From nobody Wed May 15 22:12:51 2024 Delivered-To: importer@patchew.org Authentication-Results: mx.zohomail.com; dkim=pass header.i=@intel.com; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass(p=none dis=none) header.from=intel.com ARC-Seal: i=1; a=rsa-sha256; t=1693392901; cv=none; d=zohomail.com; s=zohoarc; b=YjV3yQe3PUOe3eWPTPowUScMuhrva9iUVPH9WcOEDK/3i57rbo7mVglFCh6JM6jPezLmOLLuu5DA3olOd/EEqHNh3WAwm5efNeR48mk0EGg/XYIyXvqb4+4c3NpmCaGXzqEw0/OOUDVnTwxgyZozwV11FElXiblU/ofw3lJuR74= ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=zohomail.com; s=zohoarc; t=1693392901; h=Content-Transfer-Encoding:Cc:Date:From:In-Reply-To:List-Subscribe:List-Post:List-Id:List-Archive:List-Help:List-Unsubscribe:MIME-Version:Message-ID:References:Sender:Subject:To; bh=s+dgSUEl0u+bNhit7+V9lJNZDezDLHf8o6I6jWV0SDw=; b=MflvfcIlFn8ie6qIjCtJYczb2sWiXoeSo3f8vuGFjwLEansfvaPYKBpnA8BIwajk7T+2sDVYHm9j7tChfC4edmK42t6kCnXzXeOvLBYBuzYOi1FIBiSfX8sisMV0rwTt/5oZMQJoBUy0FV+FT9P+JXjYS7zn+3kzHjeMPtZAPEY= ARC-Authentication-Results: i=1; mx.zohomail.com; dkim=pass header.i=@intel.com; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass header.from= (p=none dis=none) Return-Path: Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) by mx.zohomail.com with SMTPS id 1693392901024221.2750524015603; Wed, 30 Aug 2023 03:55:01 -0700 (PDT) Received: from localhost ([::1] helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1qbIou-0004E2-I9; Wed, 30 Aug 2023 06:53:08 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1qbIot-0004DT-4O for qemu-devel@nongnu.org; Wed, 30 Aug 2023 06:53:07 -0400 Received: from mgamail.intel.com ([192.55.52.93]) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1qbIoq-0007ew-Cz for qemu-devel@nongnu.org; Wed, 30 Aug 2023 06:53:06 -0400 Received: from orsmga006.jf.intel.com ([10.7.209.51]) by fmsmga102.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 30 Aug 2023 03:52:33 -0700 Received: from duan-server-s2600bt.bj.intel.com ([10.240.192.147]) by orsmga006-auth.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 30 Aug 2023 03:52:28 -0700 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1693392785; x=1724928785; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=pwt63gHRnDoXZ3T2Mf+OvoWjXHFkcAU7xOVp2wScvx0=; b=SmKV/arrf5u5oKnk+EPs/6X7GYLGJHcJfybkfKRTrb+9of8fe4W++Muu jmqjQrUxieRdJbHmYcYSfexIIyC2OSl/8Hjq6GtXVuy0JD1qVeokLb8Ow TAzJCjtHea4V02QuJcNkcZJvRA4bcXh+KNCXn+GGpCHMNf3fFAw/MAnw3 Z3ySix6IjKJkNJIBJCYHVkyKmvMIT93KyGVMeVNcHXQqSgsyfW/o1Qdvl XUdOEfp9+oxygUWUONprrJDjdTYqV1c+/cCkLtFDB0bjGnLajSdtuPj4I O/W3UyMLrCA367GjJtatVbtFpVbwmNMnR9uc+l4uOucmyY9cUE0UMXyXG Q==; X-IronPort-AV: E=McAfee;i="6600,9927,10817"; a="373016370" X-IronPort-AV: E=Sophos;i="6.02,213,1688454000"; d="scan'208";a="373016370" X-ExtLoop1: 1 X-IronPort-AV: E=McAfee;i="6600,9927,10817"; a="715865946" X-IronPort-AV: E=Sophos;i="6.02,213,1688454000"; d="scan'208";a="715865946" From: Zhenzhong Duan To: qemu-devel@nongnu.org Cc: alex.williamson@redhat.com, clg@redhat.com, jgg@nvidia.com, nicolinc@nvidia.com, joao.m.martins@oracle.com, eric.auger@redhat.com, peterx@redhat.com, jasowang@redhat.com, kevin.tian@intel.com, yi.l.liu@intel.com, yi.y.sun@intel.com, chao.p.peng@intel.com, Zhenzhong Duan , "Michael S. Tsirkin" , Cornelia Huck , Paolo Bonzini Subject: [PATCH v1 01/22] scripts/update-linux-headers: Add iommufd.h Date: Wed, 30 Aug 2023 18:37:33 +0800 Message-Id: <20230830103754.36461-2-zhenzhong.duan@intel.com> X-Mailer: git-send-email 2.34.1 In-Reply-To: <20230830103754.36461-1-zhenzhong.duan@intel.com> References: <20230830103754.36461-1-zhenzhong.duan@intel.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Received-SPF: pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) client-ip=209.51.188.17; envelope-from=qemu-devel-bounces+importer=patchew.org@nongnu.org; helo=lists.gnu.org; Received-SPF: pass client-ip=192.55.52.93; envelope-from=zhenzhong.duan@intel.com; helo=mgamail.intel.com X-Spam_score_int: -43 X-Spam_score: -4.4 X-Spam_bar: ---- X-Spam_report: (-4.4 / 5.0 requ) BAYES_00=-1.9, DKIMWL_WL_HIGH=-0.001, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, RCVD_IN_DNSWL_MED=-2.3, SPF_HELO_NONE=0.001, SPF_PASS=-0.001 autolearn=ham autolearn_force=no X-Spam_action: no action X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: qemu-devel-bounces+importer=patchew.org@nongnu.org Sender: qemu-devel-bounces+importer=patchew.org@nongnu.org X-ZohoMail-DKIM: pass (identity @intel.com) X-ZM-MESSAGEID: 1693392940698100019 Content-Type: text/plain; charset="utf-8" From: Eric Auger Update the script to import iommufd.h Signed-off-by: Eric Auger Signed-off-by: Yi Liu Signed-off-by: Zhenzhong Duan --- scripts/update-linux-headers.sh | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/scripts/update-linux-headers.sh b/scripts/update-linux-headers= .sh index 35a64bb501..34295c0fe5 100755 --- a/scripts/update-linux-headers.sh +++ b/scripts/update-linux-headers.sh @@ -161,7 +161,8 @@ done rm -rf "$output/linux-headers/linux" mkdir -p "$output/linux-headers/linux" for header in const.h stddef.h kvm.h vfio.h vfio_ccw.h vfio_zdev.h vhost.h= \ - psci.h psp-sev.h userfaultfd.h memfd.h mman.h nvme_ioctl.h v= duse.h; do + psci.h psp-sev.h userfaultfd.h memfd.h mman.h nvme_ioctl.h \ + vduse.h iommufd.h; do cp "$tmpdir/include/linux/$header" "$output/linux-headers/linux" done =20 --=20 2.34.1 From nobody Wed May 15 22:12:51 2024 Delivered-To: importer@patchew.org Authentication-Results: mx.zohomail.com; dkim=pass header.i=@intel.com; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass(p=none dis=none) header.from=intel.com ARC-Seal: i=1; a=rsa-sha256; t=1693393001; cv=none; d=zohomail.com; s=zohoarc; b=HN1N/8zRAuGE058V26zAifp5PY5qnVnnSfw6wxPHRrnFFU6AAxxwHZlmMaOhOj8jzZ3uM22QeFhMQWEdevnxVjt3SyuYddIVDGT19GfztS7pAXMh+byBDD8YhBr185oBp6EilziV56lX1yCWYJ/FCQYKIa0aqaDtXALYsVvaKtU= ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=zohomail.com; s=zohoarc; t=1693393001; h=Content-Transfer-Encoding:Cc:Date:From:In-Reply-To:List-Subscribe:List-Post:List-Id:List-Archive:List-Help:List-Unsubscribe:MIME-Version:Message-ID:References:Sender:Subject:To; bh=+LwZfGu54hjDbSIsXpiqdl5NpXtfsTZmyOpyIKokOGM=; b=ibk8mqVH3svmVQTy7N9/quZEV0QEbPsNf+O2CQ2wb3xo4iJITto9OTKF4Q8hOw/4UptgW12Fk9gT7WPu3kl+FDY/Y+bUlYCOCSWfs/9bi+e+gzsHPjt2VDLDdcADFr1Yqjm5tgz685d62G+eXNBK5Glq2MNuiXw+q6Fr/W6moVI= ARC-Authentication-Results: i=1; mx.zohomail.com; dkim=pass header.i=@intel.com; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass header.from= (p=none dis=none) Return-Path: Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) by mx.zohomail.com with SMTPS id 1693393001375612.1960897322654; Wed, 30 Aug 2023 03:56:41 -0700 (PDT) Received: from localhost ([::1] helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1qbIon-0004Cj-6y; Wed, 30 Aug 2023 06:53:01 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1qbIoh-0004CN-G1 for qemu-devel@nongnu.org; Wed, 30 Aug 2023 06:52:55 -0400 Received: from mgamail.intel.com ([192.55.52.93]) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1qbIoc-0007eJ-4x for qemu-devel@nongnu.org; Wed, 30 Aug 2023 06:52:54 -0400 Received: from orsmga006.jf.intel.com ([10.7.209.51]) by fmsmga102.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 30 Aug 2023 03:52:41 -0700 Received: from duan-server-s2600bt.bj.intel.com ([10.240.192.147]) by orsmga006-auth.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 30 Aug 2023 03:52:33 -0700 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1693392770; x=1724928770; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=LKYQcQHOMOvKh9iQVujrZ29uLyWWbA45EUA3lP33byU=; b=CudMBwHcGJHzmEfP/Vrj8nJEKHtrTn9ZMZmUftyH8wm0H4240MTpyY9y CPUsnDOs0J09Y7SF6FIvoIyJYQ3mbmam82AiQXp4Pq61AmmbmOowMiU2X TtP+1JGHkZAy6KiJXtzYuKXxLQZhobw/s97aXYCrkrQghufy127DjnrS1 uIce86pLm64owOdPL2wCobOIwjG0b4i6w78+tqEg0xeZTb3H8t/tC7P+F sY03h7N4As5S7T8/0Bb6diBJuTolhT/mzf3G7kJI2jJblAu8GxzVZYSOl nQVEwxsuJYjQGAZ7gnccZKhaSdvE4hQ0cri05sBOjAWz8VePz5z+b6f+M w==; X-IronPort-AV: E=McAfee;i="6600,9927,10817"; a="373016383" X-IronPort-AV: E=Sophos;i="6.02,213,1688454000"; d="scan'208,223";a="373016383" X-ExtLoop1: 1 X-IronPort-AV: E=McAfee;i="6600,9927,10817"; a="715865988" X-IronPort-AV: E=Sophos;i="6.02,213,1688454000"; d="scan'208,223";a="715865988" From: Zhenzhong Duan To: qemu-devel@nongnu.org Cc: alex.williamson@redhat.com, clg@redhat.com, jgg@nvidia.com, nicolinc@nvidia.com, joao.m.martins@oracle.com, eric.auger@redhat.com, peterx@redhat.com, jasowang@redhat.com, kevin.tian@intel.com, yi.l.liu@intel.com, yi.y.sun@intel.com, chao.p.peng@intel.com, Zhenzhong Duan , "Michael S. Tsirkin" , Cornelia Huck , Paolo Bonzini , kvm@vger.kernel.org (open list:Overall KVM CPUs) Subject: [PATCH v1 02/22] Update linux-header to support iommufd cdev and hwpt alloc Date: Wed, 30 Aug 2023 18:37:34 +0800 Message-Id: <20230830103754.36461-3-zhenzhong.duan@intel.com> X-Mailer: git-send-email 2.34.1 In-Reply-To: <20230830103754.36461-1-zhenzhong.duan@intel.com> References: <20230830103754.36461-1-zhenzhong.duan@intel.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Received-SPF: pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) client-ip=209.51.188.17; envelope-from=qemu-devel-bounces+importer=patchew.org@nongnu.org; helo=lists.gnu.org; Received-SPF: pass client-ip=192.55.52.93; envelope-from=zhenzhong.duan@intel.com; helo=mgamail.intel.com X-Spam_score_int: -43 X-Spam_score: -4.4 X-Spam_bar: ---- X-Spam_report: (-4.4 / 5.0 requ) BAYES_00=-1.9, DKIMWL_WL_HIGH=-0.001, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, RCVD_IN_DNSWL_MED=-2.3, SPF_HELO_NONE=0.001, SPF_PASS=-0.001 autolearn=ham autolearn_force=no X-Spam_action: no action X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: qemu-devel-bounces+importer=patchew.org@nongnu.org Sender: qemu-devel-bounces+importer=patchew.org@nongnu.org X-ZohoMail-DKIM: pass (identity @intel.com) X-ZM-MESSAGEID: 1693393002736100001 Content-Type: text/plain; charset="utf-8" From https://git.kernel.org/pub/scm/linux/kernel/git/jgg/iommufd.git branch: for_next commit id: eb501c2d96cfce6b42528e8321ea085ec605e790 Signed-off-by: Zhenzhong Duan --- Note this is a placeholder patch. include/standard-headers/linux/fuse.h | 3 + linux-headers/linux/iommufd.h | 444 ++++++++++++++++++++++++++ linux-headers/linux/kvm.h | 13 +- linux-headers/linux/vfio.h | 148 ++++++++- 4 files changed, 604 insertions(+), 4 deletions(-) create mode 100644 linux-headers/linux/iommufd.h diff --git a/include/standard-headers/linux/fuse.h b/include/standard-heade= rs/linux/fuse.h index 35c131a107..2c8b8de9c2 100644 --- a/include/standard-headers/linux/fuse.h +++ b/include/standard-headers/linux/fuse.h @@ -206,6 +206,7 @@ * - add extension header * - add FUSE_EXT_GROUPS * - add FUSE_CREATE_SUPP_GROUP + * - add FUSE_HAS_EXPIRE_ONLY */ =20 #ifndef _LINUX_FUSE_H @@ -365,6 +366,7 @@ struct fuse_file_lock { * FUSE_HAS_INODE_DAX: use per inode DAX * FUSE_CREATE_SUPP_GROUP: add supplementary group info to create, mkdir, * symlink and mknod (single group that matches parent) + * FUSE_HAS_EXPIRE_ONLY: kernel supports expiry-only entry invalidation */ #define FUSE_ASYNC_READ (1 << 0) #define FUSE_POSIX_LOCKS (1 << 1) @@ -402,6 +404,7 @@ struct fuse_file_lock { #define FUSE_SECURITY_CTX (1ULL << 32) #define FUSE_HAS_INODE_DAX (1ULL << 33) #define FUSE_CREATE_SUPP_GROUP (1ULL << 34) +#define FUSE_HAS_EXPIRE_ONLY (1ULL << 35) =20 /** * CUSE INIT request/reply flags diff --git a/linux-headers/linux/iommufd.h b/linux-headers/linux/iommufd.h new file mode 100644 index 0000000000..218bf7ac98 --- /dev/null +++ b/linux-headers/linux/iommufd.h @@ -0,0 +1,444 @@ +/* SPDX-License-Identifier: GPL-2.0 WITH Linux-syscall-note */ +/* Copyright (c) 2021-2022, NVIDIA CORPORATION & AFFILIATES. + */ +#ifndef _IOMMUFD_H +#define _IOMMUFD_H + +#include +#include + +#define IOMMUFD_TYPE (';') + +/** + * DOC: General ioctl format + * + * The ioctl interface follows a general format to allow for extensibility= . Each + * ioctl is passed in a structure pointer as the argument providing the si= ze of + * the structure in the first u32. The kernel checks that any structure sp= ace + * beyond what it understands is 0. This allows userspace to use the backw= ard + * compatible portion while consistently using the newer, larger, structur= es. + * + * ioctls use a standard meaning for common errnos: + * + * - ENOTTY: The IOCTL number itself is not supported at all + * - E2BIG: The IOCTL number is supported, but the provided structure has + * non-zero in a part the kernel does not understand. + * - EOPNOTSUPP: The IOCTL number is supported, and the structure is + * understood, however a known field has a value the kernel does not + * understand or support. + * - EINVAL: Everything about the IOCTL was understood, but a field is not + * correct. + * - ENOENT: An ID or IOVA provided does not exist. + * - ENOMEM: Out of memory. + * - EOVERFLOW: Mathematics overflowed. + * + * As well as additional errnos, within specific ioctls. + */ +enum { + IOMMUFD_CMD_BASE =3D 0x80, + IOMMUFD_CMD_DESTROY =3D IOMMUFD_CMD_BASE, + IOMMUFD_CMD_IOAS_ALLOC, + IOMMUFD_CMD_IOAS_ALLOW_IOVAS, + IOMMUFD_CMD_IOAS_COPY, + IOMMUFD_CMD_IOAS_IOVA_RANGES, + IOMMUFD_CMD_IOAS_MAP, + IOMMUFD_CMD_IOAS_UNMAP, + IOMMUFD_CMD_OPTION, + IOMMUFD_CMD_VFIO_IOAS, + IOMMUFD_CMD_HWPT_ALLOC, + IOMMUFD_CMD_GET_HW_INFO, +}; + +/** + * struct iommu_destroy - ioctl(IOMMU_DESTROY) + * @size: sizeof(struct iommu_destroy) + * @id: iommufd object ID to destroy. Can be any destroyable object type. + * + * Destroy any object held within iommufd. + */ +struct iommu_destroy { + __u32 size; + __u32 id; +}; +#define IOMMU_DESTROY _IO(IOMMUFD_TYPE, IOMMUFD_CMD_DESTROY) + +/** + * struct iommu_ioas_alloc - ioctl(IOMMU_IOAS_ALLOC) + * @size: sizeof(struct iommu_ioas_alloc) + * @flags: Must be 0 + * @out_ioas_id: Output IOAS ID for the allocated object + * + * Allocate an IO Address Space (IOAS) which holds an IO Virtual Address (= IOVA) + * to memory mapping. + */ +struct iommu_ioas_alloc { + __u32 size; + __u32 flags; + __u32 out_ioas_id; +}; +#define IOMMU_IOAS_ALLOC _IO(IOMMUFD_TYPE, IOMMUFD_CMD_IOAS_ALLOC) + +/** + * struct iommu_iova_range - ioctl(IOMMU_IOVA_RANGE) + * @start: First IOVA + * @last: Inclusive last IOVA + * + * An interval in IOVA space. + */ +struct iommu_iova_range { + __aligned_u64 start; + __aligned_u64 last; +}; + +/** + * struct iommu_ioas_iova_ranges - ioctl(IOMMU_IOAS_IOVA_RANGES) + * @size: sizeof(struct iommu_ioas_iova_ranges) + * @ioas_id: IOAS ID to read ranges from + * @num_iovas: Input/Output total number of ranges in the IOAS + * @__reserved: Must be 0 + * @allowed_iovas: Pointer to the output array of struct iommu_iova_range + * @out_iova_alignment: Minimum alignment required for mapping IOVA + * + * Query an IOAS for ranges of allowed IOVAs. Mapping IOVA outside these r= anges + * is not allowed. num_iovas will be set to the total number of iovas and + * the allowed_iovas[] will be filled in as space permits. + * + * The allowed ranges are dependent on the HW path the DMA operation takes= , and + * can change during the lifetime of the IOAS. A fresh empty IOAS will hav= e a + * full range, and each attached device will narrow the ranges based on th= at + * device's HW restrictions. Detaching a device can widen the ranges. User= space + * should query ranges after every attach/detach to know what IOVAs are va= lid + * for mapping. + * + * On input num_iovas is the length of the allowed_iovas array. On output = it is + * the total number of iovas filled in. The ioctl will return -EMSGSIZE an= d set + * num_iovas to the required value if num_iovas is too small. In this case= the + * caller should allocate a larger output array and re-issue the ioctl. + * + * out_iova_alignment returns the minimum IOVA alignment that can be given + * to IOMMU_IOAS_MAP/COPY. IOVA's must satisfy:: + * + * starting_iova % out_iova_alignment =3D=3D 0 + * (starting_iova + length) % out_iova_alignment =3D=3D 0 + * + * out_iova_alignment can be 1 indicating any IOVA is allowed. It cannot + * be higher than the system PAGE_SIZE. + */ +struct iommu_ioas_iova_ranges { + __u32 size; + __u32 ioas_id; + __u32 num_iovas; + __u32 __reserved; + __aligned_u64 allowed_iovas; + __aligned_u64 out_iova_alignment; +}; +#define IOMMU_IOAS_IOVA_RANGES _IO(IOMMUFD_TYPE, IOMMUFD_CMD_IOAS_IOVA_RAN= GES) + +/** + * struct iommu_ioas_allow_iovas - ioctl(IOMMU_IOAS_ALLOW_IOVAS) + * @size: sizeof(struct iommu_ioas_allow_iovas) + * @ioas_id: IOAS ID to allow IOVAs from + * @num_iovas: Input/Output total number of ranges in the IOAS + * @__reserved: Must be 0 + * @allowed_iovas: Pointer to array of struct iommu_iova_range + * + * Ensure a range of IOVAs are always available for allocation. If this ca= ll + * succeeds then IOMMU_IOAS_IOVA_RANGES will never return a list of IOVA r= anges + * that are narrower than the ranges provided here. This call will fail if + * IOMMU_IOAS_IOVA_RANGES is currently narrower than the given ranges. + * + * When an IOAS is first created the IOVA_RANGES will be maximally sized, = and as + * devices are attached the IOVA will narrow based on the device restricti= ons. + * When an allowed range is specified any narrowing will be refused, ie de= vice + * attachment can fail if the device requires limiting within the allowed = range. + * + * Automatic IOVA allocation is also impacted by this call. MAP will only + * allocate within the allowed IOVAs if they are present. + * + * This call replaces the entire allowed list with the given list. + */ +struct iommu_ioas_allow_iovas { + __u32 size; + __u32 ioas_id; + __u32 num_iovas; + __u32 __reserved; + __aligned_u64 allowed_iovas; +}; +#define IOMMU_IOAS_ALLOW_IOVAS _IO(IOMMUFD_TYPE, IOMMUFD_CMD_IOAS_ALLOW_IO= VAS) + +/** + * enum iommufd_ioas_map_flags - Flags for map and copy + * @IOMMU_IOAS_MAP_FIXED_IOVA: If clear the kernel will compute an appropr= iate + * IOVA to place the mapping at + * @IOMMU_IOAS_MAP_WRITEABLE: DMA is allowed to write to this mapping + * @IOMMU_IOAS_MAP_READABLE: DMA is allowed to read from this mapping + */ +enum iommufd_ioas_map_flags { + IOMMU_IOAS_MAP_FIXED_IOVA =3D 1 << 0, + IOMMU_IOAS_MAP_WRITEABLE =3D 1 << 1, + IOMMU_IOAS_MAP_READABLE =3D 1 << 2, +}; + +/** + * struct iommu_ioas_map - ioctl(IOMMU_IOAS_MAP) + * @size: sizeof(struct iommu_ioas_map) + * @flags: Combination of enum iommufd_ioas_map_flags + * @ioas_id: IOAS ID to change the mapping of + * @__reserved: Must be 0 + * @user_va: Userspace pointer to start mapping from + * @length: Number of bytes to map + * @iova: IOVA the mapping was placed at. If IOMMU_IOAS_MAP_FIXED_IOVA is = set + * then this must be provided as input. + * + * Set an IOVA mapping from a user pointer. If FIXED_IOVA is specified the= n the + * mapping will be established at iova, otherwise a suitable location base= d on + * the reserved and allowed lists will be automatically selected and retur= ned in + * iova. + * + * If IOMMU_IOAS_MAP_FIXED_IOVA is specified then the iova range must curr= ently + * be unused, existing IOVA cannot be replaced. + */ +struct iommu_ioas_map { + __u32 size; + __u32 flags; + __u32 ioas_id; + __u32 __reserved; + __aligned_u64 user_va; + __aligned_u64 length; + __aligned_u64 iova; +}; +#define IOMMU_IOAS_MAP _IO(IOMMUFD_TYPE, IOMMUFD_CMD_IOAS_MAP) + +/** + * struct iommu_ioas_copy - ioctl(IOMMU_IOAS_COPY) + * @size: sizeof(struct iommu_ioas_copy) + * @flags: Combination of enum iommufd_ioas_map_flags + * @dst_ioas_id: IOAS ID to change the mapping of + * @src_ioas_id: IOAS ID to copy from + * @length: Number of bytes to copy and map + * @dst_iova: IOVA the mapping was placed at. If IOMMU_IOAS_MAP_FIXED_IOVA= is + * set then this must be provided as input. + * @src_iova: IOVA to start the copy + * + * Copy an already existing mapping from src_ioas_id and establish it in + * dst_ioas_id. The src iova/length must exactly match a range used with + * IOMMU_IOAS_MAP. + * + * This may be used to efficiently clone a subset of an IOAS to another, o= r as a + * kind of 'cache' to speed up mapping. Copy has an efficiency advantage o= ver + * establishing equivalent new mappings, as internal resources are shared,= and + * the kernel will pin the user memory only once. + */ +struct iommu_ioas_copy { + __u32 size; + __u32 flags; + __u32 dst_ioas_id; + __u32 src_ioas_id; + __aligned_u64 length; + __aligned_u64 dst_iova; + __aligned_u64 src_iova; +}; +#define IOMMU_IOAS_COPY _IO(IOMMUFD_TYPE, IOMMUFD_CMD_IOAS_COPY) + +/** + * struct iommu_ioas_unmap - ioctl(IOMMU_IOAS_UNMAP) + * @size: sizeof(struct iommu_ioas_unmap) + * @ioas_id: IOAS ID to change the mapping of + * @iova: IOVA to start the unmapping at + * @length: Number of bytes to unmap, and return back the bytes unmapped + * + * Unmap an IOVA range. The iova/length must be a superset of a previously + * mapped range used with IOMMU_IOAS_MAP or IOMMU_IOAS_COPY. Splitting or + * truncating ranges is not allowed. The values 0 to U64_MAX will unmap + * everything. + */ +struct iommu_ioas_unmap { + __u32 size; + __u32 ioas_id; + __aligned_u64 iova; + __aligned_u64 length; +}; +#define IOMMU_IOAS_UNMAP _IO(IOMMUFD_TYPE, IOMMUFD_CMD_IOAS_UNMAP) + +/** + * enum iommufd_option - ioctl(IOMMU_OPTION_RLIMIT_MODE) and + * ioctl(IOMMU_OPTION_HUGE_PAGES) + * @IOMMU_OPTION_RLIMIT_MODE: + * Change how RLIMIT_MEMLOCK accounting works. The caller must have pri= vilege + * to invoke this. Value 0 (default) is user based accouting, 1 uses pr= ocess + * based accounting. Global option, object_id must be 0 + * @IOMMU_OPTION_HUGE_PAGES: + * Value 1 (default) allows contiguous pages to be combined when genera= ting + * iommu mappings. Value 0 disables combining, everything is mapped to + * PAGE_SIZE. This can be useful for benchmarking. This is a per-IOAS + * option, the object_id must be the IOAS ID. + */ +enum iommufd_option { + IOMMU_OPTION_RLIMIT_MODE =3D 0, + IOMMU_OPTION_HUGE_PAGES =3D 1, +}; + +/** + * enum iommufd_option_ops - ioctl(IOMMU_OPTION_OP_SET) and + * ioctl(IOMMU_OPTION_OP_GET) + * @IOMMU_OPTION_OP_SET: Set the option's value + * @IOMMU_OPTION_OP_GET: Get the option's value + */ +enum iommufd_option_ops { + IOMMU_OPTION_OP_SET =3D 0, + IOMMU_OPTION_OP_GET =3D 1, +}; + +/** + * struct iommu_option - iommu option multiplexer + * @size: sizeof(struct iommu_option) + * @option_id: One of enum iommufd_option + * @op: One of enum iommufd_option_ops + * @__reserved: Must be 0 + * @object_id: ID of the object if required + * @val64: Option value to set or value returned on get + * + * Change a simple option value. This multiplexor allows controlling optio= ns + * on objects. IOMMU_OPTION_OP_SET will load an option and IOMMU_OPTION_OP= _GET + * will return the current value. + */ +struct iommu_option { + __u32 size; + __u32 option_id; + __u16 op; + __u16 __reserved; + __u32 object_id; + __aligned_u64 val64; +}; +#define IOMMU_OPTION _IO(IOMMUFD_TYPE, IOMMUFD_CMD_OPTION) + +/** + * enum iommufd_vfio_ioas_op - IOMMU_VFIO_IOAS_* ioctls + * @IOMMU_VFIO_IOAS_GET: Get the current compatibility IOAS + * @IOMMU_VFIO_IOAS_SET: Change the current compatibility IOAS + * @IOMMU_VFIO_IOAS_CLEAR: Disable VFIO compatibility + */ +enum iommufd_vfio_ioas_op { + IOMMU_VFIO_IOAS_GET =3D 0, + IOMMU_VFIO_IOAS_SET =3D 1, + IOMMU_VFIO_IOAS_CLEAR =3D 2, +}; + +/** + * struct iommu_vfio_ioas - ioctl(IOMMU_VFIO_IOAS) + * @size: sizeof(struct iommu_vfio_ioas) + * @ioas_id: For IOMMU_VFIO_IOAS_SET the input IOAS ID to set + * For IOMMU_VFIO_IOAS_GET will output the IOAS ID + * @op: One of enum iommufd_vfio_ioas_op + * @__reserved: Must be 0 + * + * The VFIO compatibility support uses a single ioas because VFIO APIs do = not + * support the ID field. Set or Get the IOAS that VFIO compatibility will = use. + * When VFIO_GROUP_SET_CONTAINER is used on an iommufd it will get the + * compatibility ioas, either by taking what is already set, or auto creat= ing + * one. From then on VFIO will continue to use that ioas and is not effect= ed by + * this ioctl. SET or CLEAR does not destroy any auto-created IOAS. + */ +struct iommu_vfio_ioas { + __u32 size; + __u32 ioas_id; + __u16 op; + __u16 __reserved; +}; +#define IOMMU_VFIO_IOAS _IO(IOMMUFD_TYPE, IOMMUFD_CMD_VFIO_IOAS) + +/** + * struct iommu_hwpt_alloc - ioctl(IOMMU_HWPT_ALLOC) + * @size: sizeof(struct iommu_hwpt_alloc) + * @flags: Must be 0 + * @dev_id: The device to allocate this HWPT for + * @pt_id: The IOAS to connect this HWPT to + * @out_hwpt_id: The ID of the new HWPT + * @__reserved: Must be 0 + * + * Explicitly allocate a hardware page table object. This is the same obje= ct + * type that is returned by iommufd_device_attach() and represents the + * underlying iommu driver's iommu_domain kernel object. + * + * A HWPT will be created with the IOVA mappings from the given IOAS. + */ +struct iommu_hwpt_alloc { + __u32 size; + __u32 flags; + __u32 dev_id; + __u32 pt_id; + __u32 out_hwpt_id; + __u32 __reserved; +}; +#define IOMMU_HWPT_ALLOC _IO(IOMMUFD_TYPE, IOMMUFD_CMD_HWPT_ALLOC) + +/** + * struct iommu_hw_info_vtd - Intel VT-d hardware information + * + * @flags: Must be 0 + * @__reserved: Must be 0 + * + * @cap_reg: Value of Intel VT-d capability register defined in VT-d spec + * section 11.4.2 Capability Register. + * @ecap_reg: Value of Intel VT-d capability register defined in VT-d spec + * section 11.4.3 Extended Capability Register. + * + * User needs to understand the Intel VT-d specification to decode the + * register value. + */ +struct iommu_hw_info_vtd { + __u32 flags; + __u32 __reserved; + __aligned_u64 cap_reg; + __aligned_u64 ecap_reg; +}; + +/** + * enum iommu_hw_info_type - IOMMU Hardware Info Types + * @IOMMU_HW_INFO_TYPE_NONE: Used by the drivers that do not report hardwa= re + * info + * @IOMMU_HW_INFO_TYPE_INTEL_VTD: Intel VT-d iommu info type + */ +enum iommu_hw_info_type { + IOMMU_HW_INFO_TYPE_NONE, + IOMMU_HW_INFO_TYPE_INTEL_VTD, +}; + +/** + * struct iommu_hw_info - ioctl(IOMMU_GET_HW_INFO) + * @size: sizeof(struct iommu_hw_info) + * @flags: Must be 0 + * @dev_id: The device bound to the iommufd + * @data_len: Input the length of a user buffer in bytes. Output the lengt= h of + * data that kernel supports + * @data_uptr: User pointer to a user-space buffer used by the kernel to f= ill + * the iommu type specific hardware information data + * @out_data_type: Output the iommu hardware info type as defined in the e= num + * iommu_hw_info_type. + * @__reserved: Must be 0 + * + * Query an iommu type specific hardware information data from an iommu be= hind + * a given device that has been bound to iommufd. This hardware info data = will + * be used to sync capabilities between the virtual iommu and the physical + * iommu, e.g. a nested translation setup needs to check the hardware info= , so + * a guest stage-1 page table can be compatible with the physical iommu. + * + * To capture an iommu type specific hardware information data, @data_uptr= and + * its length @data_len must be provided. Trailing bytes will be zeroed if= the + * user buffer is larger than the data that kernel has. Otherwise, kernel = only + * fills the buffer using the given length in @data_len. If the ioctl succ= eeds, + * @data_len will be updated to the length that kernel actually supports, + * @out_data_type will be filled to decode the data filled in the buffer + * pointed by @data_uptr. Input @data_len =3D=3D zero is allowed. + */ +struct iommu_hw_info { + __u32 size; + __u32 flags; + __u32 dev_id; + __u32 data_len; + __aligned_u64 data_uptr; + __u32 out_data_type; + __u32 __reserved; +}; +#define IOMMU_GET_HW_INFO _IO(IOMMUFD_TYPE, IOMMUFD_CMD_GET_HW_INFO) +#endif diff --git a/linux-headers/linux/kvm.h b/linux-headers/linux/kvm.h index 1f3f3333a4..0d74ee999a 100644 --- a/linux-headers/linux/kvm.h +++ b/linux-headers/linux/kvm.h @@ -1414,9 +1414,16 @@ struct kvm_device_attr { __u64 addr; /* userspace address of attr data */ }; =20 -#define KVM_DEV_VFIO_GROUP 1 -#define KVM_DEV_VFIO_GROUP_ADD 1 -#define KVM_DEV_VFIO_GROUP_DEL 2 +#define KVM_DEV_VFIO_FILE 1 + +#define KVM_DEV_VFIO_FILE_ADD 1 +#define KVM_DEV_VFIO_FILE_DEL 2 + +/* KVM_DEV_VFIO_GROUP aliases are for compile time uapi compatibility */ +#define KVM_DEV_VFIO_GROUP KVM_DEV_VFIO_FILE + +#define KVM_DEV_VFIO_GROUP_ADD KVM_DEV_VFIO_FILE_ADD +#define KVM_DEV_VFIO_GROUP_DEL KVM_DEV_VFIO_FILE_DEL #define KVM_DEV_VFIO_GROUP_SET_SPAPR_TCE 3 =20 enum kvm_device_type { diff --git a/linux-headers/linux/vfio.h b/linux-headers/linux/vfio.h index 16db89071e..7326ace436 100644 --- a/linux-headers/linux/vfio.h +++ b/linux-headers/linux/vfio.h @@ -677,11 +677,60 @@ enum { * VFIO_DEVICE_GET_PCI_HOT_RESET_INFO - _IOWR(VFIO_TYPE, VFIO_BASE + 12, * struct vfio_pci_hot_reset_info) * + * This command is used to query the affected devices in the hot reset for + * a given device. + * + * This command always reports the segment, bus, and devfn information for + * each affected device, and selectively reports the group_id or devid per + * the way how the calling device is opened. + * + * - If the calling device is opened via the traditional group/container + * API, group_id is reported. User should check if it has owned all + * the affected devices and provides a set of group fds to prove the + * ownership in VFIO_DEVICE_PCI_HOT_RESET ioctl. + * + * - If the calling device is opened as a cdev, devid is reported. + * Flag VFIO_PCI_HOT_RESET_FLAG_DEV_ID is set to indicate this + * data type. All the affected devices should be represented in + * the dev_set, ex. bound to a vfio driver, and also be owned by + * this interface which is determined by the following conditions: + * 1) Has a valid devid within the iommufd_ctx of the calling device. + * Ownership cannot be determined across separate iommufd_ctx and + * the cdev calling conventions do not support a proof-of-ownership + * model as provided in the legacy group interface. In this case + * valid devid with value greater than zero is provided in the return + * structure. + * 2) Does not have a valid devid within the iommufd_ctx of the calling + * device, but belongs to the same IOMMU group as the calling device + * or another opened device that has a valid devid within the + * iommufd_ctx of the calling device. This provides implicit ownersh= ip + * for devices within the same DMA isolation context. In this case + * the devid value of VFIO_PCI_DEVID_OWNED is provided in the return + * structure. + * + * A devid value of VFIO_PCI_DEVID_NOT_OWNED is provided in the return + * structure for affected devices where device is NOT represented in the + * dev_set or ownership is not available. Such devices prevent the use + * of VFIO_DEVICE_PCI_HOT_RESET ioctl outside of the proof-of-ownership + * calling conventions (ie. via legacy group accessed devices). Flag + * VFIO_PCI_HOT_RESET_FLAG_DEV_ID_OWNED would be set when all the + * affected devices are represented in the dev_set and also owned by + * the user. This flag is available only when + * flag VFIO_PCI_HOT_RESET_FLAG_DEV_ID is set, otherwise reserved. + * When set, user could invoke VFIO_DEVICE_PCI_HOT_RESET with a zero + * length fd array on the calling device as the ownership is validated + * by iommufd_ctx. + * * Return: 0 on success, -errno on failure: * -enospc =3D insufficient buffer, -enodev =3D unsupported for device. */ struct vfio_pci_dependent_device { - __u32 group_id; + union { + __u32 group_id; + __u32 devid; +#define VFIO_PCI_DEVID_OWNED 0 +#define VFIO_PCI_DEVID_NOT_OWNED -1 + }; __u16 segment; __u8 bus; __u8 devfn; /* Use PCI_SLOT/PCI_FUNC */ @@ -690,6 +739,8 @@ struct vfio_pci_dependent_device { struct vfio_pci_hot_reset_info { __u32 argsz; __u32 flags; +#define VFIO_PCI_HOT_RESET_FLAG_DEV_ID (1 << 0) +#define VFIO_PCI_HOT_RESET_FLAG_DEV_ID_OWNED (1 << 1) __u32 count; struct vfio_pci_dependent_device devices[]; }; @@ -700,6 +751,24 @@ struct vfio_pci_hot_reset_info { * VFIO_DEVICE_PCI_HOT_RESET - _IOW(VFIO_TYPE, VFIO_BASE + 13, * struct vfio_pci_hot_reset) * + * A PCI hot reset results in either a bus or slot reset which may affect + * other devices sharing the bus/slot. The calling user must have + * ownership of the full set of affected devices as determined by the + * VFIO_DEVICE_GET_PCI_HOT_RESET_INFO ioctl. + * + * When called on a device file descriptor acquired through the vfio + * group interface, the user is required to provide proof of ownership + * of those affected devices via the group_fds array in struct + * vfio_pci_hot_reset. + * + * When called on a direct cdev opened vfio device, the flags field of + * struct vfio_pci_hot_reset_info reports the ownership status of the + * affected devices and this ioctl must be called with an empty group_fds + * array. See above INFO ioctl definition for ownership requirements. + * + * Mixed usage of legacy groups and cdevs across the set of affected + * devices is not supported. + * * Return: 0 on success, -errno on failure. */ struct vfio_pci_hot_reset { @@ -828,6 +897,83 @@ struct vfio_device_feature { =20 #define VFIO_DEVICE_FEATURE _IO(VFIO_TYPE, VFIO_BASE + 17) =20 +/* + * VFIO_DEVICE_BIND_IOMMUFD - _IOR(VFIO_TYPE, VFIO_BASE + 18, + * struct vfio_device_bind_iommufd) + * @argsz: User filled size of this data. + * @flags: Must be 0. + * @iommufd: iommufd to bind. + * @out_devid: The device id generated by this bind. devid is a handle for + * this device/iommufd bond and can be used in IOMMUFD commands. + * + * Bind a vfio_device to the specified iommufd. + * + * User is restricted from accessing the device before the binding operati= on + * is completed. Only allowed on cdev fds. + * + * Unbind is automatically conducted when device fd is closed. + * + * Return: 0 on success, -errno on failure. + */ +struct vfio_device_bind_iommufd { + __u32 argsz; + __u32 flags; + __s32 iommufd; + __u32 out_devid; +}; + +#define VFIO_DEVICE_BIND_IOMMUFD _IO(VFIO_TYPE, VFIO_BASE + 18) + +/* + * VFIO_DEVICE_ATTACH_IOMMUFD_PT - _IOW(VFIO_TYPE, VFIO_BASE + 19, + * struct vfio_device_attach_iommufd_pt) + * @argsz: User filled size of this data. + * @flags: Must be 0. + * @pt_id: Input the target id which can represent an ioas or a hwpt + * allocated via iommufd subsystem. + * Output the input ioas id or the attached hwpt id which could + * be the specified hwpt itself or a hwpt automatically created + * for the specified ioas by kernel during the attachment. + * + * Associate the device with an address space within the bound iommufd. + * Undo by VFIO_DEVICE_DETACH_IOMMUFD_PT or device fd close. This is only + * allowed on cdev fds. + * + * If a vfio device is currently attached to a valid hw_pagetable, without= doing + * a VFIO_DEVICE_DETACH_IOMMUFD_PT, a second VFIO_DEVICE_ATTACH_IOMMUFD_PT= ioctl + * passing in another hw_pagetable (hwpt) id is allowed. This action, also= known + * as a hw_pagetable replacement, will replace the device's currently atta= ched + * hw_pagetable with a new hw_pagetable corresponding to the given pt_id. + * + * Return: 0 on success, -errno on failure. + */ +struct vfio_device_attach_iommufd_pt { + __u32 argsz; + __u32 flags; + __u32 pt_id; +}; + +#define VFIO_DEVICE_ATTACH_IOMMUFD_PT _IO(VFIO_TYPE, VFIO_BASE + 19) + +/* + * VFIO_DEVICE_DETACH_IOMMUFD_PT - _IOW(VFIO_TYPE, VFIO_BASE + 20, + * struct vfio_device_detach_iommufd_pt) + * @argsz: User filled size of this data. + * @flags: Must be 0. + * + * Remove the association of the device and its current associated address + * space. After it, the device should be in a blocking DMA state. This i= s only + * allowed on cdev fds. + * + * Return: 0 on success, -errno on failure. + */ +struct vfio_device_detach_iommufd_pt { + __u32 argsz; + __u32 flags; +}; + +#define VFIO_DEVICE_DETACH_IOMMUFD_PT _IO(VFIO_TYPE, VFIO_BASE + 20) + /* * Provide support for setting a PCI VF Token, which is used as a shared * secret between PF and VF drivers. This feature may only be set on a --=20 2.34.1 From nobody Wed May 15 22:12:51 2024 Delivered-To: importer@patchew.org Authentication-Results: mx.zohomail.com; dkim=pass header.i=@intel.com; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass(p=none dis=none) header.from=intel.com ARC-Seal: i=1; a=rsa-sha256; t=1693392842; cv=none; d=zohomail.com; s=zohoarc; b=BNXwEnPl927QlFWijOtONPkq2mUV54beuOIa4QljeYLEMDohuhv5+pzKfoZf5UQR6CetlOxHSNq2+0cGjOH+v55mFCG0jwT0z7FU4BiW6h9GCrtK24l5p+qSeJovOLDbcLWP+ipgtNo1Sgc7ImV/tvjHsGTeqWFQOCb+f0MGvGo= ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=zohomail.com; s=zohoarc; t=1693392842; h=Content-Transfer-Encoding:Cc:Date:From:In-Reply-To:List-Subscribe:List-Post:List-Id:List-Archive:List-Help:List-Unsubscribe:MIME-Version:Message-ID:References:Sender:Subject:To; bh=i5m9826QhjQUZX+IbFLbNfqwyd/PcFJSo7+4kmecw0Y=; b=gadK1AHFHqWyHVC+XxmASp3ArlM2l1WIMI0ONQ+7qT5iH6IOt6dq68Jeun+6Xy+/prPKFZvdW6hPl9JJ2HZ14kdRIizGkiQRx4j7oTL90fd+6LsDSOW/EE/ULoxoYYV2TfkyH6Dh0vgG2exl5k/m3mOtoaN0G1tIuarbIjlTdMM= ARC-Authentication-Results: i=1; mx.zohomail.com; dkim=pass header.i=@intel.com; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass header.from= (p=none dis=none) Return-Path: Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) by mx.zohomail.com with SMTPS id 1693392842236390.3242861946096; Wed, 30 Aug 2023 03:54:02 -0700 (PDT) Received: from localhost ([::1] helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1qbIp0-0004FK-6d; Wed, 30 Aug 2023 06:53:14 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1qbIox-0004Eo-W1 for qemu-devel@nongnu.org; Wed, 30 Aug 2023 06:53:12 -0400 Received: from mgamail.intel.com ([192.55.52.93]) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1qbIos-0007eq-LN for qemu-devel@nongnu.org; Wed, 30 Aug 2023 06:53:11 -0400 Received: from orsmga006.jf.intel.com ([10.7.209.51]) by fmsmga102.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 30 Aug 2023 03:52:44 -0700 Received: from duan-server-s2600bt.bj.intel.com ([10.240.192.147]) by orsmga006-auth.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 30 Aug 2023 03:52:38 -0700 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1693392786; x=1724928786; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=mAxINrmbhPPumv/EnDokLXHI5hJ3vUqibb7HnMNAodQ=; b=X8Wzs8FicZZAzZP65/xmrLAJBZUfj5Fvl1jpawMsLiEZ//w39d4+iYdh /+L6NVy/PVu6q5A2hgRJ6ydXVPIZbgkJXoJr3fYw680wZvlVZCNgGkKBh 2sx9SfT1vYGZOxr+m3EgvkInKTMoUPlRoaFrl2xTe7ISUkRj3arHGIxE3 YvIS9f21ElTX3kVJkepPRi8VrUZ9Y8LfhKnxRHHasPbB7tVGfiaRS4xp3 4B4T/1pLm/YrtQURsZHmPedVaDoIgnsAinkGmn7cwsPcV8HStmvDfb9fm PexVsSOnbWMsCheX7s+AyDhk8gDrR3wqCFG3LYLTWXHR3hTZVqcmbCJxt A==; X-IronPort-AV: E=McAfee;i="6600,9927,10817"; a="373016391" X-IronPort-AV: E=Sophos;i="6.02,213,1688454000"; d="scan'208";a="373016391" X-ExtLoop1: 1 X-IronPort-AV: E=McAfee;i="6600,9927,10817"; a="715866015" X-IronPort-AV: E=Sophos;i="6.02,213,1688454000"; d="scan'208";a="715866015" From: Zhenzhong Duan To: qemu-devel@nongnu.org Cc: alex.williamson@redhat.com, clg@redhat.com, jgg@nvidia.com, nicolinc@nvidia.com, joao.m.martins@oracle.com, eric.auger@redhat.com, peterx@redhat.com, jasowang@redhat.com, kevin.tian@intel.com, yi.l.liu@intel.com, yi.y.sun@intel.com, chao.p.peng@intel.com, Yi Sun , Zhenzhong Duan Subject: [PATCH v1 03/22] vfio/common: Move IOMMU agnostic helpers to a separate file Date: Wed, 30 Aug 2023 18:37:35 +0800 Message-Id: <20230830103754.36461-4-zhenzhong.duan@intel.com> X-Mailer: git-send-email 2.34.1 In-Reply-To: <20230830103754.36461-1-zhenzhong.duan@intel.com> References: <20230830103754.36461-1-zhenzhong.duan@intel.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Received-SPF: pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) client-ip=209.51.188.17; envelope-from=qemu-devel-bounces+importer=patchew.org@nongnu.org; helo=lists.gnu.org; Received-SPF: pass client-ip=192.55.52.93; envelope-from=zhenzhong.duan@intel.com; helo=mgamail.intel.com X-Spam_score_int: -43 X-Spam_score: -4.4 X-Spam_bar: ---- X-Spam_report: (-4.4 / 5.0 requ) BAYES_00=-1.9, DKIMWL_WL_HIGH=-0.001, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, RCVD_IN_DNSWL_MED=-2.3, SPF_HELO_NONE=0.001, SPF_PASS=-0.001 autolearn=ham autolearn_force=no X-Spam_action: no action X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: qemu-devel-bounces+importer=patchew.org@nongnu.org Sender: qemu-devel-bounces+importer=patchew.org@nongnu.org X-ZohoMail-DKIM: pass (identity @intel.com) X-ZM-MESSAGEID: 1693392914212100005 Content-Type: text/plain; charset="utf-8" From: Yi Liu Move low-level iommu agnostic helpers to a separate helpers.c file. They relate to regions, interrupts and device/region capabilities. Signed-off-by: Eric Auger Signed-off-by: Yi Sun Signed-off-by: Yi Liu Signed-off-by: Zhenzhong Duan --- hw/vfio/common.c | 569 -------------------------------- hw/vfio/helpers.c | 598 ++++++++++++++++++++++++++++++++++ hw/vfio/meson.build | 1 + include/hw/vfio/vfio-common.h | 2 + 4 files changed, 601 insertions(+), 569 deletions(-) create mode 100644 hw/vfio/helpers.c diff --git a/hw/vfio/common.c b/hw/vfio/common.c index 9aac21abb7..9ca695837f 100644 --- a/hw/vfio/common.c +++ b/hw/vfio/common.c @@ -61,84 +61,6 @@ static QLIST_HEAD(, VFIOAddressSpace) vfio_address_space= s =3D static int vfio_kvm_device_fd =3D -1; #endif =20 -/* - * Common VFIO interrupt disable - */ -void vfio_disable_irqindex(VFIODevice *vbasedev, int index) -{ - struct vfio_irq_set irq_set =3D { - .argsz =3D sizeof(irq_set), - .flags =3D VFIO_IRQ_SET_DATA_NONE | VFIO_IRQ_SET_ACTION_TRIGGER, - .index =3D index, - .start =3D 0, - .count =3D 0, - }; - - ioctl(vbasedev->fd, VFIO_DEVICE_SET_IRQS, &irq_set); -} - -void vfio_unmask_single_irqindex(VFIODevice *vbasedev, int index) -{ - struct vfio_irq_set irq_set =3D { - .argsz =3D sizeof(irq_set), - .flags =3D VFIO_IRQ_SET_DATA_NONE | VFIO_IRQ_SET_ACTION_UNMASK, - .index =3D index, - .start =3D 0, - .count =3D 1, - }; - - ioctl(vbasedev->fd, VFIO_DEVICE_SET_IRQS, &irq_set); -} - -void vfio_mask_single_irqindex(VFIODevice *vbasedev, int index) -{ - struct vfio_irq_set irq_set =3D { - .argsz =3D sizeof(irq_set), - .flags =3D VFIO_IRQ_SET_DATA_NONE | VFIO_IRQ_SET_ACTION_MASK, - .index =3D index, - .start =3D 0, - .count =3D 1, - }; - - ioctl(vbasedev->fd, VFIO_DEVICE_SET_IRQS, &irq_set); -} - -static inline const char *action_to_str(int action) -{ - switch (action) { - case VFIO_IRQ_SET_ACTION_MASK: - return "MASK"; - case VFIO_IRQ_SET_ACTION_UNMASK: - return "UNMASK"; - case VFIO_IRQ_SET_ACTION_TRIGGER: - return "TRIGGER"; - default: - return "UNKNOWN ACTION"; - } -} - -static const char *index_to_str(VFIODevice *vbasedev, int index) -{ - if (vbasedev->type !=3D VFIO_DEVICE_TYPE_PCI) { - return NULL; - } - - switch (index) { - case VFIO_PCI_INTX_IRQ_INDEX: - return "INTX"; - case VFIO_PCI_MSI_IRQ_INDEX: - return "MSI"; - case VFIO_PCI_MSIX_IRQ_INDEX: - return "MSIX"; - case VFIO_PCI_ERR_IRQ_INDEX: - return "ERR"; - case VFIO_PCI_REQ_IRQ_INDEX: - return "REQ"; - default: - return NULL; - } -} - static int vfio_ram_block_discard_disable(VFIOContainer *container, bool s= tate) { switch (container->iommu_type) { @@ -162,160 +84,6 @@ static int vfio_ram_block_discard_disable(VFIOContaine= r *container, bool state) } } =20 -int vfio_set_irq_signaling(VFIODevice *vbasedev, int index, int subindex, - int action, int fd, Error **errp) -{ - struct vfio_irq_set *irq_set; - int argsz, ret =3D 0; - const char *name; - int32_t *pfd; - - argsz =3D sizeof(*irq_set) + sizeof(*pfd); - - irq_set =3D g_malloc0(argsz); - irq_set->argsz =3D argsz; - irq_set->flags =3D VFIO_IRQ_SET_DATA_EVENTFD | action; - irq_set->index =3D index; - irq_set->start =3D subindex; - irq_set->count =3D 1; - pfd =3D (int32_t *)&irq_set->data; - *pfd =3D fd; - - if (ioctl(vbasedev->fd, VFIO_DEVICE_SET_IRQS, irq_set)) { - ret =3D -errno; - } - g_free(irq_set); - - if (!ret) { - return 0; - } - - error_setg_errno(errp, -ret, "VFIO_DEVICE_SET_IRQS failure"); - - name =3D index_to_str(vbasedev, index); - if (name) { - error_prepend(errp, "%s-%d: ", name, subindex); - } else { - error_prepend(errp, "index %d-%d: ", index, subindex); - } - error_prepend(errp, - "Failed to %s %s eventfd signaling for interrupt ", - fd < 0 ? "tear down" : "set up", action_to_str(action)); - return ret; -} - -/* - * IO Port/MMIO - Beware of the endians, VFIO is always little endian - */ -void vfio_region_write(void *opaque, hwaddr addr, - uint64_t data, unsigned size) -{ - VFIORegion *region =3D opaque; - VFIODevice *vbasedev =3D region->vbasedev; - union { - uint8_t byte; - uint16_t word; - uint32_t dword; - uint64_t qword; - } buf; - - switch (size) { - case 1: - buf.byte =3D data; - break; - case 2: - buf.word =3D cpu_to_le16(data); - break; - case 4: - buf.dword =3D cpu_to_le32(data); - break; - case 8: - buf.qword =3D cpu_to_le64(data); - break; - default: - hw_error("vfio: unsupported write size, %u bytes", size); - break; - } - - if (pwrite(vbasedev->fd, &buf, size, region->fd_offset + addr) !=3D si= ze) { - error_report("%s(%s:region%d+0x%"HWADDR_PRIx", 0x%"PRIx64 - ",%d) failed: %m", - __func__, vbasedev->name, region->nr, - addr, data, size); - } - - trace_vfio_region_write(vbasedev->name, region->nr, addr, data, size); - - /* - * A read or write to a BAR always signals an INTx EOI. This will - * do nothing if not pending (including not in INTx mode). We assume - * that a BAR access is in response to an interrupt and that BAR - * accesses will service the interrupt. Unfortunately, we don't know - * which access will service the interrupt, so we're potentially - * getting quite a few host interrupts per guest interrupt. - */ - vbasedev->ops->vfio_eoi(vbasedev); -} - -uint64_t vfio_region_read(void *opaque, - hwaddr addr, unsigned size) -{ - VFIORegion *region =3D opaque; - VFIODevice *vbasedev =3D region->vbasedev; - union { - uint8_t byte; - uint16_t word; - uint32_t dword; - uint64_t qword; - } buf; - uint64_t data =3D 0; - - if (pread(vbasedev->fd, &buf, size, region->fd_offset + addr) !=3D siz= e) { - error_report("%s(%s:region%d+0x%"HWADDR_PRIx", %d) failed: %m", - __func__, vbasedev->name, region->nr, - addr, size); - return (uint64_t)-1; - } - switch (size) { - case 1: - data =3D buf.byte; - break; - case 2: - data =3D le16_to_cpu(buf.word); - break; - case 4: - data =3D le32_to_cpu(buf.dword); - break; - case 8: - data =3D le64_to_cpu(buf.qword); - break; - default: - hw_error("vfio: unsupported read size, %u bytes", size); - break; - } - - trace_vfio_region_read(vbasedev->name, region->nr, addr, size, data); - - /* Same as write above */ - vbasedev->ops->vfio_eoi(vbasedev); - - return data; -} - -const MemoryRegionOps vfio_region_ops =3D { - .read =3D vfio_region_read, - .write =3D vfio_region_write, - .endianness =3D DEVICE_LITTLE_ENDIAN, - .valid =3D { - .min_access_size =3D 1, - .max_access_size =3D 8, - }, - .impl =3D { - .min_access_size =3D 1, - .max_access_size =3D 8, - }, -}; - /* * Device state interfaces */ @@ -1916,30 +1684,6 @@ static void vfio_listener_release(VFIOContainer *con= tainer) } } =20 -static struct vfio_info_cap_header * -vfio_get_cap(void *ptr, uint32_t cap_offset, uint16_t id) -{ - struct vfio_info_cap_header *hdr; - - for (hdr =3D ptr + cap_offset; hdr !=3D ptr; hdr =3D ptr + hdr->next) { - if (hdr->id =3D=3D id) { - return hdr; - } - } - - return NULL; -} - -struct vfio_info_cap_header * -vfio_get_region_info_cap(struct vfio_region_info *info, uint16_t id) -{ - if (!(info->flags & VFIO_REGION_INFO_FLAG_CAPS)) { - return NULL; - } - - return vfio_get_cap((void *)info, info->cap_offset, id); -} - static struct vfio_info_cap_header * vfio_get_iommu_type1_info_cap(struct vfio_iommu_type1_info *info, uint16_t= id) { @@ -1950,16 +1694,6 @@ vfio_get_iommu_type1_info_cap(struct vfio_iommu_type= 1_info *info, uint16_t id) return vfio_get_cap((void *)info, info->cap_offset, id); } =20 -struct vfio_info_cap_header * -vfio_get_device_info_cap(struct vfio_device_info *info, uint16_t id) -{ - if (!(info->flags & VFIO_DEVICE_FLAGS_CAPS)) { - return NULL; - } - - return vfio_get_cap((void *)info, info->cap_offset, id); -} - bool vfio_get_info_dma_avail(struct vfio_iommu_type1_info *info, unsigned int *avail) { @@ -1981,232 +1715,6 @@ bool vfio_get_info_dma_avail(struct vfio_iommu_type= 1_info *info, return true; } =20 -static int vfio_setup_region_sparse_mmaps(VFIORegion *region, - struct vfio_region_info *info) -{ - struct vfio_info_cap_header *hdr; - struct vfio_region_info_cap_sparse_mmap *sparse; - int i, j; - - hdr =3D vfio_get_region_info_cap(info, VFIO_REGION_INFO_CAP_SPARSE_MMA= P); - if (!hdr) { - return -ENODEV; - } - - sparse =3D container_of(hdr, struct vfio_region_info_cap_sparse_mmap, = header); - - trace_vfio_region_sparse_mmap_header(region->vbasedev->name, - region->nr, sparse->nr_areas); - - region->mmaps =3D g_new0(VFIOMmap, sparse->nr_areas); - - for (i =3D 0, j =3D 0; i < sparse->nr_areas; i++) { - if (sparse->areas[i].size) { - trace_vfio_region_sparse_mmap_entry(i, sparse->areas[i].offset, - sparse->areas[i].offset + - sparse->areas[i].size - 1); - region->mmaps[j].offset =3D sparse->areas[i].offset; - region->mmaps[j].size =3D sparse->areas[i].size; - j++; - } - } - - region->nr_mmaps =3D j; - region->mmaps =3D g_realloc(region->mmaps, j * sizeof(VFIOMmap)); - - return 0; -} - -int vfio_region_setup(Object *obj, VFIODevice *vbasedev, VFIORegion *regio= n, - int index, const char *name) -{ - struct vfio_region_info *info; - int ret; - - ret =3D vfio_get_region_info(vbasedev, index, &info); - if (ret) { - return ret; - } - - region->vbasedev =3D vbasedev; - region->flags =3D info->flags; - region->size =3D info->size; - region->fd_offset =3D info->offset; - region->nr =3D index; - - if (region->size) { - region->mem =3D g_new0(MemoryRegion, 1); - memory_region_init_io(region->mem, obj, &vfio_region_ops, - region, name, region->size); - - if (!vbasedev->no_mmap && - region->flags & VFIO_REGION_INFO_FLAG_MMAP) { - - ret =3D vfio_setup_region_sparse_mmaps(region, info); - - if (ret) { - region->nr_mmaps =3D 1; - region->mmaps =3D g_new0(VFIOMmap, region->nr_mmaps); - region->mmaps[0].offset =3D 0; - region->mmaps[0].size =3D region->size; - } - } - } - - g_free(info); - - trace_vfio_region_setup(vbasedev->name, index, name, - region->flags, region->fd_offset, region->size= ); - return 0; -} - -static void vfio_subregion_unmap(VFIORegion *region, int index) -{ - trace_vfio_region_unmap(memory_region_name(®ion->mmaps[index].mem), - region->mmaps[index].offset, - region->mmaps[index].offset + - region->mmaps[index].size - 1); - memory_region_del_subregion(region->mem, ®ion->mmaps[index].mem); - munmap(region->mmaps[index].mmap, region->mmaps[index].size); - object_unparent(OBJECT(®ion->mmaps[index].mem)); - region->mmaps[index].mmap =3D NULL; -} - -int vfio_region_mmap(VFIORegion *region) -{ - int i, prot =3D 0; - char *name; - - if (!region->mem) { - return 0; - } - - prot |=3D region->flags & VFIO_REGION_INFO_FLAG_READ ? PROT_READ : 0; - prot |=3D region->flags & VFIO_REGION_INFO_FLAG_WRITE ? PROT_WRITE : 0; - - for (i =3D 0; i < region->nr_mmaps; i++) { - region->mmaps[i].mmap =3D mmap(NULL, region->mmaps[i].size, prot, - MAP_SHARED, region->vbasedev->fd, - region->fd_offset + - region->mmaps[i].offset); - if (region->mmaps[i].mmap =3D=3D MAP_FAILED) { - int ret =3D -errno; - - trace_vfio_region_mmap_fault(memory_region_name(region->mem), = i, - region->fd_offset + - region->mmaps[i].offset, - region->fd_offset + - region->mmaps[i].offset + - region->mmaps[i].size - 1, ret); - - region->mmaps[i].mmap =3D NULL; - - for (i--; i >=3D 0; i--) { - vfio_subregion_unmap(region, i); - } - - return ret; - } - - name =3D g_strdup_printf("%s mmaps[%d]", - memory_region_name(region->mem), i); - memory_region_init_ram_device_ptr(®ion->mmaps[i].mem, - memory_region_owner(region->mem), - name, region->mmaps[i].size, - region->mmaps[i].mmap); - g_free(name); - memory_region_add_subregion(region->mem, region->mmaps[i].offset, - ®ion->mmaps[i].mem); - - trace_vfio_region_mmap(memory_region_name(®ion->mmaps[i].mem), - region->mmaps[i].offset, - region->mmaps[i].offset + - region->mmaps[i].size - 1); - } - - return 0; -} - -void vfio_region_unmap(VFIORegion *region) -{ - int i; - - if (!region->mem) { - return; - } - - for (i =3D 0; i < region->nr_mmaps; i++) { - if (region->mmaps[i].mmap) { - vfio_subregion_unmap(region, i); - } - } -} - -void vfio_region_exit(VFIORegion *region) -{ - int i; - - if (!region->mem) { - return; - } - - for (i =3D 0; i < region->nr_mmaps; i++) { - if (region->mmaps[i].mmap) { - memory_region_del_subregion(region->mem, ®ion->mmaps[i].mem= ); - } - } - - trace_vfio_region_exit(region->vbasedev->name, region->nr); -} - -void vfio_region_finalize(VFIORegion *region) -{ - int i; - - if (!region->mem) { - return; - } - - for (i =3D 0; i < region->nr_mmaps; i++) { - if (region->mmaps[i].mmap) { - munmap(region->mmaps[i].mmap, region->mmaps[i].size); - object_unparent(OBJECT(®ion->mmaps[i].mem)); - } - } - - object_unparent(OBJECT(region->mem)); - - g_free(region->mem); - g_free(region->mmaps); - - trace_vfio_region_finalize(region->vbasedev->name, region->nr); - - region->mem =3D NULL; - region->mmaps =3D NULL; - region->nr_mmaps =3D 0; - region->size =3D 0; - region->flags =3D 0; - region->nr =3D 0; -} - -void vfio_region_mmaps_set_enabled(VFIORegion *region, bool enabled) -{ - int i; - - if (!region->mem) { - return; - } - - for (i =3D 0; i < region->nr_mmaps; i++) { - if (region->mmaps[i].mmap) { - memory_region_set_enabled(®ion->mmaps[i].mem, enabled); - } - } - - trace_vfio_region_mmaps_set_enabled(memory_region_name(region->mem), - enabled); -} - void vfio_reset_handler(void *opaque) { VFIOGroup *group; @@ -2905,83 +2413,6 @@ void vfio_put_base_device(VFIODevice *vbasedev) close(vbasedev->fd); } =20 -int vfio_get_region_info(VFIODevice *vbasedev, int index, - struct vfio_region_info **info) -{ - size_t argsz =3D sizeof(struct vfio_region_info); - - *info =3D g_malloc0(argsz); - - (*info)->index =3D index; -retry: - (*info)->argsz =3D argsz; - - if (ioctl(vbasedev->fd, VFIO_DEVICE_GET_REGION_INFO, *info)) { - g_free(*info); - *info =3D NULL; - return -errno; - } - - if ((*info)->argsz > argsz) { - argsz =3D (*info)->argsz; - *info =3D g_realloc(*info, argsz); - - goto retry; - } - - return 0; -} - -int vfio_get_dev_region_info(VFIODevice *vbasedev, uint32_t type, - uint32_t subtype, struct vfio_region_info **i= nfo) -{ - int i; - - for (i =3D 0; i < vbasedev->num_regions; i++) { - struct vfio_info_cap_header *hdr; - struct vfio_region_info_cap_type *cap_type; - - if (vfio_get_region_info(vbasedev, i, info)) { - continue; - } - - hdr =3D vfio_get_region_info_cap(*info, VFIO_REGION_INFO_CAP_TYPE); - if (!hdr) { - g_free(*info); - continue; - } - - cap_type =3D container_of(hdr, struct vfio_region_info_cap_type, h= eader); - - trace_vfio_get_dev_region(vbasedev->name, i, - cap_type->type, cap_type->subtype); - - if (cap_type->type =3D=3D type && cap_type->subtype =3D=3D subtype= ) { - return 0; - } - - g_free(*info); - } - - *info =3D NULL; - return -ENODEV; -} - -bool vfio_has_region_cap(VFIODevice *vbasedev, int region, uint16_t cap_ty= pe) -{ - struct vfio_region_info *info =3D NULL; - bool ret =3D false; - - if (!vfio_get_region_info(vbasedev, region, &info)) { - if (vfio_get_region_info_cap(info, cap_type)) { - ret =3D true; - } - g_free(info); - } - - return ret; -} - /* * Interfaces for IBM EEH (Enhanced Error Handling) */ diff --git a/hw/vfio/helpers.c b/hw/vfio/helpers.c new file mode 100644 index 0000000000..4338456b08 --- /dev/null +++ b/hw/vfio/helpers.c @@ -0,0 +1,598 @@ +/* + * low level and IOMMU backend agnostic helpers used by VFIO devices, + * related to regions, interrupts, capabilities + * + * Copyright Red Hat, Inc. 2012 + * + * Authors: + * Alex Williamson + * + * This work is licensed under the terms of the GNU GPL, version 2. See + * the COPYING file in the top-level directory. + * + * Based on qemu-kvm device-assignment: + * Adapted for KVM by Qumranet. + * Copyright (c) 2007, Neocleus, Alex Novik (alex@neocleus.com) + * Copyright (c) 2007, Neocleus, Guy Zana (guy@neocleus.com) + * Copyright (C) 2008, Qumranet, Amit Shah (amit.shah@qumranet.com) + * Copyright (C) 2008, Red Hat, Amit Shah (amit.shah@redhat.com) + * Copyright (C) 2008, IBM, Muli Ben-Yehuda (muli@il.ibm.com) + */ + +#include "qemu/osdep.h" +#include + +#include "hw/vfio/vfio-common.h" +#include "hw/vfio/vfio.h" +#include "hw/hw.h" +#include "trace.h" +#include "qapi/error.h" + +/* + * Common VFIO interrupt disable + */ +void vfio_disable_irqindex(VFIODevice *vbasedev, int index) +{ + struct vfio_irq_set irq_set =3D { + .argsz =3D sizeof(irq_set), + .flags =3D VFIO_IRQ_SET_DATA_NONE | VFIO_IRQ_SET_ACTION_TRIGGER, + .index =3D index, + .start =3D 0, + .count =3D 0, + }; + + ioctl(vbasedev->fd, VFIO_DEVICE_SET_IRQS, &irq_set); +} + +void vfio_unmask_single_irqindex(VFIODevice *vbasedev, int index) +{ + struct vfio_irq_set irq_set =3D { + .argsz =3D sizeof(irq_set), + .flags =3D VFIO_IRQ_SET_DATA_NONE | VFIO_IRQ_SET_ACTION_UNMASK, + .index =3D index, + .start =3D 0, + .count =3D 1, + }; + + ioctl(vbasedev->fd, VFIO_DEVICE_SET_IRQS, &irq_set); +} + +void vfio_mask_single_irqindex(VFIODevice *vbasedev, int index) +{ + struct vfio_irq_set irq_set =3D { + .argsz =3D sizeof(irq_set), + .flags =3D VFIO_IRQ_SET_DATA_NONE | VFIO_IRQ_SET_ACTION_MASK, + .index =3D index, + .start =3D 0, + .count =3D 1, + }; + + ioctl(vbasedev->fd, VFIO_DEVICE_SET_IRQS, &irq_set); +} + +static inline const char *action_to_str(int action) +{ + switch (action) { + case VFIO_IRQ_SET_ACTION_MASK: + return "MASK"; + case VFIO_IRQ_SET_ACTION_UNMASK: + return "UNMASK"; + case VFIO_IRQ_SET_ACTION_TRIGGER: + return "TRIGGER"; + default: + return "UNKNOWN ACTION"; + } +} + +static const char *index_to_str(VFIODevice *vbasedev, int index) +{ + if (vbasedev->type !=3D VFIO_DEVICE_TYPE_PCI) { + return NULL; + } + + switch (index) { + case VFIO_PCI_INTX_IRQ_INDEX: + return "INTX"; + case VFIO_PCI_MSI_IRQ_INDEX: + return "MSI"; + case VFIO_PCI_MSIX_IRQ_INDEX: + return "MSIX"; + case VFIO_PCI_ERR_IRQ_INDEX: + return "ERR"; + case VFIO_PCI_REQ_IRQ_INDEX: + return "REQ"; + default: + return NULL; + } +} + +int vfio_set_irq_signaling(VFIODevice *vbasedev, int index, int subindex, + int action, int fd, Error **errp) +{ + struct vfio_irq_set *irq_set; + int argsz, ret =3D 0; + const char *name; + int32_t *pfd; + + argsz =3D sizeof(*irq_set) + sizeof(*pfd); + + irq_set =3D g_malloc0(argsz); + irq_set->argsz =3D argsz; + irq_set->flags =3D VFIO_IRQ_SET_DATA_EVENTFD | action; + irq_set->index =3D index; + irq_set->start =3D subindex; + irq_set->count =3D 1; + pfd =3D (int32_t *)&irq_set->data; + *pfd =3D fd; + + if (ioctl(vbasedev->fd, VFIO_DEVICE_SET_IRQS, irq_set)) { + ret =3D -errno; + } + g_free(irq_set); + + if (!ret) { + return 0; + } + + error_setg_errno(errp, -ret, "VFIO_DEVICE_SET_IRQS failure"); + + name =3D index_to_str(vbasedev, index); + if (name) { + error_prepend(errp, "%s-%d: ", name, subindex); + } else { + error_prepend(errp, "index %d-%d: ", index, subindex); + } + error_prepend(errp, + "Failed to %s %s eventfd signaling for interrupt ", + fd < 0 ? "tear down" : "set up", action_to_str(action)); + return ret; +} + +/* + * IO Port/MMIO - Beware of the endians, VFIO is always little endian + */ +void vfio_region_write(void *opaque, hwaddr addr, + uint64_t data, unsigned size) +{ + VFIORegion *region =3D opaque; + VFIODevice *vbasedev =3D region->vbasedev; + union { + uint8_t byte; + uint16_t word; + uint32_t dword; + uint64_t qword; + } buf; + + switch (size) { + case 1: + buf.byte =3D data; + break; + case 2: + buf.word =3D cpu_to_le16(data); + break; + case 4: + buf.dword =3D cpu_to_le32(data); + break; + case 8: + buf.qword =3D cpu_to_le64(data); + break; + default: + hw_error("vfio: unsupported write size, %u bytes", size); + break; + } + + if (pwrite(vbasedev->fd, &buf, size, region->fd_offset + addr) !=3D si= ze) { + error_report("%s(%s:region%d+0x%"HWADDR_PRIx", 0x%"PRIx64 + ",%d) failed: %m", + __func__, vbasedev->name, region->nr, + addr, data, size); + } + + trace_vfio_region_write(vbasedev->name, region->nr, addr, data, size); + + /* + * A read or write to a BAR always signals an INTx EOI. This will + * do nothing if not pending (including not in INTx mode). We assume + * that a BAR access is in response to an interrupt and that BAR + * accesses will service the interrupt. Unfortunately, we don't know + * which access will service the interrupt, so we're potentially + * getting quite a few host interrupts per guest interrupt. + */ + vbasedev->ops->vfio_eoi(vbasedev); +} + +uint64_t vfio_region_read(void *opaque, + hwaddr addr, unsigned size) +{ + VFIORegion *region =3D opaque; + VFIODevice *vbasedev =3D region->vbasedev; + union { + uint8_t byte; + uint16_t word; + uint32_t dword; + uint64_t qword; + } buf; + uint64_t data =3D 0; + + if (pread(vbasedev->fd, &buf, size, region->fd_offset + addr) !=3D siz= e) { + error_report("%s(%s:region%d+0x%"HWADDR_PRIx", %d) failed: %m", + __func__, vbasedev->name, region->nr, + addr, size); + return (uint64_t)-1; + } + switch (size) { + case 1: + data =3D buf.byte; + break; + case 2: + data =3D le16_to_cpu(buf.word); + break; + case 4: + data =3D le32_to_cpu(buf.dword); + break; + case 8: + data =3D le64_to_cpu(buf.qword); + break; + default: + hw_error("vfio: unsupported read size, %u bytes", size); + break; + } + + trace_vfio_region_read(vbasedev->name, region->nr, addr, size, data); + + /* Same as write above */ + vbasedev->ops->vfio_eoi(vbasedev); + + return data; +} + +const MemoryRegionOps vfio_region_ops =3D { + .read =3D vfio_region_read, + .write =3D vfio_region_write, + .endianness =3D DEVICE_LITTLE_ENDIAN, + .valid =3D { + .min_access_size =3D 1, + .max_access_size =3D 8, + }, + .impl =3D { + .min_access_size =3D 1, + .max_access_size =3D 8, + }, +}; + +struct vfio_info_cap_header * +vfio_get_cap(void *ptr, uint32_t cap_offset, uint16_t id) +{ + struct vfio_info_cap_header *hdr; + + for (hdr =3D ptr + cap_offset; hdr !=3D ptr; hdr =3D ptr + hdr->next) { + if (hdr->id =3D=3D id) { + return hdr; + } + } + + return NULL; +} + +struct vfio_info_cap_header * +vfio_get_region_info_cap(struct vfio_region_info *info, uint16_t id) +{ + if (!(info->flags & VFIO_REGION_INFO_FLAG_CAPS)) { + return NULL; + } + + return vfio_get_cap((void *)info, info->cap_offset, id); +} + +struct vfio_info_cap_header * +vfio_get_device_info_cap(struct vfio_device_info *info, uint16_t id) +{ + if (!(info->flags & VFIO_DEVICE_FLAGS_CAPS)) { + return NULL; + } + + return vfio_get_cap((void *)info, info->cap_offset, id); +} + +static int vfio_setup_region_sparse_mmaps(VFIORegion *region, + struct vfio_region_info *info) +{ + struct vfio_info_cap_header *hdr; + struct vfio_region_info_cap_sparse_mmap *sparse; + int i, j; + + hdr =3D vfio_get_region_info_cap(info, VFIO_REGION_INFO_CAP_SPARSE_MMA= P); + if (!hdr) { + return -ENODEV; + } + + sparse =3D container_of(hdr, struct vfio_region_info_cap_sparse_mmap, = header); + + trace_vfio_region_sparse_mmap_header(region->vbasedev->name, + region->nr, sparse->nr_areas); + + region->mmaps =3D g_new0(VFIOMmap, sparse->nr_areas); + + for (i =3D 0, j =3D 0; i < sparse->nr_areas; i++) { + if (sparse->areas[i].size) { + trace_vfio_region_sparse_mmap_entry(i, sparse->areas[i].offset, + sparse->areas[i].offset + + sparse->areas[i].size - 1); + region->mmaps[j].offset =3D sparse->areas[i].offset; + region->mmaps[j].size =3D sparse->areas[i].size; + j++; + } + } + + region->nr_mmaps =3D j; + region->mmaps =3D g_realloc(region->mmaps, j * sizeof(VFIOMmap)); + + return 0; +} + +int vfio_region_setup(Object *obj, VFIODevice *vbasedev, VFIORegion *regio= n, + int index, const char *name) +{ + struct vfio_region_info *info; + int ret; + + ret =3D vfio_get_region_info(vbasedev, index, &info); + if (ret) { + return ret; + } + + region->vbasedev =3D vbasedev; + region->flags =3D info->flags; + region->size =3D info->size; + region->fd_offset =3D info->offset; + region->nr =3D index; + + if (region->size) { + region->mem =3D g_new0(MemoryRegion, 1); + memory_region_init_io(region->mem, obj, &vfio_region_ops, + region, name, region->size); + + if (!vbasedev->no_mmap && + region->flags & VFIO_REGION_INFO_FLAG_MMAP) { + + ret =3D vfio_setup_region_sparse_mmaps(region, info); + + if (ret) { + region->nr_mmaps =3D 1; + region->mmaps =3D g_new0(VFIOMmap, region->nr_mmaps); + region->mmaps[0].offset =3D 0; + region->mmaps[0].size =3D region->size; + } + } + } + + g_free(info); + + trace_vfio_region_setup(vbasedev->name, index, name, + region->flags, region->fd_offset, region->size= ); + return 0; +} + +static void vfio_subregion_unmap(VFIORegion *region, int index) +{ + trace_vfio_region_unmap(memory_region_name(®ion->mmaps[index].mem), + region->mmaps[index].offset, + region->mmaps[index].offset + + region->mmaps[index].size - 1); + memory_region_del_subregion(region->mem, ®ion->mmaps[index].mem); + munmap(region->mmaps[index].mmap, region->mmaps[index].size); + object_unparent(OBJECT(®ion->mmaps[index].mem)); + region->mmaps[index].mmap =3D NULL; +} + +int vfio_region_mmap(VFIORegion *region) +{ + int i, prot =3D 0; + char *name; + + if (!region->mem) { + return 0; + } + + prot |=3D region->flags & VFIO_REGION_INFO_FLAG_READ ? PROT_READ : 0; + prot |=3D region->flags & VFIO_REGION_INFO_FLAG_WRITE ? PROT_WRITE : 0; + + for (i =3D 0; i < region->nr_mmaps; i++) { + region->mmaps[i].mmap =3D mmap(NULL, region->mmaps[i].size, prot, + MAP_SHARED, region->vbasedev->fd, + region->fd_offset + + region->mmaps[i].offset); + if (region->mmaps[i].mmap =3D=3D MAP_FAILED) { + int ret =3D -errno; + + trace_vfio_region_mmap_fault(memory_region_name(region->mem), = i, + region->fd_offset + + region->mmaps[i].offset, + region->fd_offset + + region->mmaps[i].offset + + region->mmaps[i].size - 1, ret); + + region->mmaps[i].mmap =3D NULL; + + for (i--; i >=3D 0; i--) { + vfio_subregion_unmap(region, i); + } + + return ret; + } + + name =3D g_strdup_printf("%s mmaps[%d]", + memory_region_name(region->mem), i); + memory_region_init_ram_device_ptr(®ion->mmaps[i].mem, + memory_region_owner(region->mem), + name, region->mmaps[i].size, + region->mmaps[i].mmap); + g_free(name); + memory_region_add_subregion(region->mem, region->mmaps[i].offset, + ®ion->mmaps[i].mem); + + trace_vfio_region_mmap(memory_region_name(®ion->mmaps[i].mem), + region->mmaps[i].offset, + region->mmaps[i].offset + + region->mmaps[i].size - 1); + } + + return 0; +} + +void vfio_region_unmap(VFIORegion *region) +{ + int i; + + if (!region->mem) { + return; + } + + for (i =3D 0; i < region->nr_mmaps; i++) { + if (region->mmaps[i].mmap) { + vfio_subregion_unmap(region, i); + } + } +} + +void vfio_region_exit(VFIORegion *region) +{ + int i; + + if (!region->mem) { + return; + } + + for (i =3D 0; i < region->nr_mmaps; i++) { + if (region->mmaps[i].mmap) { + memory_region_del_subregion(region->mem, ®ion->mmaps[i].mem= ); + } + } + + trace_vfio_region_exit(region->vbasedev->name, region->nr); +} + +void vfio_region_finalize(VFIORegion *region) +{ + int i; + + if (!region->mem) { + return; + } + + for (i =3D 0; i < region->nr_mmaps; i++) { + if (region->mmaps[i].mmap) { + munmap(region->mmaps[i].mmap, region->mmaps[i].size); + object_unparent(OBJECT(®ion->mmaps[i].mem)); + } + } + + object_unparent(OBJECT(region->mem)); + + g_free(region->mem); + g_free(region->mmaps); + + trace_vfio_region_finalize(region->vbasedev->name, region->nr); + + region->mem =3D NULL; + region->mmaps =3D NULL; + region->nr_mmaps =3D 0; + region->size =3D 0; + region->flags =3D 0; + region->nr =3D 0; +} + +void vfio_region_mmaps_set_enabled(VFIORegion *region, bool enabled) +{ + int i; + + if (!region->mem) { + return; + } + + for (i =3D 0; i < region->nr_mmaps; i++) { + if (region->mmaps[i].mmap) { + memory_region_set_enabled(®ion->mmaps[i].mem, enabled); + } + } + + trace_vfio_region_mmaps_set_enabled(memory_region_name(region->mem), + enabled); +} + +int vfio_get_region_info(VFIODevice *vbasedev, int index, + struct vfio_region_info **info) +{ + size_t argsz =3D sizeof(struct vfio_region_info); + + *info =3D g_malloc0(argsz); + + (*info)->index =3D index; +retry: + (*info)->argsz =3D argsz; + + if (ioctl(vbasedev->fd, VFIO_DEVICE_GET_REGION_INFO, *info)) { + g_free(*info); + *info =3D NULL; + return -errno; + } + + if ((*info)->argsz > argsz) { + argsz =3D (*info)->argsz; + *info =3D g_realloc(*info, argsz); + + goto retry; + } + + return 0; +} + +int vfio_get_dev_region_info(VFIODevice *vbasedev, uint32_t type, + uint32_t subtype, struct vfio_region_info **i= nfo) +{ + int i; + + for (i =3D 0; i < vbasedev->num_regions; i++) { + struct vfio_info_cap_header *hdr; + struct vfio_region_info_cap_type *cap_type; + + if (vfio_get_region_info(vbasedev, i, info)) { + continue; + } + + hdr =3D vfio_get_region_info_cap(*info, VFIO_REGION_INFO_CAP_TYPE); + if (!hdr) { + g_free(*info); + continue; + } + + cap_type =3D container_of(hdr, struct vfio_region_info_cap_type, h= eader); + + trace_vfio_get_dev_region(vbasedev->name, i, + cap_type->type, cap_type->subtype); + + if (cap_type->type =3D=3D type && cap_type->subtype =3D=3D subtype= ) { + return 0; + } + + g_free(*info); + } + + *info =3D NULL; + return -ENODEV; +} + +bool vfio_has_region_cap(VFIODevice *vbasedev, int region, uint16_t cap_ty= pe) +{ + struct vfio_region_info *info =3D NULL; + bool ret =3D false; + + if (!vfio_get_region_info(vbasedev, region, &info)) { + if (vfio_get_region_info_cap(info, cap_type)) { + ret =3D true; + } + g_free(info); + } + + return ret; +} diff --git a/hw/vfio/meson.build b/hw/vfio/meson.build index da9af297a0..3746c9f984 100644 --- a/hw/vfio/meson.build +++ b/hw/vfio/meson.build @@ -1,5 +1,6 @@ vfio_ss =3D ss.source_set() vfio_ss.add(files( + 'helpers.c', 'common.c', 'spapr.c', 'migration.c', diff --git a/include/hw/vfio/vfio-common.h b/include/hw/vfio/vfio-common.h index da43d27352..5e376c436e 100644 --- a/include/hw/vfio/vfio-common.h +++ b/include/hw/vfio/vfio-common.h @@ -243,6 +243,8 @@ bool vfio_get_info_dma_avail(struct vfio_iommu_type1_in= fo *info, unsigned int *avail); struct vfio_info_cap_header * vfio_get_device_info_cap(struct vfio_device_info *info, uint16_t id); +struct vfio_info_cap_header * +vfio_get_cap(void *ptr, uint32_t cap_offset, uint16_t id); #endif extern const MemoryListener vfio_prereg_listener; =20 --=20 2.34.1 From nobody Wed May 15 22:12:51 2024 Delivered-To: importer@patchew.org Authentication-Results: mx.zohomail.com; dkim=pass header.i=@intel.com; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass(p=none dis=none) header.from=intel.com ARC-Seal: i=1; a=rsa-sha256; t=1693392848; cv=none; d=zohomail.com; s=zohoarc; b=Il3wSqsfNjkKhmoFrS+9u0PwStusFfhDIrok7S8c/q3NZvk16J3nNxT4z8hO0T0qTBWxcVfuPHvBDayGnNfJ/mLWtdwMfH+kX478MYGzhdmkVUlrcqy3d1yiz2sUFP96zBlwphgHwoMHUMqStsAUT6d+eLc11kCLlGuDg+7zGXE= ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=zohomail.com; s=zohoarc; t=1693392848; h=Content-Transfer-Encoding:Cc:Date:From:In-Reply-To:List-Subscribe:List-Post:List-Id:List-Archive:List-Help:List-Unsubscribe:MIME-Version:Message-ID:References:Sender:Subject:To; bh=GpGEyyrqHim/6xMp7yrszzP8BUA5F0FUv1rNELTy6Lw=; b=VEg1cyJOWgzUsh/4oyRulA6nPU3P0bHs2qNspk/5iVmYS9G5mF33/8UgrmFuY6HUwEdTfyxanLLmli1Xe19Op9z9jNAL9P5B47ymu3BTjowauk3WUNcFfq8+i/rpdUZYi+nl92R9M58/MriXHpGqOni+IgbGkhTetjmvmdgJvgY= ARC-Authentication-Results: i=1; mx.zohomail.com; dkim=pass header.i=@intel.com; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass header.from= (p=none dis=none) Return-Path: Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) by mx.zohomail.com with SMTPS id 169339284897646.74597905782514; Wed, 30 Aug 2023 03:54:08 -0700 (PDT) Received: from localhost ([::1] helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1qbIoy-0004FA-Ux; Wed, 30 Aug 2023 06:53:12 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1qbIow-0004EQ-AL for qemu-devel@nongnu.org; Wed, 30 Aug 2023 06:53:10 -0400 Received: from mgamail.intel.com ([192.55.52.93]) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1qbIos-0007fH-R3 for qemu-devel@nongnu.org; Wed, 30 Aug 2023 06:53:10 -0400 Received: from orsmga006.jf.intel.com ([10.7.209.51]) by fmsmga102.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 30 Aug 2023 03:52:47 -0700 Received: from duan-server-s2600bt.bj.intel.com ([10.240.192.147]) by orsmga006-auth.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 30 Aug 2023 03:52:43 -0700 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1693392786; x=1724928786; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=AZlTzsz7sjLkxbo98T5TeReGMRoYap7+PFgc2Cox5bQ=; b=bUR8tadD8mVE1sYumayC7hDdCQitO8mEqprTep6kW+FThFwmmxtatejC 0qNwxkCdFGj7k9atkOkiXfqtpqe9C5OcvpGJByD1lbWRprGaGbgFH8TXa Oz4E7kWmhRWRlTWSYPDE5mYdQaOqyghsWQ96RFCDHv3ye4EmaNAX8KGIA Wx1NMHsAWPlvoxHI7cEmYLNnDC/ZpaSqHbyOMIyMEbxWdZCBZYkCAN9dK p3a8bbsxMHgmJ3KcwgKIFwepMiPbgLs6nqz+sw+ltSZGbLkORp6uO4nBx KYW/fQZTDZOGVEv1Oa58pHPXDZzRrFcUXh1zmSAlJ5vUReIkpO3Z6+ta7 g==; X-IronPort-AV: E=McAfee;i="6600,9927,10817"; a="373016403" X-IronPort-AV: E=Sophos;i="6.02,213,1688454000"; d="scan'208";a="373016403" X-ExtLoop1: 1 X-IronPort-AV: E=McAfee;i="6600,9927,10817"; a="715866051" X-IronPort-AV: E=Sophos;i="6.02,213,1688454000"; d="scan'208";a="715866051" From: Zhenzhong Duan To: qemu-devel@nongnu.org Cc: alex.williamson@redhat.com, clg@redhat.com, jgg@nvidia.com, nicolinc@nvidia.com, joao.m.martins@oracle.com, eric.auger@redhat.com, peterx@redhat.com, jasowang@redhat.com, kevin.tian@intel.com, yi.l.liu@intel.com, yi.y.sun@intel.com, chao.p.peng@intel.com, Zhenzhong Duan Subject: [PATCH v1 04/22] vfio/common: Introduce vfio_container_add|del_section_window() Date: Wed, 30 Aug 2023 18:37:36 +0800 Message-Id: <20230830103754.36461-5-zhenzhong.duan@intel.com> X-Mailer: git-send-email 2.34.1 In-Reply-To: <20230830103754.36461-1-zhenzhong.duan@intel.com> References: <20230830103754.36461-1-zhenzhong.duan@intel.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Received-SPF: pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) client-ip=209.51.188.17; envelope-from=qemu-devel-bounces+importer=patchew.org@nongnu.org; helo=lists.gnu.org; Received-SPF: pass client-ip=192.55.52.93; envelope-from=zhenzhong.duan@intel.com; helo=mgamail.intel.com X-Spam_score_int: -43 X-Spam_score: -4.4 X-Spam_bar: ---- X-Spam_report: (-4.4 / 5.0 requ) BAYES_00=-1.9, DKIMWL_WL_HIGH=-0.001, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, RCVD_IN_DNSWL_MED=-2.3, SPF_HELO_NONE=0.001, SPF_PASS=-0.001 autolearn=ham autolearn_force=no X-Spam_action: no action X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: qemu-devel-bounces+importer=patchew.org@nongnu.org Sender: qemu-devel-bounces+importer=patchew.org@nongnu.org X-ZohoMail-DKIM: pass (identity @intel.com) X-ZM-MESSAGEID: 1693392933425100001 Content-Type: text/plain; charset="utf-8" From: Eric Auger Introduce helper functions that isolate the code used for VFIO_SPAPR_TCE_v2_IOMMU. This code reliance is IOMMU backend specific whereas the rest of the code in the callers, ie. vfio_listener_region_add|del is not. Signed-off-by: Eric Auger Signed-off-by: Yi Liu Signed-off-by: Zhenzhong Duan --- hw/vfio/common.c | 156 +++++++++++++++++++++++++++-------------------- 1 file changed, 89 insertions(+), 67 deletions(-) diff --git a/hw/vfio/common.c b/hw/vfio/common.c index 9ca695837f..67150e4575 100644 --- a/hw/vfio/common.c +++ b/hw/vfio/common.c @@ -796,6 +796,92 @@ static bool vfio_get_section_iova_range(VFIOContainer = *container, return true; } =20 +static int vfio_container_add_section_window(VFIOContainer *container, + MemoryRegionSection *section, + Error **errp) +{ + VFIOHostDMAWindow *hostwin; + hwaddr pgsize =3D 0; + int ret; + + if (container->iommu_type !=3D VFIO_SPAPR_TCE_v2_IOMMU) { + return 0; + } + + /* For now intersections are not allowed, we may relax this later */ + QLIST_FOREACH(hostwin, &container->hostwin_list, hostwin_next) { + if (ranges_overlap(hostwin->min_iova, + hostwin->max_iova - hostwin->min_iova + 1, + section->offset_within_address_space, + int128_get64(section->size))) { + error_setg(errp, + "region [0x%"PRIx64",0x%"PRIx64"] overlaps with existing" + "host DMA window [0x%"PRIx64",0x%"PRIx64"]", + section->offset_within_address_space, + section->offset_within_address_space + + int128_get64(section->size) - 1, + hostwin->min_iova, hostwin->max_iova); + return -EINVAL; + } + } + + ret =3D vfio_spapr_create_window(container, section, &pgsize); + if (ret) { + error_setg_errno(errp, -ret, "Failed to create SPAPR window"); + return ret; + } + + vfio_host_win_add(container, section->offset_within_address_space, + section->offset_within_address_space + + int128_get64(section->size) - 1, pgsize); +#ifdef CONFIG_KVM + if (kvm_enabled()) { + VFIOGroup *group; + IOMMUMemoryRegion *iommu_mr =3D IOMMU_MEMORY_REGION(section->mr); + struct kvm_vfio_spapr_tce param; + struct kvm_device_attr attr =3D { + .group =3D KVM_DEV_VFIO_GROUP, + .attr =3D KVM_DEV_VFIO_GROUP_SET_SPAPR_TCE, + .addr =3D (uint64_t)(unsigned long)¶m, + }; + + if (!memory_region_iommu_get_attr(iommu_mr, IOMMU_ATTR_SPAPR_TCE_F= D, + ¶m.tablefd)) { + QLIST_FOREACH(group, &container->group_list, container_next) { + param.groupfd =3D group->fd; + if (ioctl(vfio_kvm_device_fd, KVM_SET_DEVICE_ATTR, &attr))= { + error_report("vfio: failed to setup fd %d " + "for a group with fd %d: %s", + param.tablefd, param.groupfd, + strerror(errno)); + return 0; + } + trace_vfio_spapr_group_attach(param.groupfd, param.tablefd= ); + } + } + } +#endif + return 0; +} + +static void vfio_container_del_section_window(VFIOContainer *container, + MemoryRegionSection *section) +{ + if (container->iommu_type !=3D VFIO_SPAPR_TCE_v2_IOMMU) { + return; + } + + vfio_spapr_remove_window(container, + section->offset_within_address_space); + if (vfio_host_win_del(container, + section->offset_within_address_space, + section->offset_within_address_space + + int128_get64(section->size) - 1) < 0) { + hw_error("%s: Cannot delete missing window at %"HWADDR_PRIx, + __func__, section->offset_within_address_space); + } +} + static void vfio_listener_region_add(MemoryListener *listener, MemoryRegionSection *section) { @@ -822,62 +908,8 @@ static void vfio_listener_region_add(MemoryListener *l= istener, return; } =20 - if (container->iommu_type =3D=3D VFIO_SPAPR_TCE_v2_IOMMU) { - hwaddr pgsize =3D 0; - - /* For now intersections are not allowed, we may relax this later = */ - QLIST_FOREACH(hostwin, &container->hostwin_list, hostwin_next) { - if (ranges_overlap(hostwin->min_iova, - hostwin->max_iova - hostwin->min_iova + 1, - section->offset_within_address_space, - int128_get64(section->size))) { - error_setg(&err, - "region [0x%"PRIx64",0x%"PRIx64"] overlaps with existi= ng" - "host DMA window [0x%"PRIx64",0x%"PRIx64"]", - section->offset_within_address_space, - section->offset_within_address_space + - int128_get64(section->size) - 1, - hostwin->min_iova, hostwin->max_iova); - goto fail; - } - } - - ret =3D vfio_spapr_create_window(container, section, &pgsize); - if (ret) { - error_setg_errno(&err, -ret, "Failed to create SPAPR window"); - goto fail; - } - - vfio_host_win_add(container, section->offset_within_address_space, - section->offset_within_address_space + - int128_get64(section->size) - 1, pgsize); -#ifdef CONFIG_KVM - if (kvm_enabled()) { - VFIOGroup *group; - IOMMUMemoryRegion *iommu_mr =3D IOMMU_MEMORY_REGION(section->m= r); - struct kvm_vfio_spapr_tce param; - struct kvm_device_attr attr =3D { - .group =3D KVM_DEV_VFIO_GROUP, - .attr =3D KVM_DEV_VFIO_GROUP_SET_SPAPR_TCE, - .addr =3D (uint64_t)(unsigned long)¶m, - }; - - if (!memory_region_iommu_get_attr(iommu_mr, IOMMU_ATTR_SPAPR_T= CE_FD, - ¶m.tablefd)) { - QLIST_FOREACH(group, &container->group_list, container_nex= t) { - param.groupfd =3D group->fd; - if (ioctl(vfio_kvm_device_fd, KVM_SET_DEVICE_ATTR, &at= tr)) { - error_report("vfio: failed to setup fd %d " - "for a group with fd %d: %s", - param.tablefd, param.groupfd, - strerror(errno)); - return; - } - trace_vfio_spapr_group_attach(param.groupfd, param.tab= lefd); - } - } - } -#endif + if (vfio_container_add_section_window(container, section, &err)) { + goto fail; } =20 hostwin =3D vfio_find_hostwin(container, iova, end); @@ -1094,17 +1126,7 @@ static void vfio_listener_region_del(MemoryListener = *listener, =20 memory_region_unref(section->mr); =20 - if (container->iommu_type =3D=3D VFIO_SPAPR_TCE_v2_IOMMU) { - vfio_spapr_remove_window(container, - section->offset_within_address_space); - if (vfio_host_win_del(container, - section->offset_within_address_space, - section->offset_within_address_space + - int128_get64(section->size) - 1) < 0) { - hw_error("%s: Cannot delete missing window at %"HWADDR_PRIx, - __func__, section->offset_within_address_space); - } - } + vfio_container_del_section_window(container, section); } =20 static int vfio_set_dirty_page_tracking(VFIOContainer *container, bool sta= rt) --=20 2.34.1 From nobody Wed May 15 22:12:51 2024 Delivered-To: importer@patchew.org Authentication-Results: mx.zohomail.com; dkim=pass header.i=@intel.com; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass(p=none dis=none) header.from=intel.com ARC-Seal: i=1; a=rsa-sha256; t=1693392917; cv=none; d=zohomail.com; s=zohoarc; b=Q5fh8vBXqtHhAbm/C8TiD+GFpJbkr4zsnnZJGAmVRsufQilCiCeHRmvkeW0xCgCPO0kxzGau6g7vrtdnfS5snuhnNcmX19h7KHXy3dyoS9zpxRRENVBDUO1i2LCrg2yiuWluj1oNmP2TQy3Xiyab1TjTahFSrJGe5YmnqMwx5j4= ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=zohomail.com; s=zohoarc; t=1693392917; h=Content-Transfer-Encoding:Cc:Date:From:In-Reply-To:List-Subscribe:List-Post:List-Id:List-Archive:List-Help:List-Unsubscribe:MIME-Version:Message-ID:References:Sender:Subject:To; bh=oz1zT5HvIcNhNlD8V+S8K/ZakZGZFYpsK3fQHKbFc1g=; b=F6wfFECw3Pu3q5o5WkXSuPsd4nV70S86kUOhkc0SnON8I8NpQ6/efpywXzrgs6GStk3QEts83q77mSbJqlUN4Te6s1FLFSPR8BrO2iGZ947LilaCM5yWAcA6otP5BF2+Q5Lo9jmUyClov73mfFIzbB1pMa7sTkhn+t4dB0rWaYs= ARC-Authentication-Results: i=1; mx.zohomail.com; dkim=pass header.i=@intel.com; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass header.from= (p=none dis=none) Return-Path: Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) by mx.zohomail.com with SMTPS id 1693392917658462.10594103758524; Wed, 30 Aug 2023 03:55:17 -0700 (PDT) Received: from localhost ([::1] helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1qbIou-0004Di-9A; Wed, 30 Aug 2023 06:53:08 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1qbIos-0004DL-9K for qemu-devel@nongnu.org; Wed, 30 Aug 2023 06:53:06 -0400 Received: from mgamail.intel.com ([192.55.52.93]) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1qbIop-0007eq-7q for qemu-devel@nongnu.org; Wed, 30 Aug 2023 06:53:06 -0400 Received: from orsmga006.jf.intel.com ([10.7.209.51]) by fmsmga102.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 30 Aug 2023 03:52:52 -0700 Received: from duan-server-s2600bt.bj.intel.com ([10.240.192.147]) by orsmga006-auth.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 30 Aug 2023 03:52:47 -0700 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1693392783; x=1724928783; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=D9aTIKHB/HvRfwXvd1jYJrNHVmQZAyYwCbrHCKAztgM=; b=ZP7XwARG085Iimh5Lezh2D784VLtHwq12G3TbwEbQWKXFh4UDO/XLn9s B/vtl01yX6dzj2cYIchROkbbOG2MKk4qPDwdKHtFSg5gu1Uxc60e9WLDl uO2u0nnnNvqLM9KEjaTBQtofyA1bi4NWXhnqW1JPHHor80XpXK70+F4SI o2tdHg909kEsU8r2PwYta98QjwqyCKAsVkcnwys3RKm2S8yOMziQxNyAu RclNfqaRSAXt0boB5uHXqtR/8wczR+bOQJTdoVRFxRM/a65yZzTCUWJUo rvkcVkGVfhT3GxjeHVJWYyvoQvR+WOacrvp7VxpkzHFWUG03Fw0XMYXOZ A==; X-IronPort-AV: E=McAfee;i="6600,9927,10817"; a="373016421" X-IronPort-AV: E=Sophos;i="6.02,213,1688454000"; d="scan'208";a="373016421" X-ExtLoop1: 1 X-IronPort-AV: E=McAfee;i="6600,9927,10817"; a="715866064" X-IronPort-AV: E=Sophos;i="6.02,213,1688454000"; d="scan'208";a="715866064" From: Zhenzhong Duan To: qemu-devel@nongnu.org Cc: alex.williamson@redhat.com, clg@redhat.com, jgg@nvidia.com, nicolinc@nvidia.com, joao.m.martins@oracle.com, eric.auger@redhat.com, peterx@redhat.com, jasowang@redhat.com, kevin.tian@intel.com, yi.l.liu@intel.com, yi.y.sun@intel.com, chao.p.peng@intel.com, Zhenzhong Duan Subject: [PATCH v1 05/22] vfio/common: Extract out vfio_kvm_device_[add/del]_fd Date: Wed, 30 Aug 2023 18:37:37 +0800 Message-Id: <20230830103754.36461-6-zhenzhong.duan@intel.com> X-Mailer: git-send-email 2.34.1 In-Reply-To: <20230830103754.36461-1-zhenzhong.duan@intel.com> References: <20230830103754.36461-1-zhenzhong.duan@intel.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Received-SPF: pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) client-ip=209.51.188.17; envelope-from=qemu-devel-bounces+importer=patchew.org@nongnu.org; helo=lists.gnu.org; Received-SPF: pass client-ip=192.55.52.93; envelope-from=zhenzhong.duan@intel.com; helo=mgamail.intel.com X-Spam_score_int: -43 X-Spam_score: -4.4 X-Spam_bar: ---- X-Spam_report: (-4.4 / 5.0 requ) BAYES_00=-1.9, DKIMWL_WL_HIGH=-0.001, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, RCVD_IN_DNSWL_MED=-2.3, SPF_HELO_NONE=0.001, SPF_PASS=-0.001 autolearn=ham autolearn_force=no X-Spam_action: no action X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: qemu-devel-bounces+importer=patchew.org@nongnu.org Sender: qemu-devel-bounces+importer=patchew.org@nongnu.org X-ZohoMail-DKIM: pass (identity @intel.com) X-ZM-MESSAGEID: 1693392919319100003 Content-Type: text/plain; charset="utf-8" ...which will be used by both legacy and iommufd backend. Signed-off-by: Yi Liu Signed-off-by: Zhenzhong Duan --- hw/vfio/common.c | 44 +++++++++++++++++++++++------------ include/hw/vfio/vfio-common.h | 3 +++ 2 files changed, 32 insertions(+), 15 deletions(-) diff --git a/hw/vfio/common.c b/hw/vfio/common.c index 67150e4575..949ad6714a 100644 --- a/hw/vfio/common.c +++ b/hw/vfio/common.c @@ -1759,17 +1759,17 @@ void vfio_reset_handler(void *opaque) } } =20 -static void vfio_kvm_device_add_group(VFIOGroup *group) +int vfio_kvm_device_add_fd(int fd) { #ifdef CONFIG_KVM struct kvm_device_attr attr =3D { - .group =3D KVM_DEV_VFIO_GROUP, - .attr =3D KVM_DEV_VFIO_GROUP_ADD, - .addr =3D (uint64_t)(unsigned long)&group->fd, + .group =3D KVM_DEV_VFIO_FILE, + .attr =3D KVM_DEV_VFIO_FILE_ADD, + .addr =3D (uint64_t)(unsigned long)&fd, }; =20 if (!kvm_enabled()) { - return; + return 0; } =20 if (vfio_kvm_device_fd < 0) { @@ -1779,37 +1779,51 @@ static void vfio_kvm_device_add_group(VFIOGroup *gr= oup) =20 if (kvm_vm_ioctl(kvm_state, KVM_CREATE_DEVICE, &cd)) { error_report("Failed to create KVM VFIO device: %m"); - return; + return -ENODEV; } =20 vfio_kvm_device_fd =3D cd.fd; } =20 if (ioctl(vfio_kvm_device_fd, KVM_SET_DEVICE_ATTR, &attr)) { - error_report("Failed to add group %d to KVM VFIO device: %m", - group->groupid); + error_report("Failed to add fd %d to KVM VFIO device: %m", + fd); + return -errno; } #endif + return 0; } =20 -static void vfio_kvm_device_del_group(VFIOGroup *group) +static void vfio_kvm_device_add_group(VFIOGroup *group) +{ + vfio_kvm_device_add_fd(group->fd); +} + +int vfio_kvm_device_del_fd(int fd) { #ifdef CONFIG_KVM struct kvm_device_attr attr =3D { - .group =3D KVM_DEV_VFIO_GROUP, - .attr =3D KVM_DEV_VFIO_GROUP_DEL, - .addr =3D (uint64_t)(unsigned long)&group->fd, + .group =3D KVM_DEV_VFIO_FILE, + .attr =3D KVM_DEV_VFIO_FILE_DEL, + .addr =3D (uint64_t)(unsigned long)&fd, }; =20 if (vfio_kvm_device_fd < 0) { - return; + return -EINVAL; } =20 if (ioctl(vfio_kvm_device_fd, KVM_SET_DEVICE_ATTR, &attr)) { - error_report("Failed to remove group %d from KVM VFIO device: %m", - group->groupid); + error_report("Failed to remove fd %d from KVM VFIO device: %m", + fd); + return -EBADF; } #endif + return 0; +} + +static void vfio_kvm_device_del_group(VFIOGroup *group) +{ + vfio_kvm_device_del_fd(group->fd); } =20 static VFIOAddressSpace *vfio_get_address_space(AddressSpace *as) diff --git a/include/hw/vfio/vfio-common.h b/include/hw/vfio/vfio-common.h index 5e376c436e..598c3ce079 100644 --- a/include/hw/vfio/vfio-common.h +++ b/include/hw/vfio/vfio-common.h @@ -220,6 +220,9 @@ struct vfio_device_info *vfio_get_device_info(int fd); int vfio_get_device(VFIOGroup *group, const char *name, VFIODevice *vbasedev, Error **errp); =20 +int vfio_kvm_device_add_fd(int fd); +int vfio_kvm_device_del_fd(int fd); + extern const MemoryRegionOps vfio_region_ops; typedef QLIST_HEAD(VFIOGroupList, VFIOGroup) VFIOGroupList; extern VFIOGroupList vfio_group_list; --=20 2.34.1 From nobody Wed May 15 22:12:51 2024 Delivered-To: importer@patchew.org Authentication-Results: mx.zohomail.com; dkim=pass header.i=@intel.com; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass(p=none dis=none) header.from=intel.com ARC-Seal: i=1; a=rsa-sha256; t=1693392867; cv=none; d=zohomail.com; s=zohoarc; b=S7uemzkgsXSB4fVXStQRTDQQU4/bb4Y2i18cdhZoCm4CCI5bBz2HpVt1cT3YD4q2VPL2C0zb2eihQm94Cm+hooboIGZVaPhchyWExjco+F6GSWxLP+oHu0as2I+DDz1gYRtu8NjC8F7hueg683IOkJd6Uliw2ZUyr+Fevsf8jcE= ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=zohomail.com; s=zohoarc; t=1693392867; h=Content-Transfer-Encoding:Cc:Date:From:In-Reply-To:List-Subscribe:List-Post:List-Id:List-Archive:List-Help:List-Unsubscribe:MIME-Version:Message-ID:References:Sender:Subject:To; bh=dhgadYjQ009+ak/lWfl0E67lfD+VUvVFW5N/ZZP7S2Y=; b=Mr+elQDQZqYV2rHs2rap41Pn2liUaxHU/eFWfPzYh6YnmBBp+f8FflAGhV4QeRlU5lVdvaFIvz+aVS+B/A21KygTlqh8YA+UuUa2nMh9kkDMaSwEyyCOEosQAj9j9IzEF0jrgMnU1+3l7+IT6C5aJy8tLvsUR+8hLwFW51wtfsQ= ARC-Authentication-Results: i=1; mx.zohomail.com; dkim=pass header.i=@intel.com; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass header.from= (p=none dis=none) Return-Path: Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) by mx.zohomail.com with SMTPS id 1693392866872178.016284921982; Wed, 30 Aug 2023 03:54:26 -0700 (PDT) Received: from localhost ([::1] helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1qbIp0-0004FP-GJ; Wed, 30 Aug 2023 06:53:14 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1qbIoz-0004FC-2F for qemu-devel@nongnu.org; Wed, 30 Aug 2023 06:53:13 -0400 Received: from mgamail.intel.com ([192.55.52.93]) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1qbIou-0007ew-IW for qemu-devel@nongnu.org; Wed, 30 Aug 2023 06:53:12 -0400 Received: from orsmga006.jf.intel.com ([10.7.209.51]) by fmsmga102.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 30 Aug 2023 03:52:54 -0700 Received: from duan-server-s2600bt.bj.intel.com ([10.240.192.147]) by orsmga006-auth.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 30 Aug 2023 03:52:51 -0700 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1693392788; x=1724928788; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=xYI/at+9CgjXgWNnHqSRpor8zU7rItGxGWW4cdE/7ec=; b=GDEsNbyObUO68SVNr4cnwVfxWVjbarSoGNW3M7mU/0OA5YZrWruOlYPO 4q99y7/EZz0Ko8QzVAhdTLwTeWYnB9NRmBCi8UQ3JTxRYyYQftB4khAJG 3eHFLJAGU46Eni1v2WpAvw6GP1uSJqzFYYGr6CBqUR1Mh36zKPCOtpvhc hTOFAenDhqUI5GwXD27u8YbA6KckTod0G8sFEXvpmYtHpiafAXyrrs+mt Une9mTc817nvToE7Ua3T9HdyL28Bd1BEBgRqDRTqSuWHqqg6XlFgQwjFi +SvAl1iF7k1ZgZqO2D4MckY6pYIhsW+xWHFQqjWa26oxZ2XQ+6LPOLwVt g==; X-IronPort-AV: E=McAfee;i="6600,9927,10817"; a="373016441" X-IronPort-AV: E=Sophos;i="6.02,213,1688454000"; d="scan'208";a="373016441" X-ExtLoop1: 1 X-IronPort-AV: E=McAfee;i="6600,9927,10817"; a="715866070" X-IronPort-AV: E=Sophos;i="6.02,213,1688454000"; d="scan'208";a="715866070" From: Zhenzhong Duan To: qemu-devel@nongnu.org Cc: alex.williamson@redhat.com, clg@redhat.com, jgg@nvidia.com, nicolinc@nvidia.com, joao.m.martins@oracle.com, eric.auger@redhat.com, peterx@redhat.com, jasowang@redhat.com, kevin.tian@intel.com, yi.l.liu@intel.com, yi.y.sun@intel.com, chao.p.peng@intel.com, Zhenzhong Duan Subject: [PATCH v1 06/22] vfio/common: Add a vfio device iterator Date: Wed, 30 Aug 2023 18:37:38 +0800 Message-Id: <20230830103754.36461-7-zhenzhong.duan@intel.com> X-Mailer: git-send-email 2.34.1 In-Reply-To: <20230830103754.36461-1-zhenzhong.duan@intel.com> References: <20230830103754.36461-1-zhenzhong.duan@intel.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Received-SPF: pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) client-ip=209.51.188.17; envelope-from=qemu-devel-bounces+importer=patchew.org@nongnu.org; helo=lists.gnu.org; Received-SPF: pass client-ip=192.55.52.93; envelope-from=zhenzhong.duan@intel.com; helo=mgamail.intel.com X-Spam_score_int: -43 X-Spam_score: -4.4 X-Spam_bar: ---- X-Spam_report: (-4.4 / 5.0 requ) BAYES_00=-1.9, DKIMWL_WL_HIGH=-0.001, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, RCVD_IN_DNSWL_MED=-2.3, SPF_HELO_NONE=0.001, SPF_PASS=-0.001 autolearn=ham autolearn_force=no X-Spam_action: no action X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: qemu-devel-bounces+importer=patchew.org@nongnu.org Sender: qemu-devel-bounces+importer=patchew.org@nongnu.org X-ZohoMail-DKIM: pass (identity @intel.com) X-ZM-MESSAGEID: 1693392938192100007 Content-Type: text/plain; charset="utf-8" With a vfio device iterator added, we can make some migration and reset related functions group agnostic. E.x: vfio_mig_active vfio_migratable_device_num vfio_devices_all_dirty_tracking vfio_devices_all_device_dirty_tracking vfio_devices_all_running_and_mig_active vfio_devices_dma_logging_stop vfio_devices_dma_logging_start vfio_devices_query_dirty_bitmap vfio_reset_handler Or else we need to add container specific callback variants for above functions just because they iterate devices based on group. Move the reset handler registration/unregistration to a place that is not group specific, saying first vfio address space created instead of the first group. Signed-off-by: Zhenzhong Duan --- hw/vfio/common.c | 224 ++++++++++++++++++++++++++--------------------- 1 file changed, 122 insertions(+), 102 deletions(-) diff --git a/hw/vfio/common.c b/hw/vfio/common.c index 949ad6714a..51c6e7598e 100644 --- a/hw/vfio/common.c +++ b/hw/vfio/common.c @@ -84,6 +84,26 @@ static int vfio_ram_block_discard_disable(VFIOContainer = *container, bool state) } } =20 +static VFIODevice *vfio_container_dev_iter_next(VFIOContainer *container, + VFIODevice *curr) +{ + VFIOGroup *group; + + if (!curr) { + group =3D QLIST_FIRST(&container->group_list); + } else { + if (curr->next.le_next) { + return curr->next.le_next; + } + group =3D curr->group->container_next.le_next; + } + + if (!group) { + return NULL; + } + return QLIST_FIRST(&group->device_list); +} + /* * Device state interfaces */ @@ -112,17 +132,22 @@ static int vfio_get_dirty_bitmap(VFIOContainer *conta= iner, uint64_t iova, =20 bool vfio_mig_active(void) { - VFIOGroup *group; + VFIOAddressSpace *space; + VFIOContainer *container; VFIODevice *vbasedev; =20 - if (QLIST_EMPTY(&vfio_group_list)) { + if (QLIST_EMPTY(&vfio_address_spaces)) { return false; } =20 - QLIST_FOREACH(group, &vfio_group_list, next) { - QLIST_FOREACH(vbasedev, &group->device_list, next) { - if (vbasedev->migration_blocker) { - return false; + QLIST_FOREACH(space, &vfio_address_spaces, list) { + QLIST_FOREACH(container, &space->containers, next) { + vbasedev =3D NULL; + while ((vbasedev =3D vfio_container_dev_iter_next(container, + vbasedev))) { + if (vbasedev->migration_blocker) { + return false; + } } } } @@ -133,14 +158,19 @@ static Error *multiple_devices_migration_blocker; =20 static unsigned int vfio_migratable_device_num(void) { - VFIOGroup *group; + VFIOAddressSpace *space; + VFIOContainer *container; VFIODevice *vbasedev; unsigned int device_num =3D 0; =20 - QLIST_FOREACH(group, &vfio_group_list, next) { - QLIST_FOREACH(vbasedev, &group->device_list, next) { - if (vbasedev->migration) { - device_num++; + QLIST_FOREACH(space, &vfio_address_spaces, list) { + QLIST_FOREACH(container, &space->containers, next) { + vbasedev =3D NULL; + while ((vbasedev =3D vfio_container_dev_iter_next(container, + vbasedev))) { + if (vbasedev->migration) { + device_num++; + } } } } @@ -207,8 +237,7 @@ static void vfio_set_migration_error(int err) =20 static bool vfio_devices_all_dirty_tracking(VFIOContainer *container) { - VFIOGroup *group; - VFIODevice *vbasedev; + VFIODevice *vbasedev =3D NULL; MigrationState *ms =3D migrate_get_current(); =20 if (ms->state !=3D MIGRATION_STATUS_ACTIVE && @@ -216,19 +245,17 @@ static bool vfio_devices_all_dirty_tracking(VFIOConta= iner *container) return false; } =20 - QLIST_FOREACH(group, &container->group_list, container_next) { - QLIST_FOREACH(vbasedev, &group->device_list, next) { - VFIOMigration *migration =3D vbasedev->migration; + while ((vbasedev =3D vfio_container_dev_iter_next(container, vbasedev)= )) { + VFIOMigration *migration =3D vbasedev->migration; =20 - if (!migration) { - return false; - } + if (!migration) { + return false; + } =20 - if (vbasedev->pre_copy_dirty_page_tracking =3D=3D ON_OFF_AUTO_= OFF && - (migration->device_state =3D=3D VFIO_DEVICE_STATE_RUNNING = || - migration->device_state =3D=3D VFIO_DEVICE_STATE_PRE_COPY= )) { - return false; - } + if (vbasedev->pre_copy_dirty_page_tracking =3D=3D ON_OFF_AUTO_OFF = && + (migration->device_state =3D=3D VFIO_DEVICE_STATE_RUNNING || + migration->device_state =3D=3D VFIO_DEVICE_STATE_PRE_COPY)) { + return false; } } return true; @@ -236,14 +263,11 @@ static bool vfio_devices_all_dirty_tracking(VFIOConta= iner *container) =20 static bool vfio_devices_all_device_dirty_tracking(VFIOContainer *containe= r) { - VFIOGroup *group; - VFIODevice *vbasedev; + VFIODevice *vbasedev =3D NULL; =20 - QLIST_FOREACH(group, &container->group_list, container_next) { - QLIST_FOREACH(vbasedev, &group->device_list, next) { - if (!vbasedev->dirty_pages_supported) { - return false; - } + while ((vbasedev =3D vfio_container_dev_iter_next(container, vbasedev)= )) { + if (!vbasedev->dirty_pages_supported) { + return false; } } =20 @@ -256,27 +280,24 @@ static bool vfio_devices_all_device_dirty_tracking(VF= IOContainer *container) */ static bool vfio_devices_all_running_and_mig_active(VFIOContainer *contain= er) { - VFIOGroup *group; - VFIODevice *vbasedev; + VFIODevice *vbasedev =3D NULL; =20 if (!migration_is_active(migrate_get_current())) { return false; } =20 - QLIST_FOREACH(group, &container->group_list, container_next) { - QLIST_FOREACH(vbasedev, &group->device_list, next) { - VFIOMigration *migration =3D vbasedev->migration; + while ((vbasedev =3D vfio_container_dev_iter_next(container, vbasedev)= )) { + VFIOMigration *migration =3D vbasedev->migration; =20 - if (!migration) { - return false; - } + if (!migration) { + return false; + } =20 - if (migration->device_state =3D=3D VFIO_DEVICE_STATE_RUNNING || - migration->device_state =3D=3D VFIO_DEVICE_STATE_PRE_COPY)= { - continue; - } else { - return false; - } + if (migration->device_state =3D=3D VFIO_DEVICE_STATE_RUNNING || + migration->device_state =3D=3D VFIO_DEVICE_STATE_PRE_COPY) { + continue; + } else { + return false; } } return true; @@ -1243,25 +1264,22 @@ static void vfio_devices_dma_logging_stop(VFIOConta= iner *container) uint64_t buf[DIV_ROUND_UP(sizeof(struct vfio_device_feature), sizeof(uint64_t))] =3D {}; struct vfio_device_feature *feature =3D (struct vfio_device_feature *)= buf; - VFIODevice *vbasedev; - VFIOGroup *group; + VFIODevice *vbasedev =3D NULL; =20 feature->argsz =3D sizeof(buf); feature->flags =3D VFIO_DEVICE_FEATURE_SET | VFIO_DEVICE_FEATURE_DMA_LOGGING_STOP; =20 - QLIST_FOREACH(group, &container->group_list, container_next) { - QLIST_FOREACH(vbasedev, &group->device_list, next) { - if (!vbasedev->dirty_tracking) { - continue; - } + while ((vbasedev =3D vfio_container_dev_iter_next(container, vbasedev)= )) { + if (!vbasedev->dirty_tracking) { + continue; + } =20 - if (ioctl(vbasedev->fd, VFIO_DEVICE_FEATURE, feature)) { - warn_report("%s: Failed to stop DMA logging, err %d (%s)", - vbasedev->name, -errno, strerror(errno)); - } - vbasedev->dirty_tracking =3D false; + if (ioctl(vbasedev->fd, VFIO_DEVICE_FEATURE, feature)) { + warn_report("%s: Failed to stop DMA logging, err %d (%s)", + vbasedev->name, -errno, strerror(errno)); } + vbasedev->dirty_tracking =3D false; } } =20 @@ -1336,8 +1354,7 @@ static int vfio_devices_dma_logging_start(VFIOContain= er *container) { struct vfio_device_feature *feature; VFIODirtyRanges ranges; - VFIODevice *vbasedev; - VFIOGroup *group; + VFIODevice *vbasedev =3D NULL; int ret =3D 0; =20 vfio_dirty_tracking_init(container, &ranges); @@ -1347,21 +1364,19 @@ static int vfio_devices_dma_logging_start(VFIOConta= iner *container) return -errno; } =20 - QLIST_FOREACH(group, &container->group_list, container_next) { - QLIST_FOREACH(vbasedev, &group->device_list, next) { - if (vbasedev->dirty_tracking) { - continue; - } + while ((vbasedev =3D vfio_container_dev_iter_next(container, vbasedev)= )) { + if (vbasedev->dirty_tracking) { + continue; + } =20 - ret =3D ioctl(vbasedev->fd, VFIO_DEVICE_FEATURE, feature); - if (ret) { - ret =3D -errno; - error_report("%s: Failed to start DMA logging, err %d (%s)= ", - vbasedev->name, ret, strerror(errno)); - goto out; - } - vbasedev->dirty_tracking =3D true; + ret =3D ioctl(vbasedev->fd, VFIO_DEVICE_FEATURE, feature); + if (ret) { + ret =3D -errno; + error_report("%s: Failed to start DMA logging, err %d (%s)", + vbasedev->name, ret, strerror(errno)); + goto out; } + vbasedev->dirty_tracking =3D true; } =20 out: @@ -1440,22 +1455,19 @@ static int vfio_devices_query_dirty_bitmap(VFIOCont= ainer *container, VFIOBitmap *vbmap, hwaddr iova, hwaddr size) { - VFIODevice *vbasedev; - VFIOGroup *group; + VFIODevice *vbasedev =3D NULL; int ret; =20 - QLIST_FOREACH(group, &container->group_list, container_next) { - QLIST_FOREACH(vbasedev, &group->device_list, next) { - ret =3D vfio_device_dma_logging_report(vbasedev, iova, size, - vbmap->bitmap); - if (ret) { - error_report("%s: Failed to get DMA logging report, iova: " - "0x%" HWADDR_PRIx ", size: 0x%" HWADDR_PRIx - ", err: %d (%s)", - vbasedev->name, iova, size, ret, strerror(-re= t)); + while ((vbasedev =3D vfio_container_dev_iter_next(container, vbasedev)= )) { + ret =3D vfio_device_dma_logging_report(vbasedev, iova, size, + vbmap->bitmap); + if (ret) { + error_report("%s: Failed to get DMA logging report, iova: " + "0x%" HWADDR_PRIx ", size: 0x%" HWADDR_PRIx + ", err: %d (%s)", + vbasedev->name, iova, size, ret, strerror(-ret)); =20 - return ret; - } + return ret; } } =20 @@ -1739,21 +1751,30 @@ bool vfio_get_info_dma_avail(struct vfio_iommu_type= 1_info *info, =20 void vfio_reset_handler(void *opaque) { - VFIOGroup *group; + VFIOAddressSpace *space; + VFIOContainer *container; VFIODevice *vbasedev; =20 - QLIST_FOREACH(group, &vfio_group_list, next) { - QLIST_FOREACH(vbasedev, &group->device_list, next) { - if (vbasedev->dev->realized) { - vbasedev->ops->vfio_compute_needs_reset(vbasedev); + QLIST_FOREACH(space, &vfio_address_spaces, list) { + QLIST_FOREACH(container, &space->containers, next) { + vbasedev =3D NULL; + while ((vbasedev =3D vfio_container_dev_iter_next(container, + vbasedev))) { + if (vbasedev->dev->realized) { + vbasedev->ops->vfio_compute_needs_reset(vbasedev); + } } } } =20 - QLIST_FOREACH(group, &vfio_group_list, next) { - QLIST_FOREACH(vbasedev, &group->device_list, next) { - if (vbasedev->dev->realized && vbasedev->needs_reset) { - vbasedev->ops->vfio_hot_reset_multi(vbasedev); + QLIST_FOREACH(space, &vfio_address_spaces, list) { + QLIST_FOREACH(container, &space->containers, next) { + vbasedev =3D NULL; + while ((vbasedev =3D vfio_container_dev_iter_next(container, + vbasedev))) { + if (vbasedev->dev->realized && vbasedev->needs_reset) { + vbasedev->ops->vfio_hot_reset_multi(vbasedev); + } } } } @@ -1841,6 +1862,10 @@ static VFIOAddressSpace *vfio_get_address_space(Addr= essSpace *as) space->as =3D as; QLIST_INIT(&space->containers); =20 + if (QLIST_EMPTY(&vfio_address_spaces)) { + qemu_register_reset(vfio_reset_handler, NULL); + } + QLIST_INSERT_HEAD(&vfio_address_spaces, space, list); =20 return space; @@ -1852,6 +1877,9 @@ static void vfio_put_address_space(VFIOAddressSpace *= space) QLIST_REMOVE(space, list); g_free(space); } + if (QLIST_EMPTY(&vfio_address_spaces)) { + qemu_unregister_reset(vfio_reset_handler, NULL); + } } =20 /* @@ -2317,10 +2345,6 @@ VFIOGroup *vfio_get_group(int groupid, AddressSpace = *as, Error **errp) goto close_fd_exit; } =20 - if (QLIST_EMPTY(&vfio_group_list)) { - qemu_register_reset(vfio_reset_handler, NULL); - } - QLIST_INSERT_HEAD(&vfio_group_list, group, next); =20 return group; @@ -2349,10 +2373,6 @@ void vfio_put_group(VFIOGroup *group) trace_vfio_put_group(group->fd); close(group->fd); g_free(group); - - if (QLIST_EMPTY(&vfio_group_list)) { - qemu_unregister_reset(vfio_reset_handler, NULL); - } } =20 struct vfio_device_info *vfio_get_device_info(int fd) --=20 2.34.1 From nobody Wed May 15 22:12:51 2024 Delivered-To: importer@patchew.org Authentication-Results: mx.zohomail.com; dkim=pass header.i=@intel.com; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass(p=none dis=none) header.from=intel.com ARC-Seal: i=1; a=rsa-sha256; t=1693392877; cv=none; d=zohomail.com; s=zohoarc; b=FxTg4Dy6rB/9T/9m0tAOB1HMFacXA9nTkJfn+6zFDcKDrK5fIp4e56jwtTJCrxzmxanuacTveXKJc1nEsV2mWtXphf9vPbLPVyM6g1hzHdPX6VFnVO8lJpLgTuGJqo97i51zY/E4doFilRMb8Jnuywji9GRbqwnO+F1CBQsTWWc= ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=zohomail.com; s=zohoarc; t=1693392877; h=Content-Transfer-Encoding:Cc:Date:From:In-Reply-To:List-Subscribe:List-Post:List-Id:List-Archive:List-Help:List-Unsubscribe:MIME-Version:Message-ID:References:Sender:Subject:To; bh=KPKwjdDgbWzdUBTOkcTKvKrRV2tR34NRTMMhG+IVAyg=; b=oJVaReGWmtcDgr2TWttwEVaHtRoGR5q+y06WVCqm/yACd4W2Y0jgEtl6T7fVIPQ5K7d6Dv/R+pnMETakvy1a/Fsu4iKtpR2dQj/e2Q5qG/rZMHl0yezRUF1pd88EznFZVmX2wJ+jg3Ws461GR3elfRZELYvKt3aAevYOqWCIizI= ARC-Authentication-Results: i=1; mx.zohomail.com; dkim=pass header.i=@intel.com; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass header.from= (p=none dis=none) Return-Path: Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) by mx.zohomail.com with SMTPS id 1693392877215303.64480356974525; Wed, 30 Aug 2023 03:54:37 -0700 (PDT) Received: from localhost ([::1] helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1qbIp1-0004G4-II; Wed, 30 Aug 2023 06:53:15 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1qbIp0-0004FQ-FM for qemu-devel@nongnu.org; Wed, 30 Aug 2023 06:53:14 -0400 Received: from mgamail.intel.com ([192.55.52.93]) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1qbIox-0007fH-Vk for qemu-devel@nongnu.org; Wed, 30 Aug 2023 06:53:14 -0400 Received: from orsmga006.jf.intel.com ([10.7.209.51]) by fmsmga102.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 30 Aug 2023 03:52:59 -0700 Received: from duan-server-s2600bt.bj.intel.com ([10.240.192.147]) by orsmga006-auth.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 30 Aug 2023 03:52:55 -0700 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1693392792; x=1724928792; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=YbYyfmsmWybT6aQTjEyEx7ZGZgqUEt/i2jtdPlGMfjE=; b=bXuU5LhSrTf4ytpPSgYM/cLTutCIccH6sZM7jtlfv7x+t6KuLImteR+H i6OWwfRmWjU8v/u4FQL79cLxTvixLjUk5BrYhAbXmUiC166M1qwb8gPgk JclFmpTw93//CfrcIwlRShIo8/LG/U9I15lMF+XmVkE1cZqHGB58iZXia dd7lE3080GQEDwAL+ZAQhsaGWHf2M2fOCiudtK6auv7tUsPVn/PBbJvLV fIXqN87YiE/DyQj7vLnUIgUSnjs8e8HbW2Rle6qf1Qt4yBFsGZwkhvDhR uouXg8EcAs4ly+0ABP0OzWyYE37xbwYdPwJDbZP/8/Il0bh01QK9uQWOS Q==; X-IronPort-AV: E=McAfee;i="6600,9927,10817"; a="373016467" X-IronPort-AV: E=Sophos;i="6.02,213,1688454000"; d="scan'208";a="373016467" X-ExtLoop1: 1 X-IronPort-AV: E=McAfee;i="6600,9927,10817"; a="715866078" X-IronPort-AV: E=Sophos;i="6.02,213,1688454000"; d="scan'208";a="715866078" From: Zhenzhong Duan To: qemu-devel@nongnu.org Cc: alex.williamson@redhat.com, clg@redhat.com, jgg@nvidia.com, nicolinc@nvidia.com, joao.m.martins@oracle.com, eric.auger@redhat.com, peterx@redhat.com, jasowang@redhat.com, kevin.tian@intel.com, yi.l.liu@intel.com, yi.y.sun@intel.com, chao.p.peng@intel.com, Zhenzhong Duan Subject: [PATCH v1 07/22] vfio/common: Refactor vfio_viommu_preset() to be group agnostic Date: Wed, 30 Aug 2023 18:37:39 +0800 Message-Id: <20230830103754.36461-8-zhenzhong.duan@intel.com> X-Mailer: git-send-email 2.34.1 In-Reply-To: <20230830103754.36461-1-zhenzhong.duan@intel.com> References: <20230830103754.36461-1-zhenzhong.duan@intel.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Received-SPF: pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) client-ip=209.51.188.17; envelope-from=qemu-devel-bounces+importer=patchew.org@nongnu.org; helo=lists.gnu.org; Received-SPF: pass client-ip=192.55.52.93; envelope-from=zhenzhong.duan@intel.com; helo=mgamail.intel.com X-Spam_score_int: -43 X-Spam_score: -4.4 X-Spam_bar: ---- X-Spam_report: (-4.4 / 5.0 requ) BAYES_00=-1.9, DKIMWL_WL_HIGH=-0.001, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, RCVD_IN_DNSWL_MED=-2.3, SPF_HELO_NONE=0.001, SPF_PASS=-0.001 autolearn=ham autolearn_force=no X-Spam_action: no action X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: qemu-devel-bounces+importer=patchew.org@nongnu.org Sender: qemu-devel-bounces+importer=patchew.org@nongnu.org X-ZohoMail-DKIM: pass (identity @intel.com) X-ZM-MESSAGEID: 1693392938809100009 Content-Type: text/plain; charset="utf-8" So that it doesn't need to be moved into container.c as done in following patch. Signed-off-by: Zhenzhong Duan --- hw/vfio/common.c | 17 ++++++++++++++++- 1 file changed, 16 insertions(+), 1 deletion(-) diff --git a/hw/vfio/common.c b/hw/vfio/common.c index 51c6e7598e..fda5fc87b9 100644 --- a/hw/vfio/common.c +++ b/hw/vfio/common.c @@ -219,7 +219,22 @@ void vfio_unblock_multiple_devices_migration(void) =20 bool vfio_viommu_preset(VFIODevice *vbasedev) { - return vbasedev->group->container->space->as !=3D &address_space_memor= y; + VFIOAddressSpace *space; + VFIOContainer *container; + VFIODevice *tmp_dev; + + QLIST_FOREACH(space, &vfio_address_spaces, list) { + QLIST_FOREACH(container, &space->containers, next) { + tmp_dev =3D NULL; + while ((tmp_dev =3D vfio_container_dev_iter_next(container, + tmp_dev))) { + if (vbasedev =3D=3D tmp_dev) { + return space->as !=3D &address_space_memory; + } + } + } + } + g_assert_not_reached(); } =20 static void vfio_set_migration_error(int err) --=20 2.34.1 From nobody Wed May 15 22:12:51 2024 Delivered-To: importer@patchew.org Authentication-Results: mx.zohomail.com; dkim=pass header.i=@intel.com; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass(p=none dis=none) header.from=intel.com ARC-Seal: i=1; a=rsa-sha256; t=1693392860; cv=none; d=zohomail.com; s=zohoarc; b=V8iaAwdNmy9He96mfkv5NGLs9mmClR6YZ/LfwLxhWWWYiqQfWEtXButXswrDt2Neo+EEMq2y0pGtnh9YJhTkPflpNvRAlmeoisB0zwZsKZOUFtmLQVuf+g/ypaWL6Ne/3cFMkTaCCEDf2VyiIO1L1m0AN5LkcRnIX/3MLi3H2TY= ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=zohomail.com; s=zohoarc; t=1693392860; h=Content-Transfer-Encoding:Cc:Date:From:In-Reply-To:List-Subscribe:List-Post:List-Id:List-Archive:List-Help:List-Unsubscribe:MIME-Version:Message-ID:References:Sender:Subject:To; bh=q40T9eQ2I/N6bNZME8NgBfRzRIzr0luCBt84iLmATlQ=; b=nNLlNnho4pOvLwZGZbBT3HhzCl+Bc3M8d4jsvqIB51S6SsKThrL7fXn9/1ImWEtKYI5/64qZNruo2vSswisB1SsW+lRIDFDEOGHfXC4E0SDX7tmKYKLChFe+vn697696ttmu3rjdahKm/RQq1kZIO1mrLm0RvFbO0v2ct7DRNFQ= ARC-Authentication-Results: i=1; mx.zohomail.com; dkim=pass header.i=@intel.com; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass header.from= (p=none dis=none) Return-Path: Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) by mx.zohomail.com with SMTPS id 1693392860194612.2584081321622; Wed, 30 Aug 2023 03:54:20 -0700 (PDT) Received: from localhost ([::1] helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1qbIpD-0004Hj-5w; Wed, 30 Aug 2023 06:53:27 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1qbIp9-0004Gv-Qa for qemu-devel@nongnu.org; Wed, 30 Aug 2023 06:53:23 -0400 Received: from mgamail.intel.com ([192.55.52.93]) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1qbIp4-0007fH-Ao for qemu-devel@nongnu.org; Wed, 30 Aug 2023 06:53:23 -0400 Received: from orsmga006.jf.intel.com ([10.7.209.51]) by fmsmga102.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 30 Aug 2023 03:53:05 -0700 Received: from duan-server-s2600bt.bj.intel.com ([10.240.192.147]) by orsmga006-auth.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 30 Aug 2023 03:52:59 -0700 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1693392798; x=1724928798; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=ZOq5RSwhUYIyKkECuWRj7Izg7HRic3vJKy7s5TEpKHI=; b=hOP5uHR0Ebb9uZ1D9ShWsADgz8a1XD0xhPEtZ7uG92wWI0VE1VHFHfhA 0nfBX4V16oYDC0SyEtMzRXBaBGNT5ljN1FfcoEUFcacPcl3r74EMRP2O2 lGENfMQsRb/msUhBgkFeoY+FHFdHHSq7WUMP08RUvpNYVMBH9GXkEFz1w GC2lq3SQzWa5+oQyCWvTuP1nciUn+jofaFrtYW7m4RpGPWFGwk0/sm/HQ FiMyXZb/SfjtUTuzLNUAAUvsoOiuCS17mBQIYs+KngrQZbM+k5RTFRIgy PlJZ4HV+5mrPjuvxrnU/P4918RzmjHzecEK2JpUwfQU41IUKmVbykJRtE Q==; X-IronPort-AV: E=McAfee;i="6600,9927,10817"; a="373016487" X-IronPort-AV: E=Sophos;i="6.02,213,1688454000"; d="scan'208";a="373016487" X-ExtLoop1: 1 X-IronPort-AV: E=McAfee;i="6600,9927,10817"; a="715866092" X-IronPort-AV: E=Sophos;i="6.02,213,1688454000"; d="scan'208";a="715866092" From: Zhenzhong Duan To: qemu-devel@nongnu.org Cc: alex.williamson@redhat.com, clg@redhat.com, jgg@nvidia.com, nicolinc@nvidia.com, joao.m.martins@oracle.com, eric.auger@redhat.com, peterx@redhat.com, jasowang@redhat.com, kevin.tian@intel.com, yi.l.liu@intel.com, yi.y.sun@intel.com, chao.p.peng@intel.com, Zhenzhong Duan Subject: [PATCH v1 08/22] vfio/common: Move legacy VFIO backend code into separate container.c Date: Wed, 30 Aug 2023 18:37:40 +0800 Message-Id: <20230830103754.36461-9-zhenzhong.duan@intel.com> X-Mailer: git-send-email 2.34.1 In-Reply-To: <20230830103754.36461-1-zhenzhong.duan@intel.com> References: <20230830103754.36461-1-zhenzhong.duan@intel.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Received-SPF: pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) client-ip=209.51.188.17; envelope-from=qemu-devel-bounces+importer=patchew.org@nongnu.org; helo=lists.gnu.org; Received-SPF: pass client-ip=192.55.52.93; envelope-from=zhenzhong.duan@intel.com; helo=mgamail.intel.com X-Spam_score_int: -43 X-Spam_score: -4.4 X-Spam_bar: ---- X-Spam_report: (-4.4 / 5.0 requ) BAYES_00=-1.9, DKIMWL_WL_HIGH=-0.001, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, RCVD_IN_DNSWL_MED=-2.3, SPF_HELO_NONE=0.001, SPF_PASS=-0.001 autolearn=ham autolearn_force=no X-Spam_action: no action X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: qemu-devel-bounces+importer=patchew.org@nongnu.org Sender: qemu-devel-bounces+importer=patchew.org@nongnu.org X-ZohoMail-DKIM: pass (identity @intel.com) X-ZM-MESSAGEID: 1693392936431100005 Content-Type: text/plain; charset="utf-8" From: Yi Liu Move all the code really dependent on the legacy VFIO container/group into a separate file: container.c. What does remain in common.c is the code related to VFIOAddressSpace, MemoryListeners, migration and all other general operations. Move struct VFIOBitmap declaration to vfio-common.h also for containter.c usage. Signed-off-by: Eric Auger Signed-off-by: Yi Liu Signed-off-by: Zhenzhong Duan --- --- hw/vfio/common.c | 1085 +-------------------------------- hw/vfio/container.c | 1085 +++++++++++++++++++++++++++++++++ hw/vfio/meson.build | 1 + include/hw/vfio/vfio-common.h | 45 ++ 4 files changed, 1147 insertions(+), 1069 deletions(-) create mode 100644 hw/vfio/container.c diff --git a/hw/vfio/common.c b/hw/vfio/common.c index fda5fc87b9..044710fc1f 100644 --- a/hw/vfio/common.c +++ b/hw/vfio/common.c @@ -45,8 +45,6 @@ #include "migration/qemu-file.h" #include "sysemu/tpm.h" =20 -VFIOGroupList vfio_group_list =3D - QLIST_HEAD_INITIALIZER(vfio_group_list); static QLIST_HEAD(, VFIOAddressSpace) vfio_address_spaces =3D QLIST_HEAD_INITIALIZER(vfio_address_spaces); =20 @@ -58,63 +56,14 @@ static QLIST_HEAD(, VFIOAddressSpace) vfio_address_spac= es =3D * initialized, this file descriptor is only released on QEMU exit and * we'll re-use it should another vfio device be attached before then. */ -static int vfio_kvm_device_fd =3D -1; +int vfio_kvm_device_fd =3D -1; #endif =20 -static int vfio_ram_block_discard_disable(VFIOContainer *container, bool s= tate) -{ - switch (container->iommu_type) { - case VFIO_TYPE1v2_IOMMU: - case VFIO_TYPE1_IOMMU: - /* - * We support coordinated discarding of RAM via the RamDiscardMana= ger. - */ - return ram_block_uncoordinated_discard_disable(state); - default: - /* - * VFIO_SPAPR_TCE_IOMMU most probably works just fine with - * RamDiscardManager, however, it is completely untested. - * - * VFIO_SPAPR_TCE_v2_IOMMU with "DMA memory preregistering" does - * completely the opposite of managing mapping/pinning dynamically= as - * required by RamDiscardManager. We would have to special-case se= ctions - * with a RamDiscardManager. - */ - return ram_block_discard_disable(state); - } -} - -static VFIODevice *vfio_container_dev_iter_next(VFIOContainer *container, - VFIODevice *curr) -{ - VFIOGroup *group; - - if (!curr) { - group =3D QLIST_FIRST(&container->group_list); - } else { - if (curr->next.le_next) { - return curr->next.le_next; - } - group =3D curr->group->container_next.le_next; - } - - if (!group) { - return NULL; - } - return QLIST_FIRST(&group->device_list); -} - /* * Device state interfaces */ =20 -typedef struct { - unsigned long *bitmap; - hwaddr size; - hwaddr pages; -} VFIOBitmap; - -static int vfio_bitmap_alloc(VFIOBitmap *vbmap, hwaddr size) +int vfio_bitmap_alloc(VFIOBitmap *vbmap, hwaddr size) { vbmap->pages =3D REAL_HOST_PAGE_ALIGN(size) / qemu_real_host_page_size= (); vbmap->size =3D ROUND_UP(vbmap->pages, sizeof(__u64) * BITS_PER_BYTE) / @@ -127,9 +76,6 @@ static int vfio_bitmap_alloc(VFIOBitmap *vbmap, hwaddr s= ize) return 0; } =20 -static int vfio_get_dirty_bitmap(VFIOContainer *container, uint64_t iova, - uint64_t size, ram_addr_t ram_addr); - bool vfio_mig_active(void) { VFIOAddressSpace *space; @@ -276,7 +222,7 @@ static bool vfio_devices_all_dirty_tracking(VFIOContain= er *container) return true; } =20 -static bool vfio_devices_all_device_dirty_tracking(VFIOContainer *containe= r) +bool vfio_devices_all_device_dirty_tracking(VFIOContainer *container) { VFIODevice *vbasedev =3D NULL; =20 @@ -293,7 +239,7 @@ static bool vfio_devices_all_device_dirty_tracking(VFIO= Container *container) * Check if all VFIO devices are running and migration is active, which is * essentially equivalent to the migration being in pre-copy phase. */ -static bool vfio_devices_all_running_and_mig_active(VFIOContainer *contain= er) +bool vfio_devices_all_running_and_mig_active(VFIOContainer *container) { VFIODevice *vbasedev =3D NULL; =20 @@ -318,150 +264,8 @@ static bool vfio_devices_all_running_and_mig_active(V= FIOContainer *container) return true; } =20 -static int vfio_dma_unmap_bitmap(VFIOContainer *container, - hwaddr iova, ram_addr_t size, - IOMMUTLBEntry *iotlb) -{ - struct vfio_iommu_type1_dma_unmap *unmap; - struct vfio_bitmap *bitmap; - VFIOBitmap vbmap; - int ret; - - ret =3D vfio_bitmap_alloc(&vbmap, size); - if (ret) { - return ret; - } - - unmap =3D g_malloc0(sizeof(*unmap) + sizeof(*bitmap)); - - unmap->argsz =3D sizeof(*unmap) + sizeof(*bitmap); - unmap->iova =3D iova; - unmap->size =3D size; - unmap->flags |=3D VFIO_DMA_UNMAP_FLAG_GET_DIRTY_BITMAP; - bitmap =3D (struct vfio_bitmap *)&unmap->data; - - /* - * cpu_physical_memory_set_dirty_lebitmap() supports pages in bitmap of - * qemu_real_host_page_size to mark those dirty. Hence set bitmap_pgsi= ze - * to qemu_real_host_page_size. - */ - bitmap->pgsize =3D qemu_real_host_page_size(); - bitmap->size =3D vbmap.size; - bitmap->data =3D (__u64 *)vbmap.bitmap; - - if (vbmap.size > container->max_dirty_bitmap_size) { - error_report("UNMAP: Size of bitmap too big 0x%"PRIx64, vbmap.size= ); - ret =3D -E2BIG; - goto unmap_exit; - } - - ret =3D ioctl(container->fd, VFIO_IOMMU_UNMAP_DMA, unmap); - if (!ret) { - cpu_physical_memory_set_dirty_lebitmap(vbmap.bitmap, - iotlb->translated_addr, vbmap.pages); - } else { - error_report("VFIO_UNMAP_DMA with DIRTY_BITMAP : %m"); - } - -unmap_exit: - g_free(unmap); - g_free(vbmap.bitmap); - - return ret; -} - -/* - * DMA - Mapping and unmapping for the "type1" IOMMU interface used on x86 - */ -static int vfio_dma_unmap(VFIOContainer *container, - hwaddr iova, ram_addr_t size, - IOMMUTLBEntry *iotlb) -{ - struct vfio_iommu_type1_dma_unmap unmap =3D { - .argsz =3D sizeof(unmap), - .flags =3D 0, - .iova =3D iova, - .size =3D size, - }; - bool need_dirty_sync =3D false; - int ret; - - if (iotlb && vfio_devices_all_running_and_mig_active(container)) { - if (!vfio_devices_all_device_dirty_tracking(container) && - container->dirty_pages_supported) { - return vfio_dma_unmap_bitmap(container, iova, size, iotlb); - } - - need_dirty_sync =3D true; - } - - while (ioctl(container->fd, VFIO_IOMMU_UNMAP_DMA, &unmap)) { - /* - * The type1 backend has an off-by-one bug in the kernel (71a7d3d7= 8e3c - * v4.15) where an overflow in its wrap-around check prevents us f= rom - * unmapping the last page of the address space. Test for the err= or - * condition and re-try the unmap excluding the last page. The - * expectation is that we've never mapped the last page anyway and= this - * unmap request comes via vIOMMU support which also makes it unli= kely - * that this page is used. This bug was introduced well after typ= e1 v2 - * support was introduced, so we shouldn't need to test for v1. A= fix - * is queued for kernel v5.0 so this workaround can be removed once - * affected kernels are sufficiently deprecated. - */ - if (errno =3D=3D EINVAL && unmap.size && !(unmap.iova + unmap.size= ) && - container->iommu_type =3D=3D VFIO_TYPE1v2_IOMMU) { - trace_vfio_dma_unmap_overflow_workaround(); - unmap.size -=3D 1ULL << ctz64(container->pgsizes); - continue; - } - error_report("VFIO_UNMAP_DMA failed: %s", strerror(errno)); - return -errno; - } - - if (need_dirty_sync) { - ret =3D vfio_get_dirty_bitmap(container, iova, size, - iotlb->translated_addr); - if (ret) { - return ret; - } - } - - return 0; -} - -static int vfio_dma_map(VFIOContainer *container, hwaddr iova, - ram_addr_t size, void *vaddr, bool readonly) -{ - struct vfio_iommu_type1_dma_map map =3D { - .argsz =3D sizeof(map), - .flags =3D VFIO_DMA_MAP_FLAG_READ, - .vaddr =3D (__u64)(uintptr_t)vaddr, - .iova =3D iova, - .size =3D size, - }; - - if (!readonly) { - map.flags |=3D VFIO_DMA_MAP_FLAG_WRITE; - } - - /* - * Try the mapping, if it fails with EBUSY, unmap the region and try - * again. This shouldn't be necessary, but we sometimes see it in - * the VGA ROM space. - */ - if (ioctl(container->fd, VFIO_IOMMU_MAP_DMA, &map) =3D=3D 0 || - (errno =3D=3D EBUSY && vfio_dma_unmap(container, iova, size, NULL)= =3D=3D 0 && - ioctl(container->fd, VFIO_IOMMU_MAP_DMA, &map) =3D=3D 0)) { - return 0; - } - - error_report("VFIO_MAP_DMA failed: %s", strerror(errno)); - return -errno; -} - -static void vfio_host_win_add(VFIOContainer *container, - hwaddr min_iova, hwaddr max_iova, - uint64_t iova_pgsizes) +void vfio_host_win_add(VFIOContainer *container, hwaddr min_iova, + hwaddr max_iova, uint64_t iova_pgsizes) { VFIOHostDMAWindow *hostwin; =20 @@ -482,8 +286,8 @@ static void vfio_host_win_add(VFIOContainer *container, QLIST_INSERT_HEAD(&container->hostwin_list, hostwin, hostwin_next); } =20 -static int vfio_host_win_del(VFIOContainer *container, hwaddr min_iova, - hwaddr max_iova) +int vfio_host_win_del(VFIOContainer *container, + hwaddr min_iova, hwaddr max_iova) { VFIOHostDMAWindow *hostwin; =20 @@ -832,92 +636,6 @@ static bool vfio_get_section_iova_range(VFIOContainer = *container, return true; } =20 -static int vfio_container_add_section_window(VFIOContainer *container, - MemoryRegionSection *section, - Error **errp) -{ - VFIOHostDMAWindow *hostwin; - hwaddr pgsize =3D 0; - int ret; - - if (container->iommu_type !=3D VFIO_SPAPR_TCE_v2_IOMMU) { - return 0; - } - - /* For now intersections are not allowed, we may relax this later */ - QLIST_FOREACH(hostwin, &container->hostwin_list, hostwin_next) { - if (ranges_overlap(hostwin->min_iova, - hostwin->max_iova - hostwin->min_iova + 1, - section->offset_within_address_space, - int128_get64(section->size))) { - error_setg(errp, - "region [0x%"PRIx64",0x%"PRIx64"] overlaps with existing" - "host DMA window [0x%"PRIx64",0x%"PRIx64"]", - section->offset_within_address_space, - section->offset_within_address_space + - int128_get64(section->size) - 1, - hostwin->min_iova, hostwin->max_iova); - return -EINVAL; - } - } - - ret =3D vfio_spapr_create_window(container, section, &pgsize); - if (ret) { - error_setg_errno(errp, -ret, "Failed to create SPAPR window"); - return ret; - } - - vfio_host_win_add(container, section->offset_within_address_space, - section->offset_within_address_space + - int128_get64(section->size) - 1, pgsize); -#ifdef CONFIG_KVM - if (kvm_enabled()) { - VFIOGroup *group; - IOMMUMemoryRegion *iommu_mr =3D IOMMU_MEMORY_REGION(section->mr); - struct kvm_vfio_spapr_tce param; - struct kvm_device_attr attr =3D { - .group =3D KVM_DEV_VFIO_GROUP, - .attr =3D KVM_DEV_VFIO_GROUP_SET_SPAPR_TCE, - .addr =3D (uint64_t)(unsigned long)¶m, - }; - - if (!memory_region_iommu_get_attr(iommu_mr, IOMMU_ATTR_SPAPR_TCE_F= D, - ¶m.tablefd)) { - QLIST_FOREACH(group, &container->group_list, container_next) { - param.groupfd =3D group->fd; - if (ioctl(vfio_kvm_device_fd, KVM_SET_DEVICE_ATTR, &attr))= { - error_report("vfio: failed to setup fd %d " - "for a group with fd %d: %s", - param.tablefd, param.groupfd, - strerror(errno)); - return 0; - } - trace_vfio_spapr_group_attach(param.groupfd, param.tablefd= ); - } - } - } -#endif - return 0; -} - -static void vfio_container_del_section_window(VFIOContainer *container, - MemoryRegionSection *section) -{ - if (container->iommu_type !=3D VFIO_SPAPR_TCE_v2_IOMMU) { - return; - } - - vfio_spapr_remove_window(container, - section->offset_within_address_space); - if (vfio_host_win_del(container, - section->offset_within_address_space, - section->offset_within_address_space + - int128_get64(section->size) - 1) < 0) { - hw_error("%s: Cannot delete missing window at %"HWADDR_PRIx, - __func__, section->offset_within_address_space); - } -} - static void vfio_listener_region_add(MemoryListener *listener, MemoryRegionSection *section) { @@ -1165,33 +883,6 @@ static void vfio_listener_region_del(MemoryListener *= listener, vfio_container_del_section_window(container, section); } =20 -static int vfio_set_dirty_page_tracking(VFIOContainer *container, bool sta= rt) -{ - int ret; - struct vfio_iommu_type1_dirty_bitmap dirty =3D { - .argsz =3D sizeof(dirty), - }; - - if (!container->dirty_pages_supported) { - return 0; - } - - if (start) { - dirty.flags =3D VFIO_IOMMU_DIRTY_PAGES_FLAG_START; - } else { - dirty.flags =3D VFIO_IOMMU_DIRTY_PAGES_FLAG_STOP; - } - - ret =3D ioctl(container->fd, VFIO_IOMMU_DIRTY_PAGES, &dirty); - if (ret) { - ret =3D -errno; - error_report("Failed to set dirty tracking flag 0x%x errno: %d", - dirty.flags, errno); - } - - return ret; -} - typedef struct VFIODirtyRanges { hwaddr min32; hwaddr max32; @@ -1466,9 +1157,9 @@ static int vfio_device_dma_logging_report(VFIODevice = *vbasedev, hwaddr iova, return 0; } =20 -static int vfio_devices_query_dirty_bitmap(VFIOContainer *container, - VFIOBitmap *vbmap, hwaddr iova, - hwaddr size) +int vfio_devices_query_dirty_bitmap(VFIOContainer *container, + VFIOBitmap *vbmap, hwaddr iova, + hwaddr size) { VFIODevice *vbasedev =3D NULL; int ret; @@ -1489,45 +1180,8 @@ static int vfio_devices_query_dirty_bitmap(VFIOConta= iner *container, return 0; } =20 -static int vfio_query_dirty_bitmap(VFIOContainer *container, VFIOBitmap *v= bmap, - hwaddr iova, hwaddr size) -{ - struct vfio_iommu_type1_dirty_bitmap *dbitmap; - struct vfio_iommu_type1_dirty_bitmap_get *range; - int ret; - - dbitmap =3D g_malloc0(sizeof(*dbitmap) + sizeof(*range)); - - dbitmap->argsz =3D sizeof(*dbitmap) + sizeof(*range); - dbitmap->flags =3D VFIO_IOMMU_DIRTY_PAGES_FLAG_GET_BITMAP; - range =3D (struct vfio_iommu_type1_dirty_bitmap_get *)&dbitmap->data; - range->iova =3D iova; - range->size =3D size; - - /* - * cpu_physical_memory_set_dirty_lebitmap() supports pages in bitmap of - * qemu_real_host_page_size to mark those dirty. Hence set bitmap's pg= size - * to qemu_real_host_page_size. - */ - range->bitmap.pgsize =3D qemu_real_host_page_size(); - range->bitmap.size =3D vbmap->size; - range->bitmap.data =3D (__u64 *)vbmap->bitmap; - - ret =3D ioctl(container->fd, VFIO_IOMMU_DIRTY_PAGES, dbitmap); - if (ret) { - ret =3D -errno; - error_report("Failed to get dirty bitmap for iova: 0x%"PRIx64 - " size: 0x%"PRIx64" err: %d", (uint64_t)range->iova, - (uint64_t)range->size, errno); - } - - g_free(dbitmap); - - return ret; -} - -static int vfio_get_dirty_bitmap(VFIOContainer *container, uint64_t iova, - uint64_t size, ram_addr_t ram_addr) +int vfio_get_dirty_bitmap(VFIOContainer *container, uint64_t iova, + uint64_t size, ram_addr_t ram_addr) { bool all_device_dirty_tracking =3D vfio_devices_all_device_dirty_tracking(container); @@ -1716,7 +1370,7 @@ static void vfio_listener_log_sync(MemoryListener *li= stener, } } =20 -static const MemoryListener vfio_memory_listener =3D { +const MemoryListener vfio_memory_listener =3D { .name =3D "vfio", .region_add =3D vfio_listener_region_add, .region_del =3D vfio_listener_region_del, @@ -1725,45 +1379,6 @@ static const MemoryListener vfio_memory_listener =3D= { .log_sync =3D vfio_listener_log_sync, }; =20 -static void vfio_listener_release(VFIOContainer *container) -{ - memory_listener_unregister(&container->listener); - if (container->iommu_type =3D=3D VFIO_SPAPR_TCE_v2_IOMMU) { - memory_listener_unregister(&container->prereg_listener); - } -} - -static struct vfio_info_cap_header * -vfio_get_iommu_type1_info_cap(struct vfio_iommu_type1_info *info, uint16_t= id) -{ - if (!(info->flags & VFIO_IOMMU_INFO_CAPS)) { - return NULL; - } - - return vfio_get_cap((void *)info, info->cap_offset, id); -} - -bool vfio_get_info_dma_avail(struct vfio_iommu_type1_info *info, - unsigned int *avail) -{ - struct vfio_info_cap_header *hdr; - struct vfio_iommu_type1_info_dma_avail *cap; - - /* If the capability cannot be found, assume no DMA limiting */ - hdr =3D vfio_get_iommu_type1_info_cap(info, - VFIO_IOMMU_TYPE1_INFO_DMA_AVAIL); - if (hdr =3D=3D NULL) { - return false; - } - - if (avail !=3D NULL) { - cap =3D (void *) hdr; - *avail =3D cap->avail; - } - - return true; -} - void vfio_reset_handler(void *opaque) { VFIOAddressSpace *space; @@ -1830,11 +1445,6 @@ int vfio_kvm_device_add_fd(int fd) return 0; } =20 -static void vfio_kvm_device_add_group(VFIOGroup *group) -{ - vfio_kvm_device_add_fd(group->fd); -} - int vfio_kvm_device_del_fd(int fd) { #ifdef CONFIG_KVM @@ -1857,12 +1467,7 @@ int vfio_kvm_device_del_fd(int fd) return 0; } =20 -static void vfio_kvm_device_del_group(VFIOGroup *group) -{ - vfio_kvm_device_del_fd(group->fd); -} - -static VFIOAddressSpace *vfio_get_address_space(AddressSpace *as) +VFIOAddressSpace *vfio_get_address_space(AddressSpace *as) { VFIOAddressSpace *space; =20 @@ -1886,7 +1491,7 @@ static VFIOAddressSpace *vfio_get_address_space(Addre= ssSpace *as) return space; } =20 -static void vfio_put_address_space(VFIOAddressSpace *space) +void vfio_put_address_space(VFIOAddressSpace *space) { if (QLIST_EMPTY(&space->containers)) { QLIST_REMOVE(space, list); @@ -1897,499 +1502,6 @@ static void vfio_put_address_space(VFIOAddressSpace= *space) } } =20 -/* - * vfio_get_iommu_type - selects the richest iommu_type (v2 first) - */ -static int vfio_get_iommu_type(VFIOContainer *container, - Error **errp) -{ - int iommu_types[] =3D { VFIO_TYPE1v2_IOMMU, VFIO_TYPE1_IOMMU, - VFIO_SPAPR_TCE_v2_IOMMU, VFIO_SPAPR_TCE_IOMMU }; - int i; - - for (i =3D 0; i < ARRAY_SIZE(iommu_types); i++) { - if (ioctl(container->fd, VFIO_CHECK_EXTENSION, iommu_types[i])) { - return iommu_types[i]; - } - } - error_setg(errp, "No available IOMMU models"); - return -EINVAL; -} - -static int vfio_init_container(VFIOContainer *container, int group_fd, - Error **errp) -{ - int iommu_type, ret; - - iommu_type =3D vfio_get_iommu_type(container, errp); - if (iommu_type < 0) { - return iommu_type; - } - - ret =3D ioctl(group_fd, VFIO_GROUP_SET_CONTAINER, &container->fd); - if (ret) { - error_setg_errno(errp, errno, "Failed to set group container"); - return -errno; - } - - while (ioctl(container->fd, VFIO_SET_IOMMU, iommu_type)) { - if (iommu_type =3D=3D VFIO_SPAPR_TCE_v2_IOMMU) { - /* - * On sPAPR, despite the IOMMU subdriver always advertises v1 = and - * v2, the running platform may not support v2 and there is no - * way to guess it until an IOMMU group gets added to the cont= ainer. - * So in case it fails with v2, try v1 as a fallback. - */ - iommu_type =3D VFIO_SPAPR_TCE_IOMMU; - continue; - } - error_setg_errno(errp, errno, "Failed to set iommu for container"); - return -errno; - } - - container->iommu_type =3D iommu_type; - return 0; -} - -static int vfio_get_iommu_info(VFIOContainer *container, - struct vfio_iommu_type1_info **info) -{ - - size_t argsz =3D sizeof(struct vfio_iommu_type1_info); - - *info =3D g_new0(struct vfio_iommu_type1_info, 1); -again: - (*info)->argsz =3D argsz; - - if (ioctl(container->fd, VFIO_IOMMU_GET_INFO, *info)) { - g_free(*info); - *info =3D NULL; - return -errno; - } - - if (((*info)->argsz > argsz)) { - argsz =3D (*info)->argsz; - *info =3D g_realloc(*info, argsz); - goto again; - } - - return 0; -} - -static struct vfio_info_cap_header * -vfio_get_iommu_info_cap(struct vfio_iommu_type1_info *info, uint16_t id) -{ - struct vfio_info_cap_header *hdr; - void *ptr =3D info; - - if (!(info->flags & VFIO_IOMMU_INFO_CAPS)) { - return NULL; - } - - for (hdr =3D ptr + info->cap_offset; hdr !=3D ptr; hdr =3D ptr + hdr->= next) { - if (hdr->id =3D=3D id) { - return hdr; - } - } - - return NULL; -} - -static void vfio_get_iommu_info_migration(VFIOContainer *container, - struct vfio_iommu_type1_info *inf= o) -{ - struct vfio_info_cap_header *hdr; - struct vfio_iommu_type1_info_cap_migration *cap_mig; - - hdr =3D vfio_get_iommu_info_cap(info, VFIO_IOMMU_TYPE1_INFO_CAP_MIGRAT= ION); - if (!hdr) { - return; - } - - cap_mig =3D container_of(hdr, struct vfio_iommu_type1_info_cap_migrati= on, - header); - - /* - * cpu_physical_memory_set_dirty_lebitmap() supports pages in bitmap of - * qemu_real_host_page_size to mark those dirty. - */ - if (cap_mig->pgsize_bitmap & qemu_real_host_page_size()) { - container->dirty_pages_supported =3D true; - container->max_dirty_bitmap_size =3D cap_mig->max_dirty_bitmap_siz= e; - container->dirty_pgsizes =3D cap_mig->pgsize_bitmap; - } -} - -static int vfio_connect_container(VFIOGroup *group, AddressSpace *as, - Error **errp) -{ - VFIOContainer *container; - int ret, fd; - VFIOAddressSpace *space; - - space =3D vfio_get_address_space(as); - - /* - * VFIO is currently incompatible with discarding of RAM insofar as the - * madvise to purge (zap) the page from QEMU's address space does not - * interact with the memory API and therefore leaves stale virtual to - * physical mappings in the IOMMU if the page was previously pinned. = We - * therefore set discarding broken for each group added to a container, - * whether the container is used individually or shared. This provides - * us with options to allow devices within a group to opt-in and allow - * discarding, so long as it is done consistently for a group (for ins= tance - * if the device is an mdev device where it is known that the host ven= dor - * driver will never pin pages outside of the working set of the guest - * driver, which would thus not be discarding candidates). - * - * The first opportunity to induce pinning occurs here where we attemp= t to - * attach the group to existing containers within the AddressSpace. I= f any - * pages are already zapped from the virtual address space, such as fr= om - * previous discards, new pinning will cause valid mappings to be - * re-established. Likewise, when the overall MemoryListener for a new - * container is registered, a replay of mappings within the AddressSpa= ce - * will occur, re-establishing any previously zapped pages as well. - * - * Especially virtio-balloon is currently only prevented from discardi= ng - * new memory, it will not yet set ram_block_discard_set_required() and - * therefore, neither stops us here or deals with the sudden memory - * consumption of inflated memory. - * - * We do support discarding of memory coordinated via the RamDiscardMa= nager - * with some IOMMU types. vfio_ram_block_discard_disable() handles the - * details once we know which type of IOMMU we are using. - */ - - QLIST_FOREACH(container, &space->containers, next) { - if (!ioctl(group->fd, VFIO_GROUP_SET_CONTAINER, &container->fd)) { - ret =3D vfio_ram_block_discard_disable(container, true); - if (ret) { - error_setg_errno(errp, -ret, - "Cannot set discarding of RAM broken"); - if (ioctl(group->fd, VFIO_GROUP_UNSET_CONTAINER, - &container->fd)) { - error_report("vfio: error disconnecting group %d from" - " container", group->groupid); - } - return ret; - } - group->container =3D container; - QLIST_INSERT_HEAD(&container->group_list, group, container_nex= t); - vfio_kvm_device_add_group(group); - return 0; - } - } - - fd =3D qemu_open_old("/dev/vfio/vfio", O_RDWR); - if (fd < 0) { - error_setg_errno(errp, errno, "failed to open /dev/vfio/vfio"); - ret =3D -errno; - goto put_space_exit; - } - - ret =3D ioctl(fd, VFIO_GET_API_VERSION); - if (ret !=3D VFIO_API_VERSION) { - error_setg(errp, "supported vfio version: %d, " - "reported version: %d", VFIO_API_VERSION, ret); - ret =3D -EINVAL; - goto close_fd_exit; - } - - container =3D g_malloc0(sizeof(*container)); - container->space =3D space; - container->fd =3D fd; - container->error =3D NULL; - container->dirty_pages_supported =3D false; - container->dma_max_mappings =3D 0; - QLIST_INIT(&container->giommu_list); - QLIST_INIT(&container->hostwin_list); - QLIST_INIT(&container->vrdl_list); - - ret =3D vfio_init_container(container, group->fd, errp); - if (ret) { - goto free_container_exit; - } - - ret =3D vfio_ram_block_discard_disable(container, true); - if (ret) { - error_setg_errno(errp, -ret, "Cannot set discarding of RAM broken"= ); - goto free_container_exit; - } - - switch (container->iommu_type) { - case VFIO_TYPE1v2_IOMMU: - case VFIO_TYPE1_IOMMU: - { - struct vfio_iommu_type1_info *info; - - ret =3D vfio_get_iommu_info(container, &info); - if (ret) { - error_setg_errno(errp, -ret, "Failed to get VFIO IOMMU info"); - goto enable_discards_exit; - } - - if (info->flags & VFIO_IOMMU_INFO_PGSIZES) { - container->pgsizes =3D info->iova_pgsizes; - } else { - container->pgsizes =3D qemu_real_host_page_size(); - } - - if (!vfio_get_info_dma_avail(info, &container->dma_max_mappings)) { - container->dma_max_mappings =3D 65535; - } - vfio_get_iommu_info_migration(container, info); - g_free(info); - - /* - * FIXME: We should parse VFIO_IOMMU_TYPE1_INFO_CAP_IOVA_RANGE - * information to get the actual window extent rather than assume - * a 64-bit IOVA address space. - */ - vfio_host_win_add(container, 0, (hwaddr)-1, container->pgsizes); - - break; - } - case VFIO_SPAPR_TCE_v2_IOMMU: - case VFIO_SPAPR_TCE_IOMMU: - { - struct vfio_iommu_spapr_tce_info info; - bool v2 =3D container->iommu_type =3D=3D VFIO_SPAPR_TCE_v2_IOMMU; - - /* - * The host kernel code implementing VFIO_IOMMU_DISABLE is called - * when container fd is closed so we do not call it explicitly - * in this file. - */ - if (!v2) { - ret =3D ioctl(fd, VFIO_IOMMU_ENABLE); - if (ret) { - error_setg_errno(errp, errno, "failed to enable container"= ); - ret =3D -errno; - goto enable_discards_exit; - } - } else { - container->prereg_listener =3D vfio_prereg_listener; - - memory_listener_register(&container->prereg_listener, - &address_space_memory); - if (container->error) { - memory_listener_unregister(&container->prereg_listener); - ret =3D -1; - error_propagate_prepend(errp, container->error, - "RAM memory listener initialization failed: "); - goto enable_discards_exit; - } - } - - info.argsz =3D sizeof(info); - ret =3D ioctl(fd, VFIO_IOMMU_SPAPR_TCE_GET_INFO, &info); - if (ret) { - error_setg_errno(errp, errno, - "VFIO_IOMMU_SPAPR_TCE_GET_INFO failed"); - ret =3D -errno; - if (v2) { - memory_listener_unregister(&container->prereg_listener); - } - goto enable_discards_exit; - } - - if (v2) { - container->pgsizes =3D info.ddw.pgsizes; - /* - * There is a default window in just created container. - * To make region_add/del simpler, we better remove this - * window now and let those iommu_listener callbacks - * create/remove them when needed. - */ - ret =3D vfio_spapr_remove_window(container, info.dma32_window_= start); - if (ret) { - error_setg_errno(errp, -ret, - "failed to remove existing window"); - goto enable_discards_exit; - } - } else { - /* The default table uses 4K pages */ - container->pgsizes =3D 0x1000; - vfio_host_win_add(container, info.dma32_window_start, - info.dma32_window_start + - info.dma32_window_size - 1, - 0x1000); - } - } - } - - vfio_kvm_device_add_group(group); - - QLIST_INIT(&container->group_list); - QLIST_INSERT_HEAD(&space->containers, container, next); - - group->container =3D container; - QLIST_INSERT_HEAD(&container->group_list, group, container_next); - - container->listener =3D vfio_memory_listener; - - memory_listener_register(&container->listener, container->space->as); - - if (container->error) { - ret =3D -1; - error_propagate_prepend(errp, container->error, - "memory listener initialization failed: "); - goto listener_release_exit; - } - - container->initialized =3D true; - - return 0; -listener_release_exit: - QLIST_REMOVE(group, container_next); - QLIST_REMOVE(container, next); - vfio_kvm_device_del_group(group); - vfio_listener_release(container); - -enable_discards_exit: - vfio_ram_block_discard_disable(container, false); - -free_container_exit: - g_free(container); - -close_fd_exit: - close(fd); - -put_space_exit: - vfio_put_address_space(space); - - return ret; -} - -static void vfio_disconnect_container(VFIOGroup *group) -{ - VFIOContainer *container =3D group->container; - - QLIST_REMOVE(group, container_next); - group->container =3D NULL; - - /* - * Explicitly release the listener first before unset container, - * since unset may destroy the backend container if it's the last - * group. - */ - if (QLIST_EMPTY(&container->group_list)) { - vfio_listener_release(container); - } - - if (ioctl(group->fd, VFIO_GROUP_UNSET_CONTAINER, &container->fd)) { - error_report("vfio: error disconnecting group %d from container", - group->groupid); - } - - if (QLIST_EMPTY(&container->group_list)) { - VFIOAddressSpace *space =3D container->space; - VFIOGuestIOMMU *giommu, *tmp; - VFIOHostDMAWindow *hostwin, *next; - - QLIST_REMOVE(container, next); - - QLIST_FOREACH_SAFE(giommu, &container->giommu_list, giommu_next, t= mp) { - memory_region_unregister_iommu_notifier( - MEMORY_REGION(giommu->iommu_mr), &giommu->n); - QLIST_REMOVE(giommu, giommu_next); - g_free(giommu); - } - - QLIST_FOREACH_SAFE(hostwin, &container->hostwin_list, hostwin_next, - next) { - QLIST_REMOVE(hostwin, hostwin_next); - g_free(hostwin); - } - - trace_vfio_disconnect_container(container->fd); - close(container->fd); - g_free(container); - - vfio_put_address_space(space); - } -} - -VFIOGroup *vfio_get_group(int groupid, AddressSpace *as, Error **errp) -{ - VFIOGroup *group; - char path[32]; - struct vfio_group_status status =3D { .argsz =3D sizeof(status) }; - - QLIST_FOREACH(group, &vfio_group_list, next) { - if (group->groupid =3D=3D groupid) { - /* Found it. Now is it already in the right context? */ - if (group->container->space->as =3D=3D as) { - return group; - } else { - error_setg(errp, "group %d used in multiple address spaces= ", - group->groupid); - return NULL; - } - } - } - - group =3D g_malloc0(sizeof(*group)); - - snprintf(path, sizeof(path), "/dev/vfio/%d", groupid); - group->fd =3D qemu_open_old(path, O_RDWR); - if (group->fd < 0) { - error_setg_errno(errp, errno, "failed to open %s", path); - goto free_group_exit; - } - - if (ioctl(group->fd, VFIO_GROUP_GET_STATUS, &status)) { - error_setg_errno(errp, errno, "failed to get group %d status", gro= upid); - goto close_fd_exit; - } - - if (!(status.flags & VFIO_GROUP_FLAGS_VIABLE)) { - error_setg(errp, "group %d is not viable", groupid); - error_append_hint(errp, - "Please ensure all devices within the iommu_grou= p " - "are bound to their vfio bus driver.\n"); - goto close_fd_exit; - } - - group->groupid =3D groupid; - QLIST_INIT(&group->device_list); - - if (vfio_connect_container(group, as, errp)) { - error_prepend(errp, "failed to setup container for group %d: ", - groupid); - goto close_fd_exit; - } - - QLIST_INSERT_HEAD(&vfio_group_list, group, next); - - return group; - -close_fd_exit: - close(group->fd); - -free_group_exit: - g_free(group); - - return NULL; -} - -void vfio_put_group(VFIOGroup *group) -{ - if (!group || !QLIST_EMPTY(&group->device_list)) { - return; - } - - if (!group->ram_block_discard_allowed) { - vfio_ram_block_discard_disable(group->container, false); - } - vfio_kvm_device_del_group(group); - vfio_disconnect_container(group); - QLIST_REMOVE(group, next); - trace_vfio_put_group(group->fd); - close(group->fd); - g_free(group); -} - struct vfio_device_info *vfio_get_device_info(int fd) { struct vfio_device_info *info; @@ -2413,168 +1525,3 @@ retry: =20 return info; } - -int vfio_get_device(VFIOGroup *group, const char *name, - VFIODevice *vbasedev, Error **errp) -{ - g_autofree struct vfio_device_info *info =3D NULL; - int fd; - - fd =3D ioctl(group->fd, VFIO_GROUP_GET_DEVICE_FD, name); - if (fd < 0) { - error_setg_errno(errp, errno, "error getting device from group %d", - group->groupid); - error_append_hint(errp, - "Verify all devices in group %d are bound to vfio- " - "or pci-stub and not already in use\n", group->group= id); - return fd; - } - - info =3D vfio_get_device_info(fd); - if (!info) { - error_setg_errno(errp, errno, "error getting device info"); - close(fd); - return -1; - } - - /* - * Set discarding of RAM as not broken for this group if the driver kn= ows - * the device operates compatibly with discarding. Setting must be - * consistent per group, but since compatibility is really only possib= le - * with mdev currently, we expect singleton groups. - */ - if (vbasedev->ram_block_discard_allowed !=3D - group->ram_block_discard_allowed) { - if (!QLIST_EMPTY(&group->device_list)) { - error_setg(errp, "Inconsistent setting of support for discardi= ng " - "RAM (e.g., balloon) within group"); - close(fd); - return -1; - } - - if (!group->ram_block_discard_allowed) { - group->ram_block_discard_allowed =3D true; - vfio_ram_block_discard_disable(group->container, false); - } - } - - vbasedev->fd =3D fd; - vbasedev->group =3D group; - QLIST_INSERT_HEAD(&group->device_list, vbasedev, next); - - vbasedev->num_irqs =3D info->num_irqs; - vbasedev->num_regions =3D info->num_regions; - vbasedev->flags =3D info->flags; - - trace_vfio_get_device(name, info->flags, info->num_regions, info->num_= irqs); - - vbasedev->reset_works =3D !!(info->flags & VFIO_DEVICE_FLAGS_RESET); - - return 0; -} - -void vfio_put_base_device(VFIODevice *vbasedev) -{ - if (!vbasedev->group) { - return; - } - QLIST_REMOVE(vbasedev, next); - vbasedev->group =3D NULL; - trace_vfio_put_base_device(vbasedev->fd); - close(vbasedev->fd); -} - -/* - * Interfaces for IBM EEH (Enhanced Error Handling) - */ -static bool vfio_eeh_container_ok(VFIOContainer *container) -{ - /* - * As of 2016-03-04 (linux-4.5) the host kernel EEH/VFIO - * implementation is broken if there are multiple groups in a - * container. The hardware works in units of Partitionable - * Endpoints (=3D=3D IOMMU groups) and the EEH operations naively - * iterate across all groups in the container, without any logic - * to make sure the groups have their state synchronized. For - * certain operations (ENABLE) that might be ok, until an error - * occurs, but for others (GET_STATE) it's clearly broken. - */ - - /* - * XXX Once fixed kernels exist, test for them here - */ - - if (QLIST_EMPTY(&container->group_list)) { - return false; - } - - if (QLIST_NEXT(QLIST_FIRST(&container->group_list), container_next)) { - return false; - } - - return true; -} - -static int vfio_eeh_container_op(VFIOContainer *container, uint32_t op) -{ - struct vfio_eeh_pe_op pe_op =3D { - .argsz =3D sizeof(pe_op), - .op =3D op, - }; - int ret; - - if (!vfio_eeh_container_ok(container)) { - error_report("vfio/eeh: EEH_PE_OP 0x%x: " - "kernel requires a container with exactly one group",= op); - return -EPERM; - } - - ret =3D ioctl(container->fd, VFIO_EEH_PE_OP, &pe_op); - if (ret < 0) { - error_report("vfio/eeh: EEH_PE_OP 0x%x failed: %m", op); - return -errno; - } - - return ret; -} - -static VFIOContainer *vfio_eeh_as_container(AddressSpace *as) -{ - VFIOAddressSpace *space =3D vfio_get_address_space(as); - VFIOContainer *container =3D NULL; - - if (QLIST_EMPTY(&space->containers)) { - /* No containers to act on */ - goto out; - } - - container =3D QLIST_FIRST(&space->containers); - - if (QLIST_NEXT(container, next)) { - /* We don't yet have logic to synchronize EEH state across - * multiple containers */ - container =3D NULL; - goto out; - } - -out: - vfio_put_address_space(space); - return container; -} - -bool vfio_eeh_as_ok(AddressSpace *as) -{ - VFIOContainer *container =3D vfio_eeh_as_container(as); - - return (container !=3D NULL) && vfio_eeh_container_ok(container); -} - -int vfio_eeh_as_op(AddressSpace *as, uint32_t op) -{ - VFIOContainer *container =3D vfio_eeh_as_container(as); - - if (!container) { - return -ENODEV; - } - return vfio_eeh_container_op(container, op); -} diff --git a/hw/vfio/container.c b/hw/vfio/container.c new file mode 100644 index 0000000000..175cdbbdff --- /dev/null +++ b/hw/vfio/container.c @@ -0,0 +1,1085 @@ +/* + * generic functions used by VFIO devices + * + * Copyright Red Hat, Inc. 2012 + * + * Authors: + * Alex Williamson + * + * This work is licensed under the terms of the GNU GPL, version 2. See + * the COPYING file in the top-level directory. + * + * Based on qemu-kvm device-assignment: + * Adapted for KVM by Qumranet. + * Copyright (c) 2007, Neocleus, Alex Novik (alex@neocleus.com) + * Copyright (c) 2007, Neocleus, Guy Zana (guy@neocleus.com) + * Copyright (C) 2008, Qumranet, Amit Shah (amit.shah@qumranet.com) + * Copyright (C) 2008, Red Hat, Amit Shah (amit.shah@redhat.com) + * Copyright (C) 2008, IBM, Muli Ben-Yehuda (muli@il.ibm.com) + */ + +#include "qemu/osdep.h" +#include +#ifdef CONFIG_KVM +#include +#endif +#include + +#include "hw/vfio/vfio-common.h" +#include "hw/vfio/vfio.h" +#include "exec/address-spaces.h" +#include "exec/memory.h" +#include "exec/ram_addr.h" +#include "hw/hw.h" +#include "qemu/error-report.h" +#include "qemu/range.h" +#include "sysemu/kvm.h" +#include "sysemu/reset.h" +#include "trace.h" +#include "qapi/error.h" +#include "migration/migration.h" + +VFIOGroupList vfio_group_list =3D + QLIST_HEAD_INITIALIZER(vfio_group_list); + +static int vfio_ram_block_discard_disable(VFIOContainer *container, bool s= tate) +{ + switch (container->iommu_type) { + case VFIO_TYPE1v2_IOMMU: + case VFIO_TYPE1_IOMMU: + /* + * We support coordinated discarding of RAM via the RamDiscardMana= ger. + */ + return ram_block_uncoordinated_discard_disable(state); + default: + /* + * VFIO_SPAPR_TCE_IOMMU most probably works just fine with + * RamDiscardManager, however, it is completely untested. + * + * VFIO_SPAPR_TCE_v2_IOMMU with "DMA memory preregistering" does + * completely the opposite of managing mapping/pinning dynamically= as + * required by RamDiscardManager. We would have to special-case se= ctions + * with a RamDiscardManager. + */ + return ram_block_discard_disable(state); + } +} + +VFIODevice *vfio_container_dev_iter_next(VFIOContainer *container, + VFIODevice *curr) +{ + VFIOGroup *group; + + if (!curr) { + group =3D QLIST_FIRST(&container->group_list); + } else { + if (curr->next.le_next) { + return curr->next.le_next; + } + group =3D curr->group->container_next.le_next; + } + + if (!group) { + return NULL; + } + return QLIST_FIRST(&group->device_list); +} + +static int vfio_dma_unmap_bitmap(VFIOContainer *container, + hwaddr iova, ram_addr_t size, + IOMMUTLBEntry *iotlb) +{ + struct vfio_iommu_type1_dma_unmap *unmap; + struct vfio_bitmap *bitmap; + VFIOBitmap vbmap; + int ret; + + ret =3D vfio_bitmap_alloc(&vbmap, size); + if (ret) { + return ret; + } + + unmap =3D g_malloc0(sizeof(*unmap) + sizeof(*bitmap)); + + unmap->argsz =3D sizeof(*unmap) + sizeof(*bitmap); + unmap->iova =3D iova; + unmap->size =3D size; + unmap->flags |=3D VFIO_DMA_UNMAP_FLAG_GET_DIRTY_BITMAP; + bitmap =3D (struct vfio_bitmap *)&unmap->data; + + /* + * cpu_physical_memory_set_dirty_lebitmap() supports pages in bitmap of + * qemu_real_host_page_size to mark those dirty. Hence set bitmap_pgsi= ze + * to qemu_real_host_page_size. + */ + bitmap->pgsize =3D qemu_real_host_page_size(); + bitmap->size =3D vbmap.size; + bitmap->data =3D (__u64 *)vbmap.bitmap; + + if (vbmap.size > container->max_dirty_bitmap_size) { + error_report("UNMAP: Size of bitmap too big 0x%"PRIx64, vbmap.size= ); + ret =3D -E2BIG; + goto unmap_exit; + } + + ret =3D ioctl(container->fd, VFIO_IOMMU_UNMAP_DMA, unmap); + if (!ret) { + cpu_physical_memory_set_dirty_lebitmap(vbmap.bitmap, + iotlb->translated_addr, vbmap.pages); + } else { + error_report("VFIO_UNMAP_DMA with DIRTY_BITMAP : %m"); + } + +unmap_exit: + g_free(unmap); + g_free(vbmap.bitmap); + + return ret; +} + +/* + * DMA - Mapping and unmapping for the "type1" IOMMU interface used on x86 + */ +int vfio_dma_unmap(VFIOContainer *container, hwaddr iova, + ram_addr_t size, IOMMUTLBEntry *iotlb) +{ + struct vfio_iommu_type1_dma_unmap unmap =3D { + .argsz =3D sizeof(unmap), + .flags =3D 0, + .iova =3D iova, + .size =3D size, + }; + bool need_dirty_sync =3D false; + int ret; + + if (iotlb && vfio_devices_all_running_and_mig_active(container)) { + if (!vfio_devices_all_device_dirty_tracking(container) && + container->dirty_pages_supported) { + return vfio_dma_unmap_bitmap(container, iova, size, iotlb); + } + + need_dirty_sync =3D true; + } + + while (ioctl(container->fd, VFIO_IOMMU_UNMAP_DMA, &unmap)) { + /* + * The type1 backend has an off-by-one bug in the kernel (71a7d3d7= 8e3c + * v4.15) where an overflow in its wrap-around check prevents us f= rom + * unmapping the last page of the address space. Test for the err= or + * condition and re-try the unmap excluding the last page. The + * expectation is that we've never mapped the last page anyway and= this + * unmap request comes via vIOMMU support which also makes it unli= kely + * that this page is used. This bug was introduced well after typ= e1 v2 + * support was introduced, so we shouldn't need to test for v1. A= fix + * is queued for kernel v5.0 so this workaround can be removed once + * affected kernels are sufficiently deprecated. + */ + if (errno =3D=3D EINVAL && unmap.size && !(unmap.iova + unmap.size= ) && + container->iommu_type =3D=3D VFIO_TYPE1v2_IOMMU) { + trace_vfio_dma_unmap_overflow_workaround(); + unmap.size -=3D 1ULL << ctz64(container->pgsizes); + continue; + } + error_report("VFIO_UNMAP_DMA failed: %s", strerror(errno)); + return -errno; + } + + if (need_dirty_sync) { + ret =3D vfio_get_dirty_bitmap(container, iova, size, + iotlb->translated_addr); + if (ret) { + return ret; + } + } + + return 0; +} + +int vfio_dma_map(VFIOContainer *container, hwaddr iova, + ram_addr_t size, void *vaddr, bool readonly) +{ + struct vfio_iommu_type1_dma_map map =3D { + .argsz =3D sizeof(map), + .flags =3D VFIO_DMA_MAP_FLAG_READ, + .vaddr =3D (__u64)(uintptr_t)vaddr, + .iova =3D iova, + .size =3D size, + }; + + if (!readonly) { + map.flags |=3D VFIO_DMA_MAP_FLAG_WRITE; + } + + /* + * Try the mapping, if it fails with EBUSY, unmap the region and try + * again. This shouldn't be necessary, but we sometimes see it in + * the VGA ROM space. + */ + if (ioctl(container->fd, VFIO_IOMMU_MAP_DMA, &map) =3D=3D 0 || + (errno =3D=3D EBUSY && vfio_dma_unmap(container, iova, size, NULL)= =3D=3D 0 && + ioctl(container->fd, VFIO_IOMMU_MAP_DMA, &map) =3D=3D 0)) { + return 0; + } + + error_report("VFIO_MAP_DMA failed: %s", strerror(errno)); + return -errno; +} + +int vfio_set_dirty_page_tracking(VFIOContainer *container, bool start) +{ + int ret; + struct vfio_iommu_type1_dirty_bitmap dirty =3D { + .argsz =3D sizeof(dirty), + }; + + if (!container->dirty_pages_supported) { + return 0; + } + + if (start) { + dirty.flags =3D VFIO_IOMMU_DIRTY_PAGES_FLAG_START; + } else { + dirty.flags =3D VFIO_IOMMU_DIRTY_PAGES_FLAG_STOP; + } + + ret =3D ioctl(container->fd, VFIO_IOMMU_DIRTY_PAGES, &dirty); + if (ret) { + ret =3D -errno; + error_report("Failed to set dirty tracking flag 0x%x errno: %d", + dirty.flags, errno); + } + + return ret; +} + +int vfio_query_dirty_bitmap(VFIOContainer *container, VFIOBitmap *vbmap, + hwaddr iova, hwaddr size) +{ + struct vfio_iommu_type1_dirty_bitmap *dbitmap; + struct vfio_iommu_type1_dirty_bitmap_get *range; + int ret; + + dbitmap =3D g_malloc0(sizeof(*dbitmap) + sizeof(*range)); + + dbitmap->argsz =3D sizeof(*dbitmap) + sizeof(*range); + dbitmap->flags =3D VFIO_IOMMU_DIRTY_PAGES_FLAG_GET_BITMAP; + range =3D (struct vfio_iommu_type1_dirty_bitmap_get *)&dbitmap->data; + range->iova =3D iova; + range->size =3D size; + + /* + * cpu_physical_memory_set_dirty_lebitmap() supports pages in bitmap of + * qemu_real_host_page_size to mark those dirty. Hence set bitmap's pg= size + * to qemu_real_host_page_size. + */ + range->bitmap.pgsize =3D qemu_real_host_page_size(); + range->bitmap.size =3D vbmap->size; + range->bitmap.data =3D (__u64 *)vbmap->bitmap; + + ret =3D ioctl(container->fd, VFIO_IOMMU_DIRTY_PAGES, dbitmap); + if (ret) { + ret =3D -errno; + error_report("Failed to get dirty bitmap for iova: 0x%"PRIx64 + " size: 0x%"PRIx64" err: %d", (uint64_t)range->iova, + (uint64_t)range->size, errno); + } + + g_free(dbitmap); + + return ret; +} + +static void vfio_listener_release(VFIOContainer *container) +{ + memory_listener_unregister(&container->listener); + if (container->iommu_type =3D=3D VFIO_SPAPR_TCE_v2_IOMMU) { + memory_listener_unregister(&container->prereg_listener); + } +} + +int vfio_container_add_section_window(VFIOContainer *container, + MemoryRegionSection *section, + Error **errp) +{ + VFIOHostDMAWindow *hostwin; + hwaddr pgsize =3D 0; + int ret; + + if (container->iommu_type !=3D VFIO_SPAPR_TCE_v2_IOMMU) { + return 0; + } + + /* For now intersections are not allowed, we may relax this later */ + QLIST_FOREACH(hostwin, &container->hostwin_list, hostwin_next) { + if (ranges_overlap(hostwin->min_iova, + hostwin->max_iova - hostwin->min_iova + 1, + section->offset_within_address_space, + int128_get64(section->size))) { + error_setg(errp, + "region [0x%"PRIx64",0x%"PRIx64"] overlaps with existing" + "host DMA window [0x%"PRIx64",0x%"PRIx64"]", + section->offset_within_address_space, + section->offset_within_address_space + + int128_get64(section->size) - 1, + hostwin->min_iova, hostwin->max_iova); + return -EINVAL; + } + } + + ret =3D vfio_spapr_create_window(container, section, &pgsize); + if (ret) { + error_setg_errno(errp, -ret, "Failed to create SPAPR window"); + return ret; + } + + vfio_host_win_add(container, section->offset_within_address_space, + section->offset_within_address_space + + int128_get64(section->size) - 1, pgsize); +#ifdef CONFIG_KVM + if (kvm_enabled()) { + VFIOGroup *group; + IOMMUMemoryRegion *iommu_mr =3D IOMMU_MEMORY_REGION(section->mr); + struct kvm_vfio_spapr_tce param; + struct kvm_device_attr attr =3D { + .group =3D KVM_DEV_VFIO_GROUP, + .attr =3D KVM_DEV_VFIO_GROUP_SET_SPAPR_TCE, + .addr =3D (uint64_t)(unsigned long)¶m, + }; + + if (!memory_region_iommu_get_attr(iommu_mr, IOMMU_ATTR_SPAPR_TCE_F= D, + ¶m.tablefd)) { + QLIST_FOREACH(group, &container->group_list, container_next) { + param.groupfd =3D group->fd; + if (ioctl(vfio_kvm_device_fd, KVM_SET_DEVICE_ATTR, &attr))= { + error_report("vfio: failed to setup fd %d " + "for a group with fd %d: %s", + param.tablefd, param.groupfd, + strerror(errno)); + return 0; + } + trace_vfio_spapr_group_attach(param.groupfd, param.tablefd= ); + } + } + } +#endif + return 0; +} + +void vfio_container_del_section_window(VFIOContainer *container, + MemoryRegionSection *section) +{ + if (container->iommu_type !=3D VFIO_SPAPR_TCE_v2_IOMMU) { + return; + } + + vfio_spapr_remove_window(container, + section->offset_within_address_space); + if (vfio_host_win_del(container, + section->offset_within_address_space, + section->offset_within_address_space + + int128_get64(section->size) - 1) < 0) { + hw_error("%s: Cannot delete missing window at %"HWADDR_PRIx, + __func__, section->offset_within_address_space); + } +} + +static struct vfio_info_cap_header * +vfio_get_iommu_type1_info_cap(struct vfio_iommu_type1_info *info, uint16_t= id) +{ + if (!(info->flags & VFIO_IOMMU_INFO_CAPS)) { + return NULL; + } + + return vfio_get_cap((void *)info, info->cap_offset, id); +} + +bool vfio_get_info_dma_avail(struct vfio_iommu_type1_info *info, + unsigned int *avail) +{ + struct vfio_info_cap_header *hdr; + struct vfio_iommu_type1_info_dma_avail *cap; + + /* If the capability cannot be found, assume no DMA limiting */ + hdr =3D vfio_get_iommu_type1_info_cap(info, + VFIO_IOMMU_TYPE1_INFO_DMA_AVAIL); + if (hdr =3D=3D NULL) { + return false; + } + + if (avail !=3D NULL) { + cap =3D (void *) hdr; + *avail =3D cap->avail; + } + + return true; +} + +static void vfio_kvm_device_add_group(VFIOGroup *group) +{ + vfio_kvm_device_add_fd(group->fd); +} + +static void vfio_kvm_device_del_group(VFIOGroup *group) +{ + vfio_kvm_device_del_fd(group->fd); +} + +/* + * vfio_get_iommu_type - selects the richest iommu_type (v2 first) + */ +static int vfio_get_iommu_type(VFIOContainer *container, + Error **errp) +{ + int iommu_types[] =3D { VFIO_TYPE1v2_IOMMU, VFIO_TYPE1_IOMMU, + VFIO_SPAPR_TCE_v2_IOMMU, VFIO_SPAPR_TCE_IOMMU }; + int i; + + for (i =3D 0; i < ARRAY_SIZE(iommu_types); i++) { + if (ioctl(container->fd, VFIO_CHECK_EXTENSION, iommu_types[i])) { + return iommu_types[i]; + } + } + error_setg(errp, "No available IOMMU models"); + return -EINVAL; +} + +static int vfio_init_container(VFIOContainer *container, int group_fd, + Error **errp) +{ + int iommu_type, ret; + + iommu_type =3D vfio_get_iommu_type(container, errp); + if (iommu_type < 0) { + return iommu_type; + } + + ret =3D ioctl(group_fd, VFIO_GROUP_SET_CONTAINER, &container->fd); + if (ret) { + error_setg_errno(errp, errno, "Failed to set group container"); + return -errno; + } + + while (ioctl(container->fd, VFIO_SET_IOMMU, iommu_type)) { + if (iommu_type =3D=3D VFIO_SPAPR_TCE_v2_IOMMU) { + /* + * On sPAPR, despite the IOMMU subdriver always advertises v1 = and + * v2, the running platform may not support v2 and there is no + * way to guess it until an IOMMU group gets added to the cont= ainer. + * So in case it fails with v2, try v1 as a fallback. + */ + iommu_type =3D VFIO_SPAPR_TCE_IOMMU; + continue; + } + error_setg_errno(errp, errno, "Failed to set iommu for container"); + return -errno; + } + + container->iommu_type =3D iommu_type; + return 0; +} + +static int vfio_get_iommu_info(VFIOContainer *container, + struct vfio_iommu_type1_info **info) +{ + + size_t argsz =3D sizeof(struct vfio_iommu_type1_info); + + *info =3D g_new0(struct vfio_iommu_type1_info, 1); +again: + (*info)->argsz =3D argsz; + + if (ioctl(container->fd, VFIO_IOMMU_GET_INFO, *info)) { + g_free(*info); + *info =3D NULL; + return -errno; + } + + if (((*info)->argsz > argsz)) { + argsz =3D (*info)->argsz; + *info =3D g_realloc(*info, argsz); + goto again; + } + + return 0; +} + +static struct vfio_info_cap_header * +vfio_get_iommu_info_cap(struct vfio_iommu_type1_info *info, uint16_t id) +{ + struct vfio_info_cap_header *hdr; + void *ptr =3D info; + + if (!(info->flags & VFIO_IOMMU_INFO_CAPS)) { + return NULL; + } + + for (hdr =3D ptr + info->cap_offset; hdr !=3D ptr; hdr =3D ptr + hdr->= next) { + if (hdr->id =3D=3D id) { + return hdr; + } + } + + return NULL; +} + +static void vfio_get_iommu_info_migration(VFIOContainer *container, + struct vfio_iommu_type1_info *inf= o) +{ + struct vfio_info_cap_header *hdr; + struct vfio_iommu_type1_info_cap_migration *cap_mig; + + hdr =3D vfio_get_iommu_info_cap(info, VFIO_IOMMU_TYPE1_INFO_CAP_MIGRAT= ION); + if (!hdr) { + return; + } + + cap_mig =3D container_of(hdr, struct vfio_iommu_type1_info_cap_migrati= on, + header); + + /* + * cpu_physical_memory_set_dirty_lebitmap() supports pages in bitmap of + * qemu_real_host_page_size to mark those dirty. + */ + if (cap_mig->pgsize_bitmap & qemu_real_host_page_size()) { + container->dirty_pages_supported =3D true; + container->max_dirty_bitmap_size =3D cap_mig->max_dirty_bitmap_siz= e; + container->dirty_pgsizes =3D cap_mig->pgsize_bitmap; + } +} + +static int vfio_connect_container(VFIOGroup *group, AddressSpace *as, + Error **errp) +{ + VFIOContainer *container; + int ret, fd; + VFIOAddressSpace *space; + + space =3D vfio_get_address_space(as); + + /* + * VFIO is currently incompatible with discarding of RAM insofar as the + * madvise to purge (zap) the page from QEMU's address space does not + * interact with the memory API and therefore leaves stale virtual to + * physical mappings in the IOMMU if the page was previously pinned. = We + * therefore set discarding broken for each group added to a container, + * whether the container is used individually or shared. This provides + * us with options to allow devices within a group to opt-in and allow + * discarding, so long as it is done consistently for a group (for ins= tance + * if the device is an mdev device where it is known that the host ven= dor + * driver will never pin pages outside of the working set of the guest + * driver, which would thus not be discarding candidates). + * + * The first opportunity to induce pinning occurs here where we attemp= t to + * attach the group to existing containers within the AddressSpace. I= f any + * pages are already zapped from the virtual address space, such as fr= om + * previous discards, new pinning will cause valid mappings to be + * re-established. Likewise, when the overall MemoryListener for a new + * container is registered, a replay of mappings within the AddressSpa= ce + * will occur, re-establishing any previously zapped pages as well. + * + * Especially virtio-balloon is currently only prevented from discardi= ng + * new memory, it will not yet set ram_block_discard_set_required() and + * therefore, neither stops us here or deals with the sudden memory + * consumption of inflated memory. + * + * We do support discarding of memory coordinated via the RamDiscardMa= nager + * with some IOMMU types. vfio_ram_block_discard_disable() handles the + * details once we know which type of IOMMU we are using. + */ + + QLIST_FOREACH(container, &space->containers, next) { + if (!ioctl(group->fd, VFIO_GROUP_SET_CONTAINER, &container->fd)) { + ret =3D vfio_ram_block_discard_disable(container, true); + if (ret) { + error_setg_errno(errp, -ret, + "Cannot set discarding of RAM broken"); + if (ioctl(group->fd, VFIO_GROUP_UNSET_CONTAINER, + &container->fd)) { + error_report("vfio: error disconnecting group %d from" + " container", group->groupid); + } + return ret; + } + group->container =3D container; + QLIST_INSERT_HEAD(&container->group_list, group, container_nex= t); + vfio_kvm_device_add_group(group); + return 0; + } + } + + fd =3D qemu_open_old("/dev/vfio/vfio", O_RDWR); + if (fd < 0) { + error_setg_errno(errp, errno, "failed to open /dev/vfio/vfio"); + ret =3D -errno; + goto put_space_exit; + } + + ret =3D ioctl(fd, VFIO_GET_API_VERSION); + if (ret !=3D VFIO_API_VERSION) { + error_setg(errp, "supported vfio version: %d, " + "reported version: %d", VFIO_API_VERSION, ret); + ret =3D -EINVAL; + goto close_fd_exit; + } + + container =3D g_malloc0(sizeof(*container)); + container->space =3D space; + container->fd =3D fd; + container->error =3D NULL; + container->dirty_pages_supported =3D false; + container->dma_max_mappings =3D 0; + QLIST_INIT(&container->giommu_list); + QLIST_INIT(&container->hostwin_list); + QLIST_INIT(&container->vrdl_list); + + ret =3D vfio_init_container(container, group->fd, errp); + if (ret) { + goto free_container_exit; + } + + ret =3D vfio_ram_block_discard_disable(container, true); + if (ret) { + error_setg_errno(errp, -ret, "Cannot set discarding of RAM broken"= ); + goto free_container_exit; + } + + switch (container->iommu_type) { + case VFIO_TYPE1v2_IOMMU: + case VFIO_TYPE1_IOMMU: + { + struct vfio_iommu_type1_info *info; + + ret =3D vfio_get_iommu_info(container, &info); + if (ret) { + error_setg_errno(errp, -ret, "Failed to get VFIO IOMMU info"); + goto enable_discards_exit; + } + + if (info->flags & VFIO_IOMMU_INFO_PGSIZES) { + container->pgsizes =3D info->iova_pgsizes; + } else { + container->pgsizes =3D qemu_real_host_page_size(); + } + + if (!vfio_get_info_dma_avail(info, &container->dma_max_mappings)) { + container->dma_max_mappings =3D 65535; + } + vfio_get_iommu_info_migration(container, info); + g_free(info); + + /* + * FIXME: We should parse VFIO_IOMMU_TYPE1_INFO_CAP_IOVA_RANGE + * information to get the actual window extent rather than assume + * a 64-bit IOVA address space. + */ + vfio_host_win_add(container, 0, (hwaddr)-1, container->pgsizes); + + break; + } + case VFIO_SPAPR_TCE_v2_IOMMU: + case VFIO_SPAPR_TCE_IOMMU: + { + struct vfio_iommu_spapr_tce_info info; + bool v2 =3D container->iommu_type =3D=3D VFIO_SPAPR_TCE_v2_IOMMU; + + /* + * The host kernel code implementing VFIO_IOMMU_DISABLE is called + * when container fd is closed so we do not call it explicitly + * in this file. + */ + if (!v2) { + ret =3D ioctl(fd, VFIO_IOMMU_ENABLE); + if (ret) { + error_setg_errno(errp, errno, "failed to enable container"= ); + ret =3D -errno; + goto enable_discards_exit; + } + } else { + container->prereg_listener =3D vfio_prereg_listener; + + memory_listener_register(&container->prereg_listener, + &address_space_memory); + if (container->error) { + memory_listener_unregister(&container->prereg_listener); + ret =3D -1; + error_propagate_prepend(errp, container->error, + "RAM memory listener initialization failed: "); + goto enable_discards_exit; + } + } + + info.argsz =3D sizeof(info); + ret =3D ioctl(fd, VFIO_IOMMU_SPAPR_TCE_GET_INFO, &info); + if (ret) { + error_setg_errno(errp, errno, + "VFIO_IOMMU_SPAPR_TCE_GET_INFO failed"); + ret =3D -errno; + if (v2) { + memory_listener_unregister(&container->prereg_listener); + } + goto enable_discards_exit; + } + + if (v2) { + container->pgsizes =3D info.ddw.pgsizes; + /* + * There is a default window in just created container. + * To make region_add/del simpler, we better remove this + * window now and let those iommu_listener callbacks + * create/remove them when needed. + */ + ret =3D vfio_spapr_remove_window(container, info.dma32_window_= start); + if (ret) { + error_setg_errno(errp, -ret, + "failed to remove existing window"); + goto enable_discards_exit; + } + } else { + /* The default table uses 4K pages */ + container->pgsizes =3D 0x1000; + vfio_host_win_add(container, info.dma32_window_start, + info.dma32_window_start + + info.dma32_window_size - 1, + 0x1000); + } + } + } + + vfio_kvm_device_add_group(group); + + QLIST_INIT(&container->group_list); + QLIST_INSERT_HEAD(&space->containers, container, next); + + group->container =3D container; + QLIST_INSERT_HEAD(&container->group_list, group, container_next); + + container->listener =3D vfio_memory_listener; + + memory_listener_register(&container->listener, container->space->as); + + if (container->error) { + ret =3D -1; + error_propagate_prepend(errp, container->error, + "memory listener initialization failed: "); + goto listener_release_exit; + } + + container->initialized =3D true; + + return 0; +listener_release_exit: + QLIST_REMOVE(group, container_next); + QLIST_REMOVE(container, next); + vfio_kvm_device_del_group(group); + vfio_listener_release(container); + +enable_discards_exit: + vfio_ram_block_discard_disable(container, false); + +free_container_exit: + g_free(container); + +close_fd_exit: + close(fd); + +put_space_exit: + vfio_put_address_space(space); + + return ret; +} + +static void vfio_disconnect_container(VFIOGroup *group) +{ + VFIOContainer *container =3D group->container; + + QLIST_REMOVE(group, container_next); + group->container =3D NULL; + + /* + * Explicitly release the listener first before unset container, + * since unset may destroy the backend container if it's the last + * group. + */ + if (QLIST_EMPTY(&container->group_list)) { + vfio_listener_release(container); + } + + if (ioctl(group->fd, VFIO_GROUP_UNSET_CONTAINER, &container->fd)) { + error_report("vfio: error disconnecting group %d from container", + group->groupid); + } + + if (QLIST_EMPTY(&container->group_list)) { + VFIOAddressSpace *space =3D container->space; + VFIOGuestIOMMU *giommu, *tmp; + VFIOHostDMAWindow *hostwin, *next; + + QLIST_REMOVE(container, next); + + QLIST_FOREACH_SAFE(giommu, &container->giommu_list, giommu_next, t= mp) { + memory_region_unregister_iommu_notifier( + MEMORY_REGION(giommu->iommu_mr), &giommu->n); + QLIST_REMOVE(giommu, giommu_next); + g_free(giommu); + } + + QLIST_FOREACH_SAFE(hostwin, &container->hostwin_list, hostwin_next, + next) { + QLIST_REMOVE(hostwin, hostwin_next); + g_free(hostwin); + } + + trace_vfio_disconnect_container(container->fd); + close(container->fd); + g_free(container); + + vfio_put_address_space(space); + } +} + +VFIOGroup *vfio_get_group(int groupid, AddressSpace *as, Error **errp) +{ + VFIOGroup *group; + char path[32]; + struct vfio_group_status status =3D { .argsz =3D sizeof(status) }; + + QLIST_FOREACH(group, &vfio_group_list, next) { + if (group->groupid =3D=3D groupid) { + /* Found it. Now is it already in the right context? */ + if (group->container->space->as =3D=3D as) { + return group; + } else { + error_setg(errp, "group %d used in multiple address spaces= ", + group->groupid); + return NULL; + } + } + } + + group =3D g_malloc0(sizeof(*group)); + + snprintf(path, sizeof(path), "/dev/vfio/%d", groupid); + group->fd =3D qemu_open_old(path, O_RDWR); + if (group->fd < 0) { + error_setg_errno(errp, errno, "failed to open %s", path); + goto free_group_exit; + } + + if (ioctl(group->fd, VFIO_GROUP_GET_STATUS, &status)) { + error_setg_errno(errp, errno, "failed to get group %d status", gro= upid); + goto close_fd_exit; + } + + if (!(status.flags & VFIO_GROUP_FLAGS_VIABLE)) { + error_setg(errp, "group %d is not viable", groupid); + error_append_hint(errp, + "Please ensure all devices within the iommu_grou= p " + "are bound to their vfio bus driver.\n"); + goto close_fd_exit; + } + + group->groupid =3D groupid; + QLIST_INIT(&group->device_list); + + if (vfio_connect_container(group, as, errp)) { + error_prepend(errp, "failed to setup container for group %d: ", + groupid); + goto close_fd_exit; + } + + QLIST_INSERT_HEAD(&vfio_group_list, group, next); + + return group; + +close_fd_exit: + close(group->fd); + +free_group_exit: + g_free(group); + + return NULL; +} + +void vfio_put_group(VFIOGroup *group) +{ + if (!group || !QLIST_EMPTY(&group->device_list)) { + return; + } + + if (!group->ram_block_discard_allowed) { + vfio_ram_block_discard_disable(group->container, false); + } + vfio_kvm_device_del_group(group); + vfio_disconnect_container(group); + QLIST_REMOVE(group, next); + trace_vfio_put_group(group->fd); + close(group->fd); + g_free(group); +} + +int vfio_get_device(VFIOGroup *group, const char *name, + VFIODevice *vbasedev, Error **errp) +{ + g_autofree struct vfio_device_info *info =3D NULL; + int fd; + + fd =3D ioctl(group->fd, VFIO_GROUP_GET_DEVICE_FD, name); + if (fd < 0) { + error_setg_errno(errp, errno, "error getting device from group %d", + group->groupid); + error_append_hint(errp, + "Verify all devices in group %d are bound to vfio- " + "or pci-stub and not already in use\n", group->group= id); + return fd; + } + + info =3D vfio_get_device_info(fd); + if (!info) { + error_setg_errno(errp, errno, "error getting device info"); + close(fd); + return -1; + } + + /* + * Set discarding of RAM as not broken for this group if the driver kn= ows + * the device operates compatibly with discarding. Setting must be + * consistent per group, but since compatibility is really only possib= le + * with mdev currently, we expect singleton groups. + */ + if (vbasedev->ram_block_discard_allowed !=3D + group->ram_block_discard_allowed) { + if (!QLIST_EMPTY(&group->device_list)) { + error_setg(errp, "Inconsistent setting of support for discardi= ng " + "RAM (e.g., balloon) within group"); + close(fd); + return -1; + } + + if (!group->ram_block_discard_allowed) { + group->ram_block_discard_allowed =3D true; + vfio_ram_block_discard_disable(group->container, false); + } + } + + vbasedev->fd =3D fd; + vbasedev->group =3D group; + QLIST_INSERT_HEAD(&group->device_list, vbasedev, next); + + vbasedev->num_irqs =3D info->num_irqs; + vbasedev->num_regions =3D info->num_regions; + vbasedev->flags =3D info->flags; + + trace_vfio_get_device(name, info->flags, info->num_regions, info->num_= irqs); + + vbasedev->reset_works =3D !!(info->flags & VFIO_DEVICE_FLAGS_RESET); + + return 0; +} + +void vfio_put_base_device(VFIODevice *vbasedev) +{ + if (!vbasedev->group) { + return; + } + QLIST_REMOVE(vbasedev, next); + vbasedev->group =3D NULL; + trace_vfio_put_base_device(vbasedev->fd); + close(vbasedev->fd); +} + +/* + * Interfaces for IBM EEH (Enhanced Error Handling) + */ +static bool vfio_eeh_container_ok(VFIOContainer *container) +{ + /* + * As of 2016-03-04 (linux-4.5) the host kernel EEH/VFIO + * implementation is broken if there are multiple groups in a + * container. The hardware works in units of Partitionable + * Endpoints (=3D=3D IOMMU groups) and the EEH operations naively + * iterate across all groups in the container, without any logic + * to make sure the groups have their state synchronized. For + * certain operations (ENABLE) that might be ok, until an error + * occurs, but for others (GET_STATE) it's clearly broken. + */ + + /* + * XXX Once fixed kernels exist, test for them here + */ + + if (QLIST_EMPTY(&container->group_list)) { + return false; + } + + if (QLIST_NEXT(QLIST_FIRST(&container->group_list), container_next)) { + return false; + } + + return true; +} + +static int vfio_eeh_container_op(VFIOContainer *container, uint32_t op) +{ + struct vfio_eeh_pe_op pe_op =3D { + .argsz =3D sizeof(pe_op), + .op =3D op, + }; + int ret; + + if (!vfio_eeh_container_ok(container)) { + error_report("vfio/eeh: EEH_PE_OP 0x%x: " + "kernel requires a container with exactly one group",= op); + return -EPERM; + } + + ret =3D ioctl(container->fd, VFIO_EEH_PE_OP, &pe_op); + if (ret < 0) { + error_report("vfio/eeh: EEH_PE_OP 0x%x failed: %m", op); + return -errno; + } + + return ret; +} + +static VFIOContainer *vfio_eeh_as_container(AddressSpace *as) +{ + VFIOAddressSpace *space =3D vfio_get_address_space(as); + VFIOContainer *container =3D NULL; + + if (QLIST_EMPTY(&space->containers)) { + /* No containers to act on */ + goto out; + } + + container =3D QLIST_FIRST(&space->containers); + + if (QLIST_NEXT(container, next)) { + /* + * We don't yet have logic to synchronize EEH state across + * multiple containers + */ + container =3D NULL; + goto out; + } + +out: + vfio_put_address_space(space); + return container; +} + +bool vfio_eeh_as_ok(AddressSpace *as) +{ + VFIOContainer *container =3D vfio_eeh_as_container(as); + + return (container !=3D NULL) && vfio_eeh_container_ok(container); +} + +int vfio_eeh_as_op(AddressSpace *as, uint32_t op) +{ + VFIOContainer *container =3D vfio_eeh_as_container(as); + + if (!container) { + return -ENODEV; + } + return vfio_eeh_container_op(container, op); +} diff --git a/hw/vfio/meson.build b/hw/vfio/meson.build index 3746c9f984..2a6912c940 100644 --- a/hw/vfio/meson.build +++ b/hw/vfio/meson.build @@ -2,6 +2,7 @@ vfio_ss =3D ss.source_set() vfio_ss.add(files( 'helpers.c', 'common.c', + 'container.c', 'spapr.c', 'migration.c', )) diff --git a/include/hw/vfio/vfio-common.h b/include/hw/vfio/vfio-common.h index 598c3ce079..bb7f9fe9c4 100644 --- a/include/hw/vfio/vfio-common.h +++ b/include/hw/vfio/vfio-common.h @@ -33,6 +33,8 @@ =20 #define VFIO_MSG_PREFIX "vfio %s: " =20 +extern const MemoryListener vfio_memory_listener; + enum { VFIO_DEVICE_TYPE_PCI =3D 0, VFIO_DEVICE_TYPE_PLATFORM =3D 1, @@ -196,6 +198,38 @@ typedef struct VFIODisplay { } dmabuf; } VFIODisplay; =20 +typedef struct { + unsigned long *bitmap; + hwaddr size; + hwaddr pages; +} VFIOBitmap; + +void vfio_host_win_add(VFIOContainer *container, + hwaddr min_iova, hwaddr max_iova, + uint64_t iova_pgsizes); +int vfio_host_win_del(VFIOContainer *container, hwaddr min_iova, + hwaddr max_iova); +VFIOAddressSpace *vfio_get_address_space(AddressSpace *as); +void vfio_put_address_space(VFIOAddressSpace *space); +bool vfio_devices_all_running_and_saving(VFIOContainer *container); + +/* container->fd */ +VFIODevice *vfio_container_dev_iter_next(VFIOContainer *container, + VFIODevice *curr); +int vfio_dma_unmap(VFIOContainer *container, hwaddr iova, + ram_addr_t size, IOMMUTLBEntry *iotlb); +int vfio_dma_map(VFIOContainer *container, hwaddr iova, + ram_addr_t size, void *vaddr, bool readonly); +int vfio_set_dirty_page_tracking(VFIOContainer *container, bool start); +int vfio_query_dirty_bitmap(VFIOContainer *container, VFIOBitmap *vbmap, + hwaddr iova, hwaddr size); + +int vfio_container_add_section_window(VFIOContainer *container, + MemoryRegionSection *section, + Error **errp); +void vfio_container_del_section_window(VFIOContainer *container, + MemoryRegionSection *section); + void vfio_put_base_device(VFIODevice *vbasedev); void vfio_disable_irqindex(VFIODevice *vbasedev, int index); void vfio_unmask_single_irqindex(VFIODevice *vbasedev, int index); @@ -220,6 +254,8 @@ struct vfio_device_info *vfio_get_device_info(int fd); int vfio_get_device(VFIOGroup *group, const char *name, VFIODevice *vbasedev, Error **errp); =20 +extern int vfio_kvm_device_fd; + int vfio_kvm_device_add_fd(int fd); int vfio_kvm_device_del_fd(int fd); =20 @@ -260,4 +296,13 @@ int vfio_spapr_remove_window(VFIOContainer *container, bool vfio_migration_realize(VFIODevice *vbasedev, Error **errp); void vfio_migration_exit(VFIODevice *vbasedev); =20 +int vfio_bitmap_alloc(VFIOBitmap *vbmap, hwaddr size); +bool vfio_devices_all_running_and_mig_active(VFIOContainer *container); +bool vfio_devices_all_device_dirty_tracking(VFIOContainer *container); +int vfio_devices_query_dirty_bitmap(VFIOContainer *container, + VFIOBitmap *vbmap, hwaddr iova, + hwaddr size); +int vfio_get_dirty_bitmap(VFIOContainer *container, uint64_t iova, + uint64_t size, ram_addr_t ram_addr); + #endif /* HW_VFIO_VFIO_COMMON_H */ --=20 2.34.1 From nobody Wed May 15 22:12:51 2024 Delivered-To: importer@patchew.org Authentication-Results: mx.zohomail.com; dkim=pass header.i=@intel.com; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass(p=none dis=none) header.from=intel.com ARC-Seal: i=1; a=rsa-sha256; t=1693392845; cv=none; d=zohomail.com; s=zohoarc; b=SDkQF2fb8/bMd/oDAIfI0AMxg/KyJ00m0pXe1KwGr/fsJI9vz6ajywx5NZ5gnoWT30kjkQPtqUHyOT3R4lvwXPo4HHiEd7W0lYygLP6JMhBQQMxzfHnUO/k5mHVlF9tYp638LI0PT3hxPTHle3uBKdfBUgo2ebYoqUDb1meIyFY= ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=zohomail.com; s=zohoarc; t=1693392845; h=Content-Transfer-Encoding:Cc:Date:From:In-Reply-To:List-Subscribe:List-Post:List-Id:List-Archive:List-Help:List-Unsubscribe:MIME-Version:Message-ID:References:Sender:Subject:To; bh=cAviH8m72iVzYi9V0Ccm0rljm2MERKJ5WfHJVJO210M=; b=lxK5JTlinqixSUk0+ZnGqhd/HVBV7WrJ917/jKGgjQ6e9021+eglM1CWW7zkf4yNvfL/bQdAUFchYd0nColYvRW490E9fxaAbZF9igLQgTYJK2iH+qb4RSq2Db402RnwCqxKkm2xSlCYSM/jFpAjCrl1anm1dwGnjtz1eI1kT/c= ARC-Authentication-Results: i=1; mx.zohomail.com; dkim=pass header.i=@intel.com; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass header.from= (p=none dis=none) Return-Path: Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) by mx.zohomail.com with SMTPS id 1693392845589742.9630259951373; Wed, 30 Aug 2023 03:54:05 -0700 (PDT) Received: from localhost ([::1] helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1qbIpJ-0004M7-2v; Wed, 30 Aug 2023 06:53:33 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1qbIpB-0004HK-MT for qemu-devel@nongnu.org; Wed, 30 Aug 2023 06:53:25 -0400 Received: from mgamail.intel.com ([192.55.52.93]) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1qbIp8-0007gi-CN for qemu-devel@nongnu.org; Wed, 30 Aug 2023 06:53:24 -0400 Received: from orsmga006.jf.intel.com ([10.7.209.51]) by fmsmga102.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 30 Aug 2023 03:53:08 -0700 Received: from duan-server-s2600bt.bj.intel.com ([10.240.192.147]) by orsmga006-auth.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 30 Aug 2023 03:53:04 -0700 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1693392802; x=1724928802; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=j3l5fzbriBBFSfDLiWnYcZf/PKOqSMFOKrOrLx/Rjfc=; b=QAKVnjeTusgVnwD33aJ3qjltjG3wVdIgwOBbm8fu/067w7LiNibpJEI2 ZVtXKdZFXEZ7ptQC9K+NS8hcOlukQb1gLLZJbsZFWOxX7LPhJRPmAoCWG e6wyGfM+qd05Hx1JPmDNmpAc3onc2qNLT3rJDsUuNEGdvUJOCy1nFy/7s 4l6NHSV+UR4Wh9W72jDXEU7NjAZp70BsOswg3vWb2N49qdfkJNAIIwmBy nUTb9EYk+WPere1//2BZgGCq2s4KCYFSlO31V0TS0NqA++5wXDEiQ8nj/ 3S1p1rE+Xhr6i67XgyY89n1mzJhEiTQIlJlMnitwuRQfW8ukQCDjvjE19 A==; X-IronPort-AV: E=McAfee;i="6600,9927,10817"; a="373016510" X-IronPort-AV: E=Sophos;i="6.02,213,1688454000"; d="scan'208";a="373016510" X-ExtLoop1: 1 X-IronPort-AV: E=McAfee;i="6600,9927,10817"; a="715866102" X-IronPort-AV: E=Sophos;i="6.02,213,1688454000"; d="scan'208";a="715866102" From: Zhenzhong Duan To: qemu-devel@nongnu.org Cc: alex.williamson@redhat.com, clg@redhat.com, jgg@nvidia.com, nicolinc@nvidia.com, joao.m.martins@oracle.com, eric.auger@redhat.com, peterx@redhat.com, jasowang@redhat.com, kevin.tian@intel.com, yi.l.liu@intel.com, yi.y.sun@intel.com, chao.p.peng@intel.com, Zhenzhong Duan Subject: [PATCH v1 09/22] vfio/container: Introduce vfio_[attach/detach]_device Date: Wed, 30 Aug 2023 18:37:41 +0800 Message-Id: <20230830103754.36461-10-zhenzhong.duan@intel.com> X-Mailer: git-send-email 2.34.1 In-Reply-To: <20230830103754.36461-1-zhenzhong.duan@intel.com> References: <20230830103754.36461-1-zhenzhong.duan@intel.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Received-SPF: pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) client-ip=209.51.188.17; envelope-from=qemu-devel-bounces+importer=patchew.org@nongnu.org; helo=lists.gnu.org; Received-SPF: pass client-ip=192.55.52.93; envelope-from=zhenzhong.duan@intel.com; helo=mgamail.intel.com X-Spam_score_int: -43 X-Spam_score: -4.4 X-Spam_bar: ---- X-Spam_report: (-4.4 / 5.0 requ) BAYES_00=-1.9, DKIMWL_WL_HIGH=-0.001, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, RCVD_IN_DNSWL_MED=-2.3, SPF_HELO_NONE=0.001, SPF_PASS=-0.001 autolearn=ham autolearn_force=no X-Spam_action: no action X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: qemu-devel-bounces+importer=patchew.org@nongnu.org Sender: qemu-devel-bounces+importer=patchew.org@nongnu.org X-ZohoMail-DKIM: pass (identity @intel.com) X-ZM-MESSAGEID: 1693392925891100003 Content-Type: text/plain; charset="utf-8" From: Eric Auger We want the VFIO devices to be able to use two different IOMMU callbacks, the legacy VFIO one and the new iommufd one. Introduce vfio_[attach/detach]_device which aim at hiding the underlying IOMMU backend (IOCTLs, datatypes, ...). Once vfio_attach_device completes, the device is attached to a security context and its fd can be used. Conversely When vfio_detach_device completes, the device has been detached to the security context. In this patch, only the vfio-pci device gets converted to use the new API. Subsequent patches will handle other devices. Signed-off-by: Eric Auger Signed-off-by: Yi Liu Signed-off-by: Zhenzhong Duan --- hw/vfio/container.c | 66 +++++++++++++++++++++++++++++++++++ hw/vfio/pci.c | 50 ++++---------------------- hw/vfio/trace-events | 2 +- include/hw/vfio/vfio-common.h | 3 ++ 4 files changed, 76 insertions(+), 45 deletions(-) diff --git a/hw/vfio/container.c b/hw/vfio/container.c index 175cdbbdff..74556da0c7 100644 --- a/hw/vfio/container.c +++ b/hw/vfio/container.c @@ -1083,3 +1083,69 @@ int vfio_eeh_as_op(AddressSpace *as, uint32_t op) } return vfio_eeh_container_op(container, op); } + +static int vfio_device_groupid(VFIODevice *vbasedev, Error **errp) +{ + char *tmp, group_path[PATH_MAX], *group_name; + int ret, groupid; + ssize_t len; + + tmp =3D g_strdup_printf("%s/iommu_group", vbasedev->sysfsdev); + len =3D readlink(tmp, group_path, sizeof(group_path)); + g_free(tmp); + + if (len <=3D 0 || len >=3D sizeof(group_path)) { + ret =3D len < 0 ? -errno : -ENAMETOOLONG; + error_setg_errno(errp, -ret, "no iommu_group found"); + return ret; + } + + group_path[len] =3D 0; + + group_name =3D basename(group_path); + if (sscanf(group_name, "%d", &groupid) !=3D 1) { + error_setg_errno(errp, errno, "failed to read %s", group_path); + return -errno; + } + return groupid; +} + +int vfio_attach_device(char *name, VFIODevice *vbasedev, + AddressSpace *as, Error **errp) +{ + int groupid =3D vfio_device_groupid(vbasedev, errp); + VFIODevice *vbasedev_iter; + VFIOGroup *group; + int ret; + + if (groupid < 0) { + return groupid; + } + + group =3D vfio_get_group(groupid, as, errp); + if (!group) { + return -ENOENT; + } + + QLIST_FOREACH(vbasedev_iter, &group->device_list, next) { + if (strcmp(vbasedev_iter->name, vbasedev->name) =3D=3D 0) { + error_setg(errp, "device is already attached"); + vfio_put_group(group); + return -EBUSY; + } + } + ret =3D vfio_get_device(group, name, vbasedev, errp); + if (ret) { + vfio_put_group(group); + } + + return ret; +} + +void vfio_detach_device(VFIODevice *vbasedev) +{ + VFIOGroup *group =3D vbasedev->group; + + vfio_put_base_device(vbasedev); + vfio_put_group(group); +} diff --git a/hw/vfio/pci.c b/hw/vfio/pci.c index a205c6b113..34f65ecd17 100644 --- a/hw/vfio/pci.c +++ b/hw/vfio/pci.c @@ -2828,10 +2828,10 @@ static void vfio_populate_device(VFIOPCIDevice *vde= v, Error **errp) =20 static void vfio_put_device(VFIOPCIDevice *vdev) { + vfio_detach_device(&vdev->vbasedev); + g_free(vdev->vbasedev.name); g_free(vdev->msix); - - vfio_put_base_device(&vdev->vbasedev); } =20 static void vfio_err_notifier_handler(void *opaque) @@ -2978,13 +2978,9 @@ static void vfio_realize(PCIDevice *pdev, Error **er= rp) { VFIOPCIDevice *vdev =3D VFIO_PCI(pdev); VFIODevice *vbasedev =3D &vdev->vbasedev; - VFIODevice *vbasedev_iter; - VFIOGroup *group; - char *tmp, *subsys, group_path[PATH_MAX], *group_name; + char *tmp, *subsys; Error *err =3D NULL; - ssize_t len; struct stat st; - int groupid; int i, ret; bool is_mdev; char uuid[UUID_FMT_LEN]; @@ -3015,38 +3011,7 @@ static void vfio_realize(PCIDevice *pdev, Error **er= rp) vbasedev->type =3D VFIO_DEVICE_TYPE_PCI; vbasedev->dev =3D DEVICE(vdev); =20 - tmp =3D g_strdup_printf("%s/iommu_group", vbasedev->sysfsdev); - len =3D readlink(tmp, group_path, sizeof(group_path)); - g_free(tmp); - - if (len <=3D 0 || len >=3D sizeof(group_path)) { - error_setg_errno(errp, len < 0 ? errno : ENAMETOOLONG, - "no iommu_group found"); - goto error; - } - - group_path[len] =3D 0; - - group_name =3D basename(group_path); - if (sscanf(group_name, "%d", &groupid) !=3D 1) { - error_setg_errno(errp, errno, "failed to read %s", group_path); - goto error; - } - - trace_vfio_realize(vbasedev->name, groupid); - - group =3D vfio_get_group(groupid, pci_device_iommu_address_space(pdev)= , errp); - if (!group) { - goto error; - } - - QLIST_FOREACH(vbasedev_iter, &group->device_list, next) { - if (strcmp(vbasedev_iter->name, vbasedev->name) =3D=3D 0) { - error_setg(errp, "device is already attached"); - vfio_put_group(group); - goto error; - } - } + trace_vfio_realize(vbasedev->name); =20 /* * Mediated devices *might* operate compatibly with discarding of RAM,= but @@ -3065,7 +3030,6 @@ static void vfio_realize(PCIDevice *pdev, Error **err= p) if (vbasedev->ram_block_discard_allowed && !is_mdev) { error_setg(errp, "x-balloon-allowed only potentially compatible " "with mdev devices"); - vfio_put_group(group); goto error; } =20 @@ -3076,10 +3040,10 @@ static void vfio_realize(PCIDevice *pdev, Error **e= rrp) name =3D g_strdup(vbasedev->name); } =20 - ret =3D vfio_get_device(group, name, vbasedev, errp); + ret =3D vfio_attach_device(name, vbasedev, + pci_device_iommu_address_space(pdev), errp); g_free(name); if (ret) { - vfio_put_group(group); goto error; } =20 @@ -3318,7 +3282,6 @@ error: static void vfio_instance_finalize(Object *obj) { VFIOPCIDevice *vdev =3D VFIO_PCI(obj); - VFIOGroup *group =3D vdev->vbasedev.group; =20 vfio_display_finalize(vdev); vfio_bars_finalize(vdev); @@ -3332,7 +3295,6 @@ static void vfio_instance_finalize(Object *obj) * g_free(vdev->igd_opregion); */ vfio_put_device(vdev); - vfio_put_group(group); } =20 static void vfio_exitfn(PCIDevice *pdev) diff --git a/hw/vfio/trace-events b/hw/vfio/trace-events index ee7509e68e..8016d9f0d2 100644 --- a/hw/vfio/trace-events +++ b/hw/vfio/trace-events @@ -37,7 +37,7 @@ vfio_pci_hot_reset_dep_devices(int domain, int bus, int s= lot, int function, int vfio_pci_hot_reset_result(const char *name, const char *result) "%s hot re= set: %s" vfio_populate_device_config(const char *name, unsigned long size, unsigned= long offset, unsigned long flags) "Device %s config:\n size: 0x%lx, offse= t: 0x%lx, flags: 0x%lx" vfio_populate_device_get_irq_info_failure(const char *errstr) "VFIO_DEVICE= _GET_IRQ_INFO failure: %s" -vfio_realize(const char *name, int group_id) " (%s) group %d" +vfio_realize(const char *name) " (%s)" vfio_mdev(const char *name, bool is_mdev) " (%s) is_mdev %d" vfio_add_ext_cap_dropped(const char *name, uint16_t cap, uint16_t offset) = "%s 0x%x@0x%x" vfio_pci_reset(const char *name) " (%s)" diff --git a/include/hw/vfio/vfio-common.h b/include/hw/vfio/vfio-common.h index bb7f9fe9c4..a29dfe7723 100644 --- a/include/hw/vfio/vfio-common.h +++ b/include/hw/vfio/vfio-common.h @@ -253,6 +253,9 @@ void vfio_put_group(VFIOGroup *group); struct vfio_device_info *vfio_get_device_info(int fd); int vfio_get_device(VFIOGroup *group, const char *name, VFIODevice *vbasedev, Error **errp); +int vfio_attach_device(char *name, VFIODevice *vbasedev, + AddressSpace *as, Error **errp); +void vfio_detach_device(VFIODevice *vbasedev); =20 extern int vfio_kvm_device_fd; =20 --=20 2.34.1 From nobody Wed May 15 22:12:51 2024 Delivered-To: importer@patchew.org Authentication-Results: mx.zohomail.com; dkim=pass header.i=@intel.com; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass(p=none dis=none) header.from=intel.com ARC-Seal: i=1; a=rsa-sha256; t=1693392834; cv=none; d=zohomail.com; s=zohoarc; b=EQ2Ez1NUz+3sONp4rY2GgjVXWzmDToxGzXjfs8DG0ketYqXg6HbvlZI2hWmFOny9tUfX9cqcCYjswuN/BX+WBx39zYYork2OGyWNWy3UCsl9M2zOPvug5FGaN7HXNjyo0d4YQ9VbtLuDPBIsOsD36X+htg5ME8MpbIj1uucc4t8= ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=zohomail.com; s=zohoarc; t=1693392834; h=Content-Transfer-Encoding:Cc:Date:From:In-Reply-To:List-Subscribe:List-Post:List-Id:List-Archive:List-Help:List-Unsubscribe:MIME-Version:Message-ID:References:Sender:Subject:To; bh=HpR5foyMRf/iD32dLjg0q2Tqx4eR2Zg8y6pE5mL7ra8=; b=ZY2pnWPbrrtDTWJzxFFnyt5d1XeDKHf/7TuFfcNcAqBvCJL3JRoKpOY3gMAnDsMVAcoJ+hFNKJBlOp0nzotWNMRBqKBXrbwwzRAWod/nCmLIV5PFaAwz/ai4UMWPpoRkwaUXytrytE27Nc/9Tm+IdBPlDAwXOo6JoeoWgKYn8/M= ARC-Authentication-Results: i=1; mx.zohomail.com; dkim=pass header.i=@intel.com; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass header.from= (p=none dis=none) Return-Path: Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) by mx.zohomail.com with SMTPS id 1693392834019436.816939810339; Wed, 30 Aug 2023 03:53:54 -0700 (PDT) Received: from localhost ([::1] helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1qbIpF-0004I3-V0; Wed, 30 Aug 2023 06:53:31 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1qbIpC-0004Ha-Il for qemu-devel@nongnu.org; Wed, 30 Aug 2023 06:53:26 -0400 Received: from mgamail.intel.com ([192.55.52.93]) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1qbIpA-0007fH-8c for qemu-devel@nongnu.org; Wed, 30 Aug 2023 06:53:26 -0400 Received: from orsmga006.jf.intel.com ([10.7.209.51]) by fmsmga102.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 30 Aug 2023 03:53:13 -0700 Received: from duan-server-s2600bt.bj.intel.com ([10.240.192.147]) by orsmga006-auth.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 30 Aug 2023 03:53:08 -0700 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1693392804; x=1724928804; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=SvlSSQEEskjTcJO2Ju6WBWVkXRHLlMua+97gnXK3DqQ=; b=Ym/rllxaVjDMfhti5w9LCb9uUg4NxUnVEa7fqxtrir3zBac88xjmUJ16 5+ORLFE5OuP4iAUFiON8zsHE39x+5HeQ5WRW07FiqFguB6foJ2D23yFMq uPO/tTdXbfk+04JhdDMTIUytNwv7Mk8ggkRfo4aZ+hjwsRvSiFnO2IJNI lP8QN3RowJV+deikGyyk0Fa4tGfE5frzKuWyzEYMhVy4sFNKQP6n8FHNP ibUvm5NYSpzmVJAa6k4sVoASRwtvILbbegu3/MT95gc8YTWThiIL1qF7K RRfEBCWhVwK2PhF9E0vz9oZ4SyvuGU6NAmpgVqqPJDCIddZ3swnDiXetD A==; X-IronPort-AV: E=McAfee;i="6600,9927,10817"; a="373016540" X-IronPort-AV: E=Sophos;i="6.02,213,1688454000"; d="scan'208";a="373016540" X-ExtLoop1: 1 X-IronPort-AV: E=McAfee;i="6600,9927,10817"; a="715866123" X-IronPort-AV: E=Sophos;i="6.02,213,1688454000"; d="scan'208";a="715866123" From: Zhenzhong Duan To: qemu-devel@nongnu.org Cc: alex.williamson@redhat.com, clg@redhat.com, jgg@nvidia.com, nicolinc@nvidia.com, joao.m.martins@oracle.com, eric.auger@redhat.com, peterx@redhat.com, jasowang@redhat.com, kevin.tian@intel.com, yi.l.liu@intel.com, yi.y.sun@intel.com, chao.p.peng@intel.com, Zhenzhong Duan Subject: [PATCH v1 10/22] vfio/platform: Use vfio_[attach/detach]_device Date: Wed, 30 Aug 2023 18:37:42 +0800 Message-Id: <20230830103754.36461-11-zhenzhong.duan@intel.com> X-Mailer: git-send-email 2.34.1 In-Reply-To: <20230830103754.36461-1-zhenzhong.duan@intel.com> References: <20230830103754.36461-1-zhenzhong.duan@intel.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Received-SPF: pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) client-ip=209.51.188.17; envelope-from=qemu-devel-bounces+importer=patchew.org@nongnu.org; helo=lists.gnu.org; Received-SPF: pass client-ip=192.55.52.93; envelope-from=zhenzhong.duan@intel.com; helo=mgamail.intel.com X-Spam_score_int: -43 X-Spam_score: -4.4 X-Spam_bar: ---- X-Spam_report: (-4.4 / 5.0 requ) BAYES_00=-1.9, DKIMWL_WL_HIGH=-0.001, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, RCVD_IN_DNSWL_MED=-2.3, SPF_HELO_NONE=0.001, SPF_PASS=-0.001 autolearn=ham autolearn_force=no X-Spam_action: no action X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: qemu-devel-bounces+importer=patchew.org@nongnu.org Sender: qemu-devel-bounces+importer=patchew.org@nongnu.org X-ZohoMail-DKIM: pass (identity @intel.com) X-ZM-MESSAGEID: 1693392904344100001 Content-Type: text/plain; charset="utf-8" From: Eric Auger Let the vfio-platform device use vfio_attach_device() and vfio_detach_device(), hence hiding the details of the used IOMMU backend. Signed-off-by: Eric Auger Signed-off-by: Yi Liu Signed-off-by: Zhenzhong Duan --- hw/vfio/platform.c | 43 ++++--------------------------------------- hw/vfio/trace-events | 2 +- 2 files changed, 5 insertions(+), 40 deletions(-) diff --git a/hw/vfio/platform.c b/hw/vfio/platform.c index 5af73f9287..5c08c39315 100644 --- a/hw/vfio/platform.c +++ b/hw/vfio/platform.c @@ -529,12 +529,7 @@ static VFIODeviceOps vfio_platform_ops =3D { */ static int vfio_base_device_init(VFIODevice *vbasedev, Error **errp) { - VFIOGroup *group; - VFIODevice *vbasedev_iter; - char *tmp, group_path[PATH_MAX], *group_name; - ssize_t len; struct stat st; - int groupid; int ret; =20 /* @sysfsdev takes precedence over @host */ @@ -557,47 +552,17 @@ static int vfio_base_device_init(VFIODevice *vbasedev= , Error **errp) return -errno; } =20 - tmp =3D g_strdup_printf("%s/iommu_group", vbasedev->sysfsdev); - len =3D readlink(tmp, group_path, sizeof(group_path)); - g_free(tmp); + trace_vfio_platform_base_device_init(vbasedev->name); =20 - if (len < 0 || len >=3D sizeof(group_path)) { - ret =3D len < 0 ? -errno : -ENAMETOOLONG; - error_setg_errno(errp, -ret, "no iommu_group found"); - return ret; - } - - group_path[len] =3D 0; - - group_name =3D basename(group_path); - if (sscanf(group_name, "%d", &groupid) !=3D 1) { - error_setg_errno(errp, errno, "failed to read %s", group_path); - return -errno; - } - - trace_vfio_platform_base_device_init(vbasedev->name, groupid); - - group =3D vfio_get_group(groupid, &address_space_memory, errp); - if (!group) { - return -ENOENT; - } - - QLIST_FOREACH(vbasedev_iter, &group->device_list, next) { - if (strcmp(vbasedev_iter->name, vbasedev->name) =3D=3D 0) { - error_setg(errp, "device is already attached"); - vfio_put_group(group); - return -EBUSY; - } - } - ret =3D vfio_get_device(group, vbasedev->name, vbasedev, errp); + ret =3D vfio_attach_device(vbasedev->name, vbasedev, + &address_space_memory, errp); if (ret) { - vfio_put_group(group); return ret; } =20 ret =3D vfio_populate_device(vbasedev, errp); if (ret) { - vfio_put_group(group); + vfio_detach_device(vbasedev); } =20 return ret; diff --git a/hw/vfio/trace-events b/hw/vfio/trace-events index 8016d9f0d2..bd32970854 100644 --- a/hw/vfio/trace-events +++ b/hw/vfio/trace-events @@ -124,7 +124,7 @@ vfio_get_dirty_bitmap(int fd, uint64_t iova, uint64_t s= ize, uint64_t bitmap_size vfio_iommu_map_dirty_notify(uint64_t iova_start, uint64_t iova_end) "iommu= dirty @ 0x%"PRIx64" - 0x%"PRIx64 =20 # platform.c -vfio_platform_base_device_init(char *name, int groupid) "%s belongs to gro= up #%d" +vfio_platform_base_device_init(char *name) "%s" vfio_platform_realize(char *name, char *compat) "vfio device %s, compat = =3D %s" vfio_platform_eoi(int pin, int fd) "EOI IRQ pin %d (fd=3D%d)" vfio_platform_intp_mmap_enable(int pin) "IRQ #%d still active, stay in slo= w path" --=20 2.34.1 From nobody Wed May 15 22:12:51 2024 Delivered-To: importer@patchew.org Authentication-Results: mx.zohomail.com; dkim=pass header.i=@intel.com; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass(p=none dis=none) header.from=intel.com ARC-Seal: i=1; a=rsa-sha256; t=1693392877; cv=none; d=zohomail.com; s=zohoarc; b=nANapldVkv8nOT+3RbagYHFL/86xVrqBkPn+aAMbv3o5LJII4gkgfIE9cuHJUAIumhaNrvepVp//RkqJPv/iBGtM2kJ/77YH10N6Bp6jK8CmBYhAanWTbzvwMeh7JyI4214wBZmVTEIzi0C1FL1hgeVWbGagPt9vtqkK1Bxt+Ec= ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=zohomail.com; s=zohoarc; t=1693392877; h=Content-Transfer-Encoding:Cc:Date:From:In-Reply-To:List-Subscribe:List-Post:List-Id:List-Archive:List-Help:List-Unsubscribe:MIME-Version:Message-ID:References:Sender:Subject:To; bh=lyCj/h3cc/1i/jn2ZwwJQiCYBmEXvRhhTLlZ6xoEEi0=; b=kdHPHYR+m3B46oSGsXlIz+zDz265v5iii4COltWGwSSNr0wVQKdo4YcAfwrTq37fU6nsA0tLLb/ZZVT1wtdDb2/Crj7/h0Iwhti0HVpVklUk/SyrMEon0wwsJR5my5yo+5hLMU/WJEU7irdfMl9Dbdgh3O8NUa9F1VF5k/43eK8= ARC-Authentication-Results: i=1; mx.zohomail.com; dkim=pass header.i=@intel.com; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass header.from= (p=none dis=none) Return-Path: Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) by mx.zohomail.com with SMTPS id 1693392877209200.12178296812863; Wed, 30 Aug 2023 03:54:37 -0700 (PDT) Received: from localhost ([::1] helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1qbIpM-0004O8-4N; Wed, 30 Aug 2023 06:53:36 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1qbIpJ-0004Mw-ND; Wed, 30 Aug 2023 06:53:33 -0400 Received: from mgamail.intel.com ([192.55.52.93]) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1qbIpH-0007fH-4Y; Wed, 30 Aug 2023 06:53:33 -0400 Received: from orsmga006.jf.intel.com ([10.7.209.51]) by fmsmga102.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 30 Aug 2023 03:53:19 -0700 Received: from duan-server-s2600bt.bj.intel.com ([10.240.192.147]) by orsmga006-auth.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 30 Aug 2023 03:53:12 -0700 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1693392811; x=1724928811; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=sMYo/7cpo6pQSEDRVcxQXWkRt9bc8yQALxYISmMeMZY=; b=fH0rcSL0vXO1dZmkjH3fonWR/tALJbR7t6SK7vARdp3GHzs+5cyUd4Pu DGoeCT2JGqICO/78bj8v0sb+CBTHAkrelKLMpKc3AEazNh4b0zjyMgJx1 w7Z6Wo3LVpZg7i4KgBgmzIGkSPXf4tjVqPbGXj9btdTk+vGX2KxXMNShc QasUkIQbzAU1DQboZi/DBDp5a4z2GHYUxQEm7SXb+6ftg27lt+zTm0TF2 0PGKC6imc+In/q/2Fau+PPjptZ6yWJXpYHsK5rTeMcZJOeb1jCbMTV1W1 0agQuIsZ+yJTUSReYClYiuEYQONGipJZTxoqD7KVhi/XSu7+tXO05CQjU A==; X-IronPort-AV: E=McAfee;i="6600,9927,10817"; a="373016597" X-IronPort-AV: E=Sophos;i="6.02,213,1688454000"; d="scan'208";a="373016597" X-ExtLoop1: 1 X-IronPort-AV: E=McAfee;i="6600,9927,10817"; a="715866153" X-IronPort-AV: E=Sophos;i="6.02,213,1688454000"; d="scan'208";a="715866153" From: Zhenzhong Duan To: qemu-devel@nongnu.org Cc: alex.williamson@redhat.com, clg@redhat.com, jgg@nvidia.com, nicolinc@nvidia.com, joao.m.martins@oracle.com, eric.auger@redhat.com, peterx@redhat.com, jasowang@redhat.com, kevin.tian@intel.com, yi.l.liu@intel.com, yi.y.sun@intel.com, chao.p.peng@intel.com, Zhenzhong Duan , Tony Krowiak , Halil Pasic , Jason Herne , Thomas Huth , qemu-s390x@nongnu.org (open list:vfio-ap) Subject: [PATCH v1 11/22] vfio/ap: Use vfio_[attach/detach]_device Date: Wed, 30 Aug 2023 18:37:43 +0800 Message-Id: <20230830103754.36461-12-zhenzhong.duan@intel.com> X-Mailer: git-send-email 2.34.1 In-Reply-To: <20230830103754.36461-1-zhenzhong.duan@intel.com> References: <20230830103754.36461-1-zhenzhong.duan@intel.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Received-SPF: pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) client-ip=209.51.188.17; envelope-from=qemu-devel-bounces+importer=patchew.org@nongnu.org; helo=lists.gnu.org; Received-SPF: pass client-ip=192.55.52.93; envelope-from=zhenzhong.duan@intel.com; helo=mgamail.intel.com X-Spam_score_int: -43 X-Spam_score: -4.4 X-Spam_bar: ---- X-Spam_report: (-4.4 / 5.0 requ) BAYES_00=-1.9, DKIMWL_WL_HIGH=-0.001, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, RCVD_IN_DNSWL_MED=-2.3, SPF_HELO_NONE=0.001, SPF_PASS=-0.001 autolearn=ham autolearn_force=no X-Spam_action: no action X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: qemu-devel-bounces+importer=patchew.org@nongnu.org Sender: qemu-devel-bounces+importer=patchew.org@nongnu.org X-ZohoMail-DKIM: pass (identity @intel.com) X-ZM-MESSAGEID: 1693392877975100001 Content-Type: text/plain; charset="utf-8" From: Eric Auger Let the vfio-ap device use vfio_attach_device() and vfio_detach_device(), hence hiding the details of the used IOMMU backend. Signed-off-by: Eric Auger Signed-off-by: Yi Liu Signed-off-by: Zhenzhong Duan --- hw/vfio/ap.c | 68 +++++++++------------------------------------------- 1 file changed, 11 insertions(+), 57 deletions(-) diff --git a/hw/vfio/ap.c b/hw/vfio/ap.c index 6e21d1da5a..16ea7fb3c2 100644 --- a/hw/vfio/ap.c +++ b/hw/vfio/ap.c @@ -53,40 +53,6 @@ struct VFIODeviceOps vfio_ap_ops =3D { .vfio_compute_needs_reset =3D vfio_ap_compute_needs_reset, }; =20 -static void vfio_ap_put_device(VFIOAPDevice *vapdev) -{ - g_free(vapdev->vdev.name); - vfio_put_base_device(&vapdev->vdev); -} - -static VFIOGroup *vfio_ap_get_group(VFIOAPDevice *vapdev, Error **errp) -{ - GError *gerror =3D NULL; - char *symlink, *group_path; - int groupid; - - symlink =3D g_strdup_printf("%s/iommu_group", vapdev->vdev.sysfsdev); - group_path =3D g_file_read_link(symlink, &gerror); - g_free(symlink); - - if (!group_path) { - error_setg(errp, "%s: no iommu_group found for %s: %s", - TYPE_VFIO_AP_DEVICE, vapdev->vdev.sysfsdev, gerror->mes= sage); - g_error_free(gerror); - return NULL; - } - - if (sscanf(basename(group_path), "%d", &groupid) !=3D 1) { - error_setg(errp, "vfio: failed to read %s", group_path); - g_free(group_path); - return NULL; - } - - g_free(group_path); - - return vfio_get_group(groupid, &address_space_memory, errp); -} - static void vfio_ap_req_notifier_handler(void *opaque) { VFIOAPDevice *vapdev =3D opaque; @@ -189,22 +155,15 @@ static void vfio_ap_unregister_irq_notifier(VFIOAPDev= ice *vapdev, static void vfio_ap_realize(DeviceState *dev, Error **errp) { int ret; - char *mdevid; Error *err =3D NULL; - VFIOGroup *vfio_group; APDevice *apdev =3D AP_DEVICE(dev); VFIOAPDevice *vapdev =3D VFIO_AP_DEVICE(apdev); + VFIODevice *vbasedev =3D &vapdev->vdev; =20 - vfio_group =3D vfio_ap_get_group(vapdev, errp); - if (!vfio_group) { - return; - } - - vapdev->vdev.ops =3D &vfio_ap_ops; - vapdev->vdev.type =3D VFIO_DEVICE_TYPE_AP; - mdevid =3D basename(vapdev->vdev.sysfsdev); - vapdev->vdev.name =3D g_strdup_printf("%s", mdevid); - vapdev->vdev.dev =3D dev; + vbasedev->name =3D g_path_get_basename(vbasedev->sysfsdev); + vbasedev->ops =3D &vfio_ap_ops; + vbasedev->type =3D VFIO_DEVICE_TYPE_AP; + vbasedev->dev =3D dev; =20 /* * vfio-ap devices operate in a way compatible with discarding of @@ -214,9 +173,11 @@ static void vfio_ap_realize(DeviceState *dev, Error **= errp) */ vapdev->vdev.ram_block_discard_allowed =3D true; =20 - ret =3D vfio_get_device(vfio_group, mdevid, &vapdev->vdev, errp); + ret =3D vfio_attach_device(vbasedev->name, vbasedev, + &address_space_memory, errp); if (ret) { - goto out_get_dev_err; + g_free(vbasedev->name); + return; } =20 vfio_ap_register_irq_notifier(vapdev, VFIO_AP_REQ_IRQ_INDEX, &err); @@ -227,23 +188,16 @@ static void vfio_ap_realize(DeviceState *dev, Error *= *errp) */ error_report_err(err); } - - return; - -out_get_dev_err: - vfio_ap_put_device(vapdev); - vfio_put_group(vfio_group); } =20 static void vfio_ap_unrealize(DeviceState *dev) { APDevice *apdev =3D AP_DEVICE(dev); VFIOAPDevice *vapdev =3D VFIO_AP_DEVICE(apdev); - VFIOGroup *group =3D vapdev->vdev.group; =20 vfio_ap_unregister_irq_notifier(vapdev, VFIO_AP_REQ_IRQ_INDEX); - vfio_ap_put_device(vapdev); - vfio_put_group(group); + vfio_detach_device(&vapdev->vdev); + g_free(vapdev->vdev.name); } =20 static Property vfio_ap_properties[] =3D { --=20 2.34.1 From nobody Wed May 15 22:12:51 2024 Delivered-To: importer@patchew.org Authentication-Results: mx.zohomail.com; dkim=pass header.i=@intel.com; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass(p=none dis=none) header.from=intel.com ARC-Seal: i=1; a=rsa-sha256; t=1693392885; cv=none; d=zohomail.com; s=zohoarc; b=QPLYXhkK5zXxCktRKD9e3W7IeKBWza3RCOGlgp4lj2ey7cv6xZ68tKT+M3xwpBGtlSAIs4LvVmhxRTlNtZgHqBSO46E5NpkRW1FqdzJi2L6eityYTCLNMVQH5g+px1kZzSzgnqQL/YGjisFAa7yh3xfCfv6oWDc0nyuuxglUuWk= ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=zohomail.com; s=zohoarc; t=1693392885; h=Content-Transfer-Encoding:Cc:Date:From:In-Reply-To:List-Subscribe:List-Post:List-Id:List-Archive:List-Help:List-Unsubscribe:MIME-Version:Message-ID:References:Sender:Subject:To; bh=ZlUON6aoGfVcz5J4AwlOPt8FzTgjSDiBb+0FVPD9eek=; b=IDVjQJfA9TLThe+OeyUTbLW65Ou8V0XK1cHfhUnh4qvAnpoFFpDEFqd2pckhyWCSos2S3icoJmDMaxe4g6OzW9+XHiiEwJElqYRydhH2GhY6+sxUldTU1w4cCcCvolNYgZ079kQo77qv6bjlgHWV1NVMABDOcpbhbImTuQM1mFs= ARC-Authentication-Results: i=1; mx.zohomail.com; dkim=pass header.i=@intel.com; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass header.from= (p=none dis=none) Return-Path: Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) by mx.zohomail.com with SMTPS id 16933928857041023.8078144277447; Wed, 30 Aug 2023 03:54:45 -0700 (PDT) Received: from localhost ([::1] helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1qbIpc-00057j-0y; Wed, 30 Aug 2023 06:53:52 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1qbIpa-0004zx-T3; Wed, 30 Aug 2023 06:53:50 -0400 Received: from mgamail.intel.com ([192.55.52.93]) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1qbIpX-0007jy-O3; Wed, 30 Aug 2023 06:53:50 -0400 Received: from orsmga006.jf.intel.com ([10.7.209.51]) by fmsmga102.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 30 Aug 2023 03:53:40 -0700 Received: from duan-server-s2600bt.bj.intel.com ([10.240.192.147]) by orsmga006-auth.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 30 Aug 2023 03:53:19 -0700 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1693392827; x=1724928827; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=q55X3uiTie+uGYX/svT6+aBQe+30ypTkskXJ6fEbSRg=; b=GDf+dh0t08jL8GUokP3hjp0IOd7GO4j6HZ5Z+1Db+JuCOja/nhghExto CKEVB89Ee0Ig4Vb9v5iYiY/uE2hRSUdXfQf74/g/4nbROQKMWiAYWUFtf L9zVKGY6FRDzsHUW9GD6BdOoxfkYXjRBLszFQBOQGs+cFObaA4HIVUKIx Q+mfM3E7f8AXeKp6nGMaR9X0BDO2U1TYoGJCA1pBGUNlHI0HCqQX8j2Wp Pv5YWJd807+BZBoLABJ8QosyEiL/ZYk+4rSd6mnpXNgiBjitzokXS9JkV 76/68cndzrVTiVwkmY22boUm4ASAfaTdqk8RsUA/DJCrUPiP1CL3bkpb3 Q==; X-IronPort-AV: E=McAfee;i="6600,9927,10817"; a="373016646" X-IronPort-AV: E=Sophos;i="6.02,213,1688454000"; d="scan'208";a="373016646" X-ExtLoop1: 1 X-IronPort-AV: E=McAfee;i="6600,9927,10817"; a="715866171" X-IronPort-AV: E=Sophos;i="6.02,213,1688454000"; d="scan'208";a="715866171" From: Zhenzhong Duan To: qemu-devel@nongnu.org Cc: alex.williamson@redhat.com, clg@redhat.com, jgg@nvidia.com, nicolinc@nvidia.com, joao.m.martins@oracle.com, eric.auger@redhat.com, peterx@redhat.com, jasowang@redhat.com, kevin.tian@intel.com, yi.l.liu@intel.com, yi.y.sun@intel.com, chao.p.peng@intel.com, Zhenzhong Duan , Eric Farman , Matthew Rosato , Thomas Huth , qemu-s390x@nongnu.org (open list:vfio-ccw) Subject: [PATCH v1 12/22] vfio/ccw: Use vfio_[attach/detach]_device Date: Wed, 30 Aug 2023 18:37:44 +0800 Message-Id: <20230830103754.36461-13-zhenzhong.duan@intel.com> X-Mailer: git-send-email 2.34.1 In-Reply-To: <20230830103754.36461-1-zhenzhong.duan@intel.com> References: <20230830103754.36461-1-zhenzhong.duan@intel.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Received-SPF: pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) client-ip=209.51.188.17; envelope-from=qemu-devel-bounces+importer=patchew.org@nongnu.org; helo=lists.gnu.org; Received-SPF: pass client-ip=192.55.52.93; envelope-from=zhenzhong.duan@intel.com; helo=mgamail.intel.com X-Spam_score_int: -43 X-Spam_score: -4.4 X-Spam_bar: ---- X-Spam_report: (-4.4 / 5.0 requ) BAYES_00=-1.9, DKIMWL_WL_HIGH=-0.001, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, RCVD_IN_DNSWL_MED=-2.3, SPF_HELO_NONE=0.001, SPF_PASS=-0.001 autolearn=ham autolearn_force=no X-Spam_action: no action X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: qemu-devel-bounces+importer=patchew.org@nongnu.org Sender: qemu-devel-bounces+importer=patchew.org@nongnu.org X-ZohoMail-DKIM: pass (identity @intel.com) X-ZM-MESSAGEID: 1693392886033100001 Content-Type: text/plain; charset="utf-8" From: Eric Auger Let the vfio-ccw device use vfio_attach_device() and vfio_detach_device(), hence hiding the details of the used IOMMU backend. Also now all the devices have been migrated to use the new vfio_attach_device/vfio_detach_device API, let's turn the legacy functions into static functions, local to container.c. Signed-off-by: Eric Auger Signed-off-by: Yi Liu Signed-off-by: Zhenzhong Duan --- hw/vfio/ccw.c | 120 ++++++++-------------------------- hw/vfio/container.c | 10 +-- include/hw/vfio/vfio-common.h | 5 -- 3 files changed, 33 insertions(+), 102 deletions(-) diff --git a/hw/vfio/ccw.c b/hw/vfio/ccw.c index 1e2fce83b0..f078e014fa 100644 --- a/hw/vfio/ccw.c +++ b/hw/vfio/ccw.c @@ -572,88 +572,15 @@ static void vfio_ccw_put_region(VFIOCCWDevice *vcdev) g_free(vcdev->io_region); } =20 -static void vfio_ccw_put_device(VFIOCCWDevice *vcdev) -{ - g_free(vcdev->vdev.name); - vfio_put_base_device(&vcdev->vdev); -} - -static void vfio_ccw_get_device(VFIOGroup *group, VFIOCCWDevice *vcdev, - Error **errp) -{ - S390CCWDevice *cdev =3D S390_CCW_DEVICE(vcdev); - char *name =3D g_strdup_printf("%x.%x.%04x", cdev->hostid.cssid, - cdev->hostid.ssid, - cdev->hostid.devid); - VFIODevice *vbasedev; - - QLIST_FOREACH(vbasedev, &group->device_list, next) { - if (strcmp(vbasedev->name, name) =3D=3D 0) { - error_setg(errp, "vfio: subchannel %s has already been attache= d", - name); - goto out_err; - } - } - - /* - * All vfio-ccw devices are believed to operate in a way compatible wi= th - * discarding of memory in RAM blocks, ie. pages pinned in the host are - * in the current working set of the guest driver and therefore never - * overlap e.g., with pages available to the guest balloon driver. Th= is - * needs to be set before vfio_get_device() for vfio common to handle - * ram_block_discard_disable(). - */ - vcdev->vdev.ram_block_discard_allowed =3D true; - - if (vfio_get_device(group, cdev->mdevid, &vcdev->vdev, errp)) { - goto out_err; - } - - vcdev->vdev.ops =3D &vfio_ccw_ops; - vcdev->vdev.type =3D VFIO_DEVICE_TYPE_CCW; - vcdev->vdev.name =3D name; - vcdev->vdev.dev =3D DEVICE(vcdev); - - return; - -out_err: - g_free(name); -} - -static VFIOGroup *vfio_ccw_get_group(S390CCWDevice *cdev, Error **errp) -{ - char *tmp, group_path[PATH_MAX]; - ssize_t len; - int groupid; - - tmp =3D g_strdup_printf("/sys/bus/css/devices/%x.%x.%04x/%s/iommu_grou= p", - cdev->hostid.cssid, cdev->hostid.ssid, - cdev->hostid.devid, cdev->mdevid); - len =3D readlink(tmp, group_path, sizeof(group_path)); - g_free(tmp); - - if (len <=3D 0 || len >=3D sizeof(group_path)) { - error_setg(errp, "vfio: no iommu_group found"); - return NULL; - } - - group_path[len] =3D 0; - - if (sscanf(basename(group_path), "%d", &groupid) !=3D 1) { - error_setg(errp, "vfio: failed to read %s", group_path); - return NULL; - } - - return vfio_get_group(groupid, &address_space_memory, errp); -} - static void vfio_ccw_realize(DeviceState *dev, Error **errp) { - VFIOGroup *group; - S390CCWDevice *cdev =3D S390_CCW_DEVICE(dev); - VFIOCCWDevice *vcdev =3D VFIO_CCW(cdev); + CcwDevice *ccw_dev =3D DO_UPCAST(CcwDevice, parent_obj, dev); + S390CCWDevice *cdev =3D DO_UPCAST(S390CCWDevice, parent_obj, ccw_dev); + VFIOCCWDevice *vcdev =3D DO_UPCAST(VFIOCCWDevice, cdev, cdev); S390CCWDeviceClass *cdc =3D S390_CCW_DEVICE_GET_CLASS(cdev); + VFIODevice *vbasedev =3D &vcdev->vdev; Error *err =3D NULL; + int ret; =20 /* Call the class init function for subchannel. */ if (cdc->realize) { @@ -663,14 +590,25 @@ static void vfio_ccw_realize(DeviceState *dev, Error = **errp) } } =20 - group =3D vfio_ccw_get_group(cdev, &err); - if (!group) { - goto out_group_err; - } + vbasedev->ops =3D &vfio_ccw_ops; + vbasedev->type =3D VFIO_DEVICE_TYPE_CCW; + vbasedev->name =3D g_strdup(cdev->mdevid); + vbasedev->dev =3D &vcdev->cdev.parent_obj.parent_obj; =20 - vfio_ccw_get_device(group, vcdev, &err); - if (err) { - goto out_device_err; + /* + * All vfio-ccw devices are believed to operate in a way compatible wi= th + * discarding of memory in RAM blocks, ie. pages pinned in the host are + * in the current working set of the guest driver and therefore never + * overlap e.g., with pages available to the guest balloon driver. Th= is + * needs to be set before vfio_get_device() for vfio common to handle + * ram_block_discard_disable(). + */ + vbasedev->ram_block_discard_allowed =3D true; + + ret =3D vfio_attach_device(vbasedev->name, vbasedev, + &address_space_memory, errp); + if (ret) { + goto out_attach_dev_err; } =20 vfio_ccw_get_region(vcdev, &err); @@ -708,10 +646,9 @@ out_irq_notifier_err: out_io_notifier_err: vfio_ccw_put_region(vcdev); out_region_err: - vfio_ccw_put_device(vcdev); -out_device_err: - vfio_put_group(group); -out_group_err: + vfio_detach_device(vbasedev); +out_attach_dev_err: + g_free(vbasedev->name); if (cdc->unrealize) { cdc->unrealize(cdev); } @@ -724,14 +661,13 @@ static void vfio_ccw_unrealize(DeviceState *dev) S390CCWDevice *cdev =3D S390_CCW_DEVICE(dev); VFIOCCWDevice *vcdev =3D VFIO_CCW(cdev); S390CCWDeviceClass *cdc =3D S390_CCW_DEVICE_GET_CLASS(cdev); - VFIOGroup *group =3D vcdev->vdev.group; =20 vfio_ccw_unregister_irq_notifier(vcdev, VFIO_CCW_REQ_IRQ_INDEX); vfio_ccw_unregister_irq_notifier(vcdev, VFIO_CCW_CRW_IRQ_INDEX); vfio_ccw_unregister_irq_notifier(vcdev, VFIO_CCW_IO_IRQ_INDEX); vfio_ccw_put_region(vcdev); - vfio_ccw_put_device(vcdev); - vfio_put_group(group); + vfio_detach_device(&vcdev->vdev); + g_free(vcdev->vdev.name); =20 if (cdc->unrealize) { cdc->unrealize(cdev); diff --git a/hw/vfio/container.c b/hw/vfio/container.c index 74556da0c7..c71fddc09a 100644 --- a/hw/vfio/container.c +++ b/hw/vfio/container.c @@ -837,7 +837,7 @@ static void vfio_disconnect_container(VFIOGroup *group) } } =20 -VFIOGroup *vfio_get_group(int groupid, AddressSpace *as, Error **errp) +static VFIOGroup *vfio_get_group(int groupid, AddressSpace *as, Error **er= rp) { VFIOGroup *group; char path[32]; @@ -900,7 +900,7 @@ free_group_exit: return NULL; } =20 -void vfio_put_group(VFIOGroup *group) +static void vfio_put_group(VFIOGroup *group) { if (!group || !QLIST_EMPTY(&group->device_list)) { return; @@ -917,8 +917,8 @@ void vfio_put_group(VFIOGroup *group) g_free(group); } =20 -int vfio_get_device(VFIOGroup *group, const char *name, - VFIODevice *vbasedev, Error **errp) +static int vfio_get_device(VFIOGroup *group, const char *name, + VFIODevice *vbasedev, Error **errp) { g_autofree struct vfio_device_info *info =3D NULL; int fd; @@ -976,7 +976,7 @@ int vfio_get_device(VFIOGroup *group, const char *name, return 0; } =20 -void vfio_put_base_device(VFIODevice *vbasedev) +static void vfio_put_base_device(VFIODevice *vbasedev) { if (!vbasedev->group) { return; diff --git a/include/hw/vfio/vfio-common.h b/include/hw/vfio/vfio-common.h index a29dfe7723..95bcafdaf6 100644 --- a/include/hw/vfio/vfio-common.h +++ b/include/hw/vfio/vfio-common.h @@ -230,7 +230,6 @@ int vfio_container_add_section_window(VFIOContainer *co= ntainer, void vfio_container_del_section_window(VFIOContainer *container, MemoryRegionSection *section); =20 -void vfio_put_base_device(VFIODevice *vbasedev); void vfio_disable_irqindex(VFIODevice *vbasedev, int index); void vfio_unmask_single_irqindex(VFIODevice *vbasedev, int index); void vfio_mask_single_irqindex(VFIODevice *vbasedev, int index); @@ -248,11 +247,7 @@ void vfio_region_unmap(VFIORegion *region); void vfio_region_exit(VFIORegion *region); void vfio_region_finalize(VFIORegion *region); void vfio_reset_handler(void *opaque); -VFIOGroup *vfio_get_group(int groupid, AddressSpace *as, Error **errp); -void vfio_put_group(VFIOGroup *group); struct vfio_device_info *vfio_get_device_info(int fd); -int vfio_get_device(VFIOGroup *group, const char *name, - VFIODevice *vbasedev, Error **errp); int vfio_attach_device(char *name, VFIODevice *vbasedev, AddressSpace *as, Error **errp); void vfio_detach_device(VFIODevice *vbasedev); --=20 2.34.1 From nobody Wed May 15 22:12:51 2024 Delivered-To: importer@patchew.org Authentication-Results: mx.zohomail.com; dkim=pass header.i=@intel.com; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass(p=none dis=none) header.from=intel.com ARC-Seal: i=1; a=rsa-sha256; t=1693392958; cv=none; d=zohomail.com; s=zohoarc; b=K5ttKjzPp4SUnql5FLwwkJEpWc6BF36JfOP0SRiQgQ2F68vGctBbA4hxPk1Ch0n+oFobW8iWGy4zMyt7DFTUkAEjfNNIf4RPlbrzsfvAPWkRDtNO799s2JyhT7NQ1GEJmyxXASpDAx0XDL0QpDUJ5ZFsn01C3hSELhpQfO1KvAI= ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=zohomail.com; s=zohoarc; t=1693392958; h=Content-Transfer-Encoding:Cc:Date:From:In-Reply-To:List-Subscribe:List-Post:List-Id:List-Archive:List-Help:List-Unsubscribe:MIME-Version:Message-ID:References:Sender:Subject:To; bh=r46pEQ5S1ymn90UpavbU3GMnzb11d8oPN8ASfEnpLTU=; b=a+Twv1C/HgIywznr1LZLTXg7fBB7X6CLYvb9dD3g2WrmVmG7wrB1fSVvDw8+a1q/lt7El1Yr+ySJpVrM23cgyc0rOoh6TesV9fr0CPnk27OcnCVDeH2T3pqbmOYJAxH9GZoQKHgzgzjStbKT6ohKnQUZIMzgAT/gXxhjokpMsj8= ARC-Authentication-Results: i=1; mx.zohomail.com; dkim=pass header.i=@intel.com; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass header.from= (p=none dis=none) Return-Path: Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) by mx.zohomail.com with SMTPS id 1693392958151462.5942287739364; Wed, 30 Aug 2023 03:55:58 -0700 (PDT) Received: from localhost ([::1] helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1qbIpz-0006Lp-4T; Wed, 30 Aug 2023 06:54:15 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1qbIpx-0006HY-Bg; Wed, 30 Aug 2023 06:54:13 -0400 Received: from mgamail.intel.com ([192.55.52.93]) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1qbIpr-0007je-Oh; Wed, 30 Aug 2023 06:54:13 -0400 Received: from orsmga006.jf.intel.com ([10.7.209.51]) by fmsmga102.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 30 Aug 2023 03:53:40 -0700 Received: from duan-server-s2600bt.bj.intel.com ([10.240.192.147]) by orsmga006-auth.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 30 Aug 2023 03:53:24 -0700 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1693392847; x=1724928847; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=nQt7vHwzkHRSCJwtFsCf1mi0ZxDzc17Y/rQGdmjbGJ8=; b=QmZbcMsMzzJIh3xcBQtAbdbiGI+zesRMGVjhqhMizHYIoR4hD9QxGiuD hTXm8RNNrvwJt8RVIfsRrSqIfPIbG/ioHQds7hwv6YYhPU+wBr52Q5cpv i5+HHYqhKzpSakZz4yo7sJb3XXQ2BLF7dw8p2F+9pP4pgkI1hXLBoxnKK 26b8BrSzOIMtNveJU0ZSk5lnEMlkBC4SiHu18QuFtRVSFT5pvQde+5Xhx RMLU65//f/ow0DqPvG4UVvrwupkClGvqJ+p6GZbO9khJC0ptoEAvTylLc b/QL3nq8FdP4P9eFUC0wmfhs0K8u9lGV95fPJhwqahIit5F88O4lIXACE w==; X-IronPort-AV: E=McAfee;i="6600,9927,10817"; a="373016654" X-IronPort-AV: E=Sophos;i="6.02,213,1688454000"; d="scan'208";a="373016654" X-ExtLoop1: 1 X-IronPort-AV: E=McAfee;i="6600,9927,10817"; a="715866210" X-IronPort-AV: E=Sophos;i="6.02,213,1688454000"; d="scan'208";a="715866210" From: Zhenzhong Duan To: qemu-devel@nongnu.org Cc: alex.williamson@redhat.com, clg@redhat.com, jgg@nvidia.com, nicolinc@nvidia.com, joao.m.martins@oracle.com, eric.auger@redhat.com, peterx@redhat.com, jasowang@redhat.com, kevin.tian@intel.com, yi.l.liu@intel.com, yi.y.sun@intel.com, chao.p.peng@intel.com, Yi Sun , Zhenzhong Duan , Daniel Henrique Barboza , David Gibson , Greg Kurz , Harsh Prateek Bora , qemu-ppc@nongnu.org (open list:sPAPR (pseries)) Subject: [PATCH v1 13/22] vfio: Add base container Date: Wed, 30 Aug 2023 18:37:45 +0800 Message-Id: <20230830103754.36461-14-zhenzhong.duan@intel.com> X-Mailer: git-send-email 2.34.1 In-Reply-To: <20230830103754.36461-1-zhenzhong.duan@intel.com> References: <20230830103754.36461-1-zhenzhong.duan@intel.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Received-SPF: pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) client-ip=209.51.188.17; envelope-from=qemu-devel-bounces+importer=patchew.org@nongnu.org; helo=lists.gnu.org; Received-SPF: pass client-ip=192.55.52.93; envelope-from=zhenzhong.duan@intel.com; helo=mgamail.intel.com X-Spam_score_int: -43 X-Spam_score: -4.4 X-Spam_bar: ---- X-Spam_report: (-4.4 / 5.0 requ) BAYES_00=-1.9, DKIMWL_WL_HIGH=-0.001, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, RCVD_IN_DNSWL_MED=-2.3, SPF_HELO_NONE=0.001, SPF_PASS=-0.001 autolearn=ham autolearn_force=no X-Spam_action: no action X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: qemu-devel-bounces+importer=patchew.org@nongnu.org Sender: qemu-devel-bounces+importer=patchew.org@nongnu.org X-ZohoMail-DKIM: pass (identity @intel.com) X-ZM-MESSAGEID: 1693392960410100003 Content-Type: text/plain; charset="utf-8" From: Yi Liu Abstract the VFIOContainer to be a base object. It is supposed to be embedded by legacy VFIO container and later on, into the new iommufd based container. The base container implements generic code such as code related to memory_listener and address space management. The VFIOContainerOps implements callbacks that depend on the kernel user space being used. 'common.c' and vfio device code only manipulates the base container with wrapper functions that calls the functions defined in VFIOContainerOpsClass. Existing 'container.c' code is converted to implement the legacy container ops functions. Below is the base container. It's named as VFIOContainer, old VFIOContainer is replaced with VFIOLegacyContainer. struct VFIOContainer { VFIOIOMMUBackendOpsClass *ops; VFIOAddressSpace *space; MemoryListener listener; Error *error; bool initialized; bool dirty_pages_supported; uint64_t dirty_pgsizes; uint64_t max_dirty_bitmap_size; unsigned long pgsizes; unsigned int dma_max_mappings; QLIST_HEAD(, VFIOGuestIOMMU) giommu_list; QLIST_HEAD(, VFIOHostDMAWindow) hostwin_list; QLIST_HEAD(, VFIORamDiscardListener) vrdl_list; QLIST_ENTRY(VFIOContainer) next; }; struct VFIOLegacyContainer { VFIOContainer bcontainer; int fd; /* /dev/vfio/vfio, empowered by the attached groups */ MemoryListener prereg_listener; unsigned iommu_type; QLIST_HEAD(, VFIOGroup) group_list; }; Co-authored-by: Eric Auger Signed-off-by: Eric Auger Signed-off-by: Yi Liu Signed-off-by: Yi Sun Signed-off-by: Zhenzhong Duan --- hw/vfio/common.c | 72 +++++--- hw/vfio/container-base.c | 160 +++++++++++++++++ hw/vfio/container.c | 247 ++++++++++++++++---------- hw/vfio/meson.build | 1 + hw/vfio/spapr.c | 22 +-- hw/vfio/trace-events | 4 +- include/hw/vfio/vfio-common.h | 85 ++------- include/hw/vfio/vfio-container-base.h | 155 ++++++++++++++++ 8 files changed, 540 insertions(+), 206 deletions(-) create mode 100644 hw/vfio/container-base.c create mode 100644 include/hw/vfio/vfio-container-base.h diff --git a/hw/vfio/common.c b/hw/vfio/common.c index 044710fc1f..86b6af5740 100644 --- a/hw/vfio/common.c +++ b/hw/vfio/common.c @@ -379,19 +379,20 @@ static void vfio_iommu_map_notify(IOMMUNotifier *n, I= OMMUTLBEntry *iotlb) * of vaddr will always be there, even if the memory object is * destroyed and its backing memory munmap-ed. */ - ret =3D vfio_dma_map(container, iova, - iotlb->addr_mask + 1, vaddr, - read_only); + ret =3D vfio_container_dma_map(container, iova, + iotlb->addr_mask + 1, vaddr, + read_only); if (ret) { - error_report("vfio_dma_map(%p, 0x%"HWADDR_PRIx", " + error_report("vfio_container_dma_map(%p, 0x%"HWADDR_PRIx", " "0x%"HWADDR_PRIx", %p) =3D %d (%s)", container, iova, iotlb->addr_mask + 1, vaddr, ret, strerror(-ret)); } } else { - ret =3D vfio_dma_unmap(container, iova, iotlb->addr_mask + 1, iotl= b); + ret =3D vfio_container_dma_unmap(container, iova, + iotlb->addr_mask + 1, iotlb); if (ret) { - error_report("vfio_dma_unmap(%p, 0x%"HWADDR_PRIx", " + error_report("vfio_container_dma_unmap(%p, 0x%"HWADDR_PRIx", " "0x%"HWADDR_PRIx") =3D %d (%s)", container, iova, iotlb->addr_mask + 1, ret, strerror(-ret)); @@ -407,14 +408,15 @@ static void vfio_ram_discard_notify_discard(RamDiscar= dListener *rdl, { VFIORamDiscardListener *vrdl =3D container_of(rdl, VFIORamDiscardListe= ner, listener); + VFIOContainer *container =3D vrdl->container; const hwaddr size =3D int128_get64(section->size); const hwaddr iova =3D section->offset_within_address_space; int ret; =20 /* Unmap with a single call. */ - ret =3D vfio_dma_unmap(vrdl->container, iova, size , NULL); + ret =3D vfio_container_dma_unmap(container, iova, size , NULL); if (ret) { - error_report("%s: vfio_dma_unmap() failed: %s", __func__, + error_report("%s: vfio_container_dma_unmap() failed: %s", __func__, strerror(-ret)); } } @@ -424,6 +426,7 @@ static int vfio_ram_discard_notify_populate(RamDiscardL= istener *rdl, { VFIORamDiscardListener *vrdl =3D container_of(rdl, VFIORamDiscardListe= ner, listener); + VFIOContainer *container =3D vrdl->container; const hwaddr end =3D section->offset_within_region + int128_get64(section->size); hwaddr start, next, iova; @@ -442,8 +445,8 @@ static int vfio_ram_discard_notify_populate(RamDiscardL= istener *rdl, section->offset_within_address_space; vaddr =3D memory_region_get_ram_ptr(section->mr) + start; =20 - ret =3D vfio_dma_map(vrdl->container, iova, next - start, - vaddr, section->readonly); + ret =3D vfio_container_dma_map(container, iova, next - start, + vaddr, section->readonly); if (ret) { /* Rollback */ vfio_ram_discard_notify_discard(rdl, section); @@ -756,10 +759,10 @@ static void vfio_listener_region_add(MemoryListener *= listener, } } =20 - ret =3D vfio_dma_map(container, iova, int128_get64(llsize), - vaddr, section->readonly); + ret =3D vfio_container_dma_map(container, iova, int128_get64(llsize), + vaddr, section->readonly); if (ret) { - error_setg(&err, "vfio_dma_map(%p, 0x%"HWADDR_PRIx", " + error_setg(&err, "vfio_container_dma_map(%p, 0x%"HWADDR_PRIx", " "0x%"HWADDR_PRIx", %p) =3D %d (%s)", container, iova, int128_get64(llsize), vaddr, ret, strerror(-ret)); @@ -775,7 +778,7 @@ static void vfio_listener_region_add(MemoryListener *li= stener, =20 fail: if (memory_region_is_ram_device(section->mr)) { - error_report("failed to vfio_dma_map. pci p2p may not work"); + error_report("failed to vfio_container_dma_map. pci p2p may not wo= rk"); return; } /* @@ -860,18 +863,20 @@ static void vfio_listener_region_del(MemoryListener *= listener, if (int128_eq(llsize, int128_2_64())) { /* The unmap ioctl doesn't accept a full 64-bit span. */ llsize =3D int128_rshift(llsize, 1); - ret =3D vfio_dma_unmap(container, iova, int128_get64(llsize), = NULL); + ret =3D vfio_container_dma_unmap(container, iova, + int128_get64(llsize), NULL); if (ret) { - error_report("vfio_dma_unmap(%p, 0x%"HWADDR_PRIx", " + error_report("vfio_container_dma_unmap(%p, 0x%"HWADDR_PRIx= ", " "0x%"HWADDR_PRIx") =3D %d (%s)", container, iova, int128_get64(llsize), ret, strerror(-ret)); } iova +=3D int128_get64(llsize); } - ret =3D vfio_dma_unmap(container, iova, int128_get64(llsize), NULL= ); + ret =3D vfio_container_dma_unmap(container, iova, + int128_get64(llsize), NULL); if (ret) { - error_report("vfio_dma_unmap(%p, 0x%"HWADDR_PRIx", " + error_report("vfio_container_dma_unmap(%p, 0x%"HWADDR_PRIx", " "0x%"HWADDR_PRIx") =3D %d (%s)", container, iova, int128_get64(llsize), ret, strerror(-ret)); @@ -1103,7 +1108,7 @@ static void vfio_listener_log_global_start(MemoryList= ener *listener) if (vfio_devices_all_device_dirty_tracking(container)) { ret =3D vfio_devices_dma_logging_start(container); } else { - ret =3D vfio_set_dirty_page_tracking(container, true); + ret =3D vfio_container_set_dirty_page_tracking(container, true); } =20 if (ret) { @@ -1121,7 +1126,7 @@ static void vfio_listener_log_global_stop(MemoryListe= ner *listener) if (vfio_devices_all_device_dirty_tracking(container)) { vfio_devices_dma_logging_stop(container); } else { - ret =3D vfio_set_dirty_page_tracking(container, false); + ret =3D vfio_container_set_dirty_page_tracking(container, false); } =20 if (ret) { @@ -1204,7 +1209,7 @@ int vfio_get_dirty_bitmap(VFIOContainer *container, u= int64_t iova, if (all_device_dirty_tracking) { ret =3D vfio_devices_query_dirty_bitmap(container, &vbmap, iova, s= ize); } else { - ret =3D vfio_query_dirty_bitmap(container, &vbmap, iova, size); + ret =3D vfio_container_query_dirty_bitmap(container, &vbmap, iova,= size); } =20 if (ret) { @@ -1214,8 +1219,7 @@ int vfio_get_dirty_bitmap(VFIOContainer *container, u= int64_t iova, dirty_pages =3D cpu_physical_memory_set_dirty_lebitmap(vbmap.bitmap, r= am_addr, vbmap.pages); =20 - trace_vfio_get_dirty_bitmap(container->fd, iova, size, vbmap.size, - ram_addr, dirty_pages); + trace_vfio_get_dirty_bitmap(iova, size, vbmap.size, ram_addr, dirty_pa= ges); out: g_free(vbmap.bitmap); =20 @@ -1525,3 +1529,25 @@ retry: =20 return info; } + +int vfio_attach_device(char *name, VFIODevice *vbasedev, + AddressSpace *as, Error **errp) +{ + const VFIOIOMMUBackendOpsClass *ops; + + ops =3D VFIO_IOMMU_BACKEND_OPS_CLASS( + object_class_by_name(TYPE_VFIO_IOMMU_BACKEND_LEGACY_OPS)= ); + if (!ops) { + error_setg(errp, "VFIO IOMMU Backend not found!"); + return -ENODEV; + } + return ops->attach_device(name, vbasedev, as, errp); +} + +void vfio_detach_device(VFIODevice *vbasedev) +{ + if (!vbasedev->container) { + return; + } + vbasedev->container->ops->detach_device(vbasedev); +} diff --git a/hw/vfio/container-base.c b/hw/vfio/container-base.c new file mode 100644 index 0000000000..876e95c6dd --- /dev/null +++ b/hw/vfio/container-base.c @@ -0,0 +1,160 @@ +/* + * VFIO BASE CONTAINER + * + * Copyright (C) 2023 Intel Corporation. + * Copyright Red Hat, Inc. 2023 + * + * Authors: Yi Liu + * Eric Auger + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + + * You should have received a copy of the GNU General Public License along + * with this program; if not, see . + */ + +#include "qemu/osdep.h" +#include "qapi/error.h" +#include "qemu/error-report.h" +#include "hw/vfio/vfio-container-base.h" + +VFIODevice *vfio_container_dev_iter_next(VFIOContainer *container, + VFIODevice *curr) +{ + if (!container->ops->dev_iter_next) { + return NULL; + } + + return container->ops->dev_iter_next(container, curr); +} + +int vfio_container_dma_map(VFIOContainer *container, + hwaddr iova, ram_addr_t size, + void *vaddr, bool readonly) +{ + if (!container->ops->dma_map) { + return -EINVAL; + } + + return container->ops->dma_map(container, iova, size, vaddr, readonly); +} + +int vfio_container_dma_unmap(VFIOContainer *container, + hwaddr iova, ram_addr_t size, + IOMMUTLBEntry *iotlb) +{ + if (!container->ops->dma_unmap) { + return -EINVAL; + } + + return container->ops->dma_unmap(container, iova, size, iotlb); +} + +int vfio_container_set_dirty_page_tracking(VFIOContainer *container, + bool start) +{ + /* Fallback to all pages dirty if dirty page sync isn't supported */ + if (!container->ops->set_dirty_page_tracking) { + return 0; + } + + return container->ops->set_dirty_page_tracking(container, start); +} + +int vfio_container_query_dirty_bitmap(VFIOContainer *container, + VFIOBitmap *vbmap, + hwaddr iova, hwaddr size) +{ + if (!container->ops->query_dirty_bitmap) { + return -EINVAL; + } + + return container->ops->query_dirty_bitmap(container, vbmap, iova, size= ); +} + +int vfio_container_add_section_window(VFIOContainer *container, + MemoryRegionSection *section, + Error **errp) +{ + if (!container->ops->add_window) { + return 0; + } + + return container->ops->add_window(container, section, errp); +} + +void vfio_container_del_section_window(VFIOContainer *container, + MemoryRegionSection *section) +{ + if (!container->ops->del_window) { + return; + } + + return container->ops->del_window(container, section); +} + +void vfio_container_init(VFIOContainer *container, + VFIOAddressSpace *space, + struct VFIOIOMMUBackendOpsClass *ops) +{ + container->ops =3D ops; + container->space =3D space; + container->error =3D NULL; + container->dirty_pages_supported =3D false; + container->dma_max_mappings =3D 0; + QLIST_INIT(&container->giommu_list); + QLIST_INIT(&container->hostwin_list); + QLIST_INIT(&container->vrdl_list); +} + +void vfio_container_destroy(VFIOContainer *container) +{ + VFIORamDiscardListener *vrdl, *vrdl_tmp; + VFIOGuestIOMMU *giommu, *tmp; + VFIOHostDMAWindow *hostwin, *next; + + QLIST_SAFE_REMOVE(container, next); + + QLIST_FOREACH_SAFE(vrdl, &container->vrdl_list, next, vrdl_tmp) { + RamDiscardManager *rdm; + + rdm =3D memory_region_get_ram_discard_manager(vrdl->mr); + ram_discard_manager_unregister_listener(rdm, &vrdl->listener); + QLIST_REMOVE(vrdl, next); + g_free(vrdl); + } + + QLIST_FOREACH_SAFE(giommu, &container->giommu_list, giommu_next, tmp) { + memory_region_unregister_iommu_notifier( + MEMORY_REGION(giommu->iommu_mr), &giommu->n); + QLIST_REMOVE(giommu, giommu_next); + g_free(giommu); + } + + QLIST_FOREACH_SAFE(hostwin, &container->hostwin_list, hostwin_next, + next) { + QLIST_REMOVE(hostwin, hostwin_next); + g_free(hostwin); + } +} + +static const TypeInfo vfio_iommu_backend_ops_type_info =3D { + .name =3D TYPE_VFIO_IOMMU_BACKEND_OPS, + .parent =3D TYPE_OBJECT, + .abstract =3D true, + .class_size =3D sizeof(VFIOIOMMUBackendOpsClass), +}; + +static void vfio_iommu_backend_ops_register_types(void) +{ + type_register_static(&vfio_iommu_backend_ops_type_info); +} +type_init(vfio_iommu_backend_ops_register_types); diff --git a/hw/vfio/container.c b/hw/vfio/container.c index c71fddc09a..bb29b3612d 100644 --- a/hw/vfio/container.c +++ b/hw/vfio/container.c @@ -42,7 +42,8 @@ VFIOGroupList vfio_group_list =3D QLIST_HEAD_INITIALIZER(vfio_group_list); =20 -static int vfio_ram_block_discard_disable(VFIOContainer *container, bool s= tate) +static int vfio_ram_block_discard_disable(VFIOLegacyContainer *container, + bool state) { switch (container->iommu_type) { case VFIO_TYPE1v2_IOMMU: @@ -65,11 +66,18 @@ static int vfio_ram_block_discard_disable(VFIOContainer= *container, bool state) } } =20 -VFIODevice *vfio_container_dev_iter_next(VFIOContainer *container, - VFIODevice *curr) +static VFIODevice *vfio_legacy_dev_iter_next(VFIOContainer *bcontainer, + VFIODevice *curr) { VFIOGroup *group; =20 + assert(object_class_dynamic_cast(OBJECT_CLASS(bcontainer->ops), + TYPE_VFIO_IOMMU_BACKEND_LEGACY_OPS)); + + VFIOLegacyContainer *container =3D container_of(bcontainer, + VFIOLegacyContainer, + bcontainer); + if (!curr) { group =3D QLIST_FIRST(&container->group_list); } else { @@ -85,10 +93,11 @@ VFIODevice *vfio_container_dev_iter_next(VFIOContainer = *container, return QLIST_FIRST(&group->device_list); } =20 -static int vfio_dma_unmap_bitmap(VFIOContainer *container, +static int vfio_dma_unmap_bitmap(VFIOLegacyContainer *container, hwaddr iova, ram_addr_t size, IOMMUTLBEntry *iotlb) { + VFIOContainer *bcontainer =3D &container->bcontainer; struct vfio_iommu_type1_dma_unmap *unmap; struct vfio_bitmap *bitmap; VFIOBitmap vbmap; @@ -116,7 +125,7 @@ static int vfio_dma_unmap_bitmap(VFIOContainer *contain= er, bitmap->size =3D vbmap.size; bitmap->data =3D (__u64 *)vbmap.bitmap; =20 - if (vbmap.size > container->max_dirty_bitmap_size) { + if (vbmap.size > bcontainer->max_dirty_bitmap_size) { error_report("UNMAP: Size of bitmap too big 0x%"PRIx64, vbmap.size= ); ret =3D -E2BIG; goto unmap_exit; @@ -140,9 +149,13 @@ unmap_exit: /* * DMA - Mapping and unmapping for the "type1" IOMMU interface used on x86 */ -int vfio_dma_unmap(VFIOContainer *container, hwaddr iova, - ram_addr_t size, IOMMUTLBEntry *iotlb) +static int vfio_legacy_dma_unmap(VFIOContainer *bcontainer, hwaddr iova, + ram_addr_t size, IOMMUTLBEntry *iotlb) { + VFIOLegacyContainer *container =3D container_of(bcontainer, + VFIOLegacyContainer, + bcontainer); + struct vfio_iommu_type1_dma_unmap unmap =3D { .argsz =3D sizeof(unmap), .flags =3D 0, @@ -152,9 +165,9 @@ int vfio_dma_unmap(VFIOContainer *container, hwaddr iov= a, bool need_dirty_sync =3D false; int ret; =20 - if (iotlb && vfio_devices_all_running_and_mig_active(container)) { - if (!vfio_devices_all_device_dirty_tracking(container) && - container->dirty_pages_supported) { + if (iotlb && vfio_devices_all_running_and_mig_active(bcontainer)) { + if (!vfio_devices_all_device_dirty_tracking(bcontainer) && + bcontainer->dirty_pages_supported) { return vfio_dma_unmap_bitmap(container, iova, size, iotlb); } =20 @@ -176,8 +189,8 @@ int vfio_dma_unmap(VFIOContainer *container, hwaddr iov= a, */ if (errno =3D=3D EINVAL && unmap.size && !(unmap.iova + unmap.size= ) && container->iommu_type =3D=3D VFIO_TYPE1v2_IOMMU) { - trace_vfio_dma_unmap_overflow_workaround(); - unmap.size -=3D 1ULL << ctz64(container->pgsizes); + trace_vfio_legacy_dma_unmap_overflow_workaround(); + unmap.size -=3D 1ULL << ctz64(bcontainer->pgsizes); continue; } error_report("VFIO_UNMAP_DMA failed: %s", strerror(errno)); @@ -185,7 +198,7 @@ int vfio_dma_unmap(VFIOContainer *container, hwaddr iov= a, } =20 if (need_dirty_sync) { - ret =3D vfio_get_dirty_bitmap(container, iova, size, + ret =3D vfio_get_dirty_bitmap(bcontainer, iova, size, iotlb->translated_addr); if (ret) { return ret; @@ -195,9 +208,13 @@ int vfio_dma_unmap(VFIOContainer *container, hwaddr io= va, return 0; } =20 -int vfio_dma_map(VFIOContainer *container, hwaddr iova, - ram_addr_t size, void *vaddr, bool readonly) +static int vfio_legacy_dma_map(VFIOContainer *bcontainer, hwaddr iova, + ram_addr_t size, void *vaddr, bool readonly) { + VFIOLegacyContainer *container =3D container_of(bcontainer, + VFIOLegacyContainer, + bcontainer); + struct vfio_iommu_type1_dma_map map =3D { .argsz =3D sizeof(map), .flags =3D VFIO_DMA_MAP_FLAG_READ, @@ -216,7 +233,8 @@ int vfio_dma_map(VFIOContainer *container, hwaddr iova, * the VGA ROM space. */ if (ioctl(container->fd, VFIO_IOMMU_MAP_DMA, &map) =3D=3D 0 || - (errno =3D=3D EBUSY && vfio_dma_unmap(container, iova, size, NULL)= =3D=3D 0 && + (errno =3D=3D EBUSY && + vfio_legacy_dma_unmap(bcontainer, iova, size, NULL) =3D=3D 0 && ioctl(container->fd, VFIO_IOMMU_MAP_DMA, &map) =3D=3D 0)) { return 0; } @@ -225,14 +243,18 @@ int vfio_dma_map(VFIOContainer *container, hwaddr iov= a, return -errno; } =20 -int vfio_set_dirty_page_tracking(VFIOContainer *container, bool start) +static int vfio_legacy_set_dirty_page_tracking(VFIOContainer *bcontainer, + bool start) { + VFIOLegacyContainer *container =3D container_of(bcontainer, + VFIOLegacyContainer, + bcontainer); int ret; struct vfio_iommu_type1_dirty_bitmap dirty =3D { .argsz =3D sizeof(dirty), }; =20 - if (!container->dirty_pages_supported) { + if (!bcontainer->dirty_pages_supported) { return 0; } =20 @@ -252,9 +274,13 @@ int vfio_set_dirty_page_tracking(VFIOContainer *contai= ner, bool start) return ret; } =20 -int vfio_query_dirty_bitmap(VFIOContainer *container, VFIOBitmap *vbmap, - hwaddr iova, hwaddr size) +static int vfio_legacy_query_dirty_bitmap(VFIOContainer *bcontainer, + VFIOBitmap *vbmap, + hwaddr iova, hwaddr size) { + VFIOLegacyContainer *container =3D container_of(bcontainer, + VFIOLegacyContainer, + bcontainer); struct vfio_iommu_type1_dirty_bitmap *dbitmap; struct vfio_iommu_type1_dirty_bitmap_get *range; int ret; @@ -289,18 +315,24 @@ int vfio_query_dirty_bitmap(VFIOContainer *container,= VFIOBitmap *vbmap, return ret; } =20 -static void vfio_listener_release(VFIOContainer *container) +static void vfio_listener_release(VFIOLegacyContainer *container) { - memory_listener_unregister(&container->listener); + VFIOContainer *bcontainer =3D &container->bcontainer; + + memory_listener_unregister(&bcontainer->listener); if (container->iommu_type =3D=3D VFIO_SPAPR_TCE_v2_IOMMU) { memory_listener_unregister(&container->prereg_listener); } } =20 -int vfio_container_add_section_window(VFIOContainer *container, - MemoryRegionSection *section, - Error **errp) +static int +vfio_legacy_add_section_window(VFIOContainer *bcontainer, + MemoryRegionSection *section, + Error **errp) { + VFIOLegacyContainer *container =3D container_of(bcontainer, + VFIOLegacyContainer, + bcontainer); VFIOHostDMAWindow *hostwin; hwaddr pgsize =3D 0; int ret; @@ -310,7 +342,7 @@ int vfio_container_add_section_window(VFIOContainer *co= ntainer, } =20 /* For now intersections are not allowed, we may relax this later */ - QLIST_FOREACH(hostwin, &container->hostwin_list, hostwin_next) { + QLIST_FOREACH(hostwin, &bcontainer->hostwin_list, hostwin_next) { if (ranges_overlap(hostwin->min_iova, hostwin->max_iova - hostwin->min_iova + 1, section->offset_within_address_space, @@ -332,7 +364,7 @@ int vfio_container_add_section_window(VFIOContainer *co= ntainer, return ret; } =20 - vfio_host_win_add(container, section->offset_within_address_space, + vfio_host_win_add(bcontainer, section->offset_within_address_space, section->offset_within_address_space + int128_get64(section->size) - 1, pgsize); #ifdef CONFIG_KVM @@ -365,16 +397,21 @@ int vfio_container_add_section_window(VFIOContainer *= container, return 0; } =20 -void vfio_container_del_section_window(VFIOContainer *container, - MemoryRegionSection *section) +static void +vfio_legacy_del_section_window(VFIOContainer *bcontainer, + MemoryRegionSection *section) { + VFIOLegacyContainer *container =3D container_of(bcontainer, + VFIOLegacyContainer, + bcontainer); + if (container->iommu_type !=3D VFIO_SPAPR_TCE_v2_IOMMU) { return; } =20 vfio_spapr_remove_window(container, section->offset_within_address_space); - if (vfio_host_win_del(container, + if (vfio_host_win_del(bcontainer, section->offset_within_address_space, section->offset_within_address_space + int128_get64(section->size) - 1) < 0) { @@ -427,7 +464,7 @@ static void vfio_kvm_device_del_group(VFIOGroup *group) /* * vfio_get_iommu_type - selects the richest iommu_type (v2 first) */ -static int vfio_get_iommu_type(VFIOContainer *container, +static int vfio_get_iommu_type(VFIOLegacyContainer *container, Error **errp) { int iommu_types[] =3D { VFIO_TYPE1v2_IOMMU, VFIO_TYPE1_IOMMU, @@ -443,7 +480,7 @@ static int vfio_get_iommu_type(VFIOContainer *container, return -EINVAL; } =20 -static int vfio_init_container(VFIOContainer *container, int group_fd, +static int vfio_init_container(VFIOLegacyContainer *container, int group_f= d, Error **errp) { int iommu_type, ret; @@ -478,7 +515,7 @@ static int vfio_init_container(VFIOContainer *container= , int group_fd, return 0; } =20 -static int vfio_get_iommu_info(VFIOContainer *container, +static int vfio_get_iommu_info(VFIOLegacyContainer *container, struct vfio_iommu_type1_info **info) { =20 @@ -522,11 +559,12 @@ vfio_get_iommu_info_cap(struct vfio_iommu_type1_info = *info, uint16_t id) return NULL; } =20 -static void vfio_get_iommu_info_migration(VFIOContainer *container, - struct vfio_iommu_type1_info *inf= o) +static void vfio_get_iommu_info_migration(VFIOLegacyContainer *container, + struct vfio_iommu_type1_info *in= fo) { struct vfio_info_cap_header *hdr; struct vfio_iommu_type1_info_cap_migration *cap_mig; + VFIOContainer *bcontainer =3D &container->bcontainer; =20 hdr =3D vfio_get_iommu_info_cap(info, VFIO_IOMMU_TYPE1_INFO_CAP_MIGRAT= ION); if (!hdr) { @@ -541,16 +579,19 @@ static void vfio_get_iommu_info_migration(VFIOContain= er *container, * qemu_real_host_page_size to mark those dirty. */ if (cap_mig->pgsize_bitmap & qemu_real_host_page_size()) { - container->dirty_pages_supported =3D true; - container->max_dirty_bitmap_size =3D cap_mig->max_dirty_bitmap_siz= e; - container->dirty_pgsizes =3D cap_mig->pgsize_bitmap; + bcontainer->dirty_pages_supported =3D true; + bcontainer->max_dirty_bitmap_size =3D cap_mig->max_dirty_bitmap_si= ze; + bcontainer->dirty_pgsizes =3D cap_mig->pgsize_bitmap; } } =20 static int vfio_connect_container(VFIOGroup *group, AddressSpace *as, Error **errp) { - VFIOContainer *container; + VFIOIOMMUBackendOpsClass *ops =3D VFIO_IOMMU_BACKEND_OPS_CLASS( + object_class_by_name(TYPE_VFIO_IOMMU_BACKEND_LEGACY_OPS)); + VFIOContainer *bcontainer; + VFIOLegacyContainer *container; int ret, fd; VFIOAddressSpace *space; =20 @@ -587,7 +628,8 @@ static int vfio_connect_container(VFIOGroup *group, Add= ressSpace *as, * details once we know which type of IOMMU we are using. */ =20 - QLIST_FOREACH(container, &space->containers, next) { + QLIST_FOREACH(bcontainer, &space->containers, next) { + container =3D container_of(bcontainer, VFIOLegacyContainer, bconta= iner); if (!ioctl(group->fd, VFIO_GROUP_SET_CONTAINER, &container->fd)) { ret =3D vfio_ram_block_discard_disable(container, true); if (ret) { @@ -623,14 +665,9 @@ static int vfio_connect_container(VFIOGroup *group, Ad= dressSpace *as, } =20 container =3D g_malloc0(sizeof(*container)); - container->space =3D space; container->fd =3D fd; - container->error =3D NULL; - container->dirty_pages_supported =3D false; - container->dma_max_mappings =3D 0; - QLIST_INIT(&container->giommu_list); - QLIST_INIT(&container->hostwin_list); - QLIST_INIT(&container->vrdl_list); + bcontainer =3D &container->bcontainer; + vfio_container_init(bcontainer, space, ops); =20 ret =3D vfio_init_container(container, group->fd, errp); if (ret) { @@ -656,13 +693,13 @@ static int vfio_connect_container(VFIOGroup *group, A= ddressSpace *as, } =20 if (info->flags & VFIO_IOMMU_INFO_PGSIZES) { - container->pgsizes =3D info->iova_pgsizes; + bcontainer->pgsizes =3D info->iova_pgsizes; } else { - container->pgsizes =3D qemu_real_host_page_size(); + bcontainer->pgsizes =3D qemu_real_host_page_size(); } =20 - if (!vfio_get_info_dma_avail(info, &container->dma_max_mappings)) { - container->dma_max_mappings =3D 65535; + if (!vfio_get_info_dma_avail(info, &bcontainer->dma_max_mappings))= { + bcontainer->dma_max_mappings =3D 65535; } vfio_get_iommu_info_migration(container, info); g_free(info); @@ -672,7 +709,7 @@ static int vfio_connect_container(VFIOGroup *group, Add= ressSpace *as, * information to get the actual window extent rather than assume * a 64-bit IOVA address space. */ - vfio_host_win_add(container, 0, (hwaddr)-1, container->pgsizes); + vfio_host_win_add(bcontainer, 0, (hwaddr)-1, bcontainer->pgsizes); =20 break; } @@ -699,10 +736,10 @@ static int vfio_connect_container(VFIOGroup *group, A= ddressSpace *as, =20 memory_listener_register(&container->prereg_listener, &address_space_memory); - if (container->error) { + if (bcontainer->error) { memory_listener_unregister(&container->prereg_listener); ret =3D -1; - error_propagate_prepend(errp, container->error, + error_propagate_prepend(errp, bcontainer->error, "RAM memory listener initialization failed: "); goto enable_discards_exit; } @@ -721,7 +758,7 @@ static int vfio_connect_container(VFIOGroup *group, Add= ressSpace *as, } =20 if (v2) { - container->pgsizes =3D info.ddw.pgsizes; + bcontainer->pgsizes =3D info.ddw.pgsizes; /* * There is a default window in just created container. * To make region_add/del simpler, we better remove this @@ -736,8 +773,8 @@ static int vfio_connect_container(VFIOGroup *group, Add= ressSpace *as, } } else { /* The default table uses 4K pages */ - container->pgsizes =3D 0x1000; - vfio_host_win_add(container, info.dma32_window_start, + bcontainer->pgsizes =3D 0x1000; + vfio_host_win_add(bcontainer, info.dma32_window_start, info.dma32_window_start + info.dma32_window_size - 1, 0x1000); @@ -748,28 +785,28 @@ static int vfio_connect_container(VFIOGroup *group, A= ddressSpace *as, vfio_kvm_device_add_group(group); =20 QLIST_INIT(&container->group_list); - QLIST_INSERT_HEAD(&space->containers, container, next); + QLIST_INSERT_HEAD(&space->containers, bcontainer, next); =20 group->container =3D container; QLIST_INSERT_HEAD(&container->group_list, group, container_next); =20 - container->listener =3D vfio_memory_listener; + bcontainer->listener =3D vfio_memory_listener; =20 - memory_listener_register(&container->listener, container->space->as); + memory_listener_register(&bcontainer->listener, bcontainer->space->as); =20 - if (container->error) { + if (bcontainer->error) { ret =3D -1; - error_propagate_prepend(errp, container->error, + error_propagate_prepend(errp, bcontainer->error, "memory listener initialization failed: "); goto listener_release_exit; } =20 - container->initialized =3D true; + bcontainer->initialized =3D true; =20 return 0; listener_release_exit: QLIST_REMOVE(group, container_next); - QLIST_REMOVE(container, next); + QLIST_REMOVE(bcontainer, next); vfio_kvm_device_del_group(group); vfio_listener_release(container); =20 @@ -790,7 +827,8 @@ put_space_exit: =20 static void vfio_disconnect_container(VFIOGroup *group) { - VFIOContainer *container =3D group->container; + VFIOLegacyContainer *container =3D group->container; + VFIOContainer *bcontainer =3D &container->bcontainer; =20 QLIST_REMOVE(group, container_next); group->container =3D NULL; @@ -810,25 +848,9 @@ static void vfio_disconnect_container(VFIOGroup *group) } =20 if (QLIST_EMPTY(&container->group_list)) { - VFIOAddressSpace *space =3D container->space; - VFIOGuestIOMMU *giommu, *tmp; - VFIOHostDMAWindow *hostwin, *next; - - QLIST_REMOVE(container, next); - - QLIST_FOREACH_SAFE(giommu, &container->giommu_list, giommu_next, t= mp) { - memory_region_unregister_iommu_notifier( - MEMORY_REGION(giommu->iommu_mr), &giommu->n); - QLIST_REMOVE(giommu, giommu_next); - g_free(giommu); - } - - QLIST_FOREACH_SAFE(hostwin, &container->hostwin_list, hostwin_next, - next) { - QLIST_REMOVE(hostwin, hostwin_next); - g_free(hostwin); - } + VFIOAddressSpace *space =3D bcontainer->space; =20 + vfio_container_destroy(bcontainer); trace_vfio_disconnect_container(container->fd); close(container->fd); g_free(container); @@ -840,13 +862,15 @@ static void vfio_disconnect_container(VFIOGroup *grou= p) static VFIOGroup *vfio_get_group(int groupid, AddressSpace *as, Error **er= rp) { VFIOGroup *group; + VFIOContainer *bcontainer; char path[32]; struct vfio_group_status status =3D { .argsz =3D sizeof(status) }; =20 QLIST_FOREACH(group, &vfio_group_list, next) { if (group->groupid =3D=3D groupid) { /* Found it. Now is it already in the right context? */ - if (group->container->space->as =3D=3D as) { + bcontainer =3D &group->container->bcontainer; + if (bcontainer->space->as =3D=3D as) { return group; } else { error_setg(errp, "group %d used in multiple address spaces= ", @@ -990,7 +1014,7 @@ static void vfio_put_base_device(VFIODevice *vbasedev) /* * Interfaces for IBM EEH (Enhanced Error Handling) */ -static bool vfio_eeh_container_ok(VFIOContainer *container) +static bool vfio_eeh_container_ok(VFIOLegacyContainer *container) { /* * As of 2016-03-04 (linux-4.5) the host kernel EEH/VFIO @@ -1018,7 +1042,7 @@ static bool vfio_eeh_container_ok(VFIOContainer *cont= ainer) return true; } =20 -static int vfio_eeh_container_op(VFIOContainer *container, uint32_t op) +static int vfio_eeh_container_op(VFIOLegacyContainer *container, uint32_t = op) { struct vfio_eeh_pe_op pe_op =3D { .argsz =3D sizeof(pe_op), @@ -1041,19 +1065,21 @@ static int vfio_eeh_container_op(VFIOContainer *con= tainer, uint32_t op) return ret; } =20 -static VFIOContainer *vfio_eeh_as_container(AddressSpace *as) +static VFIOLegacyContainer *vfio_eeh_as_container(AddressSpace *as) { VFIOAddressSpace *space =3D vfio_get_address_space(as); - VFIOContainer *container =3D NULL; + VFIOLegacyContainer *container =3D NULL; + VFIOContainer *bcontainer =3D NULL; =20 if (QLIST_EMPTY(&space->containers)) { /* No containers to act on */ goto out; } =20 - container =3D QLIST_FIRST(&space->containers); + bcontainer =3D QLIST_FIRST(&space->containers); + container =3D container_of(bcontainer, VFIOLegacyContainer, bcontainer= ); =20 - if (QLIST_NEXT(container, next)) { + if (QLIST_NEXT(bcontainer, next)) { /* * We don't yet have logic to synchronize EEH state across * multiple containers @@ -1069,14 +1095,14 @@ out: =20 bool vfio_eeh_as_ok(AddressSpace *as) { - VFIOContainer *container =3D vfio_eeh_as_container(as); + VFIOLegacyContainer *container =3D vfio_eeh_as_container(as); =20 return (container !=3D NULL) && vfio_eeh_container_ok(container); } =20 int vfio_eeh_as_op(AddressSpace *as, uint32_t op) { - VFIOContainer *container =3D vfio_eeh_as_container(as); + VFIOLegacyContainer *container =3D vfio_eeh_as_container(as); =20 if (!container) { return -ENODEV; @@ -1110,8 +1136,8 @@ static int vfio_device_groupid(VFIODevice *vbasedev, = Error **errp) return groupid; } =20 -int vfio_attach_device(char *name, VFIODevice *vbasedev, - AddressSpace *as, Error **errp) +static int vfio_legacy_attach_device(char *name, VFIODevice *vbasedev, + AddressSpace *as, Error **errp) { int groupid =3D vfio_device_groupid(vbasedev, errp); VFIODevice *vbasedev_iter; @@ -1137,15 +1163,46 @@ int vfio_attach_device(char *name, VFIODevice *vbas= edev, ret =3D vfio_get_device(group, name, vbasedev, errp); if (ret) { vfio_put_group(group); + return ret; } + vbasedev->container =3D &group->container->bcontainer; =20 return ret; } =20 -void vfio_detach_device(VFIODevice *vbasedev) +static void vfio_legacy_detach_device(VFIODevice *vbasedev) { VFIOGroup *group =3D vbasedev->group; =20 vfio_put_base_device(vbasedev); vfio_put_group(group); + vbasedev->container =3D NULL; +} + +static void vfio_iommu_backend_legacy_ops_class_init(ObjectClass *oc, + void *data) { + VFIOIOMMUBackendOpsClass *ops =3D VFIO_IOMMU_BACKEND_OPS_CLASS(oc); + + ops->dev_iter_next =3D vfio_legacy_dev_iter_next; + ops->dma_map =3D vfio_legacy_dma_map; + ops->dma_unmap =3D vfio_legacy_dma_unmap; + ops->attach_device =3D vfio_legacy_attach_device; + ops->detach_device =3D vfio_legacy_detach_device; + ops->set_dirty_page_tracking =3D vfio_legacy_set_dirty_page_tracking; + ops->query_dirty_bitmap =3D vfio_legacy_query_dirty_bitmap; + ops->add_window =3D vfio_legacy_add_section_window; + ops->del_window =3D vfio_legacy_del_section_window; +} + +static const TypeInfo vfio_iommu_backend_legacy_ops_type =3D { + .name =3D TYPE_VFIO_IOMMU_BACKEND_LEGACY_OPS, + + .parent =3D TYPE_VFIO_IOMMU_BACKEND_OPS, + .class_init =3D vfio_iommu_backend_legacy_ops_class_init, + .abstract =3D true, +}; +static void vfio_iommu_backend_legacy_ops_register_types(void) +{ + type_register_static(&vfio_iommu_backend_legacy_ops_type); } +type_init(vfio_iommu_backend_legacy_ops_register_types); diff --git a/hw/vfio/meson.build b/hw/vfio/meson.build index 2a6912c940..eb6ce6229d 100644 --- a/hw/vfio/meson.build +++ b/hw/vfio/meson.build @@ -2,6 +2,7 @@ vfio_ss =3D ss.source_set() vfio_ss.add(files( 'helpers.c', 'common.c', + 'container-base.c', 'container.c', 'spapr.c', 'migration.c', diff --git a/hw/vfio/spapr.c b/hw/vfio/spapr.c index 9ec1e95f6d..7647e7d492 100644 --- a/hw/vfio/spapr.c +++ b/hw/vfio/spapr.c @@ -39,8 +39,8 @@ static void *vfio_prereg_gpa_to_vaddr(MemoryRegionSection= *section, hwaddr gpa) static void vfio_prereg_listener_region_add(MemoryListener *listener, MemoryRegionSection *section) { - VFIOContainer *container =3D container_of(listener, VFIOContainer, - prereg_listener); + VFIOLegacyContainer *container =3D container_of(listener, VFIOLegacyCo= ntainer, + prereg_listener); const hwaddr gpa =3D section->offset_within_address_space; hwaddr end; int ret; @@ -83,9 +83,9 @@ static void vfio_prereg_listener_region_add(MemoryListene= r *listener, * can gracefully fail. Runtime, there's not much we can do other * than throw a hardware error. */ - if (!container->initialized) { - if (!container->error) { - error_setg_errno(&container->error, -ret, + if (!container->bcontainer.initialized) { + if (!container->bcontainer.error) { + error_setg_errno(&container->bcontainer.error, -ret, "Memory registering failed"); } } else { @@ -97,8 +97,8 @@ static void vfio_prereg_listener_region_add(MemoryListene= r *listener, static void vfio_prereg_listener_region_del(MemoryListener *listener, MemoryRegionSection *section) { - VFIOContainer *container =3D container_of(listener, VFIOContainer, - prereg_listener); + VFIOLegacyContainer *container =3D container_of(listener, VFIOLegacyCo= ntainer, + prereg_listener); const hwaddr gpa =3D section->offset_within_address_space; hwaddr end; int ret; @@ -141,7 +141,7 @@ const MemoryListener vfio_prereg_listener =3D { .region_del =3D vfio_prereg_listener_region_del, }; =20 -int vfio_spapr_create_window(VFIOContainer *container, +int vfio_spapr_create_window(VFIOLegacyContainer *container, MemoryRegionSection *section, hwaddr *pgsize) { @@ -159,13 +159,13 @@ int vfio_spapr_create_window(VFIOContainer *container, if (pagesize > rampagesize) { pagesize =3D rampagesize; } - pgmask =3D container->pgsizes & (pagesize | (pagesize - 1)); + pgmask =3D container->bcontainer.pgsizes & (pagesize | (pagesize - 1)); pagesize =3D pgmask ? (1ULL << (63 - clz64(pgmask))) : 0; if (!pagesize) { error_report("Host doesn't support page size 0x%"PRIx64 ", the supported mask is 0x%lx", memory_region_iommu_get_min_page_size(iommu_mr), - container->pgsizes); + container->bcontainer.pgsizes); return -EINVAL; } =20 @@ -233,7 +233,7 @@ int vfio_spapr_create_window(VFIOContainer *container, return 0; } =20 -int vfio_spapr_remove_window(VFIOContainer *container, +int vfio_spapr_remove_window(VFIOLegacyContainer *container, hwaddr offset_within_address_space) { struct vfio_iommu_spapr_tce_remove remove =3D { diff --git a/hw/vfio/trace-events b/hw/vfio/trace-events index bd32970854..1692bcd8f1 100644 --- a/hw/vfio/trace-events +++ b/hw/vfio/trace-events @@ -119,8 +119,8 @@ vfio_region_unmap(const char *name, unsigned long offse= t, unsigned long end) "Re vfio_region_sparse_mmap_header(const char *name, int index, int nr_areas) = "Device %s region %d: %d sparse mmap entries" vfio_region_sparse_mmap_entry(int i, unsigned long start, unsigned long en= d) "sparse entry %d [0x%lx - 0x%lx]" vfio_get_dev_region(const char *name, int index, uint32_t type, uint32_t s= ubtype) "%s index %d, %08x/%08x" -vfio_dma_unmap_overflow_workaround(void) "" -vfio_get_dirty_bitmap(int fd, uint64_t iova, uint64_t size, uint64_t bitma= p_size, uint64_t start, uint64_t dirty_pages) "container fd=3D%d, iova=3D0x= %"PRIx64" size=3D 0x%"PRIx64" bitmap_size=3D0x%"PRIx64" start=3D0x%"PRIx64"= dirty_pages=3D%"PRIu64 +vfio_legacy_dma_unmap_overflow_workaround(void) "" +vfio_get_dirty_bitmap(uint64_t iova, uint64_t size, uint64_t bitmap_size, = uint64_t start, uint64_t dirty_pages) "iova=3D0x%"PRIx64" size=3D 0x%"PRIx6= 4" bitmap_size=3D0x%"PRIx64" start=3D0x%"PRIx64" dirty_pages=3D%"PRIu64 vfio_iommu_map_dirty_notify(uint64_t iova_start, uint64_t iova_end) "iommu= dirty @ 0x%"PRIx64" - 0x%"PRIx64 =20 # platform.c diff --git a/include/hw/vfio/vfio-common.h b/include/hw/vfio/vfio-common.h index 95bcafdaf6..b1a76dcc9c 100644 --- a/include/hw/vfio/vfio-common.h +++ b/include/hw/vfio/vfio-common.h @@ -30,6 +30,7 @@ #include #endif #include "sysemu/sysemu.h" +#include "hw/vfio/vfio-container-base.h" =20 #define VFIO_MSG_PREFIX "vfio %s: " =20 @@ -74,64 +75,22 @@ typedef struct VFIOMigration { bool initial_data_sent; } VFIOMigration; =20 -typedef struct VFIOAddressSpace { - AddressSpace *as; - QLIST_HEAD(, VFIOContainer) containers; - QLIST_ENTRY(VFIOAddressSpace) list; -} VFIOAddressSpace; - struct VFIOGroup; =20 -typedef struct VFIOContainer { - VFIOAddressSpace *space; +typedef struct VFIOLegacyContainer { + VFIOContainer bcontainer; int fd; /* /dev/vfio/vfio, empowered by the attached groups */ - MemoryListener listener; MemoryListener prereg_listener; unsigned iommu_type; - Error *error; - bool initialized; - bool dirty_pages_supported; - uint64_t dirty_pgsizes; - uint64_t max_dirty_bitmap_size; - unsigned long pgsizes; - unsigned int dma_max_mappings; - QLIST_HEAD(, VFIOGuestIOMMU) giommu_list; - QLIST_HEAD(, VFIOHostDMAWindow) hostwin_list; QLIST_HEAD(, VFIOGroup) group_list; - QLIST_HEAD(, VFIORamDiscardListener) vrdl_list; - QLIST_ENTRY(VFIOContainer) next; -} VFIOContainer; - -typedef struct VFIOGuestIOMMU { - VFIOContainer *container; - IOMMUMemoryRegion *iommu_mr; - hwaddr iommu_offset; - IOMMUNotifier n; - QLIST_ENTRY(VFIOGuestIOMMU) giommu_next; -} VFIOGuestIOMMU; - -typedef struct VFIORamDiscardListener { - VFIOContainer *container; - MemoryRegion *mr; - hwaddr offset_within_address_space; - hwaddr size; - uint64_t granularity; - RamDiscardListener listener; - QLIST_ENTRY(VFIORamDiscardListener) next; -} VFIORamDiscardListener; - -typedef struct VFIOHostDMAWindow { - hwaddr min_iova; - hwaddr max_iova; - uint64_t iova_pgsizes; - QLIST_ENTRY(VFIOHostDMAWindow) hostwin_next; -} VFIOHostDMAWindow; +} VFIOLegacyContainer; =20 typedef struct VFIODeviceOps VFIODeviceOps; =20 typedef struct VFIODevice { QLIST_ENTRY(VFIODevice) next; struct VFIOGroup *group; + VFIOContainer *container; char *sysfsdev; char *name; DeviceState *dev; @@ -165,7 +124,7 @@ struct VFIODeviceOps { typedef struct VFIOGroup { int fd; int groupid; - VFIOContainer *container; + VFIOLegacyContainer *container; QLIST_HEAD(, VFIODevice) device_list; QLIST_ENTRY(VFIOGroup) next; QLIST_ENTRY(VFIOGroup) container_next; @@ -198,37 +157,13 @@ typedef struct VFIODisplay { } dmabuf; } VFIODisplay; =20 -typedef struct { - unsigned long *bitmap; - hwaddr size; - hwaddr pages; -} VFIOBitmap; - -void vfio_host_win_add(VFIOContainer *container, +void vfio_host_win_add(VFIOContainer *bcontainer, hwaddr min_iova, hwaddr max_iova, uint64_t iova_pgsizes); -int vfio_host_win_del(VFIOContainer *container, hwaddr min_iova, +int vfio_host_win_del(VFIOContainer *bcontainer, hwaddr min_iova, hwaddr max_iova); VFIOAddressSpace *vfio_get_address_space(AddressSpace *as); void vfio_put_address_space(VFIOAddressSpace *space); -bool vfio_devices_all_running_and_saving(VFIOContainer *container); - -/* container->fd */ -VFIODevice *vfio_container_dev_iter_next(VFIOContainer *container, - VFIODevice *curr); -int vfio_dma_unmap(VFIOContainer *container, hwaddr iova, - ram_addr_t size, IOMMUTLBEntry *iotlb); -int vfio_dma_map(VFIOContainer *container, hwaddr iova, - ram_addr_t size, void *vaddr, bool readonly); -int vfio_set_dirty_page_tracking(VFIOContainer *container, bool start); -int vfio_query_dirty_bitmap(VFIOContainer *container, VFIOBitmap *vbmap, - hwaddr iova, hwaddr size); - -int vfio_container_add_section_window(VFIOContainer *container, - MemoryRegionSection *section, - Error **errp); -void vfio_container_del_section_window(VFIOContainer *container, - MemoryRegionSection *section); =20 void vfio_disable_irqindex(VFIODevice *vbasedev, int index); void vfio_unmask_single_irqindex(VFIODevice *vbasedev, int index); @@ -285,10 +220,10 @@ vfio_get_cap(void *ptr, uint32_t cap_offset, uint16_t= id); #endif extern const MemoryListener vfio_prereg_listener; =20 -int vfio_spapr_create_window(VFIOContainer *container, +int vfio_spapr_create_window(VFIOLegacyContainer *container, MemoryRegionSection *section, hwaddr *pgsize); -int vfio_spapr_remove_window(VFIOContainer *container, +int vfio_spapr_remove_window(VFIOLegacyContainer *container, hwaddr offset_within_address_space); =20 bool vfio_migration_realize(VFIODevice *vbasedev, Error **errp); diff --git a/include/hw/vfio/vfio-container-base.h b/include/hw/vfio/vfio-c= ontainer-base.h new file mode 100644 index 0000000000..b18fa92146 --- /dev/null +++ b/include/hw/vfio/vfio-container-base.h @@ -0,0 +1,155 @@ +/* + * VFIO BASE CONTAINER + * + * Copyright (C) 2023 Intel Corporation. + * Copyright Red Hat, Inc. 2023 + * + * Authors: Yi Liu + * Eric Auger + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + + * You should have received a copy of the GNU General Public License along + * with this program; if not, see . + */ + +#ifndef HW_VFIO_VFIO_BASE_CONTAINER_H +#define HW_VFIO_VFIO_BASE_CONTAINER_H + +#include "exec/memory.h" +#ifndef CONFIG_USER_ONLY +#include "exec/hwaddr.h" +#endif + +typedef struct VFIOContainer VFIOContainer; + +typedef struct VFIOAddressSpace { + AddressSpace *as; + QLIST_HEAD(, VFIOContainer) containers; + QLIST_ENTRY(VFIOAddressSpace) list; +} VFIOAddressSpace; + +typedef struct VFIOGuestIOMMU { + VFIOContainer *container; + IOMMUMemoryRegion *iommu_mr; + hwaddr iommu_offset; + IOMMUNotifier n; + QLIST_ENTRY(VFIOGuestIOMMU) giommu_next; +} VFIOGuestIOMMU; + +typedef struct VFIORamDiscardListener { + VFIOContainer *container; + MemoryRegion *mr; + hwaddr offset_within_address_space; + hwaddr size; + uint64_t granularity; + RamDiscardListener listener; + QLIST_ENTRY(VFIORamDiscardListener) next; +} VFIORamDiscardListener; + +typedef struct VFIOHostDMAWindow { + hwaddr min_iova; + hwaddr max_iova; + uint64_t iova_pgsizes; + QLIST_ENTRY(VFIOHostDMAWindow) hostwin_next; +} VFIOHostDMAWindow; + +typedef struct { + unsigned long *bitmap; + hwaddr size; + hwaddr pages; +} VFIOBitmap; + +typedef struct VFIODevice VFIODevice; +typedef struct VFIOIOMMUBackendOpsClass VFIOIOMMUBackendOpsClass; + +/* + * This is the base object for vfio container backends + */ +struct VFIOContainer { + VFIOIOMMUBackendOpsClass *ops; + VFIOAddressSpace *space; + MemoryListener listener; + Error *error; + bool initialized; + bool dirty_pages_supported; + uint64_t dirty_pgsizes; + uint64_t max_dirty_bitmap_size; + unsigned long pgsizes; + unsigned int dma_max_mappings; + QLIST_HEAD(, VFIOGuestIOMMU) giommu_list; + QLIST_HEAD(, VFIOHostDMAWindow) hostwin_list; + QLIST_HEAD(, VFIORamDiscardListener) vrdl_list; + QLIST_ENTRY(VFIOContainer) next; +}; + +VFIODevice *vfio_container_dev_iter_next(VFIOContainer *container, + VFIODevice *curr); +int vfio_container_dma_map(VFIOContainer *container, + hwaddr iova, ram_addr_t size, + void *vaddr, bool readonly); +int vfio_container_dma_unmap(VFIOContainer *container, + hwaddr iova, ram_addr_t size, + IOMMUTLBEntry *iotlb); +bool vfio_container_devices_all_dirty_tracking(VFIOContainer *container); +int vfio_container_set_dirty_page_tracking(VFIOContainer *container, + bool start); +int vfio_container_query_dirty_bitmap(VFIOContainer *container, + VFIOBitmap *vbmap, + hwaddr iova, hwaddr size); +int vfio_container_add_section_window(VFIOContainer *container, + MemoryRegionSection *section, + Error **errp); +void vfio_container_del_section_window(VFIOContainer *container, + MemoryRegionSection *section); + +void vfio_container_init(VFIOContainer *container, + VFIOAddressSpace *space, + struct VFIOIOMMUBackendOpsClass *ops); +void vfio_container_destroy(VFIOContainer *container); + +#define TYPE_VFIO_IOMMU_BACKEND_LEGACY_OPS "vfio-iommu-backend-legacy-ops" +#define TYPE_VFIO_IOMMU_BACKEND_OPS "vfio-iommu-backend-ops" + +DECLARE_CLASS_CHECKERS(VFIOIOMMUBackendOpsClass, + VFIO_IOMMU_BACKEND_OPS, TYPE_VFIO_IOMMU_BACKEND_OPS) + +struct VFIOIOMMUBackendOpsClass { + /*< private >*/ + ObjectClass parent_class; + + /*< public >*/ + /* required */ + VFIODevice *(*dev_iter_next)(VFIOContainer *container, VFIODevice *cur= r); + int (*dma_map)(VFIOContainer *container, + hwaddr iova, ram_addr_t size, + void *vaddr, bool readonly); + int (*dma_unmap)(VFIOContainer *container, + hwaddr iova, ram_addr_t size, + IOMMUTLBEntry *iotlb); + int (*attach_device)(char *name, VFIODevice *vbasedev, + AddressSpace *as, Error **errp); + void (*detach_device)(VFIODevice *vbasedev); + /* migration feature */ + int (*set_dirty_page_tracking)(VFIOContainer *container, bool start); + int (*query_dirty_bitmap)(VFIOContainer *bcontainer, VFIOBitmap *vbmap, + hwaddr iova, hwaddr size); + + /* SPAPR specific */ + int (*add_window)(VFIOContainer *container, + MemoryRegionSection *section, + Error **errp); + void (*del_window)(VFIOContainer *container, + MemoryRegionSection *section); +}; + + +#endif /* HW_VFIO_VFIO_BASE_CONTAINER_H */ --=20 2.34.1 From nobody Wed May 15 22:12:51 2024 Delivered-To: importer@patchew.org Authentication-Results: mx.zohomail.com; dkim=pass header.i=@intel.com; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass(p=none dis=none) header.from=intel.com ARC-Seal: i=1; a=rsa-sha256; t=1693392898; cv=none; d=zohomail.com; s=zohoarc; b=cmJi4ST1Xx7sdekajNR9kaYeSSWrsZwjbUA2OuuSum9Awd3bSw/le3cGQu82Kohoj0FcAnRcoAsJS7T5bDmhg6V/5pRDx2ETEPHI+9aaG+cqfMLq0P64ZbVDNjJwSfGDU2FGytK28mcHiYEyrOceQNyGHvDWU6N5ZTrayv+Kia4= ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=zohomail.com; s=zohoarc; t=1693392898; h=Content-Transfer-Encoding:Cc:Date:From:In-Reply-To:List-Subscribe:List-Post:List-Id:List-Archive:List-Help:List-Unsubscribe:MIME-Version:Message-ID:References:Sender:Subject:To; bh=nRvX1NGYHIzud27WVeDvn/DAgmBVbq010V/Wo0gJfOE=; b=K2hy57gu1qGVZa664Jl+R1WdSFNPIAU6uwFfiSpwMDTbzL+1jl5JDmjvF7kgQU/YHOFUzZGCkX90qmr2LBEYVBfLr4szPj18QArAQz9l2uH/j+8P8sT7UfYx9yvE3QHD33SyZgDFBiMcZsGpVK/eiO2RmuQtMsQKDmESX17emzs= ARC-Authentication-Results: i=1; mx.zohomail.com; dkim=pass header.i=@intel.com; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass header.from= (p=none dis=none) Return-Path: Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) by mx.zohomail.com with SMTPS id 1693392898293612.0836669148529; Wed, 30 Aug 2023 03:54:58 -0700 (PDT) Received: from localhost ([::1] helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1qbIpX-0004bN-Ed; Wed, 30 Aug 2023 06:53:47 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1qbIpW-0004aM-Cu for qemu-devel@nongnu.org; Wed, 30 Aug 2023 06:53:46 -0400 Received: from mgamail.intel.com ([192.55.52.93]) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1qbIpT-0007je-Ok for qemu-devel@nongnu.org; Wed, 30 Aug 2023 06:53:46 -0400 Received: from orsmga006.jf.intel.com ([10.7.209.51]) by fmsmga102.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 30 Aug 2023 03:53:39 -0700 Received: from duan-server-s2600bt.bj.intel.com ([10.240.192.147]) by orsmga006-auth.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 30 Aug 2023 03:53:29 -0700 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1693392824; x=1724928824; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=FwV54K2kWvnU/H45Up13S/nT+vitO1lLqtEnEkMrTpo=; b=hPm5AuvEY8uVFHUPGhB20eNlfzIg4FRH5AhfI0H/YEeCIp0sOuUcM4dI 7QPW0k8Onq5maDX6FSVHVmq34pmX3sEcVf1bSYPsTMrtB6wiqjhSY8S/b ItLfbhMHkYSvAAt63ogGusNFUWWl6b5K1Psc3Ck8MJ0skw1u+kI4xYNhU RN3EIV8czXPADq86NJjT6TxVqjQwlr5zGWgDtuJuZ+MkSGt49aWGDvlxE Za84fAOdQ4NaygbjupCK3gYjbBoXjPIvb8cpxEZe4nu91pAwVemY3CyrE 0SzSDc7giWDbC85bg3QKRkSM5ANcMlIQ5tQ5ug3hxnmxPIPkFgrvxERDY g==; X-IronPort-AV: E=McAfee;i="6600,9927,10817"; a="373016635" X-IronPort-AV: E=Sophos;i="6.02,213,1688454000"; d="scan'208";a="373016635" X-ExtLoop1: 1 X-IronPort-AV: E=McAfee;i="6600,9927,10817"; a="715866214" X-IronPort-AV: E=Sophos;i="6.02,213,1688454000"; d="scan'208";a="715866214" From: Zhenzhong Duan To: qemu-devel@nongnu.org Cc: alex.williamson@redhat.com, clg@redhat.com, jgg@nvidia.com, nicolinc@nvidia.com, joao.m.martins@oracle.com, eric.auger@redhat.com, peterx@redhat.com, jasowang@redhat.com, kevin.tian@intel.com, yi.l.liu@intel.com, yi.y.sun@intel.com, chao.p.peng@intel.com, Zhenzhong Duan Subject: [PATCH v1 14/22] vfio/common: Simplify vfio_viommu_preset() Date: Wed, 30 Aug 2023 18:37:46 +0800 Message-Id: <20230830103754.36461-15-zhenzhong.duan@intel.com> X-Mailer: git-send-email 2.34.1 In-Reply-To: <20230830103754.36461-1-zhenzhong.duan@intel.com> References: <20230830103754.36461-1-zhenzhong.duan@intel.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Received-SPF: pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) client-ip=209.51.188.17; envelope-from=qemu-devel-bounces+importer=patchew.org@nongnu.org; helo=lists.gnu.org; Received-SPF: pass client-ip=192.55.52.93; envelope-from=zhenzhong.duan@intel.com; helo=mgamail.intel.com X-Spam_score_int: -43 X-Spam_score: -4.4 X-Spam_bar: ---- X-Spam_report: (-4.4 / 5.0 requ) BAYES_00=-1.9, DKIMWL_WL_HIGH=-0.001, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, RCVD_IN_DNSWL_MED=-2.3, SPF_HELO_NONE=0.001, SPF_PASS=-0.001 autolearn=ham autolearn_force=no X-Spam_action: no action X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: qemu-devel-bounces+importer=patchew.org@nongnu.org Sender: qemu-devel-bounces+importer=patchew.org@nongnu.org X-ZohoMail-DKIM: pass (identity @intel.com) X-ZM-MESSAGEID: 1693392939372100015 Content-Type: text/plain; charset="utf-8" Commit "vfio/container-base: Introduce [attach/detach]_device container cal= lbacks" add support to link to address space, we can utilize it to simplify vfio_viommu_preset(). Signed-off-by: Zhenzhong Duan --- hw/vfio/common.c | 17 +---------------- 1 file changed, 1 insertion(+), 16 deletions(-) diff --git a/hw/vfio/common.c b/hw/vfio/common.c index 86b6af5740..6c3e98d5fd 100644 --- a/hw/vfio/common.c +++ b/hw/vfio/common.c @@ -165,22 +165,7 @@ void vfio_unblock_multiple_devices_migration(void) =20 bool vfio_viommu_preset(VFIODevice *vbasedev) { - VFIOAddressSpace *space; - VFIOContainer *container; - VFIODevice *tmp_dev; - - QLIST_FOREACH(space, &vfio_address_spaces, list) { - QLIST_FOREACH(container, &space->containers, next) { - tmp_dev =3D NULL; - while ((tmp_dev =3D vfio_container_dev_iter_next(container, - tmp_dev))) { - if (vbasedev =3D=3D tmp_dev) { - return space->as !=3D &address_space_memory; - } - } - } - } - g_assert_not_reached(); + return vbasedev->container->space->as !=3D &address_space_memory; } =20 static void vfio_set_migration_error(int err) --=20 2.34.1 From nobody Wed May 15 22:12:51 2024 Delivered-To: importer@patchew.org Authentication-Results: mx.zohomail.com; dkim=pass header.i=@intel.com; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass(p=none dis=none) header.from=intel.com ARC-Seal: i=1; a=rsa-sha256; t=1693392966; cv=none; d=zohomail.com; s=zohoarc; b=Kl61WMpVvs6VU8FHYQkPlgAK0sswcIfPyZY/QKMT05HttcXiAzvGr80qyeiVSyxTIXAF8XPbVtjWjXy3QxOyDgJc12it9MgmsIUiPnFX//cYBmMSXM/DinR4iimyAYJDsTM5ebWuCuHZ7YBMutaWGCWNfXQAhwEg6bj/UJkVf6s= ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=zohomail.com; s=zohoarc; t=1693392966; h=Content-Transfer-Encoding:Cc:Date:From:In-Reply-To:List-Subscribe:List-Post:List-Id:List-Archive:List-Help:List-Unsubscribe:MIME-Version:Message-ID:References:Sender:Subject:To; bh=dK3g2CTKUd097p+o+oH+Wifk3C8poqnX+KrXLhSQRc8=; b=bs7B9ije05gGATDLnTPSR7bTMc223ezW/CildkP3jJtRfCMd7ps2TYXesRDcU2msR/V/k2ym+hURS8JZXFGH2c8BOzEHDKgSU5ZJOJaBXtVHohrpFxBjnaQKCCwbw39IeYSMEB32RSiTxmsJFjLGBv3vGGBiZGHoWxTP5xRFa7Q= ARC-Authentication-Results: i=1; mx.zohomail.com; dkim=pass header.i=@intel.com; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass header.from= (p=none dis=none) Return-Path: Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) by mx.zohomail.com with SMTPS id 1693392966689702.8822922337984; Wed, 30 Aug 2023 03:56:06 -0700 (PDT) Received: from localhost ([::1] helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1qbIpa-0004uI-5v; Wed, 30 Aug 2023 06:53:50 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1qbIpY-0004ej-0W for qemu-devel@nongnu.org; Wed, 30 Aug 2023 06:53:48 -0400 Received: from mgamail.intel.com ([192.55.52.93]) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1qbIpV-0007jr-EE for qemu-devel@nongnu.org; Wed, 30 Aug 2023 06:53:47 -0400 Received: from orsmga006.jf.intel.com ([10.7.209.51]) by fmsmga102.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 30 Aug 2023 03:53:41 -0700 Received: from duan-server-s2600bt.bj.intel.com ([10.240.192.147]) by orsmga006-auth.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 30 Aug 2023 03:53:33 -0700 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1693392825; x=1724928825; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=HgO/BHF4jIS0HTKN9QhvMskHyXxCXX9sam8G+fC8ujw=; b=cYS4s2rG6Jtc5th7rJKrQDFs3tktjHZ5JQNUo2aE+2/CwbrIoN4Wxzvm WdJG1+OOGsbIYPfooTfx+V6AhenbQILADM+gMoiqfyp4FYi39Xx2ef7wj cfHgfura68VD1A9dkas5zufoXSqRaNrAuEKUS9djXT/77PN/NkfVyy2UO cGzpAHzbMjle/kGHscyZYh0YdH1BcIF1dwoRStut4bXABsr1nse1OSLbO iXYsrm+ok5rHmtSxxUmzdcnTSanGXNt/JY0ff9lXyJFL9Hupt6mlAIsAW C2LrLo4s+z8rK8aw0SAqN9vLQOs2KMGumJ3wJjuU8mzo2kwAx6KtnfttB Q==; X-IronPort-AV: E=McAfee;i="6600,9927,10817"; a="373016659" X-IronPort-AV: E=Sophos;i="6.02,213,1688454000"; d="scan'208";a="373016659" X-ExtLoop1: 1 X-IronPort-AV: E=McAfee;i="6600,9927,10817"; a="715866229" X-IronPort-AV: E=Sophos;i="6.02,213,1688454000"; d="scan'208";a="715866229" From: Zhenzhong Duan To: qemu-devel@nongnu.org Cc: alex.williamson@redhat.com, clg@redhat.com, jgg@nvidia.com, nicolinc@nvidia.com, joao.m.martins@oracle.com, eric.auger@redhat.com, peterx@redhat.com, jasowang@redhat.com, kevin.tian@intel.com, yi.l.liu@intel.com, yi.y.sun@intel.com, chao.p.peng@intel.com, Zhenzhong Duan , Paolo Bonzini , =?UTF-8?q?Marc-Andr=C3=A9=20Lureau?= , =?UTF-8?q?Daniel=20P=2E=20Berrang=C3=A9?= , Thomas Huth , =?UTF-8?q?Philippe=20Mathieu-Daud=C3=A9?= Subject: [PATCH v1 15/22] Add iommufd configure option Date: Wed, 30 Aug 2023 18:37:47 +0800 Message-Id: <20230830103754.36461-16-zhenzhong.duan@intel.com> X-Mailer: git-send-email 2.34.1 In-Reply-To: <20230830103754.36461-1-zhenzhong.duan@intel.com> References: <20230830103754.36461-1-zhenzhong.duan@intel.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Received-SPF: pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) client-ip=209.51.188.17; envelope-from=qemu-devel-bounces+importer=patchew.org@nongnu.org; helo=lists.gnu.org; Received-SPF: pass client-ip=192.55.52.93; envelope-from=zhenzhong.duan@intel.com; helo=mgamail.intel.com X-Spam_score_int: -43 X-Spam_score: -4.4 X-Spam_bar: ---- X-Spam_report: (-4.4 / 5.0 requ) BAYES_00=-1.9, DKIMWL_WL_HIGH=-0.001, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, RCVD_IN_DNSWL_MED=-2.3, SPF_HELO_NONE=0.001, SPF_PASS=-0.001 autolearn=ham autolearn_force=no X-Spam_action: no action X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: qemu-devel-bounces+importer=patchew.org@nongnu.org Sender: qemu-devel-bounces+importer=patchew.org@nongnu.org X-ZohoMail-DKIM: pass (identity @intel.com) X-ZM-MESSAGEID: 1693392968816100003 Content-Type: text/plain; charset="utf-8" This adds "--enable-iommufd/--disable-iommufd" to enable or disable iommufd support, enabled by default. Signed-off-by: Zhenzhong Duan --- meson.build | 6 ++++++ meson_options.txt | 2 ++ scripts/meson-buildoptions.sh | 3 +++ 3 files changed, 11 insertions(+) diff --git a/meson.build b/meson.build index 98e68ef0b1..6526d8cc9b 100644 --- a/meson.build +++ b/meson.build @@ -574,6 +574,10 @@ have_tpm =3D get_option('tpm') \ .require(targetos !=3D 'windows', error_message: 'TPM emulation only ava= ilable on POSIX systems') \ .allowed() =20 +have_iommufd =3D get_option('iommufd') \ + .require(targetos =3D=3D 'linux', error_message: 'iommufd is supported o= nly on Linux') \ + .allowed() + # vhost have_vhost_user =3D get_option('vhost_user') \ .disable_auto_if(targetos !=3D 'linux') \ @@ -2129,6 +2133,7 @@ endif config_host_data.set('CONFIG_SNAPPY', snappy.found()) config_host_data.set('CONFIG_TPM', have_tpm) config_host_data.set('CONFIG_TSAN', get_option('tsan')) +config_host_data.set('CONFIG_IOMMUFD', have_iommufd) config_host_data.set('CONFIG_USB_LIBUSB', libusb.found()) config_host_data.set('CONFIG_VDE', vde.found()) config_host_data.set('CONFIG_VHOST_NET', have_vhost_net) @@ -4051,6 +4056,7 @@ summary_info +=3D {'vhost-user-crypto support': have_= vhost_user_crypto} summary_info +=3D {'vhost-user-blk server support': have_vhost_user_blk_se= rver} summary_info +=3D {'vhost-vdpa support': have_vhost_vdpa} summary_info +=3D {'build guest agent': have_ga} +summary_info +=3D {'iommufd support': have_iommufd} summary(summary_info, bool_yn: true, section: 'Configurable features') =20 # Compilation information diff --git a/meson_options.txt b/meson_options.txt index aaea5ddd77..aed91d173b 100644 --- a/meson_options.txt +++ b/meson_options.txt @@ -105,6 +105,8 @@ option('dbus_display', type: 'feature', value: 'auto', description: '-display dbus support') option('tpm', type : 'feature', value : 'auto', description: 'TPM support') +option('iommufd', type : 'feature', value : 'auto', + description: 'iommufd support') =20 # Do not enable it by default even for Mingw32, because it doesn't # work on Wine. diff --git a/scripts/meson-buildoptions.sh b/scripts/meson-buildoptions.sh index 9da3fe299b..719401ffb0 100644 --- a/scripts/meson-buildoptions.sh +++ b/scripts/meson-buildoptions.sh @@ -113,6 +113,7 @@ meson_options_help() { printf "%s\n" ' hax HAX acceleration support' printf "%s\n" ' hvf HVF acceleration support' printf "%s\n" ' iconv Font glyph conversion support' + printf "%s\n" ' iommufd iommufd support' printf "%s\n" ' jack JACK sound support' printf "%s\n" ' keyring Linux keyring support' printf "%s\n" ' kvm KVM acceleration support' @@ -325,6 +326,8 @@ _meson_option_parse() { --enable-install-blobs) printf "%s" -Dinstall_blobs=3Dtrue ;; --disable-install-blobs) printf "%s" -Dinstall_blobs=3Dfalse ;; --interp-prefix=3D*) quote_sh "-Dinterp_prefix=3D$2" ;; + --enable-iommufd) printf "%s" -Diommufd=3Denabled ;; + --disable-iommufd) printf "%s" -Diommufd=3Ddisabled ;; --enable-jack) printf "%s" -Djack=3Denabled ;; --disable-jack) printf "%s" -Djack=3Ddisabled ;; --enable-keyring) printf "%s" -Dkeyring=3Denabled ;; --=20 2.34.1 From nobody Wed May 15 22:12:51 2024 Delivered-To: importer@patchew.org Authentication-Results: mx.zohomail.com; dkim=pass header.i=@intel.com; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass(p=none dis=none) header.from=intel.com ARC-Seal: i=1; a=rsa-sha256; t=1693392885; cv=none; d=zohomail.com; s=zohoarc; b=CaZ1PYUXox2zpOvc3BOqwQ7sjIVh1IRnSilaCtdsyVTXDSesIktM9zNLbPXvSXsA2jWXYOIPt0d07SDkpOeMtnQ8ysYQd6xfpS5H+Ha6CwjRwShWCey6xXjJmoiTb74Jyr+XGPmtpWTEy8QrbQJIESAznhsNGqtFGA9xuXAurpM= ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=zohomail.com; s=zohoarc; t=1693392885; h=Content-Transfer-Encoding:Cc:Date:From:In-Reply-To:List-Subscribe:List-Post:List-Id:List-Archive:List-Help:List-Unsubscribe:MIME-Version:Message-ID:References:Sender:Subject:To; bh=/hYhmlipOnprNRehMohnmHk7dBccrPp796UjKiIc2ng=; b=OylGL5j+CN9UbTCXz2fgaUoyAITs5lEHE4WdT091yiFdmwEnmgSlQ9jSuT1bL4dylr7tMGLl42DiTMsSs4ClVjGi8TtWLP6FRAxbGgnm+VGduh87SVxZwIwo2V7aPS9kDq3/sJJ6ViWO5OOlmSCRaY/cyllqNkE/MY/zMHX5y5k= ARC-Authentication-Results: i=1; mx.zohomail.com; dkim=pass header.i=@intel.com; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass header.from= (p=none dis=none) Return-Path: Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) by mx.zohomail.com with SMTPS id 1693392885704514.5657763923459; Wed, 30 Aug 2023 03:54:45 -0700 (PDT) Received: from localhost ([::1] helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1qbIpm-0005qF-QK; Wed, 30 Aug 2023 06:54:02 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1qbIpl-0005oi-PN for qemu-devel@nongnu.org; Wed, 30 Aug 2023 06:54:01 -0400 Received: from mgamail.intel.com ([192.55.52.93]) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1qbIpi-0007jr-D1 for qemu-devel@nongnu.org; Wed, 30 Aug 2023 06:54:01 -0400 Received: from orsmga006.jf.intel.com ([10.7.209.51]) by fmsmga102.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 30 Aug 2023 03:53:44 -0700 Received: from duan-server-s2600bt.bj.intel.com ([10.240.192.147]) by orsmga006-auth.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 30 Aug 2023 03:53:39 -0700 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1693392838; x=1724928838; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=9LsIWNjXtA+GcRuyg+Ee7v8B3cKrH4OKtIZXbtf/DXo=; b=BMN5gV/A5in41V5C9dpYmq9oso9tarhTKl6/Cea8bC16oVaGgzecmwOp UHrQUU1PlzL+GouSlPldG52Z8LM5wBud4kJX4UyJMWJEL21HD6mafaT8n Z/b1wUSyhmOypTVhfcTxdDnjO36Wa8pjH9b6S99wIsboHMhkQQnaNoVp9 ZLwt+ET9TgxDAKQ4Z/hzl1hP89tIRdBUeTUQCnr/xbeuSTuAsN7f+hao1 Hbl0R2q2/WNWtKYN/LCn9BmPK9/h0TsWaa2JGZuTHuyvXzIa9NCEBv5KV piu1XFWAaNcSFuMi4oWEKV3xDCDm1Oe3H/fWrL2nJhrB1O1saGSCnfDKL Q==; X-IronPort-AV: E=McAfee;i="6600,9927,10817"; a="373016665" X-IronPort-AV: E=Sophos;i="6.02,213,1688454000"; d="scan'208";a="373016665" X-ExtLoop1: 1 X-IronPort-AV: E=McAfee;i="6600,9927,10817"; a="715866252" X-IronPort-AV: E=Sophos;i="6.02,213,1688454000"; d="scan'208";a="715866252" From: Zhenzhong Duan To: qemu-devel@nongnu.org Cc: alex.williamson@redhat.com, clg@redhat.com, jgg@nvidia.com, nicolinc@nvidia.com, joao.m.martins@oracle.com, eric.auger@redhat.com, peterx@redhat.com, jasowang@redhat.com, kevin.tian@intel.com, yi.l.liu@intel.com, yi.y.sun@intel.com, chao.p.peng@intel.com, Zhenzhong Duan , Paolo Bonzini , Eric Blake , Markus Armbruster , =?UTF-8?q?Daniel=20P=2E=20Berrang=C3=A9?= , Eduardo Habkost Subject: [PATCH v1 16/22] backends/iommufd: Introduce the iommufd object Date: Wed, 30 Aug 2023 18:37:48 +0800 Message-Id: <20230830103754.36461-17-zhenzhong.duan@intel.com> X-Mailer: git-send-email 2.34.1 In-Reply-To: <20230830103754.36461-1-zhenzhong.duan@intel.com> References: <20230830103754.36461-1-zhenzhong.duan@intel.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Received-SPF: pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) client-ip=209.51.188.17; envelope-from=qemu-devel-bounces+importer=patchew.org@nongnu.org; helo=lists.gnu.org; Received-SPF: pass client-ip=192.55.52.93; envelope-from=zhenzhong.duan@intel.com; helo=mgamail.intel.com X-Spam_score_int: -43 X-Spam_score: -4.4 X-Spam_bar: ---- X-Spam_report: (-4.4 / 5.0 requ) BAYES_00=-1.9, DKIMWL_WL_HIGH=-0.001, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, RCVD_IN_DNSWL_MED=-2.3, SPF_HELO_NONE=0.001, SPF_PASS=-0.001 autolearn=ham autolearn_force=no X-Spam_action: no action X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: qemu-devel-bounces+importer=patchew.org@nongnu.org Sender: qemu-devel-bounces+importer=patchew.org@nongnu.org X-ZohoMail-DKIM: pass (identity @intel.com) X-ZM-MESSAGEID: 1693392938988100012 Content-Type: text/plain; charset="utf-8" From: Eric Auger Introduce an iommufd object which allows the interaction with the host /dev/iommu device. The /dev/iommu can have been already pre-opened outside of qemu, in which case the fd can be passed directly along with the iommufd object: This allows the iommufd object to be shared accross several subsystems (VFIO, VDPA, ...). For example, libvirt would open the /dev/iommu once. If no fd is passed along with the iommufd object, the /dev/iommu is opened by the qemu code. The CONFIG_IOMMUFD option must be set to compile this new object. Suggested-by: Alex Williamson Signed-off-by: Eric Auger Signed-off-by: Yi Liu Signed-off-by: Zhenzhong Duan --- MAINTAINERS | 7 + backends/Kconfig | 4 + backends/iommufd.c | 291 +++++++++++++++++++++++++++++++++++++++ backends/meson.build | 3 + backends/trace-events | 13 ++ include/sysemu/iommufd.h | 49 +++++++ qapi/qom.json | 18 ++- qemu-options.hx | 13 ++ 8 files changed, 397 insertions(+), 1 deletion(-) create mode 100644 backends/iommufd.c create mode 100644 include/sysemu/iommufd.h diff --git a/MAINTAINERS b/MAINTAINERS index 6111b6b4d9..04663fbb6f 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -2079,6 +2079,13 @@ F: hw/vfio/ap.c F: docs/system/s390x/vfio-ap.rst L: qemu-s390x@nongnu.org =20 +iommufd +M: Yi Liu +M: Eric Auger +S: Supported +F: backends/iommufd.c +F: include/sysemu/iommufd.h + vhost M: Michael S. Tsirkin S: Supported diff --git a/backends/Kconfig b/backends/Kconfig index f35abc1609..2cb23f62fa 100644 --- a/backends/Kconfig +++ b/backends/Kconfig @@ -1 +1,5 @@ source tpm/Kconfig + +config IOMMUFD + bool + depends on VFIO diff --git a/backends/iommufd.c b/backends/iommufd.c new file mode 100644 index 0000000000..07ea434424 --- /dev/null +++ b/backends/iommufd.c @@ -0,0 +1,291 @@ +/* + * iommufd container backend + * + * Copyright (C) 2023 Intel Corporation. + * Copyright Red Hat, Inc. 2023 + * + * Authors: Yi Liu + * Eric Auger + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + + * You should have received a copy of the GNU General Public License along + * with this program; if not, see . + */ + +#include "qemu/osdep.h" +#include "sysemu/iommufd.h" +#include "qapi/error.h" +#include "qapi/qmp/qerror.h" +#include "qemu/module.h" +#include "qom/object_interfaces.h" +#include "qemu/error-report.h" +#include "monitor/monitor.h" +#include "trace.h" +#include +#include + +static void iommufd_backend_init(Object *obj) +{ + IOMMUFDBackend *be =3D IOMMUFD_BACKEND(obj); + + be->fd =3D -1; + be->users =3D 0; + be->owned =3D true; + qemu_mutex_init(&be->lock); +} + +static void iommufd_backend_finalize(Object *obj) +{ + IOMMUFDBackend *be =3D IOMMUFD_BACKEND(obj); + + if (be->owned) { + close(be->fd); + be->fd =3D -1; + } +} + +static void iommufd_backend_set_fd(Object *obj, const char *str, Error **e= rrp) +{ + IOMMUFDBackend *be =3D IOMMUFD_BACKEND(obj); + int fd =3D -1; + + fd =3D monitor_fd_param(monitor_cur(), str, errp); + if (fd =3D=3D -1) { + error_prepend(errp, "Could not parse remote object fd %s:", str); + return; + } + qemu_mutex_lock(&be->lock); + be->fd =3D fd; + be->owned =3D false; + qemu_mutex_unlock(&be->lock); + trace_iommu_backend_set_fd(be->fd); +} + +static void iommufd_backend_class_init(ObjectClass *oc, void *data) +{ + object_class_property_add_str(oc, "fd", NULL, iommufd_backend_set_fd); +} + +int iommufd_backend_connect(IOMMUFDBackend *be, Error **errp) +{ + int fd, ret =3D 0; + + qemu_mutex_lock(&be->lock); + if (be->users =3D=3D UINT32_MAX) { + error_setg(errp, "too many connections"); + ret =3D -E2BIG; + goto out; + } + if (be->owned && !be->users) { + fd =3D qemu_open_old("/dev/iommu", O_RDWR); + if (fd < 0) { + error_setg_errno(errp, errno, "/dev/iommu opening failed"); + ret =3D fd; + goto out; + } + be->fd =3D fd; + } + be->users++; +out: + trace_iommufd_backend_connect(be->fd, be->owned, + be->users, ret); + qemu_mutex_unlock(&be->lock); + return ret; +} + +void iommufd_backend_disconnect(IOMMUFDBackend *be) +{ + qemu_mutex_lock(&be->lock); + if (!be->users) { + goto out; + } + be->users--; + if (!be->users && be->owned) { + close(be->fd); + be->fd =3D -1; + } +out: + trace_iommufd_backend_disconnect(be->fd, be->users); + qemu_mutex_unlock(&be->lock); +} + +static int iommufd_backend_alloc_ioas(int fd, uint32_t *ioas) +{ + int ret; + struct iommu_ioas_alloc alloc_data =3D { + .size =3D sizeof(alloc_data), + .flags =3D 0, + }; + + ret =3D ioctl(fd, IOMMU_IOAS_ALLOC, &alloc_data); + if (ret) { + error_report("Failed to allocate ioas %m"); + } + + *ioas =3D alloc_data.out_ioas_id; + trace_iommufd_backend_alloc_ioas(fd, *ioas, ret); + + return ret; +} + +void iommufd_backend_free_id(int fd, uint32_t id) +{ + int ret; + struct iommu_destroy des =3D { + .size =3D sizeof(des), + .id =3D id, + }; + + ret =3D ioctl(fd, IOMMU_DESTROY, &des); + trace_iommufd_backend_free_id(fd, id, ret); + if (ret) { + error_report("Failed to free id: %u %m", id); + } +} + +int iommufd_backend_get_ioas(IOMMUFDBackend *be, uint32_t *ioas_id) +{ + int ret; + + ret =3D iommufd_backend_alloc_ioas(be->fd, ioas_id); + trace_iommufd_backend_get_ioas(be->fd, *ioas_id, ret); + return ret; +} + +void iommufd_backend_put_ioas(IOMMUFDBackend *be, uint32_t ioas) +{ + trace_iommufd_backend_put_ioas(be->fd, ioas); + iommufd_backend_free_id(be->fd, ioas); +} + +int iommufd_backend_unmap_dma(IOMMUFDBackend *be, uint32_t ioas, + hwaddr iova, ram_addr_t size) +{ + int ret; + struct iommu_ioas_unmap unmap =3D { + .size =3D sizeof(unmap), + .ioas_id =3D ioas, + .iova =3D iova, + .length =3D size, + }; + + ret =3D ioctl(be->fd, IOMMU_IOAS_UNMAP, &unmap); + trace_iommufd_backend_unmap_dma(be->fd, ioas, iova, size, ret); + if (ret && errno =3D=3D ENOENT) { + ret =3D 0; + } + if (ret) { + error_report("IOMMU_IOAS_UNMAP failed: %s", strerror(errno)); + } + return !ret ? 0 : -errno; +} + +int iommufd_backend_map_dma(IOMMUFDBackend *be, uint32_t ioas, hwaddr iova, + ram_addr_t size, void *vaddr, bool readonly) +{ + int ret; + struct iommu_ioas_map map =3D { + .size =3D sizeof(map), + .flags =3D IOMMU_IOAS_MAP_READABLE | + IOMMU_IOAS_MAP_FIXED_IOVA, + .ioas_id =3D ioas, + .__reserved =3D 0, + .user_va =3D (int64_t)vaddr, + .iova =3D iova, + .length =3D size, + }; + + if (!readonly) { + map.flags |=3D IOMMU_IOAS_MAP_WRITEABLE; + } + + ret =3D ioctl(be->fd, IOMMU_IOAS_MAP, &map); + trace_iommufd_backend_map_dma(be->fd, ioas, iova, size, + vaddr, readonly, ret); + if (ret) { + error_report("IOMMU_IOAS_MAP failed: %s", strerror(errno)); + } + return !ret ? 0 : -errno; +} + +int iommufd_backend_copy_dma(IOMMUFDBackend *be, uint32_t src_ioas, + uint32_t dst_ioas, hwaddr iova, + ram_addr_t size, bool readonly) +{ + int ret; + struct iommu_ioas_copy copy =3D { + .size =3D sizeof(copy), + .flags =3D IOMMU_IOAS_MAP_READABLE | + IOMMU_IOAS_MAP_FIXED_IOVA, + .dst_ioas_id =3D dst_ioas, + .src_ioas_id =3D src_ioas, + .length =3D size, + .dst_iova =3D iova, + .src_iova =3D iova, + }; + + if (!readonly) { + copy.flags |=3D IOMMU_IOAS_MAP_WRITEABLE; + } + + ret =3D ioctl(be->fd, IOMMU_IOAS_COPY, ©); + trace_iommufd_backend_copy_dma(be->fd, src_ioas, dst_ioas, + iova, size, readonly, ret); + if (ret) { + error_report("IOMMU_IOAS_COPY failed: %s", strerror(errno)); + } + return !ret ? 0 : -errno; +} + +int iommufd_backend_alloc_hwpt(int iommufd, uint32_t dev_id, + uint32_t pt_id, uint32_t *out_hwpt) +{ + int ret; + struct iommu_hwpt_alloc alloc_hwpt =3D { + .size =3D sizeof(struct iommu_hwpt_alloc), + .flags =3D 0, + .dev_id =3D dev_id, + .pt_id =3D pt_id, + .__reserved =3D 0, + }; + + ret =3D ioctl(iommufd, IOMMU_HWPT_ALLOC, &alloc_hwpt); + trace_iommufd_backend_alloc_hwpt(iommufd, dev_id, pt_id, ret); + + if (ret) { + error_report("IOMMU_HWPT_ALLOC failed: %s", strerror(errno)); + } else { + *out_hwpt =3D alloc_hwpt.out_hwpt_id; + } + return !ret ? 0 : -errno; +} + +static const TypeInfo iommufd_backend_info =3D { + .name =3D TYPE_IOMMUFD_BACKEND, + .parent =3D TYPE_OBJECT, + .instance_size =3D sizeof(IOMMUFDBackend), + .instance_init =3D iommufd_backend_init, + .instance_finalize =3D iommufd_backend_finalize, + .class_size =3D sizeof(IOMMUFDBackendClass), + .class_init =3D iommufd_backend_class_init, + .interfaces =3D (InterfaceInfo[]) { + { TYPE_USER_CREATABLE }, + { } + } +}; + +static void register_types(void) +{ + type_register_static(&iommufd_backend_info); +} + +type_init(register_types); diff --git a/backends/meson.build b/backends/meson.build index 914c7c4afb..29dc147c8e 100644 --- a/backends/meson.build +++ b/backends/meson.build @@ -20,6 +20,9 @@ if have_vhost_user system_ss.add(when: 'CONFIG_VIRTIO', if_true: files('vhost-user.c')) endif system_ss.add(when: 'CONFIG_VIRTIO_CRYPTO', if_true: files('cryptodev-vhos= t.c')) +if have_iommufd + system_ss.add(files('iommufd.c')) +endif if have_vhost_user_crypto system_ss.add(when: 'CONFIG_VIRTIO_CRYPTO', if_true: files('cryptodev-vh= ost-user.c')) endif diff --git a/backends/trace-events b/backends/trace-events index 652eb76a57..093e3eb1da 100644 --- a/backends/trace-events +++ b/backends/trace-events @@ -5,3 +5,16 @@ dbus_vmstate_pre_save(void) dbus_vmstate_post_load(int version_id) "version_id: %d" dbus_vmstate_loading(const char *id) "id: %s" dbus_vmstate_saving(const char *id) "id: %s" + +# iommufd.c +iommufd_backend_connect(int fd, bool owned, uint32_t users, int ret) "fd= =3D%d owned=3D%d users=3D%d (%d)" +iommufd_backend_disconnect(int fd, uint32_t users) "fd=3D%d users=3D%d" +iommu_backend_set_fd(int fd) "pre-opened /dev/iommu fd=3D%d" +iommufd_backend_get_ioas(int iommufd, uint32_t ioas, int ret) " iommufd=3D= %d ioas=3D%d (%d)" +iommufd_backend_put_ioas(int iommufd, uint32_t ioas) " iommufd=3D%d ioas= =3D%d" +iommufd_backend_unmap_dma(int iommufd, uint32_t ioas, uint64_t iova, uint6= 4_t size, int ret) " iommufd=3D%d ioas=3D%d iova=3D0x%"PRIx64" size=3D0x%"P= RIx64" (%d)" +iommufd_backend_map_dma(int iommufd, uint32_t ioas, uint64_t iova, uint64_= t size, void *vaddr, bool readonly, int ret) " iommufd=3D%d ioas=3D%d iova= =3D0x%"PRIx64" size=3D0x%"PRIx64" addr=3D%p readonly=3D%d (%d)" +iommufd_backend_copy_dma(int iommufd, uint32_t src_ioas, uint32_t dst_ioas= , uint64_t iova, uint64_t size, bool readonly, int ret) " iommufd=3D%d src_= ioas=3D%d dst_ioas=3D%d iova=3D0x%"PRIx64" size=3D0x%"PRIx64" readonly=3D%d= (%d)" +iommufd_backend_alloc_ioas(int iommufd, uint32_t ioas, int ret) " iommufd= =3D%d ioas=3D%d (%d)" +iommufd_backend_free_id(int iommufd, uint32_t id, int ret) " iommufd=3D%d = id=3D%d (%d)" +iommufd_backend_alloc_hwpt(int iommufd, uint32_t dev_id, uint32_t pt_id, i= nt ret) " iommufd=3D%d dev_id=3D%u pt_id=3D%u (%d)" diff --git a/include/sysemu/iommufd.h b/include/sysemu/iommufd.h new file mode 100644 index 0000000000..f3bd212170 --- /dev/null +++ b/include/sysemu/iommufd.h @@ -0,0 +1,49 @@ +#ifndef SYSEMU_IOMMUFD_H +#define SYSEMU_IOMMUFD_H + +#include "qom/object.h" +#include "qemu/thread.h" +#include "exec/hwaddr.h" +#include "exec/cpu-common.h" + +#define TYPE_IOMMUFD_BACKEND "iommufd" +OBJECT_DECLARE_TYPE(IOMMUFDBackend, IOMMUFDBackendClass, + IOMMUFD_BACKEND) +#define IOMMUFD_BACKEND(obj) \ + OBJECT_CHECK(IOMMUFDBackend, (obj), TYPE_IOMMUFD_BACKEND) +#define IOMMUFD_BACKEND_GET_CLASS(obj) \ + OBJECT_GET_CLASS(IOMMUFDBackendClass, (obj), TYPE_IOMMUFD_BACKEND) +#define IOMMUFD_BACKEND_CLASS(klass) \ + OBJECT_CLASS_CHECK(IOMMUFDBackendClass, (klass), TYPE_IOMMUFD_BACKEND) +struct IOMMUFDBackendClass { + ObjectClass parent_class; +}; + +struct IOMMUFDBackend { + Object parent; + + /*< protected >*/ + int fd; /* /dev/iommu file descriptor */ + bool owned; /* is the /dev/iommu opened internally */ + QemuMutex lock; + uint32_t users; + + /*< public >*/ +}; + +int iommufd_backend_connect(IOMMUFDBackend *be, Error **errp); +void iommufd_backend_disconnect(IOMMUFDBackend *be); + +int iommufd_backend_get_ioas(IOMMUFDBackend *be, uint32_t *ioas_id); +void iommufd_backend_put_ioas(IOMMUFDBackend *be, uint32_t ioas_id); +void iommufd_backend_free_id(int fd, uint32_t id); +int iommufd_backend_unmap_dma(IOMMUFDBackend *be, uint32_t ioas, + hwaddr iova, ram_addr_t size); +int iommufd_backend_map_dma(IOMMUFDBackend *be, uint32_t ioas, hwaddr iova, + ram_addr_t size, void *vaddr, bool readonly); +int iommufd_backend_copy_dma(IOMMUFDBackend *be, uint32_t src_ioas, + uint32_t dst_ioas, hwaddr iova, + ram_addr_t size, bool readonly); +int iommufd_backend_alloc_hwpt(int iommufd, uint32_t dev_id, + uint32_t pt_id, uint32_t *out_hwpt); +#endif diff --git a/qapi/qom.json b/qapi/qom.json index fa3e88c8e6..2646ac4cca 100644 --- a/qapi/qom.json +++ b/qapi/qom.json @@ -779,6 +779,18 @@ { 'struct': 'VfioUserServerProperties', 'data': { 'socket': 'SocketAddress', 'device': 'str' } } =20 +## +# @IOMMUFDProperties: +# +# Properties for IOMMUFDbackend objects. +# +# fd: file descriptor name +# +# Since: 7.2 +## +{ 'struct': 'IOMMUFDProperties', + 'data': { '*fd': 'str' } } + ## # @RngProperties: # @@ -933,6 +945,8 @@ 'qtest', 'rng-builtin', 'rng-egd', + { 'name': 'iommufd', + 'if': 'CONFIG_IOMMUFD' }, { 'name': 'rng-random', 'if': 'CONFIG_POSIX' }, 'secret', @@ -1014,7 +1028,9 @@ 'tls-creds-x509': 'TlsCredsX509Properties', 'tls-cipher-suites': 'TlsCredsProperties', 'x-remote-object': 'RemoteObjectProperties', - 'x-vfio-user-server': 'VfioUserServerProperties' + 'x-vfio-user-server': 'VfioUserServerProperties', + 'iommufd': { 'type': 'IOMMUFDProperties', + 'if': 'CONFIG_IOMMUFD' } } } =20 ## diff --git a/qemu-options.hx b/qemu-options.hx index 29b98c3d4c..827dd085ee 100644 --- a/qemu-options.hx +++ b/qemu-options.hx @@ -5098,6 +5098,19 @@ SRST =20 The ``share`` boolean option is on by default with memfd. =20 +#ifdef CONFIG_IOMMUFD + ``-object iommufd,id=3Did[,fd=3Dfd]`` + Creates an iommufd backend which allows control of DMA mapping + through the /dev/iommu device. + + The ``id`` parameter is a unique ID which frontends (such as + vfio-pci of vdpa) will use to connect withe the iommufd backend. + + The ``fd`` parameter is an optional pre-opened file descriptor + resulting from /dev/iommu opening. Usually the iommufd is shared + accross all subsystems, bringing the benefit of centralized + reference counting. +#endif ``-object rng-builtin,id=3Did`` Creates a random number generator backend which obtains entropy from QEMU builtin functions. The ``id`` parameter is a unique ID --=20 2.34.1 From nobody Wed May 15 22:12:51 2024 Delivered-To: importer@patchew.org Authentication-Results: mx.zohomail.com; dkim=pass header.i=@intel.com; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass(p=none dis=none) header.from=intel.com ARC-Seal: i=1; a=rsa-sha256; t=1693393010; cv=none; d=zohomail.com; s=zohoarc; b=PUkCtg/tHBTNWpEJ8EY5YJMuGC/XIZ1UXBPTSjo8+Dk6QAXBOB2TBTvz2svoo5sVwki0WA13wsyLSRoBR0jVxkLJofkqg6Oy4blaoz2GxJ1AYV33IxbcMPsBY2zDFu3r8EVeXhOTNggNAavWOTP98MdOJULOERv66HxsTLBxGmg= ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=zohomail.com; s=zohoarc; t=1693393010; h=Content-Transfer-Encoding:Cc:Date:From:In-Reply-To:List-Subscribe:List-Post:List-Id:List-Archive:List-Help:List-Unsubscribe:MIME-Version:Message-ID:References:Sender:Subject:To; bh=zkLwmDpHRhnvPFgRPIJYrvXmePL/HcW9dTxI5kZNxKo=; b=elpfU4ikhyOHRkrox7Yuvi68FpOcczN/fspyaYYzZ6sfBLqGjsFTMHYJ0iAlE8wupwdVHL023CfcY51ZcvLrhvzpWis4kF+OD0XQE3NyGX5vz/9jRAM9oeKGYowlPbksYL/Tm6ix8FF6ZiW/WzVvuNWPvf6QKcXZ9TZE8xxy5iM= ARC-Authentication-Results: i=1; mx.zohomail.com; dkim=pass header.i=@intel.com; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass header.from= (p=none dis=none) Return-Path: Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) by mx.zohomail.com with SMTPS id 1693393010906786.0217796928699; Wed, 30 Aug 2023 03:56:50 -0700 (PDT) Received: from localhost ([::1] helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1qbIpt-00063a-Bt; Wed, 30 Aug 2023 06:54:09 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1qbIps-000629-JM for qemu-devel@nongnu.org; Wed, 30 Aug 2023 06:54:08 -0400 Received: from mgamail.intel.com ([192.55.52.93]) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1qbIpp-0007jy-Rd for qemu-devel@nongnu.org; Wed, 30 Aug 2023 06:54:08 -0400 Received: from orsmga006.jf.intel.com ([10.7.209.51]) by fmsmga102.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 30 Aug 2023 03:53:48 -0700 Received: from duan-server-s2600bt.bj.intel.com ([10.240.192.147]) by orsmga006-auth.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 30 Aug 2023 03:53:44 -0700 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1693392845; x=1724928845; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=W3wp6+6pVnRTM7wP/U9b4HQwamgiP4qP/FnE9wLdssg=; b=efEcxBbo+6zB5t2t+mBBINSim1r8t9z58+FWsTgF6rsgO9EzEbub6Lra MoAn1/OcJeGWW5z6S6dbCfgWMnRlZlFfD7XhFex+F6Sm0mEi5tab9ZcoF 3zsYRVkGOvjH1z6D7700pQPNdmkDawdDARe9hJ0viCrSiJ3LUzq8SzsEJ hhSVP3MTB36vsLWoxj7ONlo61KL/p4ihptLR6xzY+NVr7y2ZTU0iTRF3W mITVdMDr9za1nWg1MxYYFWgFSrVXE6mmAFNOZutO/tDLcaUDRBZDLUVfQ hhVgoRS/G3gmE43tA0vcBg62GcMwuZjj6PUOIvGlOopox4Nj2m7/aLpP0 g==; X-IronPort-AV: E=McAfee;i="6600,9927,10817"; a="373016680" X-IronPort-AV: E=Sophos;i="6.02,213,1688454000"; d="scan'208";a="373016680" X-ExtLoop1: 1 X-IronPort-AV: E=McAfee;i="6600,9927,10817"; a="715866261" X-IronPort-AV: E=Sophos;i="6.02,213,1688454000"; d="scan'208";a="715866261" From: Zhenzhong Duan To: qemu-devel@nongnu.org Cc: alex.williamson@redhat.com, clg@redhat.com, jgg@nvidia.com, nicolinc@nvidia.com, joao.m.martins@oracle.com, eric.auger@redhat.com, peterx@redhat.com, jasowang@redhat.com, kevin.tian@intel.com, yi.l.liu@intel.com, yi.y.sun@intel.com, chao.p.peng@intel.com, Zhenzhong Duan Subject: [PATCH v1 17/22] util/char_dev: Add open_cdev() Date: Wed, 30 Aug 2023 18:37:49 +0800 Message-Id: <20230830103754.36461-18-zhenzhong.duan@intel.com> X-Mailer: git-send-email 2.34.1 In-Reply-To: <20230830103754.36461-1-zhenzhong.duan@intel.com> References: <20230830103754.36461-1-zhenzhong.duan@intel.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Received-SPF: pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) client-ip=209.51.188.17; envelope-from=qemu-devel-bounces+importer=patchew.org@nongnu.org; helo=lists.gnu.org; Received-SPF: pass client-ip=192.55.52.93; envelope-from=zhenzhong.duan@intel.com; helo=mgamail.intel.com X-Spam_score_int: -43 X-Spam_score: -4.4 X-Spam_bar: ---- X-Spam_report: (-4.4 / 5.0 requ) BAYES_00=-1.9, DKIMWL_WL_HIGH=-0.001, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, RCVD_IN_DNSWL_MED=-2.3, SPF_HELO_NONE=0.001, SPF_PASS=-0.001 autolearn=ham autolearn_force=no X-Spam_action: no action X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: qemu-devel-bounces+importer=patchew.org@nongnu.org Sender: qemu-devel-bounces+importer=patchew.org@nongnu.org X-ZohoMail-DKIM: pass (identity @intel.com) X-ZM-MESSAGEID: 1693393012726100001 Content-Type: text/plain; charset="utf-8" From: Yi Liu /dev/vfio/devices/vfioX may not exist. In that case it is still possible to open /dev/char/$major:$minor instead. Add helper function to abstract the cdev open. Suggested-by: Jason Gunthorpe Signed-off-by: Yi Liu Signed-off-by: Zhenzhong Duan --- MAINTAINERS | 6 ++++ include/qemu/char_dev.h | 16 +++++++++++ util/chardev_open.c | 61 +++++++++++++++++++++++++++++++++++++++++ util/meson.build | 1 + 4 files changed, 84 insertions(+) create mode 100644 include/qemu/char_dev.h create mode 100644 util/chardev_open.c diff --git a/MAINTAINERS b/MAINTAINERS index 04663fbb6f..74d18593fe 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -3372,6 +3372,12 @@ S: Maintained F: include/qemu/iova-tree.h F: util/iova-tree.c =20 +cdev Open +M: Yi Liu +S: Maintained +F: include/qemu/char_dev.h +F: util/chardev_open.c + elf2dmp M: Viktor Prutyanov S: Maintained diff --git a/include/qemu/char_dev.h b/include/qemu/char_dev.h new file mode 100644 index 0000000000..6580d351c6 --- /dev/null +++ b/include/qemu/char_dev.h @@ -0,0 +1,16 @@ +/* + * QEMU Chardev Helper + * + * Copyright (C) 2023 Intel Corporation. + * + * Authors: Yi Liu + * + * This work is licensed under the terms of the GNU GPL, version 2. See + * the COPYING file in the top-level directory. + */ + +#ifndef QEMU_CHARDEV_HELPERS_H +#define QEMU_CHARDEV_HELPERS_H + +int open_cdev(const char *devpath, dev_t cdev); +#endif diff --git a/util/chardev_open.c b/util/chardev_open.c new file mode 100644 index 0000000000..d03e415131 --- /dev/null +++ b/util/chardev_open.c @@ -0,0 +1,61 @@ +/* + * Copyright (C) 2023 Intel Corporation. + * Copyright (c) 2019, Mellanox Technologies. All rights reserved. + * + * Authors: Yi Liu + * + * This work is licensed under the terms of the GNU GPL, version 2. See + * the COPYING file in the top-level directory. + * + * Copied from + * https://github.com/linux-rdma/rdma-core/blob/master/util/open_cdev.c + * + */ +#ifndef _GNU_SOURCE +#define _GNU_SOURCE +#endif +#include "qemu/osdep.h" +#include "qemu/char_dev.h" + +static int open_cdev_internal(const char *path, dev_t cdev) +{ + struct stat st; + int fd; + + fd =3D qemu_open_old(path, O_RDWR); + if (fd =3D=3D -1) { + return -1; + } + if (fstat(fd, &st) || !S_ISCHR(st.st_mode) || + (cdev !=3D 0 && st.st_rdev !=3D cdev)) { + close(fd); + return -1; + } + return fd; +} + +static int open_cdev_robust(dev_t cdev) +{ + char *devpath; + int ret; + + /* + * This assumes that udev is being used and is creating the /dev/char/ + * symlinks. + */ + devpath =3D g_strdup_printf("/dev/char/%u:%u", major(cdev), minor(cdev= )); + ret =3D open_cdev_internal(devpath, cdev); + g_free(devpath); + return ret; +} + +int open_cdev(const char *devpath, dev_t cdev) +{ + int fd; + + fd =3D open_cdev_internal(devpath, cdev); + if (fd =3D=3D -1 && cdev !=3D 0) { + return open_cdev_robust(cdev); + } + return fd; +} diff --git a/util/meson.build b/util/meson.build index a375160286..d5313d858f 100644 --- a/util/meson.build +++ b/util/meson.build @@ -107,6 +107,7 @@ if have_block util_ss.add(files('filemonitor-stub.c')) endif util_ss.add(when: 'CONFIG_LINUX', if_true: files('vfio-helpers.c')) + util_ss.add(when: 'CONFIG_LINUX', if_true: files('chardev_open.c')) endif =20 if cpu =3D=3D 'aarch64' --=20 2.34.1 From nobody Wed May 15 22:12:51 2024 Delivered-To: importer@patchew.org Authentication-Results: mx.zohomail.com; dkim=pass header.i=@intel.com; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass(p=none dis=none) header.from=intel.com ARC-Seal: i=1; a=rsa-sha256; t=1693392893; cv=none; d=zohomail.com; s=zohoarc; b=Zz/8AaolCa2wlp90u0G7Uvh9uu4H8+PFD28DdotT8BPUqNvSaWko/DmPXbQPfYhEwVkRVW7pngZY2o6HzynVPAV5FYofOu3iaLYuo+9+P6vRys7uXoQgRjl7NL6+IU+XsqbvhPUbQh2HfL+qIY71eJhbscdkKRVxjltSrF5mT+I= ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=zohomail.com; s=zohoarc; t=1693392893; h=Content-Transfer-Encoding:Cc:Date:From:In-Reply-To:List-Subscribe:List-Post:List-Id:List-Archive:List-Help:List-Unsubscribe:MIME-Version:Message-ID:References:Sender:Subject:To; bh=nA7oBgMOkRNGD9X8osoU4YjUShE3YKnheNH9wVCjBHw=; b=Zk4LqP/juwzXuWKCKIWuYd3BhFZPzFviIi2ZM1cCilM1sDlSvGgHZ3mnO67HXKa93XW3tdt2khb7Jdav86cy8JONRElNLPFZOMGSjp3RyEAI56mVLN6SBS+DFshgby90Yk0IDWorCsnGH98s2+j/F5YqkSQ1WfmJAMEwIpgofPU= ARC-Authentication-Results: i=1; mx.zohomail.com; dkim=pass header.i=@intel.com; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass header.from= (p=none dis=none) Return-Path: Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) by mx.zohomail.com with SMTPS id 1693392893265678.0580158843032; Wed, 30 Aug 2023 03:54:53 -0700 (PDT) Received: from localhost ([::1] helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1qbIq2-0006QF-BH; Wed, 30 Aug 2023 06:54:18 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1qbIq0-0006Nt-2j for qemu-devel@nongnu.org; Wed, 30 Aug 2023 06:54:16 -0400 Received: from mgamail.intel.com ([192.55.52.93]) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1qbIpw-0007jr-9J for qemu-devel@nongnu.org; Wed, 30 Aug 2023 06:54:15 -0400 Received: from orsmga006.jf.intel.com ([10.7.209.51]) by fmsmga102.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 30 Aug 2023 03:53:53 -0700 Received: from duan-server-s2600bt.bj.intel.com ([10.240.192.147]) by orsmga006-auth.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 30 Aug 2023 03:53:48 -0700 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1693392852; x=1724928852; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=SjCB66+QBBn1CbZW0Uyo3MoEmPFX2rCBHPJKnH5sunU=; b=oBN0ksmhjymZKqLF8+8a/wVfU2VixxZuE0b8JDpiTcEdARIEq1A/C6rm kOM/Sr01R0fu/lumdVaQnZUG20f1FLGQ/AbZF5BxqAm7gL8Nbd8wE7aV0 aXIL5Gjwuk1+yOhSXLn1SyN5Ev5bAlg5+ZED+1/opJpZfgCAWPAUrp5rP sg7cRn6kXcXC+G9wTUib3C9CET84+KlVP8YX8lzcx11Utj/YuKqD4RbjI oVs+coitDfkvGs6pdOhnf13NaZpJd3x7aqtrh3T37qyvqA6MBLt6zXwrO dN6n1ahtmXL6MR4PiSEL+3m26+an/glyol4S1dSdgCjAu2d0ifYcOLh92 A==; X-IronPort-AV: E=McAfee;i="6600,9927,10817"; a="373016695" X-IronPort-AV: E=Sophos;i="6.02,213,1688454000"; d="scan'208";a="373016695" X-ExtLoop1: 1 X-IronPort-AV: E=McAfee;i="6600,9927,10817"; a="715866275" X-IronPort-AV: E=Sophos;i="6.02,213,1688454000"; d="scan'208";a="715866275" From: Zhenzhong Duan To: qemu-devel@nongnu.org Cc: alex.williamson@redhat.com, clg@redhat.com, jgg@nvidia.com, nicolinc@nvidia.com, joao.m.martins@oracle.com, eric.auger@redhat.com, peterx@redhat.com, jasowang@redhat.com, kevin.tian@intel.com, yi.l.liu@intel.com, yi.y.sun@intel.com, chao.p.peng@intel.com, Zhenzhong Duan Subject: [PATCH v1 18/22] vfio/iommufd: Implement the iommufd backend Date: Wed, 30 Aug 2023 18:37:50 +0800 Message-Id: <20230830103754.36461-19-zhenzhong.duan@intel.com> X-Mailer: git-send-email 2.34.1 In-Reply-To: <20230830103754.36461-1-zhenzhong.duan@intel.com> References: <20230830103754.36461-1-zhenzhong.duan@intel.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Received-SPF: pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) client-ip=209.51.188.17; envelope-from=qemu-devel-bounces+importer=patchew.org@nongnu.org; helo=lists.gnu.org; Received-SPF: pass client-ip=192.55.52.93; envelope-from=zhenzhong.duan@intel.com; helo=mgamail.intel.com X-Spam_score_int: -43 X-Spam_score: -4.4 X-Spam_bar: ---- X-Spam_report: (-4.4 / 5.0 requ) BAYES_00=-1.9, DKIMWL_WL_HIGH=-0.001, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, RCVD_IN_DNSWL_MED=-2.3, SPF_HELO_NONE=0.001, SPF_PASS=-0.001 autolearn=ham autolearn_force=no X-Spam_action: no action X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: qemu-devel-bounces+importer=patchew.org@nongnu.org Sender: qemu-devel-bounces+importer=patchew.org@nongnu.org X-ZohoMail-DKIM: pass (identity @intel.com) X-ZM-MESSAGEID: 1693392939002100013 Content-Type: text/plain; charset="utf-8" From: Yi Liu Add the iommufd backend. The IOMMUFD container class is implemented based on the new /dev/iommu user API. This backend obviously depends on CONFIG_IOMMUFD. So far, the iommufd backend doesn't support dirty page sync yet due to missing support in the host kernel. Co-authored-by: Eric Auger Signed-off-by: Eric Auger Signed-off-by: Yi Liu Signed-off-by: Zhenzhong Duan --- hw/vfio/common.c | 12 +- hw/vfio/iommufd.c | 521 ++++++++++++++++++++++++++ hw/vfio/meson.build | 3 + hw/vfio/trace-events | 12 + include/hw/vfio/vfio-common.h | 25 ++ include/hw/vfio/vfio-container-base.h | 3 + 6 files changed, 574 insertions(+), 2 deletions(-) create mode 100644 hw/vfio/iommufd.c diff --git a/hw/vfio/common.c b/hw/vfio/common.c index 6c3e98d5fd..b40cd8544d 100644 --- a/hw/vfio/common.c +++ b/hw/vfio/common.c @@ -45,7 +45,7 @@ #include "migration/qemu-file.h" #include "sysemu/tpm.h" =20 -static QLIST_HEAD(, VFIOAddressSpace) vfio_address_spaces =3D +VFIOAddressSpaceList vfio_address_spaces =3D QLIST_HEAD_INITIALIZER(vfio_address_spaces); =20 #ifdef CONFIG_KVM @@ -1520,8 +1520,16 @@ int vfio_attach_device(char *name, VFIODevice *vbase= dev, { const VFIOIOMMUBackendOpsClass *ops; =20 - ops =3D VFIO_IOMMU_BACKEND_OPS_CLASS( +#ifdef CONFIG_IOMMUFD + if (vbasedev->iommufd) { + ops =3D VFIO_IOMMU_BACKEND_OPS_CLASS( + object_class_by_name(TYPE_VFIO_IOMMU_BACKEND_IOMMUFD_OPS= )); + } else +#endif + { + ops =3D VFIO_IOMMU_BACKEND_OPS_CLASS( object_class_by_name(TYPE_VFIO_IOMMU_BACKEND_LEGACY_OPS)= ); + } if (!ops) { error_setg(errp, "VFIO IOMMU Backend not found!"); return -ENODEV; diff --git a/hw/vfio/iommufd.c b/hw/vfio/iommufd.c new file mode 100644 index 0000000000..876d0e4928 --- /dev/null +++ b/hw/vfio/iommufd.c @@ -0,0 +1,521 @@ +/* + * iommufd container backend + * + * Copyright (C) 2023 Intel Corporation. + * Copyright Red Hat, Inc. 2023 + * + * Authors: Yi Liu + * Eric Auger + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + + * You should have received a copy of the GNU General Public License along + * with this program; if not, see . + */ + +#include "qemu/osdep.h" +#include +#include +#include + +#include "hw/vfio/vfio-common.h" +#include "qemu/error-report.h" +#include "trace.h" +#include "qapi/error.h" +#include "sysemu/iommufd.h" +#include "hw/qdev-core.h" +#include "sysemu/reset.h" +#include "qemu/cutils.h" +#include "qemu/char_dev.h" + +static int iommufd_map(VFIOContainer *bcontainer, hwaddr iova, + ram_addr_t size, void *vaddr, bool readonly) +{ + VFIOIOMMUFDContainer *container =3D + container_of(bcontainer, VFIOIOMMUFDContainer, bcontainer); + + return iommufd_backend_map_dma(container->be, + container->ioas_id, + iova, size, vaddr, readonly); +} + +static int iommufd_unmap(VFIOContainer *bcontainer, + hwaddr iova, ram_addr_t size, + IOMMUTLBEntry *iotlb) +{ + VFIOIOMMUFDContainer *container =3D + container_of(bcontainer, VFIOIOMMUFDContainer, bcontainer); + + /* TODO: Handle dma_unmap_bitmap with iotlb args (migration) */ + return iommufd_backend_unmap_dma(container->be, + container->ioas_id, iova, size); +} + +static void vfio_kvm_device_add_device(VFIODevice *vbasedev) +{ + vfio_kvm_device_add_fd(vbasedev->fd); +} + +static void vfio_kvm_device_del_device(VFIODevice *vbasedev) +{ + vfio_kvm_device_del_fd(vbasedev->fd); +} + +static int iommufd_connect_and_bind(VFIODevice *vbasedev, Error **errp) +{ + IOMMUFDBackend *iommufd =3D vbasedev->iommufd; + struct vfio_device_bind_iommufd bind =3D { + .argsz =3D sizeof(bind), + .flags =3D 0, + }; + int ret; + + ret =3D iommufd_backend_connect(iommufd, errp); + if (ret) { + return ret; + } + + /* + * Add device to kvm-vfio to be prepared for the tracking + * in KVM. Especially for some emulated devices, it requires + * to have kvm information in the device open. + */ + vfio_kvm_device_add_device(vbasedev); + + /* Bind device to iommufd */ + bind.iommufd =3D iommufd->fd; + ret =3D ioctl(vbasedev->fd, VFIO_DEVICE_BIND_IOMMUFD, &bind); + if (ret) { + error_setg_errno(errp, errno, "error bind device fd=3D%d to iommuf= d=3D%d", + vbasedev->fd, bind.iommufd); + goto err_bind; + } + + vbasedev->devid =3D bind.out_devid; + trace_vfio_iommufd_bind_device(bind.iommufd, vbasedev->name, + vbasedev->fd, vbasedev->devid); + return ret; +err_bind: + vfio_kvm_device_del_device(vbasedev); + iommufd_backend_disconnect(iommufd); + return ret; +} + +static void iommufd_unbind_and_disconnect(VFIODevice *vbasedev) +{ + /* Unbind is automatically conducted when device fd is closed */ + vfio_kvm_device_del_device(vbasedev); + iommufd_backend_disconnect(vbasedev->iommufd); +} + +static int vfio_get_devicefd(const char *sysfs_path, Error **errp) +{ + long int ret =3D -ENOTTY; + char *path, *vfio_dev_path =3D NULL, *vfio_path =3D NULL; + DIR *dir =3D NULL; + struct dirent *dent; + gchar *contents; + struct stat st; + gsize length; + int major, minor; + dev_t vfio_devt; + + path =3D g_strdup_printf("%s/vfio-dev", sysfs_path); + if (stat(path, &st) < 0) { + error_setg_errno(errp, errno, "no such host device"); + goto out_free_path; + } + + dir =3D opendir(path); + if (!dir) { + error_setg_errno(errp, errno, "couldn't open dirrectory %s", path); + goto out_free_path; + } + + while ((dent =3D readdir(dir))) { + if (!strncmp(dent->d_name, "vfio", 4)) { + vfio_dev_path =3D g_strdup_printf("%s/%s/dev", path, dent->d_n= ame); + break; + } + } + + if (!vfio_dev_path) { + error_setg(errp, "failed to find vfio-dev/vfioX/dev"); + goto out_close_dir; + } + + if (!g_file_get_contents(vfio_dev_path, &contents, &length, NULL)) { + error_setg(errp, "failed to load \"%s\"", vfio_dev_path); + goto out_free_dev_path; + } + + if (sscanf(contents, "%d:%d", &major, &minor) !=3D 2) { + error_setg(errp, "failed to get major:minor for \"%s\"", vfio_dev_= path); + goto out_free_dev_path; + } + g_free(contents); + vfio_devt =3D makedev(major, minor); + + vfio_path =3D g_strdup_printf("/dev/vfio/devices/%s", dent->d_name); + ret =3D open_cdev(vfio_path, vfio_devt); + if (ret < 0) { + error_setg(errp, "Failed to open %s", vfio_path); + } + + trace_vfio_iommufd_get_devicefd(vfio_path, ret); + g_free(vfio_path); + +out_free_dev_path: + g_free(vfio_dev_path); +out_close_dir: + closedir(dir); +out_free_path: + if (*errp) { + error_prepend(errp, VFIO_MSG_PREFIX, path); + } + g_free(path); + + return ret; +} + +static VFIOIOASHwpt *vfio_container_get_hwpt(VFIOIOMMUFDContainer *contain= er, + uint32_t hwpt_id) +{ + VFIOIOASHwpt *hwpt; + + QLIST_FOREACH(hwpt, &container->hwpt_list, next) { + if (hwpt->hwpt_id =3D=3D hwpt_id) { + return hwpt; + } + } + + hwpt =3D g_malloc0(sizeof(*hwpt)); + + hwpt->hwpt_id =3D hwpt_id; + QLIST_INIT(&hwpt->device_list); + QLIST_INSERT_HEAD(&container->hwpt_list, hwpt, next); + + return hwpt; +} + +static void vfio_container_put_hwpt(IOMMUFDBackend *be, VFIOIOASHwpt *hwpt) +{ + QLIST_REMOVE(hwpt, next); + iommufd_backend_free_id(be->fd, hwpt->hwpt_id); + g_free(hwpt); +} + +static int __vfio_device_attach_hwpt(VFIODevice *vbasedev, uint32_t hwpt_i= d, + Error **errp) +{ + struct vfio_device_attach_iommufd_pt attach_data =3D { + .argsz =3D sizeof(attach_data), + .flags =3D 0, + .pt_id =3D hwpt_id, + }; + int ret; + + ret =3D ioctl(vbasedev->fd, VFIO_DEVICE_ATTACH_IOMMUFD_PT, &attach_dat= a); + if (ret) { + error_setg_errno(errp, errno, + "[iommufd=3D%d] error attach %s (%d) to hwpt_id= =3D%d", + vbasedev->iommufd->fd, vbasedev->name, vbasedev->= fd, + hwpt_id); + } + return ret; +} + +static int __vfio_device_detach_hwpt(VFIODevice *vbasedev, Error **errp) +{ + struct vfio_device_detach_iommufd_pt detach_data =3D { + .argsz =3D sizeof(detach_data), + .flags =3D 0, + }; + int ret; + + ret =3D ioctl(vbasedev->fd, VFIO_DEVICE_DETACH_IOMMUFD_PT, &detach_dat= a); + if (ret) { + error_setg_errno(errp, errno, "detach %s from ioas failed", + vbasedev->name); + } + return ret; +} + +static int vfio_device_attach_container(VFIODevice *vbasedev, + VFIOIOMMUFDContainer *container, + Error **errp) +{ + int ret, iommufd =3D vbasedev->iommufd->fd; + VFIOIOASHwpt *hwpt; + uint32_t hwpt_id; + Error *err =3D NULL; + + /* try to attach to an existing hwpt in this container */ + QLIST_FOREACH(hwpt, &container->hwpt_list, next) { + ret =3D __vfio_device_attach_hwpt(vbasedev, hwpt->hwpt_id, &err); + if (ret) { + const char *msg =3D error_get_pretty(err); + + trace_vfio_iommufd_fail_attach_existing_hwpt(msg); + error_free(err); + err =3D NULL; + } else { + goto found_hwpt; + } + } + + ret =3D iommufd_backend_alloc_hwpt(iommufd, vbasedev->devid, + container->ioas_id, &hwpt_id); + + if (ret) { + error_setg_errno(errp, errno, "error alloc shadow hwpt"); + return ret; + } + + /* Attach device to an hwpt within iommufd */ + ret =3D __vfio_device_attach_hwpt(vbasedev, hwpt_id, errp); + if (ret) { + iommufd_backend_free_id(iommufd, hwpt_id); + return ret; + } + + hwpt =3D vfio_container_get_hwpt(container, hwpt_id); +found_hwpt: + QLIST_INSERT_HEAD(&hwpt->device_list, vbasedev, next); + vbasedev->hwpt =3D hwpt; + + trace_vfio_iommufd_attach_device(iommufd, vbasedev->name, vbasedev->fd, + container->ioas_id, hwpt->hwpt_id); + return ret; +} + +static void vfio_device_detach_container(VFIODevice *vbasedev, + VFIOIOMMUFDContainer *container, + Error **errp) +{ + VFIOIOASHwpt *hwpt =3D vbasedev->hwpt; + + __vfio_device_detach_hwpt(vbasedev, errp); + QLIST_REMOVE(vbasedev, next); + vbasedev->hwpt =3D NULL; + if (QLIST_EMPTY(&hwpt->device_list)) { + vfio_container_put_hwpt(vbasedev->iommufd, hwpt); + } + + trace_vfio_iommufd_detach_device(container->be->fd, vbasedev->name, + container->ioas_id); +} + +static void vfio_iommufd_container_destroy(VFIOIOMMUFDContainer *container) +{ + VFIOContainer *bcontainer =3D &container->bcontainer; + + if (!QLIST_EMPTY(&container->hwpt_list)) { + return; + } + memory_listener_unregister(&bcontainer->listener); + vfio_container_destroy(bcontainer); + iommufd_backend_put_ioas(container->be, container->ioas_id); + g_free(container); +} + +static int vfio_ram_block_discard_disable(bool state) +{ + /* + * We support coordinated discarding of RAM via the RamDiscardManager. + */ + return ram_block_uncoordinated_discard_disable(state); +} + +static int iommufd_attach_device(char *name, VFIODevice *vbasedev, + AddressSpace *as, Error **errp) +{ + VFIOIOMMUBackendOpsClass *ops =3D VFIO_IOMMU_BACKEND_OPS_CLASS( + object_class_by_name(TYPE_VFIO_IOMMU_BACKEND_IOMMUFD_OPS)); + VFIOContainer *bcontainer; + VFIOIOMMUFDContainer *container; + VFIOAddressSpace *space; + struct vfio_device_info dev_info =3D { .argsz =3D sizeof(dev_info) }; + int ret, devfd; + uint32_t ioas_id; + Error *err =3D NULL; + + devfd =3D vfio_get_devicefd(vbasedev->sysfsdev, errp); + if (devfd < 0) { + return devfd; + } + vbasedev->fd =3D devfd; + + ret =3D iommufd_connect_and_bind(vbasedev, errp); + if (ret) { + goto err_connect_bind; + } + + space =3D vfio_get_address_space(as); + + /* try to attach to an existing container in this space */ + QLIST_FOREACH(bcontainer, &space->containers, next) { + container =3D container_of(bcontainer, VFIOIOMMUFDContainer, bcont= ainer); + if (bcontainer->ops !=3D ops || vbasedev->iommufd !=3D container->= be) { + continue; + } + if (vfio_device_attach_container(vbasedev, container, &err)) { + const char *msg =3D error_get_pretty(err); + + trace_vfio_iommufd_fail_attach_existing_container(msg); + error_free(err); + err =3D NULL; + } else { + ret =3D vfio_ram_block_discard_disable(true); + if (ret) { + error_setg(errp, + "Cannot set discarding of RAM broken (%d)", = ret); + goto err_discard_disable; + } + goto found_container; + } + } + + /* Need to allocate a new dedicated container */ + ret =3D iommufd_backend_get_ioas(vbasedev->iommufd, &ioas_id); + if (ret < 0) { + error_setg_errno(errp, errno, "Failed to alloc ioas"); + goto err_get_ioas; + } + + trace_vfio_iommufd_alloc_ioas(vbasedev->iommufd->fd, ioas_id); + + container =3D g_malloc0(sizeof(*container)); + container->be =3D vbasedev->iommufd; + container->ioas_id =3D ioas_id; + QLIST_INIT(&container->hwpt_list); + + bcontainer =3D &container->bcontainer; + vfio_container_init(bcontainer, space, ops); + + ret =3D vfio_device_attach_container(vbasedev, container, errp); + if (ret) { + goto err_attach_container; + } + + ret =3D vfio_ram_block_discard_disable(true); + if (ret) { + goto err_discard_disable; + } + + /* + * TODO: for now iommufd BE is on par with vfio iommu type1, so it's + * fine to add the whole range as window. For SPAPR, below code + * should be updated. + */ + vfio_host_win_add(bcontainer, 0, (hwaddr)-1, 4096); + bcontainer->pgsizes =3D 4096; + + QLIST_INSERT_HEAD(&space->containers, bcontainer, next); + + bcontainer->listener =3D vfio_memory_listener; + + memory_listener_register(&bcontainer->listener, bcontainer->space->as); + + bcontainer->initialized =3D true; + +found_container: + /* + * TODO: examine RAM_BLOCK_DISCARD stuff, should we do group level + * for discarding incompatibility check as well? + */ + if (vbasedev->ram_block_discard_allowed) { + vfio_ram_block_discard_disable(false); + } + + ret =3D ioctl(devfd, VFIO_DEVICE_GET_INFO, &dev_info); + if (ret) { + error_setg_errno(errp, errno, "error getting device info"); + goto err_discard_disable; + } + + vbasedev->group =3D 0; + vbasedev->num_irqs =3D dev_info.num_irqs; + vbasedev->num_regions =3D dev_info.num_regions; + vbasedev->flags =3D dev_info.flags; + vbasedev->reset_works =3D !!(dev_info.flags & VFIO_DEVICE_FLAGS_RESET); + vbasedev->container =3D bcontainer; + + trace_vfio_iommufd_device_info(vbasedev->name, devfd, vbasedev->num_ir= qs, + vbasedev->num_regions, vbasedev->flags); + return 0; + +err_discard_disable: + vfio_device_detach_container(vbasedev, container, &err); + if (err) { + error_report_err(err); + } +err_attach_container: + vfio_iommufd_container_destroy(container); +err_get_ioas: + vfio_put_address_space(space); + iommufd_unbind_and_disconnect(vbasedev); +err_connect_bind: + close(vbasedev->fd); + return ret; +} + +static void iommufd_detach_device(VFIODevice *vbasedev) +{ + VFIOContainer *bcontainer =3D vbasedev->container; + VFIOIOMMUFDContainer *container; + VFIOAddressSpace *space =3D bcontainer->space; + Error *err =3D NULL; + + if (!bcontainer) { + return; + } + + if (!vbasedev->ram_block_discard_allowed) { + vfio_ram_block_discard_disable(false); + } + + container =3D container_of(bcontainer, VFIOIOMMUFDContainer, bcontaine= r); + vfio_device_detach_container(vbasedev, container, &err); + if (err) { + error_report_err(err); + } + if (QLIST_EMPTY(&container->hwpt_list)) { + vfio_iommufd_container_destroy(container); + vfio_put_address_space(space); + } + vbasedev->container =3D NULL; + iommufd_unbind_and_disconnect(vbasedev); + close(vbasedev->fd); +} + +static void vfio_iommu_backend_iommufd_ops_class_init(ObjectClass *oc, + void *data) { + VFIOIOMMUBackendOpsClass *ops =3D VFIO_IOMMU_BACKEND_OPS_CLASS(oc); + + ops->dma_map =3D iommufd_map; + ops->dma_unmap =3D iommufd_unmap; + ops->attach_device =3D iommufd_attach_device; + ops->detach_device =3D iommufd_detach_device; +} + +static const TypeInfo vfio_iommu_backend_iommufd_ops_type =3D { + .name =3D TYPE_VFIO_IOMMU_BACKEND_IOMMUFD_OPS, + + .parent =3D TYPE_VFIO_IOMMU_BACKEND_OPS, + .class_init =3D vfio_iommu_backend_iommufd_ops_class_init, + .abstract =3D true, +}; +static void vfio_iommu_backend_iommufd_ops_register_types(void) +{ + type_register_static(&vfio_iommu_backend_iommufd_ops_type); +} +type_init(vfio_iommu_backend_iommufd_ops_register_types); diff --git a/hw/vfio/meson.build b/hw/vfio/meson.build index eb6ce6229d..9cae2c9e21 100644 --- a/hw/vfio/meson.build +++ b/hw/vfio/meson.build @@ -7,6 +7,9 @@ vfio_ss.add(files( 'spapr.c', 'migration.c', )) +if have_iommufd + vfio_ss.add(files('iommufd.c')) +endif vfio_ss.add(when: 'CONFIG_VFIO_PCI', if_true: files( 'display.c', 'pci-quirks.c', diff --git a/hw/vfio/trace-events b/hw/vfio/trace-events index 1692bcd8f1..60b56f23a1 100644 --- a/hw/vfio/trace-events +++ b/hw/vfio/trace-events @@ -167,3 +167,15 @@ vfio_save_setup(const char *name, uint64_t data_buffer= _size) " (%s) data buffer vfio_state_pending_estimate(const char *name, uint64_t precopy, uint64_t p= ostcopy, uint64_t precopy_init_size, uint64_t precopy_dirty_size) " (%s) pr= ecopy 0x%"PRIx64" postcopy 0x%"PRIx64" precopy initial size 0x%"PRIx64" pre= copy dirty size 0x%"PRIx64 vfio_state_pending_exact(const char *name, uint64_t precopy, uint64_t post= copy, uint64_t stopcopy_size, uint64_t precopy_init_size, uint64_t precopy_= dirty_size) " (%s) precopy 0x%"PRIx64" postcopy 0x%"PRIx64" stopcopy size 0= x%"PRIx64" precopy initial size 0x%"PRIx64" precopy dirty size 0x%"PRIx64 vfio_vmstate_change(const char *name, int running, const char *reason, con= st char *dev_state) " (%s) running %d reason %s device state %s" + +#iommufd.c + +vfio_iommufd_get_devicefd(const char *dev, int devfd) " %s (fd=3D%d)" +vfio_iommufd_bind_device(int iommufd, const char *name, int devfd, int dev= id) " [iommufd=3D%d] Successfully bound device %s (fd=3D%d): output devid= =3D%d" +vfio_iommufd_fail_attach_existing_hwpt(const char *msg) " %s" +vfio_iommufd_attach_device(int iommufd, const char *name, int devfd, int i= oasid, int hwptid) " [iommufd=3D%d] Successfully attached device %s (%d) to= ioasid=3D%d: output hwptd=3D%d" +vfio_iommufd_detach_device(int iommufd, const char *name, int ioasid) " [i= ommufd=3D%d] Detached %s from ioasid=3D%d" +vfio_iommufd_alloc_ioas(int iommufd, int ioas_id) " [iommufd=3D%d] new IOM= MUFD container with ioasid=3D%d" +vfio_iommufd_device_info(char *name, int devfd, int num_irqs, int num_regi= ons, int flags) " %s (%d) num_irqs=3D%d num_regions=3D%d flags=3D%d" +vfio_iommufd_fail_attach_existing_container(const char *msg) " %s" +vfio_iommufd_container_reset(char *name) " Successfully reset %s" diff --git a/include/hw/vfio/vfio-common.h b/include/hw/vfio/vfio-common.h index b1a76dcc9c..027a59a13a 100644 --- a/include/hw/vfio/vfio-common.h +++ b/include/hw/vfio/vfio-common.h @@ -85,6 +85,26 @@ typedef struct VFIOLegacyContainer { QLIST_HEAD(, VFIOGroup) group_list; } VFIOLegacyContainer; =20 +#ifdef CONFIG_IOMMUFD +typedef struct VFIOIOASHwpt { + uint32_t hwpt_id; + QLIST_HEAD(, VFIODevice) device_list; + QLIST_ENTRY(VFIOIOASHwpt) next; +} VFIOIOASHwpt; + +typedef struct IOMMUFDBackend IOMMUFDBackend; + +typedef struct VFIOIOMMUFDContainer { + VFIOContainer bcontainer; + IOMMUFDBackend *be; + uint32_t ioas_id; + QLIST_HEAD(, VFIOIOASHwpt) hwpt_list; +} VFIOIOMMUFDContainer; +#endif + +typedef QLIST_HEAD(VFIOAddressSpaceList, VFIOAddressSpace) VFIOAddressSpac= eList; +extern VFIOAddressSpaceList vfio_address_spaces; + typedef struct VFIODeviceOps VFIODeviceOps; =20 typedef struct VFIODevice { @@ -110,6 +130,11 @@ typedef struct VFIODevice { OnOffAuto pre_copy_dirty_page_tracking; bool dirty_pages_supported; bool dirty_tracking; +#ifdef CONFIG_IOMMUFD + int devid; + VFIOIOASHwpt *hwpt; + IOMMUFDBackend *iommufd; +#endif } VFIODevice; =20 struct VFIODeviceOps { diff --git a/include/hw/vfio/vfio-container-base.h b/include/hw/vfio/vfio-c= ontainer-base.h index b18fa92146..51aff4af05 100644 --- a/include/hw/vfio/vfio-container-base.h +++ b/include/hw/vfio/vfio-container-base.h @@ -117,6 +117,9 @@ void vfio_container_init(VFIOContainer *container, void vfio_container_destroy(VFIOContainer *container); =20 #define TYPE_VFIO_IOMMU_BACKEND_LEGACY_OPS "vfio-iommu-backend-legacy-ops" +#ifdef CONFIG_IOMMUFD +#define TYPE_VFIO_IOMMU_BACKEND_IOMMUFD_OPS "vfio-iommu-backend-iommufd-op= s" +#endif #define TYPE_VFIO_IOMMU_BACKEND_OPS "vfio-iommu-backend-ops" =20 DECLARE_CLASS_CHECKERS(VFIOIOMMUBackendOpsClass, --=20 2.34.1 From nobody Wed May 15 22:12:51 2024 Delivered-To: importer@patchew.org Authentication-Results: mx.zohomail.com; dkim=pass header.i=@intel.com; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass(p=none dis=none) header.from=intel.com ARC-Seal: i=1; a=rsa-sha256; t=1693392937; cv=none; d=zohomail.com; s=zohoarc; b=PbdAVwFK09b/A+4b8yVq+s+Z4QfEgE0X/o/GtsKuWMDxMO38rOpLTcNR1LqAx6Bf6+xVkHi2+qFfoswWDrCOMWyLMPwC+exrSnYNK6zCB+vQ7NPpEk3R3su4BzljDWehV6V53DTLfj00VoeUVd4xzc6+Gog6eHCZxFZ+TQ5SKl4= ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=zohomail.com; s=zohoarc; t=1693392937; h=Content-Transfer-Encoding:Cc:Date:From:In-Reply-To:List-Subscribe:List-Post:List-Id:List-Archive:List-Help:List-Unsubscribe:MIME-Version:Message-ID:References:Sender:Subject:To; bh=nC1JMEqYXQpcDOFpblAjxEzU5oSsaAF656cdTgktlfs=; b=Z35xTDpHq5mCqcEUHk8KPTOA64V50aaBi9FTmpnxQxnpDVGEHXA62QsPIWpfrRf70odtcQbNd9QJgR4XOdeV6YI701CqItDt1aB9u5K1T+fmEnsTy1JgmG5SaM2zH3OncGMWjsoK64XTxP+/MCry0Xs0sXV9kLmChoAX0JoPcFs= ARC-Authentication-Results: i=1; mx.zohomail.com; dkim=pass header.i=@intel.com; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass header.from= (p=none dis=none) Return-Path: Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) by mx.zohomail.com with SMTPS id 1693392937312153.258999785235; Wed, 30 Aug 2023 03:55:37 -0700 (PDT) Received: from localhost ([::1] helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1qbIq7-0006kC-0u; Wed, 30 Aug 2023 06:54:23 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1qbIq5-0006fr-FF for qemu-devel@nongnu.org; Wed, 30 Aug 2023 06:54:21 -0400 Received: from mgamail.intel.com ([192.55.52.93]) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1qbIq3-0007jy-5r for qemu-devel@nongnu.org; Wed, 30 Aug 2023 06:54:21 -0400 Received: from orsmga006.jf.intel.com ([10.7.209.51]) by fmsmga102.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 30 Aug 2023 03:53:56 -0700 Received: from duan-server-s2600bt.bj.intel.com ([10.240.192.147]) by orsmga006-auth.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 30 Aug 2023 03:53:52 -0700 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1693392859; x=1724928859; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=nfgDOi+Bnd7/f8ZFjHWNuRaWM7jar4LHHKIB8rQrMAk=; b=B1NSc0o8XHQaI0u/d6eIPJumOYTnIGf2XVH0uEEVFbHdJ3CvfqlSXfcc 6ZmgwjMoKMjgPKJzGncZKNjOixvEg0BznL9NGzkMjZBUfIn3R6+LU4Z87 K+YLBF/zHLKvLzyrfeZXNZz17YpzY1W/lmJXnOhQR/mWvCsbLuzN4DSQX Ra115nUNHVDjwXOqtFG0D4Y5L7ySaOzqzRUIhFfeUdxzx4edYYB/Ophxz x984e8yMW2D0D6yF6AMbekMtZB6Te2RGSieRRgnNnLbfJ2h3A0myWxgSB Ousd2gDW2uDHPBYeNPwEeeJICsPMVsL30/t0ORM5mLs+atDpt2SLTHEXk g==; X-IronPort-AV: E=McAfee;i="6600,9927,10817"; a="373016707" X-IronPort-AV: E=Sophos;i="6.02,213,1688454000"; d="scan'208";a="373016707" X-ExtLoop1: 1 X-IronPort-AV: E=McAfee;i="6600,9927,10817"; a="715866287" X-IronPort-AV: E=Sophos;i="6.02,213,1688454000"; d="scan'208";a="715866287" From: Zhenzhong Duan To: qemu-devel@nongnu.org Cc: alex.williamson@redhat.com, clg@redhat.com, jgg@nvidia.com, nicolinc@nvidia.com, joao.m.martins@oracle.com, eric.auger@redhat.com, peterx@redhat.com, jasowang@redhat.com, kevin.tian@intel.com, yi.l.liu@intel.com, yi.y.sun@intel.com, chao.p.peng@intel.com, Zhenzhong Duan Subject: [PATCH v1 19/22] vfio/iommufd: Add vfio device iterator callback for iommufd Date: Wed, 30 Aug 2023 18:37:51 +0800 Message-Id: <20230830103754.36461-20-zhenzhong.duan@intel.com> X-Mailer: git-send-email 2.34.1 In-Reply-To: <20230830103754.36461-1-zhenzhong.duan@intel.com> References: <20230830103754.36461-1-zhenzhong.duan@intel.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Received-SPF: pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) client-ip=209.51.188.17; envelope-from=qemu-devel-bounces+importer=patchew.org@nongnu.org; helo=lists.gnu.org; Received-SPF: pass client-ip=192.55.52.93; envelope-from=zhenzhong.duan@intel.com; helo=mgamail.intel.com X-Spam_score_int: -43 X-Spam_score: -4.4 X-Spam_bar: ---- X-Spam_report: (-4.4 / 5.0 requ) BAYES_00=-1.9, DKIMWL_WL_HIGH=-0.001, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, RCVD_IN_DNSWL_MED=-2.3, SPF_HELO_NONE=0.001, SPF_PASS=-0.001 autolearn=ham autolearn_force=no X-Spam_action: no action X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: qemu-devel-bounces+importer=patchew.org@nongnu.org Sender: qemu-devel-bounces+importer=patchew.org@nongnu.org X-ZohoMail-DKIM: pass (identity @intel.com) X-ZM-MESSAGEID: 1693392940752100021 Content-Type: text/plain; charset="utf-8" The way to get vfio device pointer is different between legacy container and iommufd container, with iommufd backend support added, it's time to add the iterator support for iommufd. Signed-off-by: Zhenzhong Duan --- hw/vfio/iommufd.c | 29 +++++++++++++++++++++++++++++ 1 file changed, 29 insertions(+) diff --git a/hw/vfio/iommufd.c b/hw/vfio/iommufd.c index 876d0e4928..dd24e76e39 100644 --- a/hw/vfio/iommufd.c +++ b/hw/vfio/iommufd.c @@ -36,6 +36,34 @@ #include "qemu/cutils.h" #include "qemu/char_dev.h" =20 +static VFIODevice *iommufd_dev_iter_next(VFIOContainer *bcontainer, + VFIODevice *curr) +{ + + VFIOIOASHwpt *hwpt; + + assert(object_class_dynamic_cast(OBJECT_CLASS(bcontainer->ops), + TYPE_VFIO_IOMMU_BACKEND_IOMMUFD_OPS)); + + VFIOIOMMUFDContainer *container =3D container_of(bcontainer, + VFIOIOMMUFDContainer, + bcontainer); + + if (!curr) { + hwpt =3D QLIST_FIRST(&container->hwpt_list); + } else { + if (curr->next.le_next) { + return curr->next.le_next; + } + hwpt =3D curr->hwpt->next.le_next; + } + + if (!hwpt) { + return NULL; + } + return QLIST_FIRST(&hwpt->device_list); +} + static int iommufd_map(VFIOContainer *bcontainer, hwaddr iova, ram_addr_t size, void *vaddr, bool readonly) { @@ -501,6 +529,7 @@ static void vfio_iommu_backend_iommufd_ops_class_init(O= bjectClass *oc, void *data) { VFIOIOMMUBackendOpsClass *ops =3D VFIO_IOMMU_BACKEND_OPS_CLASS(oc); =20 + ops->dev_iter_next =3D iommufd_dev_iter_next; ops->dma_map =3D iommufd_map; ops->dma_unmap =3D iommufd_unmap; ops->attach_device =3D iommufd_attach_device; --=20 2.34.1 From nobody Wed May 15 22:12:51 2024 Delivered-To: importer@patchew.org Authentication-Results: mx.zohomail.com; dkim=pass header.i=@intel.com; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass(p=none dis=none) header.from=intel.com ARC-Seal: i=1; a=rsa-sha256; t=1693392894; cv=none; d=zohomail.com; s=zohoarc; b=DdZCQk4p7KnVvDUtHJcwI8YeSGUHd2cHty6qAtc9IvgKB+FyQTr3dKB9GpM7pfpBTMspFC6Cl7mcNPXJYlC0gYkwlqnHxopBzuWBi4z27nUa6HJ3oruEKpHoOPZqBZZDPa+vnyymaMnz8IvNoUMAc2XlNmaDd8b/4BpFPfJe3Ms= ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=zohomail.com; s=zohoarc; t=1693392894; h=Content-Transfer-Encoding:Cc:Date:From:In-Reply-To:List-Subscribe:List-Post:List-Id:List-Archive:List-Help:List-Unsubscribe:MIME-Version:Message-ID:References:Sender:Subject:To; bh=OVM3ylRxHg/x9ot1IKHhvKXyQBAImJSMuzltaQgDF0Q=; b=cs1i4zgwZssVjtKn6V9oTWW+VLXXNdnhgjDT299uiRwmSY2C2X4G/VG2q6393WUmtuzlmTjF39xml11B6tL2wGoEqHv3sBgpVzQU3I+Z61Ca2pJKFhFL0A4B8kz4sNJLxRfYRYn4mwtqLEBHmAr9QqENSBCO5AMY57vxMj4gop0= ARC-Authentication-Results: i=1; mx.zohomail.com; dkim=pass header.i=@intel.com; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass header.from= (p=none dis=none) Return-Path: Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) by mx.zohomail.com with SMTPS id 1693392894649567.4102040757415; Wed, 30 Aug 2023 03:54:54 -0700 (PDT) Received: from localhost ([::1] helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1qbIqM-0007XH-0u; Wed, 30 Aug 2023 06:54:38 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1qbIqA-00074j-N2 for qemu-devel@nongnu.org; Wed, 30 Aug 2023 06:54:28 -0400 Received: from mgamail.intel.com ([192.55.52.93]) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1qbIq7-0007je-NB for qemu-devel@nongnu.org; Wed, 30 Aug 2023 06:54:26 -0400 Received: from orsmga006.jf.intel.com ([10.7.209.51]) by fmsmga102.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 30 Aug 2023 03:54:00 -0700 Received: from duan-server-s2600bt.bj.intel.com ([10.240.192.147]) by orsmga006-auth.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 30 Aug 2023 03:53:56 -0700 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1693392863; x=1724928863; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=MJ3hN+vwo4EtjM/DCbistPA7dAnDb5Lm8vWq7drftMc=; b=Jc8jsUdpq/oRgH2wcZDrum//OoHq6AsFObwAGB78PPQn4UVBE4ix26KT 9Hh1Ja3ZkGhP1eQb1FWNCag6QVFXpPdsytPkC7cCGX5S1xt0USBiYLjYx qWDPxQLWvKEt8lr+z0AWkjFuDHuLfbWiAlZ7eMJlXCRAtSyX4DCVL1IDk 7cr9i9sE5cdcmxESpC5oLmo7p7NNoSe1GHHE2aBd6Hm80jG58ZjjLpxj2 Dw40TErImLgYrJOlTaCJXs7HBS9Xv3GKeI5mf7DGXBS6QUabbhuZibG7r 63FuEZXUiy9qo1g4NlejfrMy4/8lP05lS0puQdUUUxwyCXViRrW1MjmrN w==; X-IronPort-AV: E=McAfee;i="6600,9927,10817"; a="373016714" X-IronPort-AV: E=Sophos;i="6.02,213,1688454000"; d="scan'208";a="373016714" X-ExtLoop1: 1 X-IronPort-AV: E=McAfee;i="6600,9927,10817"; a="715866295" X-IronPort-AV: E=Sophos;i="6.02,213,1688454000"; d="scan'208";a="715866295" From: Zhenzhong Duan To: qemu-devel@nongnu.org Cc: alex.williamson@redhat.com, clg@redhat.com, jgg@nvidia.com, nicolinc@nvidia.com, joao.m.martins@oracle.com, eric.auger@redhat.com, peterx@redhat.com, jasowang@redhat.com, kevin.tian@intel.com, yi.l.liu@intel.com, yi.y.sun@intel.com, chao.p.peng@intel.com, Zhenzhong Duan Subject: [PATCH v1 20/22] vfio/pci: Adapt vfio pci hot reset support with iommufd BE Date: Wed, 30 Aug 2023 18:37:52 +0800 Message-Id: <20230830103754.36461-21-zhenzhong.duan@intel.com> X-Mailer: git-send-email 2.34.1 In-Reply-To: <20230830103754.36461-1-zhenzhong.duan@intel.com> References: <20230830103754.36461-1-zhenzhong.duan@intel.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Received-SPF: pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) client-ip=209.51.188.17; envelope-from=qemu-devel-bounces+importer=patchew.org@nongnu.org; helo=lists.gnu.org; Received-SPF: pass client-ip=192.55.52.93; envelope-from=zhenzhong.duan@intel.com; helo=mgamail.intel.com X-Spam_score_int: -43 X-Spam_score: -4.4 X-Spam_bar: ---- X-Spam_report: (-4.4 / 5.0 requ) BAYES_00=-1.9, DKIMWL_WL_HIGH=-0.001, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, RCVD_IN_DNSWL_MED=-2.3, SPF_HELO_NONE=0.001, SPF_PASS=-0.001 autolearn=ham autolearn_force=no X-Spam_action: no action X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: qemu-devel-bounces+importer=patchew.org@nongnu.org Sender: qemu-devel-bounces+importer=patchew.org@nongnu.org X-ZohoMail-DKIM: pass (identity @intel.com) X-ZM-MESSAGEID: 1693392896055100003 Content-Type: text/plain; charset="utf-8" As pci hot reset path need to reference pci specific functions and data structures, adding container level callback functions for legacy and iommufd BE and referencing those pci specific func/data is no better than implementing reset support with iommufd BE directly in pci.c This way we can also share the common bus reset and system reset path for different BEs. Signed-off-by: Zhenzhong Duan --- hw/vfio/pci.c | 224 +++++++++++++++++++++++++++++++++++++++---- hw/vfio/trace-events | 1 + 2 files changed, 208 insertions(+), 17 deletions(-) diff --git a/hw/vfio/pci.c b/hw/vfio/pci.c index 34f65ecd17..3a8fee3c99 100644 --- a/hw/vfio/pci.c +++ b/hw/vfio/pci.c @@ -42,6 +42,7 @@ #include "qapi/error.h" #include "migration/blocker.h" #include "migration/qemu-file.h" +#include "linux/iommufd.h" =20 #define TYPE_VFIO_PCI_NOHOTPLUG "vfio-pci-nohotplug" =20 @@ -2378,22 +2379,13 @@ static bool vfio_pci_host_match(PCIHostDeviceAddres= s *addr, const char *name) return (strcmp(tmp, name) =3D=3D 0); } =20 -static int vfio_pci_hot_reset(VFIOPCIDevice *vdev, bool single) +static int vfio_pci_get_pci_hot_reset_info(VFIOPCIDevice *vdev, + struct vfio_pci_hot_reset_info **in= fo_p) { - VFIOGroup *group; struct vfio_pci_hot_reset_info *info; - struct vfio_pci_dependent_device *devices; - struct vfio_pci_hot_reset *reset; - int32_t *fds; - int ret, i, count; - bool multi =3D false; - - trace_vfio_pci_hot_reset(vdev->vbasedev.name, single ? "one" : "multi"= ); + int ret, count; =20 - if (!single) { - vfio_pci_pre_reset(vdev); - } - vdev->vbasedev.needs_reset =3D false; + assert(info_p && !*info_p); =20 info =3D g_malloc0(sizeof(*info)); info->argsz =3D sizeof(*info); @@ -2401,24 +2393,53 @@ static int vfio_pci_hot_reset(VFIOPCIDevice *vdev, = bool single) ret =3D ioctl(vdev->vbasedev.fd, VFIO_DEVICE_GET_PCI_HOT_RESET_INFO, i= nfo); if (ret && errno !=3D ENOSPC) { ret =3D -errno; + g_free(info); if (!vdev->has_pm_reset) { error_report("vfio: Cannot reset device %s, " "no available reset mechanism.", vdev->vbasedev.n= ame); } - goto out_single; + return ret; } =20 count =3D info->count; - info =3D g_realloc(info, sizeof(*info) + (count * sizeof(*devices))); - info->argsz =3D sizeof(*info) + (count * sizeof(*devices)); - devices =3D &info->devices[0]; + info =3D g_realloc(info, sizeof(*info) + (count * sizeof(info->devices= [0]))); + info->argsz =3D sizeof(*info) + (count * sizeof(info->devices[0])); =20 ret =3D ioctl(vdev->vbasedev.fd, VFIO_DEVICE_GET_PCI_HOT_RESET_INFO, i= nfo); if (ret) { ret =3D -errno; + g_free(info); error_report("vfio: hot reset info failed: %m"); + return ret; + } + + *info_p =3D info; + return 0; +} + +static int vfio_pci_hot_reset_legacy(VFIOPCIDevice *vdev, bool single) +{ + VFIOGroup *group; + struct vfio_pci_hot_reset_info *info =3D NULL; + struct vfio_pci_dependent_device *devices; + struct vfio_pci_hot_reset *reset; + int32_t *fds; + int ret, i, count; + bool multi =3D false; + + trace_vfio_pci_hot_reset(vdev->vbasedev.name, single ? "one" : "multi"= ); + + if (!single) { + vfio_pci_pre_reset(vdev); + } + vdev->vbasedev.needs_reset =3D false; + + ret =3D vfio_pci_get_pci_hot_reset_info(vdev, &info); + + if (ret) { goto out_single; } + devices =3D &info->devices[0]; =20 trace_vfio_pci_hot_reset_has_dep_devices(vdev->vbasedev.name); =20 @@ -2560,6 +2581,175 @@ out_single: return ret; } =20 +#ifdef CONFIG_IOMMUFD +static VFIODevice *vfio_pci_find_by_iommufd_devid(__u32 devid) +{ + VFIOAddressSpace *space; + VFIOContainer *bcontainer; + VFIOIOMMUFDContainer *container; + VFIOIOASHwpt *hwpt; + VFIODevice *vbasedev_iter; + VFIOIOMMUBackendOpsClass *ops =3D VFIO_IOMMU_BACKEND_OPS_CLASS( + object_class_by_name(TYPE_VFIO_IOMMU_BACKEND_IOMMUFD_OPS)); + + QLIST_FOREACH(space, &vfio_address_spaces, list) { + QLIST_FOREACH(bcontainer, &space->containers, next) { + if (bcontainer->ops !=3D ops) { + continue; + } + container =3D container_of(bcontainer, VFIOIOMMUFDContainer, + bcontainer); + QLIST_FOREACH(hwpt, &container->hwpt_list, next) { + QLIST_FOREACH(vbasedev_iter, &hwpt->device_list, next) { + if (devid =3D=3D vbasedev_iter->devid) { + return vbasedev_iter; + } + } + } + } + } + return NULL; +} + +static int vfio_pci_hot_reset_iommufd(VFIOPCIDevice *vdev, bool single) +{ + struct vfio_pci_hot_reset_info *info =3D NULL; + struct vfio_pci_dependent_device *devices; + struct vfio_pci_hot_reset *reset; + int ret, i; + bool multi =3D false; + + trace_vfio_pci_hot_reset(vdev->vbasedev.name, single ? "one" : "multi"= ); + + if (!single) { + vfio_pci_pre_reset(vdev); + } + vdev->vbasedev.needs_reset =3D false; + + ret =3D vfio_pci_get_pci_hot_reset_info(vdev, &info); + + if (ret) { + goto out_single; + } + + assert(info->flags & VFIO_PCI_HOT_RESET_FLAG_DEV_ID); + + devices =3D &info->devices[0]; + + if (!(info->flags & VFIO_PCI_HOT_RESET_FLAG_DEV_ID_OWNED)) { + if (!vdev->has_pm_reset) { + for (i =3D 0; i < info->count; i++) { + if (devices[i].devid =3D=3D VFIO_PCI_DEVID_NOT_OWNED) { + error_report("vfio: Cannot reset device %s, " + "depends on device %04x:%02x:%02x.%x " + "which is not owned.", + vdev->vbasedev.name, devices[i].segment, + devices[i].bus, PCI_SLOT(devices[i].devfn= ), + PCI_FUNC(devices[i].devfn)); + } + } + } + ret =3D -EPERM; + goto out_single; + } + + trace_vfio_pci_hot_reset_has_dep_devices(vdev->vbasedev.name); + + for (i =3D 0; i < info->count; i++) { + VFIOPCIDevice *tmp; + VFIODevice *vbasedev_iter; + + trace_vfio_pci_hot_reset_dep_devices_iommufd(devices[i].segment, + devices[i].bus, + PCI_SLOT(devices[i].devfn), + PCI_FUNC(devices[i].devfn), + devices[i].devid); + + /* + * If a VFIO cdev device is resettable, all the dependent devices + * are either bound to same iommufd or within same iommu_groups as + * one of the iommufd bound devices. + */ + assert(devices[i].devid !=3D VFIO_PCI_DEVID_NOT_OWNED); + + if (devices[i].devid =3D=3D vdev->vbasedev.devid || + devices[i].devid =3D=3D VFIO_PCI_DEVID_OWNED) { + continue; + } + + vbasedev_iter =3D vfio_pci_find_by_iommufd_devid(devices[i].devid); + if (!vbasedev_iter || !vbasedev_iter->dev->realized || + vbasedev_iter->type !=3D VFIO_DEVICE_TYPE_PCI) { + continue; + } + tmp =3D container_of(vbasedev_iter, VFIOPCIDevice, vbasedev); + if (single) { + ret =3D -EINVAL; + goto out_single; + } + vfio_pci_pre_reset(tmp); + tmp->vbasedev.needs_reset =3D false; + multi =3D true; + } + + if (!single && !multi) { + ret =3D -EINVAL; + goto out_single; + } + + /* Use zero length array for hot reset with iommufd backend */ + reset =3D g_malloc0(sizeof(*reset)); + reset->argsz =3D sizeof(*reset); + + /* Bus reset! */ + ret =3D ioctl(vdev->vbasedev.fd, VFIO_DEVICE_PCI_HOT_RESET, reset); + g_free(reset); + + trace_vfio_pci_hot_reset_result(vdev->vbasedev.name, + ret ? strerror(errno) : "Success"); + + /* Re-enable INTx on affected devices */ + for (i =3D 0; i < info->count; i++) { + VFIOPCIDevice *tmp; + VFIODevice *vbasedev_iter; + + if (devices[i].devid =3D=3D vdev->vbasedev.devid || + devices[i].devid =3D=3D VFIO_PCI_DEVID_OWNED) { + continue; + } + + vbasedev_iter =3D vfio_pci_find_by_iommufd_devid(devices[i].devid); + if (!vbasedev_iter || !vbasedev_iter->dev->realized || + vbasedev_iter->type !=3D VFIO_DEVICE_TYPE_PCI) { + continue; + } + tmp =3D container_of(vbasedev_iter, VFIOPCIDevice, vbasedev); + vfio_pci_post_reset(tmp); + } +out_single: + if (!single) { + vfio_pci_post_reset(vdev); + } + g_free(info); + + return ret; +} +#endif + +static int vfio_pci_hot_reset(VFIOPCIDevice *vdev, bool single) +{ +#ifdef CONFIG_IOMMUFD + if (vdev->vbasedev.iommufd) { + return vfio_pci_hot_reset_iommufd(vdev, single); + } else +#endif + { + return vfio_pci_hot_reset_legacy(vdev, single); + } +} + + + /* * We want to differentiate hot reset of multiple in-use devices vs hot re= set * of a single in-use device. VFIO_DEVICE_RESET will already handle the c= ase diff --git a/hw/vfio/trace-events b/hw/vfio/trace-events index 60b56f23a1..c4f3b337b8 100644 --- a/hw/vfio/trace-events +++ b/hw/vfio/trace-events @@ -34,6 +34,7 @@ vfio_check_af_flr(const char *name) "%s Supports FLR via = AF cap" vfio_pci_hot_reset(const char *name, const char *type) " (%s) %s" vfio_pci_hot_reset_has_dep_devices(const char *name) "%s: hot reset depend= ent devices:" vfio_pci_hot_reset_dep_devices(int domain, int bus, int slot, int function= , int group_id) "\t%04x:%02x:%02x.%x group %d" +vfio_pci_hot_reset_dep_devices_iommufd(int domain, int bus, int slot, int = function, int dev_id) "\t%04x:%02x:%02x.%x devid %d" vfio_pci_hot_reset_result(const char *name, const char *result) "%s hot re= set: %s" vfio_populate_device_config(const char *name, unsigned long size, unsigned= long offset, unsigned long flags) "Device %s config:\n size: 0x%lx, offse= t: 0x%lx, flags: 0x%lx" vfio_populate_device_get_irq_info_failure(const char *errstr) "VFIO_DEVICE= _GET_IRQ_INFO failure: %s" --=20 2.34.1 From nobody Wed May 15 22:12:51 2024 Delivered-To: importer@patchew.org Authentication-Results: mx.zohomail.com; dkim=pass header.i=@intel.com; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass(p=none dis=none) header.from=intel.com ARC-Seal: i=1; a=rsa-sha256; t=1693392911; cv=none; d=zohomail.com; s=zohoarc; b=YGI2QWdsIiaTMvNdovaPNeB+fwlqY7uOGuRpnIl+NAEyfMLegLe6vA31XkVfP6CQ6Pa+A0JpElTMU80nTIf7P4BWVv4m3rzcnkszUIFSJ8XIIa3Y44XWqvgrGulP+eAw8zgjRsv5W7V2br/DvoMcc5Se5aOZDod1lJcdtz5MuLM= ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=zohomail.com; s=zohoarc; t=1693392911; h=Content-Transfer-Encoding:Cc:Date:From:In-Reply-To:List-Subscribe:List-Post:List-Id:List-Archive:List-Help:List-Unsubscribe:MIME-Version:Message-ID:References:Sender:Subject:To; bh=GwI8CpHQgIETqoPgkNDbVzgM8XE6ZHuaIcJq9PMLHu4=; b=V+vbLg6RkWDo7brA2d7MDBbIeGKwzvZwQi39DQG+c1hZgJg7P1BZWGqrLugiA+ZzQGhOvJBFBOTBJy76zclrNAxitgApJkq7PX3JGZyPdVft5B9Y5trrIBpA1m1046n5bqWuKMO8gTRSGO4Ad+tKpm3CX+wUiuPtUIH6T4hC/9o= ARC-Authentication-Results: i=1; mx.zohomail.com; dkim=pass header.i=@intel.com; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass header.from= (p=none dis=none) Return-Path: Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) by mx.zohomail.com with SMTPS id 1693392911290625.0944555410513; Wed, 30 Aug 2023 03:55:11 -0700 (PDT) Received: from localhost ([::1] helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1qbIqQ-0007qd-J5; Wed, 30 Aug 2023 06:54:42 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1qbIqC-00077c-RF for qemu-devel@nongnu.org; Wed, 30 Aug 2023 06:54:30 -0400 Received: from mgamail.intel.com ([192.55.52.93]) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1qbIqA-0007jr-JV for qemu-devel@nongnu.org; Wed, 30 Aug 2023 06:54:28 -0400 Received: from orsmga006.jf.intel.com ([10.7.209.51]) by fmsmga102.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 30 Aug 2023 03:54:04 -0700 Received: from duan-server-s2600bt.bj.intel.com ([10.240.192.147]) by orsmga006-auth.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 30 Aug 2023 03:54:00 -0700 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1693392866; x=1724928866; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=Jtq4e94jA0ux1suwGHaBOKa+ScqEJ7SBkRfAuQrbB4E=; b=Y2a299klQS2gt8AmJtEQZ14+uba897Im0jxi+RJ9gB/7LXeYZLpWrdH7 BfjHOF/dFbJBWq+qPu6Cl7DcPpfInreyfZaLx1AegGWu9K1VTs5pIlOei LZeJTrNKNdcS3sfD1IC87+K4+jrs5NtMYjoJHbUQ3vNS89skaTWKM1BHu aj5JE9bVntfSSb5h7RMwhNIb9vtrwXP4oN509QowORfpS6GjRDNMD1urd 0+voJjbgrL6m/Pkudqdx7SrtA0uIjTJTtN6CsMcHVbbTjG77Ti4TRy8d/ QGQyMbhNQHDrWTnFIGunEGUkzYFI5HMZkdPunlDnyluIwTqViOBD/cmsB w==; X-IronPort-AV: E=McAfee;i="6600,9927,10817"; a="373016725" X-IronPort-AV: E=Sophos;i="6.02,213,1688454000"; d="scan'208";a="373016725" X-ExtLoop1: 1 X-IronPort-AV: E=McAfee;i="6600,9927,10817"; a="715866306" X-IronPort-AV: E=Sophos;i="6.02,213,1688454000"; d="scan'208";a="715866306" From: Zhenzhong Duan To: qemu-devel@nongnu.org Cc: alex.williamson@redhat.com, clg@redhat.com, jgg@nvidia.com, nicolinc@nvidia.com, joao.m.martins@oracle.com, eric.auger@redhat.com, peterx@redhat.com, jasowang@redhat.com, kevin.tian@intel.com, yi.l.liu@intel.com, yi.y.sun@intel.com, chao.p.peng@intel.com, Zhenzhong Duan Subject: [PATCH v1 21/22] vfio/pci: Allow the selection of a given iommu backend Date: Wed, 30 Aug 2023 18:37:53 +0800 Message-Id: <20230830103754.36461-22-zhenzhong.duan@intel.com> X-Mailer: git-send-email 2.34.1 In-Reply-To: <20230830103754.36461-1-zhenzhong.duan@intel.com> References: <20230830103754.36461-1-zhenzhong.duan@intel.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Received-SPF: pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) client-ip=209.51.188.17; envelope-from=qemu-devel-bounces+importer=patchew.org@nongnu.org; helo=lists.gnu.org; Received-SPF: pass client-ip=192.55.52.93; envelope-from=zhenzhong.duan@intel.com; helo=mgamail.intel.com X-Spam_score_int: -43 X-Spam_score: -4.4 X-Spam_bar: ---- X-Spam_report: (-4.4 / 5.0 requ) BAYES_00=-1.9, DKIMWL_WL_HIGH=-0.001, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, RCVD_IN_DNSWL_MED=-2.3, SPF_HELO_NONE=0.001, SPF_PASS=-0.001 autolearn=ham autolearn_force=no X-Spam_action: no action X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: qemu-devel-bounces+importer=patchew.org@nongnu.org Sender: qemu-devel-bounces+importer=patchew.org@nongnu.org X-ZohoMail-DKIM: pass (identity @intel.com) X-ZM-MESSAGEID: 1693392912586100003 Content-Type: text/plain; charset="utf-8" From: Eric Auger Now we support two types of iommu backends, let's add the capability to select one of them. This depends on whether an iommufd object has been linked with the vfio-pci device: if the user wants to use the legacy backend, it shall not link the vfio-pci device with any iommufd object: -device vfio-pci,host=3D0000:02:00.0 This is called the legacy mode/backend. If the user wants to use the iommufd backend (/dev/iommu) it shall pass an iommufd object id in the vfio-pci device options: -object iommufd,id=3Diommufd0 -device vfio-pci,host=3D0000:02:00.0,iommufd=3Diommufd0 Note the /dev/iommu device may have been pre-opened by a management tool such as libvirt. This mode is no more considered for the legacy backend. So let's remove the "TODO" comment. Suggested-by: Alex Williamson Signed-off-by: Eric Auger Signed-off-by: Yi Liu Signed-off-by: Zhenzhong Duan --- hw/vfio/pci.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/hw/vfio/pci.c b/hw/vfio/pci.c index 3a8fee3c99..99265253f8 100644 --- a/hw/vfio/pci.c +++ b/hw/vfio/pci.c @@ -43,6 +43,7 @@ #include "migration/blocker.h" #include "migration/qemu-file.h" #include "linux/iommufd.h" +#include "sysemu/iommufd.h" =20 #define TYPE_VFIO_PCI_NOHOTPLUG "vfio-pci-nohotplug" =20 @@ -3611,11 +3612,10 @@ static Property vfio_pci_dev_properties[] =3D { qdev_prop_nv_gpudirect_clique, uint8_t), DEFINE_PROP_OFF_AUTO_PCIBAR("x-msix-relocation", VFIOPCIDevice, msix_r= elo, OFF_AUTOPCIBAR_OFF), - /* - * TODO - support passed fds... is this necessary? - * DEFINE_PROP_STRING("vfiofd", VFIOPCIDevice, vfiofd_name), - * DEFINE_PROP_STRING("vfiogroupfd, VFIOPCIDevice, vfiogroupfd_name), - */ +#ifdef CONFIG_IOMMUFD + DEFINE_PROP_LINK("iommufd", VFIOPCIDevice, vbasedev.iommufd, + TYPE_IOMMUFD_BACKEND, IOMMUFDBackend *), +#endif DEFINE_PROP_END_OF_LIST(), }; =20 --=20 2.34.1 From nobody Wed May 15 22:12:51 2024 Delivered-To: importer@patchew.org Authentication-Results: mx.zohomail.com; dkim=pass header.i=@intel.com; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass(p=none dis=none) header.from=intel.com ARC-Seal: i=1; a=rsa-sha256; t=1693392915; cv=none; d=zohomail.com; s=zohoarc; b=foc1DQWrVqMdugCouhkSvopzbAwlBtqvjValmPWAvP+YyFDb8/fkoA1pAQJGij94snDAjcd7qe83QOaLxwZonlVJ+QRNCuSwYriBaBOaU7emy0ROgGvu6rtKIe4wqEoIgpkapViXZAHX8YDabYmGE2Uk/fx1vKtV0zOk4d3fsrM= ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=zohomail.com; s=zohoarc; t=1693392915; h=Content-Transfer-Encoding:Cc:Date:From:In-Reply-To:List-Subscribe:List-Post:List-Id:List-Archive:List-Help:List-Unsubscribe:MIME-Version:Message-ID:References:Sender:Subject:To; bh=YlBcPn+SY2xi8QrMlXP4rnqFM2c5Sj8HFPnVkgKElOw=; b=bmhOi+I9YH9J58QnT5CInjj7cL/9TmqGtlwyPdlIiHIiG3IF0xJq7Nn11A7OCTlrRShtJqjWFU5PB8JaZmIAYF63Aqs+Zocu4/c/itfBk/d+hp0bh9x2wfeyUIjb8hv95ghdKkHgk3FZLO6WEVLM39bgCF+YZsraQOVXQBtjAQI= ARC-Authentication-Results: i=1; mx.zohomail.com; dkim=pass header.i=@intel.com; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass header.from= (p=none dis=none) Return-Path: Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) by mx.zohomail.com with SMTPS id 1693392915696806.3217371250386; Wed, 30 Aug 2023 03:55:15 -0700 (PDT) Received: from localhost ([::1] helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1qbIqR-0007tu-Fd; Wed, 30 Aug 2023 06:54:43 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1qbIqI-0007Ou-DY for qemu-devel@nongnu.org; Wed, 30 Aug 2023 06:54:36 -0400 Received: from mgamail.intel.com ([192.55.52.93]) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1qbIqF-0007jy-QZ for qemu-devel@nongnu.org; Wed, 30 Aug 2023 06:54:34 -0400 Received: from orsmga006.jf.intel.com ([10.7.209.51]) by fmsmga102.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 30 Aug 2023 03:54:08 -0700 Received: from duan-server-s2600bt.bj.intel.com ([10.240.192.147]) by orsmga006-auth.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 30 Aug 2023 03:54:04 -0700 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1693392871; x=1724928871; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=Fe536ojnXPqBN2KauwtAnauVAuNJwCDbrghoCi5sVtQ=; b=ePwGoogv1J4J0OSpZysST9IgLWvGl+qH1e+XIM8v4bMtLmVi/JuCq8VX 7quyLFz9D3ptc2wIMHkQchYLLV28nYfhA6qm8BCmH3FFhz88fFhNWoNQ3 nJyaJMZh64GdEruvSZ23Gs48QW8CYkOGIfoQ8ompZP+8IyovxIfGeEzOT /8qIyGJc8qSgolcGfFUhAClJPZgTmVsxRFL38Oosvg+CDJOKdgHBdIFGs iQNOCmF95n+udvNAh/b7+f1z1ASN3K+ghF/vI97MS3hAWArNs/4fvemvM kmT4zQJICFWtaIIjxkRrfrN4l39g4RDhDhWv6CreVF7dlKLG92iO0F3Hp g==; X-IronPort-AV: E=McAfee;i="6600,9927,10817"; a="373016736" X-IronPort-AV: E=Sophos;i="6.02,213,1688454000"; d="scan'208";a="373016736" X-ExtLoop1: 1 X-IronPort-AV: E=McAfee;i="6600,9927,10817"; a="715866316" X-IronPort-AV: E=Sophos;i="6.02,213,1688454000"; d="scan'208";a="715866316" From: Zhenzhong Duan To: qemu-devel@nongnu.org Cc: alex.williamson@redhat.com, clg@redhat.com, jgg@nvidia.com, nicolinc@nvidia.com, joao.m.martins@oracle.com, eric.auger@redhat.com, peterx@redhat.com, jasowang@redhat.com, kevin.tian@intel.com, yi.l.liu@intel.com, yi.y.sun@intel.com, chao.p.peng@intel.com, Zhenzhong Duan Subject: [PATCH v1 22/22] vfio/pci: Make vfio cdev pre-openable by passing a file handle Date: Wed, 30 Aug 2023 18:37:54 +0800 Message-Id: <20230830103754.36461-23-zhenzhong.duan@intel.com> X-Mailer: git-send-email 2.34.1 In-Reply-To: <20230830103754.36461-1-zhenzhong.duan@intel.com> References: <20230830103754.36461-1-zhenzhong.duan@intel.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Received-SPF: pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) client-ip=209.51.188.17; envelope-from=qemu-devel-bounces+importer=patchew.org@nongnu.org; helo=lists.gnu.org; Received-SPF: pass client-ip=192.55.52.93; envelope-from=zhenzhong.duan@intel.com; helo=mgamail.intel.com X-Spam_score_int: -43 X-Spam_score: -4.4 X-Spam_bar: ---- X-Spam_report: (-4.4 / 5.0 requ) BAYES_00=-1.9, DKIMWL_WL_HIGH=-0.001, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, RCVD_IN_DNSWL_MED=-2.3, SPF_HELO_NONE=0.001, SPF_PASS=-0.001 autolearn=ham autolearn_force=no X-Spam_action: no action X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: qemu-devel-bounces+importer=patchew.org@nongnu.org Sender: qemu-devel-bounces+importer=patchew.org@nongnu.org X-ZohoMail-DKIM: pass (identity @intel.com) X-ZM-MESSAGEID: 1693392940737100020 Content-Type: text/plain; charset="utf-8" This gives management tools like libvirt a chance to open the vfio cdev with privilege and pass FD to qemu. This way qemu never needs to have privilege to open a VFIO or iommu cdev node. Add a helper function vfio_device_get_name() to check fd and get device name, it will also be used by other vfio devices. There is no easy way to check if a device is mdev with FD passing, so fail the x-balloon-allowed check unconditionally in this case. There is also no easy way to get BDF as name with FD passing, so we fake a name by VFIO_FD[fd]. Signed-off-by: Zhenzhong Duan --- hw/vfio/helpers.c | 28 ++++++++++++++++++++++++++++ hw/vfio/iommufd.c | 12 ++++++++---- hw/vfio/pci.c | 35 ++++++++++++++++++++++++++++------- include/hw/vfio/vfio-common.h | 1 + 4 files changed, 65 insertions(+), 11 deletions(-) diff --git a/hw/vfio/helpers.c b/hw/vfio/helpers.c index 4338456b08..1a27efb075 100644 --- a/hw/vfio/helpers.c +++ b/hw/vfio/helpers.c @@ -596,3 +596,31 @@ bool vfio_has_region_cap(VFIODevice *vbasedev, int reg= ion, uint16_t cap_type) =20 return ret; } + +int vfio_device_get_name(VFIODevice *vbasedev, Error **errp) +{ + struct stat st; + + if (vbasedev->fd < 0) { + if (stat(vbasedev->sysfsdev, &st) < 0) { + error_setg_errno(errp, errno, "no such host device"); + error_prepend(errp, VFIO_MSG_PREFIX, vbasedev->sysfsdev); + return -errno; + } + vbasedev->name =3D g_path_get_basename(vbasedev->sysfsdev); + } +#ifdef CONFIG_IOMMUFD + else { + if (!vbasedev->iommufd) { + error_setg(errp, "Use FD passing only with iommufd backend"); + return -EINVAL; + } + /* + * Give a name with fd so any function printing out vbasedev->name + * will not break. + */ + vbasedev->name =3D g_strdup_printf("VFIO_FD%d", vbasedev->fd); + } +#endif + return 0; +} diff --git a/hw/vfio/iommufd.c b/hw/vfio/iommufd.c index dd24e76e39..2cd2daebf4 100644 --- a/hw/vfio/iommufd.c +++ b/hw/vfio/iommufd.c @@ -376,11 +376,15 @@ static int iommufd_attach_device(char *name, VFIODevi= ce *vbasedev, uint32_t ioas_id; Error *err =3D NULL; =20 - devfd =3D vfio_get_devicefd(vbasedev->sysfsdev, errp); - if (devfd < 0) { - return devfd; + if (vbasedev->fd < 0) { + devfd =3D vfio_get_devicefd(vbasedev->sysfsdev, errp); + if (devfd < 0) { + return devfd; + } + vbasedev->fd =3D devfd; + } else { + devfd =3D vbasedev->fd; } - vbasedev->fd =3D devfd; =20 ret =3D iommufd_connect_and_bind(vbasedev, errp); if (ret) { diff --git a/hw/vfio/pci.c b/hw/vfio/pci.c index 99265253f8..eff52b5014 100644 --- a/hw/vfio/pci.c +++ b/hw/vfio/pci.c @@ -44,6 +44,7 @@ #include "migration/qemu-file.h" #include "linux/iommufd.h" #include "sysemu/iommufd.h" +#include "monitor/monitor.h" =20 #define TYPE_VFIO_PCI_NOHOTPLUG "vfio-pci-nohotplug" =20 @@ -3171,18 +3172,23 @@ static void vfio_realize(PCIDevice *pdev, Error **e= rrp) VFIODevice *vbasedev =3D &vdev->vbasedev; char *tmp, *subsys; Error *err =3D NULL; - struct stat st; int i, ret; bool is_mdev; char uuid[UUID_FMT_LEN]; char *name; =20 - if (!vbasedev->sysfsdev) { + if (vbasedev->fd < 0 && !vbasedev->sysfsdev) { if (!(~vdev->host.domain || ~vdev->host.bus || ~vdev->host.slot || ~vdev->host.function)) { error_setg(errp, "No provided host device"); +#ifdef CONFIG_IOMMUFD + error_append_hint(errp, "Use -device vfio-pci,host=3DDDDD:BB:D= D.F, " + "-device vfio-pci,sysfsdev=3DPATH_TO_DEVICE " + "or -device vfio-pci,fd=3DDEVICE_FD\n"); +#else error_append_hint(errp, "Use -device vfio-pci,host=3DDDDD:BB:D= D.F " "or -device vfio-pci,sysfsdev=3DPATH_TO_DEVI= CE\n"); +#endif return; } vbasedev->sysfsdev =3D @@ -3191,13 +3197,9 @@ static void vfio_realize(PCIDevice *pdev, Error **er= rp) vdev->host.slot, vdev->host.function); } =20 - if (stat(vbasedev->sysfsdev, &st) < 0) { - error_setg_errno(errp, errno, "no such host device"); - error_prepend(errp, VFIO_MSG_PREFIX, vbasedev->sysfsdev); + if (vfio_device_get_name(vbasedev, errp)) { return; } - - vbasedev->name =3D g_path_get_basename(vbasedev->sysfsdev); vbasedev->ops =3D &vfio_pci_ops; vbasedev->type =3D VFIO_DEVICE_TYPE_PCI; vbasedev->dev =3D DEVICE(vdev); @@ -3559,6 +3561,7 @@ static void vfio_instance_init(Object *obj) vdev->host.bus =3D ~0U; vdev->host.slot =3D ~0U; vdev->host.function =3D ~0U; + vdev->vbasedev.fd =3D -1; =20 vdev->nv_gpudirect_clique =3D 0xFF; =20 @@ -3619,6 +3622,21 @@ static Property vfio_pci_dev_properties[] =3D { DEFINE_PROP_END_OF_LIST(), }; =20 +#ifdef CONFIG_IOMMUFD +static void vfio_pci_set_fd(Object *obj, const char *str, Error **errp) +{ + VFIOPCIDevice *vdev =3D VFIO_PCI(obj); + int fd =3D -1; + + fd =3D monitor_fd_param(monitor_cur(), str, errp); + if (fd =3D=3D -1) { + error_prepend(errp, "Could not parse remote object fd %s:", str); + return; + } + vdev->vbasedev.fd =3D fd; +} +#endif + static void vfio_pci_dev_class_init(ObjectClass *klass, void *data) { DeviceClass *dc =3D DEVICE_CLASS(klass); @@ -3626,6 +3644,9 @@ static void vfio_pci_dev_class_init(ObjectClass *klas= s, void *data) =20 dc->reset =3D vfio_pci_reset; device_class_set_props(dc, vfio_pci_dev_properties); +#ifdef CONFIG_IOMMUFD + object_class_property_add_str(klass, "fd", NULL, vfio_pci_set_fd); +#endif dc->desc =3D "VFIO-based PCI device assignment"; set_bit(DEVICE_CATEGORY_MISC, dc->categories); pdc->realize =3D vfio_realize; diff --git a/include/hw/vfio/vfio-common.h b/include/hw/vfio/vfio-common.h index 027a59a13a..41c8eeaa54 100644 --- a/include/hw/vfio/vfio-common.h +++ b/include/hw/vfio/vfio-common.h @@ -242,6 +242,7 @@ struct vfio_info_cap_header * vfio_get_device_info_cap(struct vfio_device_info *info, uint16_t id); struct vfio_info_cap_header * vfio_get_cap(void *ptr, uint32_t cap_offset, uint16_t id); +int vfio_device_get_name(VFIODevice *vbasedev, Error **errp); #endif extern const MemoryListener vfio_prereg_listener; =20 --=20 2.34.1