MIPS64: Remove unaligned memory access from art generated code
Unaligned memory access was caused by sd, ld, ldc1 and sdc1
instructions. Check if offset is unaligned and replace it
with two 32 bit memory accesses, if so.
Added assembler tests for new instructions, as well as assembler
tests for LoadFromOffset, LoadFpuFromOffset, StoreToOffset and
StoreFpuToOffset.
Change-Id: I0228a4a2ce6c801eeb5b46952b8330e14468deb3
diff --git a/compiler/optimizing/code_generator_mips64.cc b/compiler/optimizing/code_generator_mips64.cc
index 3c928de..cefcb95 100644
--- a/compiler/optimizing/code_generator_mips64.cc
+++ b/compiler/optimizing/code_generator_mips64.cc
@@ -106,7 +106,7 @@
}
#define __ down_cast<CodeGeneratorMIPS64*>(codegen)->GetAssembler()->
-#define QUICK_ENTRY_POINT(x) QUICK_ENTRYPOINT_OFFSET(kMips64WordSize, x).Int32Value()
+#define QUICK_ENTRY_POINT(x) QUICK_ENTRYPOINT_OFFSET(kMips64DoublewordSize, x).Int32Value()
class BoundsCheckSlowPathMIPS64 : public SlowPathCodeMIPS64 {
public:
@@ -437,7 +437,7 @@
#undef __
#define __ down_cast<Mips64Assembler*>(GetAssembler())->
-#define QUICK_ENTRY_POINT(x) QUICK_ENTRYPOINT_OFFSET(kMips64WordSize, x).Int32Value()
+#define QUICK_ENTRY_POINT(x) QUICK_ENTRYPOINT_OFFSET(kMips64DoublewordSize, x).Int32Value()
void CodeGeneratorMIPS64::Finalize(CodeAllocator* allocator) {
// Ensure that we fix up branches.
@@ -486,12 +486,12 @@
void ParallelMoveResolverMIPS64::RestoreScratch(int reg) {
// Pop reg
__ Ld(GpuRegister(reg), SP, 0);
- __ DecreaseFrameSize(kMips64WordSize);
+ __ DecreaseFrameSize(kMips64DoublewordSize);
}
void ParallelMoveResolverMIPS64::SpillScratch(int reg) {
// Push reg
- __ IncreaseFrameSize(kMips64WordSize);
+ __ IncreaseFrameSize(kMips64DoublewordSize);
__ Sd(GpuRegister(reg), SP, 0);
}
@@ -503,7 +503,7 @@
// automatically unspilled when the scratch scope object is destroyed).
ScratchRegisterScope ensure_scratch(this, TMP, V0, codegen_->GetNumberOfCoreRegisters());
// If V0 spills onto the stack, SP-relative offsets need to be adjusted.
- int stack_offset = ensure_scratch.IsSpilled() ? kMips64WordSize : 0;
+ int stack_offset = ensure_scratch.IsSpilled() ? kMips64DoublewordSize : 0;
__ LoadFromOffset(load_type,
GpuRegister(ensure_scratch.GetRegister()),
SP,
@@ -562,7 +562,7 @@
for (int i = arraysize(kCoreCalleeSaves) - 1; i >= 0; --i) {
GpuRegister reg = kCoreCalleeSaves[i];
if (allocated_registers_.ContainsCoreRegister(reg)) {
- ofs -= kMips64WordSize;
+ ofs -= kMips64DoublewordSize;
__ Sd(reg, SP, ofs);
__ cfi().RelOffset(DWARFReg(reg), ofs);
}
@@ -571,7 +571,7 @@
for (int i = arraysize(kFpuCalleeSaves) - 1; i >= 0; --i) {
FpuRegister reg = kFpuCalleeSaves[i];
if (allocated_registers_.ContainsFloatingPointRegister(reg)) {
- ofs -= kMips64WordSize;
+ ofs -= kMips64DoublewordSize;
__ Sdc1(reg, SP, ofs);
// TODO: __ cfi().RelOffset(DWARFReg(reg), ofs);
}
@@ -609,7 +609,7 @@
FpuRegister reg = kFpuCalleeSaves[i];
if (allocated_registers_.ContainsFloatingPointRegister(reg)) {
__ Ldc1(reg, SP, ofs);
- ofs += kMips64WordSize;
+ ofs += kMips64DoublewordSize;
// TODO: __ cfi().Restore(DWARFReg(reg));
}
}
@@ -618,7 +618,7 @@
GpuRegister reg = kCoreCalleeSaves[i];
if (allocated_registers_.ContainsCoreRegister(reg)) {
__ Ld(reg, SP, ofs);
- ofs += kMips64WordSize;
+ ofs += kMips64DoublewordSize;
__ cfi().Restore(DWARFReg(reg));
}
}
@@ -976,7 +976,7 @@
__ LoadFromOffset(kLoadDoubleword,
card,
TR,
- Thread::CardTableOffset<kMips64WordSize>().Int32Value());
+ Thread::CardTableOffset<kMips64DoublewordSize>().Int32Value());
__ Dsrl(temp, object, gc::accounting::CardTable::kCardShift);
__ Daddu(temp, card, temp);
__ Sb(card, temp, 0);
@@ -994,10 +994,11 @@
blocked_core_registers_[SP] = true;
blocked_core_registers_[RA] = true;
- // AT and TMP(T8) are used as temporary/scratch registers
- // (similar to how AT is used by MIPS assemblers).
+ // AT, TMP(T8) and TMP2(T3) are used as temporary/scratch
+ // registers (similar to how AT is used by MIPS assemblers).
blocked_core_registers_[AT] = true;
blocked_core_registers_[TMP] = true;
+ blocked_core_registers_[TMP2] = true;
blocked_fpu_registers_[FTMP] = true;
// Reserve suspend and thread registers.
@@ -1021,22 +1022,22 @@
size_t CodeGeneratorMIPS64::SaveCoreRegister(size_t stack_index, uint32_t reg_id) {
__ StoreToOffset(kStoreDoubleword, GpuRegister(reg_id), SP, stack_index);
- return kMips64WordSize;
+ return kMips64DoublewordSize;
}
size_t CodeGeneratorMIPS64::RestoreCoreRegister(size_t stack_index, uint32_t reg_id) {
__ LoadFromOffset(kLoadDoubleword, GpuRegister(reg_id), SP, stack_index);
- return kMips64WordSize;
+ return kMips64DoublewordSize;
}
size_t CodeGeneratorMIPS64::SaveFloatingPointRegister(size_t stack_index, uint32_t reg_id) {
__ StoreFpuToOffset(kStoreDoubleword, FpuRegister(reg_id), SP, stack_index);
- return kMips64WordSize;
+ return kMips64DoublewordSize;
}
size_t CodeGeneratorMIPS64::RestoreFloatingPointRegister(size_t stack_index, uint32_t reg_id) {
__ LoadFpuFromOffset(kLoadDoubleword, FpuRegister(reg_id), SP, stack_index);
- return kMips64WordSize;
+ return kMips64DoublewordSize;
}
void CodeGeneratorMIPS64::DumpCoreRegister(std::ostream& stream, int reg) const {
@@ -1051,7 +1052,7 @@
HInstruction* instruction,
uint32_t dex_pc,
SlowPathCode* slow_path) {
- InvokeRuntime(GetThreadOffset<kMips64WordSize>(entrypoint).Int32Value(),
+ InvokeRuntime(GetThreadOffset<kMips64DoublewordSize>(entrypoint).Int32Value(),
instruction,
dex_pc,
slow_path);
@@ -1091,7 +1092,7 @@
__ LoadFromOffset(kLoadUnsignedHalfword,
TMP,
TR,
- Thread::ThreadFlagsOffset<kMips64WordSize>().Int32Value());
+ Thread::ThreadFlagsOffset<kMips64DoublewordSize>().Int32Value());
if (successor == nullptr) {
__ Bnezc(TMP, slow_path->GetEntryLabel());
__ Bind(slow_path->GetReturnLabel());
@@ -3014,7 +3015,7 @@
invoke->GetImtIndex() % mirror::Class::kImtSize, kMips64PointerSize).Uint32Value();
Location receiver = invoke->GetLocations()->InAt(0);
uint32_t class_offset = mirror::Object::ClassOffset().Int32Value();
- Offset entry_point = ArtMethod::EntryPointFromQuickCompiledCodeOffset(kMips64WordSize);
+ Offset entry_point = ArtMethod::EntryPointFromQuickCompiledCodeOffset(kMips64DoublewordSize);
// Set the hidden argument.
__ LoadConst32(invoke->GetLocations()->GetTemp(1).AsRegister<GpuRegister>(),
@@ -3190,7 +3191,7 @@
T9,
callee_method.AsRegister<GpuRegister>(),
ArtMethod::EntryPointFromQuickCompiledCodeOffset(
- kMips64WordSize).Int32Value());
+ kMips64DoublewordSize).Int32Value());
// T9()
__ Jalr(T9);
__ Nop();
@@ -3228,7 +3229,7 @@
size_t method_offset = mirror::Class::EmbeddedVTableEntryOffset(
invoke->GetVTableIndex(), kMips64PointerSize).SizeValue();
uint32_t class_offset = mirror::Object::ClassOffset().Int32Value();
- Offset entry_point = ArtMethod::EntryPointFromQuickCompiledCodeOffset(kMips64WordSize);
+ Offset entry_point = ArtMethod::EntryPointFromQuickCompiledCodeOffset(kMips64DoublewordSize);
// temp = object->GetClass();
__ LoadFromOffset(kLoadUnsignedWord, temp, receiver, class_offset);
@@ -3306,7 +3307,7 @@
}
static int32_t GetExceptionTlsOffset() {
- return Thread::ExceptionOffset<kMips64WordSize>().Int32Value();
+ return Thread::ExceptionOffset<kMips64DoublewordSize>().Int32Value();
}
void LocationsBuilderMIPS64::VisitLoadException(HLoadException* load) {
@@ -3546,7 +3547,8 @@
if (instruction->IsStringAlloc()) {
// String is allocated through StringFactory. Call NewEmptyString entry point.
GpuRegister temp = instruction->GetLocations()->GetTemp(0).AsRegister<GpuRegister>();
- MemberOffset code_offset = ArtMethod::EntryPointFromQuickCompiledCodeOffset(kMips64WordSize);
+ MemberOffset code_offset =
+ ArtMethod::EntryPointFromQuickCompiledCodeOffset(kMips64DoublewordSize);
__ LoadFromOffset(kLoadDoubleword, temp, TR, QUICK_ENTRY_POINT(pNewEmptyString));
__ LoadFromOffset(kLoadDoubleword, T9, temp, code_offset.Int32Value());
__ Jalr(T9);