Merge tag 'drm-next-2019-07-16' of git://anongit.freedesktop.org/drm/drm

Pull drm updates from Dave Airlie:
 "The biggest thing in this is the AMD Navi GPU support, this again
  contains a bunch of header files that are large. These are the new AMD
  RX5700 GPUs that just recently became available.

  New drivers:
   - ST-Ericsson MCDE driver
   - Ingenic JZ47xx SoC

  UAPI change:
   - HDR source metadata property

  Core:
   - HDR inforframes and EDID parsing
   - drm hdmi infoframe unpacking
   - remove prime sg_table caching into dma-buf
   - New gem vram helpers to reduce driver code
   - Lots of drmP.h removal
   - reservation fencing fix
   - documentation updates
   - drm_fb_helper_connector removed
   - mode name command handler rewrite

  fbcon:
   - Remove the fbcon notifiers

  ttm:
   - forward progress fixes

  dma-buf:
   - make mmap call optional
   - debugfs refcount fixes
   - dma-fence free with pending signals fix
   - each dma-buf gets an inode

  Panels:
   - Lots of additional panel bindings

  amdgpu:
   - initial navi10 support
   - avoid hw reset
   - HDR metadata support
   - new thermal sensors for vega asics
   - RAS fixes
   - use HMM rather than MMU notifier
   - xgmi topology via kfd
   - SR-IOV fixes
   - driver reload fixes
   - DC use a core bpc attribute
   - Aux fixes for DC
   - Bandwidth calc updates for DC
   - Clock handling refactor
   - kfd VEGAM support

  vmwgfx:
   - Coherent memory support changes

  i915:
   - HDR Support
   - HDMI i2c link
   - Icelake multi-segmented gamma support
   - GuC firmware update
   - Mule Creek Canyon PCH support for EHL
   - EHL platform updtes
   - move i915.alpha_support to i915.force_probe
   - runtime PM refactoring
   - VBT parsing refactoring
   - DSI fixes
   - struct mutex dependency reduction
   - GEM code reorg

  mali-dp:
   - Komeda driver features

  msm:
   - dsi vs EPROBE_DEFER fixes
   - msm8998 snapdragon 835 support
   - a540 gpu support
   - mdp5 and dpu interconnect support

  exynos:
   - drmP.h removal

  tegra:
   - misc fixes

  tda998x:
   - audio support improvements
   - pixel repeated mode support
   - quantisation range handling corrections
   - HDMI vendor info fix

  armada:
   - interlace support fix
   - overlay/video plane register handling refactor
   - add gamma support

  rockchip:
   - RX3328 support

  panfrost:
   - expose perf counters via hidden ioctls

  vkms:
   - enumerate CRC sources list

  ast:
   - rework BO handling

  mgag200:
   - rework BO handling

  dw-hdmi:
   - suspend/resume support

  rcar-du:
   - R8A774A1 Soc Support
   - LVDS dual-link mode support
   - Additional formats
   - Misc fixes

  omapdrm:
   - DSI command mode display support

  stm
   - fb modifier support
   - runtime PM support

  sun4i:
   - use vmap ops

  vc4:
   - binner bo binding rework

  v3d:
   - compute shader support
   - resync/sync fixes
   - job management refactoring

  lima:
   - NULL pointer in irq handler fix
   - scheduler default timeout

  virtio:
   - fence seqno support
   - trace events

  bochs:
   - misc fixes

  tc458767:
   - IRQ/HDP handling

  sii902x:
   - HDMI audio support

  atmel-hlcdc:
   - misc fixes

  meson:
   - zpos support"

* tag 'drm-next-2019-07-16' of git://anongit.freedesktop.org/drm/drm: (1815 commits)
  Revert "Merge branch 'vmwgfx-next' of git://people.freedesktop.org/~thomash/linux into drm-next"
  Revert "mm: adjust apply_to_pfn_range interface for dropped token."
  mm: adjust apply_to_pfn_range interface for dropped token.
  drm/amdgpu/navi10: add uclk activity sensor
  drm/amdgpu: properly guard the generic discovery code
  drm/amdgpu: add missing documentation on new module parameters
  drm/amdgpu: don't invalidate caches in RELEASE_MEM, only do the writeback
  drm/amd/display: avoid 64-bit division
  drm/amdgpu/psp11: simplify the ucode register logic
  drm/amdgpu: properly guard DC support in navi code
  drm/amd/powerplay: vega20: fix uninitialized variable use
  drm/amd/display: dcn20: include linux/delay.h
  amdgpu: make pmu support optional
  drm/amd/powerplay: Zero initialize current_rpm in vega20_get_fan_speed_percent
  drm/amd/powerplay: Zero initialize freq in smu_v11_0_get_current_clk_freq
  drm/amd/powerplay: Use memset to initialize metrics structs
  drm/amdgpu/mes10.1: Fix header guard
  drm/amd/powerplay: add temperature sensor support for navi10
  drm/amdgpu: fix scheduler timeout calc
  drm/amdgpu: Prepare for hmm_range_register API change (v2)
  ...

1 files changed, 84 insertions, 127 deletions

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mn.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_mn.c
index 58ed401c5996..3971c201f320 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mn.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mn.c
@@ -45,51 +45,12 @@
 
 #include <linux/firmware.h>
 #include <linux/module.h>
-#include <linux/mmu_notifier.h>
-#include <linux/interval_tree.h>
-#include <drm/drmP.h>
 #include <drm/drm.h>
 
 #include "amdgpu.h"
 #include "amdgpu_amdkfd.h"
 
 /**
- * struct amdgpu_mn
- *
- * @adev: amdgpu device pointer
- * @mm: process address space
- * @mn: MMU notifier structure
- * @type: type of MMU notifier
- * @work: destruction work item
- * @node: hash table node to find structure by adev and mn
- * @lock: rw semaphore protecting the notifier nodes
- * @objects: interval tree containing amdgpu_mn_nodes
- * @read_lock: mutex for recursive locking of @lock
- * @recursion: depth of recursion
- *
- * Data for each amdgpu device and process address space.
- */
-struct amdgpu_mn {
-	/* constant after initialisation */
-	struct amdgpu_device	*adev;
-	struct mm_struct	*mm;
-	struct mmu_notifier	mn;
-	enum amdgpu_mn_type	type;
-
-	/* only used on destruction */
-	struct work_struct	work;
-
-	/* protected by adev->mn_lock */
-	struct hlist_node	node;
-
-	/* objects protected by lock */
-	struct rw_semaphore	lock;
-	struct rb_root_cached	objects;
-	struct mutex		read_lock;
-	atomic_t		recursion;
-};
-
-/**
  * struct amdgpu_mn_node
  *
  * @it: interval node defining start-last of the affected address range
@@ -103,7 +64,7 @@ struct amdgpu_mn_node {
 };
 
 /**
- * amdgpu_mn_destroy - destroy the MMU notifier
+ * amdgpu_mn_destroy - destroy the HMM mirror
  *
  * @work: previously sheduled work item
  *
@@ -129,28 +90,26 @@ static void amdgpu_mn_destroy(struct work_struct *work)
 	}
 	up_write(&amn->lock);
 	mutex_unlock(&adev->mn_lock);
-	mmu_notifier_unregister_no_release(&amn->mn, amn->mm);
+
+	hmm_mirror_unregister(&amn->mirror);
 	kfree(amn);
 }
 
 /**
- * amdgpu_mn_release - callback to notify about mm destruction
+ * amdgpu_hmm_mirror_release - callback to notify about mm destruction
  *
- * @mn: our notifier
- * @mm: the mm this callback is about
+ * @mirror: the HMM mirror (mm) this callback is about
  *
- * Shedule a work item to lazy destroy our notifier.
+ * Shedule a work item to lazy destroy HMM mirror.
  */
-static void amdgpu_mn_release(struct mmu_notifier *mn,
-			      struct mm_struct *mm)
+static void amdgpu_hmm_mirror_release(struct hmm_mirror *mirror)
 {
-	struct amdgpu_mn *amn = container_of(mn, struct amdgpu_mn, mn);
+	struct amdgpu_mn *amn = container_of(mirror, struct amdgpu_mn, mirror);
 
 	INIT_WORK(&amn->work, amdgpu_mn_destroy);
 	schedule_work(&amn->work);
 }
 
-
 /**
  * amdgpu_mn_lock - take the write side lock for this notifier
  *
@@ -181,14 +140,10 @@ void amdgpu_mn_unlock(struct amdgpu_mn *mn)
 static int amdgpu_mn_read_lock(struct amdgpu_mn *amn, bool blockable)
 {
 	if (blockable)
-		mutex_lock(&amn->read_lock);
-	else if (!mutex_trylock(&amn->read_lock))
+		down_read(&amn->lock);
+	else if (!down_read_trylock(&amn->lock))
 		return -EAGAIN;
 
-	if (atomic_inc_return(&amn->recursion) == 1)
-		down_read_non_owner(&amn->lock);
-	mutex_unlock(&amn->read_lock);
-
 	return 0;
 }
 
@@ -199,8 +154,7 @@ static int amdgpu_mn_read_lock(struct amdgpu_mn *amn, bool blockable)
  */
 static void amdgpu_mn_read_unlock(struct amdgpu_mn *amn)
 {
-	if (atomic_dec_return(&amn->recursion) == 0)
-		up_read_non_owner(&amn->lock);
+	up_read(&amn->lock);
 }
 
 /**
@@ -229,149 +183,132 @@ static void amdgpu_mn_invalidate_node(struct amdgpu_mn_node *node,
 			true, false, MAX_SCHEDULE_TIMEOUT);
 		if (r <= 0)
 			DRM_ERROR("(%ld) failed to wait for user bo\n", r);
-
-		amdgpu_ttm_tt_mark_user_pages(bo->tbo.ttm);
 	}
 }
 
 /**
- * amdgpu_mn_invalidate_range_start_gfx - callback to notify about mm change
+ * amdgpu_mn_sync_pagetables_gfx - callback to notify about mm change
  *
- * @mn: our notifier
- * @range: mmu notifier context
+ * @mirror: the hmm_mirror (mm) is about to update
+ * @update: the update start, end address
  *
  * Block for operations on BOs to finish and mark pages as accessed and
  * potentially dirty.
  */
-static int amdgpu_mn_invalidate_range_start_gfx(struct mmu_notifier *mn,
-			const struct mmu_notifier_range *range)
+static int amdgpu_mn_sync_pagetables_gfx(struct hmm_mirror *mirror,
+			const struct hmm_update *update)
 {
-	struct amdgpu_mn *amn = container_of(mn, struct amdgpu_mn, mn);
+	struct amdgpu_mn *amn = container_of(mirror, struct amdgpu_mn, mirror);
+	unsigned long start = update->start;
+	unsigned long end = update->end;
+	bool blockable = update->blockable;
 	struct interval_tree_node *it;
-	unsigned long end;
 
 	/* notification is exclusive, but interval is inclusive */
-	end = range->end - 1;
+	end -= 1;
 
 	/* TODO we should be able to split locking for interval tree and
 	 * amdgpu_mn_invalidate_node
 	 */
-	if (amdgpu_mn_read_lock(amn, mmu_notifier_range_blockable(range)))
+	if (amdgpu_mn_read_lock(amn, blockable))
 		return -EAGAIN;
 
-	it = interval_tree_iter_first(&amn->objects, range->start, end);
+	it = interval_tree_iter_first(&amn->objects, start, end);
 	while (it) {
 		struct amdgpu_mn_node *node;
 
-		if (!mmu_notifier_range_blockable(range)) {
+		if (!blockable) {
 			amdgpu_mn_read_unlock(amn);
 			return -EAGAIN;
 		}
 
 		node = container_of(it, struct amdgpu_mn_node, it);
-		it = interval_tree_iter_next(it, range->start, end);
+		it = interval_tree_iter_next(it, start, end);
 
-		amdgpu_mn_invalidate_node(node, range->start, end);
+		amdgpu_mn_invalidate_node(node, start, end);
 	}
 
+	amdgpu_mn_read_unlock(amn);
+
 	return 0;
 }
 
 /**
- * amdgpu_mn_invalidate_range_start_hsa - callback to notify about mm change
+ * amdgpu_mn_sync_pagetables_hsa - callback to notify about mm change
  *
- * @mn: our notifier
- * @mm: the mm this callback is about
- * @start: start of updated range
- * @end: end of updated range
+ * @mirror: the hmm_mirror (mm) is about to update
+ * @update: the update start, end address
  *
  * We temporarily evict all BOs between start and end. This
  * necessitates evicting all user-mode queues of the process. The BOs
  * are restorted in amdgpu_mn_invalidate_range_end_hsa.
  */
-static int amdgpu_mn_invalidate_range_start_hsa(struct mmu_notifier *mn,
-			const struct mmu_notifier_range *range)
+static int amdgpu_mn_sync_pagetables_hsa(struct hmm_mirror *mirror,
+			const struct hmm_update *update)
 {
-	struct amdgpu_mn *amn = container_of(mn, struct amdgpu_mn, mn);
+	struct amdgpu_mn *amn = container_of(mirror, struct amdgpu_mn, mirror);
+	unsigned long start = update->start;
+	unsigned long end = update->end;
+	bool blockable = update->blockable;
 	struct interval_tree_node *it;
-	unsigned long end;
 
 	/* notification is exclusive, but interval is inclusive */
-	end = range->end - 1;
+	end -= 1;
 
-	if (amdgpu_mn_read_lock(amn, mmu_notifier_range_blockable(range)))
+	if (amdgpu_mn_read_lock(amn, blockable))
 		return -EAGAIN;
 
-	it = interval_tree_iter_first(&amn->objects, range->start, end);
+	it = interval_tree_iter_first(&amn->objects, start, end);
 	while (it) {
 		struct amdgpu_mn_node *node;
 		struct amdgpu_bo *bo;
 
-		if (!mmu_notifier_range_blockable(range)) {
+		if (!blockable) {
 			amdgpu_mn_read_unlock(amn);
 			return -EAGAIN;
 		}
 
 		node = container_of(it, struct amdgpu_mn_node, it);
-		it = interval_tree_iter_next(it, range->start, end);
+		it = interval_tree_iter_next(it, start, end);
 
 		list_for_each_entry(bo, &node->bos, mn_list) {
 			struct kgd_mem *mem = bo->kfd_bo;
 
 			if (amdgpu_ttm_tt_affect_userptr(bo->tbo.ttm,
-							 range->start,
-							 end))
-				amdgpu_amdkfd_evict_userptr(mem, range->mm);
+							 start, end))
+				amdgpu_amdkfd_evict_userptr(mem, amn->mm);
 		}
 	}
 
+	amdgpu_mn_read_unlock(amn);
+
 	return 0;
 }
 
-/**
- * amdgpu_mn_invalidate_range_end - callback to notify about mm change
- *
- * @mn: our notifier
- * @mm: the mm this callback is about
- * @start: start of updated range
- * @end: end of updated range
- *
- * Release the lock again to allow new command submissions.
+/* Low bits of any reasonable mm pointer will be unused due to struct
+ * alignment. Use these bits to make a unique key from the mm pointer
+ * and notifier type.
  */
-static void amdgpu_mn_invalidate_range_end(struct mmu_notifier *mn,
-			const struct mmu_notifier_range *range)
-{
-	struct amdgpu_mn *amn = container_of(mn, struct amdgpu_mn, mn);
-
-	amdgpu_mn_read_unlock(amn);
-}
+#define AMDGPU_MN_KEY(mm, type) ((unsigned long)(mm) + (type))
 
-static const struct mmu_notifier_ops amdgpu_mn_ops[] = {
+static struct hmm_mirror_ops amdgpu_hmm_mirror_ops[] = {
 	[AMDGPU_MN_TYPE_GFX] = {
-		.release = amdgpu_mn_release,
-		.invalidate_range_start = amdgpu_mn_invalidate_range_start_gfx,
-		.invalidate_range_end = amdgpu_mn_invalidate_range_end,
+		.sync_cpu_device_pagetables = amdgpu_mn_sync_pagetables_gfx,
+		.release = amdgpu_hmm_mirror_release
 	},
 	[AMDGPU_MN_TYPE_HSA] = {
-		.release = amdgpu_mn_release,
-		.invalidate_range_start = amdgpu_mn_invalidate_range_start_hsa,
-		.invalidate_range_end = amdgpu_mn_invalidate_range_end,
+		.sync_cpu_device_pagetables = amdgpu_mn_sync_pagetables_hsa,
+		.release = amdgpu_hmm_mirror_release
 	},
 };
 
-/* Low bits of any reasonable mm pointer will be unused due to struct
- * alignment. Use these bits to make a unique key from the mm pointer
- * and notifier type.
- */
-#define AMDGPU_MN_KEY(mm, type) ((unsigned long)(mm) + (type))
-
 /**
- * amdgpu_mn_get - create notifier context
+ * amdgpu_mn_get - create HMM mirror context
  *
  * @adev: amdgpu device pointer
  * @type: type of MMU notifier context
  *
- * Creates a notifier context for current->mm.
+ * Creates a HMM mirror context for current->mm.
  */
 struct amdgpu_mn *amdgpu_mn_get(struct amdgpu_device *adev,
 				enum amdgpu_mn_type type)
@@ -401,12 +338,10 @@ struct amdgpu_mn *amdgpu_mn_get(struct amdgpu_device *adev,
 	amn->mm = mm;
 	init_rwsem(&amn->lock);
 	amn->type = type;
-	amn->mn.ops = &amdgpu_mn_ops[type];
 	amn->objects = RB_ROOT_CACHED;
-	mutex_init(&amn->read_lock);
-	atomic_set(&amn->recursion, 0);
 
-	r = __mmu_notifier_register(&amn->mn, mm);
+	amn->mirror.ops = &amdgpu_hmm_mirror_ops[type];
+	r = hmm_mirror_register(&amn->mirror, mm);
 	if (r)
 		goto free_amn;
 
@@ -432,7 +367,7 @@ free_amn:
  * @bo: amdgpu buffer object
  * @addr: userptr addr we should monitor
  *
- * Registers an MMU notifier for the given BO at the specified address.
+ * Registers an HMM mirror for the given BO at the specified address.
  * Returns 0 on success, -ERRNO if anything goes wrong.
  */
 int amdgpu_mn_register(struct amdgpu_bo *bo, unsigned long addr)
@@ -488,11 +423,11 @@ int amdgpu_mn_register(struct amdgpu_bo *bo, unsigned long addr)
 }
 
 /**
- * amdgpu_mn_unregister - unregister a BO for notifier updates
+ * amdgpu_mn_unregister - unregister a BO for HMM mirror updates
  *
  * @bo: amdgpu buffer object
  *
- * Remove any registration of MMU notifier updates from the buffer object.
+ * Remove any registration of HMM mirror updates from the buffer object.
  */
 void amdgpu_mn_unregister(struct amdgpu_bo *bo)
 {
@@ -528,3 +463,25 @@ void amdgpu_mn_unregister(struct amdgpu_bo *bo)
 	mutex_unlock(&adev->mn_lock);
 }
 
+/* flags used by HMM internal, not related to CPU/GPU PTE flags */
+static const uint64_t hmm_range_flags[HMM_PFN_FLAG_MAX] = {
+		(1 << 0), /* HMM_PFN_VALID */
+		(1 << 1), /* HMM_PFN_WRITE */
+		0 /* HMM_PFN_DEVICE_PRIVATE */
+};
+
+static const uint64_t hmm_range_values[HMM_PFN_VALUE_MAX] = {
+		0xfffffffffffffffeUL, /* HMM_PFN_ERROR */
+		0, /* HMM_PFN_NONE */
+		0xfffffffffffffffcUL /* HMM_PFN_SPECIAL */
+};
+
+void amdgpu_hmm_init_range(struct hmm_range *range)
+{
+	if (range) {
+		range->flags = hmm_range_flags;
+		range->values = hmm_range_values;
+		range->pfn_shift = PAGE_SHIFT;
+		INIT_LIST_HEAD(&range->list);
+	}
+}