From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from gabe.freedesktop.org (gabe.freedesktop.org [131.252.210.177]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.lore.kernel.org (Postfix) with ESMTPS id BD153CF885A for ; Thu, 20 Nov 2025 13:27:50 +0000 (UTC) Received: from gabe.freedesktop.org (localhost [127.0.0.1]) by gabe.freedesktop.org (Postfix) with ESMTP id 7D80410E29D; Thu, 20 Nov 2025 13:27:50 +0000 (UTC) Authentication-Results: gabe.freedesktop.org; dkim=pass (2048-bit key; unprotected) header.d=intel.com header.i=@intel.com header.b="m+mq2dpO"; dkim-atps=neutral Received: from mgamail.intel.com (mgamail.intel.com [192.198.163.8]) by gabe.freedesktop.org (Postfix) with ESMTPS id 8364210E29D for ; Thu, 20 Nov 2025 13:27:44 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1763645264; x=1795181264; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=IcKEET7hBR4ODS94EV/iVpUY2cnqLukR7BOxCGCitks=; b=m+mq2dpOoqSEgBVemyptnUmBy2Qd404sxMaJLaEIG3q28DuXjvESuhCj BJXE6EBBm2J87/m0ZgCRMQao8/a03ms2t6qBh5K26aLNSMKAIkKX7xzS4 p5Bl7mldan6sUD9mbaX0IdUaXzwRYb32ev1P6H03AykjeSL93ypIqJ6gc lOvcO1GUfeV+GfH9Av9ZZoj0NmQ/QeGCp2jKiDLP4inZGRSnLxJUjl39s Q/anwGM2wfeIfTM9TKozLC9dHmqY2rbv9Y7MpzUyTxxgaeRtp3IeiED5V mUDbU4RW/K4+S5iJpuHfcjjcU4fxubzVP1H6mMNHHkfEK8bW+rqD8wi4u A==; X-CSE-ConnectionGUID: Vq6YsOy1Q6qkFnPNM50Plw== X-CSE-MsgGUID: ZuZKhG6SSNmReJ0b970RGA== X-IronPort-AV: E=McAfee;i="6800,10657,11618"; a="83335115" X-IronPort-AV: E=Sophos;i="6.20,213,1758610800"; d="scan'208";a="83335115" Received: from orviesa008.jf.intel.com ([10.64.159.148]) by fmvoesa102.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 20 Nov 2025 05:27:44 -0800 X-CSE-ConnectionGUID: 4MtpMdqXSoye52dT69hNqg== X-CSE-MsgGUID: riLOElLyQECSPrr7CZdSfw== X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="6.20,213,1758610800"; d="scan'208";a="191485756" Received: from egrumbac-mobl6.ger.corp.intel.com (HELO mwauld-desk.intel.com) ([10.245.244.162]) by orviesa008-auth.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 20 Nov 2025 05:27:43 -0800 From: Matthew Auld To: intel-xe@lists.freedesktop.org Cc: =?UTF-8?q?Thomas=20Hellstr=C3=B6m?= , =?UTF-8?q?Jos=C3=A9=20Roberto=20de=20Souza?= , Matthew Brost , Michal Mrozek , Carl Zhang , stable@vger.kernel.org Subject: [PATCH v3 1/2] drm/xe/uapi: disallow bind queue sharing Date: Thu, 20 Nov 2025 13:27:29 +0000 Message-ID: <20251120132727.575986-5-matthew.auld@intel.com> X-Mailer: git-send-email 2.51.1 In-Reply-To: <20251120132727.575986-4-matthew.auld@intel.com> References: <20251120132727.575986-4-matthew.auld@intel.com> MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit X-BeenThere: intel-xe@lists.freedesktop.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: Intel Xe graphics driver List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: intel-xe-bounces@lists.freedesktop.org Sender: "Intel-xe" Currently this is very broken if someone attempts to create a bind queue and share it across multiple VMs. For example currently we assume it is safe to acquire the user VM lock to protect some of the bind queue state, but if allow sharing the bind queue with multiple VMs then this quickly breaks down. To fix this reject using a bind queue with any VM that is not the same VM that was originally passed when creating the bind queue. This a uAPI change, however this was more of an oversight on kernel side that we didn't reject this, and expectation is that userspace shouldn't be using bind queues in this way, so in theory this change should go unnoticed. Based on a patch from Matt Brost. v2 (Matt B): - Hold the vm lock over queue create, to ensure it can't be closed as we attach the user_vm to the queue. - Make sure we actually check for NULL user_vm in destruction path. v3: - Fix error path handling. Fixes: dd08ebf6c352 ("drm/xe: Introduce a new DRM driver for Intel GPUs") Reported-by: Thomas Hellström Signed-off-by: Matthew Auld Cc: José Roberto de Souza Cc: Matthew Brost Cc: Michal Mrozek Cc: Carl Zhang Cc: # v6.8+ Acked-by: José Roberto de Souza --- drivers/gpu/drm/xe/xe_exec_queue.c | 32 +++++++++++++++++++++++- drivers/gpu/drm/xe/xe_exec_queue.h | 1 + drivers/gpu/drm/xe/xe_exec_queue_types.h | 6 +++++ drivers/gpu/drm/xe/xe_sriov_vf_ccs.c | 2 +- drivers/gpu/drm/xe/xe_vm.c | 7 +++++- 5 files changed, 45 insertions(+), 3 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_exec_queue.c b/drivers/gpu/drm/xe/xe_exec_queue.c index 8724f8de67e2..779d7e7e2d2e 100644 --- a/drivers/gpu/drm/xe/xe_exec_queue.c +++ b/drivers/gpu/drm/xe/xe_exec_queue.c @@ -328,6 +328,7 @@ struct xe_exec_queue *xe_exec_queue_create_class(struct xe_device *xe, struct xe * @xe: Xe device. * @tile: tile which bind exec queue belongs to. * @flags: exec queue creation flags + * @user_vm: The user VM which this exec queue belongs to * @extensions: exec queue creation extensions * * Normalize bind exec queue creation. Bind exec queue is tied to migration VM @@ -341,6 +342,7 @@ struct xe_exec_queue *xe_exec_queue_create_class(struct xe_device *xe, struct xe */ struct xe_exec_queue *xe_exec_queue_create_bind(struct xe_device *xe, struct xe_tile *tile, + struct xe_vm *user_vm, u32 flags, u64 extensions) { struct xe_gt *gt = tile->primary_gt; @@ -377,6 +379,9 @@ struct xe_exec_queue *xe_exec_queue_create_bind(struct xe_device *xe, xe_exec_queue_put(q); return ERR_PTR(err); } + + if (user_vm) + q->user_vm = xe_vm_get(user_vm); } return q; @@ -407,6 +412,11 @@ void xe_exec_queue_destroy(struct kref *ref) xe_exec_queue_put(eq); } + if (q->user_vm) { + xe_vm_put(q->user_vm); + q->user_vm = NULL; + } + q->ops->destroy(q); } @@ -742,6 +752,22 @@ int xe_exec_queue_create_ioctl(struct drm_device *dev, void *data, XE_IOCTL_DBG(xe, eci[0].engine_instance != 0)) return -EINVAL; + vm = xe_vm_lookup(xef, args->vm_id); + if (XE_IOCTL_DBG(xe, !vm)) + return -ENOENT; + + err = down_read_interruptible(&vm->lock); + if (err) { + xe_vm_put(vm); + return err; + } + + if (XE_IOCTL_DBG(xe, xe_vm_is_closed_or_banned(vm))) { + up_read(&vm->lock); + xe_vm_put(vm); + return -ENOENT; + } + for_each_tile(tile, xe, id) { struct xe_exec_queue *new; @@ -749,9 +775,11 @@ int xe_exec_queue_create_ioctl(struct drm_device *dev, void *data, if (id) flags |= EXEC_QUEUE_FLAG_BIND_ENGINE_CHILD; - new = xe_exec_queue_create_bind(xe, tile, flags, + new = xe_exec_queue_create_bind(xe, tile, vm, flags, args->extensions); if (IS_ERR(new)) { + up_read(&vm->lock); + xe_vm_put(vm); err = PTR_ERR(new); if (q) goto put_exec_queue; @@ -763,6 +791,8 @@ int xe_exec_queue_create_ioctl(struct drm_device *dev, void *data, list_add_tail(&new->multi_gt_list, &q->multi_gt_link); } + up_read(&vm->lock); + xe_vm_put(vm); } else { logical_mask = calc_validate_logical_mask(xe, eci, args->width, diff --git a/drivers/gpu/drm/xe/xe_exec_queue.h b/drivers/gpu/drm/xe/xe_exec_queue.h index fda4d4f9bda8..37a9da22f420 100644 --- a/drivers/gpu/drm/xe/xe_exec_queue.h +++ b/drivers/gpu/drm/xe/xe_exec_queue.h @@ -28,6 +28,7 @@ struct xe_exec_queue *xe_exec_queue_create_class(struct xe_device *xe, struct xe u32 flags, u64 extensions); struct xe_exec_queue *xe_exec_queue_create_bind(struct xe_device *xe, struct xe_tile *tile, + struct xe_vm *user_vm, u32 flags, u64 extensions); void xe_exec_queue_fini(struct xe_exec_queue *q); diff --git a/drivers/gpu/drm/xe/xe_exec_queue_types.h b/drivers/gpu/drm/xe/xe_exec_queue_types.h index 771ffe35cd0c..3a4263c92b3d 100644 --- a/drivers/gpu/drm/xe/xe_exec_queue_types.h +++ b/drivers/gpu/drm/xe/xe_exec_queue_types.h @@ -54,6 +54,12 @@ struct xe_exec_queue { struct kref refcount; /** @vm: VM (address space) for this exec queue */ struct xe_vm *vm; + /** + * @user_vm: User VM (address space) for this exec queue (bind queues + * only) + */ + struct xe_vm *user_vm; + /** @class: class of this exec queue */ enum xe_engine_class class; /** diff --git a/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c b/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c index 052a5071e69f..db023fb66a27 100644 --- a/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c +++ b/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c @@ -350,7 +350,7 @@ int xe_sriov_vf_ccs_init(struct xe_device *xe) flags = EXEC_QUEUE_FLAG_KERNEL | EXEC_QUEUE_FLAG_PERMANENT | EXEC_QUEUE_FLAG_MIGRATE; - q = xe_exec_queue_create_bind(xe, tile, flags, 0); + q = xe_exec_queue_create_bind(xe, tile, NULL, flags, 0); if (IS_ERR(q)) { err = PTR_ERR(q); goto err_ret; diff --git a/drivers/gpu/drm/xe/xe_vm.c b/drivers/gpu/drm/xe/xe_vm.c index f9989a7a710c..7973d654540a 100644 --- a/drivers/gpu/drm/xe/xe_vm.c +++ b/drivers/gpu/drm/xe/xe_vm.c @@ -1614,7 +1614,7 @@ struct xe_vm *xe_vm_create(struct xe_device *xe, u32 flags, struct xe_file *xef) if (!vm->pt_root[id]) continue; - q = xe_exec_queue_create_bind(xe, tile, create_flags, 0); + q = xe_exec_queue_create_bind(xe, tile, vm, create_flags, 0); if (IS_ERR(q)) { err = PTR_ERR(q); goto err_close; @@ -3571,6 +3571,11 @@ int xe_vm_bind_ioctl(struct drm_device *dev, void *data, struct drm_file *file) } } + if (XE_IOCTL_DBG(xe, q && vm != q->user_vm)) { + err = -EINVAL; + goto put_exec_queue; + } + /* Ensure all UNMAPs visible */ xe_svm_flush(vm); -- 2.51.1