rpcs3/rpcs3/Emu/RSX/GL/GLTexture.cpp

1104 lines
40 KiB
C++

#include "stdafx.h"
#include "GLTexture.h"
#include "GLCompute.h"
#include "GLRenderTargets.h"
#include "GLOverlays.h"
#include "glutils/blitter.h"
#include "glutils/ring_buffer.h"
#include "../GCM.h"
#include "../RSXThread.h"
#include "../RSXTexture.h"
#include "util/asm.hpp"
namespace gl
{
namespace debug
{
extern void set_vis_texture(texture*);
}
scratch_ring_buffer g_typeless_transfer_buffer;
legacy_ring_buffer g_upload_transfer_buffer;
scratch_ring_buffer g_compute_decode_buffer;
scratch_ring_buffer g_deswizzle_scratch_buffer;
void destroy_global_texture_resources()
{
g_typeless_transfer_buffer.remove();
g_upload_transfer_buffer.remove();
g_compute_decode_buffer.remove();
g_deswizzle_scratch_buffer.remove();
}
template <typename WordType, bool SwapBytes>
void do_deswizzle_transformation(gl::command_context& cmd, u32 block_size, buffer* dst, u32 dst_offset, buffer* src, u32 src_offset, u32 data_length, u16 width, u16 height, u16 depth)
{
switch (block_size)
{
case 4:
gl::get_compute_task<gl::cs_deswizzle_3d<u32, WordType, SwapBytes>>()->run(
cmd, dst, dst_offset, src, src_offset,
data_length, width, height, depth, 1);
break;
case 8:
gl::get_compute_task<gl::cs_deswizzle_3d<u64, WordType, SwapBytes>>()->run(
cmd, dst, dst_offset, src, src_offset,
data_length, width, height, depth, 1);
break;
case 16:
gl::get_compute_task<gl::cs_deswizzle_3d<u128, WordType, SwapBytes>>()->run(
cmd, dst, dst_offset, src, src_offset,
data_length, width, height, depth, 1);
break;
default:
fmt::throw_exception("Unreachable");
}
}
GLenum get_target(rsx::texture_dimension_extended type)
{
switch (type)
{
case rsx::texture_dimension_extended::texture_dimension_1d: return GL_TEXTURE_1D;
case rsx::texture_dimension_extended::texture_dimension_2d: return GL_TEXTURE_2D;
case rsx::texture_dimension_extended::texture_dimension_cubemap: return GL_TEXTURE_CUBE_MAP;
case rsx::texture_dimension_extended::texture_dimension_3d: return GL_TEXTURE_3D;
}
fmt::throw_exception("Unknown texture target");
}
GLenum get_sized_internal_format(u32 texture_format)
{
switch (texture_format)
{
case CELL_GCM_TEXTURE_B8: return GL_R8;
case CELL_GCM_TEXTURE_A1R5G5B5: return GL_BGR5_A1;
case CELL_GCM_TEXTURE_A4R4G4B4: return GL_RGBA4;
case CELL_GCM_TEXTURE_R5G6B5: return GL_RGB565;
case CELL_GCM_TEXTURE_A8R8G8B8: return GL_BGRA8;
case CELL_GCM_TEXTURE_G8B8: return GL_RG8;
case CELL_GCM_TEXTURE_R6G5B5: return GL_RGB565;
case CELL_GCM_TEXTURE_DEPTH24_D8: return GL_DEPTH24_STENCIL8;
case CELL_GCM_TEXTURE_DEPTH24_D8_FLOAT: return GL_DEPTH32F_STENCIL8;
case CELL_GCM_TEXTURE_DEPTH16: return GL_DEPTH_COMPONENT16;
case CELL_GCM_TEXTURE_DEPTH16_FLOAT: return GL_DEPTH_COMPONENT32F;
case CELL_GCM_TEXTURE_X16: return GL_R16;
case CELL_GCM_TEXTURE_Y16_X16: return GL_RG16;
case CELL_GCM_TEXTURE_R5G5B5A1: return GL_RGB5_A1;
case CELL_GCM_TEXTURE_W16_Z16_Y16_X16_FLOAT: return GL_RGBA16F;
case CELL_GCM_TEXTURE_W32_Z32_Y32_X32_FLOAT: return GL_RGBA32F;
case CELL_GCM_TEXTURE_X32_FLOAT: return GL_R32F;
case CELL_GCM_TEXTURE_D1R5G5B5: return GL_BGR5_A1;
case CELL_GCM_TEXTURE_D8R8G8B8: return GL_BGRA8;
case CELL_GCM_TEXTURE_Y16_X16_FLOAT: return GL_RG16F;
case CELL_GCM_TEXTURE_COMPRESSED_DXT1: return GL_COMPRESSED_RGBA_S3TC_DXT1_EXT;
case CELL_GCM_TEXTURE_COMPRESSED_DXT23: return GL_COMPRESSED_RGBA_S3TC_DXT3_EXT;
case CELL_GCM_TEXTURE_COMPRESSED_DXT45: return GL_COMPRESSED_RGBA_S3TC_DXT5_EXT;
case CELL_GCM_TEXTURE_COMPRESSED_HILO8: return GL_RG8;
case CELL_GCM_TEXTURE_COMPRESSED_HILO_S8: return GL_RG8_SNORM;
case CELL_GCM_TEXTURE_COMPRESSED_B8R8_G8R8: return GL_BGRA8;
case CELL_GCM_TEXTURE_COMPRESSED_R8B8_R8G8: return GL_BGRA8;
}
fmt::throw_exception("Unknown texture format 0x%x", texture_format);
}
std::tuple<GLenum, GLenum> get_format_type(u32 texture_format)
{
switch (texture_format)
{
case CELL_GCM_TEXTURE_B8: return std::make_tuple(GL_RED, GL_UNSIGNED_BYTE);
case CELL_GCM_TEXTURE_A1R5G5B5: return std::make_tuple(GL_BGRA, GL_UNSIGNED_SHORT_1_5_5_5_REV);
case CELL_GCM_TEXTURE_A4R4G4B4: return std::make_tuple(GL_BGRA, GL_UNSIGNED_SHORT_4_4_4_4);
case CELL_GCM_TEXTURE_R5G6B5: return std::make_tuple(GL_RGB, GL_UNSIGNED_SHORT_5_6_5);
case CELL_GCM_TEXTURE_A8R8G8B8: return std::make_tuple(GL_BGRA, GL_UNSIGNED_INT_8_8_8_8_REV);
case CELL_GCM_TEXTURE_G8B8: return std::make_tuple(GL_RG, GL_UNSIGNED_BYTE);
case CELL_GCM_TEXTURE_R6G5B5: return std::make_tuple(GL_RGB, GL_UNSIGNED_SHORT_5_6_5);
case CELL_GCM_TEXTURE_DEPTH24_D8: return std::make_tuple(GL_DEPTH_STENCIL, GL_UNSIGNED_INT_24_8);
case CELL_GCM_TEXTURE_DEPTH24_D8_FLOAT: return std::make_tuple(GL_DEPTH_STENCIL, GL_FLOAT_32_UNSIGNED_INT_24_8_REV);
case CELL_GCM_TEXTURE_DEPTH16: return std::make_tuple(GL_DEPTH_COMPONENT, GL_UNSIGNED_SHORT);
case CELL_GCM_TEXTURE_DEPTH16_FLOAT: return std::make_tuple(GL_DEPTH_COMPONENT, GL_FLOAT);
case CELL_GCM_TEXTURE_X16: return std::make_tuple(GL_RED, GL_UNSIGNED_SHORT);
case CELL_GCM_TEXTURE_Y16_X16: return std::make_tuple(GL_RG, GL_UNSIGNED_SHORT);
case CELL_GCM_TEXTURE_R5G5B5A1: return std::make_tuple(GL_RGBA, GL_UNSIGNED_SHORT_5_5_5_1);
case CELL_GCM_TEXTURE_W16_Z16_Y16_X16_FLOAT: return std::make_tuple(GL_RGBA, GL_HALF_FLOAT);
case CELL_GCM_TEXTURE_W32_Z32_Y32_X32_FLOAT: return std::make_tuple(GL_RGBA, GL_FLOAT);
case CELL_GCM_TEXTURE_X32_FLOAT: return std::make_tuple(GL_RED, GL_FLOAT);
case CELL_GCM_TEXTURE_D1R5G5B5: return std::make_tuple(GL_BGRA, GL_UNSIGNED_SHORT_1_5_5_5_REV);
case CELL_GCM_TEXTURE_D8R8G8B8: return std::make_tuple(GL_BGRA, GL_UNSIGNED_INT_8_8_8_8_REV);
case CELL_GCM_TEXTURE_Y16_X16_FLOAT: return std::make_tuple(GL_RG, GL_HALF_FLOAT);
case CELL_GCM_TEXTURE_COMPRESSED_DXT1: return std::make_tuple(GL_COMPRESSED_RGBA_S3TC_DXT1_EXT, GL_UNSIGNED_BYTE);
case CELL_GCM_TEXTURE_COMPRESSED_DXT23: return std::make_tuple(GL_COMPRESSED_RGBA_S3TC_DXT3_EXT, GL_UNSIGNED_BYTE);
case CELL_GCM_TEXTURE_COMPRESSED_DXT45: return std::make_tuple(GL_COMPRESSED_RGBA_S3TC_DXT5_EXT, GL_UNSIGNED_BYTE);
case CELL_GCM_TEXTURE_COMPRESSED_HILO8: return std::make_tuple(GL_RG, GL_UNSIGNED_BYTE);
case CELL_GCM_TEXTURE_COMPRESSED_HILO_S8: return std::make_tuple(GL_RG, GL_BYTE);
case CELL_GCM_TEXTURE_COMPRESSED_B8R8_G8R8: return std::make_tuple(GL_BGRA, GL_UNSIGNED_BYTE);
case CELL_GCM_TEXTURE_COMPRESSED_R8B8_R8G8: return std::make_tuple(GL_BGRA, GL_UNSIGNED_BYTE);
}
fmt::throw_exception("Compressed or unknown texture format 0x%x", texture_format);
}
pixel_buffer_layout get_format_type(texture::internal_format format)
{
switch (format)
{
case texture::internal_format::compressed_rgba_s3tc_dxt1:
case texture::internal_format::compressed_rgba_s3tc_dxt3:
case texture::internal_format::compressed_rgba_s3tc_dxt5:
return { GL_RGBA, GL_UNSIGNED_BYTE, 1, false };
case texture::internal_format::r8:
return { GL_RED, GL_UNSIGNED_BYTE, 1, false };
case texture::internal_format::r16:
return { GL_RED, GL_UNSIGNED_SHORT, 2, true };
case texture::internal_format::r32f:
return { GL_RED, GL_FLOAT, 4, true };
case texture::internal_format::rg8:
return { GL_RG, GL_UNSIGNED_SHORT, 2, true };
case texture::internal_format::rg16:
return { GL_RG, GL_UNSIGNED_SHORT, 2, true };
case texture::internal_format::rg16f:
return { GL_RG, GL_HALF_FLOAT, 2, true };
case texture::internal_format::rgb565:
return { GL_RGB, GL_UNSIGNED_SHORT_5_6_5, 2, true };
case texture::internal_format::rgb5a1:
return { GL_RGB, GL_UNSIGNED_SHORT_5_5_5_1, 2, true };
case texture::internal_format::bgr5a1:
return { GL_RGB, GL_UNSIGNED_SHORT_1_5_5_5_REV, 2, true };
case texture::internal_format::rgba4:
return { GL_BGRA, GL_UNSIGNED_SHORT_4_4_4_4, 2, false };
case texture::internal_format::rgba8:
return { GL_RGBA, GL_UNSIGNED_INT_8_8_8_8_REV, 4, true };
case texture::internal_format::bgra8:
return { GL_BGRA, GL_UNSIGNED_INT_8_8_8_8_REV, 4, true };
case texture::internal_format::rgba16f:
return { GL_RGBA, GL_HALF_FLOAT, 2, true };
case texture::internal_format::rgba32f:
return { GL_RGBA, GL_FLOAT, 4, true };
case texture::internal_format::depth16:
return { GL_DEPTH_COMPONENT, GL_UNSIGNED_SHORT, 2, true };
case texture::internal_format::depth32f:
return { GL_DEPTH_COMPONENT, GL_FLOAT, 2, true };
case texture::internal_format::depth24_stencil8:
case texture::internal_format::depth32f_stencil8:
return { GL_DEPTH_STENCIL, GL_UNSIGNED_INT_24_8, 4, true };
default:
fmt::throw_exception("Unexpected internal format 0x%X", static_cast<u32>(format));
}
}
pixel_buffer_layout get_format_type(const gl::texture* tex)
{
auto ret = get_format_type(tex->get_internal_format());
if (tex->format_class() == RSX_FORMAT_CLASS_DEPTH24_FLOAT_X8_PACK32)
{
ret.type = GL_FLOAT_32_UNSIGNED_INT_24_8_REV;
}
return ret;
}
std::array<GLenum, 4> get_swizzle_remap(u32 texture_format)
{
// NOTE: This must be in ARGB order in all forms below.
switch (texture_format)
{
case CELL_GCM_TEXTURE_A1R5G5B5:
case CELL_GCM_TEXTURE_R5G5B5A1:
case CELL_GCM_TEXTURE_R6G5B5:
case CELL_GCM_TEXTURE_R5G6B5:
case CELL_GCM_TEXTURE_A4R4G4B4:
case CELL_GCM_TEXTURE_A8R8G8B8:
case CELL_GCM_TEXTURE_COMPRESSED_DXT1:
case CELL_GCM_TEXTURE_COMPRESSED_DXT23:
case CELL_GCM_TEXTURE_COMPRESSED_DXT45:
case CELL_GCM_TEXTURE_COMPRESSED_B8R8_G8R8:
case CELL_GCM_TEXTURE_COMPRESSED_R8B8_R8G8:
return{ GL_ALPHA, GL_RED, GL_GREEN, GL_BLUE };
case CELL_GCM_TEXTURE_DEPTH24_D8:
case CELL_GCM_TEXTURE_DEPTH24_D8_FLOAT:
case CELL_GCM_TEXTURE_DEPTH16:
case CELL_GCM_TEXTURE_DEPTH16_FLOAT:
return{ GL_RED, GL_RED, GL_RED, GL_RED };
case CELL_GCM_TEXTURE_B8:
return{ GL_ONE, GL_RED, GL_RED, GL_RED };
case CELL_GCM_TEXTURE_X16:
return{ GL_RED, GL_ONE, GL_RED, GL_ONE };
case CELL_GCM_TEXTURE_X32_FLOAT:
return{ GL_RED, GL_RED, GL_RED, GL_RED };
case CELL_GCM_TEXTURE_G8B8:
return{ GL_GREEN, GL_RED, GL_GREEN, GL_RED };
case CELL_GCM_TEXTURE_Y16_X16:
return{ GL_GREEN, GL_RED, GL_GREEN, GL_RED };
case CELL_GCM_TEXTURE_Y16_X16_FLOAT:
return{ GL_RED, GL_GREEN, GL_RED, GL_GREEN };
case CELL_GCM_TEXTURE_W16_Z16_Y16_X16_FLOAT:
case CELL_GCM_TEXTURE_W32_Z32_Y32_X32_FLOAT:
return{ GL_ALPHA, GL_RED, GL_GREEN, GL_BLUE };
case CELL_GCM_TEXTURE_D1R5G5B5:
case CELL_GCM_TEXTURE_D8R8G8B8:
return{ GL_ONE, GL_RED, GL_GREEN, GL_BLUE };
case CELL_GCM_TEXTURE_COMPRESSED_HILO8:
case CELL_GCM_TEXTURE_COMPRESSED_HILO_S8:
return{ GL_RED, GL_GREEN, GL_RED, GL_GREEN };
}
fmt::throw_exception("Unknown format 0x%x", texture_format);
}
cs_shuffle_base* get_trivial_transform_job(const pixel_buffer_layout& pack_info)
{
if (!pack_info.swap_bytes)
{
return nullptr;
}
switch (pack_info.size)
{
case 1:
return nullptr;
case 2:
return get_compute_task<gl::cs_shuffle_16>();
case 4:
return get_compute_task<gl::cs_shuffle_32>();
default:
fmt::throw_exception("Unsupported format");
}
}
void* copy_image_to_buffer(gl::command_context& cmd, const pixel_buffer_layout& pack_info, const gl::texture* src, gl::buffer* dst,
u32 dst_offset, const int src_level, const coord3u& src_region, image_memory_requirements* mem_info)
{
auto initialize_scratch_mem = [&]() -> bool // skip_transform
{
const u64 max_mem = (mem_info->memory_required) ? mem_info->memory_required : mem_info->image_size_in_bytes;
if (!(*dst) || max_mem > static_cast<u64>(dst->size()))
{
if (*dst) dst->remove();
dst->create(buffer::target::ssbo, max_mem, nullptr, buffer::memory_type::local, GL_STATIC_COPY);
}
if (auto as_vi = dynamic_cast<const gl::viewable_image*>(src);
src->get_target() == gl::texture::target::texture2D &&
as_vi)
{
// RGBA8 <-> D24X8 bitcasts are some very common conversions due to some PS3 coding hacks & workarounds.
switch (src->get_internal_format())
{
case gl::texture::internal_format::depth24_stencil8:
gl::get_compute_task<gl::cs_d24x8_to_ssbo>()->run(cmd,
const_cast<gl::viewable_image*>(as_vi), dst, dst_offset,
{ {src_region.x, src_region.y}, {src_region.width, src_region.height} },
pack_info);
return true;
case gl::texture::internal_format::rgba8:
case gl::texture::internal_format::bgra8:
gl::get_compute_task<gl::cs_rgba8_to_ssbo>()->run(cmd,
const_cast<gl::viewable_image*>(as_vi), dst, dst_offset,
{ {src_region.x, src_region.y}, {src_region.width, src_region.height} },
pack_info);
return true;
default:
break;
}
}
dst->bind(buffer::target::pixel_pack);
src->copy_to(reinterpret_cast<void*>(static_cast<uintptr_t>(dst_offset)), static_cast<texture::format>(pack_info.format), static_cast<texture::type>(pack_info.type), src_level, src_region, {});
return false;
};
void* result = reinterpret_cast<void*>(static_cast<uintptr_t>(dst_offset));
if (src->aspect() == image_aspect::color ||
pack_info.type == GL_UNSIGNED_SHORT ||
pack_info.type == GL_UNSIGNED_INT_24_8)
{
if (!initialize_scratch_mem())
{
if (auto job = get_trivial_transform_job(pack_info))
{
job->run(cmd, dst, static_cast<u32>(mem_info->image_size_in_bytes), dst_offset);
}
}
}
else if (pack_info.type == GL_FLOAT)
{
ensure(mem_info->image_size_in_bytes == (mem_info->image_size_in_texels * 4));
mem_info->memory_required = (mem_info->image_size_in_texels * 6);
ensure(!initialize_scratch_mem());
if (pack_info.swap_bytes) [[ likely ]]
{
get_compute_task<cs_fconvert_task<f32, f16, false, true>>()->run(cmd, dst, dst_offset,
static_cast<u32>(mem_info->image_size_in_bytes), static_cast<u32>(mem_info->image_size_in_bytes));
}
else
{
get_compute_task<cs_fconvert_task<f32, f16, false, false>>()->run(cmd, dst, dst_offset,
static_cast<u32>(mem_info->image_size_in_bytes), static_cast<u32>(mem_info->image_size_in_bytes));
}
result = reinterpret_cast<void*>(mem_info->image_size_in_bytes + dst_offset);
}
else if (pack_info.type == GL_FLOAT_32_UNSIGNED_INT_24_8_REV)
{
ensure(mem_info->image_size_in_bytes == (mem_info->image_size_in_texels * 8));
mem_info->memory_required = (mem_info->image_size_in_texels * 12);
ensure(!initialize_scratch_mem());
if (pack_info.swap_bytes)
{
get_compute_task<cs_shuffle_d32fx8_to_x8d24f<true>>()->run(cmd, dst, dst_offset,
static_cast<u32>(mem_info->image_size_in_bytes), static_cast<u32>(mem_info->image_size_in_texels));
}
else
{
get_compute_task<cs_shuffle_d32fx8_to_x8d24f<false>>()->run(cmd, dst, dst_offset,
static_cast<u32>(mem_info->image_size_in_bytes), static_cast<u32>(mem_info->image_size_in_texels));
}
result = reinterpret_cast<void*>(mem_info->image_size_in_bytes + dst_offset);
}
else
{
fmt::throw_exception("Invalid depth/stencil type 0x%x", pack_info.type);
}
glMemoryBarrier(GL_SHADER_STORAGE_BARRIER_BIT | GL_PIXEL_BUFFER_BARRIER_BIT);
return result;
}
void copy_buffer_to_image(gl::command_context& cmd, const pixel_buffer_layout& unpack_info, gl::buffer* src, gl::texture* dst,
const void* src_offset, const int dst_level, const coord3u& dst_region, image_memory_requirements* mem_info)
{
buffer scratch_mem;
buffer* transfer_buf = src;
bool skip_barrier = false;
u32 in_offset = static_cast<u32>(reinterpret_cast<u64>(src_offset));
u32 out_offset = in_offset;
const auto& caps = gl::get_driver_caps();
auto initialize_scratch_mem = [&]()
{
if (in_offset >= mem_info->memory_required)
{
return;
}
const u64 max_mem = mem_info->memory_required + mem_info->image_size_in_bytes;
if ((max_mem + in_offset) <= static_cast<u64>(src->size()))
{
out_offset = static_cast<u32>(in_offset + mem_info->image_size_in_bytes);
return;
}
scratch_mem.create(buffer::target::pixel_pack, max_mem, nullptr, buffer::memory_type::local, GL_STATIC_COPY);
glMemoryBarrier(GL_BUFFER_UPDATE_BARRIER_BIT);
src->copy_to(&scratch_mem, in_offset, 0, mem_info->image_size_in_bytes);
in_offset = 0;
out_offset = static_cast<u32>(mem_info->image_size_in_bytes);
transfer_buf = &scratch_mem;
};
if ((dst->aspect() & image_aspect::stencil) == 0 || caps.ARB_shader_stencil_export_supported)
{
// We do not need to use the driver's builtin transport mechanism
glMemoryBarrier(GL_SHADER_STORAGE_BARRIER_BIT);
std::unique_ptr<gl::texture> scratch;
std::unique_ptr<gl::texture_view> scratch_view;
coordu image_region = { {dst_region.x, dst_region.y}, {dst_region.width, dst_region.height} };
switch (dst->get_target())
{
case texture::target::texture3D:
{
// Upload to splatted image and do the final copy GPU-side
image_region.height *= dst_region.depth;
scratch = std::make_unique<gl::texture>(
GL_TEXTURE_2D,
image_region.x + image_region.width, image_region.y + image_region.height, 1, 1,
static_cast<GLenum>(dst->get_internal_format()), dst->format_class());
scratch_view = std::make_unique<gl::nil_texture_view>(scratch.get());
break;
}
case texture::target::textureCUBE:
{
const subresource_range range = { image_aspect::depth | image_aspect::color, static_cast<GLuint>(dst_level), 1, dst_region.z , 1 };
scratch_view = std::make_unique<gl::texture_view>(dst, GL_TEXTURE_2D, range);
break;
}
case texture::target::texture1D:
{
scratch = std::make_unique<gl::texture>(
GL_TEXTURE_2D,
image_region.x + image_region.width, 1, 1, 1,
static_cast<GLenum>(dst->get_internal_format()), dst->format_class());
scratch_view = std::make_unique<gl::nil_texture_view>(scratch.get());
break;
}
default:
{
ensure(dst->layers() == 1);
if (dst->levels() > 1) [[ likely ]]
{
const subresource_range range = { image_aspect::depth | image_aspect::color, static_cast<GLuint>(dst_level), 1, 0 , 1 };
scratch_view = std::make_unique<gl::texture_view>(dst, GL_TEXTURE_2D, range);
}
else
{
scratch_view = std::make_unique<gl::nil_texture_view>(dst);
}
break;
}
}
// If possible, decode using a compute transform to potentially have asynchronous scheduling
bool use_compute_transform = (
dst->aspect() == gl::image_aspect::color && // Cannot use image_load_store with depth images
caps.subvendor_ATI == false); // The old AMD/ATI driver does not support image writeonly without format specifier
if (use_compute_transform)
{
switch (dst->get_internal_format())
{
case texture::internal_format::bgr5a1:
case texture::internal_format::rgb5a1:
case texture::internal_format::rgb565:
case texture::internal_format::rgba4:
// Packed formats are a problem with image_load_store
use_compute_transform = false;
break;
default:
break;
}
}
if (use_compute_transform)
{
gl::get_compute_task<gl::cs_ssbo_to_color_image>()->run(cmd, transfer_buf, scratch_view.get(), out_offset, image_region, unpack_info);
}
else
{
gl::get_overlay_pass<gl::rp_ssbo_to_generic_texture>()->run(cmd, transfer_buf, scratch_view.get(), out_offset, image_region, unpack_info);
}
glMemoryBarrier(GL_TEXTURE_FETCH_BARRIER_BIT | GL_TEXTURE_UPDATE_BARRIER_BIT);
switch (dst->get_target())
{
case texture::target::texture1D:
{
const position3u transfer_offset = { dst_region.position.x, 0, 0 };
g_hw_blitter->copy_image(cmd, scratch.get(), dst, 0, dst_level, transfer_offset, transfer_offset, { dst_region.width, 1, 1 });
break;
}
case texture::target::texture3D:
{
// Memcpy
for (u32 layer = dst_region.z, i = 0; i < dst_region.depth; ++i, ++layer)
{
const position3u src_offset = { dst_region.position.x, dst_region.position.y + (i * dst_region.height), 0 };
const position3u dst_offset = { dst_region.position.x, dst_region.position.y, layer };
g_hw_blitter->copy_image(cmd, scratch.get(), dst, 0, dst_level, src_offset, dst_offset, { dst_region.width, dst_region.height, 1 });
}
break;
}
default: break;
}
}
else
{
// Stencil format on NV. Use driver upload path
if (unpack_info.type == GL_UNSIGNED_INT_24_8)
{
if (auto job = get_trivial_transform_job(unpack_info))
{
job->run(cmd, src, static_cast<u32>(mem_info->image_size_in_bytes), in_offset);
}
else
{
skip_barrier = true;
}
}
else if (unpack_info.type == GL_FLOAT_32_UNSIGNED_INT_24_8_REV)
{
mem_info->memory_required = (mem_info->image_size_in_texels * 8);
initialize_scratch_mem();
if (unpack_info.swap_bytes)
{
get_compute_task<cs_shuffle_x8d24f_to_d32fx8<true>>()->run(cmd, transfer_buf, in_offset, out_offset, static_cast<u32>(mem_info->image_size_in_texels));
}
else
{
get_compute_task<cs_shuffle_x8d24f_to_d32fx8<false>>()->run(cmd, transfer_buf, in_offset, out_offset, static_cast<u32>(mem_info->image_size_in_texels));
}
}
else
{
fmt::throw_exception("Invalid depth/stencil type 0x%x", unpack_info.type);
}
if (!skip_barrier)
{
glMemoryBarrier(GL_PIXEL_BUFFER_BARRIER_BIT);
}
glBindBuffer(GL_SHADER_STORAGE_BUFFER, GL_NONE);
transfer_buf->bind(buffer::target::pixel_unpack);
dst->copy_from(reinterpret_cast<void*>(u64(out_offset)), static_cast<texture::format>(unpack_info.format),
static_cast<texture::type>(unpack_info.type), dst_level, dst_region, {});
}
}
gl::viewable_image* create_texture(u32 gcm_format, u16 width, u16 height, u16 depth, u16 mipmaps,
rsx::texture_dimension_extended type)
{
const GLenum target = get_target(type);
const GLenum internal_format = get_sized_internal_format(gcm_format);
const auto format_class = rsx::classify_format(gcm_format);
return new gl::viewable_image(target, width, height, depth, mipmaps, internal_format, format_class);
}
void fill_texture(gl::command_context& cmd, texture* dst, int format,
const std::vector<rsx::subresource_layout> &input_layouts,
bool is_swizzled, GLenum gl_format, GLenum gl_type, rsx::simple_array<std::byte>& staging_buffer)
{
const auto driver_caps = gl::get_driver_caps();
rsx::texture_uploader_capabilities caps
{
.supports_byteswap = true,
.supports_vtc_decoding = false,
.supports_hw_deswizzle = driver_caps.ARB_compute_shader_supported,
.supports_zero_copy = false,
.alignment = 4
};
pixel_unpack_settings unpack_settings;
unpack_settings.row_length(0).alignment(4);
glBindBuffer(GL_PIXEL_UNPACK_BUFFER, GL_NONE);
glBindBuffer(GL_PIXEL_PACK_BUFFER, GL_NONE);
if (rsx::is_compressed_host_format(format)) [[likely]]
{
caps.supports_vtc_decoding = driver_caps.vendor_NVIDIA;
unpack_settings.apply();
const GLsizei format_block_size = (format == CELL_GCM_TEXTURE_COMPRESSED_DXT1) ? 8 : 16;
for (const rsx::subresource_layout& layout : input_layouts)
{
rsx::io_buffer io_buf = staging_buffer;
upload_texture_subresource(io_buf, layout, format, is_swizzled, caps);
switch (dst->get_target())
{
case texture::target::texture1D:
{
const GLsizei size = layout.width_in_block * format_block_size;
ensure(usz(size) <= staging_buffer.size());
DSA_CALL(CompressedTextureSubImage1D, dst->id(), GL_TEXTURE_1D, layout.level, 0, layout.width_in_texel, gl_format, size, staging_buffer.data());
break;
}
case texture::target::texture2D:
{
const GLsizei size = layout.width_in_block * layout.height_in_block * format_block_size;
ensure(usz(size) <= staging_buffer.size());
DSA_CALL(CompressedTextureSubImage2D, dst->id(), GL_TEXTURE_2D, layout.level, 0, 0, layout.width_in_texel, layout.height_in_texel, gl_format, size, staging_buffer.data());
break;
}
case texture::target::textureCUBE:
{
const GLsizei size = layout.width_in_block * layout.height_in_block * format_block_size;
ensure(usz(size) <= staging_buffer.size());
if (gl::get_driver_caps().ARB_dsa_supported)
{
glCompressedTextureSubImage3D(dst->id(), layout.level, 0, 0, layout.layer, layout.width_in_texel, layout.height_in_texel, 1, gl_format, size, staging_buffer.data());
}
else
{
glCompressedTextureSubImage2DEXT(dst->id(), GL_TEXTURE_CUBE_MAP_POSITIVE_X + layout.layer, layout.level, 0, 0, layout.width_in_texel, layout.height_in_texel, gl_format, size, staging_buffer.data());
}
break;
}
case texture::target::texture3D:
{
const GLsizei size = layout.width_in_block * layout.height_in_block * layout.depth * format_block_size;
ensure(usz(size) <= staging_buffer.size());
DSA_CALL(CompressedTextureSubImage3D, dst->id(), GL_TEXTURE_3D, layout.level, 0, 0, 0, layout.width_in_texel, layout.height_in_texel, layout.depth, gl_format, size, staging_buffer.data());
break;
}
default:
{
fmt::throw_exception("Unreachable");
}
}
}
}
else
{
std::pair<void*, u32> upload_scratch_mem = {}, compute_scratch_mem = {};
image_memory_requirements mem_info;
pixel_buffer_layout mem_layout;
std::span<std::byte> dst_buffer = staging_buffer;
void* out_pointer = staging_buffer.data();
u8 block_size_in_bytes = rsx::get_format_block_size_in_bytes(format);
u64 image_linear_size = staging_buffer.size();
const auto min_required_buffer_size = std::max<u64>(utils::align(image_linear_size * 4, 0x100000), 16 * 0x100000);
if (driver_caps.ARB_compute_shader_supported)
{
if (g_upload_transfer_buffer.size() < static_cast<GLsizeiptr>(min_required_buffer_size))
{
g_upload_transfer_buffer.remove();
g_upload_transfer_buffer.create(gl::buffer::target::pixel_unpack, min_required_buffer_size);
}
if (g_compute_decode_buffer.size() < min_required_buffer_size)
{
g_compute_decode_buffer.remove();
g_compute_decode_buffer.create(gl::buffer::target::ssbo, min_required_buffer_size);
}
out_pointer = nullptr;
}
for (const rsx::subresource_layout& layout : input_layouts)
{
if (driver_caps.ARB_compute_shader_supported)
{
u64 row_pitch = rsx::align2<u64, u64>(layout.width_in_block * block_size_in_bytes, caps.alignment);
if (!rsx::is_compressed_host_format(format))
{
// Handle emulated compressed formats with host unpack (R8G8 compressed)
row_pitch = std::max<u64>(row_pitch, dst->pitch());
}
image_linear_size = row_pitch * layout.height_in_block * layout.depth;
compute_scratch_mem = { nullptr, g_compute_decode_buffer.alloc(static_cast<u32>(image_linear_size), 256) };
compute_scratch_mem.first = reinterpret_cast<void*>(static_cast<uintptr_t>(compute_scratch_mem.second));
g_upload_transfer_buffer.reserve_storage_on_heap(static_cast<u32>(image_linear_size));
upload_scratch_mem = g_upload_transfer_buffer.alloc_from_heap(static_cast<u32>(image_linear_size), 256);
dst_buffer = { reinterpret_cast<std::byte*>(upload_scratch_mem.first), image_linear_size };
}
rsx::io_buffer io_buf = dst_buffer;
caps.supports_hw_deswizzle = (is_swizzled && driver_caps.ARB_compute_shader_supported && image_linear_size > 4096);
auto op = upload_texture_subresource(io_buf, layout, format, is_swizzled, caps);
// Define upload region
coord3u region;
region.x = 0;
region.y = 0;
region.z = layout.layer;
region.width = layout.width_in_texel;
region.height = layout.height_in_texel;
region.depth = layout.depth;
if (driver_caps.ARB_compute_shader_supported)
{
// 0. Preconf
mem_layout.alignment = static_cast<u8>(caps.alignment);
mem_layout.swap_bytes = op.require_swap;
mem_layout.format = gl_format;
mem_layout.type = gl_type;
mem_layout.size = block_size_in_bytes;
// 2. Upload memory to GPU
if (!op.require_deswizzle)
{
g_upload_transfer_buffer.unmap();
g_upload_transfer_buffer.copy_to(&g_compute_decode_buffer.get(), upload_scratch_mem.second, compute_scratch_mem.second, image_linear_size);
}
else
{
// 2.1 Copy data to deswizzle buf
if (g_deswizzle_scratch_buffer.size() < min_required_buffer_size)
{
g_deswizzle_scratch_buffer.remove();
g_deswizzle_scratch_buffer.create(gl::buffer::target::ssbo, min_required_buffer_size);
}
u32 deswizzle_data_offset = g_deswizzle_scratch_buffer.alloc(static_cast<u32>(image_linear_size), 256);
g_upload_transfer_buffer.unmap();
g_upload_transfer_buffer.copy_to(&g_deswizzle_scratch_buffer.get(), upload_scratch_mem.second, deswizzle_data_offset, static_cast<u32>(image_linear_size));
// 2.2 Apply compute transform to deswizzle input and dump it in compute_scratch_mem
ensure(op.element_size == 2 || op.element_size == 4);
const auto block_size = op.element_size * op.block_length;
if (op.require_swap)
{
mem_layout.swap_bytes = false;
if (op.element_size == 4) [[ likely ]]
{
do_deswizzle_transformation<u32, true>(cmd, block_size,
&g_compute_decode_buffer.get(), compute_scratch_mem.second, &g_deswizzle_scratch_buffer.get(), deswizzle_data_offset,
static_cast<u32>(image_linear_size), layout.width_in_texel, layout.height_in_texel, layout.depth);
}
else
{
do_deswizzle_transformation<u16, true>(cmd, block_size,
&g_compute_decode_buffer.get(), compute_scratch_mem.second, &g_deswizzle_scratch_buffer.get(), deswizzle_data_offset,
static_cast<u32>(image_linear_size), layout.width_in_texel, layout.height_in_texel, layout.depth);
}
}
else
{
if (op.element_size == 4) [[ likely ]]
{
do_deswizzle_transformation<u32, false>(cmd, block_size,
&g_compute_decode_buffer.get(), compute_scratch_mem.second, &g_deswizzle_scratch_buffer.get(), deswizzle_data_offset,
static_cast<u32>(image_linear_size), layout.width_in_texel, layout.height_in_texel, layout.depth);
}
else
{
do_deswizzle_transformation<u16, false>(cmd, block_size,
&g_compute_decode_buffer.get(), compute_scratch_mem.second, &g_deswizzle_scratch_buffer.get(), deswizzle_data_offset,
static_cast<u32>(image_linear_size), layout.width_in_texel, layout.height_in_texel, layout.depth);
}
}
// Barrier
g_deswizzle_scratch_buffer.push_barrier(deswizzle_data_offset, static_cast<u32>(image_linear_size));
}
// 3. Update configuration
mem_info.image_size_in_texels = image_linear_size / block_size_in_bytes;
mem_info.image_size_in_bytes = image_linear_size;
mem_info.memory_required = 0;
// 4. Dispatch compute routines
copy_buffer_to_image(cmd, mem_layout, &g_compute_decode_buffer.get(), dst, compute_scratch_mem.first, layout.level, region, &mem_info);
// Barrier
g_compute_decode_buffer.push_barrier(compute_scratch_mem.second, static_cast<u32>(image_linear_size));
}
else
{
unpack_settings.swap_bytes(op.require_swap);
dst->copy_from(out_pointer, static_cast<texture::format>(gl_format), static_cast<texture::type>(gl_type), layout.level, region, unpack_settings);
}
}
}
}
std::array<GLenum, 4> apply_swizzle_remap(const std::array<GLenum, 4>& swizzle_remap, const std::pair<std::array<u8, 4>, std::array<u8, 4>>& decoded_remap)
{
//Remapping tables; format is A-R-G-B
//Remap input table. Contains channel index to read color from
const auto remap_inputs = decoded_remap.first;
//Remap control table. Controls whether the remap value is used, or force either 0 or 1
const auto remap_lookup = decoded_remap.second;
std::array<GLenum, 4> remap_values;
for (u8 channel = 0; channel < 4; ++channel)
{
switch (remap_lookup[channel])
{
default:
rsx_log.error("Unknown remap function 0x%X", remap_lookup[channel]);
[[fallthrough]];
case CELL_GCM_TEXTURE_REMAP_REMAP:
remap_values[channel] = swizzle_remap[remap_inputs[channel]];
break;
case CELL_GCM_TEXTURE_REMAP_ZERO:
remap_values[channel] = GL_ZERO;
break;
case CELL_GCM_TEXTURE_REMAP_ONE:
remap_values[channel] = GL_ONE;
break;
}
}
return remap_values;
}
void upload_texture(gl::command_context& cmd, texture* dst, u32 gcm_format, bool is_swizzled, const std::vector<rsx::subresource_layout>& subresources_layout)
{
// Calculate staging buffer size
rsx::simple_array<std::byte> data_upload_buf;
if (rsx::is_compressed_host_format(gcm_format))
{
const auto& desc = subresources_layout[0];
const u32 texture_data_sz = desc.width_in_block * desc.height_in_block * desc.depth * rsx::get_format_block_size_in_bytes(gcm_format);
data_upload_buf.resize(texture_data_sz);
}
else
{
const auto aligned_pitch = utils::align<u32>(dst->pitch(), 4);
const u32 texture_data_sz = dst->depth() * dst->height() * aligned_pitch;
data_upload_buf.resize(texture_data_sz);
}
// TODO: GL drivers support byteswapping and this should be used instead of doing so manually
const auto format_type = get_format_type(gcm_format);
const GLenum gl_format = std::get<0>(format_type);
const GLenum gl_type = std::get<1>(format_type);
fill_texture(cmd, dst, gcm_format, subresources_layout, is_swizzled, gl_format, gl_type, data_upload_buf);
}
u32 get_format_texel_width(GLenum format)
{
switch (format)
{
case GL_R8:
return 1;
case GL_R32F:
case GL_RG16:
case GL_RG16F:
case GL_RGBA8:
case GL_BGRA8:
case GL_COMPRESSED_RGBA_S3TC_DXT1_EXT:
case GL_COMPRESSED_RGBA_S3TC_DXT3_EXT:
case GL_COMPRESSED_RGBA_S3TC_DXT5_EXT:
return 4;
case GL_R16:
case GL_RG8:
case GL_RGB565:
return 2;
case GL_RGBA16F:
return 8;
case GL_RGBA32F:
return 16;
case GL_DEPTH_COMPONENT16:
case GL_DEPTH_COMPONENT32F:
return 2;
case GL_DEPTH24_STENCIL8:
case GL_DEPTH32F_STENCIL8:
return 4;
default:
fmt::throw_exception("Unexpected internal format 0x%X", static_cast<u32>(format));
}
}
std::pair<bool, u32> get_format_convert_flags(GLenum format)
{
const auto texel_width = get_format_texel_width(format);
return { (texel_width > 1), texel_width };
}
bool formats_are_bitcast_compatible(GLenum format1, GLenum format2)
{
if (format1 == format2) [[likely]]
{
return true;
}
// Formats are compatible if the following conditions are met:
// 1. Texel sizes must match
// 2. Both formats require no transforms (basic memcpy) or...
// 3. Both formats have the same transform (e.g RG16_UNORM to RG16_SFLOAT, both are down and uploaded with a 2-byte byteswap)
if (format1 == GL_BGRA8 || format2 == GL_BGRA8)
{
return false;
}
if (get_format_texel_width(format1) != get_format_texel_width(format2))
{
return false;
}
const auto transform_a = get_format_convert_flags(format1);
const auto transform_b = get_format_convert_flags(format2);
if (transform_a.first == transform_b.first)
{
return !transform_a.first || (transform_a.second == transform_b.second);
}
return false;
}
bool formats_are_bitcast_compatible(const texture* texture1, const texture* texture2)
{
if (const u32 transfer_class = texture1->format_class() | texture2->format_class();
transfer_class > RSX_FORMAT_CLASS_COLOR)
{
// If any one of the two images is a depth format, the other must match exactly or bust
return (texture1->format_class() == texture2->format_class());
}
return formats_are_bitcast_compatible(static_cast<GLenum>(texture1->get_internal_format()), static_cast<GLenum>(texture2->get_internal_format()));
}
void copy_typeless(gl::command_context& cmd, texture * dst, const texture * src, const coord3u& dst_region, const coord3u& src_region)
{
const auto src_bpp = src->pitch() / src->width();
const auto dst_bpp = dst->pitch() / dst->width();
image_memory_requirements src_mem = { src_region.width * src_region.height, src_region.width * src_bpp * src_region.height, 0ull };
image_memory_requirements dst_mem = { dst_region.width * dst_region.height, dst_region.width * dst_bpp * dst_region.height, 0ull };
const auto& caps = gl::get_driver_caps();
auto pack_info = get_format_type(src);
auto unpack_info = get_format_type(dst);
// D32FS8 can be read back as D24S8 or D32S8X24. In case of the latter, double memory requirements
if (pack_info.type == GL_FLOAT_32_UNSIGNED_INT_24_8_REV)
{
src_mem.image_size_in_bytes *= 2;
}
if (unpack_info.type == GL_FLOAT_32_UNSIGNED_INT_24_8_REV)
{
dst_mem.image_size_in_bytes *= 2;
}
if (caps.ARB_compute_shader_supported) [[likely]]
{
bool skip_transform = false;
if ((src->aspect() | dst->aspect()) == gl::image_aspect::color)
{
skip_transform = (pack_info.format == unpack_info.format &&
pack_info.type == unpack_info.type &&
pack_info.swap_bytes == unpack_info.swap_bytes &&
pack_info.size == unpack_info.size);
}
if (skip_transform) [[likely]]
{
// Disable byteswap to make the transport operation passthrough
pack_info.swap_bytes = false;
unpack_info.swap_bytes = false;
}
u32 scratch_offset = 0;
const u64 min_storage_requirement = src_mem.image_size_in_bytes + dst_mem.image_size_in_bytes;
const u64 min_required_buffer_size = utils::align(min_storage_requirement, 256);
if (g_typeless_transfer_buffer.size() >= min_required_buffer_size) [[ likely ]]
{
scratch_offset = g_typeless_transfer_buffer.alloc(static_cast<u32>(min_storage_requirement), 256);
}
else
{
const auto new_size = std::max(min_required_buffer_size, g_typeless_transfer_buffer.size() + (64 * 0x100000));
g_typeless_transfer_buffer.create(gl::buffer::target::ssbo, new_size);
}
void* data_ptr = copy_image_to_buffer(cmd, pack_info, src, &g_typeless_transfer_buffer.get(), scratch_offset, 0, src_region, &src_mem);
copy_buffer_to_image(cmd, unpack_info, &g_typeless_transfer_buffer.get(), dst, data_ptr, 0, dst_region, &dst_mem);
// Not truly range-accurate, but should cover most of what we care about
g_typeless_transfer_buffer.push_barrier(scratch_offset, static_cast<u32>(min_storage_requirement));
// Cleanup
// NOTE: glBindBufferRange also binds the buffer to the old-school target.
// Unbind it to avoid glitching later
glBindBuffer(GL_SHADER_STORAGE_BUFFER, GL_NONE);
glBindBuffer(GL_PIXEL_PACK_BUFFER, GL_NONE);
glBindBuffer(GL_PIXEL_UNPACK_BUFFER, GL_NONE);
}
else
{
const u64 max_mem = std::max(src_mem.image_size_in_bytes, dst_mem.image_size_in_bytes);
if (max_mem > static_cast<u64>(g_typeless_transfer_buffer.size()))
{
g_typeless_transfer_buffer.remove();
g_typeless_transfer_buffer.create(buffer::target::pixel_pack, max_mem);
}
// Simplify pack/unpack information to something OpenGL can natively digest
auto remove_depth_transformation = [](const texture* tex, pixel_buffer_layout& pack_info)
{
if (tex->aspect() & image_aspect::depth)
{
switch (pack_info.type)
{
case GL_UNSIGNED_INT_24_8:
pack_info.swap_bytes = false;
break;
case GL_FLOAT_32_UNSIGNED_INT_24_8_REV:
pack_info.type = GL_UNSIGNED_INT_24_8;
pack_info.swap_bytes = false;
break;
case GL_FLOAT:
pack_info.type = GL_HALF_FLOAT;
break;
default: break;
}
}
};
remove_depth_transformation(src, pack_info);
remove_depth_transformation(dst, unpack_info);
// Attempt to compensate for the lack of compute shader modifiers
// If crossing the aspect boundary between color and depth
// and one image is depth, invert byteswap for the other one to compensate
const auto cross_aspect_test = (image_aspect::color | image_aspect::depth);
const auto test = (src->aspect() | dst->aspect()) & cross_aspect_test;
if (test == cross_aspect_test)
{
if (src->aspect() & image_aspect::depth)
{
// Source is depth, modify unpack rule
if (pack_info.size == 4 && unpack_info.size == 4)
{
unpack_info.swap_bytes = !unpack_info.swap_bytes;
}
}
else
{
// Dest is depth, modify pack rule
if (pack_info.size == 4 && unpack_info.size == 4)
{
pack_info.swap_bytes = !pack_info.swap_bytes;
}
}
}
// Start pack operation
pixel_pack_settings pack_settings{};
pack_settings.swap_bytes(pack_info.swap_bytes);
g_typeless_transfer_buffer.get().bind(buffer::target::pixel_pack);
src->copy_to(nullptr, static_cast<texture::format>(pack_info.format), static_cast<texture::type>(pack_info.type), 0, src_region, pack_settings);
glBindBuffer(GL_PIXEL_PACK_BUFFER, GL_NONE);
// Start unpack operation
pixel_unpack_settings unpack_settings{};
unpack_settings.swap_bytes(unpack_info.swap_bytes);
g_typeless_transfer_buffer.get().bind(buffer::target::pixel_unpack);
dst->copy_from(nullptr, static_cast<texture::format>(unpack_info.format), static_cast<texture::type>(unpack_info.type), 0, dst_region, unpack_settings);
glBindBuffer(GL_PIXEL_UNPACK_BUFFER, GL_NONE);
}
}
void copy_typeless(gl::command_context& cmd, texture* dst, const texture* src)
{
const coord3u src_area = { {}, src->size3D() };
const coord3u dst_area = { {}, dst->size3D() };
copy_typeless(cmd, dst, src, dst_area, src_area);
}
}