CPU: Preparations for AArch64 recompiler

src/core/CMakeLists.txt

@@ -90,6 +90,13 @@ if(${CPU_ARCH} STREQUAL "x64")
     cpu_recompiler_code_generator_x64.cpp
   )
   message("Building x64 recompiler")
+elseif(${CPU_ARCH} STREQUAL "aarch64")
+  target_compile_definitions(core PRIVATE "WITH_RECOMPILER=1")
+  target_sources(core PRIVATE ${RECOMPILER_SRCS}
+    cpu_recompiler_code_generator_aarch64.cpp
+  )
+  target_link_libraries(core PRIVATE vixl)
+  message("Building AArch64 recompiler")
 else()
   message("Not building recompiler")
 endif()

src/core/cpu_recompiler_code_generator.cpp

@@ -5,16 +5,6 @@ Log_SetChannel(CPU::Recompiler);
 
 namespace CPU::Recompiler {
 
-CodeGenerator::CodeGenerator(Core* cpu, JitCodeBuffer* code_buffer, const ASMFunctions& asm_functions)
-  : m_cpu(cpu), m_code_buffer(code_buffer), m_asm_functions(asm_functions), m_register_cache(*this),
-    m_near_emitter(code_buffer->GetFreeCodeSpace(), code_buffer->GetFreeCodePointer()),
-    m_far_emitter(code_buffer->GetFreeFarCodeSpace(), code_buffer->GetFreeFarCodePointer()), m_emit(&m_near_emitter)
-{
-  InitHostRegs();
-}
-
-CodeGenerator::~CodeGenerator() = default;
-
 u32 CodeGenerator::CalculateRegisterOffset(Reg reg)
 {
   return uint32(offsetof(Core, m_regs.r[0]) + (static_cast<u32>(reg) * sizeof(u32)));
@@ -255,6 +245,16 @@ Value CodeGenerator::ConvertValueSize(const Value& value, RegSize size, bool sig
   return new_value;
 }
 
+Value CodeGenerator::GetValueInHostRegister(const Value& value)
+{
+  if (value.IsInHostRegister())
+    return Value(value.regcache, value.host_reg, value.size, ValueFlags::Valid | ValueFlags::InHostRegister);
+
+  Value new_value = m_register_cache.AllocateScratch(value.size);
+  EmitCopyValue(new_value.host_reg, value);
+  return new_value;
+}
+
 void CodeGenerator::ConvertValueSizeInPlace(Value* value, RegSize size, bool sign_extend)
 {
   DebugAssert(value->size != size);

src/core/cpu_recompiler_code_generator.h

@@ -156,6 +156,8 @@ private:
   Value ConvertValueSize(const Value& value, RegSize size, bool sign_extend);
   void ConvertValueSizeInPlace(Value* value, RegSize size, bool sign_extend);
 
+  Value GetValueInHostRegister(const Value& value);
+
   void SwitchToFarCode();
   void SwitchToNearCode();
   void* GetCurrentNearCodePointer() const;

src/core/cpu_recompiler_code_generator_generic.cpp

@@ -2,10 +2,6 @@
 
 namespace CPU::Recompiler {
 
-#if !defined(Y_CPU_X64)
-void CodeGenerator::AlignCodeBuffer(JitCodeBuffer* code_buffer) {}
-#endif
-
 void CodeGenerator::EmitLoadGuestRegister(HostReg host_reg, Reg guest_reg)
 {
   EmitLoadCPUStructField(host_reg, RegSize_32, CalculateRegisterOffset(guest_reg));
@@ -25,15 +21,4 @@ void CodeGenerator::EmitStoreInterpreterLoadDelay(Reg reg, const Value& value)
   m_load_delay_dirty = true;
 }
 
-#if !defined(Y_CPU_X64)
-void CodeGenerator::EmitBranch(Condition condition, Reg lr_reg, bool relative, const Value& branch_address)
-{
-  Panic("Not implemented");
-}
-void CodeGenerator::EmitRaiseException(Exception excode, Condition condition /* = Condition::Always */)
-{
-  Panic("Not implemented");
-}
-#endif
-
-} // namespace CPU::Recompiler
+} // namespace CPU::Recompiler

src/core/cpu_recompiler_code_generator_x64.cpp

@@ -74,6 +74,16 @@ static const Xbyak::Reg64 GetCPUPtrReg()
   return GetHostReg64(RCPUPTR);
 }
 
+CodeGenerator::CodeGenerator(Core* cpu, JitCodeBuffer* code_buffer, const ASMFunctions& asm_functions)
+  : m_cpu(cpu), m_code_buffer(code_buffer), m_asm_functions(asm_functions), m_register_cache(*this),
+    m_near_emitter(code_buffer->GetFreeCodeSpace(), code_buffer->GetFreeCodePointer()),
+    m_far_emitter(code_buffer->GetFreeFarCodeSpace(), code_buffer->GetFreeFarCodePointer()), m_emit(&m_near_emitter)
+{
+  InitHostRegs();
+}
+
+CodeGenerator::~CodeGenerator() = default;
+
 const char* CodeGenerator::GetHostRegName(HostReg reg, RegSize size /*= HostPointerSize*/)
 {
   static constexpr std::array<const char*, HostReg_Count> reg8_names = {

src/core/cpu_recompiler_types.h

@@ -4,6 +4,9 @@
 #if defined(Y_CPU_X64)
 #define XBYAK_NO_OP_NAMES 1
 #include "xbyak.h"
+#elif defined(Y_CPU_AARCH64)
+#include <vixl/aarch64/constants-aarch64.h>
+#include <vixl/aarch64/macro-assembler-aarch64.h>
 #endif
 
 namespace CPU {
@@ -63,6 +66,24 @@ constexpr u32 MAX_FAR_HOST_BYTES_PER_INSTRUCTION = 128;
 // Are shifts implicitly masked to 0..31?
 constexpr bool SHIFTS_ARE_IMPLICITLY_MASKED = true;
 
+#elif defined(Y_CPU_AARCH64)
+
+using HostReg = unsigned;
+using CodeEmitter = vixl::aarch64::MacroAssembler;
+enum : u32
+{
+    HostReg_Count = vixl::aarch64::kNumberOfRegisters
+};
+constexpr HostReg HostReg_Invalid = static_cast<HostReg>(HostReg_Count);
+constexpr RegSize HostPointerSize = RegSize_64;
+
+// A reasonable "maximum" number of bytes per instruction.
+constexpr u32 MAX_NEAR_HOST_BYTES_PER_INSTRUCTION = 64;
+constexpr u32 MAX_FAR_HOST_BYTES_PER_INSTRUCTION = 128;
+
+// Are shifts implicitly masked to 0..31?
+constexpr bool SHIFTS_ARE_IMPLICITLY_MASKED = true;
+
 #else
 
 using HostReg = int;
@@ -84,4 +105,4 @@ constexpr bool SHIFTS_ARE_IMPLICITLY_MASKED = false;
 
 } // namespace Recompiler
 
-} // namespace CPU
+} // namespace CPU