gpu: nvgpu: add new gr/config unit to initialize GR configuration

Add new unit gr/config to initialize GR configuration like GPC/TPC
count, MAX count and mask

Create new structure nvgpu_gr_config that stores all the configuration
and that is owned by the new unit

Move below fields from struct gr_gk20a to nvgpu_gr_config in gr/config.h
Struct gr_gk20a now only holds the pointer to struct nvgpu_gr_config

u32 max_gpc_count;
u32 max_tpc_per_gpc_count;
u32 max_zcull_per_gpc_count;
u32 max_tpc_count;

u32 gpc_count;
u32 tpc_count;
u32 ppc_count;
u32 zcb_count;

u32 pe_count_per_gpc;

u32 *gpc_tpc_count;
u32 *gpc_ppc_count;
u32 *gpc_zcb_count;
u32 *pes_tpc_count[GK20A_GR_MAX_PES_PER_GPC];

u32 *gpc_tpc_mask;
u32 *pes_tpc_mask[GK20A_GR_MAX_PES_PER_GPC];
u32 *gpc_skip_mask;

u8 *map_tiles;
u32 map_tile_count;
u32 map_row_offset;

Remove gr->sys_count since it was already no longer used

common/gr/config/gr_config.c unit now exposes the APIs to initialize
the configuration and also to query the configuration values

nvgpu_gr_config_init() is called to initialize GR configuration from
gr_gk20a_init_gr_config() and gr_gk20a_init_map_tiles() is simply
renamed as nvgpu_gr_config_init_map_tiles()

Expose new API nvgpu_gr_config_deinit() to deinit the configuration

Expose nvgpu_gr_config_get_*() APIs to query above configuration
fields stored in nvgpu_gr_config structure

Update vgpu_gr_init_gr_config() to initialize the configuration
from gr->config structure

Chip specific HALs that access GR register for initialization
are implemented in common/gr/config/gr_config_gm20b.c
Set these HALs for all GPUs

Jira NVGPU-1879

Change-Id: Ided658b43124ea61b9f273b82b73fdde4ed3c8f0
Signed-off-by: Deepak Nibade <dnibade@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2012167
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
This commit is contained in:
Deepak Nibade
2019-01-30 19:22:39 +05:30
committed by mobile promotions
parent e212e851a3
commit a5eb150635
32 changed files with 1401 additions and 863 deletions

View File

@@ -32,6 +32,7 @@
#include <nvgpu/list.h>
#include <nvgpu/clk_arb.h>
#include <nvgpu/gk20a.h>
#include <nvgpu/gr/config.h>
#include <nvgpu/channel.h>
#include <nvgpu/pmu/pmgr.h>
@@ -281,10 +282,10 @@ gk20a_ctrl_ioctl_gpu_characteristics(
gpu.L2_cache_size = g->ops.ltc.determine_L2_size_bytes(g);
gpu.on_board_video_memory_size = 0; /* integrated GPU */
gpu.num_gpc = g->gr.gpc_count;
gpu.max_gpc_count = g->gr.max_gpc_count;
gpu.num_gpc = nvgpu_gr_config_get_gpc_count(g->gr.config);
gpu.max_gpc_count = nvgpu_gr_config_get_max_gpc_count(g->gr.config);
gpu.num_tpc_per_gpc = g->gr.max_tpc_per_gpc_count;
gpu.num_tpc_per_gpc = nvgpu_gr_config_get_max_tpc_per_gpc_count(g->gr.config);
gpu.bus_type = NVGPU_GPU_BUS_TYPE_AXI; /* always AXI for now */
@@ -293,7 +294,7 @@ gk20a_ctrl_ioctl_gpu_characteristics(
if (g->ops.gr.get_gpc_mask) {
gpu.gpc_mask = g->ops.gr.get_gpc_mask(g);
} else {
gpu.gpc_mask = BIT32(g->gr.gpc_count) - 1;
gpu.gpc_mask = BIT32(gpu.num_gpc) - 1;
}
gpu.flags = nvgpu_ctrl_ioctl_gpu_characteristics_flags(g);
@@ -553,7 +554,7 @@ static int gk20a_ctrl_get_tpc_masks(struct gk20a *g,
{
struct gr_gk20a *gr = &g->gr;
int err = 0;
const u32 gpc_tpc_mask_size = sizeof(u32) * gr->max_gpc_count;
const u32 gpc_tpc_mask_size = sizeof(u32) * gr->config->max_gpc_count;
if (args->mask_buf_size > 0) {
size_t write_size = gpc_tpc_mask_size;
@@ -564,7 +565,7 @@ static int gk20a_ctrl_get_tpc_masks(struct gk20a *g,
err = copy_to_user((void __user *)(uintptr_t)
args->mask_buf_addr,
gr->gpc_tpc_mask, write_size);
gr->config->gpc_tpc_mask, write_size);
}
if (err == 0)
@@ -687,7 +688,8 @@ static int nvgpu_gpu_ioctl_wait_for_pause(struct gk20a *g,
struct nvgpu_warpstate *w_state = NULL;
u32 sm_count, ioctl_size, size, sm_id;
sm_count = g->gr.gpc_count * g->gr.tpc_count;
sm_count = nvgpu_gr_config_get_gpc_count(g->gr.config) *
nvgpu_gr_config_get_tpc_count(g->gr.config);
ioctl_size = sm_count * sizeof(struct warpstate);
ioctl_w_state = nvgpu_kzalloc(g, ioctl_size);