/*
 * Copyright 2014 Advanced Micro Devices, Inc.
 * All Rights Reserved.
 *
 * Permission is hereby granted, free of charge, to any person obtaining a
 * copy of this software and associated documentation files (the
 * "Software"), to deal in the Software without restriction, including
 * without limitation the rights to use, copy, modify, merge, publish,
 * distribute, sub license, and/or sell copies of the Software, and to
 * permit persons to whom the Software is furnished to do so, subject to
 * the following conditions:
 *
 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
 * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
 * FITNESS FOR A PARTICULAR PURPOSE AND NON-INFRINGEMENT. IN NO EVENT SHALL
 * THE COPYRIGHT HOLDERS, AUTHORS AND/OR ITS SUPPLIERS BE LIABLE FOR ANY CLAIM,
 * DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR
 * OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE
 * USE OR OTHER DEALINGS IN THE SOFTWARE.
 *
 * The above copyright notice and this permission notice (including the
 * next paragraph) shall be included in all copies or substantial portions
 * of the Software.
 *
 */
/*
 * Authors:
 *    Christian König <christian.koenig@amd.com>
 */

/**
 * DOC: MMU Notifier
 *
 * For coherent userptr handling registers an MMU notifier to inform the driver
 * about updates on the page tables of a process.
 *
 * When somebody tries to invalidate the page tables we block the update until
 * all operations on the pages in question are completed, then those pages are
 * marked as accessed and also dirty if it wasn't a read only access.
 *
 * New command submissions using the userptrs in question are delayed until all
 * page table invalidation are completed and we once more see a coherent process
 * address space.
 */

#include <linux/firmware.h>
#include <linux/module.h>
#include <drm/drm.h>

#include "amdgpu.h"
#include "amdgpu_amdkfd.h"
#include "amdgpu_hmm.h"

#define MAX_WALK_BYTE	(2UL << 30)

/**
 * amdgpu_hmm_invalidate_gfx - callback to notify about mm change
 *
 * @mni: the range (mm) is about to update
 * @range: details on the invalidation
 * @cur_seq: Value to pass to mmu_interval_set_seq()
 *
 * Block for operations on BOs to finish and mark pages as accessed and
 * potentially dirty.
 */
static bool amdgpu_hmm_invalidate_gfx(struct mmu_interval_notifier *mni,
				      const struct mmu_notifier_range *range,
				      unsigned long cur_seq)
{
	struct amdgpu_bo *bo = container_of(mni, struct amdgpu_bo, notifier);
	struct amdgpu_device *adev = amdgpu_ttm_adev(bo->tbo.bdev);
	long r;

	if (!mmu_notifier_range_blockable(range))
		return false;

	mutex_lock(&adev->notifier_lock);

	mmu_interval_set_seq(mni, cur_seq);

	r = dma_resv_wait_timeout(bo->tbo.base.resv, DMA_RESV_USAGE_BOOKKEEP,
				  false, MAX_SCHEDULE_TIMEOUT);
	mutex_unlock(&adev->notifier_lock);
	if (r <= 0)
		DRM_ERROR("(%ld) failed to wait for user bo\n", r);
	return true;
}

static const struct mmu_interval_notifier_ops amdgpu_hmm_gfx_ops = {
	.invalidate = amdgpu_hmm_invalidate_gfx,
};

/**
 * amdgpu_hmm_invalidate_hsa - callback to notify about mm change
 *
 * @mni: the range (mm) is about to update
 * @range: details on the invalidation
 * @cur_seq: Value to pass to mmu_interval_set_seq()
 *
 * We temporarily evict the BO attached to this range. This necessitates
 * evicting all user-mode queues of the process.
 */
static bool amdgpu_hmm_invalidate_hsa(struct mmu_interval_notifier *mni,
				      const struct mmu_notifier_range *range,
				      unsigned long cur_seq)
{
	struct amdgpu_bo *bo = container_of(mni, struct amdgpu_bo, notifier);

	if (!mmu_notifier_range_blockable(range))
		return false;

	amdgpu_amdkfd_evict_userptr(mni, cur_seq, bo->kfd_bo);

	return true;
}

static const struct mmu_interval_notifier_ops amdgpu_hmm_hsa_ops = {
	.invalidate = amdgpu_hmm_invalidate_hsa,
};

/**
 * amdgpu_hmm_register - register a BO for notifier updates
 *
 * @bo: amdgpu buffer object
 * @addr: userptr addr we should monitor
 *
 * Registers a mmu_notifier for the given BO at the specified address.
 * Returns 0 on success, -ERRNO if anything goes wrong.
 */
int amdgpu_hmm_register(struct amdgpu_bo *bo, unsigned long addr)
{
	int r;

	if (bo->kfd_bo)
		r = mmu_interval_notifier_insert(&bo->notifier, current->mm,
						    addr, amdgpu_bo_size(bo),
						    &amdgpu_hmm_hsa_ops);
	else
		r = mmu_interval_notifier_insert(&bo->notifier, current->mm, addr,
							amdgpu_bo_size(bo),
							&amdgpu_hmm_gfx_ops);
	if (r)
		/*
		 * Make sure amdgpu_hmm_unregister() doesn't call
		 * mmu_interval_notifier_remove() when the notifier isn't properly
		 * initialized.
		 */
		bo->notifier.mm = NULL;

	return r;
}

/**
 * amdgpu_hmm_unregister - unregister a BO for notifier updates
 *
 * @bo: amdgpu buffer object
 *
 * Remove any registration of mmu notifier updates from the buffer object.
 */
void amdgpu_hmm_unregister(struct amdgpu_bo *bo)
{
	if (!bo->notifier.mm)
		return;
	mmu_interval_notifier_remove(&bo->notifier);
	bo->notifier.mm = NULL;
}

int amdgpu_hmm_range_get_pages(struct mmu_interval_notifier *notifier,
			       uint64_t start, uint64_t npages, bool readonly,
			       void *owner,
			       struct amdgpu_hmm_range *range)
{
	unsigned long end;
	unsigned long timeout;
	unsigned long *pfns;
	int r = 0;
	struct hmm_range *hmm_range = &range->hmm_range;

	pfns = kvmalloc_array(npages, sizeof(*pfns), GFP_KERNEL);
	if (unlikely(!pfns)) {
		r = -ENOMEM;
		goto out_free_range;
	}

	hmm_range->notifier = notifier;
	hmm_range->default_flags = HMM_PFN_REQ_FAULT;
	if (!readonly)
		hmm_range->default_flags |= HMM_PFN_REQ_WRITE;
	hmm_range->hmm_pfns = pfns;
	hmm_range->start = start;
	end = start + npages * PAGE_SIZE;
	hmm_range->dev_private_owner = owner;

	do {
		hmm_range->end = min(hmm_range->start + MAX_WALK_BYTE, end);

		pr_debug("hmm range: start = 0x%lx, end = 0x%lx",
			hmm_range->start, hmm_range->end);

		timeout = jiffies + msecs_to_jiffies(HMM_RANGE_DEFAULT_TIMEOUT);

retry:
		hmm_range->notifier_seq = mmu_interval_read_begin(notifier);
		r = hmm_range_fault(hmm_range);
		if (unlikely(r)) {
			if (r == -EBUSY && !time_after(jiffies, timeout))
				goto retry;
			goto out_free_pfns;
		}

		if (hmm_range->end == end)
			break;
		hmm_range->hmm_pfns += MAX_WALK_BYTE >> PAGE_SHIFT;
		hmm_range->start = hmm_range->end;
	} while (hmm_range->end < end);

	hmm_range->start = start;
	hmm_range->hmm_pfns = pfns;

	return 0;

out_free_pfns:
	kvfree(pfns);
	hmm_range->hmm_pfns = NULL;
out_free_range:
	if (r == -EBUSY)
		r = -EAGAIN;
	return r;
}

/**
 * amdgpu_hmm_range_valid - check if an HMM range is still valid
 * @range: pointer to the &struct amdgpu_hmm_range to validate
 *
 * Determines whether the given HMM range @range is still valid by
 * checking for invalidations via the MMU notifier sequence. This is
 * typically used to verify that the range has not been invalidated
 * by concurrent address space updates before it is accessed.
 *
 * Return:
 * * true if @range is valid and can be used safely
 * * false if @range is NULL or has been invalidated
 */
bool amdgpu_hmm_range_valid(struct amdgpu_hmm_range *range)
{
	if (!range)
		return false;

	return !mmu_interval_read_retry(range->hmm_range.notifier,
					range->hmm_range.notifier_seq);
}

/**
 * amdgpu_hmm_range_alloc - allocate and initialize an AMDGPU HMM range
 * @bo: optional buffer object to associate with this HMM range
 *
 * Allocates memory for amdgpu_hmm_range and associates it with the @bo passed.
 * The reference count of the @bo is incremented.
 *
 * Return:
 * Pointer to a newly allocated struct amdgpu_hmm_range on success,
 * or NULL if memory allocation fails.
 */
struct amdgpu_hmm_range *amdgpu_hmm_range_alloc(struct amdgpu_bo *bo)
{
	struct amdgpu_hmm_range *range;

	range = kzalloc(sizeof(*range), GFP_KERNEL);
	if (!range)
		return NULL;

	range->bo = amdgpu_bo_ref(bo);
	return range;
}

/**
 * amdgpu_hmm_range_free - release an AMDGPU HMM range
 * @range: pointer to the range object to free
 *
 * Releases all resources held by @range, including the associated
 * hmm_pfns and the dropping reference of associated bo if any.
 *
 * Return: void
 */
void amdgpu_hmm_range_free(struct amdgpu_hmm_range *range)
{
	if (!range)
		return;

	kvfree(range->hmm_range.hmm_pfns);
	amdgpu_bo_unref(&range->bo);
	kfree(range);
}