From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from e28smtp02.in.ibm.com ([122.248.162.2]:55924 "EHLO e28smtp02.in.ibm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751947AbbIIFxh (ORCPT ); Wed, 9 Sep 2015 01:53:37 -0400 Received: from /spool/local by e28smtp02.in.ibm.com with IBM ESMTP SMTP Gateway: Authorized Use Only! Violators will be prosecuted for from ; Wed, 9 Sep 2015 11:23:35 +0530 Received: from d28relay02.in.ibm.com (d28relay02.in.ibm.com [9.184.220.59]) by d28dlp02.in.ibm.com (Postfix) with ESMTP id DAF7F394005A for ; Wed, 9 Sep 2015 11:23:32 +0530 (IST) Received: from d28av01.in.ibm.com (d28av01.in.ibm.com [9.184.220.63]) by d28relay02.in.ibm.com (8.14.9/8.14.9/NCO v10.0) with ESMTP id t895rV579371836 for ; Wed, 9 Sep 2015 11:23:32 +0530 Received: from d28av01.in.ibm.com (localhost [127.0.0.1]) by d28av01.in.ibm.com (8.14.4/8.14.4/NCO v10.0 AVout) with ESMTP id t895rUT4020515 for ; Wed, 9 Sep 2015 11:23:31 +0530 Date: Wed, 9 Sep 2015 13:53:29 +0800 From: Richard Yang To: Benjamin Herrenschmidt Cc: Richard Yang , Gavin Shan , aik@ozlabs.ru, linux-pci@vger.kernel.org, bhelgaas@google.com, linuxppc-dev@lists.ozlabs.org Subject: Re: [PATCH V9 11/11] powerpc/powernv: compound PE for VFs Message-ID: <20150909055329.GA7684@Richards-MacBook-Pro.local> Reply-To: Richard Yang References: <1437112961-17275-1-git-send-email-weiyang@linux.vnet.ibm.com> <1437112961-17275-12-git-send-email-weiyang@linux.vnet.ibm.com> <20150729031718.GA18069@richard> <20150909024820.GA6808@gwshan> <20150909033616.GA5148@richards-mbp.cn.ibm.com> <1441774751.2767.20.camel@au1.ibm.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii In-Reply-To: <1441774751.2767.20.camel@au1.ibm.com> Sender: linux-pci-owner@vger.kernel.org List-ID: On Wed, Sep 09, 2015 at 02:59:11PM +1000, Benjamin Herrenschmidt wrote: >On Wed, 2015-09-09 at 11:36 +0800, Richard Yang wrote: >> The VF EEH is hung since we re-designed the SRIOV. After the re >> -design, we >> don't have VF groups. >> >> My plan is to push the VF EEH patch set after the SRIOV Redesign is >> accepted. > >What do you mean taht we don't have VF groups ? > Before we redesign the SRIOV, several VFs may share one M64 segment. This introduced the compound PE for VFs. The VF group in previous mail means the compound PE composed of a master VF PE will have several slave VF PEs. >If we don't have IOMMU groups per VF that means we can't assign them to >KVM partitions -> they are completely useless. Yes, I don't mean for the IOMMU group. > >Ben. -- Richard Yang Help you, Help me