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 8370DC7EE22 for ; Tue, 9 May 2023 15:21:48 +0000 (UTC) Received: from gabe.freedesktop.org (localhost [127.0.0.1]) by gabe.freedesktop.org (Postfix) with ESMTP id 586C710E169; Tue, 9 May 2023 15:21:48 +0000 (UTC) Received: from mga17.intel.com (mga17.intel.com [192.55.52.151]) by gabe.freedesktop.org (Postfix) with ESMTPS id 28B2D10E169 for ; Tue, 9 May 2023 15:21:45 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1683645706; x=1715181706; h=message-id:date:mime-version:subject:to:cc:references: from:in-reply-to:content-transfer-encoding; bh=AXS7rEwKjhzQ3m8K5L7ikZw2ZxL4iERrn+9FrhlYr2M=; b=g/31rlTKwQCjXAB5/VRrWWBBVgQP/aK/MSkpK2sE0f/rlWXQR1k1zE8b boP65Avu9Dxidxp4f81zQO+iIS+CnF1ED61t0KGyn2ZCu40cHOlxU0w5D 9XqZc/DfRp6FEz9e1VQepo6+AxQK15Bt8yviSL+817WXvbf0PJrBBZdsd fvYaCRwxuANfvuUjUWIDegsd7GbZCIdfVuCTZV4qQtEIEGe6qmpdkAHT/ 7SuyYqWe2xx6s/hGX1CAks4XekA2RucTu9rNHI4kZCkR2VXsozCH/a5cj H/dh+KLIz72ENuPLhAlXYMUHlYtCsCGk12iClxFLfc3IynHwSjXgsK7GK Q==; X-IronPort-AV: E=McAfee;i="6600,9927,10705"; a="330323651" X-IronPort-AV: E=Sophos;i="5.99,262,1677571200"; d="scan'208";a="330323651" Received: from fmsmga001.fm.intel.com ([10.253.24.23]) by fmsmga107.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 09 May 2023 08:21:44 -0700 X-ExtLoop1: 1 X-IronPort-AV: E=McAfee;i="6600,9927,10705"; a="843152913" X-IronPort-AV: E=Sophos;i="5.99,262,1677571200"; d="scan'208";a="843152913" Received: from jparkkin-mobl.ger.corp.intel.com (HELO [10.249.254.236]) ([10.249.254.236]) by fmsmga001-auth.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 09 May 2023 08:21:43 -0700 Message-ID: Date: Tue, 9 May 2023 17:21:39 +0200 MIME-Version: 1.0 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:102.0) Gecko/20100101 Thunderbird/102.8.0 Content-Language: en-US To: Matthew Brost References: <20230502001727.3211096-1-matthew.brost@intel.com> <20230502001727.3211096-6-matthew.brost@intel.com> <90b4e835-c21a-f9b3-4b22-aa59c911f54f@linux.intel.com> From: =?UTF-8?Q?Thomas_Hellstr=c3=b6m?= In-Reply-To: Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 8bit Subject: Re: [Intel-xe] [PATCH v2 05/31] drm/xe: Long running job update 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: , Cc: intel-xe@lists.freedesktop.org Errors-To: intel-xe-bounces@lists.freedesktop.org Sender: "Intel-xe" On 5/9/23 16:56, Matthew Brost wrote: > On Mon, May 08, 2023 at 03:14:10PM +0200, Thomas Hellström wrote: >> Hi, Matthew >> >> In addition to Rodrigo's comments: >> >> On 5/2/23 02:17, Matthew Brost wrote: >>> Flow control + write ring in exec, return NULL in run_job, siganl >>> xe_hw_fence immediately, and override TDR for LR jobs. >>> >>> Signed-off-by: Matthew Brost >>> --- >>> drivers/gpu/drm/xe/xe_engine.c | 32 ++++++++ >>> drivers/gpu/drm/xe/xe_engine.h | 4 + >>> drivers/gpu/drm/xe/xe_exec.c | 8 ++ >>> drivers/gpu/drm/xe/xe_guc_engine_types.h | 2 + >>> drivers/gpu/drm/xe/xe_guc_submit.c | 95 +++++++++++++++++++++--- >>> drivers/gpu/drm/xe/xe_trace.h | 5 ++ >>> 6 files changed, 137 insertions(+), 9 deletions(-) >>> >>> diff --git a/drivers/gpu/drm/xe/xe_engine.c b/drivers/gpu/drm/xe/xe_engine.c >>> index 094ec17d3004..d1e84d7adbd4 100644 >>> --- a/drivers/gpu/drm/xe/xe_engine.c >>> +++ b/drivers/gpu/drm/xe/xe_engine.c >>> @@ -18,6 +18,7 @@ >>> #include "xe_macros.h" >>> #include "xe_migrate.h" >>> #include "xe_pm.h" >>> +#include "xe_ring_ops_types.h" >>> #include "xe_trace.h" >>> #include "xe_vm.h" >>> @@ -673,6 +674,37 @@ static void engine_kill_compute(struct xe_engine *e) >>> up_write(&e->vm->lock); >>> } >>> +/** >>> + * xe_engine_is_lr() - Whether an engine is long-running >>> + * @e: The engine >>> + * >>> + * Return: True if the engine is long-running, false otherwise. >>> + */ >>> +bool xe_engine_is_lr(struct xe_engine *e) >>> +{ >>> + return e->vm && xe_vm_no_dma_fences(e->vm) && >>> + !(e->flags & ENGINE_FLAG_VM); >>> +} >>> + >>> +static s32 xe_engine_num_job_inflight(struct xe_engine *e) >>> +{ >>> + return e->lrc->fence_ctx.next_seqno - xe_lrc_seqno(e->lrc) - 1; >>> +} >>> + >>> +/** >>> + * xe_engine_ring_full() - Whether an engine's ring is full >>> + * @e: The engine >>> + * >>> + * Return: True if the engine's ring is full, false otherwise. >>> + */ >>> +bool xe_engine_ring_full(struct xe_engine *e) >>> +{ >>> + struct xe_lrc *lrc = e->lrc; >>> + s32 max_job = lrc->ring.size / MAX_JOB_SIZE_BYTES; >>> + >>> + return xe_engine_num_job_inflight(e) >= max_job; >>> +} >>> + >>> /** >>> * xe_engine_is_idle() - Whether an engine is idle. >>> * @engine: The engine >>> diff --git a/drivers/gpu/drm/xe/xe_engine.h b/drivers/gpu/drm/xe/xe_engine.h >>> index a49cf2ab405e..2e60f6d90226 100644 >>> --- a/drivers/gpu/drm/xe/xe_engine.h >>> +++ b/drivers/gpu/drm/xe/xe_engine.h >>> @@ -42,6 +42,10 @@ static inline bool xe_engine_is_parallel(struct xe_engine *engine) >>> return engine->width > 1; >>> } >>> +bool xe_engine_is_lr(struct xe_engine *e); >>> + >>> +bool xe_engine_ring_full(struct xe_engine *e); >>> + >>> bool xe_engine_is_idle(struct xe_engine *engine); >>> void xe_engine_kill(struct xe_engine *e); >>> diff --git a/drivers/gpu/drm/xe/xe_exec.c b/drivers/gpu/drm/xe/xe_exec.c >>> index ea869f2452ef..44ea9bcd0066 100644 >>> --- a/drivers/gpu/drm/xe/xe_exec.c >>> +++ b/drivers/gpu/drm/xe/xe_exec.c >>> @@ -13,6 +13,7 @@ >>> #include "xe_device.h" >>> #include "xe_engine.h" >>> #include "xe_macros.h" >>> +#include "xe_ring_ops_types.h" >>> #include "xe_sched_job.h" >>> #include "xe_sync.h" >>> #include "xe_vm.h" >>> @@ -277,6 +278,11 @@ int xe_exec_ioctl(struct drm_device *dev, void *data, struct drm_file *file) >>> goto err_engine_end; >>> } >>> + if (xe_engine_is_lr(engine) && xe_engine_ring_full(engine)) { >>> + err = -EWOULDBLOCK; >>> + goto err_engine_end; >>> + } >>> + >>> job = xe_sched_job_create(engine, xe_engine_is_parallel(engine) ? >>> addresses : &args->address); >>> if (IS_ERR(job)) { >>> @@ -363,6 +369,8 @@ int xe_exec_ioctl(struct drm_device *dev, void *data, struct drm_file *file) >>> xe_sync_entry_signal(&syncs[i], job, >>> &job->drm.s_fence->finished); >>> + if (xe_engine_is_lr(engine)) >>> + engine->ring_ops->emit_job(job); >>> xe_sched_job_push(job); >>> xe_vm_reactivate_rebind(vm); >>> diff --git a/drivers/gpu/drm/xe/xe_guc_engine_types.h b/drivers/gpu/drm/xe/xe_guc_engine_types.h >>> index cbfb13026ec1..5d83132034a6 100644 >>> --- a/drivers/gpu/drm/xe/xe_guc_engine_types.h >>> +++ b/drivers/gpu/drm/xe/xe_guc_engine_types.h >>> @@ -31,6 +31,8 @@ struct xe_guc_engine { >>> */ >>> #define MAX_STATIC_MSG_TYPE 3 >>> struct drm_sched_msg static_msgs[MAX_STATIC_MSG_TYPE]; >>> + /** @lr_tdr: long running TDR worker */ >>> + struct work_struct lr_tdr; >>> /** @fini_async: do final fini async from this worker */ >>> struct work_struct fini_async; >>> /** @resume_time: time of last resume */ >>> diff --git a/drivers/gpu/drm/xe/xe_guc_submit.c b/drivers/gpu/drm/xe/xe_guc_submit.c >>> index 68d09e7a4cc0..0a41f5d04f6d 100644 >>> --- a/drivers/gpu/drm/xe/xe_guc_submit.c >>> +++ b/drivers/gpu/drm/xe/xe_guc_submit.c >>> @@ -500,6 +500,14 @@ static void register_engine(struct xe_engine *e) >>> parallel_write(xe, map, wq_desc.wq_status, WQ_STATUS_ACTIVE); >>> } >>> + /* >>> + * We must keep a reference for LR engines if engine is registered with >>> + * the GuC as jobs signal immediately and can't destroy an engine if the >>> + * GuC has a reference to it. >>> + */ >>> + if (xe_engine_is_lr(e)) >>> + xe_engine_get(e); >>> + >>> set_engine_registered(e); >>> trace_xe_engine_register(e); >>> if (xe_engine_is_parallel(e)) >>> @@ -662,6 +670,7 @@ guc_engine_run_job(struct drm_sched_job *drm_job) >>> { >>> struct xe_sched_job *job = to_xe_sched_job(drm_job); >>> struct xe_engine *e = job->engine; >>> + bool lr = xe_engine_is_lr(e); >>> XE_BUG_ON((engine_destroyed(e) || engine_pending_disable(e)) && >>> !engine_banned(e) && !engine_suspended(e)); >>> @@ -671,14 +680,19 @@ guc_engine_run_job(struct drm_sched_job *drm_job) >>> if (!engine_killed_or_banned(e) && !xe_sched_job_is_error(job)) { >>> if (!engine_registered(e)) >>> register_engine(e); >>> - e->ring_ops->emit_job(job); >>> + if (!lr) /* Written in IOCTL */ >> Hmm? What does "Written in IOCTL mean?" Could you rephrase to something more >> descriptive? >> > "LR jos are emitted in the IOCTL" Ah, I read it as "the lr variable was written in IOCTL." Perhaps LR jobs are already emitted at execbuf time? /Thomas > > Does that work? > > Matt > >>> + e->ring_ops->emit_job(job); >>> submit_engine(e); >>> } >>> - if (test_and_set_bit(JOB_FLAG_SUBMIT, &job->fence->flags)) >>> + if (lr) { >>> + xe_sched_job_set_error(job, -ENOTSUPP); >>> + return NULL; >>> + } else if (test_and_set_bit(JOB_FLAG_SUBMIT, &job->fence->flags)) { >>> return job->fence; >>> - else >>> + } else { >>> return dma_fence_get(job->fence); >>> + } >>> } >>> static void guc_engine_free_job(struct drm_sched_job *drm_job) >>> @@ -782,6 +796,57 @@ static void simple_error_capture(struct xe_engine *e) >>> } >>> #endif >>> +static void xe_guc_engine_trigger_cleanup(struct xe_engine *e) >>> +{ >>> + struct xe_guc *guc = engine_to_guc(e); >>> + >>> + if (xe_engine_is_lr(e)) >>> + queue_work(guc_to_gt(guc)->ordered_wq, &e->guc->lr_tdr); >>> + else >>> + drm_sched_set_timeout(&e->guc->sched, MIN_SCHED_TIMEOUT); >>> +} >>> + >>> +static void xe_guc_engine_lr_cleanup(struct work_struct *w) >>> +{ >>> + struct xe_guc_engine *ge = >>> + container_of(w, struct xe_guc_engine, lr_tdr); >>> + struct xe_engine *e = ge->engine; >>> + struct drm_gpu_scheduler *sched = &ge->sched; >>> + >>> + XE_BUG_ON(!xe_engine_is_lr(e)); >>> + trace_xe_engine_lr_cleanup(e); >>> + >>> + /* Kill the run_job / process_msg entry points */ >>> + drm_sched_run_wq_stop(sched); >>> + >>> + /* Engine state now stable, disable scheduling / deregister if needed */ >>> + if (engine_registered(e)) { >>> + struct xe_guc *guc = engine_to_guc(e); >>> + int ret; >>> + >>> + set_engine_banned(e); >>> + xe_engine_get(e); >>> + disable_scheduling_deregister(guc, e); >>> + >>> + /* >>> + * Must wait for scheduling to be disabled before signalling >>> + * any fences, if GT broken the GT reset code should signal us. >>> + */ >>> + smp_rmb(); >> wait_event() paired with wake_up() family of functions typically set the >> necessary barriers to make sure anything written prior to wake_up() is seen >> in wait_event(). So that smp_rmb() is most likely not needed. If it still >> is, its pairing smp_wmb() should be documented and pointed to as well. See >> documentation of set_current_state() vs __set_current_state(). >> >>> + ret = wait_event_timeout(guc->ct.wq, >>> + !engine_pending_disable(e) || >>> + guc_read_stopped(guc), HZ * 5); >>> + if (!ret) { >>> + XE_WARN_ON("Schedule disable failed to respond"); >>> + drm_sched_run_wq_start(sched); >>> + xe_gt_reset_async(e->gt); >>> + return; >>> + } >>> + } >>> + >>> + drm_sched_run_wq_start(sched); >>> +} >>> + >>> static enum drm_gpu_sched_stat >>> guc_engine_timedout_job(struct drm_sched_job *drm_job) >>> { >>> @@ -832,7 +897,7 @@ guc_engine_timedout_job(struct drm_sched_job *drm_job) >>> err = -EIO; >>> set_engine_banned(e); >>> xe_engine_get(e); >>> - disable_scheduling_deregister(engine_to_guc(e), e); >>> + disable_scheduling_deregister(guc, e); >>> /* >>> * Must wait for scheduling to be disabled before signalling >>> @@ -865,7 +930,7 @@ guc_engine_timedout_job(struct drm_sched_job *drm_job) >>> */ >>> list_add(&drm_job->list, &sched->pending_list); >>> drm_sched_run_wq_start(sched); >>> - drm_sched_set_timeout(&e->guc->sched, MIN_SCHED_TIMEOUT); >>> + xe_guc_engine_trigger_cleanup(e); >>> /* Mark all outstanding jobs as bad, thus completing them */ >>> spin_lock(&sched->job_list_lock); >>> @@ -889,6 +954,8 @@ static void __guc_engine_fini_async(struct work_struct *w) >>> trace_xe_engine_destroy(e); >>> + if (xe_engine_is_lr(e)) >>> + cancel_work_sync(&ge->lr_tdr); >>> if (e->flags & ENGINE_FLAG_PERSISTENT) >>> xe_device_remove_persistent_engines(gt_to_xe(e->gt), e); >>> release_guc_id(guc, e); >>> @@ -906,7 +973,7 @@ static void guc_engine_fini_async(struct xe_engine *e) >>> bool kernel = e->flags & ENGINE_FLAG_KERNEL; >>> INIT_WORK(&e->guc->fini_async, __guc_engine_fini_async); >>> - queue_work(system_unbound_wq, &e->guc->fini_async); >>> + queue_work(system_wq, &e->guc->fini_async); >>> /* We must block on kernel engines so slabs are empty on driver unload */ >>> if (kernel) { >>> @@ -1089,12 +1156,16 @@ static int guc_engine_init(struct xe_engine *e) >>> if (err) >>> goto err_free; >>> + >> Unrelated whitespace? >> >> >>> sched = &ge->sched; >>> err = drm_sched_entity_init(&ge->entity, DRM_SCHED_PRIORITY_NORMAL, >>> &sched, 1, NULL); >>> if (err) >>> goto err_sched; >>> + if (xe_engine_is_lr(e)) >>> + INIT_WORK(&e->guc->lr_tdr, xe_guc_engine_lr_cleanup); >>> + >>> mutex_lock(&guc->submission_state.lock); >>> err = alloc_guc_id(guc, e); >>> @@ -1146,7 +1217,7 @@ static void guc_engine_kill(struct xe_engine *e) >>> { >>> trace_xe_engine_kill(e); >>> set_engine_killed(e); >>> - drm_sched_set_timeout(&e->guc->sched, MIN_SCHED_TIMEOUT); >>> + xe_guc_engine_trigger_cleanup(e); >>> } >>> static void guc_engine_add_msg(struct xe_engine *e, struct drm_sched_msg *msg, >>> @@ -1296,6 +1367,9 @@ static void guc_engine_stop(struct xe_guc *guc, struct xe_engine *e) >>> /* Stop scheduling + flush any DRM scheduler operations */ >>> drm_sched_run_wq_stop(sched); >>> + if (engine_registered(e) && xe_engine_is_lr(e)) >>> + xe_engine_put(e); >>> + >>> /* Clean up lost G2H + reset engine state */ >>> if (engine_destroyed(e) && engine_registered(e)) { >>> if (engine_banned(e)) >>> @@ -1520,6 +1594,9 @@ int xe_guc_deregister_done_handler(struct xe_guc *guc, u32 *msg, u32 len) >>> trace_xe_engine_deregister_done(e); >>> clear_engine_registered(e); >>> + if (xe_engine_is_lr(e)) >>> + xe_engine_put(e); >>> + >>> if (engine_banned(e)) >>> xe_engine_put(e); >>> else >>> @@ -1557,7 +1634,7 @@ int xe_guc_engine_reset_handler(struct xe_guc *guc, u32 *msg, u32 len) >>> */ >>> set_engine_reset(e); >>> if (!engine_banned(e)) >>> - drm_sched_set_timeout(&e->guc->sched, MIN_SCHED_TIMEOUT); >>> + xe_guc_engine_trigger_cleanup(e); >>> return 0; >>> } >>> @@ -1584,7 +1661,7 @@ int xe_guc_engine_memory_cat_error_handler(struct xe_guc *guc, u32 *msg, >>> /* Treat the same as engine reset */ >>> set_engine_reset(e); >>> if (!engine_banned(e)) >>> - drm_sched_set_timeout(&e->guc->sched, MIN_SCHED_TIMEOUT); >>> + xe_guc_engine_trigger_cleanup(e); >>> return 0; >>> } >>> diff --git a/drivers/gpu/drm/xe/xe_trace.h b/drivers/gpu/drm/xe/xe_trace.h >>> index 2f8eb7ebe9a7..02861c26e145 100644 >>> --- a/drivers/gpu/drm/xe/xe_trace.h >>> +++ b/drivers/gpu/drm/xe/xe_trace.h >>> @@ -219,6 +219,11 @@ DEFINE_EVENT(xe_engine, xe_engine_resubmit, >>> TP_ARGS(e) >>> ); >>> +DEFINE_EVENT(xe_engine, xe_engine_lr_cleanup, >>> + TP_PROTO(struct xe_engine *e), >>> + TP_ARGS(e) >>> +); >>> + >>> DECLARE_EVENT_CLASS(xe_sched_job, >>> TP_PROTO(struct xe_sched_job *job), >>> TP_ARGS(job),