diff options
Diffstat (limited to '')
-rw-r--r-- | src/core/memory.cpp | 319 |
1 files changed, 176 insertions, 143 deletions
diff --git a/src/core/memory.cpp b/src/core/memory.cpp index 4397fcfb1..95e070825 100644 --- a/src/core/memory.cpp +++ b/src/core/memory.cpp @@ -35,31 +35,35 @@ struct Memory::Impl { system.ArmInterface(core_id).PageTableChanged(*current_page_table, address_space_width); } - void MapMemoryRegion(Common::PageTable& page_table, VAddr base, u64 size, PAddr target) { + void MapMemoryRegion(Common::PageTable& page_table, Common::ProcessAddress base, u64 size, + Common::PhysicalAddress target) { ASSERT_MSG((size & YUZU_PAGEMASK) == 0, "non-page aligned size: {:016X}", size); - ASSERT_MSG((base & YUZU_PAGEMASK) == 0, "non-page aligned base: {:016X}", base); - ASSERT_MSG(target >= DramMemoryMap::Base, "Out of bounds target: {:016X}", target); + ASSERT_MSG((base & YUZU_PAGEMASK) == 0, "non-page aligned base: {:016X}", GetInteger(base)); + ASSERT_MSG(target >= DramMemoryMap::Base, "Out of bounds target: {:016X}", + GetInteger(target)); MapPages(page_table, base / YUZU_PAGESIZE, size / YUZU_PAGESIZE, target, Common::PageType::Memory); if (Settings::IsFastmemEnabled()) { - system.DeviceMemory().buffer.Map(base, target - DramMemoryMap::Base, size); + system.DeviceMemory().buffer.Map(GetInteger(base), + GetInteger(target) - DramMemoryMap::Base, size); } } - void UnmapRegion(Common::PageTable& page_table, VAddr base, u64 size) { + void UnmapRegion(Common::PageTable& page_table, Common::ProcessAddress base, u64 size) { ASSERT_MSG((size & YUZU_PAGEMASK) == 0, "non-page aligned size: {:016X}", size); - ASSERT_MSG((base & YUZU_PAGEMASK) == 0, "non-page aligned base: {:016X}", base); + ASSERT_MSG((base & YUZU_PAGEMASK) == 0, "non-page aligned base: {:016X}", GetInteger(base)); MapPages(page_table, base / YUZU_PAGESIZE, size / YUZU_PAGESIZE, 0, Common::PageType::Unmapped); if (Settings::IsFastmemEnabled()) { - system.DeviceMemory().buffer.Unmap(base, size); + system.DeviceMemory().buffer.Unmap(GetInteger(base), size); } } - [[nodiscard]] u8* GetPointerFromRasterizerCachedMemory(VAddr vaddr) const { - const PAddr paddr{current_page_table->backing_addr[vaddr >> YUZU_PAGEBITS]}; + [[nodiscard]] u8* GetPointerFromRasterizerCachedMemory(u64 vaddr) const { + const Common::PhysicalAddress paddr{ + current_page_table->backing_addr[vaddr >> YUZU_PAGEBITS]}; if (!paddr) { return {}; @@ -68,8 +72,9 @@ struct Memory::Impl { return system.DeviceMemory().GetPointer<u8>(paddr) + vaddr; } - [[nodiscard]] u8* GetPointerFromDebugMemory(VAddr vaddr) const { - const PAddr paddr{current_page_table->backing_addr[vaddr >> YUZU_PAGEBITS]}; + [[nodiscard]] u8* GetPointerFromDebugMemory(u64 vaddr) const { + const Common::PhysicalAddress paddr{ + current_page_table->backing_addr[vaddr >> YUZU_PAGEBITS]}; if (paddr == 0) { return {}; @@ -78,11 +83,11 @@ struct Memory::Impl { return system.DeviceMemory().GetPointer<u8>(paddr) + vaddr; } - u8 Read8(const VAddr addr) { + u8 Read8(const Common::ProcessAddress addr) { return Read<u8>(addr); } - u16 Read16(const VAddr addr) { + u16 Read16(const Common::ProcessAddress addr) { if ((addr & 1) == 0) { return Read<u16_le>(addr); } else { @@ -92,7 +97,7 @@ struct Memory::Impl { } } - u32 Read32(const VAddr addr) { + u32 Read32(const Common::ProcessAddress addr) { if ((addr & 3) == 0) { return Read<u32_le>(addr); } else { @@ -102,7 +107,7 @@ struct Memory::Impl { } } - u64 Read64(const VAddr addr) { + u64 Read64(const Common::ProcessAddress addr) { if ((addr & 7) == 0) { return Read<u64_le>(addr); } else { @@ -112,11 +117,11 @@ struct Memory::Impl { } } - void Write8(const VAddr addr, const u8 data) { + void Write8(const Common::ProcessAddress addr, const u8 data) { Write<u8>(addr, data); } - void Write16(const VAddr addr, const u16 data) { + void Write16(const Common::ProcessAddress addr, const u16 data) { if ((addr & 1) == 0) { Write<u16_le>(addr, data); } else { @@ -125,7 +130,7 @@ struct Memory::Impl { } } - void Write32(const VAddr addr, const u32 data) { + void Write32(const Common::ProcessAddress addr, const u32 data) { if ((addr & 3) == 0) { Write<u32_le>(addr, data); } else { @@ -134,7 +139,7 @@ struct Memory::Impl { } } - void Write64(const VAddr addr, const u64 data) { + void Write64(const Common::ProcessAddress addr, const u64 data) { if ((addr & 7) == 0) { Write<u64_le>(addr, data); } else { @@ -143,23 +148,23 @@ struct Memory::Impl { } } - bool WriteExclusive8(const VAddr addr, const u8 data, const u8 expected) { + bool WriteExclusive8(const Common::ProcessAddress addr, const u8 data, const u8 expected) { return WriteExclusive<u8>(addr, data, expected); } - bool WriteExclusive16(const VAddr addr, const u16 data, const u16 expected) { + bool WriteExclusive16(const Common::ProcessAddress addr, const u16 data, const u16 expected) { return WriteExclusive<u16_le>(addr, data, expected); } - bool WriteExclusive32(const VAddr addr, const u32 data, const u32 expected) { + bool WriteExclusive32(const Common::ProcessAddress addr, const u32 data, const u32 expected) { return WriteExclusive<u32_le>(addr, data, expected); } - bool WriteExclusive64(const VAddr addr, const u64 data, const u64 expected) { + bool WriteExclusive64(const Common::ProcessAddress addr, const u64 data, const u64 expected) { return WriteExclusive<u64_le>(addr, data, expected); } - std::string ReadCString(VAddr vaddr, std::size_t max_length) { + std::string ReadCString(Common::ProcessAddress vaddr, std::size_t max_length) { std::string string; string.reserve(max_length); for (std::size_t i = 0; i < max_length; ++i) { @@ -174,8 +179,9 @@ struct Memory::Impl { return string; } - void WalkBlock(const Kernel::KProcess& process, const VAddr addr, const std::size_t size, - auto on_unmapped, auto on_memory, auto on_rasterizer, auto increment) { + void WalkBlock(const Kernel::KProcess& process, const Common::ProcessAddress addr, + const std::size_t size, auto on_unmapped, auto on_memory, auto on_rasterizer, + auto increment) { const auto& page_table = process.PageTable().PageTableImpl(); std::size_t remaining_size = size; std::size_t page_index = addr >> YUZU_PAGEBITS; @@ -185,7 +191,7 @@ struct Memory::Impl { const std::size_t copy_amount = std::min(static_cast<std::size_t>(YUZU_PAGESIZE) - page_offset, remaining_size); const auto current_vaddr = - static_cast<VAddr>((page_index << YUZU_PAGEBITS) + page_offset); + static_cast<u64>((page_index << YUZU_PAGEBITS) + page_offset); const auto [pointer, type] = page_table.pointers[page_index].PointerType(); switch (type) { @@ -220,24 +226,24 @@ struct Memory::Impl { } template <bool UNSAFE> - void ReadBlockImpl(const Kernel::KProcess& process, const VAddr src_addr, void* dest_buffer, - const std::size_t size) { + void ReadBlockImpl(const Kernel::KProcess& process, const Common::ProcessAddress src_addr, + void* dest_buffer, const std::size_t size) { WalkBlock( process, src_addr, size, [src_addr, size, &dest_buffer](const std::size_t copy_amount, - const VAddr current_vaddr) { + const Common::ProcessAddress current_vaddr) { LOG_ERROR(HW_Memory, "Unmapped ReadBlock @ 0x{:016X} (start address = 0x{:016X}, size = {})", - current_vaddr, src_addr, size); + GetInteger(current_vaddr), GetInteger(src_addr), size); std::memset(dest_buffer, 0, copy_amount); }, [&](const std::size_t copy_amount, const u8* const src_ptr) { std::memcpy(dest_buffer, src_ptr, copy_amount); }, - [&](const VAddr current_vaddr, const std::size_t copy_amount, + [&](const Common::ProcessAddress current_vaddr, const std::size_t copy_amount, const u8* const host_ptr) { if constexpr (!UNSAFE) { - system.GPU().FlushRegion(current_vaddr, copy_amount); + system.GPU().FlushRegion(GetInteger(current_vaddr), copy_amount); } std::memcpy(dest_buffer, host_ptr, copy_amount); }, @@ -246,30 +252,34 @@ struct Memory::Impl { }); } - void ReadBlock(const VAddr src_addr, void* dest_buffer, const std::size_t size) { + void ReadBlock(const Common::ProcessAddress src_addr, void* dest_buffer, + const std::size_t size) { ReadBlockImpl<false>(*system.ApplicationProcess(), src_addr, dest_buffer, size); } - void ReadBlockUnsafe(const VAddr src_addr, void* dest_buffer, const std::size_t size) { + void ReadBlockUnsafe(const Common::ProcessAddress src_addr, void* dest_buffer, + const std::size_t size) { ReadBlockImpl<true>(*system.ApplicationProcess(), src_addr, dest_buffer, size); } template <bool UNSAFE> - void WriteBlockImpl(const Kernel::KProcess& process, const VAddr dest_addr, + void WriteBlockImpl(const Kernel::KProcess& process, const Common::ProcessAddress dest_addr, const void* src_buffer, const std::size_t size) { WalkBlock( process, dest_addr, size, - [dest_addr, size](const std::size_t copy_amount, const VAddr current_vaddr) { + [dest_addr, size](const std::size_t copy_amount, + const Common::ProcessAddress current_vaddr) { LOG_ERROR(HW_Memory, "Unmapped WriteBlock @ 0x{:016X} (start address = 0x{:016X}, size = {})", - current_vaddr, dest_addr, size); + GetInteger(current_vaddr), GetInteger(dest_addr), size); }, [&](const std::size_t copy_amount, u8* const dest_ptr) { std::memcpy(dest_ptr, src_buffer, copy_amount); }, - [&](const VAddr current_vaddr, const std::size_t copy_amount, u8* const host_ptr) { + [&](const Common::ProcessAddress current_vaddr, const std::size_t copy_amount, + u8* const host_ptr) { if constexpr (!UNSAFE) { - system.GPU().InvalidateRegion(current_vaddr, copy_amount); + system.GPU().InvalidateRegion(GetInteger(current_vaddr), copy_amount); } std::memcpy(host_ptr, src_buffer, copy_amount); }, @@ -278,71 +288,77 @@ struct Memory::Impl { }); } - void WriteBlock(const VAddr dest_addr, const void* src_buffer, const std::size_t size) { + void WriteBlock(const Common::ProcessAddress dest_addr, const void* src_buffer, + const std::size_t size) { WriteBlockImpl<false>(*system.ApplicationProcess(), dest_addr, src_buffer, size); } - void WriteBlockUnsafe(const VAddr dest_addr, const void* src_buffer, const std::size_t size) { + void WriteBlockUnsafe(const Common::ProcessAddress dest_addr, const void* src_buffer, + const std::size_t size) { WriteBlockImpl<true>(*system.ApplicationProcess(), dest_addr, src_buffer, size); } - void ZeroBlock(const Kernel::KProcess& process, const VAddr dest_addr, const std::size_t size) { + void ZeroBlock(const Kernel::KProcess& process, const Common::ProcessAddress dest_addr, + const std::size_t size) { WalkBlock( process, dest_addr, size, - [dest_addr, size](const std::size_t copy_amount, const VAddr current_vaddr) { + [dest_addr, size](const std::size_t copy_amount, + const Common::ProcessAddress current_vaddr) { LOG_ERROR(HW_Memory, "Unmapped ZeroBlock @ 0x{:016X} (start address = 0x{:016X}, size = {})", - current_vaddr, dest_addr, size); + GetInteger(current_vaddr), GetInteger(dest_addr), size); }, [](const std::size_t copy_amount, u8* const dest_ptr) { std::memset(dest_ptr, 0, copy_amount); }, - [&](const VAddr current_vaddr, const std::size_t copy_amount, u8* const host_ptr) { - system.GPU().InvalidateRegion(current_vaddr, copy_amount); + [&](const Common::ProcessAddress current_vaddr, const std::size_t copy_amount, + u8* const host_ptr) { + system.GPU().InvalidateRegion(GetInteger(current_vaddr), copy_amount); std::memset(host_ptr, 0, copy_amount); }, [](const std::size_t copy_amount) {}); } - void CopyBlock(const Kernel::KProcess& process, VAddr dest_addr, VAddr src_addr, - const std::size_t size) { + void CopyBlock(const Kernel::KProcess& process, Common::ProcessAddress dest_addr, + Common::ProcessAddress src_addr, const std::size_t size) { WalkBlock( process, dest_addr, size, - [&](const std::size_t copy_amount, const VAddr current_vaddr) { + [&](const std::size_t copy_amount, const Common::ProcessAddress current_vaddr) { LOG_ERROR(HW_Memory, "Unmapped CopyBlock @ 0x{:016X} (start address = 0x{:016X}, size = {})", - current_vaddr, src_addr, size); + GetInteger(current_vaddr), GetInteger(src_addr), size); ZeroBlock(process, dest_addr, copy_amount); }, [&](const std::size_t copy_amount, const u8* const src_ptr) { WriteBlockImpl<false>(process, dest_addr, src_ptr, copy_amount); }, - [&](const VAddr current_vaddr, const std::size_t copy_amount, u8* const host_ptr) { - system.GPU().FlushRegion(current_vaddr, copy_amount); + [&](const Common::ProcessAddress current_vaddr, const std::size_t copy_amount, + u8* const host_ptr) { + system.GPU().FlushRegion(GetInteger(current_vaddr), copy_amount); WriteBlockImpl<false>(process, dest_addr, host_ptr, copy_amount); }, [&](const std::size_t copy_amount) { - dest_addr += static_cast<VAddr>(copy_amount); - src_addr += static_cast<VAddr>(copy_amount); + dest_addr += copy_amount; + src_addr += copy_amount; }); } template <typename Callback> - Result PerformCacheOperation(const Kernel::KProcess& process, VAddr dest_addr, std::size_t size, - Callback&& cb) { + Result PerformCacheOperation(const Kernel::KProcess& process, Common::ProcessAddress dest_addr, + std::size_t size, Callback&& cb) { class InvalidMemoryException : public std::exception {}; try { WalkBlock( process, dest_addr, size, - [&](const std::size_t block_size, const VAddr current_vaddr) { - LOG_ERROR(HW_Memory, "Unmapped cache maintenance @ {:#018X}", current_vaddr); + [&](const std::size_t block_size, const Common::ProcessAddress current_vaddr) { + LOG_ERROR(HW_Memory, "Unmapped cache maintenance @ {:#018X}", + GetInteger(current_vaddr)); throw InvalidMemoryException(); }, [&](const std::size_t block_size, u8* const host_ptr) {}, - [&](const VAddr current_vaddr, const std::size_t block_size, u8* const host_ptr) { - cb(current_vaddr, block_size); - }, + [&](const Common::ProcessAddress current_vaddr, const std::size_t block_size, + u8* const host_ptr) { cb(current_vaddr, block_size); }, [](const std::size_t block_size) {}); } catch (InvalidMemoryException&) { return Kernel::ResultInvalidCurrentMemory; @@ -351,34 +367,40 @@ struct Memory::Impl { return ResultSuccess; } - Result InvalidateDataCache(const Kernel::KProcess& process, VAddr dest_addr, std::size_t size) { - auto on_rasterizer = [&](const VAddr current_vaddr, const std::size_t block_size) { + Result InvalidateDataCache(const Kernel::KProcess& process, Common::ProcessAddress dest_addr, + std::size_t size) { + auto on_rasterizer = [&](const Common::ProcessAddress current_vaddr, + const std::size_t block_size) { // dc ivac: Invalidate to point of coherency // GPU flush -> CPU invalidate - system.GPU().FlushRegion(current_vaddr, block_size); + system.GPU().FlushRegion(GetInteger(current_vaddr), block_size); }; return PerformCacheOperation(process, dest_addr, size, on_rasterizer); } - Result StoreDataCache(const Kernel::KProcess& process, VAddr dest_addr, std::size_t size) { - auto on_rasterizer = [&](const VAddr current_vaddr, const std::size_t block_size) { + Result StoreDataCache(const Kernel::KProcess& process, Common::ProcessAddress dest_addr, + std::size_t size) { + auto on_rasterizer = [&](const Common::ProcessAddress current_vaddr, + const std::size_t block_size) { // dc cvac: Store to point of coherency // CPU flush -> GPU invalidate - system.GPU().InvalidateRegion(current_vaddr, block_size); + system.GPU().InvalidateRegion(GetInteger(current_vaddr), block_size); }; return PerformCacheOperation(process, dest_addr, size, on_rasterizer); } - Result FlushDataCache(const Kernel::KProcess& process, VAddr dest_addr, std::size_t size) { - auto on_rasterizer = [&](const VAddr current_vaddr, const std::size_t block_size) { + Result FlushDataCache(const Kernel::KProcess& process, Common::ProcessAddress dest_addr, + std::size_t size) { + auto on_rasterizer = [&](const Common::ProcessAddress current_vaddr, + const std::size_t block_size) { // dc civac: Store to point of coherency, and invalidate from cache // CPU flush -> GPU invalidate - system.GPU().InvalidateRegion(current_vaddr, block_size); + system.GPU().InvalidateRegion(GetInteger(current_vaddr), block_size); }; return PerformCacheOperation(process, dest_addr, size, on_rasterizer); } - void MarkRegionDebug(VAddr vaddr, u64 size, bool debug) { + void MarkRegionDebug(u64 vaddr, u64 size, bool debug) { if (vaddr == 0) { return; } @@ -434,7 +456,7 @@ struct Memory::Impl { } } - void RasterizerMarkRegionCached(VAddr vaddr, u64 size, bool cached) { + void RasterizerMarkRegionCached(u64 vaddr, u64 size, bool cached) { if (vaddr == 0) { return; } @@ -514,10 +536,12 @@ struct Memory::Impl { * @param target The target address to begin mapping from. * @param type The page type to map the memory as. */ - void MapPages(Common::PageTable& page_table, VAddr base, u64 size, PAddr target, - Common::PageType type) { - LOG_DEBUG(HW_Memory, "Mapping {:016X} onto {:016X}-{:016X}", target, base * YUZU_PAGESIZE, - (base + size) * YUZU_PAGESIZE); + void MapPages(Common::PageTable& page_table, Common::ProcessAddress base_address, u64 size, + Common::PhysicalAddress target, Common::PageType type) { + auto base = GetInteger(base_address); + + LOG_DEBUG(HW_Memory, "Mapping {:016X} onto {:016X}-{:016X}", GetInteger(target), + base * YUZU_PAGESIZE, (base + size) * YUZU_PAGESIZE); // During boot, current_page_table might not be set yet, in which case we need not flush if (system.IsPoweredOn()) { @@ -530,7 +554,7 @@ struct Memory::Impl { } } - const VAddr end = base + size; + const Common::ProcessAddress end = base + size; ASSERT_MSG(end <= page_table.pointers.size(), "out of range mapping at {:016X}", base + page_table.pointers.size()); @@ -548,7 +572,7 @@ struct Memory::Impl { while (base != end) { page_table.pointers[base].Store( system.DeviceMemory().GetPointer<u8>(target) - (base << YUZU_PAGEBITS), type); - page_table.backing_addr[base] = target - (base << YUZU_PAGEBITS); + page_table.backing_addr[base] = GetInteger(target) - (base << YUZU_PAGEBITS); ASSERT_MSG(page_table.pointers[base].Pointer(), "memory mapping base yield a nullptr within the table"); @@ -559,9 +583,9 @@ struct Memory::Impl { } } - [[nodiscard]] u8* GetPointerImpl(VAddr vaddr, auto on_unmapped, auto on_rasterizer) const { + [[nodiscard]] u8* GetPointerImpl(u64 vaddr, auto on_unmapped, auto on_rasterizer) const { // AARCH64 masks the upper 16 bit of all memory accesses - vaddr &= 0xffffffffffffULL; + vaddr = vaddr & 0xffffffffffffULL; if (vaddr >= 1uLL << current_page_table->GetAddressSpaceBits()) { on_unmapped(); @@ -593,15 +617,18 @@ struct Memory::Impl { return nullptr; } - [[nodiscard]] u8* GetPointer(const VAddr vaddr) const { + [[nodiscard]] u8* GetPointer(const Common::ProcessAddress vaddr) const { return GetPointerImpl( - vaddr, [vaddr]() { LOG_ERROR(HW_Memory, "Unmapped GetPointer @ 0x{:016X}", vaddr); }, + GetInteger(vaddr), + [vaddr]() { + LOG_ERROR(HW_Memory, "Unmapped GetPointer @ 0x{:016X}", GetInteger(vaddr)); + }, []() {}); } - [[nodiscard]] u8* GetPointerSilent(const VAddr vaddr) const { + [[nodiscard]] u8* GetPointerSilent(const Common::ProcessAddress vaddr) const { return GetPointerImpl( - vaddr, []() {}, []() {}); + GetInteger(vaddr), []() {}, []() {}); } /** @@ -616,14 +643,15 @@ struct Memory::Impl { * @returns The instance of T read from the specified virtual address. */ template <typename T> - T Read(VAddr vaddr) { + T Read(Common::ProcessAddress vaddr) { T result = 0; const u8* const ptr = GetPointerImpl( - vaddr, + GetInteger(vaddr), [vaddr]() { - LOG_ERROR(HW_Memory, "Unmapped Read{} @ 0x{:016X}", sizeof(T) * 8, vaddr); + LOG_ERROR(HW_Memory, "Unmapped Read{} @ 0x{:016X}", sizeof(T) * 8, + GetInteger(vaddr)); }, - [&]() { system.GPU().FlushRegion(vaddr, sizeof(T)); }); + [&]() { system.GPU().FlushRegion(GetInteger(vaddr), sizeof(T)); }); if (ptr) { std::memcpy(&result, ptr, sizeof(T)); } @@ -640,28 +668,28 @@ struct Memory::Impl { * is undefined. */ template <typename T> - void Write(VAddr vaddr, const T data) { + void Write(Common::ProcessAddress vaddr, const T data) { u8* const ptr = GetPointerImpl( - vaddr, + GetInteger(vaddr), [vaddr, data]() { LOG_ERROR(HW_Memory, "Unmapped Write{} @ 0x{:016X} = 0x{:016X}", sizeof(T) * 8, - vaddr, static_cast<u64>(data)); + GetInteger(vaddr), static_cast<u64>(data)); }, - [&]() { system.GPU().InvalidateRegion(vaddr, sizeof(T)); }); + [&]() { system.GPU().InvalidateRegion(GetInteger(vaddr), sizeof(T)); }); if (ptr) { std::memcpy(ptr, &data, sizeof(T)); } } template <typename T> - bool WriteExclusive(VAddr vaddr, const T data, const T expected) { + bool WriteExclusive(Common::ProcessAddress vaddr, const T data, const T expected) { u8* const ptr = GetPointerImpl( - vaddr, + GetInteger(vaddr), [vaddr, data]() { LOG_ERROR(HW_Memory, "Unmapped WriteExclusive{} @ 0x{:016X} = 0x{:016X}", - sizeof(T) * 8, vaddr, static_cast<u64>(data)); + sizeof(T) * 8, GetInteger(vaddr), static_cast<u64>(data)); }, - [&]() { system.GPU().InvalidateRegion(vaddr, sizeof(T)); }); + [&]() { system.GPU().InvalidateRegion(GetInteger(vaddr), sizeof(T)); }); if (ptr) { const auto volatile_pointer = reinterpret_cast<volatile T*>(ptr); return Common::AtomicCompareAndSwap(volatile_pointer, data, expected); @@ -669,14 +697,14 @@ struct Memory::Impl { return true; } - bool WriteExclusive128(VAddr vaddr, const u128 data, const u128 expected) { + bool WriteExclusive128(Common::ProcessAddress vaddr, const u128 data, const u128 expected) { u8* const ptr = GetPointerImpl( - vaddr, + GetInteger(vaddr), [vaddr, data]() { LOG_ERROR(HW_Memory, "Unmapped WriteExclusive128 @ 0x{:016X} = 0x{:016X}{:016X}", - vaddr, static_cast<u64>(data[1]), static_cast<u64>(data[0])); + GetInteger(vaddr), static_cast<u64>(data[1]), static_cast<u64>(data[0])); }, - [&]() { system.GPU().InvalidateRegion(vaddr, sizeof(u128)); }); + [&]() { system.GPU().InvalidateRegion(GetInteger(vaddr), sizeof(u128)); }); if (ptr) { const auto volatile_pointer = reinterpret_cast<volatile u64*>(ptr); return Common::AtomicCompareAndSwap(volatile_pointer, data, expected); @@ -702,15 +730,16 @@ void Memory::SetCurrentPageTable(Kernel::KProcess& process, u32 core_id) { impl->SetCurrentPageTable(process, core_id); } -void Memory::MapMemoryRegion(Common::PageTable& page_table, VAddr base, u64 size, PAddr target) { +void Memory::MapMemoryRegion(Common::PageTable& page_table, Common::ProcessAddress base, u64 size, + Common::PhysicalAddress target) { impl->MapMemoryRegion(page_table, base, size, target); } -void Memory::UnmapRegion(Common::PageTable& page_table, VAddr base, u64 size) { +void Memory::UnmapRegion(Common::PageTable& page_table, Common::ProcessAddress base, u64 size) { impl->UnmapRegion(page_table, base, size); } -bool Memory::IsValidVirtualAddress(const VAddr vaddr) const { +bool Memory::IsValidVirtualAddress(const Common::ProcessAddress vaddr) const { const Kernel::KProcess& process = *system.ApplicationProcess(); const auto& page_table = process.PageTable().PageTableImpl(); const size_t page = vaddr >> YUZU_PAGEBITS; @@ -722,9 +751,9 @@ bool Memory::IsValidVirtualAddress(const VAddr vaddr) const { type == Common::PageType::DebugMemory; } -bool Memory::IsValidVirtualAddressRange(VAddr base, u64 size) const { - VAddr end = base + size; - VAddr page = Common::AlignDown(base, YUZU_PAGESIZE); +bool Memory::IsValidVirtualAddressRange(Common::ProcessAddress base, u64 size) const { + Common::ProcessAddress end = base + size; + Common::ProcessAddress page = Common::AlignDown(GetInteger(base), YUZU_PAGESIZE); for (; page < end; page += YUZU_PAGESIZE) { if (!IsValidVirtualAddress(page)) { @@ -735,131 +764,135 @@ bool Memory::IsValidVirtualAddressRange(VAddr base, u64 size) const { return true; } -u8* Memory::GetPointer(VAddr vaddr) { +u8* Memory::GetPointer(Common::ProcessAddress vaddr) { return impl->GetPointer(vaddr); } -u8* Memory::GetPointerSilent(VAddr vaddr) { +u8* Memory::GetPointerSilent(Common::ProcessAddress vaddr) { return impl->GetPointerSilent(vaddr); } -const u8* Memory::GetPointer(VAddr vaddr) const { +const u8* Memory::GetPointer(Common::ProcessAddress vaddr) const { return impl->GetPointer(vaddr); } -u8 Memory::Read8(const VAddr addr) { +u8 Memory::Read8(const Common::ProcessAddress addr) { return impl->Read8(addr); } -u16 Memory::Read16(const VAddr addr) { +u16 Memory::Read16(const Common::ProcessAddress addr) { return impl->Read16(addr); } -u32 Memory::Read32(const VAddr addr) { +u32 Memory::Read32(const Common::ProcessAddress addr) { return impl->Read32(addr); } -u64 Memory::Read64(const VAddr addr) { +u64 Memory::Read64(const Common::ProcessAddress addr) { return impl->Read64(addr); } -void Memory::Write8(VAddr addr, u8 data) { +void Memory::Write8(Common::ProcessAddress addr, u8 data) { impl->Write8(addr, data); } -void Memory::Write16(VAddr addr, u16 data) { +void Memory::Write16(Common::ProcessAddress addr, u16 data) { impl->Write16(addr, data); } -void Memory::Write32(VAddr addr, u32 data) { +void Memory::Write32(Common::ProcessAddress addr, u32 data) { impl->Write32(addr, data); } -void Memory::Write64(VAddr addr, u64 data) { +void Memory::Write64(Common::ProcessAddress addr, u64 data) { impl->Write64(addr, data); } -bool Memory::WriteExclusive8(VAddr addr, u8 data, u8 expected) { +bool Memory::WriteExclusive8(Common::ProcessAddress addr, u8 data, u8 expected) { return impl->WriteExclusive8(addr, data, expected); } -bool Memory::WriteExclusive16(VAddr addr, u16 data, u16 expected) { +bool Memory::WriteExclusive16(Common::ProcessAddress addr, u16 data, u16 expected) { return impl->WriteExclusive16(addr, data, expected); } -bool Memory::WriteExclusive32(VAddr addr, u32 data, u32 expected) { +bool Memory::WriteExclusive32(Common::ProcessAddress addr, u32 data, u32 expected) { return impl->WriteExclusive32(addr, data, expected); } -bool Memory::WriteExclusive64(VAddr addr, u64 data, u64 expected) { +bool Memory::WriteExclusive64(Common::ProcessAddress addr, u64 data, u64 expected) { return impl->WriteExclusive64(addr, data, expected); } -bool Memory::WriteExclusive128(VAddr addr, u128 data, u128 expected) { +bool Memory::WriteExclusive128(Common::ProcessAddress addr, u128 data, u128 expected) { return impl->WriteExclusive128(addr, data, expected); } -std::string Memory::ReadCString(VAddr vaddr, std::size_t max_length) { +std::string Memory::ReadCString(Common::ProcessAddress vaddr, std::size_t max_length) { return impl->ReadCString(vaddr, max_length); } -void Memory::ReadBlock(const Kernel::KProcess& process, const VAddr src_addr, void* dest_buffer, - const std::size_t size) { +void Memory::ReadBlock(const Kernel::KProcess& process, const Common::ProcessAddress src_addr, + void* dest_buffer, const std::size_t size) { impl->ReadBlockImpl<false>(process, src_addr, dest_buffer, size); } -void Memory::ReadBlock(const VAddr src_addr, void* dest_buffer, const std::size_t size) { +void Memory::ReadBlock(const Common::ProcessAddress src_addr, void* dest_buffer, + const std::size_t size) { impl->ReadBlock(src_addr, dest_buffer, size); } -void Memory::ReadBlockUnsafe(const VAddr src_addr, void* dest_buffer, const std::size_t size) { +void Memory::ReadBlockUnsafe(const Common::ProcessAddress src_addr, void* dest_buffer, + const std::size_t size) { impl->ReadBlockUnsafe(src_addr, dest_buffer, size); } -void Memory::WriteBlock(const Kernel::KProcess& process, VAddr dest_addr, const void* src_buffer, - std::size_t size) { +void Memory::WriteBlock(const Kernel::KProcess& process, Common::ProcessAddress dest_addr, + const void* src_buffer, std::size_t size) { impl->WriteBlockImpl<false>(process, dest_addr, src_buffer, size); } -void Memory::WriteBlock(const VAddr dest_addr, const void* src_buffer, const std::size_t size) { +void Memory::WriteBlock(const Common::ProcessAddress dest_addr, const void* src_buffer, + const std::size_t size) { impl->WriteBlock(dest_addr, src_buffer, size); } -void Memory::WriteBlockUnsafe(const VAddr dest_addr, const void* src_buffer, +void Memory::WriteBlockUnsafe(const Common::ProcessAddress dest_addr, const void* src_buffer, const std::size_t size) { impl->WriteBlockUnsafe(dest_addr, src_buffer, size); } -void Memory::CopyBlock(const Kernel::KProcess& process, VAddr dest_addr, VAddr src_addr, - const std::size_t size) { +void Memory::CopyBlock(const Kernel::KProcess& process, Common::ProcessAddress dest_addr, + Common::ProcessAddress src_addr, const std::size_t size) { impl->CopyBlock(process, dest_addr, src_addr, size); } -void Memory::ZeroBlock(const Kernel::KProcess& process, VAddr dest_addr, const std::size_t size) { +void Memory::ZeroBlock(const Kernel::KProcess& process, Common::ProcessAddress dest_addr, + const std::size_t size) { impl->ZeroBlock(process, dest_addr, size); } -Result Memory::InvalidateDataCache(const Kernel::KProcess& process, VAddr dest_addr, - const std::size_t size) { +Result Memory::InvalidateDataCache(const Kernel::KProcess& process, + Common::ProcessAddress dest_addr, const std::size_t size) { return impl->InvalidateDataCache(process, dest_addr, size); } -Result Memory::StoreDataCache(const Kernel::KProcess& process, VAddr dest_addr, +Result Memory::StoreDataCache(const Kernel::KProcess& process, Common::ProcessAddress dest_addr, const std::size_t size) { return impl->StoreDataCache(process, dest_addr, size); } -Result Memory::FlushDataCache(const Kernel::KProcess& process, VAddr dest_addr, +Result Memory::FlushDataCache(const Kernel::KProcess& process, Common::ProcessAddress dest_addr, const std::size_t size) { return impl->FlushDataCache(process, dest_addr, size); } -void Memory::RasterizerMarkRegionCached(VAddr vaddr, u64 size, bool cached) { - impl->RasterizerMarkRegionCached(vaddr, size, cached); +void Memory::RasterizerMarkRegionCached(Common::ProcessAddress vaddr, u64 size, bool cached) { + impl->RasterizerMarkRegionCached(GetInteger(vaddr), size, cached); } -void Memory::MarkRegionDebug(VAddr vaddr, u64 size, bool debug) { - impl->MarkRegionDebug(vaddr, size, debug); +void Memory::MarkRegionDebug(Common::ProcessAddress vaddr, u64 size, bool debug) { + impl->MarkRegionDebug(GetInteger(vaddr), size, debug); } } // namespace Core::Memory |