1
0
mirror of https://github.com/RPCS3/rpcs3.git synced 2024-11-21 18:22:33 +01:00

Rewrite the way LLVM JIT does memory allocation to allow use of more than one contiguous segment (#6771)

This commit is contained in:
Rajko Stojadinovic 2019-10-28 23:01:07 +01:00 committed by Ivan
parent 42fc698186
commit b49b4c8096

View File

@ -1,4 +1,4 @@
#include "types.h"
#include "types.h"
#include "JIT.h"
#include "StrFmt.h"
#include "File.h"
@ -269,41 +269,237 @@ void asmjit::build_transaction_abort(asmjit::X86Assembler& c, unsigned char code
#include <sys/mman.h>
#endif
// Memory manager mutex
shared_mutex s_mutex;
// Size of virtual memory area reserved: 512 MB
static const u64 s_memory_size = 0x20000000;
// Try to reserve a portion of virtual memory in the first 2 GB address space beforehand, if possible.
static void* const s_memory = []() -> void*
class LLVMSegmentAllocator
{
llvm::InitializeNativeTarget();
llvm::InitializeNativeTargetAsmPrinter();
llvm::InitializeNativeTargetAsmParser();
LLVMLinkInMCJIT();
public:
// Size of virtual memory area reserved: default 512MB
static constexpr u32 DEFAULT_SEGMENT_SIZE = 0x20000000;
#ifdef MAP_32BIT
auto ptr = ::mmap(nullptr, s_memory_size, PROT_NONE, MAP_ANON | MAP_PRIVATE | MAP_32BIT, -1, 0);
if (ptr != MAP_FAILED)
return ptr;
#else
for (u64 addr = 0x10000000; addr <= 0x80000000 - s_memory_size; addr += 0x1000000)
LLVMSegmentAllocator()
{
if (auto ptr = utils::memory_reserve(s_memory_size, (void*)addr))
llvm::InitializeNativeTarget();
llvm::InitializeNativeTargetAsmPrinter();
llvm::InitializeNativeTargetAsmParser();
LLVMLinkInMCJIT();
// Try to reserve as much virtual memory in the first 2 GB address space beforehand, if possible.
Segment found_segs[16];
u32 num_segs = 0;
#ifdef MAP_32BIT
u64 max_size = 0x80000000u;
while (num_segs < 16)
{
return ptr;
auto ptr = ::mmap(nullptr, max_size, PROT_NONE, MAP_ANON | MAP_PRIVATE | MAP_32BIT, -1, 0);
if (ptr != MAP_FAILED)
found_segs[num_segs++] = Segment(ptr, u32(max_size));
else if (max_size > 0x1000000)
max_size -= 0x1000000;
else
break;
}
#else
u64 start_addr = 0x10000000;
while (num_segs < 16)
{
u64 max_addr = 0;
u64 max_size = 0x1000000;
for (u64 addr = start_addr; addr <= (0x80000000u - max_size); addr += 0x1000000)
{
for (auto curr_size = max_size; (0x80000000u - curr_size) >= addr; curr_size += 0x1000000)
{
if (auto ptr = utils::memory_reserve(curr_size, (void*)addr))
{
if (max_addr == 0 || max_size < curr_size)
{
max_addr = addr;
max_size = curr_size;
}
utils::memory_release(ptr, curr_size);
}
else
break;
}
}
if (max_addr == 0)
break;
if (auto ptr = utils::memory_reserve(max_size, (void*)max_addr))
found_segs[num_segs++] = Segment(ptr, u32(max_size));
start_addr = max_addr + max_size;
}
#endif
if (num_segs)
{
if (num_segs > 1)
{
m_segs.resize(num_segs);
for (u32 i = 0; i < num_segs; i++)
m_segs[i] = found_segs[i];
}
else
m_curr = found_segs[0];
return;
}
if (auto ptr = utils::memory_reserve(DEFAULT_SEGMENT_SIZE))
{
m_curr.addr = (u8*)ptr;
m_curr.size = DEFAULT_SEGMENT_SIZE;
m_curr.used = 0;
}
}
#endif
return utils::memory_reserve(s_memory_size);
}();
void* allocate(u32 size)
{
if (m_curr.remaining() >= size)
return m_curr.advance(size);
static void* s_next = s_memory;
if (reserve(size))
return m_curr.advance(size);
return nullptr;
}
bool reserve(u32 size)
{
if (size == 0)
return true;
store_curr();
u32 best_idx = UINT_MAX;
for (u32 i = 0, segs_size = (u32)m_segs.size(); i < segs_size; i++)
{
const auto seg_remaining = m_segs[i].remaining();
if (seg_remaining < size)
continue;
if (best_idx == UINT_MAX || m_segs[best_idx].remaining() > seg_remaining)
best_idx = i;
}
if (best_idx == UINT_MAX)
{
const auto size_to_reserve = (size > DEFAULT_SEGMENT_SIZE) ? ::align(size+4096, 4096) : DEFAULT_SEGMENT_SIZE;
if (auto ptr = utils::memory_reserve(size_to_reserve))
{
best_idx = (u32)m_segs.size();
m_segs.emplace_back(ptr, size_to_reserve);
}
else
return false;
}
const auto& best_seg = m_segs[best_idx];
if (best_seg.addr != m_curr.addr)
m_curr = best_seg;
return true;
}
std::pair<u64, u32> current_segment() const { return std::make_pair(u64(m_curr.addr), m_curr.size); }
std::pair<u64, u32> find_segment(u64 addr) const
{
for (const auto& seg: m_segs)
{
if (addr < (u64)seg.addr)
continue;
const auto end_addr = u64(seg.addr) + seg.size;
if (addr < end_addr)
return std::make_pair(u64(seg.addr), seg.size);
}
return std::make_pair(0, 0);
}
void reset()
{
if (!m_segs.size())
{
if (m_curr.addr != nullptr)
{
utils::memory_decommit(m_curr.addr, m_curr.size);
m_curr.used = 0;
}
return;
}
if (store_curr())
m_curr = Segment();
auto allocated_it = std::remove_if(m_segs.begin(), m_segs.end(), [](const Segment& seg) { return u64(seg.addr + seg.size) > 0x80000000u; });
if (allocated_it != m_segs.end())
{
for (auto it = allocated_it; it != m_segs.end(); ++it)
utils::memory_release(it->addr, it->size);
m_segs.erase(allocated_it, m_segs.end());
}
for (auto& seg : m_segs)
{
utils::memory_decommit(seg.addr, seg.size);
seg.used = 0;
}
}
private:
bool store_curr()
{
if (m_curr.addr != nullptr)
{
const auto wanted_addr = m_curr.addr;
auto existing_it = std::find_if(m_segs.begin(), m_segs.end(), [wanted_addr](const Segment& seg) { return seg.addr == wanted_addr; });
if (existing_it != m_segs.end())
existing_it->used = m_curr.used;
else
m_segs.push_back(m_curr);
return true;
}
return false;
}
struct Segment
{
Segment() {}
Segment(void* addr, u32 size) : addr((u8*)addr), size(size) {}
u8* addr = nullptr;
u32 size = 0;
u32 used = 0;
u32 remaining() const
{
if (size > used)
return size - used;
return 0;
}
void* advance(u32 offset)
{
const auto prev_used = used;
used += offset;
return &addr[prev_used];
}
};
Segment m_curr;
std::vector<Segment> m_segs;
};
// Memory manager mutex
static shared_mutex s_mutex;
// LLVM Memory allocator
static LLVMSegmentAllocator s_alloc;
#ifdef _WIN32
static std::deque<std::vector<RUNTIME_FUNCTION>> s_unwater;
static std::deque<std::pair<u64, std::vector<RUNTIME_FUNCTION>>> s_unwater;
static std::vector<std::vector<RUNTIME_FUNCTION>> s_unwind; // .pdata
#else
static std::deque<std::pair<u8*, std::size_t>> s_unfire;
@ -331,9 +527,7 @@ extern void jit_finalize()
s_unfire.clear();
#endif
utils::memory_decommit(s_memory, s_memory_size);
s_next = s_memory;
s_alloc.reset();
}
// Helper class
@ -376,17 +570,25 @@ struct MemoryManager : llvm::RTDyldMemoryManager
}
// Verify address for small code model
if ((u64)s_memory > 0x80000000 - s_memory_size ? (u64)addr - (u64)s_memory >= s_memory_size : addr >= 0x80000000)
const u64 code_start = u64(m_code_addr);
const s64 addr_diff = addr - code_start;
if (addr_diff < INT_MIN || addr_diff > INT_MAX)
{
// Lock memory manager
std::lock_guard lock(s_mutex);
// Allocate memory for trampolines
if (m_tramps)
{
const s64 tramps_diff = u64(m_tramps) - code_start;
if (tramps_diff < INT_MIN || tramps_diff > INT_MAX)
m_tramps = nullptr; //previously allocated trampoline section too far away now
}
if (!m_tramps)
{
m_tramps = reinterpret_cast<decltype(m_tramps)>(s_next);
utils::memory_commit(s_next, 4096, utils::protection::wx);
s_next = (u8*)((u64)s_next + 4096);
m_tramps = reinterpret_cast<decltype(m_tramps)>(s_alloc.allocate(4096));
utils::memory_commit(m_tramps, 4096, utils::protection::wx);
}
// Create a trampoline
@ -412,36 +614,57 @@ struct MemoryManager : llvm::RTDyldMemoryManager
return {addr, llvm::JITSymbolFlags::Exported};
}
u8* allocateCodeSection(std::uintptr_t size, uint align, uint sec_id, llvm::StringRef sec_name) override
bool needsToReserveAllocationSpace() override { return true; }
void reserveAllocationSpace(uintptr_t CodeSize, uint32_t CodeAlign, uintptr_t RODataSize, uint32_t RODataAlign, uintptr_t RWDataSize, uint32_t RWDataAlign) override
{
const u32 wanted_code_size = ::align(u32(CodeSize), std::min(4096u, CodeAlign));
const u32 wanted_rodata_size = ::align(u32(RODataSize), std::min(4096u, RODataAlign));
const u32 wanted_rwdata_size = ::align(u32(RWDataSize), std::min(4096u, RWDataAlign));
// Lock memory manager
std::lock_guard lock(s_mutex);
// Simple allocation
const u64 next = ::align((u64)s_next + size, 4096);
// Setup segment for current module if needed
s_alloc.reserve(wanted_code_size + wanted_rodata_size + wanted_rwdata_size);
}
if (next > (u64)s_memory + s_memory_size)
u8* allocateCodeSection(std::uintptr_t size, uint align, uint sec_id, llvm::StringRef sec_name) override
{
void* ptr = nullptr;
const u32 wanted_size = ::align(u32(size), 4096);
{
// Lock memory manager
std::lock_guard lock(s_mutex);
// Simple allocation
ptr = s_alloc.allocate(wanted_size);
}
if (ptr == nullptr)
{
LOG_FATAL(GENERAL, "LLVM: Out of memory (size=0x%llx, aligned 0x%x)", size, align);
return nullptr;
}
utils::memory_commit(ptr, size, utils::protection::wx);
m_code_addr = (u8*)ptr;
utils::memory_commit(s_next, size, utils::protection::wx);
m_code_addr = (u8*)s_next;
LOG_NOTICE(GENERAL, "LLVM: Code section %u '%s' allocated -> %p (size=0x%llx, aligned 0x%x)", sec_id, sec_name.data(), s_next, size, align);
return (u8*)std::exchange(s_next, (void*)next);
LOG_NOTICE(GENERAL, "LLVM: Code section %u '%s' allocated -> %p (size=0x%llx, aligned 0x%x)", sec_id, sec_name.data(), ptr, size, align);
return (u8*)ptr;
}
u8* allocateDataSection(std::uintptr_t size, uint align, uint sec_id, llvm::StringRef sec_name, bool is_ro) override
{
// Lock memory manager
std::lock_guard lock(s_mutex);
void* ptr = nullptr;
const u32 wanted_size = ::align(u32(size), 4096);
{
// Lock memory manager
std::lock_guard lock(s_mutex);
// Simple allocation
const u64 next = ::align((u64)s_next + size, 4096);
// Simple allocation
ptr = s_alloc.allocate(wanted_size);
}
if (next > (u64)s_memory + s_memory_size)
if (ptr == nullptr)
{
LOG_FATAL(GENERAL, "LLVM: Out of memory (size=0x%llx, aligned 0x%x)", size, align);
return nullptr;
@ -451,10 +674,10 @@ struct MemoryManager : llvm::RTDyldMemoryManager
{
}
utils::memory_commit(s_next, size);
utils::memory_commit(ptr, size);
LOG_NOTICE(GENERAL, "LLVM: Data section %u '%s' allocated -> %p (size=0x%llx, aligned 0x%x, %s)", sec_id, sec_name.data(), s_next, size, align, is_ro ? "ro" : "rw");
return (u8*)std::exchange(s_next, (void*)next);
LOG_NOTICE(GENERAL, "LLVM: Data section %u '%s' allocated -> %p (size=0x%llx, aligned 0x%x, %s)", sec_id, sec_name.data(), ptr, size, align, is_ro ? "ro" : "rw");
return (u8*)ptr;
}
bool finalizeMemory(std::string* = nullptr) override
@ -479,21 +702,22 @@ struct MemoryManager : llvm::RTDyldMemoryManager
#ifdef _WIN32
// Lock memory manager
std::lock_guard lock(s_mutex);
// Use s_memory as a BASE, compute the difference
const u64 unwind_diff = (u64)addr - (u64)s_memory;
// Fix RUNTIME_FUNCTION records (.pdata section)
auto pdata = std::move(s_unwater.front());
decltype(s_unwater)::value_type pdata_entry = std::move(s_unwater.front());
s_unwater.pop_front();
// Use given memory segment as a BASE, compute the difference
const u64 segment_start = pdata_entry.first;
const u64 unwind_diff = (u64)addr - segment_start;
auto& pdata = pdata_entry.second;
for (auto& rf : pdata)
{
rf.UnwindData += static_cast<DWORD>(unwind_diff);
}
// Register .xdata UNWIND_INFO structs
if (!RtlAddFunctionTable(pdata.data(), (DWORD)pdata.size(), (u64)s_memory))
if (!RtlAddFunctionTable(pdata.data(), (DWORD)pdata.size(), segment_start))
{
LOG_ERROR(GENERAL, "RtlAddFunctionTable() failed! Error %u", GetLastError());
}
@ -635,8 +859,9 @@ struct EventListener : llvm::JITEventListener
// Lock memory manager
std::lock_guard lock(s_mutex);
// Use s_memory as a BASE, compute the difference
const u64 code_diff = (u64)m_mem.m_code_addr - (u64)s_memory;
// Use current memory segment as a BASE, compute the difference
const u64 segment_start = s_alloc.current_segment().first;
const u64 code_diff = u64(m_mem.m_code_addr) - segment_start;
// Fix RUNTIME_FUNCTION records (.pdata section)
for (auto& rf : rfs)
@ -645,7 +870,7 @@ struct EventListener : llvm::JITEventListener
rf.EndAddress += static_cast<DWORD>(code_diff);
}
s_unwater.emplace_back(std::move(rfs));
s_unwater.emplace_back(segment_start, std::move(rfs));
}
}
#endif