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 CE4C1CD98DA for ; Tue, 16 Jun 2026 09:47:09 +0000 (UTC) Received: from gabe.freedesktop.org (localhost [127.0.0.1]) by gabe.freedesktop.org (Postfix) with ESMTP id 8D48310E997; Tue, 16 Jun 2026 09:47:09 +0000 (UTC) Authentication-Results: gabe.freedesktop.org; dkim=pass (2048-bit key; unprotected) header.d=intel.com header.i=@intel.com header.b="MI5XC8Yx"; dkim-atps=neutral Received: from mgamail.intel.com (mgamail.intel.com [198.175.65.18]) by gabe.freedesktop.org (Postfix) with ESMTPS id AA32610E997 for ; Tue, 16 Jun 2026 09:47:08 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1781603229; x=1813139229; h=from:to:cc:subject:date:message-id:mime-version: content-transfer-encoding; bh=kap3X0IJsXb6W2zsimCMkX+6JlzRvWfcI73JJfTOXvo=; b=MI5XC8YxOFXwsXG/de8LtyVUlBeb/7alDvMH3y1OZIzV/eCE2lDTlfsn JOb5eYVOvWOBm8rXUn0cg8sSAGyavnHwBRHDzAQPPR7JN7NgHc48EpgQR xSKYHw3flkhin1q59TCiB7N+Z6AyAqm+xc958D5ORbbJppx5JlpfBZLZV Ac0cOKLfZks8J9YZ7ix0dN5bNP2KHi5W2T9sM2YFAJ2NPsHjaO27lj1Y3 JXU6x1xcsdiKOxpLrde2451SQf6kYkd2IJoFcAEC+Gh0Qe59W3wYPaI6f WVWrQQ/2Sc2pNXLjDyiu1KrjqO/x/13NwQ/UHEY3JInsN+opoyrGVGyrQ A==; X-CSE-ConnectionGUID: euBRRBkGTrmq0oTBzhEBSA== X-CSE-MsgGUID: G7nhNXDJSeOF1XVssOlziw== X-IronPort-AV: E=McAfee;i="6800,10657,11818"; a="82426509" X-IronPort-AV: E=Sophos;i="6.24,208,1774335600"; d="scan'208";a="82426509" Received: from orviesa004.jf.intel.com ([10.64.159.144]) by orvoesa110.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 16 Jun 2026 02:47:08 -0700 X-CSE-ConnectionGUID: 2La8JnB/TY22bJcuMHSQuQ== X-CSE-MsgGUID: 64VgnWRnQF2zMAlrDxpnzg== X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="6.24,208,1774335600"; d="scan'208";a="252031161" Received: from varungup-desk.iind.intel.com ([10.190.238.71]) by orviesa004-auth.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 16 Jun 2026 02:47:07 -0700 From: Arvind Yadav To: intel-xe@lists.freedesktop.org Cc: matthew.brost@intel.com, himal.prasad.ghimiray@intel.com, thomas.hellstrom@linux.intel.com, rodrigo.vivi@intel.com, tejas.upadhyay@intel.com Subject: [PATCH v4] drm/xe/guc: Hold device ref until queue teardown completes Date: Tue, 16 Jun 2026 15:16:55 +0530 Message-ID: <20260616094655.2511-1-arvind.yadav@intel.com> X-Mailer: git-send-email 2.43.0 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" GuC exec queue destruction can run asynchronously. During queue cleanup, xe_exec_queue_fini() may drop the last references that eventually release the DRM device and run drmm cleanup actions. guc_submit_sw_fini(), registered as a drmm action, used to drain xe->destroy_wq. If DRM device release happened from a worker on xe->destroy_wq, teardown could end up draining the same workqueue from within that worker, causing a self-deadlock. Fix this by taking a drm_device reference when the queue is created and dropping it after queue teardown completes. This prevents drmm cleanup from running while queue destruction is still pending. Since GuC queue destroy work no longer uses xe->destroy_wq, remove the stale drain from guc_submit_sw_fini(). With queue-held device refs, guc_submit_sw_fini() cannot run until all GuC IDs have been released, so replace the fini wait with an assertion and remove the now-unused fini_wq. v2: - Rebase v3: - Switch to queue-lifetime drm_dev_get()/drm_dev_put() model. (Matt) - Queue async teardown on system_dfl_wq instead of xe->destroy_wq. (Matt) - Drop separate deferred drm_dev_put worker. - Remove stale drain_workqueue(xe->destroy_wq) from guc_submit_sw_fini(). v4: - Replace the guc_submit_sw_fini() wait with an assertion and remove the now-unused fini_wq. (sashiko) Fixes: 2d2be279f1ca ("drm/xe: fix UAF around queue destruction") Cc: Matthew Brost Cc: Thomas Hellström Cc: Rodrigo Vivi Cc: Himal Prasad Ghimiray Cc: Tejas Upadhyay Signed-off-by: Arvind Yadav --- drivers/gpu/drm/xe/xe_guc_submit.c | 65 +++++++++++++++++------------- drivers/gpu/drm/xe/xe_guc_types.h | 2 - 2 files changed, 37 insertions(+), 30 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_guc_submit.c b/drivers/gpu/drm/xe/xe_guc_submit.c index afe5d99cdd8b..08ad0acf06ae 100644 --- a/drivers/gpu/drm/xe/xe_guc_submit.c +++ b/drivers/gpu/drm/xe/xe_guc_submit.c @@ -10,6 +10,7 @@ #include #include +#include #include #include "abi/guc_actions_abi.h" @@ -232,17 +233,9 @@ static bool exec_queue_killed_or_banned_or_wedged(struct xe_exec_queue *q) static void guc_submit_sw_fini(struct drm_device *drm, void *arg) { struct xe_guc *guc = arg; - struct xe_device *xe = guc_to_xe(guc); struct xe_gt *gt = guc_to_gt(guc); - int ret; - - ret = wait_event_timeout(guc->submission_state.fini_wq, - xa_empty(&guc->submission_state.exec_queue_lookup), - HZ * 5); - drain_workqueue(xe->destroy_wq); - - xe_gt_assert(gt, ret); + xe_gt_assert(gt, xa_empty(&guc->submission_state.exec_queue_lookup)); xa_destroy(&guc->submission_state.exec_queue_lookup); } @@ -319,8 +312,6 @@ int xe_guc_submit_init(struct xe_guc *guc, unsigned int num_ids) xa_init(&guc->submission_state.exec_queue_lookup); - init_waitqueue_head(&guc->submission_state.fini_wq); - primelockdep(guc); guc->submission_state.initialized = true; @@ -411,9 +402,6 @@ static void __release_guc_id(struct xe_guc *guc, struct xe_exec_queue *q, xe_guc_id_mgr_release_locked(&guc->submission_state.idm, q->guc->id, q->width); - if (xa_empty(&guc->submission_state.exec_queue_lookup)) - wake_up(&guc->submission_state.fini_wq); - mutex_unlock(&guc->submission_state.lock); } @@ -1685,6 +1673,7 @@ static void guc_exec_queue_fini(struct xe_exec_queue *q) { struct xe_guc_exec_queue *ge = q->guc; struct xe_guc *guc = exec_queue_to_guc(q); + struct drm_device *drm = &guc_to_xe(guc)->drm; if (xe_exec_queue_is_multi_queue_secondary(q)) { struct xe_exec_queue_group *group = q->multi_queue.group; @@ -1703,36 +1692,52 @@ static void guc_exec_queue_fini(struct xe_exec_queue *q) * (timeline name). */ kfree_rcu(ge, rcu); + + drm_dev_put(drm); } -static void __guc_exec_queue_destroy_async(struct work_struct *w) +static void guc_exec_queue_do_destroy(struct xe_exec_queue *q) { - struct xe_guc_exec_queue *ge = - container_of(w, struct xe_guc_exec_queue, destroy_async); - struct xe_exec_queue *q = ge->q; + struct xe_guc_exec_queue *ge = q->guc; struct xe_guc *guc = exec_queue_to_guc(q); + struct xe_device *xe = guc_to_xe(guc); + struct drm_device *drm = &xe->drm; - guard(xe_pm_runtime)(guc_to_xe(guc)); - trace_xe_exec_queue_destroy(q); + /* + * guc_exec_queue_fini() drops the queue's drm_device ref. + * Keep the device alive until the PM-runtime guard unwinds. + */ + drm_dev_get(drm); - /* Confirm no work left behind accessing device structures */ - cancel_delayed_work_sync(&ge->sched.base.work_tdr); + scoped_guard(xe_pm_runtime, xe) { + trace_xe_exec_queue_destroy(q); - xe_exec_queue_fini(q); + /* Confirm no work left behind accessing device structures */ + cancel_delayed_work_sync(&ge->sched.base.work_tdr); + + xe_exec_queue_fini(q); + } + + drm_dev_put(drm); } -static void guc_exec_queue_destroy_async(struct xe_exec_queue *q) +static void __guc_exec_queue_destroy_async(struct work_struct *w) { - struct xe_guc *guc = exec_queue_to_guc(q); - struct xe_device *xe = guc_to_xe(guc); + struct xe_guc_exec_queue *ge = + container_of(w, struct xe_guc_exec_queue, destroy_async); + + guc_exec_queue_do_destroy(ge->q); +} +static void guc_exec_queue_destroy_async(struct xe_exec_queue *q) +{ INIT_WORK(&q->guc->destroy_async, __guc_exec_queue_destroy_async); /* We must block on kernel engines so slabs are empty on driver unload */ if (q->flags & EXEC_QUEUE_FLAG_PERMANENT || exec_queue_wedged(q)) - __guc_exec_queue_destroy_async(&q->guc->destroy_async); + guc_exec_queue_do_destroy(q); else - queue_work(xe->destroy_wq, &q->guc->destroy_async); + queue_work(system_dfl_wq, &q->guc->destroy_async); } static void __guc_exec_queue_destroy(struct xe_guc *guc, struct xe_exec_queue *q) @@ -1927,6 +1932,7 @@ static int guc_exec_queue_init(struct xe_exec_queue *q) { struct xe_gpu_scheduler *sched; struct xe_guc *guc = exec_queue_to_guc(q); + struct drm_device *drm = &guc_to_xe(guc)->drm; struct workqueue_struct *submit_wq = NULL; struct xe_guc_exec_queue *ge; long timeout; @@ -1938,6 +1944,8 @@ static int guc_exec_queue_init(struct xe_exec_queue *q) if (!ge) return -ENOMEM; + drm_dev_get(drm); + q->guc = ge; ge->q = q; init_rcu_head(&ge->rcu); @@ -2014,6 +2022,7 @@ static int guc_exec_queue_init(struct xe_exec_queue *q) release_guc_id(guc, q); err_free: kfree(ge); + drm_dev_put(drm); return err; } diff --git a/drivers/gpu/drm/xe/xe_guc_types.h b/drivers/gpu/drm/xe/xe_guc_types.h index c7b9642b41ba..31a2acb63ac3 100644 --- a/drivers/gpu/drm/xe/xe_guc_types.h +++ b/drivers/gpu/drm/xe/xe_guc_types.h @@ -100,8 +100,6 @@ struct xe_guc { * even initialized - before that not even the lock is valid */ bool initialized; - /** @submission_state.fini_wq: submit fini wait queue */ - wait_queue_head_t fini_wq; } submission_state; /** @hwconfig: Hardware config state */ -- 2.43.0