Merge tag 'v4.0-rc6' into drm-intel-next

Backmerge Linux 4.0-rc6 because conflicts are (again) getting out of
hand. To make sure we don't lose any bugfixes from the 4.0-rc5-rc6
flurry of patches we've applied them all to -next too.

Conflicts:
	drivers/gpu/drm/i915/intel_display.c

Always take the version from -next, we've already handled all
conflicts with explicit cherrypicking.

Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>

18 files changed, 152 insertions, 385 deletions

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
index 910ff8ab9c9c..d8135adb2238 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
@@ -645,6 +645,7 @@ static int create_sdma_queue_nocpsch(struct device_queue_manager *dqm,
 	pr_debug("     sdma queue id: %d\n", q->properties.sdma_queue_id);
 	pr_debug("     sdma engine id: %d\n", q->properties.sdma_engine_id);
 
+	init_sdma_vm(dqm, q, qpd);
 	retval = mqd->init_mqd(mqd, &q->mqd, &q->mqd_mem_obj,
 				&q->gart_mqd_addr, &q->properties);
 	if (retval != 0) {
@@ -652,7 +653,14 @@ static int create_sdma_queue_nocpsch(struct device_queue_manager *dqm,
 		return retval;
 	}
 
-	init_sdma_vm(dqm, q, qpd);
+	retval = mqd->load_mqd(mqd, q->mqd, 0,
+				0, NULL);
+	if (retval != 0) {
+		deallocate_sdma_queue(dqm, q->sdma_id);
+		mqd->uninit_mqd(mqd, q->mqd, q->mqd_mem_obj);
+		return retval;
+	}
+
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.c b/drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.c
index e415a2a9207e..c7d298e62c96 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.c
@@ -44,7 +44,7 @@ static bool initialize(struct kernel_queue *kq, struct kfd_dev *dev,
 	BUG_ON(!kq || !dev);
 	BUG_ON(type != KFD_QUEUE_TYPE_DIQ && type != KFD_QUEUE_TYPE_HIQ);
 
-	pr_debug("kfd: In func %s initializing queue type %d size %d\n",
+	pr_debug("amdkfd: In func %s initializing queue type %d size %d\n",
 			__func__, KFD_QUEUE_TYPE_HIQ, queue_size);
 
 	nop.opcode = IT_NOP;
@@ -69,12 +69,16 @@ static bool initialize(struct kernel_queue *kq, struct kfd_dev *dev,
 
 	prop.doorbell_ptr = kfd_get_kernel_doorbell(dev, &prop.doorbell_off);
 
-	if (prop.doorbell_ptr == NULL)
+	if (prop.doorbell_ptr == NULL) {
+		pr_err("amdkfd: error init doorbell");
 		goto err_get_kernel_doorbell;
+	}
 
 	retval = kfd_gtt_sa_allocate(dev, queue_size, &kq->pq);
-	if (retval != 0)
+	if (retval != 0) {
+		pr_err("amdkfd: error init pq queues size (%d)\n", queue_size);
 		goto err_pq_allocate_vidmem;
+	}
 
 	kq->pq_kernel_addr = kq->pq->cpu_ptr;
 	kq->pq_gpu_addr = kq->pq->gpu_addr;
@@ -165,10 +169,8 @@ err_rptr_allocate_vidmem:
 err_eop_allocate_vidmem:
 	kfd_gtt_sa_free(dev, kq->pq);
 err_pq_allocate_vidmem:
-	pr_err("kfd: error init pq\n");
 	kfd_release_kernel_doorbell(dev, prop.doorbell_ptr);
 err_get_kernel_doorbell:
-	pr_err("kfd: error init doorbell");
 	return false;
 
 }
@@ -187,6 +189,8 @@ static void uninitialize(struct kernel_queue *kq)
 	else if (kq->queue->properties.type == KFD_QUEUE_TYPE_DIQ)
 		kfd_gtt_sa_free(kq->dev, kq->fence_mem_obj);
 
+	kq->mqd->uninit_mqd(kq->mqd, kq->queue->mqd, kq->queue->mqd_mem_obj);
+
 	kfd_gtt_sa_free(kq->dev, kq->rptr_mem);
 	kfd_gtt_sa_free(kq->dev, kq->wptr_mem);
 	kq->ops_asic_specific.uninitialize(kq);
@@ -211,7 +215,7 @@ static int acquire_packet_buffer(struct kernel_queue *kq,
 	queue_address = (unsigned int *)kq->pq_kernel_addr;
 	queue_size_dwords = kq->queue->properties.queue_size / sizeof(uint32_t);
 
-	pr_debug("kfd: In func %s\nrptr: %d\nwptr: %d\nqueue_address 0x%p\n",
+	pr_debug("amdkfd: In func %s\nrptr: %d\nwptr: %d\nqueue_address 0x%p\n",
 			__func__, rptr, wptr, queue_address);
 
 	available_size = (rptr - 1 - wptr + queue_size_dwords) %
@@ -296,7 +300,7 @@ struct kernel_queue *kernel_queue_init(struct kfd_dev *dev,
 	}
 
 	if (kq->ops.initialize(kq, dev, type, KFD_KERNEL_QUEUE_SIZE) == false) {
-		pr_err("kfd: failed to init kernel queue\n");
+		pr_err("amdkfd: failed to init kernel queue\n");
 		kfree(kq);
 		return NULL;
 	}
@@ -319,7 +323,7 @@ static __attribute__((unused)) void test_kq(struct kfd_dev *dev)
 
 	BUG_ON(!dev);
 
-	pr_err("kfd: starting kernel queue test\n");
+	pr_err("amdkfd: starting kernel queue test\n");
 
 	kq = kernel_queue_init(dev, KFD_QUEUE_TYPE_HIQ);
 	BUG_ON(!kq);
@@ -330,7 +334,7 @@ static __attribute__((unused)) void test_kq(struct kfd_dev *dev)
 		buffer[i] = kq->nop_packet;
 	kq->ops.submit_packet(kq);
 
-	pr_err("kfd: ending kernel queue test\n");
+	pr_err("amdkfd: ending kernel queue test\n");
 }
 
 
diff --git a/drivers/gpu/drm/exynos/Kconfig b/drivers/gpu/drm/exynos/Kconfig
index a5e74612100e..0a6780367d28 100644
--- a/drivers/gpu/drm/exynos/Kconfig
+++ b/drivers/gpu/drm/exynos/Kconfig
@@ -50,7 +50,7 @@ config DRM_EXYNOS_DSI
 
 config DRM_EXYNOS_DP
 	bool "EXYNOS DRM DP driver support"
-	depends on (DRM_EXYNOS_FIMD || DRM_EXYNOS7DECON) && ARCH_EXYNOS && (DRM_PTN3460=n || DRM_PTN3460=y || DRM_PTN3460=DRM_EXYNOS)
+	depends on (DRM_EXYNOS_FIMD || DRM_EXYNOS7_DECON) && ARCH_EXYNOS && (DRM_PTN3460=n || DRM_PTN3460=y || DRM_PTN3460=DRM_EXYNOS)
 	default DRM_EXYNOS
 	select DRM_PANEL
 	help
diff --git a/drivers/gpu/drm/exynos/exynos7_drm_decon.c b/drivers/gpu/drm/exynos/exynos7_drm_decon.c
index 63f02e2380ae..970046199608 100644
--- a/drivers/gpu/drm/exynos/exynos7_drm_decon.c
+++ b/drivers/gpu/drm/exynos/exynos7_drm_decon.c
@@ -888,8 +888,8 @@ static int decon_probe(struct platform_device *pdev)
 	of_node_put(i80_if_timings);
 
 	ctx->regs = of_iomap(dev->of_node, 0);
-	if (IS_ERR(ctx->regs)) {
-		ret = PTR_ERR(ctx->regs);
+	if (!ctx->regs) {
+		ret = -ENOMEM;
 		goto err_del_component;
 	}
 
diff --git a/drivers/gpu/drm/exynos/exynos_drm_connector.c b/drivers/gpu/drm/exynos/exynos_drm_connector.c
deleted file mode 100644
index ba9b3d5ed672..000000000000
--- a/drivers/gpu/drm/exynos/exynos_drm_connector.c
+++ /dev/null
@@ -1,245 +0,0 @@
-/*
- * Copyright (c) 2011 Samsung Electronics Co., Ltd.
- * Authors:
- *	Inki Dae <inki.dae@samsung.com>
- *	Joonyoung Shim <jy0922.shim@samsung.com>
- *	Seung-Woo Kim <sw0312.kim@samsung.com>
- *
- * This program is free software; you can redistribute  it and/or modify it
- * under  the terms of  the GNU General  Public License as published by the
- * Free Software Foundation;  either version 2 of the  License, or (at your
- * option) any later version.
- */
-
-#include <drm/drmP.h>
-#include <drm/drm_crtc_helper.h>
-
-#include <drm/exynos_drm.h>
-#include "exynos_drm_drv.h"
-#include "exynos_drm_encoder.h"
-#include "exynos_drm_connector.h"
-
-#define to_exynos_connector(x)	container_of(x, struct exynos_drm_connector,\
-				drm_connector)
-
-struct exynos_drm_connector {
-	struct drm_connector		drm_connector;
-	uint32_t			encoder_id;
-	struct exynos_drm_display	*display;
-};
-
-static int exynos_drm_connector_get_modes(struct drm_connector *connector)
-{
-	struct exynos_drm_connector *exynos_connector =
-					to_exynos_connector(connector);
-	struct exynos_drm_display *display = exynos_connector->display;
-	struct edid *edid = NULL;
-	unsigned int count = 0;
-	int ret;
-
-	/*
-	 * if get_edid() exists then get_edid() callback of hdmi side
-	 * is called to get edid data through i2c interface else
-	 * get timing from the FIMD driver(display controller).
-	 *
-	 * P.S. in case of lcd panel, count is always 1 if success
-	 * because lcd panel has only one mode.
-	 */
-	if (display->ops->get_edid) {
-		edid = display->ops->get_edid(display, connector);
-		if (IS_ERR_OR_NULL(edid)) {
-			ret = PTR_ERR(edid);
-			edid = NULL;
-			DRM_ERROR("Panel operation get_edid failed %d\n", ret);
-			goto out;
-		}
-
-		count = drm_add_edid_modes(connector, edid);
-		if (!count) {
-			DRM_ERROR("Add edid modes failed %d\n", count);
-			goto out;
-		}
-
-		drm_mode_connector_update_edid_property(connector, edid);
-	} else {
-		struct exynos_drm_panel_info *panel;
-		struct drm_display_mode *mode = drm_mode_create(connector->dev);
-		if (!mode) {
-			DRM_ERROR("failed to create a new display mode.\n");
-			return 0;
-		}
-
-		if (display->ops->get_panel)
-			panel = display->ops->get_panel(display);
-		else {
-			drm_mode_destroy(connector->dev, mode);
-			return 0;
-		}
-
-		drm_display_mode_from_videomode(&panel->vm, mode);
-		mode->width_mm = panel->width_mm;
-		mode->height_mm = panel->height_mm;
-		connector->display_info.width_mm = mode->width_mm;
-		connector->display_info.height_mm = mode->height_mm;
-
-		mode->type = DRM_MODE_TYPE_DRIVER | DRM_MODE_TYPE_PREFERRED;
-		drm_mode_set_name(mode);
-		drm_mode_probed_add(connector, mode);
-
-		count = 1;
-	}
-
-out:
-	kfree(edid);
-	return count;
-}
-
-static int exynos_drm_connector_mode_valid(struct drm_connector *connector,
-					    struct drm_display_mode *mode)
-{
-	struct exynos_drm_connector *exynos_connector =
-					to_exynos_connector(connector);
-	struct exynos_drm_display *display = exynos_connector->display;
-	int ret = MODE_BAD;
-
-	DRM_DEBUG_KMS("%s\n", __FILE__);
-
-	if (display->ops->check_mode)
-		if (!display->ops->check_mode(display, mode))
-			ret = MODE_OK;
-
-	return ret;
-}
-
-static struct drm_encoder *exynos_drm_best_encoder(
-		struct drm_connector *connector)
-{
-	struct drm_device *dev = connector->dev;
-	struct exynos_drm_connector *exynos_connector =
-					to_exynos_connector(connector);
-	return drm_encoder_find(dev, exynos_connector->encoder_id);
-}
-
-static struct drm_connector_helper_funcs exynos_connector_helper_funcs = {
-	.get_modes	= exynos_drm_connector_get_modes,
-	.mode_valid	= exynos_drm_connector_mode_valid,
-	.best_encoder	= exynos_drm_best_encoder,
-};
-
-static int exynos_drm_connector_fill_modes(struct drm_connector *connector,
-				unsigned int max_width, unsigned int max_height)
-{
-	struct exynos_drm_connector *exynos_connector =
-					to_exynos_connector(connector);
-	struct exynos_drm_display *display = exynos_connector->display;
-	unsigned int width, height;
-
-	width = max_width;
-	height = max_height;
-
-	/*
-	 * if specific driver want to find desired_mode using maxmum
-	 * resolution then get max width and height from that driver.
-	 */
-	if (display->ops->get_max_resol)
-		display->ops->get_max_resol(display, &width, &height);
-
-	return drm_helper_probe_single_connector_modes(connector, width,
-							height);
-}
-
-/* get detection status of display device. */
-static enum drm_connector_status
-exynos_drm_connector_detect(struct drm_connector *connector, bool force)
-{
-	struct exynos_drm_connector *exynos_connector =
-					to_exynos_connector(connector);
-	struct exynos_drm_display *display = exynos_connector->display;
-	enum drm_connector_status status = connector_status_disconnected;
-
-	if (display->ops->is_connected) {
-		if (display->ops->is_connected(display))
-			status = connector_status_connected;
-		else
-			status = connector_status_disconnected;
-	}
-
-	return status;
-}
-
-static void exynos_drm_connector_destroy(struct drm_connector *connector)
-{
-	struct exynos_drm_connector *exynos_connector =
-		to_exynos_connector(connector);
-
-	drm_connector_unregister(connector);
-	drm_connector_cleanup(connector);
-	kfree(exynos_connector);
-}
-
-static struct drm_connector_funcs exynos_connector_funcs = {
-	.dpms		= drm_helper_connector_dpms,
-	.fill_modes	= exynos_drm_connector_fill_modes,
-	.detect		= exynos_drm_connector_detect,
-	.destroy	= exynos_drm_connector_destroy,
-};
-
-struct drm_connector *exynos_drm_connector_create(struct drm_device *dev,
-						   struct drm_encoder *encoder)
-{
-	struct exynos_drm_connector *exynos_connector;
-	struct exynos_drm_display *display = exynos_drm_get_display(encoder);
-	struct drm_connector *connector;
-	int type;
-	int err;
-
-	exynos_connector = kzalloc(sizeof(*exynos_connector), GFP_KERNEL);
-	if (!exynos_connector)
-		return NULL;
-
-	connector = &exynos_connector->drm_connector;
-
-	switch (display->type) {
-	case EXYNOS_DISPLAY_TYPE_HDMI:
-		type = DRM_MODE_CONNECTOR_HDMIA;
-		connector->interlace_allowed = true;
-		connector->polled = DRM_CONNECTOR_POLL_HPD;
-		break;
-	case EXYNOS_DISPLAY_TYPE_VIDI:
-		type = DRM_MODE_CONNECTOR_VIRTUAL;
-		connector->polled = DRM_CONNECTOR_POLL_HPD;
-		break;
-	default:
-		type = DRM_MODE_CONNECTOR_Unknown;
-		break;
-	}
-
-	drm_connector_init(dev, connector, &exynos_connector_funcs, type);
-	drm_connector_helper_add(connector, &exynos_connector_helper_funcs);
-
-	err = drm_connector_register(connector);
-	if (err)
-		goto err_connector;
-
-	exynos_connector->encoder_id = encoder->base.id;
-	exynos_connector->display = display;
-	connector->dpms = DRM_MODE_DPMS_OFF;
-	connector->encoder = encoder;
-
-	err = drm_mode_connector_attach_encoder(connector, encoder);
-	if (err) {
-		DRM_ERROR("failed to attach a connector to a encoder\n");
-		goto err_sysfs;
-	}
-
-	DRM_DEBUG_KMS("connector has been created\n");
-
-	return connector;
-
-err_sysfs:
-	drm_connector_unregister(connector);
-err_connector:
-	drm_connector_cleanup(connector);
-	kfree(exynos_connector);
-	return NULL;
-}
diff --git a/drivers/gpu/drm/exynos/exynos_drm_connector.h b/drivers/gpu/drm/exynos/exynos_drm_connector.h
deleted file mode 100644
index 4eb20d78379a..000000000000
--- a/drivers/gpu/drm/exynos/exynos_drm_connector.h
+++ /dev/null
@@ -1,20 +0,0 @@
-/*
- * Copyright (c) 2011 Samsung Electronics Co., Ltd.
- * Authors:
- *	Inki Dae <inki.dae@samsung.com>
- *	Joonyoung Shim <jy0922.shim@samsung.com>
- *	Seung-Woo Kim <sw0312.kim@samsung.com>
- *
- * This program is free software; you can redistribute  it and/or modify it
- * under  the terms of  the GNU General  Public License as published by the
- * Free Software Foundation;  either version 2 of the  License, or (at your
- * option) any later version.
- */
-
-#ifndef _EXYNOS_DRM_CONNECTOR_H_
-#define _EXYNOS_DRM_CONNECTOR_H_
-
-struct drm_connector *exynos_drm_connector_create(struct drm_device *dev,
-						   struct drm_encoder *encoder);
-
-#endif
diff --git a/drivers/gpu/drm/exynos/exynos_drm_fimd.c b/drivers/gpu/drm/exynos/exynos_drm_fimd.c
index 925fc69af1a0..c300e22da8ac 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_fimd.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_fimd.c
@@ -284,14 +284,9 @@ static void fimd_clear_channel(struct fimd_context *ctx)
 	}
 }
 
-static int fimd_ctx_initialize(struct fimd_context *ctx,
+static int fimd_iommu_attach_devices(struct fimd_context *ctx,
 			struct drm_device *drm_dev)
 {
-	struct exynos_drm_private *priv;
-	priv = drm_dev->dev_private;
-
-	ctx->drm_dev = drm_dev;
-	ctx->pipe = priv->pipe++;
 
 	/* attach this sub driver to iommu mapping if supported. */
 	if (is_drm_iommu_supported(ctx->drm_dev)) {
@@ -313,7 +308,7 @@ static int fimd_ctx_initialize(struct fimd_context *ctx,
 	return 0;
 }
 
-static void fimd_ctx_remove(struct fimd_context *ctx)
+static void fimd_iommu_detach_devices(struct fimd_context *ctx)
 {
 	/* detach this sub driver from iommu mapping if supported. */
 	if (is_drm_iommu_supported(ctx->drm_dev))
@@ -1056,25 +1051,23 @@ static int fimd_bind(struct device *dev, struct device *master, void *data)
 {
 	struct fimd_context *ctx = dev_get_drvdata(dev);
 	struct drm_device *drm_dev = data;
+	struct exynos_drm_private *priv = drm_dev->dev_private;
 	int ret;
 
-	ret = fimd_ctx_initialize(ctx, drm_dev);
-	if (ret) {
-		DRM_ERROR("fimd_ctx_initialize failed.\n");
-		return ret;
-	}
+	ctx->drm_dev = drm_dev;
+	ctx->pipe = priv->pipe++;
 
 	ctx->crtc = exynos_drm_crtc_create(drm_dev, ctx->pipe,
 					   EXYNOS_DISPLAY_TYPE_LCD,
 					   &fimd_crtc_ops, ctx);
-	if (IS_ERR(ctx->crtc)) {
-		fimd_ctx_remove(ctx);
-		return PTR_ERR(ctx->crtc);
-	}
 
 	if (ctx->display)
 		exynos_drm_create_enc_conn(drm_dev, ctx->display);
 
+	ret = fimd_iommu_attach_devices(ctx, drm_dev);
+	if (ret)
+		return ret;
+
 	return 0;
 
 }
@@ -1086,10 +1079,10 @@ static void fimd_unbind(struct device *dev, struct device *master,
 
 	fimd_dpms(ctx->crtc, DRM_MODE_DPMS_OFF);
 
+	fimd_iommu_detach_devices(ctx);
+
 	if (ctx->display)
 		exynos_dpi_remove(ctx->display);
-
-	fimd_ctx_remove(ctx);
 }
 
 static const struct component_ops fimd_component_ops = {
diff --git a/drivers/gpu/drm/exynos/exynos_drm_plane.c b/drivers/gpu/drm/exynos/exynos_drm_plane.c
index a5616872eee7..8ad5b7294eb4 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_plane.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_plane.c
@@ -175,7 +175,7 @@ static int exynos_disable_plane(struct drm_plane *plane)
 	struct exynos_drm_plane *exynos_plane = to_exynos_plane(plane);
 	struct exynos_drm_crtc *exynos_crtc = to_exynos_crtc(plane->crtc);
 
-	if (exynos_crtc->ops->win_disable)
+	if (exynos_crtc && exynos_crtc->ops->win_disable)
 		exynos_crtc->ops->win_disable(exynos_crtc,
 					      exynos_plane->zpos);
 
diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index 4ca3dab77fa4..d07c0b1fb498 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -2660,24 +2660,11 @@ i915_gem_retire_requests_ring(struct intel_engine_cs *ring)
 
 	WARN_ON(i915_verify_lists(ring->dev));
 
-	/* Move any buffers on the active list that are no longer referenced
-	 * by the ringbuffer to the flushing/inactive lists as appropriate,
-	 * before we free the context associated with the requests.
+	/* Retire requests first as we use it above for the early return.
+	 * If we retire requests last, we may use a later seqno and so clear
+	 * the requests lists without clearing the active list, leading to
+	 * confusion.
 	 */
-	while (!list_empty(&ring->active_list)) {
-		struct drm_i915_gem_object *obj;
-
-		obj = list_first_entry(&ring->active_list,
-				      struct drm_i915_gem_object,
-				      ring_list);
-
-		if (!i915_gem_request_completed(obj->last_read_req, true))
-			break;
-
-		i915_gem_object_move_to_inactive(obj);
-	}
-
-
 	while (!list_empty(&ring->request_list)) {
 		struct drm_i915_gem_request *request;
 
@@ -2700,6 +2687,23 @@ i915_gem_retire_requests_ring(struct intel_engine_cs *ring)
 		i915_gem_free_request(request);
 	}
 
+	/* Move any buffers on the active list that are no longer referenced
+	 * by the ringbuffer to the flushing/inactive lists as appropriate,
+	 * before we free the context associated with the requests.
+	 */
+	while (!list_empty(&ring->active_list)) {
+		struct drm_i915_gem_object *obj;
+
+		obj = list_first_entry(&ring->active_list,
+				      struct drm_i915_gem_object,
+				      ring_list);
+
+		if (!i915_gem_request_completed(obj->last_read_req, true))
+			break;
+
+		i915_gem_object_move_to_inactive(obj);
+	}
+
 	if (unlikely(ring->trace_irq_req &&
 		     i915_gem_request_completed(ring->trace_irq_req, true))) {
 		ring->irq_put(ring);
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c b/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c
index 29bd539af183..6efa8f38ff54 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c
@@ -340,11 +340,13 @@ nvkm_devobj_ctor(struct nvkm_object *parent, struct nvkm_object *engine,
 
 		/* switch mmio to cpu's native endianness */
 #ifndef __BIG_ENDIAN
-		if (ioread32_native(map + 0x000004) != 0x00000000)
+		if (ioread32_native(map + 0x000004) != 0x00000000) {
 #else
-		if (ioread32_native(map + 0x000004) == 0x00000000)
+		if (ioread32_native(map + 0x000004) == 0x00000000) {
 #endif
 			iowrite32_native(0x01000001, map + 0x000004);
+			ioread32_native(map);
+		}
 
 		/* read boot0 and strapping information */
 		boot0 = ioread32_native(map + 0x000000);
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/device/gm100.c b/drivers/gpu/drm/nouveau/nvkm/engine/device/gm100.c
index 539561ed3281..108d048da764 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/device/gm100.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/device/gm100.c
@@ -142,6 +142,49 @@ gm100_identify(struct nvkm_device *device)
 		device->oclass[NVDEV_ENGINE_MSPPP  ] = &gf100_msppp_oclass;
 #endif
 		break;
+	case 0x126:
+		device->cname = "GM206";
+		device->oclass[NVDEV_SUBDEV_VBIOS  ] = &nvkm_bios_oclass;
+		device->oclass[NVDEV_SUBDEV_GPIO   ] =  gk104_gpio_oclass;
+		device->oclass[NVDEV_SUBDEV_I2C    ] =  gm204_i2c_oclass;
+		device->oclass[NVDEV_SUBDEV_FUSE   ] = &gm107_fuse_oclass;
+#if 0
+		/* looks to be some non-trivial changes */
+		device->oclass[NVDEV_SUBDEV_CLK    ] = &gk104_clk_oclass;
+		/* priv ring says no to 0x10eb14 writes */
+		device->oclass[NVDEV_SUBDEV_THERM  ] = &gm107_therm_oclass;
+#endif
+		device->oclass[NVDEV_SUBDEV_MXM    ] = &nv50_mxm_oclass;
+		device->oclass[NVDEV_SUBDEV_DEVINIT] =  gm204_devinit_oclass;
+		device->oclass[NVDEV_SUBDEV_MC     ] =  gk20a_mc_oclass;
+		device->oclass[NVDEV_SUBDEV_BUS    ] =  gf100_bus_oclass;
+		device->oclass[NVDEV_SUBDEV_TIMER  ] = &gk20a_timer_oclass;
+		device->oclass[NVDEV_SUBDEV_FB     ] =  gm107_fb_oclass;
+		device->oclass[NVDEV_SUBDEV_LTC    ] =  gm107_ltc_oclass;
+		device->oclass[NVDEV_SUBDEV_IBUS   ] = &gk104_ibus_oclass;
+		device->oclass[NVDEV_SUBDEV_INSTMEM] =  nv50_instmem_oclass;
+		device->oclass[NVDEV_SUBDEV_MMU    ] = &gf100_mmu_oclass;
+		device->oclass[NVDEV_SUBDEV_BAR    ] = &gf100_bar_oclass;
+		device->oclass[NVDEV_SUBDEV_PMU    ] =  gk208_pmu_oclass;
+#if 0
+		device->oclass[NVDEV_SUBDEV_VOLT   ] = &nv40_volt_oclass;
+#endif
+		device->oclass[NVDEV_ENGINE_DMAOBJ ] =  gf110_dmaeng_oclass;
+#if 0
+		device->oclass[NVDEV_ENGINE_FIFO   ] =  gk208_fifo_oclass;
+		device->oclass[NVDEV_ENGINE_SW     ] =  gf100_sw_oclass;
+		device->oclass[NVDEV_ENGINE_GR     ] =  gm107_gr_oclass;
+#endif
+		device->oclass[NVDEV_ENGINE_DISP   ] =  gm204_disp_oclass;
+#if 0
+		device->oclass[NVDEV_ENGINE_CE0    ] = &gm204_ce0_oclass;
+		device->oclass[NVDEV_ENGINE_CE1    ] = &gm204_ce1_oclass;
+		device->oclass[NVDEV_ENGINE_CE2    ] = &gm204_ce2_oclass;
+		device->oclass[NVDEV_ENGINE_MSVLD  ] = &gk104_msvld_oclass;
+		device->oclass[NVDEV_ENGINE_MSPDEC ] = &gk104_mspdec_oclass;
+		device->oclass[NVDEV_ENGINE_MSPPP  ] = &gf100_msppp_oclass;
+#endif
+		break;
 	default:
 		nv_fatal(device, "unknown Maxwell chipset\n");
 		return -EINVAL;
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/nv04.c b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/nv04.c
index b038b6eb51db..043e4296084c 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/nv04.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/nv04.c
@@ -502,72 +502,57 @@ nv04_fifo_intr(struct nvkm_subdev *subdev)
 {
 	struct nvkm_device *device = nv_device(subdev);
 	struct nv04_fifo_priv *priv = (void *)subdev;
-	uint32_t status, reassign;
-	int cnt = 0;
+	u32 mask = nv_rd32(priv, NV03_PFIFO_INTR_EN_0);
+	u32 stat = nv_rd32(priv, NV03_PFIFO_INTR_0) & mask;
+	u32 reassign, chid, get, sem;
 
 	reassign = nv_rd32(priv, NV03_PFIFO_CACHES) & 1;
-	while ((status = nv_rd32(priv, NV03_PFIFO_INTR_0)) && (cnt++ < 100)) {
-		uint32_t chid, get;
-
-		nv_wr32(priv, NV03_PFIFO_CACHES, 0);
-
-		chid = nv_rd32(priv, NV03_PFIFO_CACHE1_PUSH1) & priv->base.max;
-		get  = nv_rd32(priv, NV03_PFIFO_CACHE1_GET);
+	nv_wr32(priv, NV03_PFIFO_CACHES, 0);
 
-		if (status & NV_PFIFO_INTR_CACHE_ERROR) {
-			nv04_fifo_cache_error(device, priv, chid, get);
-			status &= ~NV_PFIFO_INTR_CACHE_ERROR;
-		}
+	chid = nv_rd32(priv, NV03_PFIFO_CACHE1_PUSH1) & priv->base.max;
+	get  = nv_rd32(priv, NV03_PFIFO_CACHE1_GET);
 
-		if (status & NV_PFIFO_INTR_DMA_PUSHER) {
-			nv04_fifo_dma_pusher(device, priv, chid);
-			status &= ~NV_PFIFO_INTR_DMA_PUSHER;
-		}
+	if (stat & NV_PFIFO_INTR_CACHE_ERROR) {
+		nv04_fifo_cache_error(device, priv, chid, get);
+		stat &= ~NV_PFIFO_INTR_CACHE_ERROR;
+	}
 
-		if (status & NV_PFIFO_INTR_SEMAPHORE) {
-			uint32_t sem;
+	if (stat & NV_PFIFO_INTR_DMA_PUSHER) {
+		nv04_fifo_dma_pusher(device, priv, chid);
+		stat &= ~NV_PFIFO_INTR_DMA_PUSHER;
+	}
 
-			status &= ~NV_PFIFO_INTR_SEMAPHORE;
-			nv_wr32(priv, NV03_PFIFO_INTR_0,
-				NV_PFIFO_INTR_SEMAPHORE);
+	if (stat & NV_PFIFO_INTR_SEMAPHORE) {
+		stat &= ~NV_PFIFO_INTR_SEMAPHORE;
+		nv_wr32(priv, NV03_PFIFO_INTR_0, NV_PFIFO_INTR_SEMAPHORE);
 
-			sem = nv_rd32(priv, NV10_PFIFO_CACHE1_SEMAPHORE);
-			nv_wr32(priv, NV10_PFIFO_CACHE1_SEMAPHORE, sem | 0x1);
+		sem = nv_rd32(priv, NV10_PFIFO_CACHE1_SEMAPHORE);
+		nv_wr32(priv, NV10_PFIFO_CACHE1_SEMAPHORE, sem | 0x1);
 
-			nv_wr32(priv, NV03_PFIFO_CACHE1_GET, get + 4);
-			nv_wr32(priv, NV04_PFIFO_CACHE1_PULL0, 1);
-		}
+		nv_wr32(priv, NV03_PFIFO_CACHE1_GET, get + 4);
+		nv_wr32(priv, NV04_PFIFO_CACHE1_PULL0, 1);
+	}
 
-		if (device->card_type == NV_50) {
-			if (status & 0x00000010) {
-				status &= ~0x00000010;
-				nv_wr32(priv, 0x002100, 0x00000010);
-			}
-
-			if (status & 0x40000000) {
-				nv_wr32(priv, 0x002100, 0x40000000);
-				nvkm_fifo_uevent(&priv->base);
-				status &= ~0x40000000;
-			}
+	if (device->card_type == NV_50) {
+		if (stat & 0x00000010) {
+			stat &= ~0x00000010;
+			nv_wr32(priv, 0x002100, 0x00000010);
 		}
 
-		if (status) {
-			nv_warn(priv, "unknown intr 0x%08x, ch %d\n",
-				status, chid);
-			nv_wr32(priv, NV03_PFIFO_INTR_0, status);
-			status = 0;
+		if (stat & 0x40000000) {
+			nv_wr32(priv, 0x002100, 0x40000000);
+			nvkm_fifo_uevent(&priv->base);
+			stat &= ~0x40000000;
 		}
-
-		nv_wr32(priv, NV03_PFIFO_CACHES, reassign);
 	}
 
-	if (status) {
-		nv_error(priv, "still angry after %d spins, halt\n", cnt);
-		nv_wr32(priv, 0x002140, 0);
-		nv_wr32(priv, 0x000140, 0);
+	if (stat) {
+		nv_warn(priv, "unknown intr 0x%08x\n", stat);
+		nv_mask(priv, NV03_PFIFO_INTR_EN_0, stat, 0x00000000);
+		nv_wr32(priv, NV03_PFIFO_INTR_0, stat);
 	}
 
-	nv_wr32(priv, 0x000100, 0x00000100);
+	nv_wr32(priv, NV03_PFIFO_CACHES, reassign);
 }
 
 static int
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgf100.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgf100.c
index 2e7ec389eea7..57e2c5b13123 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgf100.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgf100.c
@@ -1032,9 +1032,9 @@ gf100_grctx_generate_bundle(struct gf100_grctx *info)
 	const int s = 8;
 	const int b = mmio_vram(info, impl->bundle_size, (1 << s), access);
 	mmio_refn(info, 0x408004, 0x00000000, s, b);
-	mmio_refn(info, 0x408008, 0x80000000 | (impl->bundle_size >> s), 0, b);
+	mmio_wr32(info, 0x408008, 0x80000000 | (impl->bundle_size >> s));
 	mmio_refn(info, 0x418808, 0x00000000, s, b);
-	mmio_refn(info, 0x41880c, 0x80000000 | (impl->bundle_size >> s), 0, b);
+	mmio_wr32(info, 0x41880c, 0x80000000 | (impl->bundle_size >> s));
 }
 
 void
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgk104.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgk104.c
index b52300d8861a..5e9454ba158f 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgk104.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgk104.c
@@ -851,9 +851,9 @@ gk104_grctx_generate_bundle(struct gf100_grctx *info)
 	const int s = 8;
 	const int b = mmio_vram(info, impl->bundle_size, (1 << s), access);
 	mmio_refn(info, 0x408004, 0x00000000, s, b);
-	mmio_refn(info, 0x408008, 0x80000000 | (impl->bundle_size >> s), 0, b);
+	mmio_wr32(info, 0x408008, 0x80000000 | (impl->bundle_size >> s));
 	mmio_refn(info, 0x418808, 0x00000000, s, b);
-	mmio_refn(info, 0x41880c, 0x80000000 | (impl->bundle_size >> s), 0, b);
+	mmio_wr32(info, 0x41880c, 0x80000000 | (impl->bundle_size >> s));
 	mmio_wr32(info, 0x4064c8, (state_limit << 16) | token_limit);
 }
 
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgm107.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgm107.c
index 956f4dce960c..b2fae6e389e2 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgm107.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgm107.c
@@ -871,9 +871,9 @@ gm107_grctx_generate_bundle(struct gf100_grctx *info)
 	const int s = 8;
 	const int b = mmio_vram(info, impl->bundle_size, (1 << s), access);
 	mmio_refn(info, 0x408004, 0x00000000, s, b);
-	mmio_refn(info, 0x408008, 0x80000000 | (impl->bundle_size >> s), 0, b);
+	mmio_wr32(info, 0x408008, 0x80000000 | (impl->bundle_size >> s));
 	mmio_refn(info, 0x418e24, 0x00000000, s, b);
-	mmio_refn(info, 0x418e28, 0x80000000 | (impl->bundle_size >> s), 0, b);
+	mmio_wr32(info, 0x418e28, 0x80000000 | (impl->bundle_size >> s));
 	mmio_wr32(info, 0x4064c8, (state_limit << 16) | token_limit);
 }
 
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/bios/i2c.c b/drivers/gpu/drm/nouveau/nvkm/subdev/bios/i2c.c
index d1a89b2bd5c1..c4e1f085ee10 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/bios/i2c.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/bios/i2c.c
@@ -74,7 +74,11 @@ dcb_i2c_parse(struct nvkm_bios *bios, u8 idx, struct dcb_i2c_entry *info)
 	u16 ent = dcb_i2c_entry(bios, idx, &ver, &len);
 	if (ent) {
 		if (ver >= 0x41) {
-			if (!(nv_ro32(bios, ent) & 0x80000000))
+			u32 ent_value = nv_ro32(bios, ent);
+			u8 i2c_port = (ent_value >> 27) & 0x1f;
+			u8 dpaux_port = (ent_value >> 22) & 0x1f;
+			/* value 0x1f means unused according to DCB 4.x spec */
+			if (i2c_port == 0x1f && dpaux_port == 0x1f)
 				info->type = DCB_I2C_UNUSED;
 			else
 				info->type = DCB_I2C_PMGR;
diff --git a/drivers/gpu/drm/radeon/radeon_kfd.c b/drivers/gpu/drm/radeon/radeon_kfd.c
index 061eaa9c19c7..122eb5693ba1 100644
--- a/drivers/gpu/drm/radeon/radeon_kfd.c
+++ b/drivers/gpu/drm/radeon/radeon_kfd.c
@@ -153,7 +153,7 @@ void radeon_kfd_device_init(struct radeon_device *rdev)
 			.compute_vmid_bitmap = 0xFF00,
 
 			.first_compute_pipe = 1,
-			.compute_pipe_count = 8 - 1,
+			.compute_pipe_count = 4 - 1,
 		};
 
 		radeon_doorbell_get_kfd_info(rdev,
diff --git a/drivers/gpu/drm/radeon/radeon_object.c b/drivers/gpu/drm/radeon/radeon_object.c
index 43e09942823e..318165d4855c 100644
--- a/drivers/gpu/drm/radeon/radeon_object.c
+++ b/drivers/gpu/drm/radeon/radeon_object.c
@@ -173,17 +173,6 @@ void radeon_ttm_placement_from_domain(struct radeon_bo *rbo, u32 domain)
 		else
 			rbo->placements[i].lpfn = 0;
 	}
-
-	/*
-	 * Use two-ended allocation depending on the buffer size to
-	 * improve fragmentation quality.
-	 * 512kb was measured as the most optimal number.
-	 */
-	if (rbo->tbo.mem.size > 512 * 1024) {
-		for (i = 0; i < c; i++) {
-			rbo->placements[i].flags |= TTM_PL_FLAG_TOPDOWN;
-		}
-	}
 }
 
 int radeon_bo_create(struct radeon_device *rdev,