linux-nvgpu/drivers/gpu/nvgpu/gk20a/mm_gk20a.c

/*
 * GK20A memory management
 *
 * Copyright (c) 2011-2017, NVIDIA CORPORATION.  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, sublicense,
 * 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 above copyright notice and this permission notice shall be included in
 * all copies or substantial portions of the Software.
 *
 * 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 NONINFRINGEMENT.  IN NO EVENT SHALL
 * THE AUTHORS OR COPYRIGHT HOLDERS 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.
 */

#include <trace/events/gk20a.h>

#include <nvgpu/vm.h>
#include <nvgpu/vm_area.h>
#include <nvgpu/dma.h>
#include <nvgpu/kmem.h>
#include <nvgpu/timers.h>
#include <nvgpu/pramin.h>
#include <nvgpu/list.h>
#include <nvgpu/nvgpu_mem.h>
#include <nvgpu/allocator.h>
#include <nvgpu/semaphore.h>
#include <nvgpu/page_allocator.h>
#include <nvgpu/log.h>
#include <nvgpu/bug.h>
#include <nvgpu/log2.h>
#include <nvgpu/enabled.h>
#include <nvgpu/vidmem.h>

#include "gk20a.h"
#include "platform_gk20a.h"
#include "mm_gk20a.h"
#include "fence_gk20a.h"
#include "kind_gk20a.h"
#include "bus_gk20a.h"
#include "common/linux/os_linux.h"

#include <nvgpu/hw/gk20a/hw_gmmu_gk20a.h>
#include <nvgpu/hw/gk20a/hw_ram_gk20a.h>
#include <nvgpu/hw/gk20a/hw_pram_gk20a.h>
#include <nvgpu/hw/gk20a/hw_mc_gk20a.h>
#include <nvgpu/hw/gk20a/hw_bus_gk20a.h>
#include <nvgpu/hw/gk20a/hw_flush_gk20a.h>
#include <nvgpu/hw/gk20a/hw_ltc_gk20a.h>

/*
 * GPU mapping life cycle
 * ======================
 *
 * Kernel mappings
 * ---------------
 *
 * Kernel mappings are created through vm.map(..., false):
 *
 *  - Mappings to the same allocations are reused and refcounted.
 *  - This path does not support deferred unmapping (i.e. kernel must wait for
 *    all hw operations on the buffer to complete before unmapping).
 *  - References to dmabuf are owned and managed by the (kernel) clients of
 *    the gk20a_vm layer.
 *
 *
 * User space mappings
 * -------------------
 *
 * User space mappings are created through as.map_buffer -> vm.map(..., true):
 *
 *  - Mappings to the same allocations are reused and refcounted.
 *  - This path supports deferred unmapping (i.e. we delay the actual unmapping
 *    until all hw operations have completed).
 *  - References to dmabuf are owned and managed by the vm_gk20a
 *    layer itself. vm.map acquires these refs, and sets
 *    mapped_buffer->own_mem_ref to record that we must release the refs when we
 *    actually unmap.
 *
 */

static int __must_check gk20a_init_system_vm(struct mm_gk20a *mm);
static int __must_check gk20a_init_bar1_vm(struct mm_gk20a *mm);
static int __must_check gk20a_init_hwpm(struct mm_gk20a *mm);
static int __must_check gk20a_init_cde_vm(struct mm_gk20a *mm);
static int __must_check gk20a_init_ce_vm(struct mm_gk20a *mm);

static int gk20a_init_mm_reset_enable_hw(struct gk20a *g)
{
	gk20a_dbg_fn("");
	if (g->ops.fb.reset)
		g->ops.fb.reset(g);

	if (g->ops.clock_gating.slcg_fb_load_gating_prod)
		g->ops.clock_gating.slcg_fb_load_gating_prod(g,
				g->slcg_enabled);
	if (g->ops.clock_gating.slcg_ltc_load_gating_prod)
		g->ops.clock_gating.slcg_ltc_load_gating_prod(g,
				g->slcg_enabled);
	if (g->ops.clock_gating.blcg_fb_load_gating_prod)
		g->ops.clock_gating.blcg_fb_load_gating_prod(g,
				g->blcg_enabled);
	if (g->ops.clock_gating.blcg_ltc_load_gating_prod)
		g->ops.clock_gating.blcg_ltc_load_gating_prod(g,
				g->blcg_enabled);

	if (g->ops.fb.init_fs_state)
		g->ops.fb.init_fs_state(g);

	return 0;
}

static void gk20a_remove_mm_ce_support(struct mm_gk20a *mm)
{
	struct gk20a *g = gk20a_from_mm(mm);

	if (mm->vidmem.ce_ctx_id != (u32)~0)
		gk20a_ce_delete_context_priv(g, mm->vidmem.ce_ctx_id);

	mm->vidmem.ce_ctx_id = (u32)~0;

	nvgpu_vm_put(mm->ce.vm);
}

static void gk20a_remove_mm_support(struct mm_gk20a *mm)
{
	struct gk20a *g = gk20a_from_mm(mm);

	if (g->ops.mm.fault_info_mem_destroy)
		g->ops.mm.fault_info_mem_destroy(g);

	if (g->ops.mm.remove_bar2_vm)
		g->ops.mm.remove_bar2_vm(g);

	if (g->ops.mm.is_bar1_supported(g)) {
		gk20a_free_inst_block(g, &mm->bar1.inst_block);
		nvgpu_vm_put(mm->bar1.vm);
	}

	gk20a_free_inst_block(g, &mm->pmu.inst_block);
	gk20a_free_inst_block(g, &mm->hwpm.inst_block);
	nvgpu_vm_put(mm->pmu.vm);
	nvgpu_vm_put(mm->cde.vm);

	nvgpu_semaphore_sea_destroy(g);
	nvgpu_vidmem_destroy(g);
	nvgpu_pd_cache_fini(g);
}

static int gk20a_alloc_sysmem_flush(struct gk20a *g)
{
	return nvgpu_dma_alloc_sys(g, SZ_4K, &g->mm.sysmem_flush);
}

int gk20a_init_mm_setup_sw(struct gk20a *g)
{
	struct mm_gk20a *mm = &g->mm;
	int err;

	gk20a_dbg_fn("");

	if (mm->sw_ready) {
		gk20a_dbg_fn("skip init");
		return 0;
	}

	mm->g = g;
	nvgpu_mutex_init(&mm->l2_op_lock);

	/*TBD: make channel vm size configurable */
	mm->channel.user_size = NV_MM_DEFAULT_USER_SIZE -
		NV_MM_DEFAULT_KERNEL_SIZE;
	mm->channel.kernel_size = NV_MM_DEFAULT_KERNEL_SIZE;

	gk20a_dbg_info("channel vm size: user %dMB  kernel %dMB",
		       (int)(mm->channel.user_size >> 20),
		       (int)(mm->channel.kernel_size >> 20));

	nvgpu_init_pramin(mm);

	mm->vidmem.ce_ctx_id = (u32)~0;

	err = nvgpu_vidmem_init(mm);
	if (err)
		return err;

	/*
	 * this requires fixed allocations in vidmem which must be
	 * allocated before all other buffers
	 */
	if (g->ops.pmu.alloc_blob_space
			&& !nvgpu_is_enabled(g, NVGPU_MM_UNIFIED_MEMORY)) {
		err = g->ops.pmu.alloc_blob_space(g, 0, &g->acr.ucode_blob);
		if (err)
			return err;
	}

	err = gk20a_alloc_sysmem_flush(g);
	if (err)
		return err;

	if (g->ops.mm.is_bar1_supported(g)) {
		err = gk20a_init_bar1_vm(mm);
		if (err)
			return err;
	}
	if (g->ops.mm.init_bar2_vm) {
		err = g->ops.mm.init_bar2_vm(g);
		if (err)
			return err;
	}
	err = gk20a_init_system_vm(mm);
	if (err)
		return err;

	err = gk20a_init_hwpm(mm);
	if (err)
		return err;

	err = gk20a_init_cde_vm(mm);
	if (err)
		return err;

	err = gk20a_init_ce_vm(mm);
	if (err)
		return err;

	mm->remove_support = gk20a_remove_mm_support;
	mm->remove_ce_support = gk20a_remove_mm_ce_support;

	mm->sw_ready = true;

	gk20a_dbg_fn("done");
	return 0;
}

/* make sure gk20a_init_mm_support is called before */
int gk20a_init_mm_setup_hw(struct gk20a *g)
{
	struct mm_gk20a *mm = &g->mm;
	int err;

	gk20a_dbg_fn("");

	g->ops.fb.set_mmu_page_size(g);
	if (g->ops.fb.set_use_full_comp_tag_line)
		mm->use_full_comp_tag_line =
			g->ops.fb.set_use_full_comp_tag_line(g);

	g->ops.fb.init_hw(g);

	if (g->ops.bus.bar1_bind)
		g->ops.bus.bar1_bind(g, &mm->bar1.inst_block);

	if (g->ops.mm.init_bar2_mm_hw_setup) {
		err = g->ops.mm.init_bar2_mm_hw_setup(g);
		if (err)
			return err;
	}

	if (gk20a_mm_fb_flush(g) || gk20a_mm_fb_flush(g))
		return -EBUSY;

	gk20a_dbg_fn("done");
	return 0;
}

int gk20a_init_mm_support(struct gk20a *g)
{
	u32 err;

	err = gk20a_init_mm_reset_enable_hw(g);
	if (err)
		return err;

	err = gk20a_init_mm_setup_sw(g);
	if (err)
		return err;

	if (g->ops.mm.init_mm_setup_hw)
		err = g->ops.mm.init_mm_setup_hw(g);

	return err;
}

void gk20a_init_mm_ce_context(struct gk20a *g)
{
#if defined(CONFIG_GK20A_VIDMEM)
	if (g->mm.vidmem.size && (g->mm.vidmem.ce_ctx_id == (u32)~0)) {
		g->mm.vidmem.ce_ctx_id =
			gk20a_ce_create_context_with_cb(g,
				gk20a_fifo_get_fast_ce_runlist_id(g),
				-1,
				-1,
				-1,
				NULL);

		if (g->mm.vidmem.ce_ctx_id == (u32)~0)
			nvgpu_err(g,
				"Failed to allocate CE context for vidmem page clearing support");
	}
#endif
}

int gk20a_mm_pde_coverage_bit_count(struct vm_gk20a *vm)
{
	return vm->mmu_levels[0].lo_bit[0];
}

/* for gk20a the "video memory" apertures here are misnomers. */
static inline u32 big_valid_pde0_bits(struct gk20a *g,
				      struct nvgpu_gmmu_pd *pd, u64 addr)
{
	u32 pde0_bits =
		nvgpu_aperture_mask(g, pd->mem,
		  gmmu_pde_aperture_big_sys_mem_ncoh_f(),
		  gmmu_pde_aperture_big_video_memory_f()) |
		gmmu_pde_address_big_sys_f(
			   (u32)(addr >> gmmu_pde_address_shift_v()));

	return pde0_bits;
}

static inline u32 small_valid_pde1_bits(struct gk20a *g,
					struct nvgpu_gmmu_pd *pd, u64 addr)
{
	u32 pde1_bits =
		nvgpu_aperture_mask(g, pd->mem,
		  gmmu_pde_aperture_small_sys_mem_ncoh_f(),
		  gmmu_pde_aperture_small_video_memory_f()) |
		gmmu_pde_vol_small_true_f() | /* tbd: why? */
		gmmu_pde_address_small_sys_f(
			   (u32)(addr >> gmmu_pde_address_shift_v()));

	return pde1_bits;
}

static void update_gmmu_pde_locked(struct vm_gk20a *vm,
				   const struct gk20a_mmu_level *l,
				   struct nvgpu_gmmu_pd *pd,
				   u32 pd_idx,
				   u64 virt_addr,
				   u64 phys_addr,
				   struct nvgpu_gmmu_attrs *attrs)
{
	struct gk20a *g = gk20a_from_vm(vm);
	bool small_valid, big_valid;
	u32 pd_offset = pd_offset_from_index(l, pd_idx);
	u32 pde_v[2] = {0, 0};

	small_valid = attrs->pgsz == gmmu_page_size_small;
	big_valid   = attrs->pgsz == gmmu_page_size_big;

	pde_v[0] = gmmu_pde_size_full_f();
	pde_v[0] |= big_valid ?
		big_valid_pde0_bits(g, pd, phys_addr) :
		gmmu_pde_aperture_big_invalid_f();

	pde_v[1] |= (small_valid ? small_valid_pde1_bits(g, pd, phys_addr) :
		     (gmmu_pde_aperture_small_invalid_f() |
		      gmmu_pde_vol_small_false_f()))
		|
		(big_valid ? (gmmu_pde_vol_big_true_f()) :
		 gmmu_pde_vol_big_false_f());

	pte_dbg(g, attrs,
		"PDE: i=%-4u size=%-2u offs=%-4u pgsz: %c%c | "
		"GPU %#-12llx  phys %#-12llx "
		"[0x%08x, 0x%08x]",
		pd_idx, l->entry_size, pd_offset,
		small_valid ? 'S' : '-',
		big_valid ?   'B' : '-',
		virt_addr, phys_addr,
		pde_v[1], pde_v[0]);

	pd_write(g, &vm->pdb, pd_offset + 0, pde_v[0]);
	pd_write(g, &vm->pdb, pd_offset + 1, pde_v[1]);
}

static void __update_pte_sparse(u32 *pte_w)
{
	pte_w[0]  = gmmu_pte_valid_false_f();
	pte_w[1] |= gmmu_pte_vol_true_f();
}

static void __update_pte(struct vm_gk20a *vm,
			 u32 *pte_w,
			 u64 phys_addr,
			 struct nvgpu_gmmu_attrs *attrs)
{
	struct gk20a *g = gk20a_from_vm(vm);
	u32 page_size = vm->gmmu_page_sizes[attrs->pgsz];
	u32 pte_valid = attrs->valid ?
		gmmu_pte_valid_true_f() :
		gmmu_pte_valid_false_f();
	u32 phys_shifted = phys_addr >> gmmu_pte_address_shift_v();
	u32 addr = attrs->aperture == APERTURE_SYSMEM ?
		gmmu_pte_address_sys_f(phys_shifted) :
		gmmu_pte_address_vid_f(phys_shifted);
	int ctag_shift = ilog2(g->ops.fb.compression_page_size(g));

	pte_w[0] = pte_valid | addr;

	if (attrs->priv)
		pte_w[0] |= gmmu_pte_privilege_true_f();

	pte_w[1] = __nvgpu_aperture_mask(g, attrs->aperture,
					 gmmu_pte_aperture_sys_mem_ncoh_f(),
					 gmmu_pte_aperture_video_memory_f()) |
		gmmu_pte_kind_f(attrs->kind_v) |
		gmmu_pte_comptagline_f((u32)(attrs->ctag >> ctag_shift));

	if (attrs->ctag && vm->mm->use_full_comp_tag_line &&
	    phys_addr & 0x10000)
		pte_w[1] |= gmmu_pte_comptagline_f(
			1 << (gmmu_pte_comptagline_s() - 1));

	if (attrs->rw_flag == gk20a_mem_flag_read_only) {
		pte_w[0] |= gmmu_pte_read_only_true_f();
		pte_w[1] |= gmmu_pte_write_disable_true_f();
	} else if (attrs->rw_flag == gk20a_mem_flag_write_only) {
		pte_w[1] |= gmmu_pte_read_disable_true_f();
	}

	if (!attrs->cacheable)
		pte_w[1] |= gmmu_pte_vol_true_f();

	if (attrs->ctag)
		attrs->ctag += page_size;
}

static void update_gmmu_pte_locked(struct vm_gk20a *vm,
				   const struct gk20a_mmu_level *l,
				   struct nvgpu_gmmu_pd *pd,
				   u32 pd_idx,
				   u64 virt_addr,
				   u64 phys_addr,
				   struct nvgpu_gmmu_attrs *attrs)
{
	struct gk20a *g = gk20a_from_vm(vm);
	u32 page_size  = vm->gmmu_page_sizes[attrs->pgsz];
	u32 pd_offset = pd_offset_from_index(l, pd_idx);
	u32 pte_w[2] = {0, 0};
	int ctag_shift = ilog2(g->ops.fb.compression_page_size(g));

	if (phys_addr)
		__update_pte(vm, pte_w, phys_addr, attrs);
	else if (attrs->sparse)
		__update_pte_sparse(pte_w);

	pte_dbg(g, attrs,
		"PTE: i=%-4u size=%-2u offs=%-4u | "
		"GPU %#-12llx  phys %#-12llx "
		"pgsz: %3dkb perm=%-2s kind=%#02x APT=%-6s %c%c%c%c%c "
		"ctag=0x%08x "
		"[0x%08x, 0x%08x]",
		pd_idx, l->entry_size, pd_offset,
		virt_addr, phys_addr,
		page_size >> 10,
		nvgpu_gmmu_perm_str(attrs->rw_flag),
		attrs->kind_v,
		nvgpu_aperture_str(attrs->aperture),
		attrs->cacheable ? 'C' : 'v',
		attrs->sparse    ? 'S' : '-',
		attrs->priv      ? 'P' : '-',
		attrs->coherent  ? 'c' : '-',
		attrs->valid     ? 'V' : '-',
		(u32)attrs->ctag >> ctag_shift,
		pte_w[1], pte_w[0]);

	pd_write(g, pd, pd_offset + 0, pte_w[0]);
	pd_write(g, pd, pd_offset + 1, pte_w[1]);
}

const struct gk20a_mmu_level gk20a_mm_levels_64k[] = {
	{.hi_bit = {NV_GMMU_VA_RANGE-1, NV_GMMU_VA_RANGE-1},
	 .lo_bit = {26, 26},
	 .update_entry = update_gmmu_pde_locked,
	 .entry_size = 8},
	{.hi_bit = {25, 25},
	 .lo_bit = {12, 16},
	 .update_entry = update_gmmu_pte_locked,
	 .entry_size = 8},
	{.update_entry = NULL}
};

const struct gk20a_mmu_level gk20a_mm_levels_128k[] = {
	{.hi_bit = {NV_GMMU_VA_RANGE-1, NV_GMMU_VA_RANGE-1},
	 .lo_bit = {27, 27},
	 .update_entry = update_gmmu_pde_locked,
	 .entry_size = 8},
	{.hi_bit = {26, 26},
	 .lo_bit = {12, 17},
	 .update_entry = update_gmmu_pte_locked,
	 .entry_size = 8},
	{.update_entry = NULL}
};

/*
 * Attempt to find a reserved memory area to determine PTE size for the passed
 * mapping. If no reserved area can be found use small pages.
 */
enum gmmu_pgsz_gk20a __get_pte_size_fixed_map(struct vm_gk20a *vm,
					      u64 base, u64 size)
{
	struct nvgpu_vm_area *vm_area;

	vm_area = nvgpu_vm_area_find(vm, base);
	if (!vm_area)
		return gmmu_page_size_small;

	return vm_area->pgsz_idx;
}

/*
 * This is for when the address space does not support unified address spaces.
 */
static enum gmmu_pgsz_gk20a __get_pte_size_split_addr(struct vm_gk20a *vm,
					       u64 base, u64 size)
{
	if (!base) {
		if (size >= vm->gmmu_page_sizes[gmmu_page_size_big])
			return gmmu_page_size_big;
		return gmmu_page_size_small;
	} else {
		if (base < __nv_gmmu_va_small_page_limit())
			return gmmu_page_size_small;
		else
			return gmmu_page_size_big;
	}
}

/*
 * This determines the PTE size for a given alloc. Used by both the GVA space
 * allocator and the mm core code so that agreement can be reached on how to
 * map allocations.
 *
 * The page size of a buffer is this:
 *
 *   o  If the VM doesn't support large pages then obviously small pages
 *      must be used.
 *   o  If the base address is non-zero (fixed address map):
 *      - Attempt to find a reserved memory area and use the page size
 *        based on that.
 *      - If no reserved page size is available, default to small pages.
 *   o  If the base is zero:
 *      - If the size is larger than or equal to the big page size, use big
 *        pages.
 *      - Otherwise use small pages.
 */
enum gmmu_pgsz_gk20a __get_pte_size(struct vm_gk20a *vm, u64 base, u64 size)
{
	struct gk20a *g = gk20a_from_vm(vm);

	if (!vm->big_pages)
		return gmmu_page_size_small;

	if (!nvgpu_is_enabled(g, NVGPU_MM_UNIFY_ADDRESS_SPACES))
		return __get_pte_size_split_addr(vm, base, size);

	if (base)
		return __get_pte_size_fixed_map(vm, base, size);

	if (size >= vm->gmmu_page_sizes[gmmu_page_size_big])
		return gmmu_page_size_big;
	return gmmu_page_size_small;
}

int __gk20a_vm_bind_channel(struct vm_gk20a *vm, struct channel_gk20a *ch)
{
	int err = 0;

	gk20a_dbg_fn("");

	nvgpu_vm_get(vm);
	ch->vm = vm;
	err = channel_gk20a_commit_va(ch);
	if (err)
		ch->vm = NULL;

	nvgpu_log(gk20a_from_vm(vm), gpu_dbg_map, "Binding ch=%d -> VM:%s",
		  ch->chid, vm->name);

	return err;
}

int gk20a_vm_bind_channel(struct gk20a_as_share *as_share,
			  struct channel_gk20a *ch)
{
	return __gk20a_vm_bind_channel(as_share->vm, ch);
}

int gk20a_alloc_inst_block(struct gk20a *g, struct nvgpu_mem *inst_block)
{
	int err;

	gk20a_dbg_fn("");

	err = nvgpu_dma_alloc(g, ram_in_alloc_size_v(), inst_block);
	if (err) {
		nvgpu_err(g, "%s: memory allocation failed", __func__);
		return err;
	}

	gk20a_dbg_fn("done");
	return 0;
}

void gk20a_free_inst_block(struct gk20a *g, struct nvgpu_mem *inst_block)
{
	if (inst_block->size)
		nvgpu_dma_free(g, inst_block);
}

u64 gk20a_mm_inst_block_addr(struct gk20a *g, struct nvgpu_mem *inst_block)
{
	if (g->mm.has_physical_mode)
		return nvgpu_mem_get_phys_addr(g, inst_block);
	else
		return nvgpu_mem_get_addr(g, inst_block);
}

static int gk20a_init_bar1_vm(struct mm_gk20a *mm)
{
	int err;
	struct gk20a *g = gk20a_from_mm(mm);
	struct nvgpu_mem *inst_block = &mm->bar1.inst_block;
	u32 big_page_size = g->ops.mm.get_default_big_page_size();

	mm->bar1.aperture_size = bar1_aperture_size_mb_gk20a() << 20;
	gk20a_dbg_info("bar1 vm size = 0x%x", mm->bar1.aperture_size);
	mm->bar1.vm = nvgpu_vm_init(g,
				    big_page_size,
				    SZ_4K,
				    mm->bar1.aperture_size - SZ_4K,
				    mm->bar1.aperture_size,
				    true, false,
				    "bar1");
	if (!mm->bar1.vm)
		return -ENOMEM;

	err = gk20a_alloc_inst_block(g, inst_block);
	if (err)
		goto clean_up_vm;
	g->ops.mm.init_inst_block(inst_block, mm->bar1.vm, big_page_size);

	return 0;

clean_up_vm:
	nvgpu_vm_put(mm->bar1.vm);
	return err;
}

/* pmu vm, share channel_vm interfaces */
static int gk20a_init_system_vm(struct mm_gk20a *mm)
{
	int err;
	struct gk20a *g = gk20a_from_mm(mm);
	struct nvgpu_mem *inst_block = &mm->pmu.inst_block;
	u32 big_page_size = g->ops.mm.get_default_big_page_size();
	u32 low_hole, aperture_size;

	/*
	 * No user region - so we will pass that as zero sized.
	 */
	low_hole = SZ_4K * 16;
	aperture_size = GK20A_PMU_VA_SIZE * 2;

	mm->pmu.aperture_size = GK20A_PMU_VA_SIZE;
	gk20a_dbg_info("pmu vm size = 0x%x", mm->pmu.aperture_size);

	mm->pmu.vm = nvgpu_vm_init(g, big_page_size,
				   low_hole,
				   aperture_size - low_hole,
				   aperture_size,
				   true,
				   false,
				   "system");
	if (!mm->pmu.vm)
		return -ENOMEM;

	err = gk20a_alloc_inst_block(g, inst_block);
	if (err)
		goto clean_up_vm;
	g->ops.mm.init_inst_block(inst_block, mm->pmu.vm, big_page_size);

	return 0;

clean_up_vm:
	nvgpu_vm_put(mm->pmu.vm);
	return err;
}

static int gk20a_init_hwpm(struct mm_gk20a *mm)
{
	int err;
	struct gk20a *g = gk20a_from_mm(mm);
	struct nvgpu_mem *inst_block = &mm->hwpm.inst_block;

	err = gk20a_alloc_inst_block(g, inst_block);
	if (err)
		return err;
	g->ops.mm.init_inst_block(inst_block, mm->pmu.vm, 0);

	return 0;
}

static int gk20a_init_cde_vm(struct mm_gk20a *mm)
{
	struct gk20a *g = gk20a_from_mm(mm);
	u32 big_page_size = g->ops.mm.get_default_big_page_size();

	mm->cde.vm = nvgpu_vm_init(g, big_page_size,
				   big_page_size << 10,
				   NV_MM_DEFAULT_KERNEL_SIZE,
				   NV_MM_DEFAULT_KERNEL_SIZE + NV_MM_DEFAULT_USER_SIZE,
				   false, false, "cde");
	if (!mm->cde.vm)
		return -ENOMEM;
	return 0;
}

static int gk20a_init_ce_vm(struct mm_gk20a *mm)
{
	struct gk20a *g = gk20a_from_mm(mm);
	u32 big_page_size = g->ops.mm.get_default_big_page_size();

	mm->ce.vm = nvgpu_vm_init(g, big_page_size,
				  big_page_size << 10,
				  NV_MM_DEFAULT_KERNEL_SIZE,
				  NV_MM_DEFAULT_KERNEL_SIZE + NV_MM_DEFAULT_USER_SIZE,
				  false, false, "ce");
	if (!mm->ce.vm)
		return -ENOMEM;
	return 0;
}

void gk20a_mm_init_pdb(struct gk20a *g, struct nvgpu_mem *inst_block,
		struct vm_gk20a *vm)
{
	u64 pdb_addr = nvgpu_mem_get_addr(g, vm->pdb.mem);
	u32 pdb_addr_lo = u64_lo32(pdb_addr >> ram_in_base_shift_v());
	u32 pdb_addr_hi = u64_hi32(pdb_addr);

	gk20a_dbg_info("pde pa=0x%llx", pdb_addr);

	nvgpu_mem_wr32(g, inst_block, ram_in_page_dir_base_lo_w(),
		nvgpu_aperture_mask(g, vm->pdb.mem,
		  ram_in_page_dir_base_target_sys_mem_ncoh_f(),
		  ram_in_page_dir_base_target_vid_mem_f()) |
		ram_in_page_dir_base_vol_true_f() |
		ram_in_page_dir_base_lo_f(pdb_addr_lo));

	nvgpu_mem_wr32(g, inst_block, ram_in_page_dir_base_hi_w(),
		ram_in_page_dir_base_hi_f(pdb_addr_hi));
}

void gk20a_init_inst_block(struct nvgpu_mem *inst_block, struct vm_gk20a *vm,
		u32 big_page_size)
{
	struct gk20a *g = gk20a_from_vm(vm);

	gk20a_dbg_info("inst block phys = 0x%llx, kv = 0x%p",
		gk20a_mm_inst_block_addr(g, inst_block), inst_block->cpu_va);

	g->ops.mm.init_pdb(g, inst_block, vm);

	nvgpu_mem_wr32(g, inst_block, ram_in_adr_limit_lo_w(),
		u64_lo32(vm->va_limit - 1) & ~0xfff);

	nvgpu_mem_wr32(g, inst_block, ram_in_adr_limit_hi_w(),
		ram_in_adr_limit_hi_f(u64_hi32(vm->va_limit - 1)));

	if (big_page_size && g->ops.mm.set_big_page_size)
		g->ops.mm.set_big_page_size(g, inst_block, big_page_size);
}

int gk20a_mm_fb_flush(struct gk20a *g)
{
	struct mm_gk20a *mm = &g->mm;
	u32 data;
	int ret = 0;
	struct nvgpu_timeout timeout;

	gk20a_dbg_fn("");

	gk20a_busy_noresume(g);
	if (!g->power_on) {
		gk20a_idle_nosuspend(g);
		return 0;
	}

	nvgpu_timeout_init(g, &timeout, 100, NVGPU_TIMER_RETRY_TIMER);

	nvgpu_mutex_acquire(&mm->l2_op_lock);

	/* Make sure all previous writes are committed to the L2. There's no
	   guarantee that writes are to DRAM. This will be a sysmembar internal
	   to the L2. */

	trace_gk20a_mm_fb_flush(g->name);

	gk20a_writel(g, flush_fb_flush_r(),
		flush_fb_flush_pending_busy_f());

	do {
		data = gk20a_readl(g, flush_fb_flush_r());

		if (flush_fb_flush_outstanding_v(data) ==
			flush_fb_flush_outstanding_true_v() ||
		    flush_fb_flush_pending_v(data) ==
			flush_fb_flush_pending_busy_v()) {
				gk20a_dbg_info("fb_flush 0x%x", data);
				nvgpu_udelay(5);
		} else
			break;
	} while (!nvgpu_timeout_expired(&timeout));

	if (nvgpu_timeout_peek_expired(&timeout)) {
		if (g->ops.fb.dump_vpr_wpr_info)
			g->ops.fb.dump_vpr_wpr_info(g);
		ret = -EBUSY;
	}

	trace_gk20a_mm_fb_flush_done(g->name);

	nvgpu_mutex_release(&mm->l2_op_lock);

	gk20a_idle_nosuspend(g);

	return ret;
}

static void gk20a_mm_l2_invalidate_locked(struct gk20a *g)
{
	u32 data;
	struct nvgpu_timeout timeout;

	trace_gk20a_mm_l2_invalidate(g->name);

	nvgpu_timeout_init(g, &timeout, 200, NVGPU_TIMER_RETRY_TIMER);

	/* Invalidate any clean lines from the L2 so subsequent reads go to
	   DRAM. Dirty lines are not affected by this operation. */
	gk20a_writel(g, flush_l2_system_invalidate_r(),
		flush_l2_system_invalidate_pending_busy_f());

	do {
		data = gk20a_readl(g, flush_l2_system_invalidate_r());

		if (flush_l2_system_invalidate_outstanding_v(data) ==
			flush_l2_system_invalidate_outstanding_true_v() ||
		    flush_l2_system_invalidate_pending_v(data) ==
			flush_l2_system_invalidate_pending_busy_v()) {
				gk20a_dbg_info("l2_system_invalidate 0x%x",
						data);
				nvgpu_udelay(5);
		} else
			break;
	} while (!nvgpu_timeout_expired(&timeout));

	if (nvgpu_timeout_peek_expired(&timeout))
		nvgpu_warn(g, "l2_system_invalidate too many retries");

	trace_gk20a_mm_l2_invalidate_done(g->name);
}

void gk20a_mm_l2_invalidate(struct gk20a *g)
{
	struct mm_gk20a *mm = &g->mm;
	gk20a_busy_noresume(g);
	if (g->power_on) {
		nvgpu_mutex_acquire(&mm->l2_op_lock);
		gk20a_mm_l2_invalidate_locked(g);
		nvgpu_mutex_release(&mm->l2_op_lock);
	}
	gk20a_idle_nosuspend(g);
}

void gk20a_mm_l2_flush(struct gk20a *g, bool invalidate)
{
	struct mm_gk20a *mm = &g->mm;
	u32 data;
	struct nvgpu_timeout timeout;

	gk20a_dbg_fn("");

	gk20a_busy_noresume(g);
	if (!g->power_on)
		goto hw_was_off;

	nvgpu_timeout_init(g, &timeout, 2000, NVGPU_TIMER_RETRY_TIMER);

	nvgpu_mutex_acquire(&mm->l2_op_lock);

	trace_gk20a_mm_l2_flush(g->name);

	/* Flush all dirty lines from the L2 to DRAM. Lines are left in the L2
	   as clean, so subsequent reads might hit in the L2. */
	gk20a_writel(g, flush_l2_flush_dirty_r(),
		flush_l2_flush_dirty_pending_busy_f());

	do {
		data = gk20a_readl(g, flush_l2_flush_dirty_r());

		if (flush_l2_flush_dirty_outstanding_v(data) ==
			flush_l2_flush_dirty_outstanding_true_v() ||
		    flush_l2_flush_dirty_pending_v(data) ==
			flush_l2_flush_dirty_pending_busy_v()) {
				gk20a_dbg_info("l2_flush_dirty 0x%x", data);
				nvgpu_udelay(5);
		} else
			break;
	} while (!nvgpu_timeout_expired_msg(&timeout,
					 "l2_flush_dirty too many retries"));

	trace_gk20a_mm_l2_flush_done(g->name);

	if (invalidate)
		gk20a_mm_l2_invalidate_locked(g);

	nvgpu_mutex_release(&mm->l2_op_lock);

hw_was_off:
	gk20a_idle_nosuspend(g);
}

void gk20a_mm_cbc_clean(struct gk20a *g)
{
	struct mm_gk20a *mm = &g->mm;
	u32 data;
	struct nvgpu_timeout timeout;

	gk20a_dbg_fn("");

	gk20a_busy_noresume(g);
	if (!g->power_on)
		goto hw_was_off;

	nvgpu_timeout_init(g, &timeout, 200, NVGPU_TIMER_RETRY_TIMER);

	nvgpu_mutex_acquire(&mm->l2_op_lock);

	/* Flush all dirty lines from the CBC to L2 */
	gk20a_writel(g, flush_l2_clean_comptags_r(),
		flush_l2_clean_comptags_pending_busy_f());

	do {
		data = gk20a_readl(g, flush_l2_clean_comptags_r());

		if (flush_l2_clean_comptags_outstanding_v(data) ==
			flush_l2_clean_comptags_outstanding_true_v() ||
		    flush_l2_clean_comptags_pending_v(data) ==
			flush_l2_clean_comptags_pending_busy_v()) {
				gk20a_dbg_info("l2_clean_comptags 0x%x", data);
				nvgpu_udelay(5);
		} else
			break;
	} while (!nvgpu_timeout_expired_msg(&timeout,
					 "l2_clean_comptags too many retries"));

	nvgpu_mutex_release(&mm->l2_op_lock);

hw_was_off:
	gk20a_idle_nosuspend(g);
}

int gk20a_mm_suspend(struct gk20a *g)
{
	gk20a_dbg_fn("");

#if defined(CONFIG_GK20A_VIDMEM)
	cancel_work_sync(&g->mm.vidmem.clear_mem_worker);
#endif

	g->ops.mm.cbc_clean(g);
	g->ops.mm.l2_flush(g, false);

	gk20a_dbg_fn("done");
	return 0;
}

u32 gk20a_mm_get_iommu_bit(struct gk20a *g)
{
	return 34;
}

const struct gk20a_mmu_level *gk20a_mm_get_mmu_levels(struct gk20a *g,
						      u32 big_page_size)
{
	return (big_page_size == SZ_64K) ?
		 gk20a_mm_levels_64k : gk20a_mm_levels_128k;
}