message-ID: Subject: Re: syzbot dashboard To: Linus Torvalds Cc: Eric Biggers , LKML , syzkaller Content-Type: text/plain; charset="UTF-8" Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Sat, Apr 14, 2018 at 8:25 PM, Linus Torvalds wrote: > Coming back to this now that the merge window is almost over ] > > On Mon, Mar 26, 2018 at 1:46 AM, Dmitry Vyukov wrote: >> >> I've switched emails to links instead of attachments, here are few >> recent examples: >> https://lkml.org/lkml/2018/3/25/31 >> https://lkml.org/lkml/2018/3/25/256 >> https://lkml.org/lkml/2018/3/25/257 > > Looks good to me. I notice that only the last one got any replies, though. > > I wonder if some people auto-ignore the new reports because of having > been burned by the previous "huge illegible emails" issue. > > I do see syzbot fixes in rdma, though, just not for that > cma_listen_on_all issue. So maybe that bug is nastier. "possible deadlock in handle_rx" is now marked as fixed with "vhost_net: add missing lock nesting notation". "KASAN: use-after-free Read in pppol2tp_connect (3)" is marked as fixed with "l2tp: fix races in tunnel creation". "KASAN: use-after-free Read in __list_add_valid (5)" is still open. Though, there are still 182 open bugs on dashboard: https://syzkaller.appspot.com/ 80+ have reproducers and happened last time recently enough. Few dozens without reproducers, but as history shows ~2/3 of them are fixable. Part of the remaining ones is probably fixed, but nobody told syzbot about the fixes. Part is probably still unfixed but happen episodically.