Merge branch 'drm-fixes' of git://people.freedesktop.org/~airlied/linux

Pull drm fixes from Dave Airlie:
 "Misc i915, vmwgfx and radeon fixes along with a fix for one of those
  recursive sleep mutex debug cases in the mst code"

* 'drm-fixes' of git://people.freedesktop.org/~airlied/linux:
  drm/vmwgfx: Fix an issue with the device losing its irq line on module unload
  drm/vmwgfx: Correctly NULLify dma buffer pointer on failure
  drm/vmwgfx: Reorder device takedown somewhat
  drm/vmwgfx: Fix a couple of lock dependency violations
  drm/radeon: drop setting UPLL to sleep mode
  drm/radeon: fix wait to actually occur after the signaling callback
  drm/i915: Prevent TLB error on first execution on SNB
  drm/i915: Do both mt and gen6 style forcewake reset on ivb probe
  drm/i915: Make WAIT_IOCTL negative timeouts be indefinite again
  drm/i915: use in_interrupt() not in_irq() to check context
  drm/mst: fix recursive sleep warning on qlock
  drm: Don't assign fbs for universal cursor support to files

drivers/gpu/drm/drm_crtc.c

@@ -43,9 +43,10 @@
 #include "drm_crtc_internal.h"
 #include "drm_internal.h"
 
-static struct drm_framebuffer *add_framebuffer_internal(struct drm_device *dev,
-							struct drm_mode_fb_cmd2 *r,
-							struct drm_file *file_priv);
+static struct drm_framebuffer *
+internal_framebuffer_create(struct drm_device *dev,
+			    struct drm_mode_fb_cmd2 *r,
+			    struct drm_file *file_priv);
 
 /* Avoid boilerplate.  I'm tired of typing. */
 #define DRM_ENUM_NAME_FN(fnname, list)				\
@@ -2908,13 +2909,11 @@ static int drm_mode_cursor_universal(struct drm_crtc *crtc,
 	 */
 	if (req->flags & DRM_MODE_CURSOR_BO) {
 		if (req->handle) {
-			fb = add_framebuffer_internal(dev, &fbreq, file_priv);
+			fb = internal_framebuffer_create(dev, &fbreq, file_priv);
 			if (IS_ERR(fb)) {
 				DRM_DEBUG_KMS("failed to wrap cursor buffer in drm framebuffer\n");
 				return PTR_ERR(fb);
 			}
-
-			drm_framebuffer_reference(fb);
 		} else {
 			fb = NULL;
 		}
@@ -3267,9 +3266,10 @@ static int framebuffer_check(const struct drm_mode_fb_cmd2 *r)
 	return 0;
 }
 
-static struct drm_framebuffer *add_framebuffer_internal(struct drm_device *dev,
-							struct drm_mode_fb_cmd2 *r,
-							struct drm_file *file_priv)
+static struct drm_framebuffer *
+internal_framebuffer_create(struct drm_device *dev,
+			    struct drm_mode_fb_cmd2 *r,
+			    struct drm_file *file_priv)
 {
 	struct drm_mode_config *config = &dev->mode_config;
 	struct drm_framebuffer *fb;
@@ -3301,12 +3301,6 @@ static struct drm_framebuffer *add_framebuffer_internal(struct drm_device *dev,
 		return fb;
 	}
 
-	mutex_lock(&file_priv->fbs_lock);
-	r->fb_id = fb->base.id;
-	list_add(&fb->filp_head, &file_priv->fbs);
-	DRM_DEBUG_KMS("[FB:%d]\n", fb->base.id);
-	mutex_unlock(&file_priv->fbs_lock);
-
 	return fb;
 }
 
@@ -3328,15 +3322,24 @@ static struct drm_framebuffer *add_framebuffer_internal(struct drm_device *dev,
 int drm_mode_addfb2(struct drm_device *dev,
 		    void *data, struct drm_file *file_priv)
 {
+	struct drm_mode_fb_cmd2 *r = data;
 	struct drm_framebuffer *fb;
 
 	if (!drm_core_check_feature(dev, DRIVER_MODESET))
 		return -EINVAL;
 
-	fb = add_framebuffer_internal(dev, data, file_priv);
+	fb = internal_framebuffer_create(dev, r, file_priv);
 	if (IS_ERR(fb))
 		return PTR_ERR(fb);
 
+	/* Transfer ownership to the filp for reaping on close */
+
+	DRM_DEBUG_KMS("[FB:%d]\n", fb->base.id);
+	mutex_lock(&file_priv->fbs_lock);
+	r->fb_id = fb->base.id;
+	list_add(&fb->filp_head, &file_priv->fbs);
+	mutex_unlock(&file_priv->fbs_lock);
+
 	return 0;
 }
 

drivers/gpu/drm/drm_dp_mst_topology.c

@@ -733,10 +733,14 @@ static bool check_txmsg_state(struct drm_dp_mst_topology_mgr *mgr,
 			      struct drm_dp_sideband_msg_tx *txmsg)
 {
 	bool ret;
-	mutex_lock(&mgr->qlock);
+
+	/*
+	 * All updates to txmsg->state are protected by mgr->qlock, and the two
+	 * cases we check here are terminal states. For those the barriers
+	 * provided by the wake_up/wait_event pair are enough.
+	 */
 	ret = (txmsg->state == DRM_DP_SIDEBAND_TX_RX ||
 	       txmsg->state == DRM_DP_SIDEBAND_TX_TIMEOUT);
-	mutex_unlock(&mgr->qlock);
 	return ret;
 }
 
@@ -1363,12 +1367,13 @@ static int process_single_tx_qlock(struct drm_dp_mst_topology_mgr *mgr,
 	return 0;
 }
 
-/* must be called holding qlock */
 static void process_single_down_tx_qlock(struct drm_dp_mst_topology_mgr *mgr)
 {
 	struct drm_dp_sideband_msg_tx *txmsg;
 	int ret;
 
+	WARN_ON(!mutex_is_locked(&mgr->qlock));
+
 	/* construct a chunk from the first msg in the tx_msg queue */
 	if (list_empty(&mgr->tx_msg_downq)) {
 		mgr->tx_down_in_progress = false;

drivers/gpu/drm/i915/i915_gem.c

@@ -2936,9 +2936,9 @@ i915_gem_wait_ioctl(struct drm_device *dev, void *data, struct drm_file *file)
 	req = obj->last_read_req;
 
 	/* Do this after OLR check to make sure we make forward progress polling
-	 * on this IOCTL with a timeout <=0 (like busy ioctl)
+	 * on this IOCTL with a timeout == 0 (like busy ioctl)
 	 */
-	if (args->timeout_ns <= 0) {
+	if (args->timeout_ns == 0) {
 		ret = -ETIME;
 		goto out;
 	}
@@ -2948,7 +2948,8 @@ i915_gem_wait_ioctl(struct drm_device *dev, void *data, struct drm_file *file)
 	i915_gem_request_reference(req);
 	mutex_unlock(&dev->struct_mutex);
 
-	ret = __i915_wait_request(req, reset_counter, true, &args->timeout_ns,
+	ret = __i915_wait_request(req, reset_counter, true,
+				  args->timeout_ns > 0 ? &args->timeout_ns : NULL,
 				  file->driver_priv);
 	mutex_lock(&dev->struct_mutex);
 	i915_gem_request_unreference(req);
@@ -4792,6 +4793,9 @@ i915_gem_init_hw(struct drm_device *dev)
 	if (INTEL_INFO(dev)->gen < 6 && !intel_enable_gtt())
 		return -EIO;
 
+	/* Double layer security blanket, see i915_gem_init() */
+	intel_uncore_forcewake_get(dev_priv, FORCEWAKE_ALL);
+
 	if (dev_priv->ellc_size)
 		I915_WRITE(HSW_IDICR, I915_READ(HSW_IDICR) | IDIHASHMSK(0xf));
 
@@ -4824,7 +4828,7 @@ i915_gem_init_hw(struct drm_device *dev)
 	for_each_ring(ring, dev_priv, i) {
 		ret = ring->init_hw(ring);
 		if (ret)
-			return ret;
+			goto out;
 	}
 
 	for (i = 0; i < NUM_L3_SLICES(dev); i++)
@@ -4841,9 +4845,11 @@ i915_gem_init_hw(struct drm_device *dev)
 		DRM_ERROR("Context enable failed %d\n", ret);
 		i915_gem_cleanup_ringbuffer(dev);
 
-		return ret;
+		goto out;
 	}
 
+out:
+	intel_uncore_forcewake_put(dev_priv, FORCEWAKE_ALL);
 	return ret;
 }
 
@@ -4877,6 +4883,14 @@ int i915_gem_init(struct drm_device *dev)
 		dev_priv->gt.stop_ring = intel_logical_ring_stop;
 	}
 
+	/* This is just a security blanket to placate dragons.
+	 * On some systems, we very sporadically observe that the first TLBs
+	 * used by the CS may be stale, despite us poking the TLB reset. If
+	 * we hold the forcewake during initialisation these problems
+	 * just magically go away.
+	 */
+	intel_uncore_forcewake_get(dev_priv, FORCEWAKE_ALL);
+
 	ret = i915_gem_init_userptr(dev);
 	if (ret)
 		goto out_unlock;
@@ -4903,6 +4917,7 @@ int i915_gem_init(struct drm_device *dev)
 	}
 
 out_unlock:
+	intel_uncore_forcewake_put(dev_priv, FORCEWAKE_ALL);
 	mutex_unlock(&dev->struct_mutex);
 
 	return ret;

drivers/gpu/drm/i915/intel_display.c

@@ -9716,7 +9716,7 @@ void intel_check_page_flip(struct drm_device *dev, int pipe)
 	struct drm_crtc *crtc = dev_priv->pipe_to_crtc_mapping[pipe];
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 
-	WARN_ON(!in_irq());
+	WARN_ON(!in_interrupt());
 
 	if (crtc == NULL)
 		return;

drivers/gpu/drm/i915/intel_uncore.c

@@ -1048,8 +1048,14 @@ static void intel_uncore_fw_domains_init(struct drm_device *dev)
 
 		/* We need to init first for ECOBUS access and then
 		 * determine later if we want to reinit, in case of MT access is
-		 * not working
+		 * not working. In this stage we don't know which flavour this
+		 * ivb is, so it is better to reset also the gen6 fw registers
+		 * before the ecobus check.
 		 */
+
+		__raw_i915_write32(dev_priv, FORCEWAKE, 0);
+		__raw_posting_read(dev_priv, ECOBUS);
+
 		fw_domain_init(dev_priv, FW_DOMAIN_ID_RENDER,
 			       FORCEWAKE_MT, FORCEWAKE_MT_ACK);
 

drivers/gpu/drm/radeon/radeon_fence.c

@@ -1030,37 +1030,59 @@ static inline bool radeon_test_signaled(struct radeon_fence *fence)
 	return test_bit(FENCE_FLAG_SIGNALED_BIT, &fence->base.flags);
 }
 
+struct radeon_wait_cb {
+	struct fence_cb base;
+	struct task_struct *task;
+};
+
+static void
+radeon_fence_wait_cb(struct fence *fence, struct fence_cb *cb)
+{
+	struct radeon_wait_cb *wait =
+		container_of(cb, struct radeon_wait_cb, base);
+
+	wake_up_process(wait->task);
+}
+
 static signed long radeon_fence_default_wait(struct fence *f, bool intr,
 					     signed long t)
 {
 	struct radeon_fence *fence = to_radeon_fence(f);
 	struct radeon_device *rdev = fence->rdev;
-	bool signaled;
+	struct radeon_wait_cb cb;
 
-	fence_enable_sw_signaling(&fence->base);
+	cb.task = current;
 
-	/*
-	 * This function has to return -EDEADLK, but cannot hold
-	 * exclusive_lock during the wait because some callers
-	 * may already hold it. This means checking needs_reset without
-	 * lock, and not fiddling with any gpu internals.
-	 *
-	 * The callback installed with fence_enable_sw_signaling will
-	 * run before our wait_event_*timeout call, so we will see
-	 * both the signaled fence and the changes to needs_reset.
-	 */
+	if (fence_add_callback(f, &cb.base, radeon_fence_wait_cb))
+		return t;
+
+	while (t > 0) {
+		if (intr)
+			set_current_state(TASK_INTERRUPTIBLE);
+		else
+			set_current_state(TASK_UNINTERRUPTIBLE);
+
+		/*
+		 * radeon_test_signaled must be called after
+		 * set_current_state to prevent a race with wake_up_process
+		 */
+		if (radeon_test_signaled(fence))
+			break;
+
+		if (rdev->needs_reset) {
+			t = -EDEADLK;
+			break;
+		}
+
+		t = schedule_timeout(t);
+
+		if (t > 0 && intr && signal_pending(current))
+			t = -ERESTARTSYS;
+	}
+
+	__set_current_state(TASK_RUNNING);
+	fence_remove_callback(f, &cb.base);
 
-	if (intr)
-		t = wait_event_interruptible_timeout(rdev->fence_queue,
-			((signaled = radeon_test_signaled(fence)) ||
-			 rdev->needs_reset), t);
-	else
-		t = wait_event_timeout(rdev->fence_queue,
-			((signaled = radeon_test_signaled(fence)) ||
-			 rdev->needs_reset), t);
-
-	if (t > 0 && !signaled)
-		return -EDEADLK;
 	return t;
 }
 

drivers/gpu/drm/radeon/si.c

@@ -7130,8 +7130,7 @@ int si_set_uvd_clocks(struct radeon_device *rdev, u32 vclk, u32 dclk)
 	WREG32_P(CG_UPLL_FUNC_CNTL, UPLL_BYPASS_EN_MASK, ~UPLL_BYPASS_EN_MASK);
 
 	if (!vclk || !dclk) {
-		/* keep the Bypass mode, put PLL to sleep */
-		WREG32_P(CG_UPLL_FUNC_CNTL, UPLL_SLEEP_MASK, ~UPLL_SLEEP_MASK);
+		/* keep the Bypass mode */
 		return 0;
 	}
 
@@ -7147,8 +7146,7 @@ int si_set_uvd_clocks(struct radeon_device *rdev, u32 vclk, u32 dclk)
 	/* set VCO_MODE to 1 */
 	WREG32_P(CG_UPLL_FUNC_CNTL, UPLL_VCO_MODE_MASK, ~UPLL_VCO_MODE_MASK);
 
-	/* toggle UPLL_SLEEP to 1 then back to 0 */
-	WREG32_P(CG_UPLL_FUNC_CNTL, UPLL_SLEEP_MASK, ~UPLL_SLEEP_MASK);
+	/* disable sleep mode */
 	WREG32_P(CG_UPLL_FUNC_CNTL, 0, ~UPLL_SLEEP_MASK);
 
 	/* deassert UPLL_RESET */