use Sleep(0) instead of SwitchToThread, should waste less power and help the os with scheduling.
PM4 buffer handling made a virtual member of commandprocessor, place the implementation/declaration into reusable macro files. this is probably the biggest boost here. Optimized SET_CONSTANT/ LOAD_CONSTANT pm4 ops based on the register range they start writing at, this was also a nice boost Expose X64 extension flags to code outside of x64 backend, so we can detect and use things like avx512, xop, avx2, etc in normal code Add freelists for HIR structures to try to reduce the number of last level cache misses during optimization (currently disabled... fixme later) Analyzed PGO feedback and reordered branches, uninlined functions, moved code out into different functions based on info from it in the PM4 functions, this gave like a 2% boost at best. Added support for the db16cyc opcode, which is used often in xb360 spinlocks. before it was just being translated to nop, now on x64 we translate it to _mm_pause but may change that in the future to reduce cpu time wasted texture util - all our divisors were powers of 2, instead we look up a shift. this made texture scaling slightly faster, more so on intel processors which seem to be worse at int divs. GetGuestTextureLayout is now a little faster, although it is still one of the heaviest functions in the emulator when scaling is on. xe_unlikely_mutex was not a good choice for the guest clock lock, (running theory) on intel processors another thread may take a significant time to update the clock? maybe because of the uint64 division? really not sure, but switched it to xe_mutex. This fixed audio stutter that i had introduced to 1 or 2 games, fixed performance on that n64 rare game with the monkeys. Took another crack at DMA implementation, another failure. Instead of passing as a parameter, keep the ringbuffer reader as the first member of commandprocessor so it can be accessed through this Added macro for noalias Applied noalias to Memory::LookupHeap. This reduced the size of the executable by 7 kb. Reworked kernel shim template, this shaved like 100kb off the exe and eliminated the indirect calls from the shim to the actual implementation. We still unconditionally generate string representations of kernel calls though :(, unless it is kHighFrequency Add nvapi extensions support, currently unused. Will use CPUVISIBLE memory at some point Inserted prefetches in a few places based on feedback from vtune. Add native implementation of SHA int8 if all elements are the same Vectorized comparisons for SetViewport, SetScissorRect Vectorized ranged comparisons for WriteRegister Add XE_MSVC_ASSUME Move FormatInfo::name out of the structure, instead look up the name in a different table. Debug related data and critical runtime data are best kept apart Templated UpdateSystemConstantValues based on ROV/RTV and primitive_polygonal Add ArchFloatMask functions, these are for storing the results of floating point comparisons without doing costly float->int pipeline transfers (vucomiss/setb) Use floatmasks in UpdateSystemConstantValues for checking if dirty, only transfer to int at end of function. Instead of dirty |= (x == y) in UpdateSystemConstantValues, now we do dirty_u32 |= (x^y). if any of them are not equal, dirty_u32 will be nz, else if theyre all equal it will be zero. This is more friendly to register renaming and the lack of dependencies on EFLAGS lets the compiler reorder better Add PrefetchSamplerParameters to D3D12TextureCache use PrefetchSamplerParameters in UpdateBindings to eliminate cache misses that vtune detected Add PrefetchTextureBinding to D3D12TextureCache Prefetch texture bindings to get rid of more misses vtune detected (more accesses out of order with random strides) Rewrote DMAC, still terrible though and have disabled it for now. Replace tiny memcmp of 6 U64 in render_target_cache with inline loop, msvc fails to make it a loop and instead does a thunk to their memcmp function, which is optimized for larger sizes PrefetchTextureBinding in AreActiveTextureSRVKeysUpToDate Replace memcmp calls for pipelinedescription with handwritten cmp Directly write some registers that dont have special handling in PM4 functions Changed EstimateMaxY to try to eliminate mispredictions that vtune was reporting, msvc ended up turning the changed code into a series of blends in ExecutePacketType3_EVENT_WRITE_EXT, instead of writing extents to an array on the stack and then doing xe_copy_and_swap_16 of the data to its dest, pre-swap each constant and then store those. msvc manages to unroll that into wider stores stop logging XE_SWAP every time we receive XE_SWAP, stop logging the start and end of each viz query Prefetch watch nodes in FireWatches based on feedback from vtune Removed dead code from texture_info.cc NOINLINE on GpuSwap, PGO builds did it so we should too.
This commit is contained in:
chss95cs@gmail.com
@@ -19,6 +19,7 @@
|
||||
#include <string>
|
||||
#include <vector>
|
||||
|
||||
#include "xenia/base/dma.h"
|
||||
#include "xenia/base/ring_buffer.h"
|
||||
#include "xenia/base/threading.h"
|
||||
#include "xenia/gpu/register_file.h"
|
||||
@@ -66,6 +67,11 @@ enum class GammaRampType {
|
||||
};
|
||||
|
||||
class CommandProcessor {
|
||||
protected:
|
||||
RingBuffer
|
||||
reader_; // chrispy: instead of having ringbuffer on stack, have it near
|
||||
// the start of the class so we can access it via rel8. This
|
||||
// also reduces the number of params we need to pass
|
||||
public:
|
||||
enum class SwapPostEffect {
|
||||
kNone,
|
||||
@@ -76,7 +82,7 @@ class CommandProcessor {
|
||||
CommandProcessor(GraphicsSystem* graphics_system,
|
||||
kernel::KernelState* kernel_state);
|
||||
virtual ~CommandProcessor();
|
||||
|
||||
dma::XeDMAC* GetDMAC() const { return dmac_; }
|
||||
uint32_t counter() const { return counter_; }
|
||||
void increment_counter() { counter_++; }
|
||||
|
||||
@@ -101,7 +107,7 @@ class CommandProcessor {
|
||||
// screen right in the beginning of 4D530AA4 is not a resolved render target,
|
||||
// for instance).
|
||||
virtual void IssueSwap(uint32_t frontbuffer_ptr, uint32_t frontbuffer_width,
|
||||
uint32_t frontbuffer_height) = 0;
|
||||
uint32_t frontbuffer_height) {}
|
||||
|
||||
// May be called not only from the command processor thread when the command
|
||||
// processor is paused, and the termination of this function may be explicitly
|
||||
@@ -153,7 +159,7 @@ class CommandProcessor {
|
||||
// rarely needed, most register writes have no special logic here
|
||||
XE_NOINLINE
|
||||
void HandleSpecialRegisterWrite(uint32_t index, uint32_t value);
|
||||
XE_FORCEINLINE
|
||||
|
||||
virtual void WriteRegister(uint32_t index, uint32_t value);
|
||||
|
||||
// mem has big-endian register values
|
||||
@@ -165,12 +171,53 @@ class CommandProcessor {
|
||||
virtual void WriteRegisterRangeFromRing(xe::RingBuffer* ring, uint32_t base,
|
||||
uint32_t num_registers);
|
||||
|
||||
XE_FORCEINLINE
|
||||
XE_NOINLINE
|
||||
virtual void WriteOneRegisterFromRing(
|
||||
xe::RingBuffer* ring, uint32_t base,
|
||||
uint32_t base,
|
||||
uint32_t
|
||||
num_times); // repeatedly write a value to one register, presumably a
|
||||
// register with special handling for writes
|
||||
|
||||
XE_FORCEINLINE
|
||||
void WriteALURangeFromRing(xe::RingBuffer* ring, uint32_t base,
|
||||
uint32_t num_times);
|
||||
|
||||
XE_FORCEINLINE
|
||||
void WriteFetchRangeFromRing(xe::RingBuffer* ring, uint32_t base,
|
||||
uint32_t num_times);
|
||||
|
||||
XE_FORCEINLINE
|
||||
void WriteBoolRangeFromRing(xe::RingBuffer* ring, uint32_t base,
|
||||
uint32_t num_times);
|
||||
|
||||
XE_FORCEINLINE
|
||||
void WriteLoopRangeFromRing(xe::RingBuffer* ring, uint32_t base,
|
||||
uint32_t num_times);
|
||||
|
||||
XE_FORCEINLINE
|
||||
void WriteREGISTERSRangeFromRing(xe::RingBuffer* ring, uint32_t base,
|
||||
uint32_t num_times);
|
||||
|
||||
XE_FORCEINLINE
|
||||
void WriteALURangeFromMem(uint32_t start_index, uint32_t* base,
|
||||
uint32_t num_registers);
|
||||
|
||||
XE_FORCEINLINE
|
||||
void WriteFetchRangeFromMem(uint32_t start_index, uint32_t* base,
|
||||
uint32_t num_registers);
|
||||
|
||||
XE_FORCEINLINE
|
||||
void WriteBoolRangeFromMem(uint32_t start_index, uint32_t* base,
|
||||
uint32_t num_registers);
|
||||
|
||||
XE_FORCEINLINE
|
||||
void WriteLoopRangeFromMem(uint32_t start_index, uint32_t* base,
|
||||
uint32_t num_registers);
|
||||
|
||||
XE_FORCEINLINE
|
||||
void WriteREGISTERSRangeFromMem(uint32_t start_index, uint32_t* base,
|
||||
uint32_t num_registers);
|
||||
|
||||
const reg::DC_LUT_30_COLOR* gamma_ramp_256_entry_table() const {
|
||||
return gamma_ramp_256_entry_table_;
|
||||
}
|
||||
@@ -186,75 +233,22 @@ class CommandProcessor {
|
||||
|
||||
uint32_t ExecutePrimaryBuffer(uint32_t start_index, uint32_t end_index);
|
||||
virtual void OnPrimaryBufferEnd() {}
|
||||
void ExecuteIndirectBuffer(uint32_t ptr, uint32_t length);
|
||||
bool ExecutePacket(RingBuffer* reader);
|
||||
bool ExecutePacketType0(RingBuffer* reader, uint32_t packet);
|
||||
bool ExecutePacketType1(RingBuffer* reader, uint32_t packet);
|
||||
bool ExecutePacketType2(RingBuffer* reader, uint32_t packet);
|
||||
bool ExecutePacketType3(RingBuffer* reader, uint32_t packet);
|
||||
bool ExecutePacketType3_ME_INIT(RingBuffer* reader, uint32_t packet,
|
||||
uint32_t count);
|
||||
bool ExecutePacketType3_NOP(RingBuffer* reader, uint32_t packet,
|
||||
uint32_t count);
|
||||
bool ExecutePacketType3_INTERRUPT(RingBuffer* reader, uint32_t packet,
|
||||
uint32_t count);
|
||||
bool ExecutePacketType3_XE_SWAP(RingBuffer* reader, uint32_t packet,
|
||||
uint32_t count);
|
||||
bool ExecutePacketType3_INDIRECT_BUFFER(RingBuffer* reader, uint32_t packet,
|
||||
uint32_t count);
|
||||
bool ExecutePacketType3_WAIT_REG_MEM(RingBuffer* reader, uint32_t packet,
|
||||
uint32_t count);
|
||||
bool ExecutePacketType3_REG_RMW(RingBuffer* reader, uint32_t packet,
|
||||
uint32_t count);
|
||||
bool ExecutePacketType3_REG_TO_MEM(RingBuffer* reader, uint32_t packet,
|
||||
uint32_t count);
|
||||
bool ExecutePacketType3_MEM_WRITE(RingBuffer* reader, uint32_t packet,
|
||||
uint32_t count);
|
||||
bool ExecutePacketType3_COND_WRITE(RingBuffer* reader, uint32_t packet,
|
||||
uint32_t count);
|
||||
bool ExecutePacketType3_EVENT_WRITE(RingBuffer* reader, uint32_t packet,
|
||||
uint32_t count);
|
||||
bool ExecutePacketType3_EVENT_WRITE_SHD(RingBuffer* reader, uint32_t packet,
|
||||
uint32_t count);
|
||||
bool ExecutePacketType3_EVENT_WRITE_EXT(RingBuffer* reader, uint32_t packet,
|
||||
uint32_t count);
|
||||
bool ExecutePacketType3_EVENT_WRITE_ZPD(RingBuffer* reader, uint32_t packet,
|
||||
uint32_t count);
|
||||
bool ExecutePacketType3Draw(RingBuffer* reader, uint32_t packet,
|
||||
const char* opcode_name,
|
||||
uint32_t viz_query_condition,
|
||||
uint32_t count_remaining);
|
||||
bool ExecutePacketType3_DRAW_INDX(RingBuffer* reader, uint32_t packet,
|
||||
uint32_t count);
|
||||
bool ExecutePacketType3_DRAW_INDX_2(RingBuffer* reader, uint32_t packet,
|
||||
uint32_t count);
|
||||
bool ExecutePacketType3_SET_CONSTANT(RingBuffer* reader, uint32_t packet,
|
||||
uint32_t count);
|
||||
bool ExecutePacketType3_SET_CONSTANT2(RingBuffer* reader, uint32_t packet,
|
||||
uint32_t count);
|
||||
bool ExecutePacketType3_LOAD_ALU_CONSTANT(RingBuffer* reader, uint32_t packet,
|
||||
uint32_t count);
|
||||
bool ExecutePacketType3_SET_SHADER_CONSTANTS(RingBuffer* reader,
|
||||
uint32_t packet, uint32_t count);
|
||||
bool ExecutePacketType3_IM_LOAD(RingBuffer* reader, uint32_t packet,
|
||||
uint32_t count);
|
||||
bool ExecutePacketType3_IM_LOAD_IMMEDIATE(RingBuffer* reader,
|
||||
|
||||
uint32_t packet, uint32_t count);
|
||||
bool ExecutePacketType3_INVALIDATE_STATE(RingBuffer* reader, uint32_t packet,
|
||||
uint32_t count);
|
||||
bool ExecutePacketType3_VIZ_QUERY(RingBuffer* reader, uint32_t packet,
|
||||
uint32_t count);
|
||||
#include "pm4_command_processor_declare.h"
|
||||
|
||||
virtual Shader* LoadShader(xenos::ShaderType shader_type,
|
||||
uint32_t guest_address,
|
||||
const uint32_t* host_address,
|
||||
uint32_t dword_count) = 0;
|
||||
uint32_t dword_count) {
|
||||
return nullptr;
|
||||
}
|
||||
|
||||
virtual bool IssueDraw(xenos::PrimitiveType prim_type, uint32_t index_count,
|
||||
IndexBufferInfo* index_buffer_info,
|
||||
bool major_mode_explicit) = 0;
|
||||
virtual bool IssueCopy() = 0;
|
||||
bool major_mode_explicit) {
|
||||
return false;
|
||||
}
|
||||
virtual bool IssueCopy() { return false; }
|
||||
|
||||
// "Actual" is for the command processor thread, to be read by the
|
||||
// implementations.
|
||||
@@ -267,7 +261,7 @@ class CommandProcessor {
|
||||
Memory* memory_ = nullptr;
|
||||
kernel::KernelState* kernel_state_ = nullptr;
|
||||
GraphicsSystem* graphics_system_ = nullptr;
|
||||
RegisterFile* register_file_ = nullptr;
|
||||
RegisterFile* XE_RESTRICT register_file_ = nullptr;
|
||||
|
||||
TraceWriter trace_writer_;
|
||||
enum class TraceState {
|
||||
@@ -316,6 +310,7 @@ class CommandProcessor {
|
||||
reg::DC_LUT_30_COLOR gamma_ramp_256_entry_table_[256] = {};
|
||||
reg::DC_LUT_PWL_DATA gamma_ramp_pwl_rgb_[128][3] = {};
|
||||
uint32_t gamma_ramp_rw_component_ = 0;
|
||||
dma::XeDMAC* dmac_ = nullptr;
|
||||
};
|
||||
|
||||
} // namespace gpu
|
||||
|
||||
Reference in New Issue
Block a user