Skip to content

I
imagej-elphel
Commit 2f464dff authored by Andrey Filippov's avatar Andrey Filippov
Browse files
Options

updated GPU code

parent 3c033192
Hide whitespace changes
Inline Side-by-side
Showing with 46 additions and 1217 deletions
src/main/resources/kernels/TileProcessor.cuh
View file @ 2f464dff
......@@ -1268,8 +1268,6 @@ __global__ void generate_RBGA(
#endif
/* */
textures_accumulate<<<grid_texture,threads_texture>>>(
// get rid of border tile
// border_tile, // int border_tile, // if 1 - watch for border
woi, // int * woi, // x, y, width,height
gpu_clt, // float ** gpu_clt, // [NUM_CAMS] ->[TILESY][TILESX][NUM_COLORS][DTT_SIZE*DTT_SIZE]
ntt, // size_t num_texture_tiles, // number of texture tiles to process
......@@ -2418,30 +2416,26 @@ __global__ void imclt_rbg(
// save result (back)
float * rbg_p = rbg_top;
mclt_tile = mclt_top;
if ((tileX == 0) && (tileY == 0)){
#pragma unroll
for (int i = 0; i < DTT_SIZE2; i++){
*rbg_p = 100.0f; // just testing
mclt_tile += DTT_SIZE21;
rbg_p += dstride; // DTT_SIZE2; // FIXME
}
#ifdef DBG_MARK_DBG_TILE
} else if ((tileX == DBG_TILE_X) && (tileY == DBG_TILE_Y)){
if ((tileX == DBG_TILE_X) && (tileY == DBG_TILE_Y)){
#pragma unroll
for (int i = 0; i < DTT_SIZE2; i++){
*rbg_p = (*mclt_tile) * 2.0; // just testing
mclt_tile += DTT_SIZE21;
rbg_p += dstride; // DTT_SIZE2; // FIXME
}
#endif
} else {
#endif // #ifdef DBG_MARK_DBG_TILE
#pragma unroll
for (int i = 0; i < DTT_SIZE2; i++){
*rbg_p = *mclt_tile;
mclt_tile += DTT_SIZE21;
rbg_p += dstride; // DTT_SIZE2; // FIXME
}
#ifdef DBG_MARK_DBG_TILE
}
#endif //#ifdef DBG_MARK_DBG_TILE
}
......@@ -2796,29 +2790,30 @@ __device__ void convertCorrectTile(
float window_vert_cos [2*DTT_SIZE])
{
// TODO: pass these values instead of constants to handle EO/LWIR
int max_px = IMG_WIDTH - 1; // odd
int max_py = IMG_HEIGHT - 1; // odd
int max_pxm1 = max_px - 1; // even
int max_pym1 = max_py - 1; // even
int max_kernel_hor = KERNELS_HOR -1;
int max_kernel_vert = KERNELS_VERT-1;
int ktileX, ktileY;
int kernel_index; // common for all coors
float kdx, kdy;
if (threadIdx.x == 0){
ktileX = min(KERNELS_HOR-1, max(0, ((int) lrintf(centerX * (1.0/KERNELS_STEP)+1))));
ktileY = min(KERNELS_VERT-1, max(0, ((int) lrintf(centerY * (1.0/KERNELS_STEP)+1))));
ktileX = min(max_kernel_hor, max(0, ((int) lrintf(centerX * (1.0/KERNELS_STEP)+1))));
ktileY = min(max_kernel_vert, max(0, ((int) lrintf(centerY * (1.0/KERNELS_STEP)+1))));
kdx = centerX - (ktileX << KERNELS_LSTEP) + (1 << (KERNELS_LSTEP -1)); // difference in pixel
kdy = centerY - (ktileY << KERNELS_LSTEP) + (1 << (KERNELS_LSTEP -1)); // difference in pixel
#ifdef USE_UMUL24
kernel_index = __umul24((ktileX + __umul24(ktileY, KERNELS_HOR)), NUM_COLORS);
#else
kernel_index = (ktileX + ktileY * KERNELS_HOR) * NUM_COLORS;
#endif
}
//// __syncthreads();// __syncwarp();
// broadcast kernel_index
kernel_index = __shfl_sync(
0xffffffff, // unsigned mask,
kernel_index, // T var,
0, // int srcLane,
THREADS_PER_TILE); // int width=warpSize);
//// __syncthreads();// __syncwarp(); // is it needed?
kdx = __shfl_sync(
0xffffffff, // unsigned mask,
kdx, // T var,
......@@ -2834,12 +2829,7 @@ __device__ void convertCorrectTile(
float px, py;
// copy kernel
int kernel_full_index = kernel_index + color;
#ifdef USE_UMUL24
float * kernel_src = gpu_kernels + __umul24(kernel_full_index, (DTT_SIZE * DTT_SIZE * 4));
#else
float * kernel_src = gpu_kernels + kernel_full_index* (DTT_SIZE * DTT_SIZE * 4);
#endif
float * kernelp = clt_kernels;
kernel_src += threadIdx.x; // lsb;
......@@ -2857,12 +2847,6 @@ __device__ void convertCorrectTile(
px = centerX - DTT_SIZE - (clt_extra->data_x + clt_extra->dxc_dx * kdx + clt_extra->dxc_dy * kdy) ; // fractional left corner
int itlx = (int) floorf(px +0.5f);
if (itlx < 0){
itlx &= 1; // for color - extend by pairs
}
if (itlx >= (IMG_WIDTH - DTT_SIZE)){
itlx = itlx & 1 +(IMG_WIDTH - DTT_SIZE - 2); // for color - extend by pairs
}
int_topleft [0] = itlx;
float shift_hor = itlx - px;
residual_shift[0] = shift_hor;
......@@ -2898,15 +2882,9 @@ __device__ void convertCorrectTile(
py = centerY - DTT_SIZE - (clt_extra->data_y + clt_extra->dyc_dx * kdx + clt_extra->dyc_dy * kdy) ; // fractional top corner
int itly = (int) floorf(py +0.5f);
if (itly < 0){
itly &= 1; // for color - extend by pairs
}
if (itly >= (IMG_HEIGHT - DTT_SIZE)){
itly = (itly & 1) +(IMG_HEIGHT - DTT_SIZE - 2); // for color - extend by pairs
}
int_topleft[1] = itly;
#ifdef DEBUG_OOB1
#ifdef DEBUG_OOB11
if ((int_topleft[0] < 0) || (int_topleft[1] < 0) || (int_topleft[0] >= (IMG_WIDTH - DTT_SIZE)) || (int_topleft[1] >= IMG_HEIGHT - DTT_SIZE)){
printf("Source data OOB, left=%d, top=%d\n",int_topleft[0],int_topleft[1]);
printf("\n");
......@@ -2964,68 +2942,47 @@ __device__ void convertCorrectTile(
int row_tl = int_topleft[1];
// for red, blue and green, pass 0
int local_col = ((col_tl & 1) ^ (BAYER_RED_COL ^ color0 ^ color1 ^ gpass)) + (threadIdx.x << 1); // green red row: invert column from red
// int local_row = ((row_tl & 1) ^ BAYER_RED_ROW ^ gpass); // use red row
// int local_row = ((row_tl & 1) ^ BAYER_RED_ROW ^ color0 ^ color1 ^ gpass); // use red row
int local_row = ((row_tl & 1) ^ BAYER_RED_ROW ^ color0 ^ gpass); // use red row
float hwind_cos = window_hor_cos[local_col];
float hwind_sin = window_hor_sin[local_col]; // **** only used for green
int dtt_offset = fold_indx2[local_row][local_col];
int dtt_offset_inc = fold_inc[local_row];
#ifdef USE_UMUL24
float *dct_buf = clt_tile + __umul24(gpass << 1 , DTT_SIZE * DTT_SIZE1);
float *dst_buf = clt_tile + __umul24((gpass << 1) + 1 , DTT_SIZE * DTT_SIZE1); // **** only used for green
#else
float *dct_buf = clt_tile + ((gpass << 1) * (DTT_SIZE * DTT_SIZE1));
float *dst_buf = clt_tile + (((gpass << 1) + 1) * (DTT_SIZE * DTT_SIZE1)); // **** only used for green
#endif
if ((col_tl >= 0) && ((col_tl < (IMG_WIDTH - DTT_SIZE * 2))) && (row_tl >= 0) && ((row_tl < (IMG_HEIGHT - DTT_SIZE * 2)))) {
#ifdef USE_UMUL24
float *image_p = gpu_images + __umul24(row_tl + local_row, dstride)+ col_tl + local_col;
#else
float *image_p = gpu_images + dstride * (row_tl + local_row)+ col_tl + local_col;
#endif
#pragma unroll
for (int i = 0; i < 8; i++) {
// float d = (*image_p) * window_vert_cos[local_row]; //warp illegal address (0,2,1)
float d = (*image_p);
d *= window_vert_cos[local_row]; //warp illegal address (0,2,1)
int dtt_offset1 = dtt_offset + (dtt_offset >> 3); // converting for 9-long rows (DTT_SIZE1)
dct_buf[dtt_offset1] = d * hwind_cos;
dst_buf[dtt_offset1] = d * hwind_sin; // **** only used for green
dtt_offset = ( dtt_offset + ((dtt_offset_inc & 0xf) << 3)) & 0x3f;
dtt_offset_inc >>= 4;
local_row += 2;
image_p += dstride2;
}
} else { // handling border tiles (slower)
int eff_col = (min(IMG_HEIGHT/2 -1, max(0, col_tl >> 1)) << 1) + (col_tl & 1);
int row_lsb = row_tl & 1;
int row_pair = row_tl >> 1;
#ifdef USE_UMUL24
float *image_p = gpu_images + __umul24(local_row, dstride) + (eff_col + local_col);
#else
float *image_p = gpu_images + dstride * local_row+ (eff_col + local_col);
#endif
#pragma unroll
for (int i = 0; i < 8; i++) {
int eff_row = (min(IMG_WIDTH/2 - 1, max(0, row_pair + i)) << 1) + row_lsb;
#ifdef USE_UMUL24
float d = image_p[__umul24(eff_row,dstride)] * window_vert_cos[local_row];
#else
float d = image_p[dstride * eff_row] * window_vert_cos[local_row];
#endif
int dtt_offset1 = dtt_offset + (dtt_offset >> 3); // converting for 9-long rows (DTT_SIZE1)
dct_buf[dtt_offset1] = d * hwind_cos;
dst_buf[dtt_offset1] = d * hwind_sin; // **** only used for green
dtt_offset = ( dtt_offset + ((dtt_offset_inc & 0xf) << 3)) & 0x3f;
dtt_offset_inc >>= 4;
local_row += 2;
}
int col_src = col_tl + local_col;
if (col_src < 0) {
col_src &= 1; // same Bayer
} else if (col_src > max_px){
col_src = (col_src & 1) + max_pxm1;
}
int row_src = row_tl + local_row;
int row_use = row_src;
if (row_use < 0) {
row_use &= 1; // same Bayer
} else if (row_use > max_py){
row_use = (row_use & 1) + max_pym1;
}
// float *image_p = gpu_images + dstride * (row_tl + local_row)+ col_tl + local_col;
float *image_p = gpu_images + dstride * row_use + col_src;
#pragma unroll
for (int i = 0; i < 8; i++) {
float d = (*image_p);
d *= window_vert_cos[local_row]; //warp illegal address (0,2,1)
int dtt_offset1 = dtt_offset + (dtt_offset >> 3); // converting for 9-long rows (DTT_SIZE1)
dct_buf[dtt_offset1] = d * hwind_cos;
dst_buf[dtt_offset1] = d * hwind_sin; // **** only used for green
dtt_offset = ( dtt_offset + ((dtt_offset_inc & 0xf) << 3)) & 0x3f;
dtt_offset_inc >>= 4;
local_row += 2;
row_src +=2;
if ((row_src >= 0) && (row_src <= max_pym1)){
image_p += dstride2;
}
}
}
__syncthreads();// __syncwarp();
#ifdef DEBUG2
......@@ -3058,35 +3015,6 @@ __device__ void convertCorrectTile(
dttiv_color_2d(
clt_tile,
color);
/*
dctiv_nodiverg( // all colors
clt_tile + (DTT_SIZE1 * threadIdx.x), // [0][threadIdx.x], // pointer to start of row
1); //int inc);
if (color == BAYER_GREEN){
dstiv_nodiverg( // all colors
clt_tile + DTT_SIZE1 * threadIdx.x + DTT_SIZE1 * DTT_SIZE, // clt_tile[1][threadIdx.x], // pointer to start of row
1); //int inc);
}
__syncthreads();// __syncwarp();
#ifdef DEBUG2
if ((threadIdx.x) == 0){
printf("\nDTT Tiles after horizontal pass, color=%d\n",color);
debug_print_clt1(clt_tile, color, (color== BAYER_GREEN)?3:1); // only 1 quadrant for R,B and 2 - for G
}
__syncthreads();// __syncwarp();
#endif
dctiv_nodiverg( // all colors
clt_tile + threadIdx.x, // &clt_tile[0][0][threadIdx.x], // pointer to start of column
DTT_SIZE1); // int inc,
if (color == BAYER_GREEN){
dctiv_nodiverg( // all colors
clt_tile + threadIdx.x + (DTT_SIZE1 * DTT_SIZE), // &clt_tile[1][0][threadIdx.x], // pointer to start of column
DTT_SIZE1); // int inc,
}
__syncthreads();// __syncwarp();
*/
#ifdef DEBUG2
if ((threadIdx.x) == 0){
......
src/main/resources/kernels/test_tp.cu deleted 100644 → 0
View file @ 3c033192
/**
**
** dtt8x8.cu - CPU test code to run GPU tile processor
**
** Copyright (C) 2018 Elphel, Inc.
**
** -----------------------------------------------------------------------------**
**
** dtt8x8.cu is free software: you can redistribute it and/or modify
** it under the terms of the GNU General Public License as published by
** the Free Software Foundation, either version 3 of the License, or
** (at your option) any later version.
**
** This program is distributed in the hope that it will be useful,
** but WITHOUT ANY WARRANTY; without even the implied warranty of
** MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
** GNU General Public License for more details.
**
** You should have received a copy of the GNU General Public License
** along with this program. If not, see <http://www.gnu.org/licenses/>.
**
** Additional permission under GNU GPL version 3 section 7
**
** If you modify this Program, or any covered work, by linking or
** combining it with NVIDIA Corporation's CUDA libraries from the
** NVIDIA CUDA Toolkit (or a modified version of those libraries),
** containing parts covered by the terms of NVIDIA CUDA Toolkit
** EULA, the licensors of this Program grant you additional
** permission to convey the resulting work.
** -----------------------------------------------------------------------------**
*/
#include <stdio.h>
#include <stdlib.h>
#include <math.h>
#include <cuda_runtime.h>
#include <helper_cuda.h>
#include <helper_functions.h>
// for reading binary files
#include <fstream>
#include <iterator>
#include <vector>
//#include "dtt8x8.cuh"
#include "dtt8x8.h"
#include "TileProcessor.cuh"
///#include "cuda_profiler_api.h"
//#include "cudaProfiler.h"
float * copyalloc_kernel_gpu(float * kernel_host,
int size, // size in floats
int full_size)
{
float *kernel_gpu;
checkCudaErrors(cudaMalloc((void **)&kernel_gpu, full_size * sizeof(float)));
checkCudaErrors(cudaMemcpy( // segfault
kernel_gpu,
kernel_host,
size * sizeof(float),
cudaMemcpyHostToDevice));
return kernel_gpu;
}
float * copyalloc_kernel_gpu(float * kernel_host,
int size)
{
return copyalloc_kernel_gpu(kernel_host,
size, // size in floats
size);
}
float * alloccopy_from_gpu(
float * gpu_data,
float * cpu_data, // if null, will allocate
int size)
{
if (!cpu_data) {
cpu_data = (float *)malloc(size*sizeof(float));
}
checkCudaErrors(cudaMemcpy( // segfault
cpu_data,
gpu_data,
size * sizeof(float),
cudaMemcpyDeviceToHost));
return cpu_data;
}
float * alloc_kernel_gpu(int size) // size in floats
{
float *kernel_gpu;
checkCudaErrors(cudaMalloc((void **)&kernel_gpu, size * sizeof(float)));
return kernel_gpu;
}
float ** copyalloc_pointers_gpu(float ** gpu_pointer,
int size) // number of entries (cameras)
{
float ** gpu_pointer_to_gpu_pointers;
checkCudaErrors(cudaMalloc((void **)&gpu_pointer_to_gpu_pointers, size * sizeof(float*)));
checkCudaErrors(cudaMemcpy(
gpu_pointer_to_gpu_pointers,
gpu_pointer,
size * sizeof(float*),
cudaMemcpyHostToDevice));
return gpu_pointer_to_gpu_pointers;
}
float * copyalloc_image_gpu(float * image_host,
size_t* dstride, // in floats !
int width,
int height)
{
float *image_gpu;
checkCudaErrors(cudaMallocPitch((void **)&image_gpu, dstride, width * sizeof(float), height));
checkCudaErrors(cudaMemcpy2D(
image_gpu,
*dstride, // * sizeof(float),
image_host,
width * sizeof(float), // make in 16*n?
width * sizeof(float),
height,
cudaMemcpyHostToDevice));
return image_gpu;
}
float * alloc_image_gpu(size_t* dstride, // in bytes!!
int width,
int height)
{
float *image_gpu;
checkCudaErrors(cudaMallocPitch((void **)&image_gpu, dstride, width * sizeof(float), height));
return image_gpu;
}
int readFloatsFromFile(float * data, // allocated array
const char * path) // file path
{
std::ifstream input(path, std::ios::binary );
// copies all data into buffer
std::vector<char> buffer((
std::istreambuf_iterator<char>(input)),
(std::istreambuf_iterator<char>()));
std::copy( buffer.begin(), buffer.end(), (char *) data);
return 0;
}
int writeFloatsToFile(float * data, // allocated array
int size, // length in elements
const char * path) // file path
{
// std::ifstream input(path, std::ios::binary );
std::ofstream ofile(path, std::ios::binary);
ofile.write((char *) data, size * sizeof(float));
return 0;
}
// Prepare low pass filter (64 long) to be applied to each quadrant of the CLT data
void set_clt_lpf(
float * lpf, // size*size array to be filled out
float sigma,
const int dct_size)
{
int dct_len = dct_size * dct_size;
if (sigma == 0.0f) {
lpf[0] = 1.0f;
for (int i = 1; i < dct_len; i++){
lpf[i] = 0.0;
}
} else {
for (int i = 0; i < dct_size; i++){
for (int j = 0; j < dct_size; j++){
lpf[i*dct_size+j] = exp(-(i*i+j*j)/(2*sigma));
}
}
// normalize
double sum = 0;
for (int i = 0; i < dct_size; i++){
for (int j = 0; j < dct_size; j++){
double d = lpf[i*dct_size+j];
d*=cos(M_PI*i/(2*dct_size))*cos(M_PI*j/(2*dct_size));
if (i > 0) d*= 2.0;
if (j > 0) d*= 2.0;
sum +=d;
}
}
for (int i = 0; i< dct_len; i++){
lpf[i] /= sum;
}
}
}
/**
**************************************************************************
* Program entry point
*
* \param argc [IN] - Number of command-line arguments
* \param argv [IN] - Array of command-line arguments
*
* \return Status code
*/
int main(int argc, char **argv)
{
//
// Sample initialization
//
printf("%s Starting...\n\n", argv[0]);
printf("sizeof(float*)=%d\n",(int)sizeof(float*));
//initialize CUDA
findCudaDevice(argc, (const char **)argv);
// CLT testing
const char* kernel_file[] = {
"/data_ssd/git/tile_processor_gpu/clt/main_chn0_transposed.kernel",
"/data_ssd/git/tile_processor_gpu/clt/main_chn1_transposed.kernel",
"/data_ssd/git/tile_processor_gpu/clt/main_chn2_transposed.kernel",
"/data_ssd/git/tile_processor_gpu/clt/main_chn3_transposed.kernel"};
const char* kernel_offs_file[] = {
"/data_ssd/git/tile_processor_gpu/clt/main_chn0_transposed.kernel_offsets",
"/data_ssd/git/tile_processor_gpu/clt/main_chn1_transposed.kernel_offsets",
"/data_ssd/git/tile_processor_gpu/clt/main_chn2_transposed.kernel_offsets",
"/data_ssd/git/tile_processor_gpu/clt/main_chn3_transposed.kernel_offsets"};
const char* image_files[] = {
"/data_ssd/git/tile_processor_gpu/clt/main_chn0.bayer",
"/data_ssd/git/tile_processor_gpu/clt/main_chn1.bayer",
"/data_ssd/git/tile_processor_gpu/clt/main_chn2.bayer",
"/data_ssd/git/tile_processor_gpu/clt/main_chn3.bayer"};
const char* ports_offs_xy_file[] = {
"/data_ssd/git/tile_processor_gpu/clt/main_chn0.portsxy",
"/data_ssd/git/tile_processor_gpu/clt/main_chn1.portsxy",
"/data_ssd/git/tile_processor_gpu/clt/main_chn2.portsxy",
"/data_ssd/git/tile_processor_gpu/clt/main_chn3.portsxy"};
const char* ports_clt_file[] = { // never referenced
"/data_ssd/git/tile_processor_gpu/clt/main_chn0.clt",
"/data_ssd/git/tile_processor_gpu/clt/main_chn1.clt",
"/data_ssd/git/tile_processor_gpu/clt/main_chn2.clt",
"/data_ssd/git/tile_processor_gpu/clt/main_chn3.clt"};
const char* result_rbg_file[] = {
"/data_ssd/git/tile_processor_gpu/clt/main_chn0.rbg",
"/data_ssd/git/tile_processor_gpu/clt/main_chn1.rbg",
"/data_ssd/git/tile_processor_gpu/clt/main_chn2.rbg",
"/data_ssd/git/tile_processor_gpu/clt/main_chn3.rbg"};
const char* result_corr_file = "/data_ssd/git/tile_processor_gpu/clt/main_corr.corr";
const char* result_textures_file = "/data_ssd/git/tile_processor_gpu/clt/texture.rgba";
const char* result_textures_rgba_file = "/data_ssd/git/tile_processor_gpu/clt/texture_rgba.rgba";
// not yet used
float lpf_sigmas[3] = {0.9f, 0.9f, 0.9f}; // G, B, G
float port_offsets[NUM_CAMS][2] = {// used only in textures to scale differences
{-0.5, -0.5},
{ 0.5, -0.5},
{-0.5, 0.5},
{ 0.5, 0.5}};
int keep_texture_weights = 1; // try with 0 also
int texture_colors = 3; // result will be 3+1 RGBA (for mono - 2)
/*
#define IMG_WIDTH 2592
#define IMG_HEIGHT 1936
#define NUM_CAMS 4
#define NUM_COLORS 3
#define KERNELS_STEP 16
#define KERNELS_HOR 164
#define KERNELS_VERT 123
#define KERNEL_OFFSETS 8
#define TILESX 324
#define TILESY 242
*/
/*
struct tp_task {
long task;
short ty;
short tx;
float xy[NUM_CAMS][2];
} ;
*/
int KERN_TILES = KERNELS_HOR * KERNELS_VERT * NUM_COLORS;
int KERN_SIZE = KERN_TILES * 4 * 64;
// int CORR_SIZE = (2 * DTT_SIZE -1) * (2 * DTT_SIZE -1);
int CORR_SIZE = (2 * CORR_OUT_RAD + 1) * (2 * CORR_OUT_RAD + 1);
float * host_kern_buf = (float *)malloc(KERN_SIZE * sizeof(float));
struct tp_task task_data [TILESX*TILESY]; // maximal length - each tile
int corr_indices [NUM_PAIRS*TILESX*TILESY];
// int texture_indices [TILESX*TILESY];
int texture_indices [TILESX*TILESYA];
int cpu_woi [4];
// host array of pointers to GPU memory
float * gpu_kernels_h [NUM_CAMS];
struct CltExtra * gpu_kernel_offsets_h [NUM_CAMS];
float * gpu_images_h [NUM_CAMS];
float tile_coords_h [NUM_CAMS][TILESX * TILESY][2];
float * gpu_clt_h [NUM_CAMS];
float * gpu_lpf_h [NUM_COLORS]; // never used
#ifndef NOICLT
float * gpu_corr_images_h [NUM_CAMS];
#endif
float * gpu_corrs;
int * gpu_corr_indices;
float * gpu_textures;
float * gpu_textures_rbga;
int * gpu_texture_indices;
int * gpu_woi;
int * gpu_num_texture_tiles;
float * gpu_port_offsets;
int num_corrs;
int num_textures;
int num_ports = NUM_CAMS;
// GPU pointers to GPU pointers to memory
float ** gpu_kernels; // [NUM_CAMS];
struct CltExtra ** gpu_kernel_offsets; // [NUM_CAMS];
float ** gpu_images; // [NUM_CAMS];
float ** gpu_clt; // [NUM_CAMS];
float ** gpu_lpf; // [NUM_CAMS]; // never referenced
// GPU pointers to GPU memory
// float * gpu_tasks;
struct tp_task * gpu_tasks;
size_t dstride; // in bytes !
size_t dstride_rslt; // in bytes !
size_t dstride_corr; // in bytes ! for one 2d phase correlation (padded 15x15x4 bytes)
size_t dstride_textures; // in bytes ! for one rgba/ya 16x16 tile
size_t dstride_textures_rbga; // in bytes ! for one rgba/ya 16x16 tile
float lpf_rbg[3][64]; // not used
for (int ncol = 0; ncol < 3; ncol++) {
if (lpf_sigmas[ncol] > 0.0) {
set_clt_lpf (
lpf_rbg[ncol], // float * lpf, // size*size array to be filled out
lpf_sigmas[ncol], // float sigma,
8); // int dct_size)
gpu_lpf_h[ncol] = copyalloc_kernel_gpu(lpf_rbg[ncol], 64);
} else {
gpu_lpf_h[ncol] = NULL;
}
}
for (int ncam = 0; ncam < NUM_CAMS; ncam++) {
readFloatsFromFile(
host_kern_buf, // float * data, // allocated array
kernel_file[ncam]); // char * path) // file path
gpu_kernels_h[ncam] = copyalloc_kernel_gpu(host_kern_buf, KERN_SIZE);
readFloatsFromFile(
host_kern_buf, // float * data, // allocated array
kernel_offs_file[ncam]); // char * path) // file path
gpu_kernel_offsets_h[ncam] = (struct CltExtra *) copyalloc_kernel_gpu(
host_kern_buf,
KERN_TILES * (sizeof( struct CltExtra)/sizeof(float)));
// will get results back
gpu_clt_h[ncam] = alloc_kernel_gpu(TILESY * TILESX * NUM_COLORS * 4 * DTT_SIZE * DTT_SIZE);
printf("Allocating GPU memory, 0x%x floats\n", (TILESY * TILESX * NUM_COLORS * 4 * DTT_SIZE * DTT_SIZE)) ;
// allocate result images (3x height to accommodate 3 colors
// Image is extended by 4 pixels each side to avoid checking (mclt tiles extend by 4)
//host array of pointers to GPU arrays
#ifndef NOICLT
gpu_corr_images_h[ncam] = alloc_image_gpu(
&dstride_rslt, // size_t* dstride, // in bytes!!
IMG_WIDTH + DTT_SIZE, // int width,
3*(IMG_HEIGHT + DTT_SIZE)); // int height);
#endif
}
// allocates one correlation kernel per line (15x15 floats), number of rows - number of tiles * number of pairs
gpu_corrs = alloc_image_gpu(
&dstride_corr, // in bytes ! for one 2d phase correlation (padded 15x15x4 bytes)
CORR_SIZE, // int width,
NUM_PAIRS * TILESX * TILESY); // int height);
// read channel images (assuming host_kern_buf size > image size, reusing it)
for (int ncam = 0; ncam < NUM_CAMS; ncam++) {
readFloatsFromFile(
host_kern_buf, // float * data, // allocated array
image_files[ncam]); // char * path) // file path
gpu_images_h[ncam] = copyalloc_image_gpu(
host_kern_buf, // float * image_host,
&dstride, // size_t* dstride,
IMG_WIDTH, // int width,
IMG_HEIGHT); // int height);
}
//#define DBG_TILE (174*324 +118)
for (int ncam = 0; ncam < NUM_CAMS; ncam++) {
readFloatsFromFile(
(float *) &tile_coords_h[ncam],
ports_offs_xy_file[ncam]); // char * path) // file path
}
// build TP task that processes all tiles in linescan order
for (int ty = 0; ty < TILESY; ty++){
for (int tx = 0; tx < TILESX; tx++){
int nt = ty * TILESX + tx;
task_data[nt].task = 0xf | (((1 << NUM_PAIRS)-1) << TASK_CORR_BITS);
task_data[nt].txy = tx + (ty << 16);
for (int ncam = 0; ncam < NUM_CAMS; ncam++) {
task_data[nt].xy[ncam][0] = tile_coords_h[ncam][nt][0];
task_data[nt].xy[ncam][1] = tile_coords_h[ncam][nt][1];
}
}
}
int tp_task_size = sizeof(task_data)/sizeof(struct tp_task);
#ifdef DBG0
//#define NUM_TEST_TILES 128
#define NUM_TEST_TILES 1
for (int t = 0; t < NUM_TEST_TILES; t++) {
task_data[t].task = 1;
task_data[t].txy = ((DBG_TILE + t) - 324* ((DBG_TILE + t) / 324)) + (((DBG_TILE + t) / 324)) << 16;
int nt = task_data[t].ty * TILESX + task_data[t].tx;
for (int ncam = 0; ncam < NUM_CAMS; ncam++) {
task_data[t].xy[ncam][0] = tile_coords_h[ncam][nt][0];
task_data[t].xy[ncam][1] = tile_coords_h[ncam][nt][1];
}
}
tp_task_size = NUM_TEST_TILES; // sizeof(task_data)/sizeof(float);
#endif
// segfault in the next
gpu_tasks = (struct tp_task *) copyalloc_kernel_gpu((float * ) &task_data, tp_task_size * (sizeof(struct tp_task)/sizeof(float)));
// build corr_indices
num_corrs = 0;
for (int ty = 0; ty < TILESY; ty++){
for (int tx = 0; tx < TILESX; tx++){
int nt = ty * TILESX + tx;
int cm = (task_data[nt].task >> TASK_CORR_BITS) & ((1 << NUM_PAIRS)-1);
if (cm){
for (int b = 0; b < NUM_PAIRS; b++) if ((cm & (1 << b)) != 0) {
corr_indices[num_corrs++] = (nt << CORR_NTILE_SHIFT) | b;
}
}
}
}
// num_corrs now has the total number of correlations
// copy corr_indices to gpu
// gpu_corr_indices = (int *) copyalloc_kernel_gpu((float * ) corr_indices, num_corrs);
gpu_corr_indices = (int *) copyalloc_kernel_gpu(
(float * ) corr_indices,
num_corrs,
NUM_PAIRS * TILESX * TILESY);
// build texture_indices
num_textures = 0;
for (int ty = 0; ty < TILESY; ty++){
for (int tx = 0; tx < TILESX; tx++){
int nt = ty * TILESX + tx;
// int cm = (task_data[nt].task >> TASK_TEXTURE_BIT) & 1;
int cm = task_data[nt].task & TASK_TEXTURE_BITS;
if (cm){
texture_indices[num_textures++] = (nt << CORR_NTILE_SHIFT) | (1 << LIST_TEXTURE_BIT);
}
}
}
// num_textures now has the total number of textures
// copy corr_indices to gpu
// gpu_texture_indices = (int *) copyalloc_kernel_gpu((float * ) texture_indices, num_textures);
gpu_texture_indices = (int *) copyalloc_kernel_gpu(
(float * ) texture_indices,
num_textures,
TILESX * TILESYA); // number of rows - multiple of 4
// just allocate
checkCudaErrors(cudaMalloc((void **)&gpu_woi, 4 * sizeof(float)));
checkCudaErrors(cudaMalloc((void **)&gpu_num_texture_tiles, 8 * sizeof(float))); // for each subsequence - number of non-border,
// number of border tiles
// copy port indices to gpu
gpu_port_offsets = (float *) copyalloc_kernel_gpu((float * ) port_offsets, num_ports * 2);
// int keep_texture_weights = 1; // try with 0 also
// int texture_colors = 3; // result will be 3+1 RGBA (for mono - 2)
// double [][] rgba = new double[numcol + 1 + (keep_weights?(ports + numcol + 1):0)][];
int tile_texture_size = (texture_colors + 1 + (keep_texture_weights? (NUM_CAMS + texture_colors + 1): 0)) *256;
gpu_textures = alloc_image_gpu(
&dstride_textures, // in bytes ! for one rgba/ya 16x16 tile
tile_texture_size, // int width (floats),
TILESX * TILESY); // int height);
int rgba_width = (TILESX+1) * DTT_SIZE;
int rgba_height = (TILESY+1) * DTT_SIZE;
int rbga_slices = texture_colors + 1; // 4/1
gpu_textures_rbga = alloc_image_gpu(
&dstride_textures_rbga, // in bytes ! for one rgba/ya 16x16 tile
rgba_width, // int width (floats),
rgba_height * rbga_slices); // int height);
// Now copy arrays of per-camera pointers to GPU memory to GPU itself
gpu_kernels = copyalloc_pointers_gpu (gpu_kernels_h, NUM_CAMS);
gpu_kernel_offsets = (struct CltExtra **) copyalloc_pointers_gpu ((float **) gpu_kernel_offsets_h, NUM_CAMS);
gpu_images = copyalloc_pointers_gpu (gpu_images_h, NUM_CAMS);
gpu_clt = copyalloc_pointers_gpu (gpu_clt_h, NUM_CAMS);
// gpu_corr_images = copyalloc_pointers_gpu (gpu_corr_images_h, NUM_CAMS);
//create and start CUDA timer
StopWatchInterface *timerTP = 0;
sdkCreateTimer(&timerTP);
dim3 threads_tp(THREADSX, TILES_PER_BLOCK, 1);
dim3 grid_tp((tp_task_size + TILES_PER_BLOCK -1 )/TILES_PER_BLOCK, 1);
printf("threads_tp=(%d, %d, %d)\n",threads_tp.x,threads_tp.y,threads_tp.z);
printf("grid_tp= (%d, %d, %d)\n",grid_tp.x, grid_tp.y, grid_tp.z);
#ifdef DBG_TILE
const int numIterations = 1; //0;
const int i0 = 0; // -1;
#else
const int numIterations = 10; // 0; //0;
const int i0 = -1; // 0; // -1;
#endif
cudaFuncSetCacheConfig(convert_correct_tiles, cudaFuncCachePreferShared);
/// cudaProfilerStart();
float ** fgpu_kernel_offsets = (float **) gpu_kernel_offsets; // [NUM_CAMS];
for (int i = i0; i < numIterations; i++)
{
if (i == 0)
{
checkCudaErrors(cudaDeviceSynchronize());
sdkResetTimer(&timerTP);
sdkStartTimer(&timerTP);
}
convert_correct_tiles<<<grid_tp,threads_tp>>>(
fgpu_kernel_offsets, // struct CltExtra ** gpu_kernel_offsets,
gpu_kernels, // float ** gpu_kernels,
gpu_images, // float ** gpu_images,
gpu_tasks, // struct tp_task * gpu_tasks,
gpu_clt, // float ** gpu_clt, // [NUM_CAMS][TILESY][TILESX][NUM_COLORS][DTT_SIZE*DTT_SIZE]
dstride/sizeof(float), // size_t dstride, // for gpu_images
tp_task_size, // int num_tiles) // number of tiles in task
0); // 7); // 0); // 7); // int lpf_mask) // apply lpf to colors : bit 0 - red, bit 1 - blue, bit2 - green
getLastCudaError("Kernel execution failed");
checkCudaErrors(cudaDeviceSynchronize());
printf("%d\n",i);
}
// checkCudaErrors(cudaDeviceSynchronize());
sdkStopTimer(&timerTP);
float avgTime = (float)sdkGetTimerValue(&timerTP) / (float)numIterations;
sdkDeleteTimer(&timerTP);
printf("Run time =%f ms\n", avgTime);
#ifdef SAVE_CLT
int rslt_size = (TILESY * TILESX * NUM_COLORS * 4 * DTT_SIZE * DTT_SIZE);
float * cpu_clt = (float *)malloc(rslt_size*sizeof(float));
for (int ncam = 0; ncam < NUM_CAMS; ncam++) {
checkCudaErrors(cudaMemcpy( // segfault
cpu_clt,
gpu_clt_h[ncam],
rslt_size * sizeof(float),
cudaMemcpyDeviceToHost));
#ifndef DBG_TILE
printf("Writing CLT data to %s\n", ports_clt_file[ncam]);
writeFloatsToFile(cpu_clt, // float * data, // allocated array
rslt_size, // int size, // length in elements
ports_clt_file[ncam]); // const char * path) // file path
#endif
}
#endif
#ifdef TEST_IMCLT
{
// testing imclt
dim3 threads_imclt(IMCLT_THREADS_PER_TILE, IMCLT_TILES_PER_BLOCK, 1);
dim3 grid_imclt(1,1,1);
printf("threads_imclt=(%d, %d, %d)\n",threads_imclt.x,threads_imclt.y,threads_imclt.z);
printf("grid_imclt= (%d, %d, %d)\n",grid_imclt.x, grid_imclt.y, grid_imclt.z);
for (int ncam = 0; ncam < NUM_CAMS; ncam++) {
test_imclt<<<grid_imclt,threads_imclt>>>(
gpu_clt_h[ncam], // ncam]); // // float ** gpu_clt, // [NUM_CAMS][TILESY][TILESX][NUM_COLORS][DTT_SIZE*DTT_SIZE]
ncam); // int ncam); // just for debug print
}
getLastCudaError("Kernel execution failed");
checkCudaErrors(cudaDeviceSynchronize());
printf("test_imclt() DONE\n");
}
#endif
#ifndef NOICLT
// testing imclt
dim3 threads_imclt(IMCLT_THREADS_PER_TILE, IMCLT_TILES_PER_BLOCK, 1);
printf("threads_imclt=(%d, %d, %d)\n",threads_imclt.x,threads_imclt.y,threads_imclt.z);
StopWatchInterface *timerIMCLT = 0;
sdkCreateTimer(&timerIMCLT);
for (int i = i0; i < numIterations; i++)
{
if (i == 0)
{
checkCudaErrors(cudaDeviceSynchronize());
sdkResetTimer(&timerIMCLT);
sdkStartTimer(&timerIMCLT);
}
for (int ncam = 0; ncam < NUM_CAMS; ncam++) {
for (int color = 0; color < NUM_COLORS; color++) {
#ifdef IMCLT14
for (int v_offs = 0; v_offs < 1; v_offs++){ // temporarily for debugging
for (int h_offs = 0; h_offs < 1; h_offs++){ // temporarily for debugging
#else
for (int v_offs = 0; v_offs < 2; v_offs++){
for (int h_offs = 0; h_offs < 2; h_offs++){
#endif
int tilesy_half = (TILESY + (v_offs ^ 1)) >> 1;
int tilesx_half = (TILESX + (h_offs ^ 1)) >> 1;
int tiles_in_pass = tilesy_half * tilesx_half;
dim3 grid_imclt((tiles_in_pass + IMCLT_TILES_PER_BLOCK-1) / IMCLT_TILES_PER_BLOCK,1,1);
// printf("grid_imclt= (%d, %d, %d)\n",grid_imclt.x, grid_imclt.y, grid_imclt.z);
imclt_rbg<<<grid_imclt,threads_imclt>>>(
gpu_clt_h[ncam], // float * gpu_clt, // [TILESY][TILESX][NUM_COLORS][DTT_SIZE*DTT_SIZE]
gpu_corr_images_h[ncam], // float * gpu_rbg, // WIDTH, 3 * HEIGHT
1, // int apply_lpf,
0, // int mono, // defines lpf filter
color, // int color, // defines location of clt data
v_offs, // int v_offset,
h_offs, // int h_offset,
dstride_rslt/sizeof(float)); //const size_t dstride); // in floats (pixels)
}
}
}
}
getLastCudaError("Kernel failure");
checkCudaErrors(cudaDeviceSynchronize());
printf("test pass: %d\n",i);
}
sdkStopTimer(&timerIMCLT);
float avgTimeIMCLT = (float)sdkGetTimerValue(&timerIMCLT) / (float)numIterations;
sdkDeleteTimer(&timerIMCLT);
printf("Average IMCLT run time =%f ms\n", avgTimeIMCLT);
int rslt_img_size = NUM_COLORS * (IMG_HEIGHT + DTT_SIZE) * (IMG_WIDTH + DTT_SIZE);
float * cpu_corr_image = (float *)malloc(rslt_img_size * sizeof(float));
for (int ncam = 0; ncam < NUM_CAMS; ncam++) {
checkCudaErrors(cudaMemcpy2D( // segfault
cpu_corr_image,
(IMG_WIDTH + DTT_SIZE) * sizeof(float),
gpu_corr_images_h[ncam],
dstride_rslt,
(IMG_WIDTH + DTT_SIZE) * sizeof(float),
3* (IMG_HEIGHT + DTT_SIZE),
cudaMemcpyDeviceToHost));
#ifndef DBG_TILE
printf("Writing RBG data to %s\n", result_rbg_file[ncam]);
writeFloatsToFile( // will have margins
cpu_corr_image, // float * data, // allocated array
rslt_img_size, // int size, // length in elements
result_rbg_file[ncam]); // const char * path) // file path
#endif
}
free(cpu_corr_image);
#endif
#ifndef NOCORR
// cudaProfilerStart();
// testing corr
dim3 threads_corr(CORR_THREADS_PER_TILE, CORR_TILES_PER_BLOCK, 1);
printf("threads_corr=(%d, %d, %d)\n",threads_corr.x,threads_corr.y,threads_corr.z);
StopWatchInterface *timerCORR = 0;
sdkCreateTimer(&timerCORR);
for (int i = i0; i < numIterations; i++)
{
if (i == 0)
{
checkCudaErrors(cudaDeviceSynchronize());
sdkResetTimer(&timerCORR);
sdkStartTimer(&timerCORR);
}
dim3 grid_corr((num_corrs + CORR_TILES_PER_BLOCK-1) / CORR_TILES_PER_BLOCK,1,1);
correlate2D<<<grid_corr,threads_corr>>>(
gpu_clt, // float ** gpu_clt, // [NUM_CAMS] ->[TILESY][TILESX][NUM_COLORS][DTT_SIZE*DTT_SIZE]
3, // int colors, // number of colors (3/1)
0.25, // float scale0, // scale for R
0.25, // float scale1, // scale for B
0.5, // float scale2, // scale for G
30.0, // float fat_zero, // here - absolute
num_corrs, // size_t num_corr_tiles, // number of correlation tiles to process
gpu_corr_indices, // int * gpu_corr_indices, // packed tile+pair
dstride_corr/sizeof(float), // const size_t corr_stride, // in floats
CORR_OUT_RAD, // int corr_radius, // radius of the output correlation (7 for 15x15)
gpu_corrs); // float * gpu_corrs); // correlation output data
getLastCudaError("Kernel failure");
checkCudaErrors(cudaDeviceSynchronize());
printf("test pass: %d\n",i);
}
sdkStopTimer(&timerCORR);
float avgTimeCORR = (float)sdkGetTimerValue(&timerCORR) / (float)numIterations;
sdkDeleteTimer(&timerCORR);
printf("Average CORR run time =%f ms\n", avgTimeCORR);
int corr_size = 2 * CORR_OUT_RAD + 1;
int rslt_corr_size = num_corrs * corr_size * corr_size;
float * cpu_corr = (float *)malloc(rslt_corr_size * sizeof(float));
checkCudaErrors(cudaMemcpy2D(
cpu_corr,
(corr_size * corr_size) * sizeof(float),
gpu_corrs,
dstride_corr,
(corr_size * corr_size) * sizeof(float),
num_corrs,
cudaMemcpyDeviceToHost));
#ifndef NSAVE_CORR
printf("Writing phase correlation data to %s\n", result_corr_file);
writeFloatsToFile(
cpu_corr, // float * data, // allocated array
rslt_corr_size, // int size, // length in elements
result_corr_file); // const char * path) // file path
#endif
free(cpu_corr);
#endif // ifndef NOCORR
// -----------------
#ifndef NOTEXTURES
// cudaProfilerStart();
// testing textures
dim3 threads_texture(TEXTURE_THREADS_PER_TILE, NUM_CAMS, 1); // TEXTURE_TILES_PER_BLOCK, 1);
dim3 grid_texture((num_textures + TEXTURE_TILES_PER_BLOCK-1) / TEXTURE_TILES_PER_BLOCK,1,1);
printf("threads_texture=(%d, %d, %d)\n",threads_texture.x,threads_texture.y,threads_texture.z);
printf("grid_texture=(%d, %d, %d)\n",grid_texture.x,grid_texture.y,grid_texture.z);
StopWatchInterface *timerTEXTURE = 0;
sdkCreateTimer(&timerTEXTURE);
for (int i = i0; i < numIterations; i++)
{
if (i == 0)
{
checkCudaErrors(cudaDeviceSynchronize());
sdkResetTimer(&timerTEXTURE);
sdkStartTimer(&timerTEXTURE);
}
// Channel0 weight = 0.294118
// Channel1 weight = 0.117647
// Channel2 weight = 0.588235
textures_accumulate<<<grid_texture,threads_texture>>> (
// 0, // int border_tile, // if 1 - watch for border
(int *) 0, // int * woi, // x, y, width,height
gpu_clt , // float ** gpu_clt, // [NUM_CAMS] ->[TILESY][TILESX][NUM_COLORS][DTT_SIZE*DTT_SIZE]
num_textures, // size_t num_texture_tiles, // number of texture tiles to process
gpu_texture_indices, // int * gpu_texture_indices,// packed tile + bits (now only (1 << 7)
gpu_port_offsets, // float * port_offsets, // relative ports x,y offsets - just to scale differences, may be approximate
texture_colors, // int colors, // number of colors (3/1)
(texture_colors == 1), // int is_lwir, // do not perform shot correction
10.0, // float min_shot, // 10.0
3.0, // float scale_shot, // 3.0
1.5f, // float diff_sigma, // pixel value/pixel change
10.0f, // float diff_threshold, // pixel value/pixel change
3.0, // float min_agree, // minimal number of channels to agree on a point (real number to work with fuzzy averages)
0.294118, // float weight0, // scale for R
0.117647, // float weight1, // scale for B
0.588235, // float weight2, // scale for G
1, // int dust_remove, // Do not reduce average weight when only one image differes much from the average
keep_texture_weights, // int keep_weights, // return channel weights after A in RGBA
// combining both non-overlap and overlap (each calculated if pointer is not null )
0, // const size_t texture_rbg_stride, // in floats
(float *) 0, // float * gpu_texture_rbg, // (number of colors +1 + ?)*16*16 rgba texture tiles
dstride_textures/sizeof(float), // const size_t texture_stride, // in floats (now 256*4 = 1024)
gpu_textures); // float * gpu_texture_tiles); // 4*16*16 rgba texture tiles
getLastCudaError("Kernel failure");
checkCudaErrors(cudaDeviceSynchronize());
printf("test pass: %d\n",i);
}
/// cudaProfilerStop();
sdkStopTimer(&timerTEXTURE);
float avgTimeTEXTURES = (float)sdkGetTimerValue(&timerTEXTURE) / (float)numIterations;
sdkDeleteTimer(&timerTEXTURE);
printf("Average Texture run time =%f ms\n", avgTimeTEXTURES);
int rslt_texture_size = num_textures * tile_texture_size;
float * cpu_textures = (float *)malloc(rslt_texture_size * sizeof(float));
checkCudaErrors(cudaMemcpy2D(
cpu_textures,
tile_texture_size * sizeof(float),
gpu_textures,
dstride_textures,
tile_texture_size * sizeof(float),
num_textures,
cudaMemcpyDeviceToHost));
#ifndef NSAVE_TEXTURES
printf("Writing phase texture data to %s\n", result_textures_file);
writeFloatsToFile(
cpu_textures, // float * data, // allocated array
rslt_texture_size, // int size, // length in elements
result_textures_file); // const char * path) // file path
//DBG_TILE
#ifdef DEBUG10
int texture_offset = DBG_TILE * tile_texture_size;
int chn = 0;
for (int i = 0; i < tile_texture_size; i++){
if ((i % 256) == 0){
printf("\nchn = %d\n", chn++);
}
printf("%10.4f", *(cpu_textures + texture_offset + i));
if (((i + 1) % 16) == 0){
printf("\n");
} else {
printf(" ");
}
}
// int tile_texture_size = (texture_colors + 1 + (keep_texture_weights? (NUM_CAMS + texture_colors + 1): 0)) *256;
#endif // DEBUG9
#endif
free(cpu_textures);
#endif // ifndef NOTEXTURES
#define GEN_TEXTURE_LIST
#ifdef GEN_TEXTURE_LIST
dim3 threads_list(1,1, 1); // TEXTURE_TILES_PER_BLOCK, 1);
dim3 grid_list (1,1,1);
printf("threads_list=(%d, %d, %d)\n",threads_list.x,threads_list.y,threads_list.z);
printf("grid_list=(%d, %d, %d)\n",grid_list.x,grid_list.y,grid_list.z);
StopWatchInterface *timerTEXTURELIST = 0;
sdkCreateTimer(&timerTEXTURELIST);
for (int i = i0; i < numIterations; i++)
{
if (i == 0)
{
checkCudaErrors(cudaDeviceSynchronize());
sdkResetTimer(&timerTEXTURELIST);
sdkStartTimer(&timerTEXTURELIST);
}
prepare_texture_list<<<grid_list,threads_list>>> (
gpu_tasks, // struct tp_task * gpu_tasks,
tp_task_size, // int num_tiles, // number of tiles in task list
gpu_texture_indices, // int * gpu_texture_indices,// packed tile + bits (now only (1 << 7)
gpu_num_texture_tiles, // int * num_texture_tiles, // number of texture tiles to process (8 elements)
gpu_woi, // int * woi, // x,y,width,height of the woi
TILESX, // int width, // <= TILESX, use for faster processing of LWIR images (should be actual + 1)
TILESY); // int height); // <= TILESY, use for faster processing of LWIR images
getLastCudaError("Kernel failure");
checkCudaErrors(cudaDeviceSynchronize());
printf("test pass: %d\n",i);
}
/// cudaProfilerStop();
sdkStopTimer(&timerTEXTURELIST);
float avgTimeTEXTURESLIST = (float)sdkGetTimerValue(&timerTEXTURELIST) / (float)numIterations;
sdkDeleteTimer(&timerTEXTURELIST);
printf("Average TextureList run time =%f ms\n", avgTimeTEXTURESLIST);
int cpu_num_texture_tiles[8];
checkCudaErrors(cudaMemcpy(
cpu_woi,
gpu_woi,
4 * sizeof(float),
cudaMemcpyDeviceToHost));
printf("WOI x=%d, y=%d, width=%d, height=%d\n", cpu_woi[0], cpu_woi[1], cpu_woi[2], cpu_woi[3]);
checkCudaErrors(cudaMemcpy(
cpu_num_texture_tiles,
gpu_num_texture_tiles,
8 * sizeof(float), // 8 sequences (0,2,4,6 - non-border, growing up;
//1,3,5,7 - border, growing down from the end of the corresponding non-border buffers
cudaMemcpyDeviceToHost));
printf("cpu_num_texture_tiles=(%d(%d), %d(%d), %d(%d), %d(%d) -> %d tp_task_size=%d)\n",
cpu_num_texture_tiles[0], cpu_num_texture_tiles[1],
cpu_num_texture_tiles[2], cpu_num_texture_tiles[3],
cpu_num_texture_tiles[4], cpu_num_texture_tiles[5],
cpu_num_texture_tiles[6], cpu_num_texture_tiles[7],
cpu_num_texture_tiles[0] + cpu_num_texture_tiles[1] +
cpu_num_texture_tiles[2] + cpu_num_texture_tiles[3] +
cpu_num_texture_tiles[4] + cpu_num_texture_tiles[5] +
cpu_num_texture_tiles[6] + cpu_num_texture_tiles[7],
tp_task_size
);
for (int q = 0; q < 4; q++) {
checkCudaErrors(cudaMemcpy(
texture_indices + q * TILESX * (TILESYA >> 2),
gpu_texture_indices + q * TILESX * (TILESYA >> 2),
cpu_num_texture_tiles[q] * sizeof(float), // change to cpu_num_texture_tiles when ready
cudaMemcpyDeviceToHost));
}
for (int q = 0; q < 4; q++) {
printf("%d: %3x:%3x %3x:%3x %3x:%3x %3x:%3x %3x:%3x %3x:%3x %3x:%3x %3x:%3x \n",q,
(texture_indices[q * TILESX * (TILESYA >> 2) + 0] >> 8) / TILESX, (texture_indices[q * TILESX * (TILESYA >> 2) + 0] >> 8) % TILESX,
(texture_indices[q * TILESX * (TILESYA >> 2) + 1] >> 8) / TILESX, (texture_indices[q * TILESX * (TILESYA >> 2) + 1] >> 8) % TILESX,
(texture_indices[q * TILESX * (TILESYA >> 2) + 2] >> 8) / TILESX, (texture_indices[q * TILESX * (TILESYA >> 2) + 2] >> 8) % TILESX,
(texture_indices[q * TILESX * (TILESYA >> 2) + 3] >> 8) / TILESX, (texture_indices[q * TILESX * (TILESYA >> 2) + 3] >> 8) % TILESX,
(texture_indices[q * TILESX * (TILESYA >> 2) + 4] >> 8) / TILESX, (texture_indices[q * TILESX * (TILESYA >> 2) + 4] >> 8) % TILESX,
(texture_indices[q * TILESX * (TILESYA >> 2) + 5] >> 8) / TILESX, (texture_indices[q * TILESX * (TILESYA >> 2) + 5] >> 8) % TILESX,
(texture_indices[q * TILESX * (TILESYA >> 2) + 6] >> 8) / TILESX, (texture_indices[q * TILESX * (TILESYA >> 2) + 6] >> 8) % TILESX,
(texture_indices[q * TILESX * (TILESYA >> 2) + 7] >> 8) / TILESX, (texture_indices[q * TILESX * (TILESYA >> 2) + 7] >> 8) % TILESX);
}
#endif //GEN_TEXTURE_LIST
#ifndef NOTEXTURE_RGBA
dim3 threads_rgba(1, 1, 1);
dim3 grid_rgba(1,1,1);
printf("threads_rgba=(%d, %d, %d)\n", threads_rgba.x,threads_rgba.y,threads_rgba.z);
printf("grid_rgba=(%d, %d, %d)\n", grid_rgba.x,grid_rgba.y,grid_rgba.z);
StopWatchInterface *timerRGBA = 0;
sdkCreateTimer(&timerRGBA);
for (int i = i0; i < numIterations; i++)
{
if (i == 0)
{
checkCudaErrors(cudaDeviceSynchronize());
sdkResetTimer(&timerRGBA);
sdkStartTimer(&timerRGBA);
}
generate_RBGA<<<grid_rgba,threads_rgba>>> (
// Parameters to generate texture tasks
gpu_tasks, // struct tp_task * gpu_tasks,
tp_task_size, // int num_tiles, // number of tiles in task list
// declare arrays in device code?
gpu_texture_indices, // int * gpu_texture_indices,// packed tile + bits (now only (1 << 7)
gpu_num_texture_tiles, // int * num_texture_tiles, // number of texture tiles to process (8 elements)
gpu_woi, // int * woi, // x,y,width,height of the woi
TILESX, // int width, // <= TILESX, use for faster processing of LWIR images (should be actual + 1)
TILESY, // int height); // <= TILESY, use for faster processing of LWIR images
// Parameters for the texture generation
gpu_clt , // float ** gpu_clt, // [NUM_CAMS] ->[TILESY][TILESX][NUM_COLORS][DTT_SIZE*DTT_SIZE]
gpu_port_offsets, // float * port_offsets, // relative ports x,y offsets - just to scale differences, may be approximate
texture_colors, // int colors, // number of colors (3/1)
(texture_colors == 1), // int is_lwir, // do not perform shot correction
10.0, // float min_shot, // 10.0
3.0, // float scale_shot, // 3.0
1.5f, // float diff_sigma, // pixel value/pixel change
10.0f, // float diff_threshold, // pixel value/pixel change
3.0, // float min_agree, // minimal number of channels to agree on a point (real number to work with fuzzy averages)
0.294118, // float weight0, // scale for R
0.117647, // float weight1, // scale for B
0.588235, // float weight2, // scale for G
1, // int dust_remove, // Do not reduce average weight when only one image differes much from the average
0, // int keep_weights, // return channel weights after A in RGBA
dstride_textures_rbga/sizeof(float), // const size_t texture_rbga_stride, // in floats
gpu_textures_rbga); // float * gpu_texture_tiles) // (number of colors +1 + ?)*16*16 rgba texture tiles
getLastCudaError("Kernel failure");
checkCudaErrors(cudaDeviceSynchronize());
printf("test pass: %d\n",i);
}
sdkStopTimer(&timerRGBA);
float avgTimeRGBA = (float)sdkGetTimerValue(&timerRGBA) / (float)numIterations;
sdkDeleteTimer(&timerRGBA);
printf("Average Texture run time =%f ms\n", avgTimeRGBA);
checkCudaErrors(cudaMemcpy(
cpu_woi,
gpu_woi,
4 * sizeof(float),
cudaMemcpyDeviceToHost));
printf("WOI x=%d, y=%d, width=%d, height=%d\n", cpu_woi[0], cpu_woi[1], cpu_woi[2], cpu_woi[3]);
// temporarily use larger array (4 pixels each size, switch to cudaMemcpy2DFromArray()
int rgba_woi_width = (cpu_woi[2] + 1) * DTT_SIZE;
int rgba_woi_height = (cpu_woi[3] + 1)* DTT_SIZE;
int rslt_rgba_size = rgba_woi_width * rgba_woi_height * rbga_slices;
float * cpu_textures_rgba = (float *)malloc(rslt_rgba_size * sizeof(float));
checkCudaErrors(cudaMemcpy2D(
cpu_textures_rgba,
rgba_width * sizeof(float),
gpu_textures_rbga,
dstride_textures_rbga,
rgba_width * sizeof(float),
rgba_height * rbga_slices,
cudaMemcpyDeviceToHost));
#ifndef NSAVE_TEXTURES
printf("Writing RBGA texture slices to %s\n", result_textures_rgba_file);
writeFloatsToFile(
cpu_textures_rgba, // float * data, // allocated array
rslt_rgba_size, // int size, // length in elements
result_textures_rgba_file); // const char * path) // file path
#endif
#ifdef DEBUG11
int rgba_offset = (DBG_TILE_Y - cpu_woi[1]) * DTT_SIZE * rgba_woi_width + (DBG_TILE_X - cpu_woi[0]);
for (int chn = 0; chn < rbga_slices; chn++){
printf("\nchn = %d\n", chn);
int rgba_offset_chn = rgba_offset + chn * rgba_woi_width * rgba_woi_height;
for (int i = 0; i < 8; i++){
for (int j = 0; j < 8; j++){
printf("%10.4f ", *(cpu_textures_rgba + rgba_offset_chn + i * rgba_woi_width + j));
}
printf("\n");
}
}
#endif // DEBUG11
free(cpu_textures_rgba);
#endif // ifndef NOTEXTURES
#ifdef SAVE_CLT
free(cpu_clt);
#endif
free (host_kern_buf);
// TODO: move somewhere when all is done
for (int ncam = 0; ncam < NUM_CAMS; ncam++) {
checkCudaErrors(cudaFree(gpu_kernels_h[ncam]));
checkCudaErrors(cudaFree(gpu_kernel_offsets_h[ncam]));
checkCudaErrors(cudaFree(gpu_images_h[ncam]));
checkCudaErrors(cudaFree(gpu_clt_h[ncam]));
#ifndef NOICLT
checkCudaErrors(cudaFree(gpu_corr_images_h[ncam]));
#endif
}
checkCudaErrors(cudaFree(gpu_tasks));
checkCudaErrors(cudaFree(gpu_kernels));
checkCudaErrors(cudaFree(gpu_kernel_offsets));
checkCudaErrors(cudaFree(gpu_images));
checkCudaErrors(cudaFree(gpu_clt));
// checkCudaErrors(cudaFree(gpu_corr_images));
checkCudaErrors(cudaFree(gpu_corrs));
checkCudaErrors(cudaFree(gpu_corr_indices));
checkCudaErrors(cudaFree(gpu_texture_indices));
checkCudaErrors(cudaFree(gpu_port_offsets));
checkCudaErrors(cudaFree(gpu_textures));
checkCudaErrors(cudaFree(gpu_textures_rbga));
checkCudaErrors(cudaFree(gpu_woi));
checkCudaErrors(cudaFree(gpu_num_texture_tiles));
exit(0);
}
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment