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 948F7C3DA49 for ; Thu, 18 Jul 2024 17:12:16 +0000 (UTC) Received: from gabe.freedesktop.org (localhost [127.0.0.1]) by gabe.freedesktop.org (Postfix) with ESMTP id 53F5510EA73; Thu, 18 Jul 2024 17:12:16 +0000 (UTC) Authentication-Results: gabe.freedesktop.org; dkim=pass (2048-bit key; unprotected) header.d=intel.com header.i=@intel.com header.b="IkYxU2ix"; dkim-atps=neutral Received: from mgamail.intel.com (mgamail.intel.com [198.175.65.9]) by gabe.freedesktop.org (Postfix) with ESMTPS id 7129510E16B for ; Thu, 18 Jul 2024 17:12:14 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1721322735; x=1752858735; h=message-id:date:mime-version:subject:to:cc:references: from:in-reply-to:content-transfer-encoding; bh=E0R68w84wvSUXwelpgyCPe6PKw8jrvzj9YwO30ZHxh8=; b=IkYxU2ixPM/unbLGDzGtO+iiO4yf8isAUWQmnRXfiOKK1x2tDT/NY2HP YuW6zxgllf/DvOCOMQqgvIa/a77jvqytpNOX43sb+t6tDr0+7bC4KYNSg rO72r7FhOOqHKveF3hLP7Vf6AhhFMRh64u6GKb72SorCU9A09xUraLry4 uZtE1fox3z07WUbqGztVWAZqY3QbiB/WjNLtnUmW1PMkDx5pCOLK96afL zTvxc9qoe5XHUmhv8iunieJOpo/H7D6D1uivBK8JWyeVBUhFmqKyPhby9 wb7ywNZGJHIVVORXKD0lXQNGCS/T6PbE9zTcfXXkyvfRB8pOK/CW00Y0A Q==; X-CSE-ConnectionGUID: 1uAxGqnlRk+0zryDgQqhLQ== X-CSE-MsgGUID: xJBdfiKZR2G7yUWWUdgxdQ== X-IronPort-AV: E=McAfee;i="6700,10204,11137"; a="41428537" X-IronPort-AV: E=Sophos;i="6.09,218,1716274800"; d="scan'208";a="41428537" Received: from orviesa005.jf.intel.com ([10.64.159.145]) by orvoesa101.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 18 Jul 2024 10:12:14 -0700 X-CSE-ConnectionGUID: A/vLdppQSlKAiyvpf3Arsg== X-CSE-MsgGUID: 9yCpLAgPQS+jUac9E5oQ9Q== X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="6.09,218,1716274800"; d="scan'208";a="55692832" Received: from oandoniu-mobl3.ger.corp.intel.com (HELO [10.245.244.207]) ([10.245.244.207]) by orviesa005-auth.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 18 Jul 2024 10:12:12 -0700 Message-ID: Date: Thu, 18 Jul 2024 18:12:10 +0100 MIME-Version: 1.0 User-Agent: Mozilla Thunderbird Subject: Re: [PATCH v2 1/1] drm/xe: Build PM into GuC CT layer To: Matthew Brost , intel-xe@lists.freedesktop.org Cc: nirmoy.das@intel.com, rodrigo.vivi@intel.com References: <20240718155245.1481788-1-matthew.brost@intel.com> <20240718155245.1481788-2-matthew.brost@intel.com> Content-Language: en-GB From: Matthew Auld In-Reply-To: <20240718155245.1481788-2-matthew.brost@intel.com> Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit 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" On 18/07/2024 16:52, Matthew Brost wrote: > Take PM ref when a G2H is outstanding, drop when it is received. > > To safely ensure we have PM ref when in the GuC CT layer, a PM ref needs > to be held when scheduler messages are pending too. > > v2: > - Add outer PM protections to xe_file_close (CI) > > Cc: Rodrigo Vivi > Cc: Nirmoy Das > Signed-off-by: Matthew Brost > --- > drivers/gpu/drm/xe/xe_device.c | 4 ++++ > drivers/gpu/drm/xe/xe_guc_ct.c | 5 +++++ > drivers/gpu/drm/xe/xe_guc_submit.c | 4 ++++ > 3 files changed, 13 insertions(+) > > diff --git a/drivers/gpu/drm/xe/xe_device.c b/drivers/gpu/drm/xe/xe_device.c > index 06cebaffb451..b68ab474e1a0 100644 > --- a/drivers/gpu/drm/xe/xe_device.c > +++ b/drivers/gpu/drm/xe/xe_device.c > @@ -101,6 +101,8 @@ static void xe_file_close(struct drm_device *dev, struct drm_file *file) > struct xe_exec_queue *q; > unsigned long idx; > > + xe_pm_runtime_get(xe); > + > /* > * No need for exec_queue.lock here as there is no contention for it > * when FD is closing as IOCTLs presumably can't be modifying the > @@ -126,6 +128,8 @@ static void xe_file_close(struct drm_device *dev, struct drm_file *file) > > xe_drm_client_put(xef->client); > kfree(xef); > + > + xe_pm_runtime_put(xe); > } > > static const struct drm_ioctl_desc xe_ioctls[] = { > diff --git a/drivers/gpu/drm/xe/xe_guc_ct.c b/drivers/gpu/drm/xe/xe_guc_ct.c > index 7d2e937da1d8..560f7e4bf290 100644 > --- a/drivers/gpu/drm/xe/xe_guc_ct.c > +++ b/drivers/gpu/drm/xe/xe_guc_ct.c > @@ -327,6 +327,8 @@ static void xe_guc_ct_set_state(struct xe_guc_ct *ct, > xe_gt_assert(ct_to_gt(ct), ct->g2h_outstanding == 0 || > state == XE_GUC_CT_STATE_STOPPED); > > + while (ct->g2h_outstanding--) > + xe_pm_runtime_put(ct_to_xe(ct)); > ct->g2h_outstanding = 0; > ct->state = state; > > @@ -501,6 +503,8 @@ static void __g2h_reserve_space(struct xe_guc_ct *ct, u32 g2h_len, u32 num_g2h) > > ct->ctbs.g2h.info.space -= g2h_len; > ct->g2h_outstanding += num_g2h; > + while (num_g2h--) > + xe_pm_runtime_get_noresume(ct_to_xe(ct)); Could potentially just grab on the 0->1 and only release on the 1->0? There is a device level lock being grabbed in the rpm code on every call, but not sure how big of a deal that is here. > } > } > > @@ -512,6 +516,7 @@ static void __g2h_release_space(struct xe_guc_ct *ct, u32 g2h_len) > > ct->ctbs.g2h.info.space += g2h_len; > --ct->g2h_outstanding; > + xe_pm_runtime_put(ct_to_xe(ct)); > } > > static void g2h_release_space(struct xe_guc_ct *ct, u32 g2h_len) > diff --git a/drivers/gpu/drm/xe/xe_guc_submit.c b/drivers/gpu/drm/xe/xe_guc_submit.c > index 860405527115..993d0344dc88 100644 > --- a/drivers/gpu/drm/xe/xe_guc_submit.c > +++ b/drivers/gpu/drm/xe/xe_guc_submit.c > @@ -1402,6 +1402,8 @@ static void guc_exec_queue_process_msg(struct xe_sched_msg *msg) > default: > XE_WARN_ON("Unknown message type"); > } > + > + xe_pm_runtime_put(guc_to_xe(exec_queue_to_guc(msg->private_data))); > } > > static const struct drm_sched_backend_ops drm_sched_ops = { > @@ -1492,6 +1494,8 @@ static void guc_exec_queue_kill(struct xe_exec_queue *q) > static void guc_exec_queue_add_msg(struct xe_exec_queue *q, struct xe_sched_msg *msg, > u32 opcode) > { > + xe_pm_runtime_get_noresume(guc_to_xe(exec_queue_to_guc(q))); > + > INIT_LIST_HEAD(&msg->link); > msg->opcode = opcode; > msg->private_data = q;