xiaomi-sm8450/android_kernel_xiaomi_sm8450
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

drm/amdkfd: Eliminate ops_asic_specific in kernel queue

Browse Source 
The ops_asic_specific function pointers are actually quite generic after
using a simple if condition. Eliminate it by code refactoring.

Signed-off-by: Yong Zhao <Yong.Zhao@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
This commit is contained in:
Yong Zhao
2019-11-07 23:59:43 -05:00
committed by Alex Deucher
parent 84ce6c4867
commit ccdef35d07
4 changed files with 26 additions and 125 deletions
Download Patch File Download Diff File Expand all files Collapse all files

63
drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.c
View File

@@ -87,9 +87,17 @@ static bool initialize(struct kernel_queue *kq, struct kfd_dev *dev,
kq->pq_kernel_addr = kq->pq->cpu_ptr; kq->pq_kernel_addr = kq->pq->cpu_ptr;
kq->pq_gpu_addr = kq->pq->gpu_addr; kq->pq_gpu_addr = kq->pq->gpu_addr;
retval = kq->ops_asic_specific.initialize(kq, dev, type, queue_size); /* For CIK family asics, kq->eop_mem is not needed */
if (!retval) if (dev->device_info->asic_family > CHIP_HAWAII) {
goto err_eop_allocate_vidmem; retval = kfd_gtt_sa_allocate(dev, PAGE_SIZE, &kq->eop_mem);
if (retval != 0)
goto err_eop_allocate_vidmem;
kq->eop_gpu_addr = kq->eop_mem->gpu_addr;
kq->eop_kernel_addr = kq->eop_mem->cpu_ptr;
memset(kq->eop_kernel_addr, 0, PAGE_SIZE);
}
retval = kfd_gtt_sa_allocate(dev, sizeof(*kq->rptr_kernel), retval = kfd_gtt_sa_allocate(dev, sizeof(*kq->rptr_kernel),
&kq->rptr_mem); &kq->rptr_mem);
@@ -200,7 +208,12 @@ static void uninitialize(struct kernel_queue *kq)
kfd_gtt_sa_free(kq->dev, kq->rptr_mem); kfd_gtt_sa_free(kq->dev, kq->rptr_mem);
kfd_gtt_sa_free(kq->dev, kq->wptr_mem); kfd_gtt_sa_free(kq->dev, kq->wptr_mem);
kq->ops_asic_specific.uninitialize(kq);
/* For CIK family asics, kq->eop_mem is Null, kfd_gtt_sa_free()
* is able to handle NULL properly.
*/
kfd_gtt_sa_free(kq->dev, kq->eop_mem);
kfd_gtt_sa_free(kq->dev, kq->pq); kfd_gtt_sa_free(kq->dev, kq->pq);
kfd_release_kernel_doorbell(kq->dev, kfd_release_kernel_doorbell(kq->dev,
kq->queue->properties.doorbell_ptr); kq->queue->properties.doorbell_ptr);
@@ -280,8 +293,15 @@ static void submit_packet(struct kernel_queue *kq)
} }
pr_debug("\n"); pr_debug("\n");
#endif #endif
if (kq->dev->device_info->doorbell_size == 8) {
kq->ops_asic_specific.submit_packet(kq); *kq->wptr64_kernel = kq->pending_wptr64;
write_kernel_doorbell64(kq->queue->properties.doorbell_ptr,
kq->pending_wptr64);
} else {
*kq->wptr_kernel = kq->pending_wptr;
write_kernel_doorbell(kq->queue->properties.doorbell_ptr,
kq->pending_wptr);
}
} }
static void rollback_packet(struct kernel_queue *kq) static void rollback_packet(struct kernel_queue *kq)
@@ -310,42 +330,11 @@ struct kernel_queue *kernel_queue_init(struct kfd_dev *dev,
kq->ops.submit_packet = submit_packet; kq->ops.submit_packet = submit_packet;
kq->ops.rollback_packet = rollback_packet; kq->ops.rollback_packet = rollback_packet;
switch (dev->device_info->asic_family) {
case CHIP_KAVERI:
case CHIP_HAWAII:
case CHIP_CARRIZO:
case CHIP_TONGA:
case CHIP_FIJI:
case CHIP_POLARIS10:
case CHIP_POLARIS11:
case CHIP_POLARIS12:
case CHIP_VEGAM:
kernel_queue_init_vi(&kq->ops_asic_specific);
break;
case CHIP_VEGA10:
case CHIP_VEGA12:
case CHIP_VEGA20:
case CHIP_RAVEN:
case CHIP_RENOIR:
case CHIP_ARCTURUS:
case CHIP_NAVI10:
case CHIP_NAVI12:
case CHIP_NAVI14:
kernel_queue_init_v9(&kq->ops_asic_specific);
break;
default:
WARN(1, "Unexpected ASIC family %u",
dev->device_info->asic_family);
goto out_free;
}
if (kq->ops.initialize(kq, dev, type, KFD_KERNEL_QUEUE_SIZE)) if (kq->ops.initialize(kq, dev, type, KFD_KERNEL_QUEUE_SIZE))
return kq; return kq;
pr_err("Failed to init kernel queue\n"); pr_err("Failed to init kernel queue\n");
out_free:
kfree(kq); kfree(kq);
return NULL; return NULL;
} }

4
drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.h
View File

@@ -66,7 +66,6 @@ struct kernel_queue_ops {
struct kernel_queue { struct kernel_queue {
struct kernel_queue_ops ops; struct kernel_queue_ops ops;
struct kernel_queue_ops ops_asic_specific;
/* data */ /* data */
struct kfd_dev *dev; struct kfd_dev *dev;
@@ -99,7 +98,4 @@ struct kernel_queue {
struct list_head list; struct list_head list;
}; };
void kernel_queue_init_vi(struct kernel_queue_ops *ops);
void kernel_queue_init_v9(struct kernel_queue_ops *ops);
#endif /* KFD_KERNEL_QUEUE_H_ */ #endif /* KFD_KERNEL_QUEUE_H_ */

36
drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue_v9.c
View File

@@ -27,42 +27,6 @@
#include "kfd_pm4_opcodes.h" #include "kfd_pm4_opcodes.h"
#include "gc/gc_10_1_0_sh_mask.h" #include "gc/gc_10_1_0_sh_mask.h"
static bool initialize_v9(struct kernel_queue *kq, struct kfd_dev *dev,
enum kfd_queue_type type, unsigned int queue_size)
{
int retval;
retval = kfd_gtt_sa_allocate(dev, PAGE_SIZE, &kq->eop_mem);
if (retval)
return false;
kq->eop_gpu_addr = kq->eop_mem->gpu_addr;
kq->eop_kernel_addr = kq->eop_mem->cpu_ptr;
memset(kq->eop_kernel_addr, 0, PAGE_SIZE);
return true;
}
static void uninitialize_v9(struct kernel_queue *kq)
{
kfd_gtt_sa_free(kq->dev, kq->eop_mem);
}
static void submit_packet_v9(struct kernel_queue *kq)
{
*kq->wptr64_kernel = kq->pending_wptr64;
write_kernel_doorbell64(kq->queue->properties.doorbell_ptr,
kq->pending_wptr64);
}
void kernel_queue_init_v9(struct kernel_queue_ops *ops)
{
ops->initialize = initialize_v9;
ops->uninitialize = uninitialize_v9;
ops->submit_packet = submit_packet_v9;
}
static int pm_map_process_v9(struct packet_manager *pm, static int pm_map_process_v9(struct packet_manager *pm,
uint32_t *buffer, struct qcm_process_device *qpd) uint32_t *buffer, struct qcm_process_device *qpd)
{ {

48
drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue_vi.c
View File

@@ -26,54 +26,6 @@
#include "kfd_pm4_headers_vi.h" #include "kfd_pm4_headers_vi.h"
#include "kfd_pm4_opcodes.h" #include "kfd_pm4_opcodes.h"
static bool initialize_vi(struct kernel_queue *kq, struct kfd_dev *dev,
enum kfd_queue_type type, unsigned int queue_size);
static void uninitialize_vi(struct kernel_queue *kq);
static void submit_packet_vi(struct kernel_queue *kq);
void kernel_queue_init_vi(struct kernel_queue_ops *ops)
{
ops->initialize = initialize_vi;
ops->uninitialize = uninitialize_vi;
ops->submit_packet = submit_packet_vi;
}
static bool initialize_vi(struct kernel_queue *kq, struct kfd_dev *dev,
enum kfd_queue_type type, unsigned int queue_size)
{
int retval;
/*For CIK family asics, kq->eop_mem is not needed */
if (dev->device_info->asic_family <= CHIP_MULLINS)
return true;
retval = kfd_gtt_sa_allocate(dev, PAGE_SIZE, &kq->eop_mem);
if (retval != 0)
return false;
kq->eop_gpu_addr = kq->eop_mem->gpu_addr;
kq->eop_kernel_addr = kq->eop_mem->cpu_ptr;
memset(kq->eop_kernel_addr, 0, PAGE_SIZE);
return true;
}
static void uninitialize_vi(struct kernel_queue *kq)
{
/* For CIK family asics, kq->eop_mem is Null, kfd_gtt_sa_free()
* is able to handle NULL properly.
*/
kfd_gtt_sa_free(kq->dev, kq->eop_mem);
}
static void submit_packet_vi(struct kernel_queue *kq)
{
*kq->wptr_kernel = kq->pending_wptr;
write_kernel_doorbell(kq->queue->properties.doorbell_ptr,
kq->pending_wptr);
}
unsigned int pm_build_pm4_header(unsigned int opcode, size_t packet_size) unsigned int pm_build_pm4_header(unsigned int opcode, size_t packet_size)
{ {
union PM4_MES_TYPE_3_HEADER header; union PM4_MES_TYPE_3_HEADER header;