linux_dsm_epyc7002/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
Junwei Zhang bb812f1ea8 drm/amdgpu: allocate gart memory when it's required (v3)
Instead of calling gart address space on every bo pin,
allocates it on demand

v2: fix error handling
v3: drop the change on amdgpu_amdkfd_gpuvm.c, not needed.

Signed-off-by: Junwei Zhang <Jerry.Zhang@amd.com>
Acked-by: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2018-07-10 14:15:58 -05:00

514 lines
13 KiB
C

/*
* Copyright 2014 Advanced Micro Devices, Inc.
*
* 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 COPYRIGHT HOLDER(S) OR AUTHOR(S) 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 "amdgpu_amdkfd.h"
#include "amd_shared.h"
#include <drm/drmP.h>
#include "amdgpu.h"
#include "amdgpu_gfx.h"
#include <linux/module.h>
const struct kgd2kfd_calls *kgd2kfd;
bool (*kgd2kfd_init_p)(unsigned int, const struct kgd2kfd_calls**);
static const unsigned int compute_vmid_bitmap = 0xFF00;
int amdgpu_amdkfd_init(void)
{
int ret;
#if defined(CONFIG_HSA_AMD_MODULE)
int (*kgd2kfd_init_p)(unsigned int, const struct kgd2kfd_calls**);
kgd2kfd_init_p = symbol_request(kgd2kfd_init);
if (kgd2kfd_init_p == NULL)
return -ENOENT;
ret = kgd2kfd_init_p(KFD_INTERFACE_VERSION, &kgd2kfd);
if (ret) {
symbol_put(kgd2kfd_init);
kgd2kfd = NULL;
}
#elif defined(CONFIG_HSA_AMD)
ret = kgd2kfd_init(KFD_INTERFACE_VERSION, &kgd2kfd);
if (ret)
kgd2kfd = NULL;
#else
kgd2kfd = NULL;
ret = -ENOENT;
#endif
#if defined(CONFIG_HSA_AMD_MODULE) || defined(CONFIG_HSA_AMD)
amdgpu_amdkfd_gpuvm_init_mem_limits();
#endif
return ret;
}
void amdgpu_amdkfd_fini(void)
{
if (kgd2kfd) {
kgd2kfd->exit();
symbol_put(kgd2kfd_init);
}
}
void amdgpu_amdkfd_device_probe(struct amdgpu_device *adev)
{
const struct kfd2kgd_calls *kfd2kgd;
if (!kgd2kfd)
return;
switch (adev->asic_type) {
#ifdef CONFIG_DRM_AMDGPU_CIK
case CHIP_KAVERI:
case CHIP_HAWAII:
kfd2kgd = amdgpu_amdkfd_gfx_7_get_functions();
break;
#endif
case CHIP_CARRIZO:
case CHIP_TONGA:
case CHIP_FIJI:
case CHIP_POLARIS10:
case CHIP_POLARIS11:
kfd2kgd = amdgpu_amdkfd_gfx_8_0_get_functions();
break;
case CHIP_VEGA10:
case CHIP_RAVEN:
kfd2kgd = amdgpu_amdkfd_gfx_9_0_get_functions();
break;
default:
dev_info(adev->dev, "kfd not supported on this ASIC\n");
return;
}
adev->kfd = kgd2kfd->probe((struct kgd_dev *)adev,
adev->pdev, kfd2kgd);
}
/**
* amdgpu_doorbell_get_kfd_info - Report doorbell configuration required to
* setup amdkfd
*
* @adev: amdgpu_device pointer
* @aperture_base: output returning doorbell aperture base physical address
* @aperture_size: output returning doorbell aperture size in bytes
* @start_offset: output returning # of doorbell bytes reserved for amdgpu.
*
* amdgpu and amdkfd share the doorbell aperture. amdgpu sets it up,
* takes doorbells required for its own rings and reports the setup to amdkfd.
* amdgpu reserved doorbells are at the start of the doorbell aperture.
*/
static void amdgpu_doorbell_get_kfd_info(struct amdgpu_device *adev,
phys_addr_t *aperture_base,
size_t *aperture_size,
size_t *start_offset)
{
/*
* The first num_doorbells are used by amdgpu.
* amdkfd takes whatever's left in the aperture.
*/
if (adev->doorbell.size > adev->doorbell.num_doorbells * sizeof(u32)) {
*aperture_base = adev->doorbell.base;
*aperture_size = adev->doorbell.size;
*start_offset = adev->doorbell.num_doorbells * sizeof(u32);
} else {
*aperture_base = 0;
*aperture_size = 0;
*start_offset = 0;
}
}
void amdgpu_amdkfd_device_init(struct amdgpu_device *adev)
{
int i;
int last_valid_bit;
if (adev->kfd) {
struct kgd2kfd_shared_resources gpu_resources = {
.compute_vmid_bitmap = compute_vmid_bitmap,
.num_pipe_per_mec = adev->gfx.mec.num_pipe_per_mec,
.num_queue_per_pipe = adev->gfx.mec.num_queue_per_pipe,
.gpuvm_size = min(adev->vm_manager.max_pfn
<< AMDGPU_GPU_PAGE_SHIFT,
AMDGPU_VA_HOLE_START),
.drm_render_minor = adev->ddev->render->index
};
/* this is going to have a few of the MSBs set that we need to
* clear */
bitmap_complement(gpu_resources.queue_bitmap,
adev->gfx.mec.queue_bitmap,
KGD_MAX_QUEUES);
/* remove the KIQ bit as well */
if (adev->gfx.kiq.ring.ready)
clear_bit(amdgpu_gfx_queue_to_bit(adev,
adev->gfx.kiq.ring.me - 1,
adev->gfx.kiq.ring.pipe,
adev->gfx.kiq.ring.queue),
gpu_resources.queue_bitmap);
/* According to linux/bitmap.h we shouldn't use bitmap_clear if
* nbits is not compile time constant */
last_valid_bit = 1 /* only first MEC can have compute queues */
* adev->gfx.mec.num_pipe_per_mec
* adev->gfx.mec.num_queue_per_pipe;
for (i = last_valid_bit; i < KGD_MAX_QUEUES; ++i)
clear_bit(i, gpu_resources.queue_bitmap);
amdgpu_doorbell_get_kfd_info(adev,
&gpu_resources.doorbell_physical_address,
&gpu_resources.doorbell_aperture_size,
&gpu_resources.doorbell_start_offset);
if (adev->asic_type >= CHIP_VEGA10) {
/* On SOC15 the BIF is involved in routing
* doorbells using the low 12 bits of the
* address. Communicate the assignments to
* KFD. KFD uses two doorbell pages per
* process in case of 64-bit doorbells so we
* can use each doorbell assignment twice.
*/
gpu_resources.sdma_doorbell[0][0] =
AMDGPU_DOORBELL64_sDMA_ENGINE0;
gpu_resources.sdma_doorbell[0][1] =
AMDGPU_DOORBELL64_sDMA_ENGINE0 + 0x200;
gpu_resources.sdma_doorbell[1][0] =
AMDGPU_DOORBELL64_sDMA_ENGINE1;
gpu_resources.sdma_doorbell[1][1] =
AMDGPU_DOORBELL64_sDMA_ENGINE1 + 0x200;
/* Doorbells 0x0f0-0ff and 0x2f0-2ff are reserved for
* SDMA, IH and VCN. So don't use them for the CP.
*/
gpu_resources.reserved_doorbell_mask = 0x1f0;
gpu_resources.reserved_doorbell_val = 0x0f0;
}
kgd2kfd->device_init(adev->kfd, &gpu_resources);
}
}
void amdgpu_amdkfd_device_fini(struct amdgpu_device *adev)
{
if (adev->kfd) {
kgd2kfd->device_exit(adev->kfd);
adev->kfd = NULL;
}
}
void amdgpu_amdkfd_interrupt(struct amdgpu_device *adev,
const void *ih_ring_entry)
{
if (adev->kfd)
kgd2kfd->interrupt(adev->kfd, ih_ring_entry);
}
void amdgpu_amdkfd_suspend(struct amdgpu_device *adev)
{
if (adev->kfd)
kgd2kfd->suspend(adev->kfd);
}
int amdgpu_amdkfd_resume(struct amdgpu_device *adev)
{
int r = 0;
if (adev->kfd)
r = kgd2kfd->resume(adev->kfd);
return r;
}
int alloc_gtt_mem(struct kgd_dev *kgd, size_t size,
void **mem_obj, uint64_t *gpu_addr,
void **cpu_ptr)
{
struct amdgpu_device *adev = (struct amdgpu_device *)kgd;
struct amdgpu_bo *bo = NULL;
struct amdgpu_bo_param bp;
int r;
void *cpu_ptr_tmp = NULL;
memset(&bp, 0, sizeof(bp));
bp.size = size;
bp.byte_align = PAGE_SIZE;
bp.domain = AMDGPU_GEM_DOMAIN_GTT;
bp.flags = AMDGPU_GEM_CREATE_CPU_GTT_USWC;
bp.type = ttm_bo_type_kernel;
bp.resv = NULL;
r = amdgpu_bo_create(adev, &bp, &bo);
if (r) {
dev_err(adev->dev,
"failed to allocate BO for amdkfd (%d)\n", r);
return r;
}
/* map the buffer */
r = amdgpu_bo_reserve(bo, true);
if (r) {
dev_err(adev->dev, "(%d) failed to reserve bo for amdkfd\n", r);
goto allocate_mem_reserve_bo_failed;
}
r = amdgpu_bo_pin(bo, AMDGPU_GEM_DOMAIN_GTT);
if (r) {
dev_err(adev->dev, "(%d) failed to pin bo for amdkfd\n", r);
goto allocate_mem_pin_bo_failed;
}
r = amdgpu_ttm_alloc_gart(&bo->tbo);
if (r) {
dev_err(adev->dev, "%p bind failed\n", bo);
goto allocate_mem_kmap_bo_failed;
}
r = amdgpu_bo_kmap(bo, &cpu_ptr_tmp);
if (r) {
dev_err(adev->dev,
"(%d) failed to map bo to kernel for amdkfd\n", r);
goto allocate_mem_kmap_bo_failed;
}
*mem_obj = bo;
*gpu_addr = amdgpu_bo_gpu_offset(bo);
*cpu_ptr = cpu_ptr_tmp;
amdgpu_bo_unreserve(bo);
return 0;
allocate_mem_kmap_bo_failed:
amdgpu_bo_unpin(bo);
allocate_mem_pin_bo_failed:
amdgpu_bo_unreserve(bo);
allocate_mem_reserve_bo_failed:
amdgpu_bo_unref(&bo);
return r;
}
void free_gtt_mem(struct kgd_dev *kgd, void *mem_obj)
{
struct amdgpu_bo *bo = (struct amdgpu_bo *) mem_obj;
amdgpu_bo_reserve(bo, true);
amdgpu_bo_kunmap(bo);
amdgpu_bo_unpin(bo);
amdgpu_bo_unreserve(bo);
amdgpu_bo_unref(&(bo));
}
void get_local_mem_info(struct kgd_dev *kgd,
struct kfd_local_mem_info *mem_info)
{
struct amdgpu_device *adev = (struct amdgpu_device *)kgd;
uint64_t address_mask = adev->dev->dma_mask ? ~*adev->dev->dma_mask :
~((1ULL << 32) - 1);
resource_size_t aper_limit = adev->gmc.aper_base + adev->gmc.aper_size;
memset(mem_info, 0, sizeof(*mem_info));
if (!(adev->gmc.aper_base & address_mask || aper_limit & address_mask)) {
mem_info->local_mem_size_public = adev->gmc.visible_vram_size;
mem_info->local_mem_size_private = adev->gmc.real_vram_size -
adev->gmc.visible_vram_size;
} else {
mem_info->local_mem_size_public = 0;
mem_info->local_mem_size_private = adev->gmc.real_vram_size;
}
mem_info->vram_width = adev->gmc.vram_width;
pr_debug("Address base: %pap limit %pap public 0x%llx private 0x%llx\n",
&adev->gmc.aper_base, &aper_limit,
mem_info->local_mem_size_public,
mem_info->local_mem_size_private);
if (amdgpu_sriov_vf(adev))
mem_info->mem_clk_max = adev->clock.default_mclk / 100;
else if (adev->powerplay.pp_funcs)
mem_info->mem_clk_max = amdgpu_dpm_get_mclk(adev, false) / 100;
else
mem_info->mem_clk_max = 100;
}
uint64_t get_gpu_clock_counter(struct kgd_dev *kgd)
{
struct amdgpu_device *adev = (struct amdgpu_device *)kgd;
if (adev->gfx.funcs->get_gpu_clock_counter)
return adev->gfx.funcs->get_gpu_clock_counter(adev);
return 0;
}
uint32_t get_max_engine_clock_in_mhz(struct kgd_dev *kgd)
{
struct amdgpu_device *adev = (struct amdgpu_device *)kgd;
/* the sclk is in quantas of 10kHz */
if (amdgpu_sriov_vf(adev))
return adev->clock.default_sclk / 100;
else if (adev->powerplay.pp_funcs)
return amdgpu_dpm_get_sclk(adev, false) / 100;
else
return 100;
}
void get_cu_info(struct kgd_dev *kgd, struct kfd_cu_info *cu_info)
{
struct amdgpu_device *adev = (struct amdgpu_device *)kgd;
struct amdgpu_cu_info acu_info = adev->gfx.cu_info;
memset(cu_info, 0, sizeof(*cu_info));
if (sizeof(cu_info->cu_bitmap) != sizeof(acu_info.bitmap))
return;
cu_info->cu_active_number = acu_info.number;
cu_info->cu_ao_mask = acu_info.ao_cu_mask;
memcpy(&cu_info->cu_bitmap[0], &acu_info.bitmap[0],
sizeof(acu_info.bitmap));
cu_info->num_shader_engines = adev->gfx.config.max_shader_engines;
cu_info->num_shader_arrays_per_engine = adev->gfx.config.max_sh_per_se;
cu_info->num_cu_per_sh = adev->gfx.config.max_cu_per_sh;
cu_info->simd_per_cu = acu_info.simd_per_cu;
cu_info->max_waves_per_simd = acu_info.max_waves_per_simd;
cu_info->wave_front_size = acu_info.wave_front_size;
cu_info->max_scratch_slots_per_cu = acu_info.max_scratch_slots_per_cu;
cu_info->lds_size = acu_info.lds_size;
}
uint64_t amdgpu_amdkfd_get_vram_usage(struct kgd_dev *kgd)
{
struct amdgpu_device *adev = (struct amdgpu_device *)kgd;
return amdgpu_vram_mgr_usage(&adev->mman.bdev.man[TTM_PL_VRAM]);
}
int amdgpu_amdkfd_submit_ib(struct kgd_dev *kgd, enum kgd_engine_type engine,
uint32_t vmid, uint64_t gpu_addr,
uint32_t *ib_cmd, uint32_t ib_len)
{
struct amdgpu_device *adev = (struct amdgpu_device *)kgd;
struct amdgpu_job *job;
struct amdgpu_ib *ib;
struct amdgpu_ring *ring;
struct dma_fence *f = NULL;
int ret;
switch (engine) {
case KGD_ENGINE_MEC1:
ring = &adev->gfx.compute_ring[0];
break;
case KGD_ENGINE_SDMA1:
ring = &adev->sdma.instance[0].ring;
break;
case KGD_ENGINE_SDMA2:
ring = &adev->sdma.instance[1].ring;
break;
default:
pr_err("Invalid engine in IB submission: %d\n", engine);
ret = -EINVAL;
goto err;
}
ret = amdgpu_job_alloc(adev, 1, &job, NULL);
if (ret)
goto err;
ib = &job->ibs[0];
memset(ib, 0, sizeof(struct amdgpu_ib));
ib->gpu_addr = gpu_addr;
ib->ptr = ib_cmd;
ib->length_dw = ib_len;
/* This works for NO_HWS. TODO: need to handle without knowing VMID */
job->vmid = vmid;
ret = amdgpu_ib_schedule(ring, 1, ib, job, &f);
if (ret) {
DRM_ERROR("amdgpu: failed to schedule IB.\n");
goto err_ib_sched;
}
ret = dma_fence_wait(f, false);
err_ib_sched:
dma_fence_put(f);
amdgpu_job_free(job);
err:
return ret;
}
bool amdgpu_amdkfd_is_kfd_vmid(struct amdgpu_device *adev, u32 vmid)
{
if (adev->kfd) {
if ((1 << vmid) & compute_vmid_bitmap)
return true;
}
return false;
}
#if !defined(CONFIG_HSA_AMD_MODULE) && !defined(CONFIG_HSA_AMD)
bool amdkfd_fence_check_mm(struct dma_fence *f, struct mm_struct *mm)
{
return false;
}
void amdgpu_amdkfd_unreserve_system_memory_limit(struct amdgpu_bo *bo)
{
}
void amdgpu_amdkfd_gpuvm_destroy_cb(struct amdgpu_device *adev,
struct amdgpu_vm *vm)
{
}
struct amdgpu_amdkfd_fence *to_amdgpu_amdkfd_fence(struct dma_fence *f)
{
return NULL;
}
int amdgpu_amdkfd_evict_userptr(struct kgd_mem *mem, struct mm_struct *mm)
{
return 0;
}
struct kfd2kgd_calls *amdgpu_amdkfd_gfx_7_get_functions(void)
{
return NULL;
}
struct kfd2kgd_calls *amdgpu_amdkfd_gfx_8_0_get_functions(void)
{
return NULL;
}
struct kfd2kgd_calls *amdgpu_amdkfd_gfx_9_0_get_functions(void)
{
return NULL;
}
#endif