2018-06-10 22:02:33 +00:00
|
|
|
// Copyright 2018 yuzu Emulator Project
|
|
|
|
// Licensed under GPLv2 or any later version
|
|
|
|
// Refer to the license.txt file included.
|
|
|
|
|
2018-11-06 20:26:27 +00:00
|
|
|
#include "core/core.h"
|
2018-06-10 22:02:33 +00:00
|
|
|
#include "core/memory.h"
|
2018-11-06 20:26:27 +00:00
|
|
|
#include "video_core/engines/maxwell_3d.h"
|
2018-06-10 22:02:33 +00:00
|
|
|
#include "video_core/engines/maxwell_dma.h"
|
2018-10-18 01:29:10 +00:00
|
|
|
#include "video_core/rasterizer_interface.h"
|
2018-06-10 22:02:33 +00:00
|
|
|
#include "video_core/textures/decoders.h"
|
|
|
|
|
2018-10-20 19:58:06 +00:00
|
|
|
namespace Tegra::Engines {
|
2018-06-10 22:02:33 +00:00
|
|
|
|
2018-10-18 01:29:10 +00:00
|
|
|
MaxwellDMA::MaxwellDMA(VideoCore::RasterizerInterface& rasterizer, MemoryManager& memory_manager)
|
|
|
|
: memory_manager(memory_manager), rasterizer{rasterizer} {}
|
2018-06-10 22:02:33 +00:00
|
|
|
|
|
|
|
void MaxwellDMA::WriteReg(u32 method, u32 value) {
|
|
|
|
ASSERT_MSG(method < Regs::NUM_REGS,
|
|
|
|
"Invalid MaxwellDMA register, increase the size of the Regs structure");
|
|
|
|
|
|
|
|
regs.reg_array[method] = value;
|
|
|
|
|
|
|
|
#define MAXWELLDMA_REG_INDEX(field_name) \
|
|
|
|
(offsetof(Tegra::Engines::MaxwellDMA::Regs, field_name) / sizeof(u32))
|
|
|
|
|
|
|
|
switch (method) {
|
|
|
|
case MAXWELLDMA_REG_INDEX(exec): {
|
|
|
|
HandleCopy();
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
#undef MAXWELLDMA_REG_INDEX
|
|
|
|
}
|
|
|
|
|
|
|
|
void MaxwellDMA::HandleCopy() {
|
2018-07-02 16:13:26 +00:00
|
|
|
LOG_WARNING(HW_GPU, "Requested a DMA copy");
|
2018-06-10 22:02:33 +00:00
|
|
|
|
|
|
|
const GPUVAddr source = regs.src_address.Address();
|
|
|
|
const GPUVAddr dest = regs.dst_address.Address();
|
|
|
|
|
|
|
|
const VAddr source_cpu = *memory_manager.GpuToCpuAddress(source);
|
|
|
|
const VAddr dest_cpu = *memory_manager.GpuToCpuAddress(dest);
|
|
|
|
|
|
|
|
// TODO(Subv): Perform more research and implement all features of this engine.
|
|
|
|
ASSERT(regs.exec.enable_swizzle == 0);
|
|
|
|
ASSERT(regs.exec.query_mode == Regs::QueryMode::None);
|
|
|
|
ASSERT(regs.exec.query_intr == Regs::QueryIntr::None);
|
|
|
|
ASSERT(regs.exec.copy_mode == Regs::CopyMode::Unk2);
|
|
|
|
ASSERT(regs.dst_params.pos_x == 0);
|
|
|
|
ASSERT(regs.dst_params.pos_y == 0);
|
2018-07-02 14:46:33 +00:00
|
|
|
|
2018-10-18 01:29:10 +00:00
|
|
|
if (!regs.exec.is_dst_linear && !regs.exec.is_src_linear) {
|
|
|
|
// If both the source and the destination are in block layout, assert.
|
|
|
|
UNREACHABLE_MSG("Tiled->Tiled DMA transfers are not yet implemented");
|
|
|
|
return;
|
|
|
|
}
|
2018-09-08 21:02:16 +00:00
|
|
|
|
2018-11-06 20:26:27 +00:00
|
|
|
// All copies here update the main memory, so mark all rasterizer states as invalid.
|
|
|
|
Core::System::GetInstance().GPU().Maxwell3D().dirty_flags.OnMemoryWrite();
|
|
|
|
|
2018-10-18 01:29:10 +00:00
|
|
|
if (regs.exec.is_dst_linear && regs.exec.is_src_linear) {
|
2018-09-08 21:02:16 +00:00
|
|
|
// When the enable_2d bit is disabled, the copy is performed as if we were copying a 1D
|
2018-10-18 01:29:10 +00:00
|
|
|
// buffer of length `x_count`, otherwise we copy a 2D image of dimensions (x_count,
|
|
|
|
// y_count).
|
|
|
|
if (!regs.exec.enable_2d) {
|
|
|
|
Memory::CopyBlock(dest_cpu, source_cpu, regs.x_count);
|
|
|
|
return;
|
2018-09-08 21:02:16 +00:00
|
|
|
}
|
|
|
|
|
2018-10-18 01:29:10 +00:00
|
|
|
// If both the source and the destination are in linear layout, perform a line-by-line
|
|
|
|
// copy. We're going to take a subrect of size (x_count, y_count) from the source
|
|
|
|
// rectangle. There is no need to manually flush/invalidate the regions because
|
|
|
|
// CopyBlock does that for us.
|
|
|
|
for (u32 line = 0; line < regs.y_count; ++line) {
|
|
|
|
const VAddr source_line = source_cpu + line * regs.src_pitch;
|
|
|
|
const VAddr dest_line = dest_cpu + line * regs.dst_pitch;
|
|
|
|
Memory::CopyBlock(dest_line, source_line, regs.x_count);
|
|
|
|
}
|
2018-07-02 14:46:33 +00:00
|
|
|
return;
|
|
|
|
}
|
2018-06-10 22:02:33 +00:00
|
|
|
|
2018-09-08 21:02:16 +00:00
|
|
|
ASSERT(regs.exec.enable_2d == 1);
|
2018-10-18 01:29:10 +00:00
|
|
|
|
2018-10-20 19:55:58 +00:00
|
|
|
const std::size_t copy_size = regs.x_count * regs.y_count;
|
2018-10-18 01:29:10 +00:00
|
|
|
|
2018-10-20 19:54:43 +00:00
|
|
|
const auto FlushAndInvalidate = [&](u32 src_size, u64 dst_size) {
|
2018-10-18 01:29:10 +00:00
|
|
|
// TODO(Subv): For now, manually flush the regions until we implement GPU-accelerated
|
|
|
|
// copying.
|
|
|
|
rasterizer.FlushRegion(source_cpu, src_size);
|
|
|
|
|
|
|
|
// We have to invalidate the destination region to evict any outdated surfaces from the
|
|
|
|
// cache. We do this before actually writing the new data because the destination address
|
|
|
|
// might contain a dirty surface that will have to be written back to memory.
|
|
|
|
rasterizer.InvalidateRegion(dest_cpu, dst_size);
|
|
|
|
};
|
|
|
|
|
2018-06-10 22:02:33 +00:00
|
|
|
if (regs.exec.is_dst_linear && !regs.exec.is_src_linear) {
|
2018-10-18 01:29:10 +00:00
|
|
|
ASSERT(regs.src_params.size_z == 1);
|
2018-06-10 22:02:33 +00:00
|
|
|
// If the input is tiled and the output is linear, deswizzle the input and copy it over.
|
2018-10-18 01:29:10 +00:00
|
|
|
|
2018-10-20 19:55:58 +00:00
|
|
|
const u32 src_bytes_per_pixel = regs.src_pitch / regs.src_params.size_x;
|
2018-10-18 01:29:10 +00:00
|
|
|
|
|
|
|
FlushAndInvalidate(regs.src_pitch * regs.src_params.size_y,
|
|
|
|
copy_size * src_bytes_per_pixel);
|
|
|
|
|
|
|
|
Texture::UnswizzleSubrect(regs.x_count, regs.y_count, regs.dst_pitch,
|
|
|
|
regs.src_params.size_x, src_bytes_per_pixel, source_cpu, dest_cpu,
|
|
|
|
regs.src_params.BlockHeight(), regs.src_params.pos_x,
|
|
|
|
regs.src_params.pos_y);
|
2018-06-10 22:02:33 +00:00
|
|
|
} else {
|
2018-10-18 01:29:10 +00:00
|
|
|
ASSERT(regs.dst_params.size_z == 1);
|
|
|
|
ASSERT(regs.src_pitch == regs.x_count);
|
|
|
|
|
2018-10-20 19:55:58 +00:00
|
|
|
const u32 src_bpp = regs.src_pitch / regs.x_count;
|
2018-10-18 01:29:10 +00:00
|
|
|
|
|
|
|
FlushAndInvalidate(regs.src_pitch * regs.y_count,
|
|
|
|
regs.dst_params.size_x * regs.dst_params.size_y * src_bpp);
|
|
|
|
|
2018-06-10 22:02:33 +00:00
|
|
|
// If the input is linear and the output is tiled, swizzle the input and copy it over.
|
2018-10-18 01:29:10 +00:00
|
|
|
Texture::SwizzleSubrect(regs.x_count, regs.y_count, regs.src_pitch, regs.dst_params.size_x,
|
|
|
|
src_bpp, dest_cpu, source_cpu, regs.dst_params.BlockHeight());
|
2018-06-10 22:02:33 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2018-10-20 19:58:06 +00:00
|
|
|
} // namespace Tegra::Engines
|