Skip to content

Commit

Permalink
SPU LLVM: regain some efficiency
Browse files Browse the repository at this point in the history 
Avoid returns from the recompiler gateway, favoring tail calls.
This may address performance degradation after RPCS3#5923.
  • Loading branch information
@Nekotekina
Nekotekina committed May 12, 2019
1 parent 2578916 commit 5bf921a
Showing 1 changed file with 72 additions and 23 deletions.
95 changes: 72 additions & 23 deletions rpcs3/Emu/Cell/SPURecompiler.cpp
View file
Expand Up @@ -3153,6 +3153,9 @@ class spu_llvm_recompiler : public spu_recompiler_base, public cpu_translator
// Function for check_state execution
llvm::Function* m_test_state{};

// Chunk for external tail call (dispatch)
llvm::Function* m_dispatch{};

llvm::MDNode* m_md_unlikely;
llvm::MDNode* m_md_likely;

Expand Down Expand Up @@ -3219,11 +3222,15 @@ class spu_llvm_recompiler : public spu_recompiler_base, public cpu_translator
}

// Chunk function type
// 0. Result (void)
// 0. Result (tail call target)
// 1. Thread context
// 2. Local storage pointer
// 3.
#ifdef _WIN32
const auto chunk_type = get_ftype<u8*, u8*, u8*, u32>();
#else
const auto chunk_type = get_ftype<void, u8*, u8*, u32>();
#endif

// Get function chunk name
const std::string name = fmt::format("spu-chunk-0x%05x", addr);
Expand Down Expand Up @@ -3273,9 +3280,18 @@ class spu_llvm_recompiler : public spu_recompiler_base, public cpu_translator
void tail_chunk(llvm::Value* chunk, llvm::Value* base_pc = nullptr)
{
auto call = m_ir->CreateCall(chunk, {m_thread, m_lsptr, base_pc ? base_pc : m_base_pc});
call->setCallingConv(m_finfo ? m_finfo->chunk->getCallingConv() : llvm::cast<llvm::Function>(chunk)->getCallingConv());
auto func = m_finfo ? m_finfo->chunk : llvm::cast<llvm::Function>(chunk);
call->setCallingConv(func->getCallingConv());
call->setTailCall();
m_ir->CreateRetVoid();

if (func->getReturnType() == get_type<void>())
{
m_ir->CreateRetVoid();
}
else
{
m_ir->CreateRet(call);
}
}

// Call the real function
Expand Down Expand Up @@ -3451,7 +3467,7 @@ class spu_llvm_recompiler : public spu_recompiler_base, public cpu_translator
const auto result = llvm::BasicBlock::Create(m_context, "", m_function);
m_ir->SetInsertPoint(result);
update_pc(target);
m_ir->CreateRetVoid();
tail_chunk(m_dispatch);
m_ir->SetInsertPoint(cblock);
return result;
}
Expand Down Expand Up @@ -4103,7 +4119,7 @@ class spu_llvm_recompiler : public spu_recompiler_base, public cpu_translator
m_fake_global1 = new llvm::GlobalVariable(*m_module, get_type<bool>(), false, llvm::GlobalValue::InternalLinkage, m_ir->getFalse());

// Add entry function (contains only state/code check)
const auto main_func = llvm::cast<llvm::Function>(m_module->getOrInsertFunction(hash, get_ftype<void, u8*, u8*, u8*>()).getCallee());
const auto main_func = llvm::cast<llvm::Function>(m_module->getOrInsertFunction(hash, get_ftype<void, u8*, u8*, u64>()).getCallee());
const auto main_arg2 = &*(main_func->arg_begin() + 2);
main_func->setCallingConv(CallingConv::GHC);
set_function(main_func);
Expand Down Expand Up @@ -4235,7 +4251,30 @@ class spu_llvm_recompiler : public spu_recompiler_base, public cpu_translator

// Call the entry function chunk
const auto entry_chunk = add_function(m_pos);
tail_chunk(entry_chunk->chunk);
const auto entry_call = m_ir->CreateCall(entry_chunk->chunk, {m_thread, m_lsptr, m_base_pc});
entry_call->setCallingConv(entry_chunk->chunk->getCallingConv());

#ifdef _WIN32
// TODO: fix this mess
const auto dispatcher = m_ir->CreateIntToPtr(m_ir->getInt64((u64)+spu_runtime::g_dispatcher), get_type<u8**>());
#else
const auto dispatcher = new llvm::GlobalVariable(*m_module, get_type<u8*>(), true, GlobalValue::ExternalLinkage, nullptr, "spu_dispatcher");
m_engine->addGlobalMapping("spu_dispatcher", (u64)+spu_runtime::g_dispatcher);
#endif

// Proceed to the next code
if (entry_chunk->chunk->getReturnType() != get_type<void>())
{
const auto next_call = m_ir->CreateCall(m_ir->CreateBitCast(entry_call, main_func->getType()), {m_thread, m_lsptr, m_ir->getInt64(0)});
next_call->setCallingConv(main_func->getCallingConv());
next_call->setTailCall();
}
else
{
entry_call->setTailCall();
}

m_ir->CreateRetVoid();

m_ir->SetInsertPoint(label_stop);
m_ir->CreateRetVoid();
Expand All @@ -4246,7 +4285,7 @@ class spu_llvm_recompiler : public spu_recompiler_base, public cpu_translator
{
const auto pbfail = spu_ptr<u64>(&spu_thread::block_failure);
m_ir->CreateStore(m_ir->CreateAdd(m_ir->CreateLoad(pbfail), m_ir->getInt64(1)), pbfail);
const auto dispci = call("spu_dispatch", spu_runtime::tr_dispatch, m_thread, m_ir->getInt32(0), main_arg2);
const auto dispci = call("spu_dispatch", spu_runtime::tr_dispatch, m_thread, m_lsptr, main_arg2);
dispci->setCallingConv(CallingConv::GHC);
dispci->setTailCall();
m_ir->CreateRetVoid();
Expand All @@ -4256,6 +4295,24 @@ class spu_llvm_recompiler : public spu_recompiler_base, public cpu_translator
m_ir->CreateUnreachable();
}

m_dispatch = cast<Function>(module->getOrInsertFunction("spu-null", entry_chunk->chunk->getFunctionType()).getCallee());
m_dispatch->setLinkage(llvm::GlobalValue::InternalLinkage);
m_dispatch->setCallingConv(entry_chunk->chunk->getCallingConv());
set_function(m_dispatch);

if (entry_chunk->chunk->getReturnType() == get_type<void>())
{
const auto next_func = m_ir->CreateLoad(dispatcher);
const auto next_call = m_ir->CreateCall(m_ir->CreateBitCast(next_func, main_func->getType()), {m_thread, m_lsptr, m_ir->getInt64(0)});
next_call->setCallingConv(main_func->getCallingConv());
next_call->setTailCall();
m_ir->CreateRetVoid();
}
else
{
m_ir->CreateRet(m_ir->CreateLoad(dispatcher));
}

// Longjmp analogue (restore saved host thread's stack pointer)
const auto escape = llvm::cast<llvm::Function>(m_module->getOrInsertFunction("spu_escape", get_ftype<void, u8*>()).getCallee());
escape->setLinkage(GlobalValue::InternalLinkage);
Expand Down Expand Up @@ -4475,19 +4532,13 @@ class spu_llvm_recompiler : public spu_recompiler_base, public cpu_translator
std::vector<llvm::Constant*> chunks;
chunks.reserve(m_size / 4);

const auto null = cast<Function>(module->getOrInsertFunction("spu-null", entry_chunk->chunk->getFunctionType()).getCallee());
null->setLinkage(llvm::GlobalValue::InternalLinkage);
null->setCallingConv(entry_chunk->chunk->getCallingConv());
set_function(null);
m_ir->CreateRetVoid();

for (u32 i = start; i < end; i += 4)
{
const auto found = m_functions.find(i);

if (found == m_functions.end())
{
chunks.push_back(null);
chunks.push_back(m_dispatch);
continue;
}

Expand Down Expand Up @@ -5018,7 +5069,6 @@ class spu_llvm_recompiler : public spu_recompiler_base, public cpu_translator
m_block->block_end = m_ir->GetInsertBlock();
update_pc();
call("spu_unknown", &exec_unk, m_thread, m_ir->getInt32(op_unk.opcode));
m_ir->CreateRetVoid();
}

static bool exec_stop(spu_thread* _spu, u32 code)
Expand Down Expand Up @@ -5054,12 +5104,11 @@ class spu_llvm_recompiler : public spu_recompiler_base, public cpu_translator
{
m_block->block_end = m_ir->GetInsertBlock();
update_pc(m_pos + 4);
m_ir->CreateRetVoid();
}
else
{
check_state(m_pos + 4);
tail_chunk(m_dispatch);
return;
}

check_state(m_pos + 4);
}

void STOPD(spu_opcode_t op) //
Expand Down Expand Up @@ -5847,7 +5896,7 @@ class spu_llvm_recompiler : public spu_recompiler_base, public cpu_translator
{
m_block->block_end = m_ir->GetInsertBlock();
update_pc(m_pos + 4);
m_ir->CreateRetVoid();
tail_chunk(m_dispatch);
}
}

Expand Down Expand Up @@ -7675,7 +7724,7 @@ class spu_llvm_recompiler : public spu_recompiler_base, public cpu_translator
m_ir->SetInsertPoint(fail);
}

m_ir->CreateRetVoid();
tail_chunk(m_dispatch);
m_ir->SetInsertPoint(cblock);
return result;
}
Expand Down Expand Up @@ -7814,7 +7863,7 @@ class spu_llvm_recompiler : public spu_recompiler_base, public cpu_translator
}
else
{
m_ir->CreateRetVoid();
tail_chunk(m_dispatch);
}
}
else
Expand Down

0 comments on commit 5bf921a

Please sign in to comment.