Merge tag 'drm-fixes-2019-08-02-1' of git://anongit.freedesktop.org/drm/drm

Pull more drm fixes from Daniel Vetter:
 "Dave sends his pull, everyone realizes they've been asleep at the
  wheel and hits send on their own pulls :-/

  Normally I'd just ignore these all because w/e for me and Dave. But
  this time around the latecomers also included drm-intel-fixes, which
  failed to send out a -fixes pull thus far for this release (screwed up
  vacation coverage, despite that 2/3 maintainers were around ... they
  all look appropriately guilty), and that really is overdue to get
  landed.

  And since I had to do a pull request anyway I pulled the other two
  late ones too.

  intel fixes (didn't have any ever since the main merge window pull):
   - gvt fixes (2 cc: stable)
   - fix gpu reset vs mm-shrinker vs wakeup fun (needed a few patches)
   - two gem locking fixes (one cc: stable)
   - pile of misc fixes all over with minor impact, 6 cc: stable, others
     from this window

  exynos:
   - misc minor fixes

  misc:
   - some build/Kconfig fixes
   - regression fix for vm scalability perf test which seems to mostly
     exercise dmesg/console logging ...
   - the vgem cache flush fix for arm64 broke the world on x86, so
     that's reverted again

* tag 'drm-fixes-2019-08-02-1' of git://anongit.freedesktop.org/drm/drm: (42 commits)
  Revert "drm/vgem: fix cache synchronization on arm/arm64"
  drm/exynos: fix missing decrement of retry counter
  drm/exynos: add CONFIG_MMU dependency
  drm/exynos: remove redundant assignment to pointer 'node'
  drm/exynos: using dev_get_drvdata directly
  drm/bochs: Use shadow buffer for bochs framebuffer console
  drm/fb-helper: Instanciate shadow FB if configured in device's mode_config
  drm/fb-helper: Map DRM client buffer only when required
  drm/client: Support unmapping of DRM client buffers
  drm/i915: Only recover active engines
  drm/i915: Add a wakeref getter for iff the wakeref is already active
  drm/i915: Lift intel_engines_resume() to callers
  drm/vgem: fix cache synchronization on arm/arm64
  drm/i810: Use CONFIG_PREEMPTION
  drm/bridge: tc358764: Fix build error
  drm/bridge: lvds-encoder: Fix build error while CONFIG_DRM_KMS_HELPER=m
  drm/i915/gvt: Adding ppgtt to GVT GEM context after shadow pdps settled.
  drm/i915/gvt: grab runtime pm first for forcewake use
  drm/i915/gvt: fix incorrect cache entry for guest page mapping
  drm/i915/gvt: Checking workload's gma earlier
  ...

46 files changed, 517 insertions, 256 deletions

diff --git a/drivers/gpu/drm/Kconfig b/drivers/gpu/drm/Kconfig
index 1d80222587ad..3c88420e3497 100644
--- a/drivers/gpu/drm/Kconfig
+++ b/drivers/gpu/drm/Kconfig
@@ -394,7 +394,7 @@ config DRM_R128
 config DRM_I810
 	tristate "Intel I810"
 	# !PREEMPT because of missing ioctl locking
-	depends on DRM && AGP && AGP_INTEL && (!PREEMPT || BROKEN)
+	depends on DRM && AGP && AGP_INTEL && (!PREEMPTION || BROKEN)
 	help
 	  Choose this option if you have an Intel I810 graphics card.  If M is
 	  selected, the module will be called i810.  AGP support is required
diff --git a/drivers/gpu/drm/bochs/bochs_kms.c b/drivers/gpu/drm/bochs/bochs_kms.c
index bc19dbd531ef..359030d5d818 100644
--- a/drivers/gpu/drm/bochs/bochs_kms.c
+++ b/drivers/gpu/drm/bochs/bochs_kms.c
@@ -191,6 +191,7 @@ int bochs_kms_init(struct bochs_device *bochs)
 	bochs->dev->mode_config.fb_base = bochs->fb_base;
 	bochs->dev->mode_config.preferred_depth = 24;
 	bochs->dev->mode_config.prefer_shadow = 0;
+	bochs->dev->mode_config.prefer_shadow_fbdev = 1;
 	bochs->dev->mode_config.quirk_addfb_prefer_host_byte_order = true;
 
 	bochs->dev->mode_config.funcs = &bochs_mode_funcs;
diff --git a/drivers/gpu/drm/bridge/Kconfig b/drivers/gpu/drm/bridge/Kconfig
index ee777469293a..e4e22bbae2a7 100644
--- a/drivers/gpu/drm/bridge/Kconfig
+++ b/drivers/gpu/drm/bridge/Kconfig
@@ -48,6 +48,7 @@ config DRM_DUMB_VGA_DAC
 config DRM_LVDS_ENCODER
 	tristate "Transparent parallel to LVDS encoder support"
 	depends on OF
+	select DRM_KMS_HELPER
 	select DRM_PANEL_BRIDGE
 	help
 	  Support for transparent parallel to LVDS encoders that don't require
@@ -116,9 +117,10 @@ config DRM_THINE_THC63LVD1024
 
 config DRM_TOSHIBA_TC358764
 	tristate "TC358764 DSI/LVDS bridge"
-	depends on DRM && DRM_PANEL
 	depends on OF
 	select DRM_MIPI_DSI
+	select DRM_KMS_HELPER
+	select DRM_PANEL
 	help
 	  Toshiba TC358764 DSI/LVDS bridge driver.
 
diff --git a/drivers/gpu/drm/drm_client.c b/drivers/gpu/drm/drm_client.c
index 410572f14257..e1dafb0cc5e2 100644
--- a/drivers/gpu/drm/drm_client.c
+++ b/drivers/gpu/drm/drm_client.c
@@ -254,7 +254,6 @@ drm_client_buffer_create(struct drm_client_dev *client, u32 width, u32 height, u
 	struct drm_device *dev = client->dev;
 	struct drm_client_buffer *buffer;
 	struct drm_gem_object *obj;
-	void *vaddr;
 	int ret;
 
 	buffer = kzalloc(sizeof(*buffer), GFP_KERNEL);
@@ -281,6 +280,36 @@ drm_client_buffer_create(struct drm_client_dev *client, u32 width, u32 height, u
 
 	buffer->gem = obj;
 
+	return buffer;
+
+err_delete:
+	drm_client_buffer_delete(buffer);
+
+	return ERR_PTR(ret);
+}
+
+/**
+ * drm_client_buffer_vmap - Map DRM client buffer into address space
+ * @buffer: DRM client buffer
+ *
+ * This function maps a client buffer into kernel address space. If the
+ * buffer is already mapped, it returns the mapping's address.
+ *
+ * Client buffer mappings are not ref'counted. Each call to
+ * drm_client_buffer_vmap() should be followed by a call to
+ * drm_client_buffer_vunmap(); or the client buffer should be mapped
+ * throughout its lifetime.
+ *
+ * Returns:
+ *	The mapped memory's address
+ */
+void *drm_client_buffer_vmap(struct drm_client_buffer *buffer)
+{
+	void *vaddr;
+
+	if (buffer->vaddr)
+		return buffer->vaddr;
+
 	/*
 	 * FIXME: The dependency on GEM here isn't required, we could
 	 * convert the driver handle to a dma-buf instead and use the
@@ -289,21 +318,30 @@ drm_client_buffer_create(struct drm_client_dev *client, u32 width, u32 height, u
 	 * fd_install step out of the driver backend hooks, to make that
 	 * final step optional for internal users.
 	 */
-	vaddr = drm_gem_vmap(obj);
-	if (IS_ERR(vaddr)) {
-		ret = PTR_ERR(vaddr);
-		goto err_delete;
-	}
+	vaddr = drm_gem_vmap(buffer->gem);
+	if (IS_ERR(vaddr))
+		return vaddr;
 
 	buffer->vaddr = vaddr;
 
-	return buffer;
-
-err_delete:
-	drm_client_buffer_delete(buffer);
+	return vaddr;
+}
+EXPORT_SYMBOL(drm_client_buffer_vmap);
 
-	return ERR_PTR(ret);
+/**
+ * drm_client_buffer_vunmap - Unmap DRM client buffer
+ * @buffer: DRM client buffer
+ *
+ * This function removes a client buffer's memory mapping. Calling this
+ * function is only required by clients that manage their buffer mappings
+ * by themselves.
+ */
+void drm_client_buffer_vunmap(struct drm_client_buffer *buffer)
+{
+	drm_gem_vunmap(buffer->gem, buffer->vaddr);
+	buffer->vaddr = NULL;
 }
+EXPORT_SYMBOL(drm_client_buffer_vunmap);
 
 static void drm_client_buffer_rmfb(struct drm_client_buffer *buffer)
 {
diff --git a/drivers/gpu/drm/drm_fb_helper.c b/drivers/gpu/drm/drm_fb_helper.c
index 1984e5c54d58..a7ba5b4902d6 100644
--- a/drivers/gpu/drm/drm_fb_helper.c
+++ b/drivers/gpu/drm/drm_fb_helper.c
@@ -403,6 +403,7 @@ static void drm_fb_helper_dirty_work(struct work_struct *work)
 	struct drm_clip_rect *clip = &helper->dirty_clip;
 	struct drm_clip_rect clip_copy;
 	unsigned long flags;
+	void *vaddr;
 
 	spin_lock_irqsave(&helper->dirty_lock, flags);
 	clip_copy = *clip;
@@ -412,10 +413,20 @@ static void drm_fb_helper_dirty_work(struct work_struct *work)
 
 	/* call dirty callback only when it has been really touched */
 	if (clip_copy.x1 < clip_copy.x2 && clip_copy.y1 < clip_copy.y2) {
+
 		/* Generic fbdev uses a shadow buffer */
-		if (helper->buffer)
+		if (helper->buffer) {
+			vaddr = drm_client_buffer_vmap(helper->buffer);
+			if (IS_ERR(vaddr))
+				return;
 			drm_fb_helper_dirty_blit_real(helper, &clip_copy);
-		helper->fb->funcs->dirty(helper->fb, NULL, 0, 0, &clip_copy, 1);
+		}
+		if (helper->fb->funcs->dirty)
+			helper->fb->funcs->dirty(helper->fb, NULL, 0, 0,
+						 &clip_copy, 1);
+
+		if (helper->buffer)
+			drm_client_buffer_vunmap(helper->buffer);
 	}
 }
 
@@ -604,6 +615,16 @@ void drm_fb_helper_unlink_fbi(struct drm_fb_helper *fb_helper)
 }
 EXPORT_SYMBOL(drm_fb_helper_unlink_fbi);
 
+static bool drm_fbdev_use_shadow_fb(struct drm_fb_helper *fb_helper)
+{
+	struct drm_device *dev = fb_helper->dev;
+	struct drm_framebuffer *fb = fb_helper->fb;
+
+	return dev->mode_config.prefer_shadow_fbdev ||
+	       dev->mode_config.prefer_shadow ||
+	       fb->funcs->dirty;
+}
+
 static void drm_fb_helper_dirty(struct fb_info *info, u32 x, u32 y,
 				u32 width, u32 height)
 {
@@ -611,7 +632,7 @@ static void drm_fb_helper_dirty(struct fb_info *info, u32 x, u32 y,
 	struct drm_clip_rect *clip = &helper->dirty_clip;
 	unsigned long flags;
 
-	if (!helper->fb->funcs->dirty)
+	if (!drm_fbdev_use_shadow_fb(helper))
 		return;
 
 	spin_lock_irqsave(&helper->dirty_lock, flags);
@@ -2178,6 +2199,7 @@ int drm_fb_helper_generic_probe(struct drm_fb_helper *fb_helper,
 	struct drm_framebuffer *fb;
 	struct fb_info *fbi;
 	u32 format;
+	void *vaddr;
 
 	DRM_DEBUG_KMS("surface width(%d), height(%d) and bpp(%d)\n",
 		      sizes->surface_width, sizes->surface_height,
@@ -2200,16 +2222,10 @@ int drm_fb_helper_generic_probe(struct drm_fb_helper *fb_helper,
 	fbi->fbops = &drm_fbdev_fb_ops;
 	fbi->screen_size = fb->height * fb->pitches[0];
 	fbi->fix.smem_len = fbi->screen_size;
-	fbi->screen_buffer = buffer->vaddr;
-	/* Shamelessly leak the physical address to user-space */
-#if IS_ENABLED(CONFIG_DRM_FBDEV_LEAK_PHYS_SMEM)
-	if (drm_leak_fbdev_smem && fbi->fix.smem_start == 0)
-		fbi->fix.smem_start =
-			page_to_phys(virt_to_page(fbi->screen_buffer));
-#endif
+
 	drm_fb_helper_fill_info(fbi, fb_helper, sizes);
 
-	if (fb->funcs->dirty) {
+	if (drm_fbdev_use_shadow_fb(fb_helper)) {
 		struct fb_ops *fbops;
 		void *shadow;
 
@@ -2231,6 +2247,19 @@ int drm_fb_helper_generic_probe(struct drm_fb_helper *fb_helper,
 		fbi->fbdefio = &drm_fbdev_defio;
 
 		fb_deferred_io_init(fbi);
+	} else {
+		/* buffer is mapped for HW framebuffer */
+		vaddr = drm_client_buffer_vmap(fb_helper->buffer);
+		if (IS_ERR(vaddr))
+			return PTR_ERR(vaddr);
+
+		fbi->screen_buffer = vaddr;
+		/* Shamelessly leak the physical address to user-space */
+#if IS_ENABLED(CONFIG_DRM_FBDEV_LEAK_PHYS_SMEM)
+		if (drm_leak_fbdev_smem && fbi->fix.smem_start == 0)
+			fbi->fix.smem_start =
+				page_to_phys(virt_to_page(fbi->screen_buffer));
+#endif
 	}
 
 	return 0;
diff --git a/drivers/gpu/drm/exynos/Kconfig b/drivers/gpu/drm/exynos/Kconfig
index 60ce4a8ad9e1..6f7d3b3b3628 100644
--- a/drivers/gpu/drm/exynos/Kconfig
+++ b/drivers/gpu/drm/exynos/Kconfig
@@ -2,6 +2,7 @@
 config DRM_EXYNOS
 	tristate "DRM Support for Samsung SoC EXYNOS Series"
 	depends on OF && DRM && (ARCH_S3C64XX || ARCH_S5PV210 || ARCH_EXYNOS || ARCH_MULTIPLATFORM || COMPILE_TEST)
+	depends on MMU
 	select DRM_KMS_HELPER
 	select VIDEOMODE_HELPERS
 	select SND_SOC_HDMI_CODEC if SND_SOC
diff --git a/drivers/gpu/drm/exynos/exynos_drm_fimc.c b/drivers/gpu/drm/exynos/exynos_drm_fimc.c
index a594ab7be2c0..164d914cbe9a 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_fimc.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_fimc.c
@@ -44,7 +44,7 @@ static unsigned int fimc_mask = 0xc;
 module_param_named(fimc_devs, fimc_mask, uint, 0644);
 MODULE_PARM_DESC(fimc_devs, "Alias mask for assigning FIMC devices to Exynos DRM");
 
-#define get_fimc_context(dev)	platform_get_drvdata(to_platform_device(dev))
+#define get_fimc_context(dev)	dev_get_drvdata(dev)
 
 enum {
 	FIMC_CLK_LCLK,
diff --git a/drivers/gpu/drm/exynos/exynos_drm_g2d.c b/drivers/gpu/drm/exynos/exynos_drm_g2d.c
index 50904eee96f7..2a3382d43bc9 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_g2d.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_g2d.c
@@ -267,7 +267,7 @@ static inline void g2d_hw_reset(struct g2d_data *g2d)
 static int g2d_init_cmdlist(struct g2d_data *g2d)
 {
 	struct device *dev = g2d->dev;
-	struct g2d_cmdlist_node *node = g2d->cmdlist_node;
+	struct g2d_cmdlist_node *node;
 	int nr;
 	int ret;
 	struct g2d_buf_info *buf_info;
diff --git a/drivers/gpu/drm/exynos/exynos_drm_gsc.c b/drivers/gpu/drm/exynos/exynos_drm_gsc.c
index 1e4b21c49a06..1c524db9570f 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_gsc.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_gsc.c
@@ -58,7 +58,7 @@
 #define GSC_COEF_DEPTH	3
 #define GSC_AUTOSUSPEND_DELAY		2000
 
-#define get_gsc_context(dev)	platform_get_drvdata(to_platform_device(dev))
+#define get_gsc_context(dev)	dev_get_drvdata(dev)
 #define gsc_read(offset)		readl(ctx->regs + (offset))
 #define gsc_write(cfg, offset)	writel(cfg, ctx->regs + (offset))
 
diff --git a/drivers/gpu/drm/exynos/exynos_drm_scaler.c b/drivers/gpu/drm/exynos/exynos_drm_scaler.c
index 9af096479e1c..b24ba948b725 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_scaler.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_scaler.c
@@ -94,12 +94,12 @@ static inline int scaler_reset(struct scaler_context *scaler)
 	scaler_write(SCALER_CFG_SOFT_RESET, SCALER_CFG);
 	do {
 		cpu_relax();
-	} while (retry > 1 &&
+	} while (--retry > 1 &&
 		 scaler_read(SCALER_CFG) & SCALER_CFG_SOFT_RESET);
 	do {
 		cpu_relax();
 		scaler_write(1, SCALER_INT_EN);
-	} while (retry > 0 && scaler_read(SCALER_INT_EN) != 1);
+	} while (--retry > 0 && scaler_read(SCALER_INT_EN) != 1);
 
 	return retry ? 0 : -EIO;
 }
diff --git a/drivers/gpu/drm/i915/display/intel_bios.c b/drivers/gpu/drm/i915/display/intel_bios.c
index c4710889cb32..3ef4e9f573cf 100644
--- a/drivers/gpu/drm/i915/display/intel_bios.c
+++ b/drivers/gpu/drm/i915/display/intel_bios.c
@@ -765,7 +765,7 @@ parse_psr(struct drm_i915_private *dev_priv, const struct bdb_header *bdb)
 	}
 
 	if (bdb->version >= 226) {
-		u32 wakeup_time = psr_table->psr2_tp2_tp3_wakeup_time;
+		u32 wakeup_time = psr->psr2_tp2_tp3_wakeup_time;
 
 		wakeup_time = (wakeup_time >> (2 * panel_type)) & 0x3;
 		switch (wakeup_time) {
diff --git a/drivers/gpu/drm/i915/display/intel_bw.c b/drivers/gpu/drm/i915/display/intel_bw.c
index 753ac3165061..7b908e10d32e 100644
--- a/drivers/gpu/drm/i915/display/intel_bw.c
+++ b/drivers/gpu/drm/i915/display/intel_bw.c
@@ -178,6 +178,8 @@ static int icl_get_bw_info(struct drm_i915_private *dev_priv)
 		clpchgroup = (sa->deburst * deinterleave / num_channels) << i;
 		bi->num_planes = (ipqdepth - clpchgroup) / clpchgroup + 1;
 
+		bi->num_qgv_points = qi.num_points;
+
 		for (j = 0; j < qi.num_points; j++) {
 			const struct intel_qgv_point *sp = &qi.points[j];
 			int ct, bw;
@@ -195,7 +197,7 @@ static int icl_get_bw_info(struct drm_i915_private *dev_priv)
 			bi->deratedbw[j] = min(maxdebw,
 					       bw * 9 / 10); /* 90% */
 
-			DRM_DEBUG_KMS("BW%d / QGV %d: num_planes=%d deratedbw=%d\n",
+			DRM_DEBUG_KMS("BW%d / QGV %d: num_planes=%d deratedbw=%u\n",
 				      i, j, bi->num_planes, bi->deratedbw[j]);
 		}
 
@@ -211,14 +213,17 @@ static unsigned int icl_max_bw(struct drm_i915_private *dev_priv,
 {
 	int i;
 
-	/* Did we initialize the bw limits successfully? */
-	if (dev_priv->max_bw[0].num_planes == 0)
-		return UINT_MAX;
-
 	for (i = 0; i < ARRAY_SIZE(dev_priv->max_bw); i++) {
 		const struct intel_bw_info *bi =
 			&dev_priv->max_bw[i];
 
+		/*
+		 * Pcode will not expose all QGV points when
+		 * SAGV is forced to off/min/med/max.
+		 */
+		if (qgv_point >= bi->num_qgv_points)
+			return UINT_MAX;
+
 		if (num_planes >= bi->num_planes)
 			return bi->deratedbw[qgv_point];
 	}
diff --git a/drivers/gpu/drm/i915/display/intel_cdclk.c b/drivers/gpu/drm/i915/display/intel_cdclk.c
index 8993ab283562..0d19bbd08122 100644
--- a/drivers/gpu/drm/i915/display/intel_cdclk.c
+++ b/drivers/gpu/drm/i915/display/intel_cdclk.c
@@ -2240,6 +2240,17 @@ int intel_crtc_compute_min_cdclk(const struct intel_crtc_state *crtc_state)
 		min_cdclk = max(2 * 96000, min_cdclk);
 
 	/*
+	 * "For DP audio configuration, cdclk frequency shall be set to
+	 *  meet the following requirements:
+	 *  DP Link Frequency(MHz) | Cdclk frequency(MHz)
+	 *  270                    | 320 or higher
+	 *  162                    | 200 or higher"
+	 */
+	if ((IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv)) &&
+	    intel_crtc_has_dp_encoder(crtc_state) && crtc_state->has_audio)
+		min_cdclk = max(crtc_state->port_clock, min_cdclk);
+
+	/*
 	 * On Valleyview some DSI panels lose (v|h)sync when the clock is lower
 	 * than 320000KHz.
 	 */
diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c
index 30b97ded6fdd..592b92782fab 100644
--- a/drivers/gpu/drm/i915/display/intel_display.c
+++ b/drivers/gpu/drm/i915/display/intel_display.c
@@ -1839,7 +1839,7 @@ static void intel_enable_pipe(const struct intel_crtc_state *new_crtc_state)
 		/* FIXME: assert CPU port conditions for SNB+ */
 	}
 
-	trace_intel_pipe_enable(dev_priv, pipe);
+	trace_intel_pipe_enable(crtc);
 
 	reg = PIPECONF(cpu_transcoder);
 	val = I915_READ(reg);
@@ -1880,7 +1880,7 @@ static void intel_disable_pipe(const struct intel_crtc_state *old_crtc_state)
 	 */
 	assert_planes_disabled(crtc);
 
-	trace_intel_pipe_disable(dev_priv, pipe);
+	trace_intel_pipe_disable(crtc);
 
 	reg = PIPECONF(cpu_transcoder);
 	val = I915_READ(reg);
diff --git a/drivers/gpu/drm/i915/display/intel_display_power.c b/drivers/gpu/drm/i915/display/intel_display_power.c
index c93ad512014c..2d1939db108f 100644
--- a/drivers/gpu/drm/i915/display/intel_display_power.c
+++ b/drivers/gpu/drm/i915/display/intel_display_power.c
@@ -438,16 +438,23 @@ icl_combo_phy_aux_power_well_disable(struct drm_i915_private *dev_priv,
 #define ICL_AUX_PW_TO_CH(pw_idx)	\
 	((pw_idx) - ICL_PW_CTL_IDX_AUX_A + AUX_CH_A)
 
+#define ICL_TBT_AUX_PW_TO_CH(pw_idx)	\
+	((pw_idx) - ICL_PW_CTL_IDX_AUX_TBT1 + AUX_CH_C)
+
 static void
 icl_tc_phy_aux_power_well_enable(struct drm_i915_private *dev_priv,
 				 struct i915_power_well *power_well)
 {
-	enum aux_ch aux_ch = ICL_AUX_PW_TO_CH(power_well->desc->hsw.idx);
+	int pw_idx = power_well->desc->hsw.idx;
+	bool is_tbt = power_well->desc->hsw.is_tc_tbt;
+	enum aux_ch aux_ch;
 	u32 val;
 
+	aux_ch = is_tbt ? ICL_TBT_AUX_PW_TO_CH(pw_idx) :
+			  ICL_AUX_PW_TO_CH(pw_idx);
 	val = I915_READ(DP_AUX_CH_CTL(aux_ch));
 	val &= ~DP_AUX_CH_CTL_TBT_IO;
-	if (power_well->desc->hsw.is_tc_tbt)
+	if (is_tbt)
 		val |= DP_AUX_CH_CTL_TBT_IO;
 	I915_WRITE(DP_AUX_CH_CTL(aux_ch), val);
 
diff --git a/drivers/gpu/drm/i915/display/intel_vbt_defs.h b/drivers/gpu/drm/i915/display/intel_vbt_defs.h
index 2f4894e9a03d..5ddbe71ab423 100644
--- a/drivers/gpu/drm/i915/display/intel_vbt_defs.h
+++ b/drivers/gpu/drm/i915/display/intel_vbt_defs.h
@@ -478,13 +478,13 @@ struct psr_table {
 	/* TP wake up time in multiple of 100 */
 	u16 tp1_wakeup_time;
 	u16 tp2_tp3_wakeup_time;
-
-	/* PSR2 TP2/TP3 wakeup time for 16 panels */
-	u32 psr2_tp2_tp3_wakeup_time;
 } __packed;
 
 struct bdb_psr {
 	struct psr_table psr_table[16];
+
+	/* PSR2 TP2/TP3 wakeup time for 16 panels */
+	u32 psr2_tp2_tp3_wakeup_time;
 } __packed;
 
 /*
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_pm.c b/drivers/gpu/drm/i915/gem/i915_gem_pm.c
index 05011d4a3b88..914b5d4112bb 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_pm.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_pm.c
@@ -253,14 +253,15 @@ void i915_gem_resume(struct drm_i915_private *i915)
 	i915_gem_restore_gtt_mappings(i915);
 	i915_gem_restore_fences(i915);
 
+	if (i915_gem_init_hw(i915))
+		goto err_wedged;
+
 	/*
 	 * 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.
 	 */
-	intel_gt_resume(i915);
-
-	if (i915_gem_init_hw(i915))
+	if (intel_gt_resume(i915))
 		goto err_wedged;
 
 	intel_uc_resume(i915);
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_userptr.c b/drivers/gpu/drm/i915/gem/i915_gem_userptr.c
index 528b61678334..2caa594322bc 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_userptr.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_userptr.c
@@ -664,7 +664,15 @@ i915_gem_userptr_put_pages(struct drm_i915_gem_object *obj,
 
 	for_each_sgt_page(page, sgt_iter, pages) {
 		if (obj->mm.dirty)
-			set_page_dirty(page);
+			/*
+			 * As this may not be anonymous memory (e.g. shmem)
+			 * but exist on a real mapping, we have to lock
+			 * the page in order to dirty it -- holding
+			 * the page reference is not sufficient to
+			 * prevent the inode from being truncated.
+			 * Play safe and take the lock.
+			 */
+			set_page_dirty_lock(page);
 
 		mark_page_accessed(page);
 		put_page(page);
diff --git a/drivers/gpu/drm/i915/gt/intel_context.c b/drivers/gpu/drm/i915/gt/intel_context.c
index 2c454f227c2e..23120901c55f 100644
--- a/drivers/gpu/drm/i915/gt/intel_context.c
+++ b/drivers/gpu/drm/i915/gt/intel_context.c
@@ -126,6 +126,7 @@ static void intel_context_retire(struct i915_active *active)
 	if (ce->state)
 		__context_unpin_state(ce->state);
 
+	intel_ring_unpin(ce->ring);
 	intel_context_put(ce);
 }
 
@@ -160,27 +161,35 @@ int intel_context_active_acquire(struct intel_context *ce, unsigned long flags)
 
 	intel_context_get(ce);
 
+	err = intel_ring_pin(ce->ring);
+	if (err)
+		goto err_put;
+
 	if (!ce->state)
 		return 0;
 
 	err = __context_pin_state(ce->state, flags);
-	if (err) {
-		i915_active_cancel(&ce->active);
-		intel_context_put(ce);
-		return err;
-	}
+	if (err)
+		goto err_ring;
 
 	/* Preallocate tracking nodes */
 	if (!i915_gem_context_is_kernel(ce->gem_context)) {
 		err = i915_active_acquire_preallocate_barrier(&ce->active,
 							      ce->engine);
-		if (err) {
-			i915_active_release(&ce->active);
-			return err;
-		}
+		if (err)
+			goto err_state;
 	}
 
 	return 0;
+
+err_state:
+	__context_unpin_state(ce->state);
+err_ring:
+	intel_ring_unpin(ce->ring);
+err_put:
+	intel_context_put(ce);
+	i915_active_cancel(&ce->active);
+	return err;
 }
 
 void intel_context_active_release(struct intel_context *ce)
diff --git a/drivers/gpu/drm/i915/gt/intel_engine_cs.c b/drivers/gpu/drm/i915/gt/intel_engine_cs.c
index 7fd33e81c2d9..f25632c9b292 100644
--- a/drivers/gpu/drm/i915/gt/intel_engine_cs.c
+++ b/drivers/gpu/drm/i915/gt/intel_engine_cs.c
@@ -969,9 +969,14 @@ const char *i915_cache_level_str(struct drm_i915_private *i915, int type)
 u32 intel_calculate_mcr_s_ss_select(struct drm_i915_private *dev_priv)
 {
 	const struct sseu_dev_info *sseu = &RUNTIME_INFO(dev_priv)->sseu;
+	unsigned int slice = fls(sseu->slice_mask) - 1;
+	unsigned int subslice;
 	u32 mcr_s_ss_select;
-	u32 slice = fls(sseu->slice_mask);
-	u32 subslice = fls(sseu->subslice_mask[slice]);
+
+	GEM_BUG_ON(slice >= ARRAY_SIZE(sseu->subslice_mask));
+	subslice = fls(sseu->subslice_mask[slice]);
+	GEM_BUG_ON(!subslice);
+	subslice--;
 
 	if (IS_GEN(dev_priv, 10))
 		mcr_s_ss_select = GEN8_MCR_SLICE(slice) |
@@ -1471,6 +1476,7 @@ void intel_engine_dump(struct intel_engine_cs *engine,
 	struct i915_gpu_error * const error = &engine->i915->gpu_error;
 	struct i915_request *rq;
 	intel_wakeref_t wakeref;
+	unsigned long flags;
 
 	if (header) {
 		va_list ap;
@@ -1490,10 +1496,9 @@ void intel_engine_dump(struct intel_engine_cs *engine,
 		   i915_reset_engine_count(error, engine),
 		   i915_reset_count(error));
 
-	rcu_read_lock();
-
 	drm_printf(m, "\tRequests:\n");
 
+	spin_lock_irqsave(&engine->active.lock, flags);
 	rq = intel_engine_find_active_request(engine);
 	if (rq) {
 		print_request(m, rq, "\t\tactive ");
@@ -1513,8 +1518,7 @@ void intel_engine_dump(struct intel_engine_cs *engine,
 
 		print_request_ring(m, rq);
 	}
-
-	rcu_read_unlock();
+	spin_unlock_irqrestore(&engine->active.lock, flags);
 
 	wakeref = intel_runtime_pm_get_if_in_use(&engine->i915->runtime_pm);
 	if (wakeref) {
@@ -1672,7 +1676,6 @@ struct i915_request *
 intel_engine_find_active_request(struct intel_engine_cs *engine)
 {
 	struct i915_request *request, *active = NULL;
-	unsigned long flags;
 
 	/*
 	 * We are called by the error capture, reset and to dump engine
@@ -1685,7 +1688,7 @@ intel_engine_find_active_request(struct intel_engine_cs *engine)
 	 * At all other times, we must assume the GPU is still running, but
 	 * we only care about the snapshot of this moment.
 	 */
-	spin_lock_irqsave(&engine->active.lock, flags);
+	lockdep_assert_held(&engine->active.lock);
 	list_for_each_entry(request, &engine->active.requests, sched.link) {
 		if (i915_request_completed(request))
 			continue;
@@ -1700,7 +1703,6 @@ intel_engine_find_active_request(struct intel_engine_cs *engine)
 		active = request;
 		break;
 	}
-	spin_unlock_irqrestore(&engine->active.lock, flags);
 
 	return active;
 }
diff --git a/drivers/gpu/drm/i915/gt/intel_engine_pm.c b/drivers/gpu/drm/i915/gt/intel_engine_pm.c
index 2ce00d3dc42a..ae5b6baf6dff 100644
--- a/drivers/gpu/drm/i915/gt/intel_engine_pm.c
+++ b/drivers/gpu/drm/i915/gt/intel_engine_pm.c
@@ -142,27 +142,3 @@ void intel_engine_init__pm(struct intel_engine_cs *engine)
 {
 	intel_wakeref_init(&engine->wakeref);
 }
-
-int intel_engines_resume(struct drm_i915_private *i915)
-{
-	struct intel_engine_cs *engine;
-	enum intel_engine_id id;
-	int err = 0;
-
-	intel_gt_pm_get(i915);
-	for_each_engine(engine, i915, id) {
-		intel_engine_pm_get(engine);
-		engine->serial++; /* kernel context lost */
-		err = engine->resume(engine);
-		intel_engine_pm_put(engine);
-		if (err) {
-			dev_err(i915->drm.dev,
-				"Failed to restart %s (%d)\n",
-				engine->name, err);
-			break;
-		}
-	}
-	intel_gt_pm_put(i915);
-
-	return err;
-}
diff --git a/drivers/gpu/drm/i915/gt/intel_engine_pm.h b/drivers/gpu/drm/i915/gt/intel_engine_pm.h
index b326cd993d60..a11c893f64c6 100644
--- a/drivers/gpu/drm/i915/gt/intel_engine_pm.h
+++ b/drivers/gpu/drm/i915/gt/intel_engine_pm.h
@@ -7,16 +7,22 @@
 #ifndef INTEL_ENGINE_PM_H
 #define INTEL_ENGINE_PM_H
 
+#include "intel_engine_types.h"
+#include "intel_wakeref.h"
+
 struct drm_i915_private;
-struct intel_engine_cs;
 
 void intel_engine_pm_get(struct intel_engine_cs *engine);
 void intel_engine_pm_put(struct intel_engine_cs *engine);
 
+static inline bool
+intel_engine_pm_get_if_awake(struct intel_engine_cs *engine)
+{
+	return intel_wakeref_get_if_active(&engine->wakeref);
+}
+
 void intel_engine_park(struct intel_engine_cs *engine);
 
 void intel_engine_init__pm(struct intel_engine_cs *engine);
 
-int intel_engines_resume(struct drm_i915_private *i915);
-
 #endif /* INTEL_ENGINE_PM_H */
diff --git a/drivers/gpu/drm/i915/gt/intel_engine_types.h b/drivers/gpu/drm/i915/gt/intel_engine_types.h
index 868b220214f8..43e975a26016 100644
--- a/drivers/gpu/drm/i915/gt/intel_engine_types.h
+++ b/drivers/gpu/drm/i915/gt/intel_engine_types.h
@@ -70,6 +70,18 @@ struct intel_ring {
 	struct list_head request_list;
 	struct list_head active_link;
 
+	/*
+	 * As we have two types of rings, one global to the engine used
+	 * by ringbuffer submission and those that are exclusive to a
+	 * context used by execlists, we have to play safe and allow
+	 * atomic updates to the pin_count. However, the actual pinning
+	 * of the context is either done during initialisation for
+	 * ringbuffer submission or serialised as part of the context
+	 * pinning for execlists, and so we do not need a mutex ourselves
+	 * to serialise intel_ring_pin/intel_ring_unpin.
+	 */
+	atomic_t pin_count;
+
 	u32 head;
 	u32 tail;
 	u32 emit;
diff --git a/drivers/gpu/drm/i915/gt/intel_gt_pm.c b/drivers/gpu/drm/i915/gt/intel_gt_pm.c
index 7b5967751762..9f8f7f54191f 100644
--- a/drivers/gpu/drm/i915/gt/intel_gt_pm.c
+++ b/drivers/gpu/drm/i915/gt/intel_gt_pm.c
@@ -5,6 +5,7 @@
  */
 
 #include "i915_drv.h"
+#include "intel_engine_pm.h"
 #include "intel_gt_pm.h"
 #include "intel_pm.h"
 #include "intel_wakeref.h"
@@ -118,10 +119,11 @@ void intel_gt_sanitize(struct drm_i915_private *i915, bool force)
 		intel_engine_reset(engine, false);
 }
 
-void intel_gt_resume(struct drm_i915_private *i915)
+int intel_gt_resume(struct drm_i915_private *i915)
 {
 	struct intel_engine_cs *engine;
 	enum intel_engine_id id;
+	int err = 0;
 
 	/*
 	 * After resume, we may need to poke into the pinned kernel
@@ -129,9 +131,12 @@ void intel_gt_resume(struct drm_i915_private *i915)
 	 * Only the kernel contexts should remain pinned over suspend,
 	 * allowing us to fixup the user contexts on their first pin.
 	 */
+	intel_gt_pm_get(i915);
 	for_each_engine(engine, i915, id) {
 		struct intel_context *ce;
 
+		intel_engine_pm_get(engine);
+
 		ce = engine->kernel_context;
 		if (ce)
 			ce->ops->reset(ce);
@@ -139,5 +144,19 @@ void intel_gt_resume(struct drm_i915_private *i915)
 		ce = engine->preempt_context;
 		if (ce)
 			ce->ops->reset(ce);
+
+		engine->serial++; /* kernel context lost */
+		err = engine->resume(engine);
+
+		intel_engine_pm_put(engine);
+		if (err) {
+			dev_err(i915->drm.dev,
+				"Failed to restart %s (%d)\n",
+				engine->name, err);
+			break;
+		}
 	}
+	intel_gt_pm_put(i915);
+
+	return err;
 }
diff --git a/drivers/gpu/drm/i915/gt/intel_gt_pm.h b/drivers/gpu/drm/i915/gt/intel_gt_pm.h
index 7dd1130a19a4..53f342b20181 100644
--- a/drivers/gpu/drm/i915/gt/intel_gt_pm.h
+++ b/drivers/gpu/drm/i915/gt/intel_gt_pm.h
@@ -22,6 +22,6 @@ void intel_gt_pm_put(struct drm_i915_private *i915);
 void intel_gt_pm_init(struct drm_i915_private *i915);
 
 void intel_gt_sanitize(struct drm_i915_private *i915, bool force);
-void intel_gt_resume(struct drm_i915_private *i915);
+int intel_gt_resume(struct drm_i915_private *i915);
 
 #endif /* INTEL_GT_PM_H */
diff --git a/drivers/gpu/drm/i915/gt/intel_lrc.c b/drivers/gpu/drm/i915/gt/intel_lrc.c
index b42b5f158295..82b7ace62d97 100644
--- a/drivers/gpu/drm/i915/gt/intel_lrc.c
+++ b/drivers/gpu/drm/i915/gt/intel_lrc.c
@@ -1414,6 +1414,7 @@ static void execlists_context_destroy(struct kref *kref)
 {
 	struct intel_context *ce = container_of(kref, typeof(*ce), ref);
 
+	GEM_BUG_ON(!i915_active_is_idle(&ce->active));
 	GEM_BUG_ON(intel_context_is_pinned(ce));
 
 	if (ce->state)
@@ -1426,7 +1427,6 @@ static void execlists_context_unpin(struct intel_context *ce)
 {
 	i915_gem_context_unpin_hw_id(ce->gem_context);
 	i915_gem_object_unpin_map(ce->state->obj);
-	intel_ring_unpin(ce->ring);
 }
 
 static void
@@ -1478,13 +1478,9 @@ __execlists_context_pin(struct intel_context *ce,
 		goto unpin_active;
 	}
 
-	ret = intel_ring_pin(ce->ring);
-	if (ret)
-		goto unpin_map;
-
 	ret = i915_gem_context_pin_hw_id(ce->gem_context);
 	if (ret)
-		goto unpin_ring;
+		goto unpin_map;
 
 	ce->lrc_desc = lrc_descriptor(ce, engine);
 	ce->lrc_reg_state = vaddr + LRC_STATE_PN * PAGE_SIZE;
@@ -1492,8 +1488,6 @@ __execlists_context_pin(struct intel_context *ce,
 
 	return 0;
 
-unpin_ring:
-	intel_ring_unpin(ce->ring);
 unpin_map:
 	i915_gem_object_unpin_map(ce->state->obj);
 unpin_active:
diff --git a/drivers/gpu/drm/i915/gt/intel_reset.c b/drivers/gpu/drm/i915/gt/intel_reset.c
index 4c478b38e420..3f907701ef4d 100644
--- a/drivers/gpu/drm/i915/gt/intel_reset.c
+++ b/drivers/gpu/drm/i915/gt/intel_reset.c
@@ -687,7 +687,6 @@ static void reset_prepare_engine(struct intel_engine_cs *engine)
 	 * written to the powercontext is undefined and so we may lose
 	 * GPU state upon resume, i.e. fail to restart after a reset.
 	 */
-	intel_engine_pm_get(engine);
 	intel_uncore_forcewake_get(engine->uncore, FORCEWAKE_ALL);
 	engine->reset.prepare(engine);
 }
@@ -718,16 +717,21 @@ static void revoke_mmaps(struct drm_i915_private *i915)
 	}
 }
 
-static void reset_prepare(struct drm_i915_private *i915)
+static intel_engine_mask_t reset_prepare(struct drm_i915_private *i915)
 {
 	struct intel_engine_cs *engine;
+	intel_engine_mask_t awake = 0;
 	enum intel_engine_id id;
 
-	intel_gt_pm_get(i915);
-	for_each_engine(engine, i915, id)
+	for_each_engine(engine, i915, id) {
+		if (intel_engine_pm_get_if_awake(engine))
+			awake |= engine->mask;
 		reset_prepare_engine(engine);
+	}
 
 	intel_uc_reset_prepare(i915);
+
+	return awake;
 }
 
 static void gt_revoke(struct drm_i915_private *i915)
@@ -761,20 +765,22 @@ static int gt_reset(struct drm_i915_private *i915,
 static void reset_finish_engine(struct intel_engine_cs *engine)
 {
 	engine->reset.finish(engine);
-	intel_engine_pm_put(engine);
 	intel_uncore_forcewake_put(engine->uncore, FORCEWAKE_ALL);
+
+	intel_engine_signal_breadcrumbs(engine);
 }
 
-static void reset_finish(struct drm_i915_private *i915)
+static void reset_finish(struct drm_i915_private *i915,
+			 intel_engine_mask_t awake)
 {
 	struct intel_engine_cs *engine;
 	enum intel_engine_id id;
 
 	for_each_engine(engine, i915, id) {
 		reset_finish_engine(engine);
-		intel_engine_signal_breadcrumbs(engine);
+		if (awake & engine->mask)
+			intel_engine_pm_put(engine);
 	}
-	intel_gt_pm_put(i915);
 }
 
 static void nop_submit_request(struct i915_request *request)
@@ -798,6 +804,7 @@ static void __i915_gem_set_wedged(struct drm_i915_private *i915)
 {
 	struct i915_gpu_error *error = &i915->gpu_error;
 	struct intel_engine_cs *engine;
+	intel_engine_mask_t awake;
 	enum intel_engine_id id;
 
 	if (test_bit(I915_WEDGED, &error->flags))
@@ -817,7 +824,7 @@ static void __i915_gem_set_wedged(struct drm_i915_private *i915)
 	 * rolling the global seqno forward (since this would complete requests
 	 * for which we haven't set the fence error to EIO yet).
 	 */
-	reset_prepare(i915);
+	awake = reset_prepare(i915);
 
 	/* Even if the GPU reset fails, it should still stop the engines */
 	if (!INTEL_INFO(i915)->gpu_reset_clobbers_display)
@@ -841,7 +848,7 @@ static void __i915_gem_set_wedged(struct drm_i915_private *i915)
 	for_each_engine(engine, i915, id)
 		engine->cancel_requests(engine);
 
-	reset_finish(i915);
+	reset_finish(i915, awake);
 
 	GEM_TRACE("end\n");
 }
@@ -951,6 +958,21 @@ static int do_reset(struct drm_i915_private *i915,
 	return gt_reset(i915, stalled_mask);
 }
 
+static int resume(struct drm_i915_private *i915)
+{
+	struct intel_engine_cs *engine;
+	enum intel_engine_id id;
+	int ret;
+
+	for_each_engine(engine, i915, id) {
+		ret = engine->resume(engine);
+		if (ret)
+			return ret;
+	}
+
+	return 0;
+}
+
 /**
  * i915_reset - reset chip after a hang
  * @i915: #drm_i915_private to reset
@@ -973,6 +995,7 @@ void i915_reset(struct drm_i915_private *i915,
 		const char *reason)
 {
 	struct i915_gpu_error *error = &i915->gpu_error;
+	intel_engine_mask_t awake;
 	int ret;
 
 	GEM_TRACE("flags=%lx\n", error->flags);
@@ -989,7 +1012,7 @@ void i915_reset(struct drm_i915_private *i915,
 		dev_notice(i915->drm.dev, "Resetting chip for %s\n", reason);
 	error->reset_count++;
 
-	reset_prepare(i915);
+	awake = reset_prepare(i915);
 
 	if (!intel_has_gpu_reset(i915)) {
 		if (i915_modparams.reset)
@@ -1024,13 +1047,17 @@ void i915_reset(struct drm_i915_private *i915,
 	if (ret) {
 		DRM_ERROR("Failed to initialise HW following reset (%d)\n",
 			  ret);
-		goto error;
+		goto taint;
 	}
 
+	ret = resume(i915);
+	if (ret)
+		goto taint;
+
 	i915_queue_hangcheck(i915);
 
 finish:
-	reset_finish(i915);
+	reset_finish(i915, awake);
 unlock:
 	mutex_unlock(&error->wedge_mutex);
 	return;
@@ -1081,7 +1108,7 @@ int i915_reset_engine(struct intel_engine_cs *engine, const char *msg)
 	GEM_TRACE("%s flags=%lx\n", engine->name, error->flags);
 	GEM_BUG_ON(!test_bit(I915_RESET_ENGINE + engine->id, &error->flags));
 
-	if (!intel_wakeref_active(&engine->wakeref))
+	if (!intel_engine_pm_get_if_awake(engine))
 		return 0;
 
 	reset_prepare_engine(engine);
@@ -1116,12 +1143,11 @@ int i915_reset_engine(struct intel_engine_cs *engine, const char *msg)
 	 * process to program RING_MODE, HWSP and re-enable submission.
 	 */
 	ret = engine->resume(engine);
-	if (ret)
-		goto out;
 
 out:
 	intel_engine_cancel_stop_cs(engine);
 	reset_finish_engine(engine);
+	intel_engine_pm_put(engine);
 	return ret;
 }
 
diff --git a/drivers/gpu/drm/i915/gt/intel_ringbuffer.c b/drivers/gpu/drm/i915/gt/intel_ringbuffer.c
index c6023bc9452d..12010e798868 100644
--- a/drivers/gpu/drm/i915/gt/intel_ringbuffer.c
+++ b/drivers/gpu/drm/i915/gt/intel_ringbuffer.c
@@ -1149,16 +1149,16 @@ i915_emit_bb_start(struct i915_request *rq,
 int intel_ring_pin(struct intel_ring *ring)
 {
 	struct i915_vma *vma = ring->vma;
-	enum i915_map_type map = i915_coherent_map_type(vma->vm->i915);
 	unsigned int flags;
 	void *addr;
 	int ret;
 
-	GEM_BUG_ON(ring->vaddr);
+	if (atomic_fetch_inc(&ring->pin_count))
+		return 0;
 
 	ret = i915_timeline_pin(ring->timeline);
 	if (ret)
-		return ret;
+		goto err_unpin;
 
 	flags = PIN_GLOBAL;
 
@@ -1172,26 +1172,31 @@ int intel_ring_pin(struct intel_ring *ring)
 
 	ret = i915_vma_pin(vma, 0, 0, flags);
 	if (unlikely(ret))
-		goto unpin_timeline;
+		goto err_timeline;
 
 	if (i915_vma_is_map_and_fenceable(vma))
 		addr = (void __force *)i915_vma_pin_iomap(vma);
 	else
-		addr = i915_gem_object_pin_map(vma->obj, map);
+		addr = i915_gem_object_pin_map(vma->obj,
+					       i915_coherent_map_type(vma->vm->i915));
 	if (IS_ERR(addr)) {
 		ret = PTR_ERR(addr);
-		goto unpin_ring;
+		goto err_ring;
 	}
 
 	vma->obj->pin_global++;
 
+	GEM_BUG_ON(ring->vaddr);
 	ring->vaddr = addr;
+
 	return 0;
 
-unpin_ring:
+err_ring:
 	i915_vma_unpin(vma);
-unpin_timeline:
+err_timeline:
 	i915_timeline_unpin(ring->timeline);
+err_unpin:
+	atomic_dec(&ring->pin_count);
 	return ret;
 }
 
@@ -1207,16 +1212,19 @@ void intel_ring_reset(struct intel_ring *ring, u32 tail)
 
 void intel_ring_unpin(struct intel_ring *ring)
 {
-	GEM_BUG_ON(!ring->vma);
-	GEM_BUG_ON(!ring->vaddr);
+	if (!atomic_dec_and_test(&ring->pin_count))
+		return;
 
 	/* Discard any unused bytes beyond that submitted to hw. */
 	intel_ring_reset(ring, ring->tail);
 
+	GEM_BUG_ON(!ring->vma);
 	if (i915_vma_is_map_and_fenceable(ring->vma))
 		i915_vma_unpin_iomap(ring->vma);
 	else
 		i915_gem_object_unpin_map(ring->vma->obj);
+
+	GEM_BUG_ON(!ring->vaddr);
 	ring->vaddr = NULL;
 
 	ring->vma->obj->pin_global--;
@@ -2081,10 +2089,11 @@ static void ring_destroy(struct intel_engine_cs *engine)
 	WARN_ON(INTEL_GEN(dev_priv) > 2 &&
 		(ENGINE_READ(engine, RING_MI_MODE) & MODE_IDLE) == 0);
 
+	intel_engine_cleanup_common(engine);
+
 	intel_ring_unpin(engine->buffer);
 	intel_ring_put(engine->buffer);
 
-	intel_engine_cleanup_common(engine);
 	kfree(engine);
 }
 
diff --git a/drivers/gpu/drm/i915/gt/intel_workarounds.c b/drivers/gpu/drm/i915/gt/intel_workarounds.c
index 15e90fd2cfdc..98dfb086320f 100644
--- a/drivers/gpu/drm/i915/gt/intel_workarounds.c
+++ b/drivers/gpu/drm/i915/gt/intel_workarounds.c
@@ -1098,10 +1098,25 @@ static void glk_whitelist_build(struct intel_engine_cs *engine)
 
 static void cfl_whitelist_build(struct intel_engine_cs *engine)
 {
+	struct i915_wa_list *w = &engine->whitelist;
+
 	if (engine->class != RENDER_CLASS)
 		return;
 
-	gen9_whitelist_build(&engine->whitelist);
+	gen9_whitelist_build(w);
+
+	/*
+	 * WaAllowPMDepthAndInvocationCountAccessFromUMD:cfl,whl,cml,aml
+	 *
+	 * This covers 4 register which are next to one another :
+	 *   - PS_INVOCATION_COUNT
+	 *   - PS_INVOCATION_COUNT_UDW
+	 *   - PS_DEPTH_COUNT
+	 *   - PS_DEPTH_COUNT_UDW
+	 */
+	whitelist_reg_ext(w, PS_INVOCATION_COUNT,
+			  RING_FORCE_TO_NONPRIV_RD |
+			  RING_FORCE_TO_NONPRIV_RANGE_4);
 }
 
 static void cnl_whitelist_build(struct intel_engine_cs *engine)
@@ -1129,6 +1144,19 @@ static void icl_whitelist_build(struct intel_engine_cs *engine)
 
 		/* WaEnableStateCacheRedirectToCS:icl */
 		whitelist_reg(w, GEN9_SLICE_COMMON_ECO_CHICKEN1);
+
+		/*
+		 * WaAllowPMDepthAndInvocationCountAccessFromUMD:icl
+		 *
+		 * This covers 4 register which are next to one another :
+		 *   - PS_INVOCATION_COUNT
+		 *   - PS_INVOCATION_COUNT_UDW
+		 *   - PS_DEPTH_COUNT
+		 *   - PS_DEPTH_COUNT_UDW
+		 */
+		whitelist_reg_ext(w, PS_INVOCATION_COUNT,
+				  RING_FORCE_TO_NONPRIV_RD |
+				  RING_FORCE_TO_NONPRIV_RANGE_4);
 		break;
 
 	case VIDEO_DECODE_CLASS:
@@ -1258,8 +1286,12 @@ rcs_engine_wa_init(struct intel_engine_cs *engine, struct i915_wa_list *wal)
 		if (IS_ICL_REVID(i915, ICL_REVID_A0, ICL_REVID_B0))
 			wa_write_or(wal,
 				    GEN7_SARCHKMD,
-				    GEN7_DISABLE_DEMAND_PREFETCH |
-				    GEN7_DISABLE_SAMPLER_PREFETCH);
+				    GEN7_DISABLE_DEMAND_PREFETCH);
+
+		/* Wa_1606682166:icl */
+		wa_write_or(wal,
+			    GEN7_SARCHKMD,
+			    GEN7_DISABLE_SAMPLER_PREFETCH);
 	}
 
 	if (IS_GEN_RANGE(i915, 9, 11)) {
diff --git a/drivers/gpu/drm/i915/gt/mock_engine.c b/drivers/gpu/drm/i915/gt/mock_engine.c
index 086801b51441..486c6953dcb1 100644
--- a/drivers/gpu/drm/i915/gt/mock_engine.c
+++ b/drivers/gpu/drm/i915/gt/mock_engine.c
@@ -66,6 +66,7 @@ static struct intel_ring *mock_ring(struct intel_engine_cs *engine)
 	ring->base.effective_size = sz;
 	ring->base.vaddr = (void *)(ring + 1);
 	ring->base.timeline = &ring->timeline;
+	atomic_set(&ring->base.pin_count, 1);
 
 	INIT_LIST_HEAD(&ring->base.request_list);
 	intel_ring_update_space(&ring->base);
diff --git a/drivers/gpu/drm/i915/gt/selftest_reset.c b/drivers/gpu/drm/i915/gt/selftest_reset.c
index 89da9e7cc1ba..b5c590c9ccba 100644
--- a/drivers/gpu/drm/i915/gt/selftest_reset.c
+++ b/drivers/gpu/drm/i915/gt/selftest_reset.c
@@ -71,13 +71,16 @@ static int igt_atomic_reset(void *arg)
 		goto unlock;
 
 	for (p = igt_atomic_phases; p->name; p++) {
+		intel_engine_mask_t awake;
+
 		GEM_TRACE("intel_gpu_reset under %s\n", p->name);
 
+		awake = reset_prepare(i915);
 		p->critical_section_begin();
 		reset_prepare(i915);
 		err = intel_gpu_reset(i915, ALL_ENGINES);
-		reset_finish(i915);
 		p->critical_section_end();
+		reset_finish(i915, awake);
 
 		if (err) {
 			pr_err("intel_gpu_reset failed under %s\n", p->name);
diff --git a/drivers/gpu/drm/i915/gt/selftest_workarounds.c b/drivers/gpu/drm/i915/gt/selftest_workarounds.c
index 9eaf030affd0..44becd9538be 100644
--- a/drivers/gpu/drm/i915/gt/selftest_workarounds.c
+++ b/drivers/gpu/drm/i915/gt/selftest_workarounds.c
@@ -925,7 +925,12 @@ check_whitelisted_registers(struct intel_engine_cs *engine,
 
 	err = 0;
 	for (i = 0; i < engine->whitelist.count; i++) {
-		if (!fn(engine, a[i], b[i], engine->whitelist.list[i].reg))
+		const struct i915_wa *wa = &engine->whitelist.list[i];
+
+		if (i915_mmio_reg_offset(wa->reg) & RING_FORCE_TO_NONPRIV_RD)
+			continue;
+
+		if (!fn(engine, a[i], b[i], wa->reg))
 			err = -EINVAL;
 	}
 
diff --git a/drivers/gpu/drm/i915/gvt/cmd_parser.c b/drivers/gpu/drm/i915/gvt/cmd_parser.c
index 6ea88270c818..b09dc315e2da 100644
--- a/drivers/gpu/drm/i915/gvt/cmd_parser.c
+++ b/drivers/gpu/drm/i915/gvt/cmd_parser.c
@@ -2674,11 +2674,6 @@ static int scan_workload(struct intel_vgpu_workload *workload)
 		gma_head == gma_tail)
 		return 0;
 
-	if (!intel_gvt_ggtt_validate_range(s.vgpu, s.ring_start, s.ring_size)) {
-		ret = -EINVAL;
-		goto out;
-	}
-
 	ret = ip_gma_set(&s, gma_head);
 	if (ret)
 		goto out;
@@ -2724,11 +2719,6 @@ static int scan_wa_ctx(struct intel_shadow_wa_ctx *wa_ctx)
 	s.workload = workload;
 	s.is_ctx_wa = true;
 
-	if (!intel_gvt_ggtt_validate_range(s.vgpu, s.ring_start, s.ring_size)) {
-		ret = -EINVAL;
-		goto out;
-	}
-
 	ret = ip_gma_set(&s, gma_head);
 	if (ret)
 		goto out;
diff --git a/drivers/gpu/drm/i915/gvt/fb_decoder.c b/drivers/gpu/drm/i915/gvt/fb_decoder.c
index 65e847392aea..8bb292b01271 100644
--- a/drivers/gpu/drm/i915/gvt/fb_decoder.c
+++ b/drivers/gpu/drm/i915/gvt/fb_decoder.c
@@ -245,7 +245,7 @@ int intel_vgpu_decode_primary_plane(struct intel_vgpu *vgpu,
 	plane->hw_format = fmt;
 
 	plane->base = vgpu_vreg_t(vgpu, DSPSURF(pipe)) & I915_GTT_PAGE_MASK;
-	if (!intel_gvt_ggtt_validate_range(vgpu, plane->base, 0))
+	if (!vgpu_gmadr_is_valid(vgpu, plane->base))
 		return  -EINVAL;
 
 	plane->base_gpa = intel_vgpu_gma_to_gpa(vgpu->gtt.ggtt_mm, plane->base);
@@ -368,7 +368,7 @@ int intel_vgpu_decode_cursor_plane(struct intel_vgpu *vgpu,
 			alpha_plane, alpha_force);
 
 	plane->base = vgpu_vreg_t(vgpu, CURBASE(pipe)) & I915_GTT_PAGE_MASK;
-	if (!intel_gvt_ggtt_validate_range(vgpu, plane->base, 0))
+	if (!vgpu_gmadr_is_valid(vgpu, plane->base))
 		return  -EINVAL;
 
 	plane->base_gpa = intel_vgpu_gma_to_gpa(vgpu->gtt.ggtt_mm, plane->base);
@@ -472,7 +472,7 @@ int intel_vgpu_decode_sprite_plane(struct intel_vgpu *vgpu,
 	plane->drm_format = drm_format;
 
 	plane->base = vgpu_vreg_t(vgpu, SPRSURF(pipe)) & I915_GTT_PAGE_MASK;
-	if (!intel_gvt_ggtt_validate_range(vgpu, plane->base, 0))
+	if (!vgpu_gmadr_is_valid(vgpu, plane->base))
 		return  -EINVAL;
 
 	plane->base_gpa = intel_vgpu_gma_to_gpa(vgpu->gtt.ggtt_mm, plane->base);
diff --git a/drivers/gpu/drm/i915/gvt/gtt.c b/drivers/gpu/drm/i915/gvt/gtt.c
index 53115bdae12b..4b04af569c05 100644
--- a/drivers/gpu/drm/i915/gvt/gtt.c
+++ b/drivers/gpu/drm/i915/gvt/gtt.c
@@ -2141,11 +2141,20 @@ static int emulate_ggtt_mmio_read(struct intel_vgpu *vgpu,
 	struct intel_vgpu_mm *ggtt_mm = vgpu->gtt.ggtt_mm;
 	const struct intel_gvt_device_info *info = &vgpu->gvt->device_info;
 	unsigned long index = off >> info->gtt_entry_size_shift;
+	unsigned long gma;
 	struct intel_gvt_gtt_entry e;
 
 	if (bytes != 4 && bytes != 8)
 		return -EINVAL;
 
+	gma = index << I915_GTT_PAGE_SHIFT;
+	if (!intel_gvt_ggtt_validate_range(vgpu,
+					   gma, 1 << I915_GTT_PAGE_SHIFT)) {
+		gvt_dbg_mm("read invalid ggtt at 0x%lx\n", gma);
+		memset(p_data, 0, bytes);
+		return 0;
+	}
+
 	ggtt_get_guest_entry(ggtt_mm, &e, index);
 	memcpy(p_data, (void *)&e.val64 + (off & (info->gtt_entry_size - 1)),
 			bytes);
diff --git a/drivers/gpu/drm/i915/gvt/kvmgt.c b/drivers/gpu/drm/i915/gvt/kvmgt.c
index 144301b778df..23aa3e50cbf8 100644
--- a/drivers/gpu/drm/i915/gvt/kvmgt.c
+++ b/drivers/gpu/drm/i915/gvt/kvmgt.c
@@ -1911,6 +1911,18 @@ static int kvmgt_dma_map_guest_page(unsigned long handle, unsigned long gfn,
 		ret = __gvt_cache_add(info->vgpu, gfn, *dma_addr, size);
 		if (ret)
 			goto err_unmap;
+	} else if (entry->size != size) {
+		/* the same gfn with different size: unmap and re-map */
+		gvt_dma_unmap_page(vgpu, gfn, entry->dma_addr, entry->size);
+		__gvt_cache_remove_entry(vgpu, entry);
+
+		ret = gvt_dma_map_page(vgpu, gfn, dma_addr, size);
+		if (ret)
+			goto err_unlock;
+
+		ret = __gvt_cache_add(info->vgpu, gfn, *dma_addr, size);
+		if (ret)
+			goto err_unmap;
 	} else {
 		kref_get(&entry->ref);
 		*dma_addr = entry->dma_addr;
diff --git a/drivers/gpu/drm/i915/gvt/scheduler.c b/drivers/gpu/drm/i915/gvt/scheduler.c
index 2144fb46d0e1..9f3fd7d96a69 100644
--- a/drivers/gpu/drm/i915/gvt/scheduler.c
+++ b/drivers/gpu/drm/i915/gvt/scheduler.c
@@ -364,16 +364,13 @@ static void release_shadow_wa_ctx(struct intel_shadow_wa_ctx *wa_ctx)
 	wa_ctx->indirect_ctx.shadow_va = NULL;
 }
 
-static int set_context_ppgtt_from_shadow(struct intel_vgpu_workload *workload,
-					 struct i915_gem_context *ctx)
+static void set_context_ppgtt_from_shadow(struct intel_vgpu_workload *workload,
+					  struct i915_gem_context *ctx)
 {
 	struct intel_vgpu_mm *mm = workload->shadow_mm;
 	struct i915_ppgtt *ppgtt = i915_vm_to_ppgtt(ctx->vm);
 	int i = 0;
 
-	if (mm->type != INTEL_GVT_MM_PPGTT || !mm->ppgtt_mm.shadowed)
-		return -EINVAL;
-
 	if (mm->ppgtt_mm.root_entry_type == GTT_TYPE_PPGTT_ROOT_L4_ENTRY) {
 		px_dma(ppgtt->pd) = mm->ppgtt_mm.shadow_pdps[0];
 	} else {
@@ -384,8 +381,6 @@ static int set_context_ppgtt_from_shadow(struct intel_vgpu_workload *workload,
 			px_dma(pd) = mm->ppgtt_mm.shadow_pdps[i];
 		}
 	}
-
-	return 0;
 }
 
 static int
@@ -614,6 +609,8 @@ static void release_shadow_batch_buffer(struct intel_vgpu_workload *workload)
 static int prepare_workload(struct intel_vgpu_workload *workload)
 {
 	struct intel_vgpu *vgpu = workload->vgpu;
+	struct intel_vgpu_submission *s = &vgpu->submission;
+	int ring = workload->ring_id;
 	int ret = 0;
 
 	ret = intel_vgpu_pin_mm(workload->shadow_mm);
@@ -622,8 +619,16 @@ static int prepare_workload(struct intel_vgpu_workload *workload)
 		return ret;
 	}
 
+	if (workload->shadow_mm->type != INTEL_GVT_MM_PPGTT ||
+	    !workload->shadow_mm->ppgtt_mm.shadowed) {
+		gvt_vgpu_err("workload shadow ppgtt isn't ready\n");
+		return -EINVAL;
+	}
+
 	update_shadow_pdps(workload);
 
+	set_context_ppgtt_from_shadow(workload, s->shadow[ring]->gem_context);
+
 	ret = intel_vgpu_sync_oos_pages(workload->vgpu);
 	if (ret) {
 		gvt_vgpu_err("fail to vgpu sync oos pages\n");
@@ -674,7 +679,6 @@ static int dispatch_workload(struct intel_vgpu_workload *workload)
 {
 	struct intel_vgpu *vgpu = workload->vgpu;
 	struct drm_i915_private *dev_priv = vgpu->gvt->dev_priv;
-	struct intel_vgpu_submission *s = &vgpu->submission;
 	struct i915_request *rq;
 	int ring_id = workload->ring_id;
 	int ret;
@@ -685,13 +689,6 @@ static int dispatch_workload(struct intel_vgpu_workload *workload)
 	mutex_lock(&vgpu->vgpu_lock);
 	mutex_lock(&dev_priv->drm.struct_mutex);
 
-	ret = set_context_ppgtt_from_shadow(workload,
-					    s->shadow[ring_id]->gem_context);
-	if (ret < 0) {
-		gvt_vgpu_err("workload shadow ppgtt isn't ready\n");
-		goto err_req;
-	}
-
 	ret = intel_gvt_workload_req_alloc(workload);
 	if (ret)
 		goto err_req;
@@ -990,6 +987,7 @@ static int workload_thread(void *priv)
 	int ret;
 	bool need_force_wake = (INTEL_GEN(gvt->dev_priv) >= 9);
 	DEFINE_WAIT_FUNC(wait, woken_wake_function);
+	struct intel_runtime_pm *rpm = &gvt->dev_priv->runtime_pm;
 
 	kfree(p);
 
@@ -1013,6 +1011,8 @@ static int workload_thread(void *priv)
 				workload->ring_id, workload,
 				workload->vgpu->id);
 
+		intel_runtime_pm_get(rpm);
+
 		gvt_dbg_sched("ring id %d will dispatch workload %p\n",
 				workload->ring_id, workload);
 
@@ -1042,6 +1042,7 @@ complete:
 			intel_uncore_forcewake_put(&gvt->dev_priv->uncore,
 					FORCEWAKE_ALL);
 
+		intel_runtime_pm_put_unchecked(rpm);
 		if (ret && (vgpu_is_vm_unhealthy(ret)))
 			enter_failsafe_mode(vgpu, GVT_FAILSAFE_GUEST_ERR);
 	}
@@ -1492,6 +1493,12 @@ intel_vgpu_create_workload(struct intel_vgpu *vgpu, int ring_id,
 	intel_gvt_hypervisor_read_gpa(vgpu, ring_context_gpa +
 			RING_CTX_OFF(ctx_ctrl.val), &ctx_ctl, 4);
 
+	if (!intel_gvt_ggtt_validate_range(vgpu, start,
+				_RING_CTL_BUF_SIZE(ctl))) {
+		gvt_vgpu_err("context contain invalid rb at: 0x%x\n", start);
+		return ERR_PTR(-EINVAL);
+	}
+
 	workload = alloc_workload(vgpu);
 	if (IS_ERR(workload))
 		return workload;
@@ -1516,9 +1523,31 @@ intel_vgpu_create_workload(struct intel_vgpu *vgpu, int ring_id,
 		workload->wa_ctx.indirect_ctx.size =
 			(indirect_ctx & INDIRECT_CTX_SIZE_MASK) *
 			CACHELINE_BYTES;
+
+		if (workload->wa_ctx.indirect_ctx.size != 0) {
+			if (!intel_gvt_ggtt_validate_range(vgpu,
+				workload->wa_ctx.indirect_ctx.guest_gma,
+				workload->wa_ctx.indirect_ctx.size)) {
+				kmem_cache_free(s->workloads, workload);
+				gvt_vgpu_err("invalid wa_ctx at: 0x%lx\n",
+				    workload->wa_ctx.indirect_ctx.guest_gma);
+				return ERR_PTR(-EINVAL);
+			}
+		}
+
 		workload->wa_ctx.per_ctx.guest_gma =
 			per_ctx & PER_CTX_ADDR_MASK;
 		workload->wa_ctx.per_ctx.valid = per_ctx & 1;
+		if (workload->wa_ctx.per_ctx.valid) {
+			if (!intel_gvt_ggtt_validate_range(vgpu,
+				workload->wa_ctx.per_ctx.guest_gma,
+				CACHELINE_BYTES)) {
+				kmem_cache_free(s->workloads, workload);
+				gvt_vgpu_err("invalid per_ctx at: 0x%lx\n",
+					workload->wa_ctx.per_ctx.guest_gma);
+				return ERR_PTR(-EINVAL);
+			}
+		}
 	}
 
 	gvt_dbg_el("workload %p ring id %d head %x tail %x start %x ctl %x\n",
diff --git a/drivers/gpu/drm/i915/gvt/trace_points.c b/drivers/gpu/drm/i915/gvt/trace_points.c
index a3deed692b9c..fe552e877e09 100644
--- a/drivers/gpu/drm/i915/gvt/trace_points.c
+++ b/drivers/gpu/drm/i915/gvt/trace_points.c
@@ -28,8 +28,6 @@
  *
  */
 
-#include "trace.h"
-
 #ifndef __CHECKER__
 #define CREATE_TRACE_POINTS
 #include "trace.h"
diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index bc909ec5d9c3..fe7a6ec2c199 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -1674,8 +1674,9 @@ struct drm_i915_private {
 	} dram_info;
 
 	struct intel_bw_info {
-		int num_planes;
-		int deratedbw[3];
+		unsigned int deratedbw[3]; /* for each QGV point */
+		u8 num_qgv_points;
+		u8 num_planes;
 	} max_bw[6];
 
 	struct drm_private_obj bw_obj;
diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index 190ad54fb072..8a659d3d7435 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -46,7 +46,6 @@
 #include "gem/i915_gem_ioctls.h"
 #include "gem/i915_gem_pm.h"
 #include "gem/i915_gemfs.h"
-#include "gt/intel_engine_pm.h"
 #include "gt/intel_gt_pm.h"
 #include "gt/intel_mocs.h"
 #include "gt/intel_reset.h"
@@ -1307,21 +1306,13 @@ int i915_gem_init_hw(struct drm_i915_private *dev_priv)
 
 	intel_mocs_init_l3cc_table(dev_priv);
 
-	/* Only when the HW is re-initialised, can we replay the requests */
-	ret = intel_engines_resume(dev_priv);
-	if (ret)
-		goto cleanup_uc;
-
 	intel_uncore_forcewake_put(&dev_priv->uncore, FORCEWAKE_ALL);
 
 	intel_engines_set_scheduler_caps(dev_priv);
 	return 0;
 
-cleanup_uc:
-	intel_uc_fini_hw(dev_priv);
 out:
 	intel_uncore_forcewake_put(&dev_priv->uncore, FORCEWAKE_ALL);
-
 	return ret;
 }
 
@@ -1580,6 +1571,11 @@ int i915_gem_init(struct drm_i915_private *dev_priv)
 	if (ret)
 		goto err_uc_init;
 
+	/* Only when the HW is re-initialised, can we replay the requests */
+	ret = intel_gt_resume(dev_priv);
+	if (ret)
+		goto err_init_hw;
+
 	/*
 	 * Despite its name intel_init_clock_gating applies both display
 	 * clock gating workarounds; GT mmio workarounds and the occasional
@@ -1593,20 +1589,20 @@ int i915_gem_init(struct drm_i915_private *dev_priv)
 
 	ret = intel_engines_verify_workarounds(dev_priv);
 	if (ret)
-		goto err_init_hw;
+		goto err_gt;
 
 	ret = __intel_engines_record_defaults(dev_priv);
 	if (ret)
-		goto err_init_hw;
+		goto err_gt;
 
 	if (i915_inject_load_failure()) {
 		ret = -ENODEV;
-		goto err_init_hw;
+		goto err_gt;
 	}
 
 	if (i915_inject_load_failure()) {
 		ret = -EIO;
-		goto err_init_hw;
+		goto err_gt;
 	}
 
 	intel_uncore_forcewake_put(&dev_priv->uncore, FORCEWAKE_ALL);
@@ -1620,7 +1616,7 @@ int i915_gem_init(struct drm_i915_private *dev_priv)
 	 * HW as irrevisibly wedged, but keep enough state around that the
 	 * driver doesn't explode during runtime.
 	 */
-err_init_hw:
+err_gt:
 	mutex_unlock(&dev_priv->drm.struct_mutex);
 
 	i915_gem_set_wedged(dev_priv);
@@ -1630,6 +1626,7 @@ err_init_hw:
 	i915_gem_drain_workqueue(dev_priv);
 
 	mutex_lock(&dev_priv->drm.struct_mutex);
+err_init_hw:
 	intel_uc_fini_hw(dev_priv);
 err_uc_init:
 	intel_uc_fini(dev_priv);
diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c
index 8ab820145ea6..7015a97b1097 100644
--- a/drivers/gpu/drm/i915/i915_gem_gtt.c
+++ b/drivers/gpu/drm/i915/i915_gem_gtt.c
@@ -1444,9 +1444,11 @@ unwind_pd:
 	spin_lock(&pdp->lock);
 	if (atomic_dec_and_test(&pd->used)) {
 		gen8_ppgtt_set_pdpe(pdp, vm->scratch_pd, pdpe);
+		pdp->entry[pdpe] = vm->scratch_pd;
 		GEM_BUG_ON(!atomic_read(&pdp->used));
 		atomic_dec(&pdp->used);
-		free_pd(vm, pd);
+		GEM_BUG_ON(alloc);
+		alloc = pd; /* defer the free to after the lock */
 	}
 	spin_unlock(&pdp->lock);
 unwind:
@@ -1515,7 +1517,9 @@ unwind_pdp:
 	spin_lock(&pml4->lock);
 	if (atomic_dec_and_test(&pdp->used)) {
 		gen8_ppgtt_set_pml4e(pml4, vm->scratch_pdp, pml4e);
-		free_pd(vm, pdp);
+		pml4->entry[pml4e] = vm->scratch_pdp;
+		GEM_BUG_ON(alloc);
+		alloc = pdp; /* defer the free until after the lock */
 	}
 	spin_unlock(&pml4->lock);
 unwind:
diff --git a/drivers/gpu/drm/i915/i915_gpu_error.c b/drivers/gpu/drm/i915/i915_gpu_error.c
index 41a511d5267f..8bc76fcff70d 100644
--- a/drivers/gpu/drm/i915/i915_gpu_error.c
+++ b/drivers/gpu/drm/i915/i915_gpu_error.c
@@ -1418,6 +1418,7 @@ static void gem_record_rings(struct i915_gpu_state *error)
 		struct intel_engine_cs *engine = i915->engine[i];
 		struct drm_i915_error_engine *ee = &error->engine[i];
 		struct i915_request *request;
+		unsigned long flags;
 
 		ee->engine_id = -1;
 
@@ -1429,10 +1430,11 @@ static void gem_record_rings(struct i915_gpu_state *error)
 		error_record_engine_registers(error, engine, ee);
 		error_record_engine_execlists(engine, ee);
 
+		spin_lock_irqsave(&engine->active.lock, flags);
 		request = intel_engine_find_active_request(engine);
 		if (request) {
 			struct i915_gem_context *ctx = request->gem_context;
-			struct intel_ring *ring;
+			struct intel_ring *ring = request->ring;
 
 			ee->vm = ctx->vm ?: &ggtt->vm;
 
@@ -1462,7 +1464,6 @@ static void gem_record_rings(struct i915_gpu_state *error)
 			ee->rq_post = request->postfix;
 			ee->rq_tail = request->tail;
 
-			ring = request->ring;
 			ee->cpu_ring_head = ring->head;
 			ee->cpu_ring_tail = ring->tail;
 			ee->ringbuffer =
@@ -1470,6 +1471,7 @@ static void gem_record_rings(struct i915_gpu_state *error)
 
 			engine_record_requests(engine, request, ee);
 		}
+		spin_unlock_irqrestore(&engine->active.lock, flags);
 
 		ee->hws_page =
 			i915_error_object_create(i915,
diff --git a/drivers/gpu/drm/i915/i915_perf.c b/drivers/gpu/drm/i915/i915_perf.c
index a700c5c3d167..5140017f9a39 100644
--- a/drivers/gpu/drm/i915/i915_perf.c
+++ b/drivers/gpu/drm/i915/i915_perf.c
@@ -1567,28 +1567,10 @@ static void config_oa_regs(struct drm_i915_private *dev_priv,
 	}
 }
 
-static int hsw_enable_metric_set(struct i915_perf_stream *stream)
+static void delay_after_mux(void)
 {
-	struct drm_i915_private *dev_priv = stream->dev_priv;
-	const struct i915_oa_config *oa_config = stream->oa_config;
-
-	/* PRM:
-	 *
-	 * OA unit is using “crclk” for its functionality. When trunk
-	 * level clock gating takes place, OA clock would be gated,
-	 * unable to count the events from non-render clock domain.
-	 * Render clock gating must be disabled when OA is enabled to
-	 * count the events from non-render domain. Unit level clock
-	 * gating for RCS should also be disabled.
-	 */
-	I915_WRITE(GEN7_MISCCPCTL, (I915_READ(GEN7_MISCCPCTL) &
-				    ~GEN7_DOP_CLOCK_GATE_ENABLE));
-	I915_WRITE(GEN6_UCGCTL1, (I915_READ(GEN6_UCGCTL1) |
-				  GEN6_CSUNIT_CLOCK_GATE_DISABLE));
-
-	config_oa_regs(dev_priv, oa_config->mux_regs, oa_config->mux_regs_len);
-
-	/* It apparently takes a fairly long time for a new MUX
+	/*
+	 * It apparently takes a fairly long time for a new MUX
 	 * configuration to be be applied after these register writes.
 	 * This delay duration was derived empirically based on the
 	 * render_basic config but hopefully it covers the maximum
@@ -1610,6 +1592,30 @@ static int hsw_enable_metric_set(struct i915_perf_stream *stream)
 	 * a delay at this location would mitigate any invalid reports.
 	 */
 	usleep_range(15000, 20000);
+}
+
+static int hsw_enable_metric_set(struct i915_perf_stream *stream)
+{
+	struct drm_i915_private *dev_priv = stream->dev_priv;
+	const struct i915_oa_config *oa_config = stream->oa_config;
+
+	/*
+	 * PRM:
+	 *
+	 * OA unit is using “crclk” for its functionality. When trunk
+	 * level clock gating takes place, OA clock would be gated,
+	 * unable to count the events from non-render clock domain.
+	 * Render clock gating must be disabled when OA is enabled to
+	 * count the events from non-render domain. Unit level clock
+	 * gating for RCS should also be disabled.
+	 */
+	I915_WRITE(GEN7_MISCCPCTL, (I915_READ(GEN7_MISCCPCTL) &
+				    ~GEN7_DOP_CLOCK_GATE_ENABLE));
+	I915_WRITE(GEN6_UCGCTL1, (I915_READ(GEN6_UCGCTL1) |
+				  GEN6_CSUNIT_CLOCK_GATE_DISABLE));
+
+	config_oa_regs(dev_priv, oa_config->mux_regs, oa_config->mux_regs_len);
+	delay_after_mux();
 
 	config_oa_regs(dev_priv, oa_config->b_counter_regs,
 		       oa_config->b_counter_regs_len);
@@ -1835,6 +1841,7 @@ static int gen8_enable_metric_set(struct i915_perf_stream *stream)
 		return ret;
 
 	config_oa_regs(dev_priv, oa_config->mux_regs, oa_config->mux_regs_len);
+	delay_after_mux();
 
 	config_oa_regs(dev_priv, oa_config->b_counter_regs,
 		       oa_config->b_counter_regs_len);
@@ -2515,6 +2522,9 @@ static int i915_perf_release(struct inode *inode, struct file *file)
 	i915_perf_destroy_locked(stream);
 	mutex_unlock(&dev_priv->perf.lock);
 
+	/* Release the reference the perf stream kept on the driver. */
+	drm_dev_put(&dev_priv->drm);
+
 	return 0;
 }
 
@@ -2650,6 +2660,11 @@ i915_perf_open_ioctl_locked(struct drm_i915_private *dev_priv,
 	if (!(param->flags & I915_PERF_FLAG_DISABLED))
 		i915_perf_enable_locked(stream);
 
+	/* Take a reference on the driver that will be kept with stream_fd
+	 * until its release.
+	 */
+	drm_dev_get(&dev_priv->drm);
+
 	return stream_fd;
 
 err_open:
@@ -3477,9 +3492,13 @@ void i915_perf_init(struct drm_i915_private *dev_priv)
 			dev_priv->perf.oa.ops.enable_metric_set = gen8_enable_metric_set;
 			dev_priv->perf.oa.ops.disable_metric_set = gen10_disable_metric_set;
 
-			dev_priv->perf.oa.ctx_oactxctrl_offset = 0x128;
-			dev_priv->perf.oa.ctx_flexeu0_offset = 0x3de;
-
+			if (IS_GEN(dev_priv, 10)) {
+				dev_priv->perf.oa.ctx_oactxctrl_offset = 0x128;
+				dev_priv->perf.oa.ctx_flexeu0_offset = 0x3de;
+			} else {
+				dev_priv->perf.oa.ctx_oactxctrl_offset = 0x124;
+				dev_priv->perf.oa.ctx_flexeu0_offset = 0x78e;
+			}
 			dev_priv->perf.oa.gen8_valid_ctx_bit = (1<<16);
 		}
 	}
diff --git a/drivers/gpu/drm/i915/i915_trace.h b/drivers/gpu/drm/i915/i915_trace.h
index f4ce643b3bc3..cce426b23a24 100644
--- a/drivers/gpu/drm/i915/i915_trace.h
+++ b/drivers/gpu/drm/i915/i915_trace.h
@@ -21,24 +21,22 @@
 /* watermark/fifo updates */
 
 TRACE_EVENT(intel_pipe_enable,
-	    TP_PROTO(struct drm_i915_private *dev_priv, enum pipe pipe),
-	    TP_ARGS(dev_priv, pipe),
+	    TP_PROTO(struct intel_crtc *crtc),
+	    TP_ARGS(crtc),
 
 	    TP_STRUCT__entry(
 			     __array(u32, frame, 3)
 			     __array(u32, scanline, 3)
 			     __field(enum pipe, pipe)
 			     ),
-
 	    TP_fast_assign(
-			   enum pipe _pipe;
-			   for_each_pipe(dev_priv, _pipe) {
-				   __entry->frame[_pipe] =
-					   dev_priv->drm.driver->get_vblank_counter(&dev_priv->drm, _pipe);
-				   __entry->scanline[_pipe] =
-					   intel_get_crtc_scanline(intel_get_crtc_for_pipe(dev_priv, _pipe));
+			   struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
+			   struct intel_crtc *it__;
+			   for_each_intel_crtc(&dev_priv->drm, it__) {
+				   __entry->frame[it__->pipe] = intel_crtc_get_vblank_counter(it__);
+				   __entry->scanline[it__->pipe] = intel_get_crtc_scanline(it__);
 			   }
-			   __entry->pipe = pipe;
+			   __entry->pipe = crtc->pipe;
 			   ),
 
 	    TP_printk("pipe %c enable, pipe A: frame=%u, scanline=%u, pipe B: frame=%u, scanline=%u, pipe C: frame=%u, scanline=%u",
@@ -49,8 +47,8 @@ TRACE_EVENT(intel_pipe_enable,
 );
 
 TRACE_EVENT(intel_pipe_disable,
-	    TP_PROTO(struct drm_i915_private *dev_priv, enum pipe pipe),
-	    TP_ARGS(dev_priv, pipe),
+	    TP_PROTO(struct intel_crtc *crtc),
+	    TP_ARGS(crtc),
 
 	    TP_STRUCT__entry(
 			     __array(u32, frame, 3)
@@ -59,14 +57,13 @@ TRACE_EVENT(intel_pipe_disable,
 			     ),
 
 	    TP_fast_assign(
-			   enum pipe _pipe;
-			   for_each_pipe(dev_priv, _pipe) {
-				   __entry->frame[_pipe] =
-					   dev_priv->drm.driver->get_vblank_counter(&dev_priv->drm, _pipe);
-				   __entry->scanline[_pipe] =
-					   intel_get_crtc_scanline(intel_get_crtc_for_pipe(dev_priv, _pipe));
+			   struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
+			   struct intel_crtc *it__;
+			   for_each_intel_crtc(&dev_priv->drm, it__) {
+				   __entry->frame[it__->pipe] = intel_crtc_get_vblank_counter(it__);
+				   __entry->scanline[it__->pipe] = intel_get_crtc_scanline(it__);
 			   }
-			   __entry->pipe = pipe;
+			   __entry->pipe = crtc->pipe;
 			   ),
 
 	    TP_printk("pipe %c disable, pipe A: frame=%u, scanline=%u, pipe B: frame=%u, scanline=%u, pipe C: frame=%u, scanline=%u",
@@ -89,8 +86,7 @@ TRACE_EVENT(intel_pipe_crc,
 
 	    TP_fast_assign(
 			   __entry->pipe = crtc->pipe;
-			   __entry->frame = crtc->base.dev->driver->get_vblank_counter(crtc->base.dev,
-										       crtc->pipe);
+			   __entry->frame = intel_crtc_get_vblank_counter(crtc);
 			   __entry->scanline = intel_get_crtc_scanline(crtc);
 			   memcpy(__entry->crcs, crcs, sizeof(__entry->crcs));
 			   ),
@@ -112,9 +108,10 @@ TRACE_EVENT(intel_cpu_fifo_underrun,
 			     ),
 
 	    TP_fast_assign(
+			    struct intel_crtc *crtc = intel_get_crtc_for_pipe(dev_priv, pipe);
 			   __entry->pipe = pipe;
-			   __entry->frame = dev_priv->drm.driver->get_vblank_counter(&dev_priv->drm, pipe);
-			   __entry->scanline = intel_get_crtc_scanline(intel_get_crtc_for_pipe(dev_priv, pipe));
+			   __entry->frame = intel_crtc_get_vblank_counter(crtc);
+			   __entry->scanline = intel_get_crtc_scanline(crtc);
 			   ),
 
 	    TP_printk("pipe %c, frame=%u, scanline=%u",
@@ -134,9 +131,10 @@ TRACE_EVENT(intel_pch_fifo_underrun,
 
 	    TP_fast_assign(
 			   enum pipe pipe = pch_transcoder;
+			   struct intel_crtc *crtc = intel_get_crtc_for_pipe(dev_priv, pipe);
 			   __entry->pipe = pipe;
-			   __entry->frame = dev_priv->drm.driver->get_vblank_counter(&dev_priv->drm, pipe);
-			   __entry->scanline = intel_get_crtc_scanline(intel_get_crtc_for_pipe(dev_priv, pipe));
+			   __entry->frame = intel_crtc_get_vblank_counter(crtc);
+			   __entry->scanline = intel_get_crtc_scanline(crtc);
 			   ),
 
 	    TP_printk("pch transcoder %c, frame=%u, scanline=%u",
@@ -156,12 +154,10 @@ TRACE_EVENT(intel_memory_cxsr,
 			     ),
 
 	    TP_fast_assign(
-			   enum pipe pipe;
-			   for_each_pipe(dev_priv, pipe) {
-				   __entry->frame[pipe] =
-					   dev_priv->drm.driver->get_vblank_counter(&dev_priv->drm, pipe);
-				   __entry->scanline[pipe] =
-					   intel_get_crtc_scanline(intel_get_crtc_for_pipe(dev_priv, pipe));
+			   struct intel_crtc *crtc;
+			   for_each_intel_crtc(&dev_priv->drm, crtc) {
+				   __entry->frame[crtc->pipe] = intel_crtc_get_vblank_counter(crtc);
+				   __entry->scanline[crtc->pipe] = intel_get_crtc_scanline(crtc);
 			   }
 			   __entry->old = old;
 			   __entry->new = new;
@@ -198,8 +194,7 @@ TRACE_EVENT(g4x_wm,
 
 	    TP_fast_assign(
 			   __entry->pipe = crtc->pipe;
-			   __entry->frame = crtc->base.dev->driver->get_vblank_counter(crtc->base.dev,
-										       crtc->pipe);
+			   __entry->frame = intel_crtc_get_vblank_counter(crtc);
 			   __entry->scanline = intel_get_crtc_scanline(crtc);
 			   __entry->primary = wm->pipe[crtc->pipe].plane[PLANE_PRIMARY];
 			   __entry->sprite = wm->pipe[crtc->pipe].plane[PLANE_SPRITE0];
@@ -243,8 +238,7 @@ TRACE_EVENT(vlv_wm,
 
 	    TP_fast_assign(
 			   __entry->pipe = crtc->pipe;
-			   __entry->frame = crtc->base.dev->driver->get_vblank_counter(crtc->base.dev,
-										       crtc->pipe);
+			   __entry->frame = intel_crtc_get_vblank_counter(crtc);
 			   __entry->scanline = intel_get_crtc_scanline(crtc);
 			   __entry->level = wm->level;
 			   __entry->cxsr = wm->cxsr;
@@ -278,8 +272,7 @@ TRACE_EVENT(vlv_fifo_size,
 
 	    TP_fast_assign(
 			   __entry->pipe = crtc->pipe;
-			   __entry->frame = crtc->base.dev->driver->get_vblank_counter(crtc->base.dev,
-										       crtc->pipe);
+			   __entry->frame = intel_crtc_get_vblank_counter(crtc);
 			   __entry->scanline = intel_get_crtc_scanline(crtc);
 			   __entry->sprite0_start = sprite0_start;
 			   __entry->sprite1_start = sprite1_start;
@@ -310,8 +303,7 @@ TRACE_EVENT(intel_update_plane,
 	    TP_fast_assign(
 			   __entry->pipe = crtc->pipe;
 			   __entry->name = plane->name;
-			   __entry->frame = crtc->base.dev->driver->get_vblank_counter(crtc->base.dev,
-										       crtc->pipe);
+			   __entry->frame = intel_crtc_get_vblank_counter(crtc);
 			   __entry->scanline = intel_get_crtc_scanline(crtc);
 			   memcpy(__entry->src, &plane->state->src, sizeof(__entry->src));
 			   memcpy(__entry->dst, &plane->state->dst, sizeof(__entry->dst));
@@ -338,8 +330,7 @@ TRACE_EVENT(intel_disable_plane,
 	    TP_fast_assign(
 			   __entry->pipe = crtc->pipe;
 			   __entry->name = plane->name;
-			   __entry->frame = crtc->base.dev->driver->get_vblank_counter(crtc->base.dev,
-										       crtc->pipe);
+			   __entry->frame = intel_crtc_get_vblank_counter(crtc);
 			   __entry->scanline = intel_get_crtc_scanline(crtc);
 			   ),
 
@@ -364,8 +355,7 @@ TRACE_EVENT(i915_pipe_update_start,
 
 	    TP_fast_assign(
 			   __entry->pipe = crtc->pipe;
-			   __entry->frame = crtc->base.dev->driver->get_vblank_counter(crtc->base.dev,
-										       crtc->pipe);
+			   __entry->frame = intel_crtc_get_vblank_counter(crtc);
 			   __entry->scanline = intel_get_crtc_scanline(crtc);
 			   __entry->min = crtc->debug.min_vbl;
 			   __entry->max = crtc->debug.max_vbl;
diff --git a/drivers/gpu/drm/i915/intel_runtime_pm.c b/drivers/gpu/drm/i915/intel_runtime_pm.c
index 502c54428570..8d1aebc3e857 100644
--- a/drivers/gpu/drm/i915/intel_runtime_pm.c
+++ b/drivers/gpu/drm/i915/intel_runtime_pm.c
@@ -221,13 +221,11 @@ __untrack_all_wakerefs(struct intel_runtime_pm_debug *debug,
 static void
 dump_and_free_wakeref_tracking(struct intel_runtime_pm_debug *debug)
 {
-	struct drm_printer p;
+	if (debug->count) {
+		struct drm_printer p = drm_debug_printer("i915");
 
-	if (!debug->count)
-		return;
-
-	p = drm_debug_printer("i915");
-	__print_intel_runtime_pm_wakeref(&p, debug);
+		__print_intel_runtime_pm_wakeref(&p, debug);
+	}
 
 	kfree(debug->owners);
 }
diff --git a/drivers/gpu/drm/i915/intel_wakeref.h b/drivers/gpu/drm/i915/intel_wakeref.h
index 9cbb2ebf575b..38275310b196 100644
--- a/drivers/gpu/drm/i915/intel_wakeref.h
+++ b/drivers/gpu/drm/i915/intel_wakeref.h
@@ -66,6 +66,21 @@ intel_wakeref_get(struct intel_runtime_pm *rpm,
 }
 
 /**
+ * intel_wakeref_get_if_in_use: Acquire the wakeref
+ * @wf: the wakeref
+ *
+ * Acquire a hold on the wakeref, but only if the wakeref is already
+ * active.
+ *
+ * Returns: true if the wakeref was acquired, false otherwise.
+ */
+static inline bool
+intel_wakeref_get_if_active(struct intel_wakeref *wf)
+{
+	return atomic_inc_not_zero(&wf->count);
+}
+
+/**
  * intel_wakeref_put: Release the wakeref
  * @i915: the drm_i915_private device
  * @wf: the wakeref