From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from ozlabs.org (ozlabs.org [IPv6:2401:3900:2:1::2]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by lists.ozlabs.org (Postfix) with ESMTPS id 3rVX3Q5J8kzDqld for ; Thu, 16 Jun 2016 15:28:18 +1000 (AEST) Received: from mail-pa0-x229.google.com (mail-pa0-x229.google.com [IPv6:2607:f8b0:400e:c03::229]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ozlabs.org (Postfix) with ESMTPS id 3rVX3P5CnSz9t0v for ; Thu, 16 Jun 2016 15:28:17 +1000 (AEST) Received: by mail-pa0-x229.google.com with SMTP id b5so14544508pas.3 for ; Wed, 15 Jun 2016 22:28:16 -0700 (PDT) Subject: Re: kernel bug in "Drop WIMG in favour of new constants"? To: "Darrick J. Wong" , "Aneesh Kumar K.V" References: <20160616043340.GB22590@birch.djwong.org> Cc: linuxppc-dev@ozlabs.org From: Balbir Singh Message-ID: Date: Thu, 16 Jun 2016 15:28:10 +1000 MIME-Version: 1.0 In-Reply-To: <20160616043340.GB22590@birch.djwong.org> Content-Type: text/plain; charset=utf-8 List-Id: Linux on PowerPC Developers Mail List List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , On 16/06/16 14:33, Darrick J. Wong wrote: > Hi Aneesh, > > I noticed when trying out 4.7-rc3 on qemu-2.5 that the kernel no longer > boots. 4.6 booted just fine, so I bisected the kernel to the commit > 30bda41aba4efb2370c97e2cbe7385de93ccc372, which is "powerpc/mm: Drop WIMG in > favour of new constants". The changelog suggests that the KVM changes need > closer review, and here's an actual crash: > > (I can send libvirt's machine xml, .config, and full dmesg if that helps.) > > --Darrick Looking at the address - 0xd000080080000010, looks like map_kernel_page followed by a fault here did not do the right thing. This EA range belongs to the io_remap range, IIUC. The only suspicious bits I could find are + /* + * if host is mapped cache inhibited, make sure hptel also have + * cache inhibited. + */ + if (wimg & HPTE_R_W) /* FIXME!! is this ok for all guest. ? */ + return false; Aneesh also posted some changes to hpte_update and hpte_updatepp that might be interesting Balbir