1 files changed, 43 insertions, 37 deletions

diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c b/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c
index f3dcae4b9d45..0eaa1064242c 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c
@@ -398,7 +398,7 @@ static inline void set_context_guc_id_invalid(struct intel_context *ce)
 
 static inline struct intel_guc *ce_to_guc(struct intel_context *ce)
 {
-	return &ce->engine->gt->uc.guc;
+	return gt_to_guc(ce->engine->gt);
 }
 
 static inline struct i915_priolist *to_priolist(struct rb_node *rb)
@@ -1246,7 +1246,7 @@ static void __get_engine_usage_record(struct intel_engine_cs *engine,
 static void guc_update_engine_gt_clks(struct intel_engine_cs *engine)
 {
 	struct intel_engine_guc_stats *stats = &engine->stats.guc;
-	struct intel_guc *guc = &engine->gt->uc.guc;
+	struct intel_guc *guc = gt_to_guc(engine->gt);
 	u32 last_switch, ctx_id, total;
 
 	lockdep_assert_held(&guc->timestamp.lock);
@@ -1311,7 +1311,7 @@ static ktime_t guc_engine_busyness(struct intel_engine_cs *engine, ktime_t *now)
 	struct intel_engine_guc_stats stats_saved, *stats = &engine->stats.guc;
 	struct i915_gpu_error *gpu_error = &engine->i915->gpu_error;
 	struct intel_gt *gt = engine->gt;
-	struct intel_guc *guc = &gt->uc.guc;
+	struct intel_guc *guc = gt_to_guc(gt);
 	u64 total, gt_stamp_saved;
 	unsigned long flags;
 	u32 reset_count;
@@ -1403,14 +1403,17 @@ static void guc_cancel_busyness_worker(struct intel_guc *guc)
 	 * Trying to pass a 'need_sync' or 'in_reset' flag all the way down through
 	 * every possible call stack is unfeasible. It would be too intrusive to many
 	 * areas that really don't care about the GuC backend. However, there is the
-	 * 'reset_in_progress' flag available, so just use that.
+	 * I915_RESET_BACKOFF flag and the gt->reset.mutex can be tested for is_locked.
+	 * So just use those. Note that testing both is required due to the hideously
+	 * complex nature of the i915 driver's reset code paths.
 	 *
 	 * And note that in the case of a reset occurring during driver unload
-	 * (wedge_on_fini), skipping the cancel in _prepare (when the reset flag is set
-	 * is fine because there is another cancel in _finish (when the reset flag is
-	 * not).
+	 * (wedged_on_fini), skipping the cancel in reset_prepare/reset_fini (when the
+	 * reset flag/mutex are set) is fine because there is another explicit cancel in
+	 * intel_guc_submission_fini (when the reset flag/mutex are not).
 	 */
-	if (guc_to_gt(guc)->uc.reset_in_progress)
+	if (mutex_is_locked(&guc_to_gt(guc)->reset.mutex) ||
+	    test_bit(I915_RESET_BACKOFF, &guc_to_gt(guc)->reset.flags))
 		cancel_delayed_work(&guc->timestamp.work);
 	else
 		cancel_delayed_work_sync(&guc->timestamp.work);
@@ -1424,8 +1427,6 @@ static void __reset_guc_busyness_stats(struct intel_guc *guc)
 	unsigned long flags;
 	ktime_t unused;
 
-	guc_cancel_busyness_worker(guc);
-
 	spin_lock_irqsave(&guc->timestamp.lock, flags);
 
 	guc_update_pm_timestamp(guc, &unused);
@@ -1576,7 +1577,7 @@ static void guc_fini_engine_stats(struct intel_guc *guc)
 
 void intel_guc_busyness_park(struct intel_gt *gt)
 {
-	struct intel_guc *guc = &gt->uc.guc;
+	struct intel_guc *guc = gt_to_guc(gt);
 
 	if (!guc_submission_initialized(guc))
 		return;
@@ -1603,7 +1604,7 @@ void intel_guc_busyness_park(struct intel_gt *gt)
 
 void intel_guc_busyness_unpark(struct intel_gt *gt)
 {
-	struct intel_guc *guc = &gt->uc.guc;
+	struct intel_guc *guc = gt_to_guc(gt);
 	unsigned long flags;
 	ktime_t unused;
 
@@ -2004,13 +2005,6 @@ void intel_guc_submission_cancel_requests(struct intel_guc *guc)
 
 void intel_guc_submission_reset_finish(struct intel_guc *guc)
 {
-	/*
-	 * Ensure the busyness worker gets cancelled even on a fatal wedge.
-	 * Note that reset_prepare is not allowed to because it confuses lockdep.
-	 */
-	if (guc_submission_initialized(guc))
-		guc_cancel_busyness_worker(guc);
-
 	/* Reset called during driver load or during wedge? */
 	if (unlikely(!guc_submission_initialized(guc) ||
 		     !intel_guc_is_fw_running(guc) ||
@@ -2136,6 +2130,7 @@ void intel_guc_submission_fini(struct intel_guc *guc)
 	if (!guc->submission_initialized)
 		return;
 
+	guc_fini_engine_stats(guc);
 	guc_flush_destroyed_contexts(guc);
 	guc_lrc_desc_pool_destroy_v69(guc);
 	i915_sched_engine_put(guc->sched_engine);
@@ -2194,7 +2189,7 @@ static bool need_tasklet(struct intel_guc *guc, struct i915_request *rq)
 static void guc_submit_request(struct i915_request *rq)
 {
 	struct i915_sched_engine *sched_engine = rq->engine->sched_engine;
-	struct intel_guc *guc = &rq->engine->gt->uc.guc;
+	struct intel_guc *guc = gt_to_guc(rq->engine->gt);
 	unsigned long flags;
 
 	/* Will be called from irq-context when using foreign fences. */
@@ -2220,11 +2215,10 @@ static int new_guc_id(struct intel_guc *guc, struct intel_context *ce)
 					      order_base_2(ce->parallel.number_children
 							   + 1));
 	else
-		ret = ida_simple_get(&guc->submission_state.guc_ids,
-				     NUMBER_MULTI_LRC_GUC_ID(guc),
-				     guc->submission_state.num_guc_ids,
-				     GFP_KERNEL | __GFP_RETRY_MAYFAIL |
-				     __GFP_NOWARN);
+		ret = ida_alloc_range(&guc->submission_state.guc_ids,
+				      NUMBER_MULTI_LRC_GUC_ID(guc),
+				      guc->submission_state.num_guc_ids - 1,
+				      GFP_KERNEL | __GFP_RETRY_MAYFAIL | __GFP_NOWARN);
 	if (unlikely(ret < 0))
 		return ret;
 
@@ -2247,8 +2241,8 @@ static void __release_guc_id(struct intel_guc *guc, struct intel_context *ce)
 							   + 1));
 		} else {
 			--guc->submission_state.guc_ids_in_use;
-			ida_simple_remove(&guc->submission_state.guc_ids,
-					  ce->guc_id.id);
+			ida_free(&guc->submission_state.guc_ids,
+				 ce->guc_id.id);
 		}
 		clr_ctx_id_mapping(guc, ce->guc_id.id);
 		set_context_guc_id_invalid(ce);
@@ -2645,6 +2639,7 @@ MAKE_CONTEXT_POLICY_ADD(execution_quantum, EXECUTION_QUANTUM)
 MAKE_CONTEXT_POLICY_ADD(preemption_timeout, PREEMPTION_TIMEOUT)
 MAKE_CONTEXT_POLICY_ADD(priority, SCHEDULING_PRIORITY)
 MAKE_CONTEXT_POLICY_ADD(preempt_to_idle, PREEMPT_TO_IDLE_ON_QUANTUM_EXPIRY)
+MAKE_CONTEXT_POLICY_ADD(slpc_ctx_freq_req, SLPM_GT_FREQUENCY)
 
 #undef MAKE_CONTEXT_POLICY_ADD
 
@@ -2660,10 +2655,11 @@ static int __guc_context_set_context_policies(struct intel_guc *guc,
 static int guc_context_policy_init_v70(struct intel_context *ce, bool loop)
 {
 	struct intel_engine_cs *engine = ce->engine;
-	struct intel_guc *guc = &engine->gt->uc.guc;
+	struct intel_guc *guc = gt_to_guc(engine->gt);
 	struct context_policy policy;
 	u32 execution_quantum;
 	u32 preemption_timeout;
+	u32 slpc_ctx_freq_req = 0;
 	unsigned long flags;
 	int ret;
 
@@ -2675,11 +2671,15 @@ static int guc_context_policy_init_v70(struct intel_context *ce, bool loop)
 	execution_quantum = engine->props.timeslice_duration_ms * 1000;
 	preemption_timeout = engine->props.preempt_timeout_ms * 1000;
 
+	if (ce->flags & BIT(CONTEXT_LOW_LATENCY))
+		slpc_ctx_freq_req |= SLPC_CTX_FREQ_REQ_IS_COMPUTE;
+
 	__guc_context_policy_start_klv(&policy, ce->guc_id.id);
 
 	__guc_context_policy_add_priority(&policy, ce->guc_state.prio);
 	__guc_context_policy_add_execution_quantum(&policy, execution_quantum);
 	__guc_context_policy_add_preemption_timeout(&policy, preemption_timeout);
+	__guc_context_policy_add_slpc_ctx_freq_req(&policy, slpc_ctx_freq_req);
 
 	if (engine->flags & I915_ENGINE_WANT_FORCED_PREEMPTION)
 		__guc_context_policy_add_preempt_to_idle(&policy, 1);
@@ -2736,7 +2736,7 @@ static u32 map_guc_prio_to_lrc_desc_prio(u8 prio)
 static void prepare_context_registration_info_v69(struct intel_context *ce)
 {
 	struct intel_engine_cs *engine = ce->engine;
-	struct intel_guc *guc = &engine->gt->uc.guc;
+	struct intel_guc *guc = gt_to_guc(engine->gt);
 	u32 ctx_id = ce->guc_id.id;
 	struct guc_lrc_desc_v69 *desc;
 	struct intel_context *child;
@@ -2805,7 +2805,7 @@ static void prepare_context_registration_info_v70(struct intel_context *ce,
 						  struct guc_ctxt_registration_info *info)
 {
 	struct intel_engine_cs *engine = ce->engine;
-	struct intel_guc *guc = &engine->gt->uc.guc;
+	struct intel_guc *guc = gt_to_guc(engine->gt);
 	u32 ctx_id = ce->guc_id.id;
 
 	GEM_BUG_ON(!engine->mask);
@@ -2868,7 +2868,7 @@ static int try_context_registration(struct intel_context *ce, bool loop)
 {
 	struct intel_engine_cs *engine = ce->engine;
 	struct intel_runtime_pm *runtime_pm = engine->uncore->rpm;
-	struct intel_guc *guc = &engine->gt->uc.guc;
+	struct intel_guc *guc = gt_to_guc(engine->gt);
 	intel_wakeref_t wakeref;
 	u32 ctx_id = ce->guc_id.id;
 	bool context_registered;
@@ -4496,7 +4496,13 @@ static void guc_default_vfuncs(struct intel_engine_cs *engine)
 	if (engine->class == COMPUTE_CLASS)
 		if (IS_GFX_GT_IP_STEP(engine->gt, IP_VER(12, 70), STEP_A0, STEP_B0) ||
 		    IS_DG2(engine->i915))
-			engine->flags |= I915_ENGINE_USES_WA_HOLD_CCS_SWITCHOUT;
+			engine->flags |= I915_ENGINE_USES_WA_HOLD_SWITCHOUT;
+
+	/* Wa_16019325821 */
+	/* Wa_14019159160 */
+	if ((engine->class == COMPUTE_CLASS || engine->class == RENDER_CLASS) &&
+	    IS_GFX_GT_IP_RANGE(engine->gt, IP_VER(12, 70), IP_VER(12, 71)))
+		engine->flags |= I915_ENGINE_USES_WA_HOLD_SWITCHOUT;
 
 	/*
 	 * TODO: GuC supports timeslicing and semaphores as well, but they're
@@ -4507,7 +4513,7 @@ static void guc_default_vfuncs(struct intel_engine_cs *engine)
 	 */
 
 	engine->emit_bb_start = gen8_emit_bb_start;
-	if (GRAPHICS_VER_FULL(engine->i915) >= IP_VER(12, 50))
+	if (GRAPHICS_VER_FULL(engine->i915) >= IP_VER(12, 55))
 		engine->emit_bb_start = xehp_emit_bb_start;
 }
 
@@ -4549,7 +4555,7 @@ static void guc_sched_engine_destroy(struct kref *kref)
 int intel_guc_submission_setup(struct intel_engine_cs *engine)
 {
 	struct drm_i915_private *i915 = engine->i915;
-	struct intel_guc *guc = &engine->gt->uc.guc;
+	struct intel_guc *guc = gt_to_guc(engine->gt);
 
 	/*
 	 * The setup relies on several assumptions (e.g. irqs always enabled)
@@ -5308,7 +5314,7 @@ int intel_guc_engine_failure_process_msg(struct intel_guc *guc,
 
 void intel_guc_find_hung_context(struct intel_engine_cs *engine)
 {
-	struct intel_guc *guc = &engine->gt->uc.guc;
+	struct intel_guc *guc = gt_to_guc(engine->gt);
 	struct intel_context *ce;
 	struct i915_request *rq;
 	unsigned long index;
@@ -5370,7 +5376,7 @@ void intel_guc_dump_active_requests(struct intel_engine_cs *engine,
 				    struct i915_request *hung_rq,
 				    struct drm_printer *m)
 {
-	struct intel_guc *guc = &engine->gt->uc.guc;
+	struct intel_guc *guc = gt_to_guc(engine->gt);
 	struct intel_context *ce;
 	unsigned long index;
 	unsigned long flags;
@@ -5822,7 +5828,7 @@ guc_create_virtual(struct intel_engine_cs **siblings, unsigned int count,
 	if (!ve)
 		return ERR_PTR(-ENOMEM);
 
-	guc = &siblings[0]->gt->uc.guc;
+	guc = gt_to_guc(siblings[0]->gt);
 
 	ve->base.i915 = siblings[0]->i915;
 	ve->base.gt = siblings[0]->gt;