drm/i915: Remove redundant intel_autoenable_gt_powersave()

Now that we always execute a context switch upon module load, there is
no need to queue a delayed task for doing so. The purpose of the delayed
task is to enable GT powersaving, for which we need the HW state to be
valid (i.e. having loaded a context and initialised basic state). We
used to defer this operation as historically it was slow (due to slow
register polling, fixed with commit 1758b90e ("drm/i915: Use a hybrid
scheme for fast register waits")) but now we have a requirement to save
the default HW state.

v2: Load the kernel context (to provide the power context) upon resume.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20171112112738.1463-3-chris@chris-wilson.co.uk

drivers/gpu/drm/i915/i915_drv.c

@@ -1683,8 +1683,6 @@ static int i915_drm_resume(struct drm_device *dev)
 
 	intel_csr_ucode_resume(dev_priv);
 
-	i915_gem_resume(dev_priv);
-
 	i915_restore_state(dev_priv);
 	intel_pps_unlock_regs_wa(dev_priv);
 	intel_opregion_setup(dev_priv);
@@ -1705,12 +1703,7 @@ static int i915_drm_resume(struct drm_device *dev)
 
 	drm_mode_config_reset(dev);
 
-	mutex_lock(&dev->struct_mutex);
-	if (i915_gem_init_hw(dev_priv)) {
-		DRM_ERROR("failed to re-initialize GPU, declaring wedged!\n");
-		i915_gem_set_wedged(dev_priv);
-	}
-	mutex_unlock(&dev->struct_mutex);
+	i915_gem_resume(dev_priv);
 
 	intel_guc_resume(dev_priv);
 
@@ -1745,8 +1738,6 @@ static int i915_drm_resume(struct drm_device *dev)
 
 	intel_opregion_notify_adapter(dev_priv, PCI_D0);
 
-	intel_autoenable_gt_powersave(dev_priv);
-
 	enable_rpm_wakeref_asserts(dev_priv);
 
 	return 0;

drivers/gpu/drm/i915/i915_drv.h

@@ -1392,7 +1392,6 @@ struct intel_gen6_power_mgmt {
 	struct intel_rps rps;
 	struct intel_rc6 rc6;
 	struct intel_llc_pstate llc_pstate;
-	struct delayed_work autoenable_work;
 };
 
 /* defined intel_pm.c */

drivers/gpu/drm/i915/i915_gem.c

@@ -4806,23 +4806,38 @@ int i915_gem_suspend(struct drm_i915_private *dev_priv)
 	return ret;
 }
 
-void i915_gem_resume(struct drm_i915_private *dev_priv)
+void i915_gem_resume(struct drm_i915_private *i915)
 {
-	struct drm_device *dev = &dev_priv->drm;
+	WARN_ON(i915->gt.awake);
 
-	WARN_ON(dev_priv->gt.awake);
+	mutex_lock(&i915->drm.struct_mutex);
+	intel_uncore_forcewake_get(i915, FORCEWAKE_ALL);
 
-	mutex_lock(&dev->struct_mutex);
-	i915_gem_restore_gtt_mappings(dev_priv);
-	i915_gem_restore_fences(dev_priv);
+	i915_gem_restore_gtt_mappings(i915);
+	i915_gem_restore_fences(i915);
 
 	/* As we didn't flush the kernel context before suspend, we cannot
 	 * guarantee that the context image is complete. So let's just reset
 	 * it and start again.
 	 */
-	dev_priv->gt.resume(dev_priv);
+	i915->gt.resume(i915);
 
-	mutex_unlock(&dev->struct_mutex);
+	if (i915_gem_init_hw(i915))
+		goto err_wedged;
+
+	/* Always reload a context for powersaving. */
+	if (i915_gem_switch_to_kernel_context(i915))
+		goto err_wedged;
+
+out_unlock:
+	intel_uncore_forcewake_put(i915, FORCEWAKE_ALL);
+	mutex_unlock(&i915->drm.struct_mutex);
+	return;
+
+err_wedged:
+	DRM_ERROR("failed to re-initialize GPU, declaring wedged!\n");
+	i915_gem_set_wedged(i915);
+	goto out_unlock;
 }
 
 void i915_gem_init_swizzling(struct drm_i915_private *dev_priv)

drivers/gpu/drm/i915/intel_drv.h

@@ -1880,7 +1880,6 @@ void intel_init_gt_powersave(struct drm_i915_private *dev_priv);
 void intel_cleanup_gt_powersave(struct drm_i915_private *dev_priv);
 void intel_sanitize_gt_powersave(struct drm_i915_private *dev_priv);
 void intel_enable_gt_powersave(struct drm_i915_private *dev_priv);
-void intel_autoenable_gt_powersave(struct drm_i915_private *dev_priv);
 void intel_disable_gt_powersave(struct drm_i915_private *dev_priv);
 void intel_suspend_gt_powersave(struct drm_i915_private *dev_priv);
 void gen6_rps_busy(struct drm_i915_private *dev_priv);

drivers/gpu/drm/i915/intel_pm.c

@@ -7959,8 +7959,6 @@ void intel_init_gt_powersave(struct drm_i915_private *dev_priv)
 	rps->boost_freq = rps->max_freq;
 
 	mutex_unlock(&dev_priv->pcu_lock);
-
-	intel_autoenable_gt_powersave(dev_priv);
 }
 
 void intel_cleanup_gt_powersave(struct drm_i915_private *dev_priv)
@@ -7985,9 +7983,6 @@ void intel_suspend_gt_powersave(struct drm_i915_private *dev_priv)
 	if (INTEL_GEN(dev_priv) < 6)
 		return;
 
-	if (cancel_delayed_work_sync(&dev_priv->gt_pm.autoenable_work))
-		intel_runtime_pm_put(dev_priv);
-
 	/* gen6_rps_idle() will be called later to disable interrupts */
 }
 
@@ -8146,65 +8141,6 @@ void intel_enable_gt_powersave(struct drm_i915_private *dev_priv)
 	mutex_unlock(&dev_priv->pcu_lock);
 }
 
-static void __intel_autoenable_gt_powersave(struct work_struct *work)
-{
-	struct drm_i915_private *dev_priv =
-		container_of(work,
-			     typeof(*dev_priv),
-			     gt_pm.autoenable_work.work);
-	struct intel_engine_cs *rcs;
-	struct drm_i915_gem_request *req;
-
-	rcs = dev_priv->engine[RCS];
-	if (rcs->last_retired_context)
-		goto out;
-
-	if (!rcs->init_context)
-		goto out;
-
-	mutex_lock(&dev_priv->drm.struct_mutex);
-
-	req = i915_gem_request_alloc(rcs, dev_priv->kernel_context);
-	if (IS_ERR(req))
-		goto unlock;
-
-	if (!i915_modparams.enable_execlists && i915_switch_context(req) == 0)
-		rcs->init_context(req);
-
-	/* Mark the device busy, calling intel_enable_gt_powersave() */
-	i915_add_request(req);
-
-unlock:
-	mutex_unlock(&dev_priv->drm.struct_mutex);
-out:
-	intel_runtime_pm_put(dev_priv);
-}
-
-void intel_autoenable_gt_powersave(struct drm_i915_private *dev_priv)
-{
-	if (IS_IRONLAKE_M(dev_priv)) {
-		ironlake_enable_drps(dev_priv);
-		intel_init_emon(dev_priv);
-	} else if (INTEL_INFO(dev_priv)->gen >= 6) {
-		/*
-		 * PCU communication is slow and this doesn't need to be
-		 * done at any specific time, so do this out of our fast path
-		 * to make resume and init faster.
-		 *
-		 * We depend on the HW RC6 power context save/restore
-		 * mechanism when entering D3 through runtime PM suspend. So
-		 * disable RPM until RPS/RC6 is properly setup. We can only
-		 * get here via the driver load/system resume/runtime resume
-		 * paths, so the _noresume version is enough (and in case of
-		 * runtime resume it's necessary).
-		 */
-		if (queue_delayed_work(dev_priv->wq,
-				       &dev_priv->gt_pm.autoenable_work,
-				       round_jiffies_up_relative(HZ)))
-			intel_runtime_pm_get_noresume(dev_priv);
-	}
-}
-
 static void ibx_init_clock_gating(struct drm_i915_private *dev_priv)
 {
 	/*
@@ -9435,8 +9371,6 @@ void intel_pm_setup(struct drm_i915_private *dev_priv)
 {
 	mutex_init(&dev_priv->pcu_lock);
 
-	INIT_DELAYED_WORK(&dev_priv->gt_pm.autoenable_work,
-			  __intel_autoenable_gt_powersave);
 	atomic_set(&dev_priv->gt_pm.rps.num_waiters, 0);
 
 	dev_priv->runtime_pm.suspended = false;