From patchwork Fri Oct 20 13:23:52 2017 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: "Burakov, Anatoly" X-Patchwork-Id: 30644 Return-Path: X-Original-To: patchwork@dpdk.org Delivered-To: patchwork@dpdk.org Received: from [92.243.14.124] (localhost [127.0.0.1]) by dpdk.org (Postfix) with ESMTP id C420A1B247; Fri, 20 Oct 2017 15:23:57 +0200 (CEST) Received: from mga06.intel.com (mga06.intel.com [134.134.136.31]) by dpdk.org (Postfix) with ESMTP id CC41F1B246 for ; Fri, 20 Oct 2017 15:23:55 +0200 (CEST) Received: from fmsmga005.fm.intel.com ([10.253.24.32]) by orsmga104.jf.intel.com with ESMTP; 20 Oct 2017 06:23:54 -0700 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.43,405,1503385200"; d="scan'208";a="164989892" Received: from irvmail001.ir.intel.com ([163.33.26.43]) by fmsmga005.fm.intel.com with ESMTP; 20 Oct 2017 06:23:53 -0700 Received: from sivswdev02.ir.intel.com (sivswdev02.ir.intel.com [10.237.217.46]) by irvmail001.ir.intel.com (8.14.3/8.13.6/MailSET/Hub) with ESMTP id v9KDNqdI005576; Fri, 20 Oct 2017 14:23:52 +0100 Received: from sivswdev02.ir.intel.com (localhost [127.0.0.1]) by sivswdev02.ir.intel.com with ESMTP id v9KDNqFL005482; Fri, 20 Oct 2017 14:23:52 +0100 Received: (from aburakov@localhost) by sivswdev02.ir.intel.com with LOCAL id v9KDNqNX005478; Fri, 20 Oct 2017 14:23:52 +0100 From: Anatoly Burakov To: dev@dpdk.org Cc: alejandro.lucero@netronome.com Date: Fri, 20 Oct 2017 14:23:52 +0100 Message-Id: <1508505832-5373-1-git-send-email-anatoly.burakov@intel.com> X-Mailer: git-send-email 1.7.0.7 Subject: [dpdk-dev] [PATCH] vfio: fix secondary process initialization X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: DPDK patches and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dev-bounces@dpdk.org Sender: "dev" When getting group fd from primary process, secondary wasn't storing the fd anywhere, leading to a (harmless) error message in EAL logs, and (not so harmless) potential problems when hot-unplugging devices managed by VFIO in a secondary process. Fix it by actually storing the group fd whenever we get a valid one from the secondary process. Fixes: 94c0776b1bad ("vfio: support hotplug") Cc: alejandro.lucero@netronome.com Signed-off-by: Anatoly Burakov --- lib/librte_eal/linuxapp/eal/eal_vfio.c | 15 +++++++++------ 1 file changed, 9 insertions(+), 6 deletions(-) diff --git a/lib/librte_eal/linuxapp/eal/eal_vfio.c b/lib/librte_eal/linuxapp/eal/eal_vfio.c index 9d66247..9def446 100644 --- a/lib/librte_eal/linuxapp/eal/eal_vfio.c +++ b/lib/librte_eal/linuxapp/eal/eal_vfio.c @@ -68,8 +68,8 @@ vfio_get_group_fd(int iommu_group_no) { int i; int vfio_group_fd; - int group_idx = -1; char filename[PATH_MAX]; + struct vfio_group *cur_grp; /* check if we already have the group descriptor open */ for (i = 0; i < VFIO_MAX_GROUPS; i++) @@ -85,12 +85,12 @@ vfio_get_group_fd(int iommu_group_no) /* Now lets get an index for the new group */ for (i = 0; i < VFIO_MAX_GROUPS; i++) if (vfio_cfg.vfio_groups[i].group_no == -1) { - group_idx = i; + cur_grp = &vfio_cfg.vfio_groups[i]; break; } /* This should not happen */ - if (group_idx == -1) { + if (i == VFIO_MAX_GROUPS) { RTE_LOG(ERR, EAL, "No VFIO group free slot found\n"); return -1; } @@ -123,8 +123,8 @@ vfio_get_group_fd(int iommu_group_no) /* noiommu group found */ } - vfio_cfg.vfio_groups[group_idx].group_no = iommu_group_no; - vfio_cfg.vfio_groups[group_idx].fd = vfio_group_fd; + cur_grp->group_no = iommu_group_no; + cur_grp->fd = vfio_group_fd; vfio_cfg.vfio_active_groups++; return vfio_group_fd; } @@ -157,9 +157,12 @@ vfio_get_group_fd(int iommu_group_no) return 0; case SOCKET_OK: vfio_group_fd = vfio_mp_sync_receive_fd(socket_fd); - /* if we got the fd, return it */ + /* if we got the fd, store it and return it */ if (vfio_group_fd > 0) { close(socket_fd); + cur_grp->group_no = iommu_group_no; + cur_grp->fd = vfio_group_fd; + vfio_cfg.vfio_active_groups++; return vfio_group_fd; } /* fall-through on error */