OpenCL: More work on centralization (need SCons reflection)

git-svn-id: https://dolphin-emu.googlecode.com/svn/trunk@4351 8ced0084-cf51-0410-be5f-012b33b47a6e
This commit is contained in:
XTra.KrazzY 2009-09-30 21:01:34 +00:00
parent 1285ba4401
commit c6f0335b62
6 changed files with 240 additions and 121 deletions

View File

@ -724,6 +724,14 @@
RelativePath=".\Src\MsgHandler.h"
>
</File>
<File
RelativePath=".\Src\OpenCL.cpp"
>
</File>
<File
RelativePath=".\Src\OpenCL.h"
>
</File>
<File
RelativePath=".\Src\SConscript"
>

View File

@ -0,0 +1,95 @@
// Copyright (C) 2003 Dolphin Project.
// This program 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, version 2.0.
// 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 2.0 for more details.
// A copy of the GPL 2.0 should have been included with the program.
// If not, see http://www.gnu.org/licenses/
// Official SVN repository and contact information can be found at
// http://code.google.com/p/dolphin-emu/
// TODO: Make a more centralized version of this (for now every plugin that will use it will create its own context, which is weird). An object maybe?
#include "OpenCL.h"
#include "Common.h"
namespace OpenCL {
cl_context g_context = NULL;
cl_command_queue g_cmdq = NULL;
bool Initialize() {
if(g_context)
return false;
#if defined(HAVE_OPENCL) && HAVE_OPENCL
int err; // error code returned from api calls
cl_device_id device_id;
// Connect to a compute device
//
int gpu = 1; // I think we should use CL_DEVICE_TYPE_ALL
err = clGetDeviceIDs(NULL, gpu ? CL_DEVICE_TYPE_GPU : CL_DEVICE_TYPE_CPU, 1, &device_id, NULL);
if (err != CL_SUCCESS)
{
PanicAlert("Error: Failed to create a device group!\n");
return false;
}
// Create a compute context
//
g_context = clCreateContext(0, 1, &device_id, NULL, NULL, &err);
if (!g_context)
{
PanicAlert("Error: Failed to create a compute context!\n");
return false;
}
// Create a command commands
//
g_cmdq = clCreateCommandQueue(g_context, device_id, 0, &err);
if (!g_cmdq)
{
PanicAlert("Error: Failed to create a command commands!\n");
return false;
}
return true;
#else
return false;
#endif
}
cl_context GetInstance() {
return g_context;
}
cl_command_queue GetCommandQueue() {
return g_cmdq;
}
cl_program CompileProgram(const char *program, unsigned int size) {
// TODO
return NULL;
}
void Destroy() {
if(!g_context)
return;
#if defined(HAVE_OPENCL) && HAVE_OPENCL
clReleaseCommandQueue(g_cmdq);
clReleaseContext(g_context);
#endif
}
};

View File

@ -0,0 +1,57 @@
// Copyright (C) 2003 Dolphin Project.
// This program 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, version 2.0.
// 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 2.0 for more details.
// A copy of the GPL 2.0 should have been included with the program.
// If not, see http://www.gnu.org/licenses/
// Official SVN repository and contact information can be found at
// http://code.google.com/p/dolphin-emu/
#ifndef __OPENCL_H__
#define __OPENCL_H__
// Change to #if 1 if you want to test OpenCL (and you have it) on Windows
#if 0
#pragma comment(lib, "OpenCL.lib")
#define HAVE_OPENCL 1
#endif
#if defined(HAVE_OPENCL) && HAVE_OPENCL
#ifdef __APPLE__
#include <OpenCL/opencl.h>
#else
#include <CL/cl.h>
#endif
#else
typedef void *cl_context;
typedef void *cl_command_queue;
typedef void *cl_program;
#endif
namespace OpenCL {
bool Initialize();
cl_context GetContext();
cl_command_queue GetCommandQueue();
void Destroy();
cl_program CompileProgram(const char *program, unsigned int size);
};
#endif

View File

@ -41,6 +41,8 @@
#include "Fifo.h"
#include "DataReader.h"
#include "OpenCL.h"
u8* g_pVideoData = 0;
extern u8* FAKE_GetFifoStartPtr();
@ -377,11 +379,18 @@ static void DecodeSemiNop()
void OpcodeDecoder_Init()
{
g_pVideoData = FAKE_GetFifoStartPtr();
#if defined(HAVE_OPENCL) && HAVE_OPENCL
OpenCL::Initialize();
#endif
}
void OpcodeDecoder_Shutdown()
{
#if defined(HAVE_OPENCL) && HAVE_OPENCL
OpenCL::Destroy();
#endif
}
void OpcodeDecoder_Run(bool skipped_frame)

View File

@ -17,11 +17,7 @@
#include "TextureDecoder.h"
#ifdef __APPLE__
#include <OpenCL/opencl.h>
#else
#include <CL/cl.h>
#endif
#include "OpenCL.h"
#include <fcntl.h>
#include <stdio.h>
@ -32,12 +28,8 @@
#include <sys/types.h>
#include <sys/stat.h>
size_t global; // global domain size for our calculation
size_t local; // local domain size for our calculation
cl_device_id device_id; // compute device id
cl_context context; // compute context
cl_command_queue commands; // compute command queue
size_t global; // global domain size for our calculation
size_t local; // local domain size for our calculation
struct sDecoders
@ -46,105 +38,71 @@ struct sDecoders
cl_kernel kernel; // compute kernel
const char **cKernel;
};
const char *Kernel = " \
__kernel void Decode(__local unsigned char *dst, \
__local const unsigned char *src, \
int width, int height) \
{ \
int id = get_global_id(0); \
for (int xy = 0; xy < height*width; xy += 4) \
for (int iy = 0; iy < 4; iy++, src += 8) { \
u16 *ptr = (u16 *)dst + ((xy / width) + iy) * \
width + (xy % width); \
u16 *s = (u16 *)src; \
for(int j = 0; j < 4; j++) \
*ptr++ = Common::swap16(*s++); \
}
const char *Kernel = " \
kernel void Decode(global uchar *dst, \
const global uchar *src, \
int width, int height) \
{ \
int x = get_global_id(0), y = get_global_id(1); \
\
for (int iy = 0; iy < 4; iy++, src += 8) { \
u16 *ptr = (u16 *)dst + ((x + (y / width)) + iy) * \
width + ((x * width + y) % width); \
u16 *s = (u16 *)src; \
for(int j = 0; j < 4; j++) \
*ptr++ = Common::swap16(*s++); \
} \
}";
sDecoders Decoders[] = { {NULL, NULL, &Kernel},
bool Inited = false;
// TODO: Deinit (clRelease...)
bool Init_OpenCL()
{
int err; // error code returned from api calls
// Connect to a compute device
//
int gpu = 1;
err = clGetDeviceIDs(NULL, gpu ? CL_DEVICE_TYPE_GPU : CL_DEVICE_TYPE_CPU, 1, &device_id, NULL);
if (err != CL_SUCCESS)
{
printf("Error: Failed to create a device group!\n");
return EXIT_FAILURE;
}
// Create a compute context
//
context = clCreateContext(0, 1, &device_id, NULL, NULL, &err);
if (!context)
{
printf("Error: Failed to create a compute context!\n");
return EXIT_FAILURE;
}
// Create a command commands
//
commands = clCreateCommandQueue(context, device_id, 0, &err);
if (!commands)
{
printf("Error: Failed to create a command commands!\n");
return EXIT_FAILURE;
}
// Create the compute program from the source buffer
//
/*
program = clCreateProgramWithSource(context, 1, (const char **) & KernelSource, NULL, &err);
if (!program)
{
printf("Error: Failed to create compute program!\n");
return EXIT_FAILURE;
}
// Build the program executable
//
err = clBuildProgram(program, 0, NULL, NULL, NULL, NULL);
if (err != CL_SUCCESS)
{
size_t len;
char buffer[2048];
printf("Error: Failed to build program executable!\n");
clGetProgramBuildInfo(program, device_id, CL_PROGRAM_BUILD_LOG, sizeof(buffer), buffer, &len);
printf("%s\n", buffer);
exit(1);
}
// Create the compute kernel in the program we wish to run
//
kernel = clCreateKernel(program, "Decoder", &err);
if (!kernel || err != CL_SUCCESS)
{
printf("Error: Failed to create compute kernel!\n");
exit(1);
} */
}
bool g_Inited = false;
PC_TexFormat TexDecoder_Decode_OpenCL(u8 *dst, const u8 *src, int width, int height, int texformat, int tlutaddr, int tlutfmt)
{
if(!Inited)
if(!g_Inited)
{
// Not yet inited, let's init now
// Need to make a init function later
if(!Init_OpenCL())
PanicAlert("OpenCL could not initialize successfully");
}
g_Inited = true;
// clEnqueueNDRangeKernel
#if defined(HAVE_OPENCL) && HAVE_OPENCL
// TODO: Compile the program
// Create the compute program from the source buffer
//
/*
program = clCreateProgramWithSource(context, 1, (const char **) & KernelSource, NULL, &err);
if (!program)
{
printf("Error: Failed to create compute program!\n");
return EXIT_FAILURE;
}
// Build the program executable
//
err = clBuildProgram(program, 0, NULL, NULL, NULL, NULL);
if (err != CL_SUCCESS)
{
size_t len;
char buffer[2048];
printf("Error: Failed to build program executable!\n");
clGetProgramBuildInfo(program, device_id, CL_PROGRAM_BUILD_LOG, sizeof(buffer), buffer, &len);
printf("%s\n", buffer);
exit(1);
}
// Create the compute kernel in the program we wish to run
//
kernel = clCreateKernel(program, "Decoder", &err);
if (!kernel || err != CL_SUCCESS)
{
printf("Error: Failed to create compute kernel!\n");
exit(1);
} */
#endif
}
// TODO: clEnqueueNDRangeKernel
/*switch (texformat)
@ -279,25 +237,6 @@ PC_TexFormat TexDecoder_Decode_OpenCL(u8 *dst, const u8 *src, int width, int hei
case GX_TF_CMPR: // speed critical
// The metroid games use this format almost exclusively.
{
#if 0 // TODO - currently does not handle transparency correctly and causes problems when texture dimensions are not multiples of 8
// 11111111 22222222 55555555 66666666
// 33333333 44444444 77777777 88888888
for (int y = 0; y < height; y += 8)
{
for (int x = 0; x < width; x += 8)
{
copyDXTBlock(dst+(y/2)*width+x*2, src);
src += 8;
copyDXTBlock(dst+(y/2)*width+x*2+8, src);
src += 8;
copyDXTBlock(dst+(y/2+2)*width+x*2, src);
src += 8;
copyDXTBlock(dst+(y/2+2)*width+x*2+8, src);
src += 8;
}
}
return PC_TEX_FMT_DXT1;
#else
for (int y = 0; y < height; y += 8)
{
for (int x = 0; x < width; x += 8)
@ -312,7 +251,6 @@ PC_TexFormat TexDecoder_Decode_OpenCL(u8 *dst, const u8 *src, int width, int hei
src += sizeof(DXTBlock);
}
}
#endif
return PC_TEX_FMT_BGRA32;
}
}

View File

@ -688,6 +688,18 @@
RelativePath=".\Src\TextureDecoder.h"
>
</File>
<Filter
Name="OpenCL"
>
<File
RelativePath=".\Src\OpenCL\TextureDecoder.cpp"
>
</File>
<File
RelativePath=".\Src\OpenCL\TextureDecoder.h"
>
</File>
</Filter>
</Filter>
<File
RelativePath=".\Src\GlobalControl.cpp"