file_sys: Override missing mutating functions to be stubbed out for ReadOnlyVfsDirectory by default

Ensures that read only indeed means read only.

src/core/file_sys/control_metadata.cpp

@@ -8,13 +8,23 @@
 
 namespace FileSys {
 
-const std::array<const char*, 15> LANGUAGE_NAMES = {
-    "AmericanEnglish", "BritishEnglish", "Japanese",
-    "French",          "German",         "LatinAmericanSpanish",
-    "Spanish",         "Italian",        "Dutch",
-    "CanadianFrench",  "Portugese",      "Russian",
-    "Korean",          "Taiwanese",      "Chinese",
-};
+const std::array<const char*, 15> LANGUAGE_NAMES{{
+    "AmericanEnglish",
+    "BritishEnglish",
+    "Japanese",
+    "French",
+    "German",
+    "LatinAmericanSpanish",
+    "Spanish",
+    "Italian",
+    "Dutch",
+    "CanadianFrench",
+    "Portuguese",
+    "Russian",
+    "Korean",
+    "Taiwanese",
+    "Chinese",
+}};
 
 std::string LanguageEntry::GetApplicationName() const {
     return Common::StringFromFixedZeroTerminatedBuffer(application_name.data(),

src/core/file_sys/registered_cache.cpp

@@ -381,22 +381,22 @@ std::vector<RegisteredCacheEntry> RegisteredCache::ListEntriesFilter(
     return out;
 }
 
-static std::shared_ptr<NCA> GetNCAFromNSPForID(std::shared_ptr<NSP> nsp, const NcaID& id) {
-    const auto file = nsp->GetFile(fmt::format("{}.nca", Common::HexArrayToString(id, false)));
+static std::shared_ptr<NCA> GetNCAFromNSPForID(const NSP& nsp, const NcaID& id) {
+    const auto file = nsp.GetFile(fmt::format("{}.nca", Common::HexArrayToString(id, false)));
     if (file == nullptr)
         return nullptr;
     return std::make_shared<NCA>(file);
 }
 
-InstallResult RegisteredCache::InstallEntry(std::shared_ptr<XCI> xci, bool overwrite_if_exists,
+InstallResult RegisteredCache::InstallEntry(const XCI& xci, bool overwrite_if_exists,
                                             const VfsCopyFunction& copy) {
-    return InstallEntry(xci->GetSecurePartitionNSP(), overwrite_if_exists, copy);
+    return InstallEntry(*xci.GetSecurePartitionNSP(), overwrite_if_exists, copy);
 }
 
-InstallResult RegisteredCache::InstallEntry(std::shared_ptr<NSP> nsp, bool overwrite_if_exists,
+InstallResult RegisteredCache::InstallEntry(const NSP& nsp, bool overwrite_if_exists,
                                             const VfsCopyFunction& copy) {
-    const auto& ncas = nsp->GetNCAsCollapsed();
-    const auto& meta_iter = std::find_if(ncas.begin(), ncas.end(), [](std::shared_ptr<NCA> nca) {
+    const auto ncas = nsp.GetNCAsCollapsed();
+    const auto meta_iter = std::find_if(ncas.begin(), ncas.end(), [](const auto& nca) {
         return nca->GetType() == NCAContentType::Meta;
     });
 
@@ -410,7 +410,7 @@ InstallResult RegisteredCache::InstallEntry(std::shared_ptr<NSP> nsp, bool overw
     const auto meta_id_raw = (*meta_iter)->GetName().substr(0, 32);
     const auto meta_id = Common::HexStringToArray<16>(meta_id_raw);
 
-    const auto res = RawInstallNCA(*meta_iter, copy, overwrite_if_exists, meta_id);
+    const auto res = RawInstallNCA(**meta_iter, copy, overwrite_if_exists, meta_id);
     if (res != InstallResult::Success)
         return res;
 
@@ -422,7 +422,7 @@ InstallResult RegisteredCache::InstallEntry(std::shared_ptr<NSP> nsp, bool overw
         const auto nca = GetNCAFromNSPForID(nsp, record.nca_id);
         if (nca == nullptr)
             return InstallResult::ErrorCopyFailed;
-        const auto res2 = RawInstallNCA(nca, copy, overwrite_if_exists, record.nca_id);
+        const auto res2 = RawInstallNCA(*nca, copy, overwrite_if_exists, record.nca_id);
         if (res2 != InstallResult::Success)
             return res2;
     }
@@ -431,21 +431,21 @@ InstallResult RegisteredCache::InstallEntry(std::shared_ptr<NSP> nsp, bool overw
     return InstallResult::Success;
 }
 
-InstallResult RegisteredCache::InstallEntry(std::shared_ptr<NCA> nca, TitleType type,
+InstallResult RegisteredCache::InstallEntry(const NCA& nca, TitleType type,
                                             bool overwrite_if_exists, const VfsCopyFunction& copy) {
     CNMTHeader header{
-        nca->GetTitleId(), ///< Title ID
-        0,                 ///< Ignore/Default title version
-        type,              ///< Type
-        {},                ///< Padding
-        0x10,              ///< Default table offset
-        1,                 ///< 1 Content Entry
-        0,                 ///< No Meta Entries
-        {},                ///< Padding
+        nca.GetTitleId(), ///< Title ID
+        0,                ///< Ignore/Default title version
+        type,             ///< Type
+        {},               ///< Padding
+        0x10,             ///< Default table offset
+        1,                ///< 1 Content Entry
+        0,                ///< No Meta Entries
+        {},               ///< Padding
     };
     OptionalHeader opt_header{0, 0};
-    ContentRecord c_rec{{}, {}, {}, GetCRTypeFromNCAType(nca->GetType()), {}};
-    const auto& data = nca->GetBaseFile()->ReadBytes(0x100000);
+    ContentRecord c_rec{{}, {}, {}, GetCRTypeFromNCAType(nca.GetType()), {}};
+    const auto& data = nca.GetBaseFile()->ReadBytes(0x100000);
     mbedtls_sha256(data.data(), data.size(), c_rec.hash.data(), 0);
     memcpy(&c_rec.nca_id, &c_rec.hash, 16);
     const CNMT new_cnmt(header, opt_header, {c_rec}, {});
@@ -454,10 +454,10 @@ InstallResult RegisteredCache::InstallEntry(std::shared_ptr<NCA> nca, TitleType
     return RawInstallNCA(nca, copy, overwrite_if_exists, c_rec.nca_id);
 }
 
-InstallResult RegisteredCache::RawInstallNCA(std::shared_ptr<NCA> nca, const VfsCopyFunction& copy,
+InstallResult RegisteredCache::RawInstallNCA(const NCA& nca, const VfsCopyFunction& copy,
                                              bool overwrite_if_exists,
                                              std::optional<NcaID> override_id) {
-    const auto in = nca->GetBaseFile();
+    const auto in = nca.GetBaseFile();
     Core::Crypto::SHA256Hash hash{};
 
     // Calculate NcaID

src/core/file_sys/registered_cache.h

@@ -6,7 +6,6 @@
 
 #include <array>
 #include <functional>
-#include <map>
 #include <memory>
 #include <string>
 #include <vector>
@@ -104,17 +103,16 @@ public:
 
     // Raw copies all the ncas from the xci/nsp to the csache. Does some quick checks to make sure
     // there is a meta NCA and all of them are accessible.
-    InstallResult InstallEntry(std::shared_ptr<XCI> xci, bool overwrite_if_exists = false,
+    InstallResult InstallEntry(const XCI& xci, bool overwrite_if_exists = false,
                                const VfsCopyFunction& copy = &VfsRawCopy);
-    InstallResult InstallEntry(std::shared_ptr<NSP> nsp, bool overwrite_if_exists = false,
+    InstallResult InstallEntry(const NSP& nsp, bool overwrite_if_exists = false,
                                const VfsCopyFunction& copy = &VfsRawCopy);
 
     // Due to the fact that we must use Meta-type NCAs to determine the existance of files, this
     // poses quite a challenge. Instead of creating a new meta NCA for this file, yuzu will create a
     // dir inside the NAND called 'yuzu_meta' and store the raw CNMT there.
     // TODO(DarkLordZach): Author real meta-type NCAs and install those.
-    InstallResult InstallEntry(std::shared_ptr<NCA> nca, TitleType type,
-                               bool overwrite_if_exists = false,
+    InstallResult InstallEntry(const NCA& nca, TitleType type, bool overwrite_if_exists = false,
                                const VfsCopyFunction& copy = &VfsRawCopy);
 
 private:
@@ -128,7 +126,7 @@ private:
     std::optional<NcaID> GetNcaIDFromMetadata(u64 title_id, ContentRecordType type) const;
     VirtualFile GetFileAtID(NcaID id) const;
     VirtualFile OpenFileOrDirectoryConcat(const VirtualDir& dir, std::string_view path) const;
-    InstallResult RawInstallNCA(std::shared_ptr<NCA> nca, const VfsCopyFunction& copy,
+    InstallResult RawInstallNCA(const NCA& nca, const VfsCopyFunction& copy,
                                 bool overwrite_if_exists, std::optional<NcaID> override_id = {});
     bool RawInstallYuzuMeta(const CNMT& cnmt);
 

src/core/file_sys/vfs.cpp

@@ -384,6 +384,28 @@ bool VfsDirectory::DeleteSubdirectoryRecursive(std::string_view name) {
     return success;
 }
 
+bool VfsDirectory::CleanSubdirectoryRecursive(std::string_view name) {
+    auto dir = GetSubdirectory(name);
+    if (dir == nullptr) {
+        return false;
+    }
+
+    bool success = true;
+    for (const auto& file : dir->GetFiles()) {
+        if (!dir->DeleteFile(file->GetName())) {
+            success = false;
+        }
+    }
+
+    for (const auto& sdir : dir->GetSubdirectories()) {
+        if (!dir->DeleteSubdirectoryRecursive(sdir->GetName())) {
+            success = false;
+        }
+    }
+
+    return success;
+}
+
 bool VfsDirectory::Copy(std::string_view src, std::string_view dest) {
     const auto f1 = GetFile(src);
     auto f2 = CreateFile(dest);
@@ -431,10 +453,34 @@ std::shared_ptr<VfsFile> ReadOnlyVfsDirectory::CreateFile(std::string_view name)
     return nullptr;
 }
 
+std::shared_ptr<VfsFile> ReadOnlyVfsDirectory::CreateFileAbsolute(std::string_view path) {
+    return nullptr;
+}
+
+std::shared_ptr<VfsFile> ReadOnlyVfsDirectory::CreateFileRelative(std::string_view path) {
+    return nullptr;
+}
+
+std::shared_ptr<VfsDirectory> ReadOnlyVfsDirectory::CreateDirectoryAbsolute(std::string_view path) {
+    return nullptr;
+}
+
+std::shared_ptr<VfsDirectory> ReadOnlyVfsDirectory::CreateDirectoryRelative(std::string_view path) {
+    return nullptr;
+}
+
 bool ReadOnlyVfsDirectory::DeleteSubdirectory(std::string_view name) {
     return false;
 }
 
+bool ReadOnlyVfsDirectory::DeleteSubdirectoryRecursive(std::string_view name) {
+    return false;
+}
+
+bool ReadOnlyVfsDirectory::CleanSubdirectoryRecursive(std::string_view name) {
+    return false;
+}
+
 bool ReadOnlyVfsDirectory::DeleteFile(std::string_view name) {
     return false;
 }

src/core/file_sys/vfs.h

@@ -245,12 +245,18 @@ public:
     // any failure.
     virtual std::shared_ptr<VfsDirectory> CreateDirectoryAbsolute(std::string_view path);
 
-    // Deletes the subdirectory with name and returns true on success.
+    // Deletes the subdirectory with the given name and returns true on success.
     virtual bool DeleteSubdirectory(std::string_view name) = 0;
-    // Deletes all subdirectories and files of subdirectory with name recirsively and then deletes
-    // the subdirectory. Returns true on success.
+
+    // Deletes all subdirectories and files within the provided directory and then deletes
+    // the directory itself. Returns true on success.
     virtual bool DeleteSubdirectoryRecursive(std::string_view name);
-    // Returnes whether or not the file with name name was deleted successfully.
+
+    // Deletes all subdirectories and files within the provided directory.
+    // Unlike DeleteSubdirectoryRecursive, this does not delete the provided directory.
+    virtual bool CleanSubdirectoryRecursive(std::string_view name);
+
+    // Returns whether or not the file with name name was deleted successfully.
     virtual bool DeleteFile(std::string_view name) = 0;
 
     // Returns whether or not this directory was renamed to name.
@@ -276,7 +282,13 @@ public:
     bool IsReadable() const override;
     std::shared_ptr<VfsDirectory> CreateSubdirectory(std::string_view name) override;
     std::shared_ptr<VfsFile> CreateFile(std::string_view name) override;
+    std::shared_ptr<VfsFile> CreateFileAbsolute(std::string_view path) override;
+    std::shared_ptr<VfsFile> CreateFileRelative(std::string_view path) override;
+    std::shared_ptr<VfsDirectory> CreateDirectoryAbsolute(std::string_view path) override;
+    std::shared_ptr<VfsDirectory> CreateDirectoryRelative(std::string_view path) override;
     bool DeleteSubdirectory(std::string_view name) override;
+    bool DeleteSubdirectoryRecursive(std::string_view name) override;
+    bool CleanSubdirectoryRecursive(std::string_view name) override;
     bool DeleteFile(std::string_view name) override;
     bool Rename(std::string_view name) override;
 };

src/core/gdbstub/gdbstub.cpp

@@ -282,7 +282,7 @@ static void FpuWrite(std::size_t id, u128 val, Kernel::Thread* thread = nullptr)
     if (id >= UC_ARM64_REG_Q0 && id < FPCR_REGISTER) {
         thread_context.vector_registers[id - UC_ARM64_REG_Q0] = val;
     } else if (id == FPCR_REGISTER) {
-        thread_context.fpcr = val[0];
+        thread_context.fpcr = static_cast<u32>(val[0]);
     }
 }
 

src/core/hle/kernel/resource_limit.h

@@ -14,7 +14,7 @@ namespace Kernel {
 
 class KernelCore;
 
-enum class ResourceType {
+enum class ResourceType : u32 {
     PhysicalMemory,
     Threads,
     Events,
@@ -25,6 +25,10 @@ enum class ResourceType {
     ResourceTypeCount
 };
 
+constexpr bool IsValidResourceType(ResourceType type) {
+    return type < ResourceType::ResourceTypeCount;
+}
+
 class ResourceLimit final : public Object {
 public:
     /**

src/core/hle/kernel/svc.cpp

@@ -63,56 +63,129 @@ bool IsInsideNewMapRegion(const VMManager& vm, VAddr address, u64 size) {
                                 vm.GetNewMapRegionEndAddress());
 }
 
+// 8 GiB
+constexpr u64 MAIN_MEMORY_SIZE = 0x200000000;
+
 // Helper function that performs the common sanity checks for svcMapMemory
 // and svcUnmapMemory. This is doable, as both functions perform their sanitizing
 // in the same order.
 ResultCode MapUnmapMemorySanityChecks(const VMManager& vm_manager, VAddr dst_addr, VAddr src_addr,
                                       u64 size) {
-    if (!Common::Is4KBAligned(dst_addr) || !Common::Is4KBAligned(src_addr)) {
+    if (!Common::Is4KBAligned(dst_addr)) {
+        LOG_ERROR(Kernel_SVC, "Destination address is not aligned to 4KB, 0x{:016X}", dst_addr);
         return ERR_INVALID_ADDRESS;
     }
 
-    if (size == 0 || !Common::Is4KBAligned(size)) {
+    if (!Common::Is4KBAligned(src_addr)) {
+        LOG_ERROR(Kernel_SVC, "Source address is not aligned to 4KB, 0x{:016X}", src_addr);
+        return ERR_INVALID_SIZE;
+    }
+
+    if (size == 0) {
+        LOG_ERROR(Kernel_SVC, "Size is 0");
+        return ERR_INVALID_SIZE;
+    }
+
+    if (!Common::Is4KBAligned(size)) {
+        LOG_ERROR(Kernel_SVC, "Size is not aligned to 4KB, 0x{:016X}", size);
         return ERR_INVALID_SIZE;
     }
 
     if (!IsValidAddressRange(dst_addr, size)) {
+        LOG_ERROR(Kernel_SVC,
+                  "Destination is not a valid address range, addr=0x{:016X}, size=0x{:016X}",
+                  dst_addr, size);
         return ERR_INVALID_ADDRESS_STATE;
     }
 
     if (!IsValidAddressRange(src_addr, size)) {
+        LOG_ERROR(Kernel_SVC, "Source is not a valid address range, addr=0x{:016X}, size=0x{:016X}",
+                  src_addr, size);
         return ERR_INVALID_ADDRESS_STATE;
     }
 
     if (!IsInsideAddressSpace(vm_manager, src_addr, size)) {
+        LOG_ERROR(Kernel_SVC,
+                  "Source is not within the address space, addr=0x{:016X}, size=0x{:016X}",
+                  src_addr, size);
         return ERR_INVALID_ADDRESS_STATE;
     }
 
     if (!IsInsideNewMapRegion(vm_manager, dst_addr, size)) {
+        LOG_ERROR(Kernel_SVC,
+                  "Destination is not within the new map region, addr=0x{:016X}, size=0x{:016X}",
+                  dst_addr, size);
         return ERR_INVALID_MEMORY_RANGE;
     }
 
     const VAddr dst_end_address = dst_addr + size;
     if (dst_end_address > vm_manager.GetHeapRegionBaseAddress() &&
         vm_manager.GetHeapRegionEndAddress() > dst_addr) {
+        LOG_ERROR(Kernel_SVC,
+                  "Destination does not fit within the heap region, addr=0x{:016X}, "
+                  "size=0x{:016X}, end_addr=0x{:016X}",
+                  dst_addr, size, dst_end_address);
         return ERR_INVALID_MEMORY_RANGE;
     }
 
     if (dst_end_address > vm_manager.GetMapRegionBaseAddress() &&
         vm_manager.GetMapRegionEndAddress() > dst_addr) {
+        LOG_ERROR(Kernel_SVC,
+                  "Destination does not fit within the map region, addr=0x{:016X}, "
+                  "size=0x{:016X}, end_addr=0x{:016X}",
+                  dst_addr, size, dst_end_address);
         return ERR_INVALID_MEMORY_RANGE;
     }
 
     return RESULT_SUCCESS;
 }
+
+enum class ResourceLimitValueType {
+    CurrentValue,
+    LimitValue,
+};
+
+ResultVal<s64> RetrieveResourceLimitValue(Handle resource_limit, u32 resource_type,
+                                          ResourceLimitValueType value_type) {
+    const auto type = static_cast<ResourceType>(resource_type);
+    if (!IsValidResourceType(type)) {
+        LOG_ERROR(Kernel_SVC, "Invalid resource limit type: '{}'", resource_type);
+        return ERR_INVALID_ENUM_VALUE;
+    }
+
+    const auto& kernel = Core::System::GetInstance().Kernel();
+    const auto* const current_process = kernel.CurrentProcess();
+    ASSERT(current_process != nullptr);
+
+    const auto resource_limit_object =
+        current_process->GetHandleTable().Get<ResourceLimit>(resource_limit);
+    if (!resource_limit_object) {
+        LOG_ERROR(Kernel_SVC, "Handle to non-existent resource limit instance used. Handle={:08X}",
+                  resource_limit);
+        return ERR_INVALID_HANDLE;
+    }
+
+    if (value_type == ResourceLimitValueType::CurrentValue) {
+        return MakeResult(resource_limit_object->GetCurrentResourceValue(type));
+    }
+
+    return MakeResult(resource_limit_object->GetMaxResourceValue(type));
+}
 } // Anonymous namespace
 
 /// Set the process heap to a given Size. It can both extend and shrink the heap.
 static ResultCode SetHeapSize(VAddr* heap_addr, u64 heap_size) {
     LOG_TRACE(Kernel_SVC, "called, heap_size=0x{:X}", heap_size);
 
-    // Size must be a multiple of 0x200000 (2MB) and be equal to or less than 4GB.
-    if ((heap_size & 0xFFFFFFFE001FFFFF) != 0) {
+    // Size must be a multiple of 0x200000 (2MB) and be equal to or less than 8GB.
+    if ((heap_size % 0x200000) != 0) {
+        LOG_ERROR(Kernel_SVC, "The heap size is not a multiple of 2MB, heap_size=0x{:016X}",
+                  heap_size);
+        return ERR_INVALID_SIZE;
+    }
+
+    if (heap_size >= 0x200000000) {
+        LOG_ERROR(Kernel_SVC, "The heap size is not less than 8GB, heap_size=0x{:016X}", heap_size);
         return ERR_INVALID_SIZE;
     }
 
@@ -127,20 +200,31 @@ static ResultCode SetMemoryPermission(VAddr addr, u64 size, u32 prot) {
     LOG_TRACE(Kernel_SVC, "called, addr=0x{:X}, size=0x{:X}, prot=0x{:X}", addr, size, prot);
 
     if (!Common::Is4KBAligned(addr)) {
+        LOG_ERROR(Kernel_SVC, "Address is not aligned to 4KB, addr=0x{:016X}", addr);
         return ERR_INVALID_ADDRESS;
     }
 
-    if (size == 0 || !Common::Is4KBAligned(size)) {
+    if (size == 0) {
+        LOG_ERROR(Kernel_SVC, "Size is 0");
+        return ERR_INVALID_SIZE;
+    }
+
+    if (!Common::Is4KBAligned(size)) {
+        LOG_ERROR(Kernel_SVC, "Size is not aligned to 4KB, size=0x{:016X}", size);
         return ERR_INVALID_SIZE;
     }
 
     if (!IsValidAddressRange(addr, size)) {
+        LOG_ERROR(Kernel_SVC, "Region is not a valid address range, addr=0x{:016X}, size=0x{:016X}",
+                  addr, size);
         return ERR_INVALID_ADDRESS_STATE;
     }
 
     const auto permission = static_cast<MemoryPermission>(prot);
     if (permission != MemoryPermission::None && permission != MemoryPermission::Read &&
         permission != MemoryPermission::ReadWrite) {
+        LOG_ERROR(Kernel_SVC, "Invalid memory permission specified, Got memory permission=0x{:08X}",
+                  static_cast<u32>(permission));
         return ERR_INVALID_MEMORY_PERMISSIONS;
     }
 
@@ -148,11 +232,15 @@ static ResultCode SetMemoryPermission(VAddr addr, u64 size, u32 prot) {
     auto& vm_manager = current_process->VMManager();
 
     if (!IsInsideAddressSpace(vm_manager, addr, size)) {
+        LOG_ERROR(Kernel_SVC,
+                  "Source is not within the address space, addr=0x{:016X}, size=0x{:016X}", addr,
+                  size);
         return ERR_INVALID_ADDRESS_STATE;
     }
 
     const VMManager::VMAHandle iter = vm_manager.FindVMA(addr);
     if (iter == vm_manager.vma_map.end()) {
+        LOG_ERROR(Kernel_SVC, "Unable to find VMA for address=0x{:016X}", addr);
         return ERR_INVALID_ADDRESS_STATE;
     }
 
@@ -207,6 +295,9 @@ static ResultCode UnmapMemory(VAddr dst_addr, VAddr src_addr, u64 size) {
 /// Connect to an OS service given the port name, returns the handle to the port to out
 static ResultCode ConnectToNamedPort(Handle* out_handle, VAddr port_name_address) {
     if (!Memory::IsValidVirtualAddress(port_name_address)) {
+        LOG_ERROR(Kernel_SVC,
+                  "Port Name Address is not a valid virtual address, port_name_address=0x{:016X}",
+                  port_name_address);
         return ERR_NOT_FOUND;
     }
 
@@ -214,6 +305,8 @@ static ResultCode ConnectToNamedPort(Handle* out_handle, VAddr port_name_address
     // Read 1 char beyond the max allowed port name to detect names that are too long.
     std::string port_name = Memory::ReadCString(port_name_address, PortNameMaxLength + 1);
     if (port_name.size() > PortNameMaxLength) {
+        LOG_ERROR(Kernel_SVC, "Port name is too long, expected {} but got {}", PortNameMaxLength,
+                  port_name.size());
         return ERR_OUT_OF_RANGE;
     }
 
@@ -262,6 +355,7 @@ static ResultCode GetThreadId(u32* thread_id, Handle thread_handle) {
     const auto& handle_table = Core::CurrentProcess()->GetHandleTable();
     const SharedPtr<Thread> thread = handle_table.Get<Thread>(thread_handle);
     if (!thread) {
+        LOG_ERROR(Kernel_SVC, "Thread handle does not exist, handle=0x{:08X}", thread_handle);
         return ERR_INVALID_HANDLE;
     }
 
@@ -276,6 +370,8 @@ static ResultCode GetProcessId(u32* process_id, Handle process_handle) {
     const auto& handle_table = Core::CurrentProcess()->GetHandleTable();
     const SharedPtr<Process> process = handle_table.Get<Process>(process_handle);
     if (!process) {
+        LOG_ERROR(Kernel_SVC, "Process handle does not exist, process_handle=0x{:08X}",
+                  process_handle);
         return ERR_INVALID_HANDLE;
     }
 
@@ -305,12 +401,18 @@ static ResultCode WaitSynchronization(Handle* index, VAddr handles_address, u64
     LOG_TRACE(Kernel_SVC, "called handles_address=0x{:X}, handle_count={}, nano_seconds={}",
               handles_address, handle_count, nano_seconds);
 
-    if (!Memory::IsValidVirtualAddress(handles_address))
+    if (!Memory::IsValidVirtualAddress(handles_address)) {
+        LOG_ERROR(Kernel_SVC,
+                  "Handle address is not a valid virtual address, handle_address=0x{:016X}",
+                  handles_address);
         return ERR_INVALID_POINTER;
+    }
 
     static constexpr u64 MaxHandles = 0x40;
 
     if (handle_count > MaxHandles) {
+        LOG_ERROR(Kernel_SVC, "Handle count specified is too large, expected {} but got {}",
+                  MaxHandles, handle_count);
         return ERR_OUT_OF_RANGE;
     }
 
@@ -325,6 +427,7 @@ static ResultCode WaitSynchronization(Handle* index, VAddr handles_address, u64
         const auto object = handle_table.Get<WaitObject>(handle);
 
         if (object == nullptr) {
+            LOG_ERROR(Kernel_SVC, "Object is a nullptr");
             return ERR_INVALID_HANDLE;
         }
 
@@ -348,11 +451,13 @@ static ResultCode WaitSynchronization(Handle* index, VAddr handles_address, u64
 
     // If a timeout value of 0 was provided, just return the Timeout error code instead of
     // suspending the thread.
-    if (nano_seconds == 0)
+    if (nano_seconds == 0) {
         return RESULT_TIMEOUT;
+    }
 
-    for (auto& object : objects)
+    for (auto& object : objects) {
         object->AddWaitingThread(thread);
+    }
 
     thread->SetWaitObjects(std::move(objects));
     thread->SetStatus(ThreadStatus::WaitSynchAny);
@@ -373,6 +478,8 @@ static ResultCode CancelSynchronization(Handle thread_handle) {
     const auto& handle_table = Core::CurrentProcess()->GetHandleTable();
     const SharedPtr<Thread> thread = handle_table.Get<Thread>(thread_handle);
     if (!thread) {
+        LOG_ERROR(Kernel_SVC, "Thread handle does not exist, thread_handle=0x{:08X}",
+                  thread_handle);
         return ERR_INVALID_HANDLE;
     }
 
@@ -391,10 +498,13 @@ static ResultCode ArbitrateLock(Handle holding_thread_handle, VAddr mutex_addr,
               holding_thread_handle, mutex_addr, requesting_thread_handle);
 
     if (Memory::IsKernelVirtualAddress(mutex_addr)) {
+        LOG_ERROR(Kernel_SVC, "Mutex Address is a kernel virtual address, mutex_addr={:016X}",
+                  mutex_addr);
         return ERR_INVALID_ADDRESS_STATE;
     }
 
     if (!Common::IsWordAligned(mutex_addr)) {
+        LOG_ERROR(Kernel_SVC, "Mutex Address is not word aligned, mutex_addr={:016X}", mutex_addr);
         return ERR_INVALID_ADDRESS;
     }
 
@@ -408,10 +518,13 @@ static ResultCode ArbitrateUnlock(VAddr mutex_addr) {
     LOG_TRACE(Kernel_SVC, "called mutex_addr=0x{:X}", mutex_addr);
 
     if (Memory::IsKernelVirtualAddress(mutex_addr)) {
+        LOG_ERROR(Kernel_SVC, "Mutex Address is a kernel virtual address, mutex_addr={:016X}",
+                  mutex_addr);
         return ERR_INVALID_ADDRESS_STATE;
     }
 
     if (!Common::IsWordAligned(mutex_addr)) {
+        LOG_ERROR(Kernel_SVC, "Mutex Address is not word aligned, mutex_addr={:016X}", mutex_addr);
         return ERR_INVALID_ADDRESS;
     }
 
@@ -602,10 +715,14 @@ static ResultCode GetInfo(u64* result, u64 info_id, u64 handle, u64 info_sub_id)
         break;
     case GetInfoType::RandomEntropy:
         if (handle != 0) {
+            LOG_ERROR(Kernel_SVC, "Process Handle is non zero, expected 0 result but got {:016X}",
+                      handle);
             return ERR_INVALID_HANDLE;
         }
 
         if (info_sub_id >= Process::RANDOM_ENTROPY_SIZE) {
+            LOG_ERROR(Kernel_SVC, "Entropy size is out of range, expected {} but got {}",
+                      Process::RANDOM_ENTROPY_SIZE, info_sub_id);
             return ERR_INVALID_COMBINATION;
         }
 
@@ -643,12 +760,16 @@ static ResultCode GetInfo(u64* result, u64 info_id, u64 handle, u64 info_sub_id)
     case GetInfoType::ThreadTickCount: {
         constexpr u64 num_cpus = 4;
         if (info_sub_id != 0xFFFFFFFFFFFFFFFF && info_sub_id >= num_cpus) {
+            LOG_ERROR(Kernel_SVC, "Core count is out of range, expected {} but got {}", num_cpus,
+                      info_sub_id);
             return ERR_INVALID_COMBINATION;
         }
 
         const auto thread =
             current_process->GetHandleTable().Get<Thread>(static_cast<Handle>(handle));
         if (!thread) {
+            LOG_ERROR(Kernel_SVC, "Thread handle does not exist, handle=0x{:08X}",
+                      static_cast<Handle>(handle));
             return ERR_INVALID_HANDLE;
         }
 
@@ -671,7 +792,8 @@ static ResultCode GetInfo(u64* result, u64 info_id, u64 handle, u64 info_sub_id)
         break;
     }
     default:
-        UNIMPLEMENTED();
+        LOG_WARNING(Kernel_SVC, "(STUBBED) Unimplemented svcGetInfo id=0x{:016X}", info_id);
+        return ERR_INVALID_ENUM_VALUE;
     }
 
     return RESULT_SUCCESS;
@@ -690,14 +812,22 @@ static ResultCode GetThreadContext(VAddr thread_context, Handle handle) {
     const auto* current_process = Core::CurrentProcess();
     const SharedPtr<Thread> thread = current_process->GetHandleTable().Get<Thread>(handle);
     if (!thread) {
+        LOG_ERROR(Kernel_SVC, "Thread handle does not exist, handle=0x{:08X}", handle);
         return ERR_INVALID_HANDLE;
     }
 
     if (thread->GetOwnerProcess() != current_process) {
+        LOG_ERROR(Kernel_SVC,
+                  "The current process does not own the current thread, thread_handle={:08X} "
+                  "thread_pid={}, "
+                  "current_process_pid={}",
+                  handle, thread->GetOwnerProcess()->GetProcessID(),
+                  current_process->GetProcessID());
         return ERR_INVALID_HANDLE;
     }
 
     if (thread == GetCurrentThread()) {
+        LOG_ERROR(Kernel_SVC, "The thread handle specified is the current running thread");
         return ERR_ALREADY_REGISTERED;
     }
 
@@ -718,9 +848,12 @@ static ResultCode GetThreadContext(VAddr thread_context, Handle handle) {
 
 /// Gets the priority for the specified thread
 static ResultCode GetThreadPriority(u32* priority, Handle handle) {
+    LOG_TRACE(Kernel_SVC, "called");
+
     const auto& handle_table = Core::CurrentProcess()->GetHandleTable();
     const SharedPtr<Thread> thread = handle_table.Get<Thread>(handle);
     if (!thread) {
+        LOG_ERROR(Kernel_SVC, "Thread handle does not exist, handle=0x{:08X}", handle);
         return ERR_INVALID_HANDLE;
     }
 
@@ -730,7 +863,13 @@ static ResultCode GetThreadPriority(u32* priority, Handle handle) {
 
 /// Sets the priority for the specified thread
 static ResultCode SetThreadPriority(Handle handle, u32 priority) {
+    LOG_TRACE(Kernel_SVC, "called");
+
     if (priority > THREADPRIO_LOWEST) {
+        LOG_ERROR(
+            Kernel_SVC,
+            "An invalid priority was specified, expected {} but got {} for thread_handle={:08X}",
+            THREADPRIO_LOWEST, priority, handle);
         return ERR_INVALID_THREAD_PRIORITY;
     }
 
@@ -738,6 +877,7 @@ static ResultCode SetThreadPriority(Handle handle, u32 priority) {
 
     SharedPtr<Thread> thread = current_process->GetHandleTable().Get<Thread>(handle);
     if (!thread) {
+        LOG_ERROR(Kernel_SVC, "Thread handle does not exist, handle=0x{:08X}", handle);
         return ERR_INVALID_HANDLE;
     }
 
@@ -760,32 +900,46 @@ static ResultCode MapSharedMemory(Handle shared_memory_handle, VAddr addr, u64 s
               shared_memory_handle, addr, size, permissions);
 
     if (!Common::Is4KBAligned(addr)) {
+        LOG_ERROR(Kernel_SVC, "Address is not aligned to 4KB, addr=0x{:016X}", addr);
         return ERR_INVALID_ADDRESS;
     }
 
-    if (size == 0 || !Common::Is4KBAligned(size)) {
+    if (size == 0) {
+        LOG_ERROR(Kernel_SVC, "Size is 0");
+        return ERR_INVALID_SIZE;
+    }
+
+    if (!Common::Is4KBAligned(size)) {
+        LOG_ERROR(Kernel_SVC, "Size is not aligned to 4KB, size=0x{:016X}", size);
         return ERR_INVALID_SIZE;
     }
 
     if (!IsValidAddressRange(addr, size)) {
+        LOG_ERROR(Kernel_SVC, "Region is not a valid address range, addr=0x{:016X}, size=0x{:016X}",
+                  addr, size);
         return ERR_INVALID_ADDRESS_STATE;
     }
 
     const auto permissions_type = static_cast<MemoryPermission>(permissions);
     if (permissions_type != MemoryPermission::Read &&
         permissions_type != MemoryPermission::ReadWrite) {
-        LOG_ERROR(Kernel_SVC, "Invalid permissions=0x{:08X}", permissions);
+        LOG_ERROR(Kernel_SVC, "Expected Read or ReadWrite permission but got permissions=0x{:08X}",
+                  permissions);
         return ERR_INVALID_MEMORY_PERMISSIONS;
     }
 
     auto* const current_process = Core::CurrentProcess();
     auto shared_memory = current_process->GetHandleTable().Get<SharedMemory>(shared_memory_handle);
     if (!shared_memory) {
+        LOG_ERROR(Kernel_SVC, "Shared memory does not exist, shared_memory_handle=0x{:08X}",
+                  shared_memory_handle);
         return ERR_INVALID_HANDLE;
     }
 
     const auto& vm_manager = current_process->VMManager();
     if (!vm_manager.IsWithinASLRRegion(addr, size)) {
+        LOG_ERROR(Kernel_SVC, "Region is not within the ASLR region. addr=0x{:016X}, size={:016X}",
+                  addr, size);
         return ERR_INVALID_MEMORY_RANGE;
     }
 
@@ -797,25 +951,38 @@ static ResultCode UnmapSharedMemory(Handle shared_memory_handle, VAddr addr, u64
                 shared_memory_handle, addr, size);
 
     if (!Common::Is4KBAligned(addr)) {
+        LOG_ERROR(Kernel_SVC, "Address is not aligned to 4KB, addr=0x{:016X}", addr);
         return ERR_INVALID_ADDRESS;
     }
 
-    if (size == 0 || !Common::Is4KBAligned(size)) {
+    if (size == 0) {
+        LOG_ERROR(Kernel_SVC, "Size is 0");
+        return ERR_INVALID_SIZE;
+    }
+
+    if (!Common::Is4KBAligned(size)) {
+        LOG_ERROR(Kernel_SVC, "Size is not aligned to 4KB, size=0x{:016X}", size);
         return ERR_INVALID_SIZE;
     }
 
     if (!IsValidAddressRange(addr, size)) {
+        LOG_ERROR(Kernel_SVC, "Region is not a valid address range, addr=0x{:016X}, size=0x{:016X}",
+                  addr, size);
         return ERR_INVALID_ADDRESS_STATE;
     }
 
     auto* const current_process = Core::CurrentProcess();
     auto shared_memory = current_process->GetHandleTable().Get<SharedMemory>(shared_memory_handle);
     if (!shared_memory) {
+        LOG_ERROR(Kernel_SVC, "Shared memory does not exist, shared_memory_handle=0x{:08X}",
+                  shared_memory_handle);
         return ERR_INVALID_HANDLE;
     }
 
     const auto& vm_manager = current_process->VMManager();
     if (!vm_manager.IsWithinASLRRegion(addr, size)) {
+        LOG_ERROR(Kernel_SVC, "Region is not within the ASLR region. addr=0x{:016X}, size={:016X}",
+                  addr, size);
         return ERR_INVALID_MEMORY_RANGE;
     }
 
@@ -825,9 +992,12 @@ static ResultCode UnmapSharedMemory(Handle shared_memory_handle, VAddr addr, u64
 /// Query process memory
 static ResultCode QueryProcessMemory(MemoryInfo* memory_info, PageInfo* /*page_info*/,
                                      Handle process_handle, u64 addr) {
+    LOG_TRACE(Kernel_SVC, "called process=0x{:08X} addr={:X}", process_handle, addr);
     const auto& handle_table = Core::CurrentProcess()->GetHandleTable();
     SharedPtr<Process> process = handle_table.Get<Process>(process_handle);
     if (!process) {
+        LOG_ERROR(Kernel_SVC, "Process handle does not exist, process_handle=0x{:08X}",
+                  process_handle);
         return ERR_INVALID_HANDLE;
     }
     auto vma = process->VMManager().FindVMA(addr);
@@ -843,8 +1013,6 @@ static ResultCode QueryProcessMemory(MemoryInfo* memory_info, PageInfo* /*page_i
         memory_info->size = vma->second.size;
         memory_info->type = static_cast<u32>(vma->second.meminfo_state);
     }
-
-    LOG_TRACE(Kernel_SVC, "called process=0x{:08X} addr={:X}", process_handle, addr);
     return RESULT_SUCCESS;
 }
 
@@ -873,7 +1041,14 @@ static void ExitProcess() {
 /// Creates a new thread
 static ResultCode CreateThread(Handle* out_handle, VAddr entry_point, u64 arg, VAddr stack_top,
                                u32 priority, s32 processor_id) {
+    LOG_TRACE(Kernel_SVC,
+              "called entrypoint=0x{:08X} ({}), arg=0x{:08X}, stacktop=0x{:08X}, "
+              "threadpriority=0x{:08X}, processorid=0x{:08X} : created handle=0x{:08X}",
+              entry_point, name, arg, stack_top, priority, processor_id, *out_handle);
+
     if (priority > THREADPRIO_LOWEST) {
+        LOG_ERROR(Kernel_SVC, "An invalid priority was specified, expected {} but got {}",
+                  THREADPRIO_LOWEST, priority);
         return ERR_INVALID_THREAD_PRIORITY;
     }
 
@@ -904,6 +1079,8 @@ static ResultCode CreateThread(Handle* out_handle, VAddr entry_point, u64 arg, V
 
     const auto new_guest_handle = current_process->GetHandleTable().Create(thread);
     if (new_guest_handle.Failed()) {
+        LOG_ERROR(Kernel_SVC, "Failed to create handle with error=0x{:X}",
+                  new_guest_handle.Code().raw);
         return new_guest_handle.Code();
     }
     thread->SetGuestHandle(*new_guest_handle);
@@ -911,11 +1088,6 @@ static ResultCode CreateThread(Handle* out_handle, VAddr entry_point, u64 arg, V
 
     Core::System::GetInstance().CpuCore(thread->GetProcessorID()).PrepareReschedule();
 
-    LOG_TRACE(Kernel_SVC,
-              "called entrypoint=0x{:08X} ({}), arg=0x{:08X}, stacktop=0x{:08X}, "
-              "threadpriority=0x{:08X}, processorid=0x{:08X} : created handle=0x{:08X}",
-              entry_point, name, arg, stack_top, priority, processor_id, *out_handle);
-
     return RESULT_SUCCESS;
 }
 
@@ -926,6 +1098,8 @@ static ResultCode StartThread(Handle thread_handle) {
     const auto& handle_table = Core::CurrentProcess()->GetHandleTable();
     const SharedPtr<Thread> thread = handle_table.Get<Thread>(thread_handle);
     if (!thread) {
+        LOG_ERROR(Kernel_SVC, "Thread handle does not exist, thread_handle=0x{:08X}",
+                  thread_handle);
         return ERR_INVALID_HANDLE;
     }
 
@@ -1105,10 +1279,12 @@ static ResultCode WaitForAddress(VAddr address, u32 type, s32 value, s64 timeout
                 address, type, value, timeout);
     // If the passed address is a kernel virtual address, return invalid memory state.
     if (Memory::IsKernelVirtualAddress(address)) {
+        LOG_ERROR(Kernel_SVC, "Address is a kernel virtual address, address={:016X}", address);
         return ERR_INVALID_ADDRESS_STATE;
     }
     // If the address is not properly aligned to 4 bytes, return invalid address.
-    if (address % sizeof(u32) != 0) {
+    if (!Common::IsWordAligned(address)) {
+        LOG_ERROR(Kernel_SVC, "Address is not word aligned, address={:016X}", address);
         return ERR_INVALID_ADDRESS;
     }
 
@@ -1120,6 +1296,10 @@ static ResultCode WaitForAddress(VAddr address, u32 type, s32 value, s64 timeout
     case AddressArbiter::ArbitrationType::WaitIfEqual:
         return AddressArbiter::WaitForAddressIfEqual(address, value, timeout);
     default:
+        LOG_ERROR(Kernel_SVC,
+                  "Invalid arbitration type, expected WaitIfLessThan, DecrementAndWaitIfLessThan "
+                  "or WaitIfEqual but got {}",
+                  type);
         return ERR_INVALID_ENUM_VALUE;
     }
 }
@@ -1130,10 +1310,12 @@ static ResultCode SignalToAddress(VAddr address, u32 type, s32 value, s32 num_to
                 address, type, value, num_to_wake);
     // If the passed address is a kernel virtual address, return invalid memory state.
     if (Memory::IsKernelVirtualAddress(address)) {
+        LOG_ERROR(Kernel_SVC, "Address is a kernel virtual address, address={:016X}", address);
         return ERR_INVALID_ADDRESS_STATE;
     }
     // If the address is not properly aligned to 4 bytes, return invalid address.
-    if (address % sizeof(u32) != 0) {
+    if (!Common::IsWordAligned(address)) {
+        LOG_ERROR(Kernel_SVC, "Address is not word aligned, address={:016X}", address);
         return ERR_INVALID_ADDRESS;
     }
 
@@ -1146,12 +1328,18 @@ static ResultCode SignalToAddress(VAddr address, u32 type, s32 value, s32 num_to
         return AddressArbiter::ModifyByWaitingCountAndSignalToAddressIfEqual(address, value,
                                                                              num_to_wake);
     default:
+        LOG_ERROR(Kernel_SVC,
+                  "Invalid signal type, expected Signal, IncrementAndSignalIfEqual "
+                  "or ModifyByWaitingCountAndSignalIfEqual but got {}",
+                  type);
         return ERR_INVALID_ENUM_VALUE;
     }
 }
 
 /// This returns the total CPU ticks elapsed since the CPU was powered-on
 static u64 GetSystemTick() {
+    LOG_TRACE(Kernel_SVC, "called");
+
     const u64 result{CoreTiming::GetTicks()};
 
     // Advance time to defeat dumb games that busy-wait for the frame to end.
@@ -1225,6 +1413,8 @@ static ResultCode GetThreadCoreMask(Handle thread_handle, u32* core, u64* mask)
     const auto& handle_table = Core::CurrentProcess()->GetHandleTable();
     const SharedPtr<Thread> thread = handle_table.Get<Thread>(thread_handle);
     if (!thread) {
+        LOG_ERROR(Kernel_SVC, "Thread handle does not exist, thread_handle=0x{:08X}",
+                  thread_handle);
         return ERR_INVALID_HANDLE;
     }
 
@@ -1235,12 +1425,14 @@ static ResultCode GetThreadCoreMask(Handle thread_handle, u32* core, u64* mask)
 }
 
 static ResultCode SetThreadCoreMask(Handle thread_handle, u32 core, u64 mask) {
-    LOG_DEBUG(Kernel_SVC, "called, handle=0x{:08X}, mask=0x{:16X}, core=0x{:X}", thread_handle,
+    LOG_DEBUG(Kernel_SVC, "called, handle=0x{:08X}, mask=0x{:016X}, core=0x{:X}", thread_handle,
               mask, core);
 
     const auto& handle_table = Core::CurrentProcess()->GetHandleTable();
     const SharedPtr<Thread> thread = handle_table.Get<Thread>(thread_handle);
     if (!thread) {
+        LOG_ERROR(Kernel_SVC, "Thread handle does not exist, thread_handle=0x{:08X}",
+                  thread_handle);
         return ERR_INVALID_HANDLE;
     }
 
@@ -1255,6 +1447,7 @@ static ResultCode SetThreadCoreMask(Handle thread_handle, u32 core, u64 mask) {
     }
 
     if (mask == 0) {
+        LOG_ERROR(Kernel_SVC, "Mask is 0");
         return ERR_INVALID_COMBINATION;
     }
 
@@ -1264,11 +1457,14 @@ static ResultCode SetThreadCoreMask(Handle thread_handle, u32 core, u64 mask) {
     if (core == OnlyChangeMask) {
         core = thread->GetIdealCore();
     } else if (core >= Core::NUM_CPU_CORES && core != static_cast<u32>(-1)) {
+        LOG_ERROR(Kernel_SVC, "Invalid core specified, got {}", core);
         return ERR_INVALID_PROCESSOR_ID;
     }
 
     // Error out if the input core isn't enabled in the input mask.
     if (core < Core::NUM_CPU_CORES && (mask & (1ull << core)) == 0) {
+        LOG_ERROR(Kernel_SVC, "Core is not enabled for the current mask, core={}, mask={:016X}",
+                  core, mask);
         return ERR_INVALID_COMBINATION;
     }
 
@@ -1281,21 +1477,36 @@ static ResultCode CreateSharedMemory(Handle* handle, u64 size, u32 local_permiss
                                      u32 remote_permissions) {
     LOG_TRACE(Kernel_SVC, "called, size=0x{:X}, localPerms=0x{:08X}, remotePerms=0x{:08X}", size,
               local_permissions, remote_permissions);
+    if (size == 0) {
+        LOG_ERROR(Kernel_SVC, "Size is 0");
+        return ERR_INVALID_SIZE;
+    }
+    if (!Common::Is4KBAligned(size)) {
+        LOG_ERROR(Kernel_SVC, "Size is not aligned to 4KB, 0x{:016X}", size);
+        return ERR_INVALID_SIZE;
+    }
 
-    // Size must be a multiple of 4KB and be less than or equal to
-    // approx. 8 GB (actually (1GB - 512B) * 8)
-    if (size == 0 || (size & 0xFFFFFFFE00000FFF) != 0) {
+    if (size >= MAIN_MEMORY_SIZE) {
+        LOG_ERROR(Kernel_SVC, "Size is not less than 8GB, 0x{:016X}", size);
         return ERR_INVALID_SIZE;
     }
 
     const auto local_perms = static_cast<MemoryPermission>(local_permissions);
     if (local_perms != MemoryPermission::Read && local_perms != MemoryPermission::ReadWrite) {
+        LOG_ERROR(Kernel_SVC,
+                  "Invalid local memory permissions, expected Read or ReadWrite but got "
+                  "local_permissions={}",
+                  static_cast<u32>(local_permissions));
         return ERR_INVALID_MEMORY_PERMISSIONS;
     }
 
     const auto remote_perms = static_cast<MemoryPermission>(remote_permissions);
     if (remote_perms != MemoryPermission::Read && remote_perms != MemoryPermission::ReadWrite &&
         remote_perms != MemoryPermission::DontCare) {
+        LOG_ERROR(Kernel_SVC,
+                  "Invalid remote memory permissions, expected Read, ReadWrite or DontCare but got "
+                  "remote_permissions={}",
+                  static_cast<u32>(remote_permissions));
         return ERR_INVALID_MEMORY_PERMISSIONS;
     }
 
@@ -1315,6 +1526,7 @@ static ResultCode ClearEvent(Handle handle) {
     const auto& handle_table = Core::CurrentProcess()->GetHandleTable();
     SharedPtr<Event> evt = handle_table.Get<Event>(handle);
     if (evt == nullptr) {
+        LOG_ERROR(Kernel_SVC, "Event handle does not exist, handle=0x{:08X}", handle);
         return ERR_INVALID_HANDLE;
     }
 
@@ -1333,11 +1545,14 @@ static ResultCode GetProcessInfo(u64* out, Handle process_handle, u32 type) {
     const auto& handle_table = Core::CurrentProcess()->GetHandleTable();
     const auto process = handle_table.Get<Process>(process_handle);
     if (!process) {
+        LOG_ERROR(Kernel_SVC, "Process handle does not exist, process_handle=0x{:08X}",
+                  process_handle);
         return ERR_INVALID_HANDLE;
     }
 
     const auto info_type = static_cast<InfoType>(type);
     if (info_type != InfoType::Status) {
+        LOG_ERROR(Kernel_SVC, "Expected info_type to be Status but got {} instead", type);
         return ERR_INVALID_ENUM_VALUE;
     }
 
@@ -1345,6 +1560,87 @@ static ResultCode GetProcessInfo(u64* out, Handle process_handle, u32 type) {
     return RESULT_SUCCESS;
 }
 
+static ResultCode CreateResourceLimit(Handle* out_handle) {
+    LOG_DEBUG(Kernel_SVC, "called");
+
+    auto& kernel = Core::System::GetInstance().Kernel();
+    auto resource_limit = ResourceLimit::Create(kernel);
+
+    auto* const current_process = kernel.CurrentProcess();
+    ASSERT(current_process != nullptr);
+
+    const auto handle = current_process->GetHandleTable().Create(std::move(resource_limit));
+    if (handle.Failed()) {
+        return handle.Code();
+    }
+
+    *out_handle = *handle;
+    return RESULT_SUCCESS;
+}
+
+static ResultCode GetResourceLimitLimitValue(u64* out_value, Handle resource_limit,
+                                             u32 resource_type) {
+    LOG_DEBUG(Kernel_SVC, "called. Handle={:08X}, Resource type={}", resource_limit, resource_type);
+
+    const auto limit_value = RetrieveResourceLimitValue(resource_limit, resource_type,
+                                                        ResourceLimitValueType::LimitValue);
+    if (limit_value.Failed()) {
+        return limit_value.Code();
+    }
+
+    *out_value = static_cast<u64>(*limit_value);
+    return RESULT_SUCCESS;
+}
+
+static ResultCode GetResourceLimitCurrentValue(u64* out_value, Handle resource_limit,
+                                               u32 resource_type) {
+    LOG_DEBUG(Kernel_SVC, "called. Handle={:08X}, Resource type={}", resource_limit, resource_type);
+
+    const auto current_value = RetrieveResourceLimitValue(resource_limit, resource_type,
+                                                          ResourceLimitValueType::CurrentValue);
+    if (current_value.Failed()) {
+        return current_value.Code();
+    }
+
+    *out_value = static_cast<u64>(*current_value);
+    return RESULT_SUCCESS;
+}
+
+static ResultCode SetResourceLimitLimitValue(Handle resource_limit, u32 resource_type, u64 value) {
+    LOG_DEBUG(Kernel_SVC, "called. Handle={:08X}, Resource type={}, Value={}", resource_limit,
+              resource_type, value);
+
+    const auto type = static_cast<ResourceType>(resource_type);
+    if (!IsValidResourceType(type)) {
+        LOG_ERROR(Kernel_SVC, "Invalid resource limit type: '{}'", resource_type);
+        return ERR_INVALID_ENUM_VALUE;
+    }
+
+    auto& kernel = Core::System::GetInstance().Kernel();
+    auto* const current_process = kernel.CurrentProcess();
+    ASSERT(current_process != nullptr);
+
+    auto resource_limit_object =
+        current_process->GetHandleTable().Get<ResourceLimit>(resource_limit);
+    if (!resource_limit_object) {
+        LOG_ERROR(Kernel_SVC, "Handle to non-existent resource limit instance used. Handle={:08X}",
+                  resource_limit);
+        return ERR_INVALID_HANDLE;
+    }
+
+    const auto set_result = resource_limit_object->SetLimitValue(type, static_cast<s64>(value));
+    if (set_result.IsError()) {
+        LOG_ERROR(
+            Kernel_SVC,
+            "Attempted to lower resource limit ({}) for category '{}' below its current value ({})",
+            resource_limit_object->GetMaxResourceValue(type), resource_type,
+            resource_limit_object->GetCurrentResourceValue(type));
+        return set_result;
+    }
+
+    return RESULT_SUCCESS;
+}
+
 namespace {
 struct FunctionDef {
     using Func = void();
@@ -1404,8 +1700,8 @@ static const FunctionDef SVC_Table[] = {
     {0x2D, nullptr, "UnmapPhysicalMemory"},
     {0x2E, nullptr, "GetFutureThreadInfo"},
     {0x2F, nullptr, "GetLastThreadInfo"},
-    {0x30, nullptr, "GetResourceLimitLimitValue"},
-    {0x31, nullptr, "GetResourceLimitCurrentValue"},
+    {0x30, SvcWrap<GetResourceLimitLimitValue>, "GetResourceLimitLimitValue"},
+    {0x31, SvcWrap<GetResourceLimitCurrentValue>, "GetResourceLimitCurrentValue"},
     {0x32, SvcWrap<SetThreadActivity>, "SetThreadActivity"},
     {0x33, SvcWrap<GetThreadContext>, "GetThreadContext"},
     {0x34, SvcWrap<WaitForAddress>, "WaitForAddress"},
@@ -1481,8 +1777,8 @@ static const FunctionDef SVC_Table[] = {
     {0x7A, nullptr, "StartProcess"},
     {0x7B, nullptr, "TerminateProcess"},
     {0x7C, SvcWrap<GetProcessInfo>, "GetProcessInfo"},
-    {0x7D, nullptr, "CreateResourceLimit"},
-    {0x7E, nullptr, "SetResourceLimitLimitValue"},
+    {0x7D, SvcWrap<CreateResourceLimit>, "CreateResourceLimit"},
+    {0x7E, SvcWrap<SetResourceLimitLimitValue>, "SetResourceLimitLimitValue"},
     {0x7F, nullptr, "CallSecureMonitor"},
 };
 

src/core/hle/kernel/svc_wrap.h

@@ -43,6 +43,14 @@ void SvcWrap() {
     FuncReturn(func(static_cast<u32>(Param(0)), static_cast<u32>(Param(1))).raw);
 }
 
+template <ResultCode func(u32*)>
+void SvcWrap() {
+    u32 param = 0;
+    const u32 retval = func(&param).raw;
+    Core::CurrentArmInterface().SetReg(1, param);
+    FuncReturn(retval);
+}
+
 template <ResultCode func(u32*, u32)>
 void SvcWrap() {
     u32 param_1 = 0;

src/core/hle/service/acc/acc.cpp

@@ -21,17 +21,6 @@
 
 namespace Service::Account {
 
-// TODO: RE this structure
-struct UserData {
-    INSERT_PADDING_WORDS(1);
-    u32 icon_id;
-    u8 bg_color_id;
-    INSERT_PADDING_BYTES(0x7);
-    INSERT_PADDING_BYTES(0x10);
-    INSERT_PADDING_BYTES(0x60);
-};
-static_assert(sizeof(UserData) == 0x80, "UserData structure has incorrect size");
-
 // Smallest JPEG https://github.com/mathiasbynens/small/blob/master/jpeg.jpg
 // used as a backup should the one on disk not exist
 constexpr u32 backup_jpeg_size = 107;
@@ -72,9 +61,11 @@ private:
     void Get(Kernel::HLERequestContext& ctx) {
         LOG_INFO(Service_ACC, "called user_id={}", user_id.Format());
         ProfileBase profile_base{};
-        std::array<u8, MAX_DATA> data{};
+        ProfileData data{};
         if (profile_manager.GetProfileBaseAndData(user_id, profile_base, data)) {
-            ctx.WriteBuffer(data);
+            std::array<u8, sizeof(ProfileData)> raw_data;
+            std::memcpy(raw_data.data(), &data, sizeof(ProfileData));
+            ctx.WriteBuffer(raw_data);
             IPC::ResponseBuilder rb{ctx, 16};
             rb.Push(RESULT_SUCCESS);
             rb.PushRaw(profile_base);
@@ -216,10 +207,11 @@ void Module::Interface::GetLastOpenedUser(Kernel::HLERequestContext& ctx) {
 void Module::Interface::GetProfile(Kernel::HLERequestContext& ctx) {
     IPC::RequestParser rp{ctx};
     UUID user_id = rp.PopRaw<UUID>();
+    LOG_DEBUG(Service_ACC, "called user_id={}", user_id.Format());
+
     IPC::ResponseBuilder rb{ctx, 2, 0, 1};
     rb.Push(RESULT_SUCCESS);
     rb.PushIpcInterface<IProfile>(user_id, *profile_manager);
-    LOG_DEBUG(Service_ACC, "called user_id={}", user_id.Format());
 }
 
 void Module::Interface::IsUserRegistrationRequestPermitted(Kernel::HLERequestContext& ctx) {
@@ -236,10 +228,10 @@ void Module::Interface::InitializeApplicationInfo(Kernel::HLERequestContext& ctx
 }
 
 void Module::Interface::GetBaasAccountManagerForApplication(Kernel::HLERequestContext& ctx) {
+    LOG_DEBUG(Service_ACC, "called");
     IPC::ResponseBuilder rb{ctx, 2, 0, 1};
     rb.Push(RESULT_SUCCESS);
     rb.PushIpcInterface<IManagerForApplication>();
-    LOG_DEBUG(Service_ACC, "called");
 }
 
 void Module::Interface::TrySelectUserWithoutInteraction(Kernel::HLERequestContext& ctx) {

src/core/hle/service/acc/profile_manager.cpp

@@ -18,7 +18,7 @@ struct UserRaw {
     UUID uuid2;
     u64 timestamp;
     ProfileUsername username;
-    INSERT_PADDING_BYTES(0x80);
+    ProfileData extra_data;
 };
 static_assert(sizeof(UserRaw) == 0xC8, "UserRaw has incorrect size.");
 
@@ -346,7 +346,7 @@ void ProfileManager::ParseUserSaveFile() {
             continue;
         }
 
-        AddUser({user.uuid, user.username, user.timestamp, {}, false});
+        AddUser({user.uuid, user.username, user.timestamp, user.extra_data, false});
     }
 
     std::stable_partition(profiles.begin(), profiles.end(),
@@ -361,6 +361,7 @@ void ProfileManager::WriteUserSaveFile() {
         raw.users[i].uuid2 = profiles[i].user_uuid;
         raw.users[i].uuid = profiles[i].user_uuid;
         raw.users[i].timestamp = profiles[i].creation_time;
+        raw.users[i].extra_data = profiles[i].data;
     }
 
     const auto raw_path =

src/core/hle/service/acc/profile_manager.h

@@ -13,7 +13,6 @@
 
 namespace Service::Account {
 constexpr std::size_t MAX_USERS = 8;
-constexpr std::size_t MAX_DATA = 128;
 constexpr u128 INVALID_UUID{{0, 0}};
 
 struct UUID {
@@ -50,9 +49,20 @@ static_assert(sizeof(UUID) == 16, "UUID is an invalid size!");
 
 constexpr std::size_t profile_username_size = 32;
 using ProfileUsername = std::array<u8, profile_username_size>;
-using ProfileData = std::array<u8, MAX_DATA>;
 using UserIDArray = std::array<UUID, MAX_USERS>;
 
+/// Contains extra data related to a user.
+/// TODO: RE this structure
+struct ProfileData {
+    INSERT_PADDING_WORDS(1);
+    u32 icon_id;
+    u8 bg_color_id;
+    INSERT_PADDING_BYTES(0x7);
+    INSERT_PADDING_BYTES(0x10);
+    INSERT_PADDING_BYTES(0x60);
+};
+static_assert(sizeof(ProfileData) == 0x80, "ProfileData structure has incorrect size");
+
 /// This holds general information about a users profile. This is where we store all the information
 /// based on a specific user
 struct ProfileInfo {

src/core/hle/service/am/am.cpp

@@ -217,6 +217,7 @@ ISelfController::~ISelfController() = default;
 void ISelfController::SetFocusHandlingMode(Kernel::HLERequestContext& ctx) {
     // Takes 3 input u8s with each field located immediately after the previous
     // u8, these are bool flags. No output.
+    LOG_WARNING(Service_AM, "(STUBBED) called");
 
     IPC::RequestParser rp{ctx};
 
@@ -229,44 +230,40 @@ void ISelfController::SetFocusHandlingMode(Kernel::HLERequestContext& ctx) {
 
     IPC::ResponseBuilder rb{ctx, 2};
     rb.Push(RESULT_SUCCESS);
-
-    LOG_WARNING(Service_AM, "(STUBBED) called");
 }
 
 void ISelfController::SetRestartMessageEnabled(Kernel::HLERequestContext& ctx) {
+    LOG_WARNING(Service_AM, "(STUBBED) called");
+
     IPC::ResponseBuilder rb{ctx, 2};
     rb.Push(RESULT_SUCCESS);
-
-    LOG_WARNING(Service_AM, "(STUBBED) called");
 }
 
 void ISelfController::SetPerformanceModeChangedNotification(Kernel::HLERequestContext& ctx) {
     IPC::RequestParser rp{ctx};
 
     bool flag = rp.Pop<bool>();
+    LOG_WARNING(Service_AM, "(STUBBED) called flag={}", flag);
 
     IPC::ResponseBuilder rb{ctx, 2};
     rb.Push(RESULT_SUCCESS);
-
-    LOG_WARNING(Service_AM, "(STUBBED) called flag={}", flag);
 }
 
 void ISelfController::SetScreenShotPermission(Kernel::HLERequestContext& ctx) {
+    LOG_WARNING(Service_AM, "(STUBBED) called");
+
     IPC::ResponseBuilder rb{ctx, 2};
     rb.Push(RESULT_SUCCESS);
-
-    LOG_WARNING(Service_AM, "(STUBBED) called");
 }
 
 void ISelfController::SetOperationModeChangedNotification(Kernel::HLERequestContext& ctx) {
     IPC::RequestParser rp{ctx};
 
     bool flag = rp.Pop<bool>();
+    LOG_WARNING(Service_AM, "(STUBBED) called flag={}", flag);
 
     IPC::ResponseBuilder rb{ctx, 2};
     rb.Push(RESULT_SUCCESS);
-
-    LOG_WARNING(Service_AM, "(STUBBED) called flag={}", flag);
 }
 
 void ISelfController::SetOutOfFocusSuspendingEnabled(Kernel::HLERequestContext& ctx) {
@@ -275,45 +272,45 @@ void ISelfController::SetOutOfFocusSuspendingEnabled(Kernel::HLERequestContext&
     IPC::RequestParser rp{ctx};
 
     bool enabled = rp.Pop<bool>();
+    LOG_WARNING(Service_AM, "(STUBBED) called enabled={}", enabled);
 
     IPC::ResponseBuilder rb{ctx, 2};
     rb.Push(RESULT_SUCCESS);
-
-    LOG_WARNING(Service_AM, "(STUBBED) called enabled={}", enabled);
 }
 
 void ISelfController::LockExit(Kernel::HLERequestContext& ctx) {
+    LOG_WARNING(Service_AM, "(STUBBED) called");
+
     IPC::ResponseBuilder rb{ctx, 2};
     rb.Push(RESULT_SUCCESS);
-
-    LOG_WARNING(Service_AM, "(STUBBED) called");
 }
 
 void ISelfController::UnlockExit(Kernel::HLERequestContext& ctx) {
+    LOG_WARNING(Service_AM, "(STUBBED) called");
+
     IPC::ResponseBuilder rb{ctx, 2};
     rb.Push(RESULT_SUCCESS);
-
-    LOG_WARNING(Service_AM, "(STUBBED) called");
 }
 
 void ISelfController::GetLibraryAppletLaunchableEvent(Kernel::HLERequestContext& ctx) {
+    LOG_WARNING(Service_AM, "(STUBBED) called");
+
     launchable_event->Signal();
 
     IPC::ResponseBuilder rb{ctx, 2, 1};
     rb.Push(RESULT_SUCCESS);
     rb.PushCopyObjects(launchable_event);
-
-    LOG_WARNING(Service_AM, "(STUBBED) called");
 }
 
 void ISelfController::SetScreenShotImageOrientation(Kernel::HLERequestContext& ctx) {
+    LOG_WARNING(Service_AM, "(STUBBED) called");
+
     IPC::ResponseBuilder rb{ctx, 2};
     rb.Push(RESULT_SUCCESS);
-
-    LOG_WARNING(Service_AM, "(STUBBED) called");
 }
 
 void ISelfController::CreateManagedDisplayLayer(Kernel::HLERequestContext& ctx) {
+    LOG_WARNING(Service_AM, "(STUBBED) called");
     // TODO(Subv): Find out how AM determines the display to use, for now just
     // create the layer in the Default display.
     u64 display_id = nvflinger->OpenDisplay("Default");
@@ -322,32 +319,31 @@ void ISelfController::CreateManagedDisplayLayer(Kernel::HLERequestContext& ctx)
     IPC::ResponseBuilder rb{ctx, 4};
     rb.Push(RESULT_SUCCESS);
     rb.Push(layer_id);
-
-    LOG_WARNING(Service_AM, "(STUBBED) called");
 }
 
 void ISelfController::SetHandlesRequestToDisplay(Kernel::HLERequestContext& ctx) {
+    LOG_WARNING(Service_AM, "(STUBBED) called");
+
     IPC::ResponseBuilder rb{ctx, 2};
     rb.Push(RESULT_SUCCESS);
-
-    LOG_WARNING(Service_AM, "(STUBBED) called");
 }
 
 void ISelfController::SetIdleTimeDetectionExtension(Kernel::HLERequestContext& ctx) {
     IPC::RequestParser rp{ctx};
     idle_time_detection_extension = rp.Pop<u32>();
+    LOG_WARNING(Service_AM, "(STUBBED) called idle_time_detection_extension={}",
+                idle_time_detection_extension);
+
     IPC::ResponseBuilder rb{ctx, 2};
     rb.Push(RESULT_SUCCESS);
-
-    LOG_WARNING(Service_AM, "(STUBBED) called");
 }
 
 void ISelfController::GetIdleTimeDetectionExtension(Kernel::HLERequestContext& ctx) {
+    LOG_WARNING(Service_AM, "(STUBBED) called");
+
     IPC::ResponseBuilder rb{ctx, 3};
     rb.Push(RESULT_SUCCESS);
     rb.Push<u32>(idle_time_detection_extension);
-
-    LOG_WARNING(Service_AM, "(STUBBED) called");
 }
 
 AppletMessageQueue::AppletMessageQueue() {
@@ -438,47 +434,49 @@ ICommonStateGetter::ICommonStateGetter(std::shared_ptr<AppletMessageQueue> msg_q
 ICommonStateGetter::~ICommonStateGetter() = default;
 
 void ICommonStateGetter::GetBootMode(Kernel::HLERequestContext& ctx) {
+    LOG_DEBUG(Service_AM, "called");
+
     IPC::ResponseBuilder rb{ctx, 3};
     rb.Push(RESULT_SUCCESS);
 
     rb.Push<u8>(static_cast<u8>(Service::PM::SystemBootMode::Normal)); // Normal boot mode
-
-    LOG_DEBUG(Service_AM, "called");
 }
 
 void ICommonStateGetter::GetEventHandle(Kernel::HLERequestContext& ctx) {
+    LOG_DEBUG(Service_AM, "called");
+
     IPC::ResponseBuilder rb{ctx, 2, 1};
     rb.Push(RESULT_SUCCESS);
     rb.PushCopyObjects(msg_queue->GetMesssageRecieveEvent());
-
-    LOG_DEBUG(Service_AM, "called");
 }
 
 void ICommonStateGetter::ReceiveMessage(Kernel::HLERequestContext& ctx) {
+    LOG_DEBUG(Service_AM, "called");
+
     IPC::ResponseBuilder rb{ctx, 3};
     rb.Push(RESULT_SUCCESS);
     rb.PushEnum<AppletMessageQueue::AppletMessage>(msg_queue->PopMessage());
-
-    LOG_DEBUG(Service_AM, "called");
 }
 
 void ICommonStateGetter::GetCurrentFocusState(Kernel::HLERequestContext& ctx) {
+    LOG_WARNING(Service_AM, "(STUBBED) called");
+
     IPC::ResponseBuilder rb{ctx, 3};
     rb.Push(RESULT_SUCCESS);
     rb.Push(static_cast<u8>(FocusState::InFocus));
-
-    LOG_WARNING(Service_AM, "(STUBBED) called");
 }
 
 void ICommonStateGetter::GetDefaultDisplayResolutionChangeEvent(Kernel::HLERequestContext& ctx) {
+    LOG_DEBUG(Service_AM, "called");
+
     IPC::ResponseBuilder rb{ctx, 2, 1};
     rb.Push(RESULT_SUCCESS);
     rb.PushCopyObjects(msg_queue->GetOperationModeChangedEvent());
-
-    LOG_DEBUG(Service_AM, "called");
 }
 
 void ICommonStateGetter::GetDefaultDisplayResolution(Kernel::HLERequestContext& ctx) {
+    LOG_DEBUG(Service_AM, "called");
+
     IPC::ResponseBuilder rb{ctx, 4};
     rb.Push(RESULT_SUCCESS);
 
@@ -493,8 +491,6 @@ void ICommonStateGetter::GetDefaultDisplayResolution(Kernel::HLERequestContext&
         rb.Push(static_cast<u32>(Service::VI::DisplayResolution::UndockedHeight) *
                 static_cast<u32>(Settings::values.resolution_factor));
     }
-
-    LOG_DEBUG(Service_AM, "called");
 }
 
 IStorage::IStorage(std::vector<u8> buffer)
@@ -517,21 +513,21 @@ const std::vector<u8>& IStorage::GetData() const {
 
 void ICommonStateGetter::GetOperationMode(Kernel::HLERequestContext& ctx) {
     const bool use_docked_mode{Settings::values.use_docked_mode};
+    LOG_DEBUG(Service_AM, "called, use_docked_mode={}", use_docked_mode);
+
     IPC::ResponseBuilder rb{ctx, 3};
     rb.Push(RESULT_SUCCESS);
     rb.Push(static_cast<u8>(use_docked_mode ? OperationMode::Docked : OperationMode::Handheld));
-
-    LOG_DEBUG(Service_AM, "called");
 }
 
 void ICommonStateGetter::GetPerformanceMode(Kernel::HLERequestContext& ctx) {
     const bool use_docked_mode{Settings::values.use_docked_mode};
+    LOG_DEBUG(Service_AM, "called, use_docked_mode={}", use_docked_mode);
+
     IPC::ResponseBuilder rb{ctx, 3};
     rb.Push(RESULT_SUCCESS);
     rb.Push(static_cast<u32>(use_docked_mode ? APM::PerformanceMode::Docked
                                              : APM::PerformanceMode::Handheld));
-
-    LOG_DEBUG(Service_AM, "called");
 }
 
 class ILibraryAppletAccessor final : public ServiceFramework<ILibraryAppletAccessor> {
@@ -566,32 +562,34 @@ public:
 
 private:
     void GetAppletStateChangedEvent(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_AM, "called");
+
         const auto event = applet->GetBroker().GetStateChangedEvent();
         event->Signal();
 
         IPC::ResponseBuilder rb{ctx, 2, 1};
         rb.Push(RESULT_SUCCESS);
         rb.PushCopyObjects(event);
-
-        LOG_DEBUG(Service_AM, "called");
     }
 
     void IsCompleted(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_AM, "called");
+
         IPC::ResponseBuilder rb{ctx, 3};
         rb.Push(RESULT_SUCCESS);
         rb.Push<u32>(applet->TransactionComplete());
-
-        LOG_DEBUG(Service_AM, "called");
     }
 
     void GetResult(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_AM, "called");
+
         IPC::ResponseBuilder rb{ctx, 2};
         rb.Push(applet->GetStatus());
-
-        LOG_DEBUG(Service_AM, "called");
     }
 
     void Start(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_AM, "called");
+
         ASSERT(applet != nullptr);
 
         applet->Initialize();
@@ -599,36 +597,39 @@ private:
 
         IPC::ResponseBuilder rb{ctx, 2};
         rb.Push(RESULT_SUCCESS);
-
-        LOG_DEBUG(Service_AM, "called");
     }
 
     void PushInData(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_AM, "called");
+
         IPC::RequestParser rp{ctx};
         applet->GetBroker().PushNormalDataFromGame(*rp.PopIpcInterface<IStorage>());
 
         IPC::ResponseBuilder rb{ctx, 2};
         rb.Push(RESULT_SUCCESS);
-
-        LOG_DEBUG(Service_AM, "called");
     }
 
     void PopOutData(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_AM, "called");
+
         IPC::ResponseBuilder rb{ctx, 2, 0, 1};
 
         const auto storage = applet->GetBroker().PopNormalDataToGame();
         if (storage == nullptr) {
+            LOG_ERROR(Service_AM,
+                      "storage is a nullptr. There is no data in the current normal channel");
+
             rb.Push(ERR_NO_DATA_IN_CHANNEL);
             return;
         }
 
         rb.Push(RESULT_SUCCESS);
         rb.PushIpcInterface<IStorage>(std::move(*storage));
-
-        LOG_DEBUG(Service_AM, "called");
     }
 
     void PushInteractiveInData(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_AM, "called");
+
         IPC::RequestParser rp{ctx};
         applet->GetBroker().PushInteractiveDataFromGame(*rp.PopIpcInterface<IStorage>());
 
@@ -638,51 +639,52 @@ private:
 
         IPC::ResponseBuilder rb{ctx, 2};
         rb.Push(RESULT_SUCCESS);
-
-        LOG_DEBUG(Service_AM, "called");
     }
 
     void PopInteractiveOutData(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_AM, "called");
+
         IPC::ResponseBuilder rb{ctx, 2, 0, 1};
 
         const auto storage = applet->GetBroker().PopInteractiveDataToGame();
         if (storage == nullptr) {
+            LOG_ERROR(Service_AM,
+                      "storage is a nullptr. There is no data in the current interactive channel");
+
             rb.Push(ERR_NO_DATA_IN_CHANNEL);
             return;
         }
 
         rb.Push(RESULT_SUCCESS);
         rb.PushIpcInterface<IStorage>(std::move(*storage));
-
-        LOG_DEBUG(Service_AM, "called");
     }
 
     void GetPopOutDataEvent(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_AM, "called");
+
         IPC::ResponseBuilder rb{ctx, 2, 1};
         rb.Push(RESULT_SUCCESS);
         rb.PushCopyObjects(applet->GetBroker().GetNormalDataEvent());
-
-        LOG_DEBUG(Service_AM, "called");
     }
 
     void GetPopInteractiveOutDataEvent(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_AM, "called");
+
         IPC::ResponseBuilder rb{ctx, 2, 1};
         rb.Push(RESULT_SUCCESS);
         rb.PushCopyObjects(applet->GetBroker().GetInteractiveDataEvent());
-
-        LOG_DEBUG(Service_AM, "called");
     }
 
     std::shared_ptr<Applets::Applet> applet;
 };
 
 void IStorage::Open(Kernel::HLERequestContext& ctx) {
+    LOG_DEBUG(Service_AM, "called");
+
     IPC::ResponseBuilder rb{ctx, 2, 0, 1};
 
     rb.Push(RESULT_SUCCESS);
     rb.PushIpcInterface<IStorageAccessor>(*this);
-
-    LOG_DEBUG(Service_AM, "called");
 }
 
 IStorageAccessor::IStorageAccessor(IStorage& storage)
@@ -701,21 +703,27 @@ IStorageAccessor::IStorageAccessor(IStorage& storage)
 IStorageAccessor::~IStorageAccessor() = default;
 
 void IStorageAccessor::GetSize(Kernel::HLERequestContext& ctx) {
+    LOG_DEBUG(Service_AM, "called");
+
     IPC::ResponseBuilder rb{ctx, 4};
 
     rb.Push(RESULT_SUCCESS);
     rb.Push(static_cast<u64>(backing.buffer.size()));
-
-    LOG_DEBUG(Service_AM, "called");
 }
 
 void IStorageAccessor::Write(Kernel::HLERequestContext& ctx) {
     IPC::RequestParser rp{ctx};
 
     const u64 offset{rp.Pop<u64>()};
+    LOG_DEBUG(Service_AM, "called, offset={}", offset);
+
     const std::vector<u8> data{ctx.ReadBuffer()};
 
     if (data.size() > backing.buffer.size() - offset) {
+        LOG_ERROR(Service_AM,
+                  "offset is out of bounds, backing_buffer_sz={}, data_size={}, offset={}",
+                  backing.buffer.size(), data.size(), offset);
+
         IPC::ResponseBuilder rb{ctx, 2};
         rb.Push(ERR_SIZE_OUT_OF_BOUNDS);
     }
@@ -724,17 +732,20 @@ void IStorageAccessor::Write(Kernel::HLERequestContext& ctx) {
 
     IPC::ResponseBuilder rb{ctx, 2};
     rb.Push(RESULT_SUCCESS);
-
-    LOG_DEBUG(Service_AM, "called, offset={}", offset);
 }
 
 void IStorageAccessor::Read(Kernel::HLERequestContext& ctx) {
     IPC::RequestParser rp{ctx};
 
     const u64 offset{rp.Pop<u64>()};
+    LOG_DEBUG(Service_AM, "called, offset={}", offset);
+
     const std::size_t size{ctx.GetWriteBufferSize()};
 
     if (size > backing.buffer.size() - offset) {
+        LOG_ERROR(Service_AM, "offset is out of bounds, backing_buffer_sz={}, size={}, offset={}",
+                  backing.buffer.size(), size, offset);
+
         IPC::ResponseBuilder rb{ctx, 2};
         rb.Push(ERR_SIZE_OUT_OF_BOUNDS);
     }
@@ -743,8 +754,6 @@ void IStorageAccessor::Read(Kernel::HLERequestContext& ctx) {
 
     IPC::ResponseBuilder rb{ctx, 2};
     rb.Push(RESULT_SUCCESS);
-
-    LOG_DEBUG(Service_AM, "called, offset={}", offset);
 }
 
 ILibraryAppletCreator::ILibraryAppletCreator() : ServiceFramework("ILibraryAppletCreator") {
@@ -783,6 +792,8 @@ void ILibraryAppletCreator::CreateLibraryApplet(Kernel::HLERequestContext& ctx)
     const auto applet = GetAppletFromId(applet_id);
 
     if (applet == nullptr) {
+        LOG_ERROR(Service_AM, "Applet doesn't exist! applet_id={}", static_cast<u32>(applet_id));
+
         IPC::ResponseBuilder rb{ctx, 2};
         rb.Push(ResultCode(-1));
         return;
@@ -792,23 +803,23 @@ void ILibraryAppletCreator::CreateLibraryApplet(Kernel::HLERequestContext& ctx)
 
     rb.Push(RESULT_SUCCESS);
     rb.PushIpcInterface<AM::ILibraryAppletAccessor>(applet);
-
-    LOG_DEBUG(Service_AM, "called");
 }
 
 void ILibraryAppletCreator::CreateStorage(Kernel::HLERequestContext& ctx) {
     IPC::RequestParser rp{ctx};
     const u64 size{rp.Pop<u64>()};
+    LOG_DEBUG(Service_AM, "called, size={}", size);
+
     std::vector<u8> buffer(size);
 
     IPC::ResponseBuilder rb{ctx, 2, 0, 1};
     rb.Push(RESULT_SUCCESS);
     rb.PushIpcInterface<AM::IStorage>(std::move(buffer));
-
-    LOG_DEBUG(Service_AM, "called, size={}", size);
 }
 
 void ILibraryAppletCreator::CreateTransferMemoryStorage(Kernel::HLERequestContext& ctx) {
+    LOG_DEBUG(Service_AM, "called");
+
     IPC::RequestParser rp{ctx};
 
     rp.SetCurrentOffset(3);
@@ -819,6 +830,7 @@ void ILibraryAppletCreator::CreateTransferMemoryStorage(Kernel::HLERequestContex
             handle);
 
     if (shared_mem == nullptr) {
+        LOG_ERROR(Service_AM, "shared_mem is a nullpr for handle={:08X}", handle);
         IPC::ResponseBuilder rb{ctx, 2};
         rb.Push(ResultCode(-1));
         return;
@@ -882,38 +894,45 @@ IApplicationFunctions::IApplicationFunctions() : ServiceFramework("IApplicationF
 IApplicationFunctions::~IApplicationFunctions() = default;
 
 void IApplicationFunctions::EnableApplicationCrashReport(Kernel::HLERequestContext& ctx) {
+    LOG_WARNING(Service_AM, "(STUBBED) called");
+
     IPC::ResponseBuilder rb{ctx, 2};
     rb.Push(RESULT_SUCCESS);
-    LOG_WARNING(Service_AM, "(STUBBED) called");
 }
 
 void IApplicationFunctions::BeginBlockingHomeButtonShortAndLongPressed(
     Kernel::HLERequestContext& ctx) {
+    LOG_WARNING(Service_AM, "(STUBBED) called");
+
     IPC::ResponseBuilder rb{ctx, 2};
     rb.Push(RESULT_SUCCESS);
-    LOG_WARNING(Service_AM, "(STUBBED) called");
 }
 
 void IApplicationFunctions::EndBlockingHomeButtonShortAndLongPressed(
     Kernel::HLERequestContext& ctx) {
+    LOG_WARNING(Service_AM, "(STUBBED) called");
+
     IPC::ResponseBuilder rb{ctx, 2};
     rb.Push(RESULT_SUCCESS);
-    LOG_WARNING(Service_AM, "(STUBBED) called");
 }
 
 void IApplicationFunctions::BeginBlockingHomeButton(Kernel::HLERequestContext& ctx) {
+    LOG_WARNING(Service_AM, "(STUBBED) called");
+
     IPC::ResponseBuilder rb{ctx, 2};
     rb.Push(RESULT_SUCCESS);
-    LOG_WARNING(Service_AM, "(STUBBED) called");
 }
 
 void IApplicationFunctions::EndBlockingHomeButton(Kernel::HLERequestContext& ctx) {
+    LOG_WARNING(Service_AM, "(STUBBED) called");
+
     IPC::ResponseBuilder rb{ctx, 2};
     rb.Push(RESULT_SUCCESS);
-    LOG_WARNING(Service_AM, "(STUBBED) called");
 }
 
 void IApplicationFunctions::PopLaunchParameter(Kernel::HLERequestContext& ctx) {
+    LOG_DEBUG(Service_AM, "called");
+
     LaunchParameters params{};
 
     params.magic = POP_LAUNCH_PARAMETER_MAGIC;
@@ -932,21 +951,19 @@ void IApplicationFunctions::PopLaunchParameter(Kernel::HLERequestContext& ctx) {
     std::memcpy(buffer.data(), &params, buffer.size());
 
     rb.PushIpcInterface<AM::IStorage>(buffer);
-
-    LOG_DEBUG(Service_AM, "called");
 }
 
 void IApplicationFunctions::CreateApplicationAndRequestToStartForQuest(
     Kernel::HLERequestContext& ctx) {
+    LOG_WARNING(Service_AM, "(STUBBED) called");
+
     IPC::ResponseBuilder rb{ctx, 2};
     rb.Push(RESULT_SUCCESS);
-    LOG_WARNING(Service_AM, "(STUBBED) called");
 }
 
 void IApplicationFunctions::EnsureSaveData(Kernel::HLERequestContext& ctx) {
     IPC::RequestParser rp{ctx};
     u128 uid = rp.PopRaw<u128>(); // What does this do?
-
     LOG_WARNING(Service, "(STUBBED) called uid = {:016X}{:016X}", uid[1], uid[0]);
 
     IPC::ResponseBuilder rb{ctx, 4};
@@ -961,60 +978,62 @@ void IApplicationFunctions::SetTerminateResult(Kernel::HLERequestContext& ctx) {
 
     IPC::RequestParser rp{ctx};
     u32 result = rp.Pop<u32>();
+    LOG_WARNING(Service_AM, "(STUBBED) called, result=0x{:08X}", result);
 
     IPC::ResponseBuilder rb{ctx, 2};
     rb.Push(RESULT_SUCCESS);
-
-    LOG_WARNING(Service_AM, "(STUBBED) called, result=0x{:08X}", result);
 }
 
 void IApplicationFunctions::GetDisplayVersion(Kernel::HLERequestContext& ctx) {
+    LOG_WARNING(Service_AM, "(STUBBED) called");
+
     IPC::ResponseBuilder rb{ctx, 6};
     rb.Push(RESULT_SUCCESS);
     rb.Push<u64>(1);
     rb.Push<u64>(0);
-    LOG_WARNING(Service_AM, "(STUBBED) called");
 }
 
 void IApplicationFunctions::GetDesiredLanguage(Kernel::HLERequestContext& ctx) {
     // TODO(bunnei): This should be configurable
+    LOG_DEBUG(Service_AM, "called");
+
     IPC::ResponseBuilder rb{ctx, 4};
     rb.Push(RESULT_SUCCESS);
     rb.Push(
         static_cast<u64>(Service::Set::GetLanguageCodeFromIndex(Settings::values.language_index)));
-    LOG_DEBUG(Service_AM, "called");
 }
 
 void IApplicationFunctions::InitializeGamePlayRecording(Kernel::HLERequestContext& ctx) {
+    LOG_WARNING(Service_AM, "(STUBBED) called");
+
     IPC::ResponseBuilder rb{ctx, 2};
     rb.Push(RESULT_SUCCESS);
-    LOG_WARNING(Service_AM, "(STUBBED) called");
 }
 
 void IApplicationFunctions::SetGamePlayRecordingState(Kernel::HLERequestContext& ctx) {
+    LOG_WARNING(Service_AM, "(STUBBED) called");
+
     IPC::ResponseBuilder rb{ctx, 2};
     rb.Push(RESULT_SUCCESS);
-
-    LOG_WARNING(Service_AM, "(STUBBED) called");
 }
 
 void IApplicationFunctions::NotifyRunning(Kernel::HLERequestContext& ctx) {
+    LOG_WARNING(Service_AM, "(STUBBED) called");
+
     IPC::ResponseBuilder rb{ctx, 3};
     rb.Push(RESULT_SUCCESS);
     rb.Push<u8>(0); // Unknown, seems to be ignored by official processes
-
-    LOG_WARNING(Service_AM, "(STUBBED) called");
 }
 
 void IApplicationFunctions::GetPseudoDeviceId(Kernel::HLERequestContext& ctx) {
+    LOG_WARNING(Service_AM, "(STUBBED) called");
+
     IPC::ResponseBuilder rb{ctx, 6};
     rb.Push(RESULT_SUCCESS);
 
     // Returns a 128-bit UUID
     rb.Push<u64>(0);
     rb.Push<u64>(0);
-
-    LOG_WARNING(Service_AM, "(STUBBED) called");
 }
 
 void InstallInterfaces(SM::ServiceManager& service_manager,
@@ -1051,9 +1070,10 @@ IHomeMenuFunctions::IHomeMenuFunctions() : ServiceFramework("IHomeMenuFunctions"
 IHomeMenuFunctions::~IHomeMenuFunctions() = default;
 
 void IHomeMenuFunctions::RequestToGetForeground(Kernel::HLERequestContext& ctx) {
+    LOG_WARNING(Service_AM, "(STUBBED) called");
+
     IPC::ResponseBuilder rb{ctx, 2};
     rb.Push(RESULT_SUCCESS);
-    LOG_WARNING(Service_AM, "(STUBBED) called");
 }
 
 IGlobalStateController::IGlobalStateController() : ServiceFramework("IGlobalStateController") {

src/core/hle/service/am/applet_ae.cpp

@@ -32,66 +32,75 @@ public:
 
 private:
     void GetCommonStateGetter(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_AM, "called");
+
         IPC::ResponseBuilder rb{ctx, 2, 0, 1};
         rb.Push(RESULT_SUCCESS);
         rb.PushIpcInterface<ICommonStateGetter>(msg_queue);
-        LOG_DEBUG(Service_AM, "called");
     }
 
     void GetSelfController(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_AM, "called");
+
         IPC::ResponseBuilder rb{ctx, 2, 0, 1};
         rb.Push(RESULT_SUCCESS);
         rb.PushIpcInterface<ISelfController>(nvflinger);
-        LOG_DEBUG(Service_AM, "called");
     }
 
     void GetWindowController(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_AM, "called");
+
         IPC::ResponseBuilder rb{ctx, 2, 0, 1};
         rb.Push(RESULT_SUCCESS);
         rb.PushIpcInterface<IWindowController>();
-        LOG_DEBUG(Service_AM, "called");
     }
 
     void GetAudioController(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_AM, "called");
+
         IPC::ResponseBuilder rb{ctx, 2, 0, 1};
         rb.Push(RESULT_SUCCESS);
         rb.PushIpcInterface<IAudioController>();
-        LOG_DEBUG(Service_AM, "called");
     }
 
     void GetDisplayController(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_AM, "called");
+
         IPC::ResponseBuilder rb{ctx, 2, 0, 1};
         rb.Push(RESULT_SUCCESS);
         rb.PushIpcInterface<IDisplayController>();
-        LOG_DEBUG(Service_AM, "called");
     }
 
     void GetProcessWindingController(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_AM, "called");
+
         IPC::ResponseBuilder rb{ctx, 2, 0, 1};
         rb.Push(RESULT_SUCCESS);
         rb.PushIpcInterface<IProcessWindingController>();
-        LOG_DEBUG(Service_AM, "called");
     }
 
     void GetDebugFunctions(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_AM, "called");
+
         IPC::ResponseBuilder rb{ctx, 2, 0, 1};
         rb.Push(RESULT_SUCCESS);
         rb.PushIpcInterface<IDebugFunctions>();
-        LOG_DEBUG(Service_AM, "called");
     }
 
     void GetLibraryAppletCreator(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_AM, "called");
+
         IPC::ResponseBuilder rb{ctx, 2, 0, 1};
         rb.Push(RESULT_SUCCESS);
         rb.PushIpcInterface<ILibraryAppletCreator>();
-        LOG_DEBUG(Service_AM, "called");
     }
 
     void GetApplicationFunctions(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_AM, "called");
+
         IPC::ResponseBuilder rb{ctx, 2, 0, 1};
         rb.Push(RESULT_SUCCESS);
         rb.PushIpcInterface<IApplicationFunctions>();
-        LOG_DEBUG(Service_AM, "called");
     }
 
     std::shared_ptr<NVFlinger::NVFlinger> nvflinger;
@@ -122,97 +131,110 @@ public:
 
 private:
     void GetCommonStateGetter(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_AM, "called");
+
         IPC::ResponseBuilder rb{ctx, 2, 0, 1};
         rb.Push(RESULT_SUCCESS);
         rb.PushIpcInterface<ICommonStateGetter>(msg_queue);
-        LOG_DEBUG(Service_AM, "called");
     }
 
     void GetSelfController(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_AM, "called");
+
         IPC::ResponseBuilder rb{ctx, 2, 0, 1};
         rb.Push(RESULT_SUCCESS);
         rb.PushIpcInterface<ISelfController>(nvflinger);
-        LOG_DEBUG(Service_AM, "called");
     }
 
     void GetWindowController(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_AM, "called");
+
         IPC::ResponseBuilder rb{ctx, 2, 0, 1};
         rb.Push(RESULT_SUCCESS);
         rb.PushIpcInterface<IWindowController>();
-        LOG_DEBUG(Service_AM, "called");
     }
 
     void GetAudioController(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_AM, "called");
+
         IPC::ResponseBuilder rb{ctx, 2, 0, 1};
         rb.Push(RESULT_SUCCESS);
         rb.PushIpcInterface<IAudioController>();
-        LOG_DEBUG(Service_AM, "called");
     }
 
     void GetDisplayController(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_AM, "called");
+
         IPC::ResponseBuilder rb{ctx, 2, 0, 1};
         rb.Push(RESULT_SUCCESS);
         rb.PushIpcInterface<IDisplayController>();
-        LOG_DEBUG(Service_AM, "called");
     }
 
     void GetDebugFunctions(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_AM, "called");
+
         IPC::ResponseBuilder rb{ctx, 2, 0, 1};
         rb.Push(RESULT_SUCCESS);
         rb.PushIpcInterface<IDebugFunctions>();
-        LOG_DEBUG(Service_AM, "called");
     }
 
     void GetLibraryAppletCreator(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_AM, "called");
+
         IPC::ResponseBuilder rb{ctx, 2, 0, 1};
         rb.Push(RESULT_SUCCESS);
         rb.PushIpcInterface<ILibraryAppletCreator>();
-        LOG_DEBUG(Service_AM, "called");
     }
 
     void GetHomeMenuFunctions(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_AM, "called");
+
         IPC::ResponseBuilder rb{ctx, 2, 0, 1};
         rb.Push(RESULT_SUCCESS);
         rb.PushIpcInterface<IHomeMenuFunctions>();
-        LOG_DEBUG(Service_AM, "called");
     }
 
     void GetGlobalStateController(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_AM, "called");
+
         IPC::ResponseBuilder rb{ctx, 2, 0, 1};
         rb.Push(RESULT_SUCCESS);
         rb.PushIpcInterface<IGlobalStateController>();
-        LOG_DEBUG(Service_AM, "called");
     }
 
     void GetApplicationCreator(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_AM, "called");
+
         IPC::ResponseBuilder rb{ctx, 2, 0, 1};
         rb.Push(RESULT_SUCCESS);
         rb.PushIpcInterface<IApplicationCreator>();
-        LOG_DEBUG(Service_AM, "called");
     }
     std::shared_ptr<NVFlinger::NVFlinger> nvflinger;
     std::shared_ptr<AppletMessageQueue> msg_queue;
 };
 
 void AppletAE::OpenSystemAppletProxy(Kernel::HLERequestContext& ctx) {
+    LOG_DEBUG(Service_AM, "called");
+
     IPC::ResponseBuilder rb{ctx, 2, 0, 1};
     rb.Push(RESULT_SUCCESS);
     rb.PushIpcInterface<ISystemAppletProxy>(nvflinger, msg_queue);
-    LOG_DEBUG(Service_AM, "called");
 }
 
 void AppletAE::OpenLibraryAppletProxy(Kernel::HLERequestContext& ctx) {
+    LOG_DEBUG(Service_AM, "called");
+
     IPC::ResponseBuilder rb{ctx, 2, 0, 1};
     rb.Push(RESULT_SUCCESS);
     rb.PushIpcInterface<ILibraryAppletProxy>(nvflinger, msg_queue);
-    LOG_DEBUG(Service_AM, "called");
 }
 
 void AppletAE::OpenLibraryAppletProxyOld(Kernel::HLERequestContext& ctx) {
+    LOG_DEBUG(Service_AM, "called");
+
     IPC::ResponseBuilder rb{ctx, 2, 0, 1};
     rb.Push(RESULT_SUCCESS);
     rb.PushIpcInterface<ILibraryAppletProxy>(nvflinger, msg_queue);
-    LOG_DEBUG(Service_AM, "called");
 }
 
 AppletAE::AppletAE(std::shared_ptr<NVFlinger::NVFlinger> nvflinger,

src/core/hle/service/am/applet_oe.cpp

@@ -35,59 +35,67 @@ public:
 
 private:
     void GetAudioController(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_AM, "called");
+
         IPC::ResponseBuilder rb{ctx, 2, 0, 1};
         rb.Push(RESULT_SUCCESS);
         rb.PushIpcInterface<IAudioController>();
-        LOG_DEBUG(Service_AM, "called");
     }
 
     void GetDisplayController(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_AM, "called");
+
         IPC::ResponseBuilder rb{ctx, 2, 0, 1};
         rb.Push(RESULT_SUCCESS);
         rb.PushIpcInterface<IDisplayController>();
-        LOG_DEBUG(Service_AM, "called");
     }
 
     void GetDebugFunctions(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_AM, "called");
+
         IPC::ResponseBuilder rb{ctx, 2, 0, 1};
         rb.Push(RESULT_SUCCESS);
         rb.PushIpcInterface<IDebugFunctions>();
-        LOG_DEBUG(Service_AM, "called");
     }
 
     void GetWindowController(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_AM, "called");
+
         IPC::ResponseBuilder rb{ctx, 2, 0, 1};
         rb.Push(RESULT_SUCCESS);
         rb.PushIpcInterface<IWindowController>();
-        LOG_DEBUG(Service_AM, "called");
     }
 
     void GetSelfController(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_AM, "called");
+
         IPC::ResponseBuilder rb{ctx, 2, 0, 1};
         rb.Push(RESULT_SUCCESS);
         rb.PushIpcInterface<ISelfController>(nvflinger);
-        LOG_DEBUG(Service_AM, "called");
     }
 
     void GetCommonStateGetter(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_AM, "called");
+
         IPC::ResponseBuilder rb{ctx, 2, 0, 1};
         rb.Push(RESULT_SUCCESS);
         rb.PushIpcInterface<ICommonStateGetter>(msg_queue);
-        LOG_DEBUG(Service_AM, "called");
     }
 
     void GetLibraryAppletCreator(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_AM, "called");
+
         IPC::ResponseBuilder rb{ctx, 2, 0, 1};
         rb.Push(RESULT_SUCCESS);
         rb.PushIpcInterface<ILibraryAppletCreator>();
-        LOG_DEBUG(Service_AM, "called");
     }
 
     void GetApplicationFunctions(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_AM, "called");
+
         IPC::ResponseBuilder rb{ctx, 2, 0, 1};
         rb.Push(RESULT_SUCCESS);
         rb.PushIpcInterface<IApplicationFunctions>();
-        LOG_DEBUG(Service_AM, "called");
     }
 
     std::shared_ptr<NVFlinger::NVFlinger> nvflinger;
@@ -95,10 +103,11 @@ private:
 };
 
 void AppletOE::OpenApplicationProxy(Kernel::HLERequestContext& ctx) {
+    LOG_DEBUG(Service_AM, "called");
+
     IPC::ResponseBuilder rb{ctx, 2, 0, 1};
     rb.Push(RESULT_SUCCESS);
     rb.PushIpcInterface<IApplicationProxy>(nvflinger, msg_queue);
-    LOG_DEBUG(Service_AM, "called");
 }
 
 AppletOE::AppletOE(std::shared_ptr<NVFlinger::NVFlinger> nvflinger,

src/core/hle/service/aoc/aoc_u.cpp

@@ -68,6 +68,8 @@ AOC_U::AOC_U() : ServiceFramework("aoc:u"), add_on_content(AccumulateAOCTitleIDs
 AOC_U::~AOC_U() = default;
 
 void AOC_U::CountAddOnContent(Kernel::HLERequestContext& ctx) {
+    LOG_DEBUG(Service_AOC, "called");
+
     IPC::ResponseBuilder rb{ctx, 3};
     rb.Push(RESULT_SUCCESS);
 
@@ -82,6 +84,7 @@ void AOC_U::ListAddOnContent(Kernel::HLERequestContext& ctx) {
 
     const auto offset = rp.PopRaw<u32>();
     auto count = rp.PopRaw<u32>();
+    LOG_DEBUG(Service_AOC, "called with offset={}, count={}", offset, count);
 
     const auto current = Core::System::GetInstance().CurrentProcess()->GetTitleID();
 
@@ -110,6 +113,8 @@ void AOC_U::ListAddOnContent(Kernel::HLERequestContext& ctx) {
 }
 
 void AOC_U::GetAddOnContentBaseId(Kernel::HLERequestContext& ctx) {
+    LOG_DEBUG(Service_AOC, "called");
+
     IPC::ResponseBuilder rb{ctx, 4};
     rb.Push(RESULT_SUCCESS);
     const auto title_id = Core::System::GetInstance().CurrentProcess()->GetTitleID();
@@ -128,7 +133,6 @@ void AOC_U::PrepareAddOnContent(Kernel::HLERequestContext& ctx) {
     IPC::RequestParser rp{ctx};
 
     const auto aoc_id = rp.PopRaw<u32>();
-
     LOG_WARNING(Service_AOC, "(STUBBED) called with aoc_id={:08X}", aoc_id);
 
     IPC::ResponseBuilder rb{ctx, 2};

src/core/hle/service/apm/interface.cpp

@@ -40,24 +40,22 @@ private:
 
         auto mode = static_cast<PerformanceMode>(rp.Pop<u32>());
         u32 config = rp.Pop<u32>();
+        LOG_WARNING(Service_APM, "(STUBBED) called mode={} config={}", static_cast<u32>(mode),
+                    config);
 
         IPC::ResponseBuilder rb{ctx, 2};
         rb.Push(RESULT_SUCCESS);
-
-        LOG_WARNING(Service_APM, "(STUBBED) called mode={} config={}", static_cast<u32>(mode),
-                    config);
     }
 
     void GetPerformanceConfiguration(Kernel::HLERequestContext& ctx) {
         IPC::RequestParser rp{ctx};
 
         auto mode = static_cast<PerformanceMode>(rp.Pop<u32>());
+        LOG_WARNING(Service_APM, "(STUBBED) called mode={}", static_cast<u32>(mode));
 
         IPC::ResponseBuilder rb{ctx, 3};
         rb.Push(RESULT_SUCCESS);
         rb.Push<u32>(static_cast<u32>(PerformanceConfiguration::Config1));
-
-        LOG_WARNING(Service_APM, "(STUBBED) called mode={}", static_cast<u32>(mode));
     }
 };
 
@@ -73,11 +71,11 @@ APM::APM(std::shared_ptr<Module> apm, const char* name)
 APM::~APM() = default;
 
 void APM::OpenSession(Kernel::HLERequestContext& ctx) {
+    LOG_DEBUG(Service_APM, "called");
+
     IPC::ResponseBuilder rb{ctx, 2, 0, 1};
     rb.Push(RESULT_SUCCESS);
     rb.PushIpcInterface<ISession>();
-
-    LOG_DEBUG(Service_APM, "called");
 }
 
 APM_Sys::APM_Sys() : ServiceFramework{"apm:sys"} {
@@ -98,11 +96,11 @@ APM_Sys::APM_Sys() : ServiceFramework{"apm:sys"} {
 APM_Sys::~APM_Sys() = default;
 
 void APM_Sys::GetPerformanceEvent(Kernel::HLERequestContext& ctx) {
+    LOG_DEBUG(Service_APM, "called");
+
     IPC::ResponseBuilder rb{ctx, 2, 0, 1};
     rb.Push(RESULT_SUCCESS);
     rb.PushIpcInterface<ISession>();
-
-    LOG_DEBUG(Service_APM, "called");
 }
 
 } // namespace Service::APM

src/core/hle/service/arp/arp.cpp

@@ -59,11 +59,11 @@ public:
 
 private:
     void AcquireRegistrar(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_ARP, "called");
+
         IPC::ResponseBuilder rb{ctx, 2, 0, 1};
         rb.Push(RESULT_SUCCESS);
         rb.PushIpcInterface<IRegistrar>();
-
-        LOG_DEBUG(Service_ARP, "called");
     }
 };
 

src/core/hle/service/audio/audout_u.cpp

@@ -86,6 +86,7 @@ private:
 
     void GetAudioOutState(Kernel::HLERequestContext& ctx) {
         LOG_DEBUG(Service_Audio, "called");
+
         IPC::ResponseBuilder rb{ctx, 3};
         rb.Push(RESULT_SUCCESS);
         rb.Push(static_cast<u32>(stream->IsPlaying() ? AudioState::Started : AudioState::Stopped));
@@ -148,6 +149,7 @@ private:
 
     void GetReleasedAudioOutBufferImpl(Kernel::HLERequestContext& ctx) {
         LOG_DEBUG(Service_Audio, "called {}", ctx.Description());
+
         IPC::RequestParser rp{ctx};
         const u64 max_count{ctx.GetWriteBufferSize() / sizeof(u64)};
         const auto released_buffers{audio_core.GetTagsAndReleaseBuffers(stream, max_count)};
@@ -163,6 +165,7 @@ private:
 
     void ContainsAudioOutBuffer(Kernel::HLERequestContext& ctx) {
         LOG_DEBUG(Service_Audio, "called");
+
         IPC::RequestParser rp{ctx};
         const u64 tag{rp.Pop<u64>()};
         IPC::ResponseBuilder rb{ctx, 3};
@@ -172,6 +175,7 @@ private:
 
     void GetAudioOutBufferCount(Kernel::HLERequestContext& ctx) {
         LOG_DEBUG(Service_Audio, "called");
+
         IPC::ResponseBuilder rb{ctx, 3};
         rb.Push(RESULT_SUCCESS);
         rb.Push(static_cast<u32>(stream->GetQueueSize()));
@@ -189,6 +193,7 @@ private:
 
 void AudOutU::ListAudioOutsImpl(Kernel::HLERequestContext& ctx) {
     LOG_DEBUG(Service_Audio, "called");
+
     IPC::RequestParser rp{ctx};
 
     ctx.WriteBuffer(DefaultDevice);

src/core/hle/service/audio/audren_u.cpp

@@ -52,74 +52,79 @@ private:
     }
 
     void GetSampleRate(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_Audio, "called");
+
         IPC::ResponseBuilder rb{ctx, 3};
         rb.Push(RESULT_SUCCESS);
         rb.Push<u32>(renderer->GetSampleRate());
-        LOG_DEBUG(Service_Audio, "called");
     }
 
     void GetSampleCount(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_Audio, "called");
+
         IPC::ResponseBuilder rb{ctx, 3};
         rb.Push(RESULT_SUCCESS);
         rb.Push<u32>(renderer->GetSampleCount());
-        LOG_DEBUG(Service_Audio, "called");
     }
 
     void GetState(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_Audio, "called");
+
         IPC::ResponseBuilder rb{ctx, 3};
         rb.Push(RESULT_SUCCESS);
         rb.Push<u32>(static_cast<u32>(renderer->GetStreamState()));
-        LOG_DEBUG(Service_Audio, "called");
     }
 
     void GetMixBufferCount(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_Audio, "called");
+
         IPC::ResponseBuilder rb{ctx, 3};
         rb.Push(RESULT_SUCCESS);
         rb.Push<u32>(renderer->GetMixBufferCount());
-        LOG_DEBUG(Service_Audio, "called");
     }
 
     void RequestUpdateImpl(Kernel::HLERequestContext& ctx) {
+        LOG_WARNING(Service_Audio, "(STUBBED) called");
+
         ctx.WriteBuffer(renderer->UpdateAudioRenderer(ctx.ReadBuffer()));
         IPC::ResponseBuilder rb{ctx, 2};
         rb.Push(RESULT_SUCCESS);
-        LOG_WARNING(Service_Audio, "(STUBBED) called");
     }
 
     void Start(Kernel::HLERequestContext& ctx) {
+        LOG_WARNING(Service_Audio, "(STUBBED) called");
+
         IPC::ResponseBuilder rb{ctx, 2};
 
         rb.Push(RESULT_SUCCESS);
-
-        LOG_WARNING(Service_Audio, "(STUBBED) called");
     }
 
     void Stop(Kernel::HLERequestContext& ctx) {
+        LOG_WARNING(Service_Audio, "(STUBBED) called");
+
         IPC::ResponseBuilder rb{ctx, 2};
 
         rb.Push(RESULT_SUCCESS);
-
-        LOG_WARNING(Service_Audio, "(STUBBED) called");
     }
 
     void QuerySystemEvent(Kernel::HLERequestContext& ctx) {
+        LOG_WARNING(Service_Audio, "(STUBBED) called");
+
         IPC::ResponseBuilder rb{ctx, 2, 1};
         rb.Push(RESULT_SUCCESS);
         rb.PushCopyObjects(system_event);
-
-        LOG_WARNING(Service_Audio, "(STUBBED) called");
     }
 
     void SetRenderingTimeLimit(Kernel::HLERequestContext& ctx) {
         IPC::RequestParser rp{ctx};
         rendering_time_limit_percent = rp.Pop<u32>();
+        LOG_DEBUG(Service_Audio, "called. rendering_time_limit_percent={}",
+                  rendering_time_limit_percent);
+
         ASSERT(rendering_time_limit_percent >= 0 && rendering_time_limit_percent <= 100);
 
         IPC::ResponseBuilder rb{ctx, 2};
         rb.Push(RESULT_SUCCESS);
-
-        LOG_DEBUG(Service_Audio, "called. rendering_time_limit_percent={}",
-                  rendering_time_limit_percent);
     }
 
     void GetRenderingTimeLimit(Kernel::HLERequestContext& ctx) {
@@ -211,6 +216,7 @@ private:
 
     void GetActiveChannelCount(Kernel::HLERequestContext& ctx) {
         LOG_WARNING(Service_Audio, "(STUBBED) called");
+
         IPC::ResponseBuilder rb{ctx, 3};
         rb.Push(RESULT_SUCCESS);
         rb.Push<u32>(1);
@@ -235,19 +241,20 @@ AudRenU::AudRenU() : ServiceFramework("audren:u") {
 AudRenU::~AudRenU() = default;
 
 void AudRenU::OpenAudioRenderer(Kernel::HLERequestContext& ctx) {
+    LOG_DEBUG(Service_Audio, "called");
+
     IPC::RequestParser rp{ctx};
     auto params = rp.PopRaw<AudioCore::AudioRendererParameter>();
     IPC::ResponseBuilder rb{ctx, 2, 0, 1};
 
     rb.Push(RESULT_SUCCESS);
     rb.PushIpcInterface<Audio::IAudioRenderer>(std::move(params));
-
-    LOG_DEBUG(Service_Audio, "called");
 }
 
 void AudRenU::GetAudioRendererWorkBufferSize(Kernel::HLERequestContext& ctx) {
     IPC::RequestParser rp{ctx};
     auto params = rp.PopRaw<AudioCore::AudioRendererParameter>();
+    LOG_DEBUG(Service_Audio, "called");
 
     u64 buffer_sz = Common::AlignUp(4 * params.mix_buffer_count, 0x40);
     buffer_sz += params.unknown_c * 1024;
@@ -301,26 +308,26 @@ void AudRenU::GetAudioRendererWorkBufferSize(Kernel::HLERequestContext& ctx) {
     rb.Push(RESULT_SUCCESS);
     rb.Push<u64>(output_sz);
 
-    LOG_DEBUG(Service_Audio, "called, buffer_size=0x{:X}", output_sz);
+    LOG_DEBUG(Service_Audio, "buffer_size=0x{:X}", output_sz);
 }
 
 void AudRenU::GetAudioDevice(Kernel::HLERequestContext& ctx) {
+    LOG_DEBUG(Service_Audio, "called");
+
     IPC::ResponseBuilder rb{ctx, 2, 0, 1};
 
     rb.Push(RESULT_SUCCESS);
     rb.PushIpcInterface<Audio::IAudioDevice>();
-
-    LOG_DEBUG(Service_Audio, "called");
 }
 
 void AudRenU::GetAudioDeviceServiceWithRevisionInfo(Kernel::HLERequestContext& ctx) {
+    LOG_WARNING(Service_Audio, "(STUBBED) called");
+
     IPC::ResponseBuilder rb{ctx, 2, 0, 1};
 
     rb.Push(RESULT_SUCCESS);
-    rb.PushIpcInterface<Audio::IAudioDevice>();
-
-    LOG_WARNING(Service_Audio, "(STUBBED) called"); // TODO(ogniK): Figure out what is different
-                                                    // based on the current revision
+    rb.PushIpcInterface<Audio::IAudioDevice>(); // TODO(ogniK): Figure out what is different
+                                                // based on the current revision
 }
 
 bool AudRenU::IsFeatureSupported(AudioFeatures feature, u32_le revision) const {

src/core/hle/service/audio/hwopus.cpp

@@ -46,10 +46,13 @@ public:
 
 private:
     void DecodeInterleaved(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Audio, "called");
+
         u32 consumed = 0;
         u32 sample_count = 0;
         std::vector<opus_int16> samples(ctx.GetWriteBufferSize() / sizeof(opus_int16));
         if (!Decoder_DecodeInterleaved(consumed, sample_count, ctx.ReadBuffer(), samples)) {
+            LOG_ERROR(Audio, "Failed to decode opus data");
             IPC::ResponseBuilder rb{ctx, 2};
             // TODO(ogniK): Use correct error code
             rb.Push(ResultCode(-1));
@@ -63,12 +66,15 @@ private:
     }
 
     void DecodeInterleavedWithPerformance(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Audio, "called");
+
         u32 consumed = 0;
         u32 sample_count = 0;
         u64 performance = 0;
         std::vector<opus_int16> samples(ctx.GetWriteBufferSize() / sizeof(opus_int16));
         if (!Decoder_DecodeInterleaved(consumed, sample_count, ctx.ReadBuffer(), samples,
                                        performance)) {
+            LOG_ERROR(Audio, "Failed to decode opus data");
             IPC::ResponseBuilder rb{ctx, 2};
             // TODO(ogniK): Use correct error code
             rb.Push(ResultCode(-1));
@@ -88,24 +94,39 @@ private:
         std::optional<std::reference_wrapper<u64>> performance_time = std::nullopt) {
         const auto start_time = std::chrono::high_resolution_clock::now();
         std::size_t raw_output_sz = output.size() * sizeof(opus_int16);
-        if (sizeof(OpusHeader) > input.size())
+        if (sizeof(OpusHeader) > input.size()) {
+            LOG_ERROR(Audio, "Input is smaller than the header size, header_sz={}, input_sz={}",
+                      sizeof(OpusHeader), input.size());
             return false;
+        }
         OpusHeader hdr{};
         std::memcpy(&hdr, input.data(), sizeof(OpusHeader));
         if (sizeof(OpusHeader) + static_cast<u32>(hdr.sz) > input.size()) {
+            LOG_ERROR(Audio, "Input does not fit in the opus header size. data_sz={}, input_sz={}",
+                      sizeof(OpusHeader) + static_cast<u32>(hdr.sz), input.size());
             return false;
         }
         auto frame = input.data() + sizeof(OpusHeader);
         auto decoded_sample_count = opus_packet_get_nb_samples(
             frame, static_cast<opus_int32>(input.size() - sizeof(OpusHeader)),
             static_cast<opus_int32>(sample_rate));
-        if (decoded_sample_count * channel_count * sizeof(u16) > raw_output_sz)
+        if (decoded_sample_count * channel_count * sizeof(u16) > raw_output_sz) {
+            LOG_ERROR(
+                Audio,
+                "Decoded data does not fit into the output data, decoded_sz={}, raw_output_sz={}",
+                decoded_sample_count * channel_count * sizeof(u16), raw_output_sz);
             return false;
+        }
+        const int frame_size = (static_cast<int>(raw_output_sz / sizeof(s16) / channel_count));
         auto out_sample_count =
-            opus_decode(decoder.get(), frame, hdr.sz, output.data(),
-                        (static_cast<int>(raw_output_sz / sizeof(s16) / channel_count)), 0);
-        if (out_sample_count < 0)
+            opus_decode(decoder.get(), frame, hdr.sz, output.data(), frame_size, 0);
+        if (out_sample_count < 0) {
+            LOG_ERROR(Audio,
+                      "Incorrect sample count received from opus_decode, "
+                      "output_sample_count={}, frame_size={}, data_sz_from_hdr={}",
+                      out_sample_count, frame_size, static_cast<u32>(hdr.sz));
             return false;
+        }
         const auto end_time = std::chrono::high_resolution_clock::now() - start_time;
         sample_count = out_sample_count;
         consumed = static_cast<u32>(sizeof(OpusHeader) + hdr.sz);
@@ -134,14 +155,17 @@ static std::size_t WorkerBufferSize(u32 channel_count) {
 
 void HwOpus::GetWorkBufferSize(Kernel::HLERequestContext& ctx) {
     IPC::RequestParser rp{ctx};
-    auto sample_rate = rp.Pop<u32>();
-    auto channel_count = rp.Pop<u32>();
+    const auto sample_rate = rp.Pop<u32>();
+    const auto channel_count = rp.Pop<u32>();
+    LOG_DEBUG(Audio, "called with sample_rate={}, channel_count={}", sample_rate, channel_count);
+
     ASSERT_MSG(sample_rate == 48000 || sample_rate == 24000 || sample_rate == 16000 ||
                    sample_rate == 12000 || sample_rate == 8000,
                "Invalid sample rate");
     ASSERT_MSG(channel_count == 1 || channel_count == 2, "Invalid channel count");
-    u32 worker_buffer_sz = static_cast<u32>(WorkerBufferSize(channel_count));
-    LOG_DEBUG(Audio, "called worker_buffer_sz={}", worker_buffer_sz);
+
+    const u32 worker_buffer_sz = static_cast<u32>(WorkerBufferSize(channel_count));
+    LOG_DEBUG(Audio, "worker_buffer_sz={}", worker_buffer_sz);
 
     IPC::ResponseBuilder rb{ctx, 3};
     rb.Push(RESULT_SUCCESS);
@@ -155,6 +179,7 @@ void HwOpus::OpenOpusDecoder(Kernel::HLERequestContext& ctx) {
     auto buffer_sz = rp.Pop<u32>();
     LOG_DEBUG(Audio, "called sample_rate={}, channel_count={}, buffer_size={}", sample_rate,
               channel_count, buffer_sz);
+
     ASSERT_MSG(sample_rate == 48000 || sample_rate == 24000 || sample_rate == 16000 ||
                    sample_rate == 12000 || sample_rate == 8000,
                "Invalid sample rate");
@@ -164,7 +189,8 @@ void HwOpus::OpenOpusDecoder(Kernel::HLERequestContext& ctx) {
     ASSERT_MSG(buffer_sz >= worker_sz, "Worker buffer too large");
     std::unique_ptr<OpusDecoder, OpusDeleter> decoder{
         static_cast<OpusDecoder*>(operator new(worker_sz))};
-    if (opus_decoder_init(decoder.get(), sample_rate, channel_count)) {
+    if (const int err = opus_decoder_init(decoder.get(), sample_rate, channel_count)) {
+        LOG_ERROR(Audio, "Failed to init opus decoder with error={}", err);
         IPC::ResponseBuilder rb{ctx, 2};
         // TODO(ogniK): Use correct error code
         rb.Push(ResultCode(-1));

src/core/hle/service/bcat/module.cpp

@@ -33,10 +33,11 @@ public:
 };
 
 void Module::Interface::CreateBcatService(Kernel::HLERequestContext& ctx) {
+    LOG_DEBUG(Service_BCAT, "called");
+
     IPC::ResponseBuilder rb{ctx, 2, 0, 1};
     rb.Push(RESULT_SUCCESS);
     rb.PushIpcInterface<IBcatService>();
-    LOG_DEBUG(Service_BCAT, "called");
 }
 
 Module::Interface::Interface(std::shared_ptr<Module> module, const char* name)

src/core/hle/service/btdrv/btdrv.cpp

@@ -34,13 +34,14 @@ public:
 
 private:
     void RegisterEvent(Kernel::HLERequestContext& ctx) {
+        LOG_WARNING(Service_BTM, "(STUBBED) called");
+
         auto& kernel = Core::System::GetInstance().Kernel();
         register_event =
             Kernel::Event::Create(kernel, Kernel::ResetType::OneShot, "BT:RegisterEvent");
         IPC::ResponseBuilder rb{ctx, 2, 1};
         rb.Push(RESULT_SUCCESS);
         rb.PushCopyObjects(register_event);
-        LOG_WARNING(Service_BTM, "(STUBBED) called");
     }
     Kernel::SharedPtr<Kernel::Event> register_event;
 };

src/core/hle/service/btm/btm.cpp

@@ -57,40 +57,44 @@ public:
 
 private:
     void GetScanEvent(Kernel::HLERequestContext& ctx) {
+        LOG_WARNING(Service_BTM, "(STUBBED) called");
+
         auto& kernel = Core::System::GetInstance().Kernel();
         scan_event =
             Kernel::Event::Create(kernel, Kernel::ResetType::OneShot, "IBtmUserCore:ScanEvent");
         IPC::ResponseBuilder rb{ctx, 2, 1};
         rb.Push(RESULT_SUCCESS);
         rb.PushCopyObjects(scan_event);
-        LOG_WARNING(Service_BTM, "(STUBBED) called");
     }
     void GetConnectionEvent(Kernel::HLERequestContext& ctx) {
+        LOG_WARNING(Service_BTM, "(STUBBED) called");
+
         auto& kernel = Core::System::GetInstance().Kernel();
         connection_event = Kernel::Event::Create(kernel, Kernel::ResetType::OneShot,
                                                  "IBtmUserCore:ConnectionEvent");
         IPC::ResponseBuilder rb{ctx, 2, 1};
         rb.Push(RESULT_SUCCESS);
         rb.PushCopyObjects(connection_event);
-        LOG_WARNING(Service_BTM, "(STUBBED) called");
     }
     void GetDiscoveryEvent(Kernel::HLERequestContext& ctx) {
+        LOG_WARNING(Service_BTM, "(STUBBED) called");
+
         auto& kernel = Core::System::GetInstance().Kernel();
         service_discovery =
             Kernel::Event::Create(kernel, Kernel::ResetType::OneShot, "IBtmUserCore:Discovery");
         IPC::ResponseBuilder rb{ctx, 2, 1};
         rb.Push(RESULT_SUCCESS);
         rb.PushCopyObjects(service_discovery);
-        LOG_WARNING(Service_BTM, "(STUBBED) called");
     }
     void GetConfigEvent(Kernel::HLERequestContext& ctx) {
+        LOG_WARNING(Service_BTM, "(STUBBED) called");
+
         auto& kernel = Core::System::GetInstance().Kernel();
         config_event =
             Kernel::Event::Create(kernel, Kernel::ResetType::OneShot, "IBtmUserCore:ConfigEvent");
         IPC::ResponseBuilder rb{ctx, 2, 1};
         rb.Push(RESULT_SUCCESS);
         rb.PushCopyObjects(config_event);
-        LOG_WARNING(Service_BTM, "(STUBBED) called");
     }
     Kernel::SharedPtr<Kernel::Event> scan_event;
     Kernel::SharedPtr<Kernel::Event> connection_event;
@@ -111,10 +115,11 @@ public:
 
 private:
     void GetCoreImpl(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_BTM, "called");
+
         IPC::ResponseBuilder rb{ctx, 2, 0, 1};
         rb.Push(RESULT_SUCCESS);
         rb.PushIpcInterface<IBtmUserCore>();
-        LOG_DEBUG(Service_BTM, "called");
     }
 };
 
@@ -209,11 +214,11 @@ public:
 
 private:
     void GetCoreImpl(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_BTM, "called");
+
         IPC::ResponseBuilder rb{ctx, 2, 0, 1};
         rb.Push(RESULT_SUCCESS);
         rb.PushIpcInterface<IBtmSystemCore>();
-
-        LOG_DEBUG(Service_BTM, "called");
     }
 };
 

src/core/hle/service/fgm/fgm.cpp

@@ -42,11 +42,11 @@ public:
 
 private:
     void Initialize(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_FGM, "called");
+
         IPC::ResponseBuilder rb{ctx, 2, 0, 1};
         rb.Push(RESULT_SUCCESS);
         rb.PushIpcInterface<IRequest>();
-
-        LOG_DEBUG(Service_FGM, "called");
     }
 };
 

src/core/hle/service/filesystem/filesystem.cpp

@@ -113,6 +113,18 @@ ResultCode VfsDirectoryServiceWrapper::DeleteDirectoryRecursively(const std::str
     return RESULT_SUCCESS;
 }
 
+ResultCode VfsDirectoryServiceWrapper::CleanDirectoryRecursively(const std::string& path) const {
+    const std::string sanitized_path(FileUtil::SanitizePath(path));
+    auto dir = GetDirectoryRelativeWrapped(backing, FileUtil::GetParentPath(sanitized_path));
+
+    if (!dir->CleanSubdirectoryRecursive(FileUtil::GetFilename(sanitized_path))) {
+        // TODO(DarkLordZach): Find a better error code for this
+        return ResultCode(-1);
+    }
+
+    return RESULT_SUCCESS;
+}
+
 ResultCode VfsDirectoryServiceWrapper::RenameFile(const std::string& src_path_,
                                                   const std::string& dest_path_) const {
     std::string src_path(FileUtil::SanitizePath(src_path_));

src/core/hle/service/filesystem/filesystem.h

@@ -113,6 +113,18 @@ public:
      */
     ResultCode DeleteDirectoryRecursively(const std::string& path) const;
 
+    /**
+     * Cleans the specified directory. This is similar to DeleteDirectoryRecursively,
+     * in that it deletes all the contents of the specified directory, however, this
+     * function does *not* delete the directory itself. It only deletes everything
+     * within it.
+     *
+     * @param path Path relative to the archive.
+     *
+     * @return Result of the operation.
+     */
+    ResultCode CleanDirectoryRecursively(const std::string& path) const;
+
     /**
      * Rename a File specified by its path
      * @param src_path Source path relative to the archive

src/core/hle/service/filesystem/fsp_srv.cpp

@@ -62,11 +62,13 @@ private:
 
         // Error checking
         if (length < 0) {
+            LOG_ERROR(Service_FS, "Length is less than 0, length={}", length);
             IPC::ResponseBuilder rb{ctx, 2};
             rb.Push(FileSys::ERROR_INVALID_SIZE);
             return;
         }
         if (offset < 0) {
+            LOG_ERROR(Service_FS, "Offset is less than 0, offset={}", offset);
             IPC::ResponseBuilder rb{ctx, 2};
             rb.Push(FileSys::ERROR_INVALID_OFFSET);
             return;
@@ -107,11 +109,13 @@ private:
 
         // Error checking
         if (length < 0) {
+            LOG_ERROR(Service_FS, "Length is less than 0, length={}", length);
             IPC::ResponseBuilder rb{ctx, 2};
             rb.Push(FileSys::ERROR_INVALID_SIZE);
             return;
         }
         if (offset < 0) {
+            LOG_ERROR(Service_FS, "Offset is less than 0, offset={}", offset);
             IPC::ResponseBuilder rb{ctx, 2};
             rb.Push(FileSys::ERROR_INVALID_OFFSET);
             return;
@@ -138,11 +142,13 @@ private:
 
         // Error checking
         if (length < 0) {
+            LOG_ERROR(Service_FS, "Length is less than 0, length={}", length);
             IPC::ResponseBuilder rb{ctx, 2};
             rb.Push(FileSys::ERROR_INVALID_SIZE);
             return;
         }
         if (offset < 0) {
+            LOG_ERROR(Service_FS, "Offset is less than 0, offset={}", offset);
             IPC::ResponseBuilder rb{ctx, 2};
             rb.Push(FileSys::ERROR_INVALID_OFFSET);
             return;
@@ -180,9 +186,10 @@ private:
     void SetSize(Kernel::HLERequestContext& ctx) {
         IPC::RequestParser rp{ctx};
         const u64 size = rp.Pop<u64>();
-        backend->Resize(size);
         LOG_DEBUG(Service_FS, "called, size={}", size);
 
+        backend->Resize(size);
+
         IPC::ResponseBuilder rb{ctx, 2};
         rb.Push(RESULT_SUCCESS);
     }
@@ -284,7 +291,7 @@ public:
             {10, &IFileSystem::Commit, "Commit"},
             {11, nullptr, "GetFreeSpaceSize"},
             {12, nullptr, "GetTotalSpaceSize"},
-            {13, nullptr, "CleanDirectoryRecursively"},
+            {13, &IFileSystem::CleanDirectoryRecursively, "CleanDirectoryRecursively"},
             {14, nullptr, "GetFileTimeStampRaw"},
             {15, nullptr, "QueryEntry"},
         };
@@ -354,6 +361,16 @@ public:
         rb.Push(backend.DeleteDirectoryRecursively(name));
     }
 
+    void CleanDirectoryRecursively(Kernel::HLERequestContext& ctx) {
+        const auto file_buffer = ctx.ReadBuffer();
+        const std::string name = Common::StringFromBuffer(file_buffer);
+
+        LOG_DEBUG(Service_FS, "called. Directory: {}", name);
+
+        IPC::ResponseBuilder rb{ctx, 2};
+        rb.Push(backend.CleanDirectoryRecursively(name));
+    }
+
     void RenameFile(Kernel::HLERequestContext& ctx) {
         IPC::RequestParser rp{ctx};
 
@@ -465,6 +482,8 @@ public:
     }
 
     void ReadSaveDataInfo(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_FS, "called");
+
         // Calculate how many entries we can fit in the output buffer
         const u64 count_entries = ctx.GetWriteBufferSize() / sizeof(SaveDataInfo);
 
@@ -703,6 +722,8 @@ void FSP_SRV::OpenFileSystemWithPatch(Kernel::HLERequestContext& ctx) {
 
     const auto type = rp.PopRaw<FileSystemType>();
     const auto title_id = rp.PopRaw<u64>();
+    LOG_WARNING(Service_FS, "(STUBBED) called with type={}, title_id={:016X}",
+                static_cast<u8>(type), title_id);
 
     IPC::ResponseBuilder rb{ctx, 2, 0, 0};
     rb.Push(ResultCode(-1));
@@ -738,6 +759,7 @@ void FSP_SRV::MountSaveData(Kernel::HLERequestContext& ctx) {
     auto space_id = rp.PopRaw<FileSys::SaveDataSpaceId>();
     auto unk = rp.Pop<u32>();
     LOG_INFO(Service_FS, "called with unknown={:08X}", unk);
+
     auto save_struct = rp.PopRaw<FileSys::SaveDataDescriptor>();
 
     auto dir = OpenSaveData(space_id, save_struct);
@@ -763,6 +785,7 @@ void FSP_SRV::OpenReadOnlySaveDataFileSystem(Kernel::HLERequestContext& ctx) {
 void FSP_SRV::OpenSaveDataInfoReaderBySaveDataSpaceId(Kernel::HLERequestContext& ctx) {
     IPC::RequestParser rp{ctx};
     const auto space = rp.PopRaw<FileSys::SaveDataSpaceId>();
+    LOG_INFO(Service_FS, "called, space={}", static_cast<u8>(space));
 
     IPC::ResponseBuilder rb{ctx, 2, 0, 1};
     rb.Push(RESULT_SUCCESS);

src/core/hle/service/hid/controllers/npad.cpp

@@ -524,6 +524,8 @@ void Controller_NPad::SetNpadMode(u32 npad_id, NPadAssignments assignment_mode)
 
 void Controller_NPad::VibrateController(const std::vector<u32>& controller_ids,
                                         const std::vector<Vibration>& vibrations) {
+    LOG_WARNING(Service_HID, "(STUBBED) called");
+
     if (!can_controllers_vibrate) {
         return;
     }
@@ -533,7 +535,6 @@ void Controller_NPad::VibrateController(const std::vector<u32>& controller_ids,
             // TODO(ogniK): Vibrate the physical controller
         }
     }
-    LOG_WARNING(Service_HID, "(STUBBED) called");
     last_processed_vibration = vibrations.back();
 }
 
@@ -575,8 +576,8 @@ void Controller_NPad::AddNewControllerAt(NPadControllerType controller, u32 npad
         return;
     }
 
-    connected_controllers[npad_id] = {controller, true};
-    InitNewlyAddedControler(npad_id);
+    connected_controllers[NPadIdToIndex(npad_id)] = {controller, true};
+    InitNewlyAddedControler(NPadIdToIndex(npad_id));
 }
 
 void Controller_NPad::ConnectNPad(u32 npad_id) {

src/core/hle/service/hid/controllers/npad.h

@@ -80,9 +80,9 @@ public:
     struct LedPattern {
         explicit LedPattern(u64 light1, u64 light2, u64 light3, u64 light4) {
             position1.Assign(light1);
-            position1.Assign(light2);
-            position1.Assign(light3);
-            position1.Assign(light4);
+            position2.Assign(light2);
+            position3.Assign(light3);
+            position4.Assign(light4);
         }
         union {
             u64 raw{};

src/core/hle/service/hid/hid.cpp

@@ -124,10 +124,11 @@ public:
 
 private:
     void GetSharedMemoryHandle(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_HID, "called");
+
         IPC::ResponseBuilder rb{ctx, 2, 1};
         rb.Push(RESULT_SUCCESS);
         rb.PushCopyObjects(shared_mem);
-        LOG_DEBUG(Service_HID, "called");
     }
 
     void UpdateControllers(u64 userdata, int cycles_late) {
@@ -163,9 +164,10 @@ public:
 
 private:
     void ActivateVibrationDevice(Kernel::HLERequestContext& ctx) {
+        LOG_WARNING(Service_HID, "(STUBBED) called");
+
         IPC::ResponseBuilder rb{ctx, 2};
         rb.Push(RESULT_SUCCESS);
-        LOG_WARNING(Service_HID, "(STUBBED) called");
     }
 };
 
@@ -303,6 +305,8 @@ private:
     std::shared_ptr<IAppletResource> applet_resource;
 
     void CreateAppletResource(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_HID, "called");
+
         if (applet_resource == nullptr) {
             applet_resource = std::make_shared<IAppletResource>();
         }
@@ -310,206 +314,228 @@ private:
         IPC::ResponseBuilder rb{ctx, 2, 0, 1};
         rb.Push(RESULT_SUCCESS);
         rb.PushIpcInterface<IAppletResource>(applet_resource);
-        LOG_DEBUG(Service_HID, "called");
     }
 
     void ActivateXpad(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_HID, "called");
+
         applet_resource->ActivateController(HidController::XPad);
         IPC::ResponseBuilder rb{ctx, 2};
         rb.Push(RESULT_SUCCESS);
-        LOG_DEBUG(Service_HID, "called");
     }
 
     void ActivateDebugPad(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_HID, "called");
+
         applet_resource->ActivateController(HidController::DebugPad);
         IPC::ResponseBuilder rb{ctx, 2};
         rb.Push(RESULT_SUCCESS);
-        LOG_DEBUG(Service_HID, "called");
     }
 
     void ActivateTouchScreen(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_HID, "called");
+
         applet_resource->ActivateController(HidController::Touchscreen);
         IPC::ResponseBuilder rb{ctx, 2};
         rb.Push(RESULT_SUCCESS);
-        LOG_DEBUG(Service_HID, "called");
     }
 
     void ActivateMouse(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_HID, "called");
+
         applet_resource->ActivateController(HidController::Mouse);
         IPC::ResponseBuilder rb{ctx, 2};
         rb.Push(RESULT_SUCCESS);
-        LOG_DEBUG(Service_HID, "called");
     }
 
     void ActivateKeyboard(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_HID, "called");
+
         applet_resource->ActivateController(HidController::Keyboard);
         IPC::ResponseBuilder rb{ctx, 2};
         rb.Push(RESULT_SUCCESS);
-        LOG_DEBUG(Service_HID, "called");
     }
 
     void ActivateGesture(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_HID, "called");
+
         applet_resource->ActivateController(HidController::Gesture);
         IPC::ResponseBuilder rb{ctx, 2};
         rb.Push(RESULT_SUCCESS);
-        LOG_DEBUG(Service_HID, "called");
     }
 
     void ActivateNpadWithRevision(Kernel::HLERequestContext& ctx) {
         // Should have no effect with how our npad sets up the data
+        LOG_DEBUG(Service_HID, "called");
+
         applet_resource->ActivateController(HidController::NPad);
         IPC::ResponseBuilder rb{ctx, 2};
         rb.Push(RESULT_SUCCESS);
-        LOG_DEBUG(Service_HID, "called");
     }
 
     void StartSixAxisSensor(Kernel::HLERequestContext& ctx) {
         IPC::RequestParser rp{ctx};
         auto handle = rp.PopRaw<u32>();
+        LOG_WARNING(Service_HID, "(STUBBED) called with handle={}", handle);
+
         IPC::ResponseBuilder rb{ctx, 2};
         rb.Push(RESULT_SUCCESS);
-        LOG_WARNING(Service_HID, "(STUBBED) called");
     }
 
     void SetGyroscopeZeroDriftMode(Kernel::HLERequestContext& ctx) {
+        LOG_WARNING(Service_HID, "(STUBBED) called");
+
         IPC::ResponseBuilder rb{ctx, 2};
         rb.Push(RESULT_SUCCESS);
-        LOG_WARNING(Service_HID, "(STUBBED) called");
     }
 
     void IsSixAxisSensorAtRest(Kernel::HLERequestContext& ctx) {
+        LOG_WARNING(Service_HID, "(STUBBED) called");
+
         IPC::ResponseBuilder rb{ctx, 3};
         rb.Push(RESULT_SUCCESS);
         // TODO (Hexagon12): Properly implement reading gyroscope values from controllers.
         rb.Push(true);
-        LOG_WARNING(Service_HID, "(STUBBED) called");
     }
 
     void SetSupportedNpadStyleSet(Kernel::HLERequestContext& ctx) {
         IPC::RequestParser rp{ctx};
         auto supported_styleset = rp.PopRaw<u32>();
+        LOG_DEBUG(Service_HID, "called with supported_styleset={}", supported_styleset);
+
         applet_resource->GetController<Controller_NPad>(HidController::NPad)
             .SetSupportedStyleSet({supported_styleset});
 
         IPC::ResponseBuilder rb{ctx, 2};
         rb.Push(RESULT_SUCCESS);
-
-        LOG_DEBUG(Service_HID, "called");
     }
 
     void GetSupportedNpadStyleSet(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_HID, "called");
+
         auto& controller = applet_resource->GetController<Controller_NPad>(HidController::NPad);
 
         IPC::ResponseBuilder rb{ctx, 3};
         rb.Push(RESULT_SUCCESS);
         rb.Push<u32>(controller.GetSupportedStyleSet().raw);
-        LOG_DEBUG(Service_HID, "called");
     }
 
     void SetSupportedNpadIdType(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_HID, "called");
+
         applet_resource->GetController<Controller_NPad>(HidController::NPad)
             .SetSupportedNPadIdTypes(ctx.ReadBuffer().data(), ctx.GetReadBufferSize());
         IPC::ResponseBuilder rb{ctx, 2};
         rb.Push(RESULT_SUCCESS);
-        LOG_DEBUG(Service_HID, "called");
     }
 
     void ActivateNpad(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_HID, "called");
+
         IPC::ResponseBuilder rb{ctx, 2};
         rb.Push(RESULT_SUCCESS);
         applet_resource->ActivateController(HidController::NPad);
-        LOG_DEBUG(Service_HID, "called");
     }
 
     void AcquireNpadStyleSetUpdateEventHandle(Kernel::HLERequestContext& ctx) {
         IPC::RequestParser rp{ctx};
         auto npad_id = rp.PopRaw<u32>();
+        LOG_DEBUG(Service_HID, "called with npad_id={}", npad_id);
+
         IPC::ResponseBuilder rb{ctx, 2, 1};
         rb.Push(RESULT_SUCCESS);
         rb.PushCopyObjects(applet_resource->GetController<Controller_NPad>(HidController::NPad)
                                .GetStyleSetChangedEvent());
-        LOG_DEBUG(Service_HID, "called");
     }
 
     void DisconnectNpad(Kernel::HLERequestContext& ctx) {
         IPC::RequestParser rp{ctx};
         auto npad_id = rp.PopRaw<u32>();
+        LOG_DEBUG(Service_HID, "called with npad_id={}", npad_id);
+
         applet_resource->GetController<Controller_NPad>(HidController::NPad)
             .DisconnectNPad(npad_id);
         IPC::ResponseBuilder rb{ctx, 2};
         rb.Push(RESULT_SUCCESS);
-        LOG_DEBUG(Service_HID, "called");
     }
 
     void GetPlayerLedPattern(Kernel::HLERequestContext& ctx) {
         IPC::RequestParser rp{ctx};
         auto npad_id = rp.PopRaw<u32>();
+        LOG_DEBUG(Service_HID, "called with npad_id={}", npad_id);
+
         IPC::ResponseBuilder rb{ctx, 4};
         rb.Push(RESULT_SUCCESS);
         rb.PushRaw<u64>(applet_resource->GetController<Controller_NPad>(HidController::NPad)
                             .GetLedPattern(npad_id)
                             .raw);
-        LOG_DEBUG(Service_HID, "called");
     }
 
     void SetNpadJoyHoldType(Kernel::HLERequestContext& ctx) {
-        auto& controller = applet_resource->GetController<Controller_NPad>(HidController::NPad);
         IPC::RequestParser rp{ctx};
         const auto hold_type = rp.PopRaw<u64>();
+        LOG_DEBUG(Service_HID, "called with hold_type={}", hold_type);
+
+        auto& controller = applet_resource->GetController<Controller_NPad>(HidController::NPad);
         controller.SetHoldType(Controller_NPad::NpadHoldType{hold_type});
 
         IPC::ResponseBuilder rb{ctx, 2};
         rb.Push(RESULT_SUCCESS);
-        LOG_DEBUG(Service_HID, "called");
     }
 
     void GetNpadJoyHoldType(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_HID, "called");
+
         const auto& controller =
             applet_resource->GetController<Controller_NPad>(HidController::NPad);
         IPC::ResponseBuilder rb{ctx, 4};
         rb.Push(RESULT_SUCCESS);
         rb.Push<u64>(static_cast<u64>(controller.GetHoldType()));
-        LOG_DEBUG(Service_HID, "called");
     }
 
     void SetNpadJoyAssignmentModeSingleByDefault(Kernel::HLERequestContext& ctx) {
         IPC::RequestParser rp{ctx};
         auto npad_id = rp.PopRaw<u32>();
+        LOG_WARNING(Service_HID, "(STUBBED) called with npad_id={}", npad_id);
+
         IPC::ResponseBuilder rb{ctx, 2};
         rb.Push(RESULT_SUCCESS);
-        LOG_WARNING(Service_HID, "(STUBBED) called");
     }
 
     void BeginPermitVibrationSession(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_HID, "called");
+
         applet_resource->GetController<Controller_NPad>(HidController::NPad)
             .SetVibrationEnabled(true);
         IPC::ResponseBuilder rb{ctx, 2};
         rb.Push(RESULT_SUCCESS);
-        LOG_DEBUG(Service_HID, "called");
     }
 
     void EndPermitVibrationSession(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_HID, "called");
+
         applet_resource->GetController<Controller_NPad>(HidController::NPad)
             .SetVibrationEnabled(false);
         IPC::ResponseBuilder rb{ctx, 2};
         rb.Push(RESULT_SUCCESS);
-        LOG_DEBUG(Service_HID, "called");
     }
 
     void SendVibrationValue(Kernel::HLERequestContext& ctx) {
         IPC::RequestParser rp{ctx};
         const auto controller_id = rp.PopRaw<u32>();
         const auto vibration_values = rp.PopRaw<Controller_NPad::Vibration>();
+        LOG_DEBUG(Service_HID, "called with controller_id={}", controller_id);
 
         IPC::ResponseBuilder rb{ctx, 2};
         rb.Push(RESULT_SUCCESS);
 
         applet_resource->GetController<Controller_NPad>(HidController::NPad)
             .VibrateController({controller_id}, {vibration_values});
-        LOG_DEBUG(Service_HID, "called");
     }
 
     void SendVibrationValues(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_HID, "called");
+
         const auto controllers = ctx.ReadBuffer(0);
         const auto vibrations = ctx.ReadBuffer(1);
 
@@ -527,86 +553,96 @@ private:
 
         IPC::ResponseBuilder rb{ctx, 2};
         rb.Push(RESULT_SUCCESS);
-        LOG_DEBUG(Service_HID, "called");
     }
 
     void GetActualVibrationValue(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_HID, "called");
+
         IPC::ResponseBuilder rb{ctx, 6};
         rb.Push(RESULT_SUCCESS);
         rb.PushRaw<Controller_NPad::Vibration>(
             applet_resource->GetController<Controller_NPad>(HidController::NPad)
                 .GetLastVibration());
-        LOG_DEBUG(Service_HID, "called");
     }
 
     void SetNpadJoyAssignmentModeDual(Kernel::HLERequestContext& ctx) {
         IPC::RequestParser rp{ctx};
         const auto npad_id = rp.PopRaw<u32>();
+        LOG_DEBUG(Service_HID, "called with npad_id={}", npad_id);
+
         auto& controller = applet_resource->GetController<Controller_NPad>(HidController::NPad);
         controller.SetNpadMode(npad_id, Controller_NPad::NPadAssignments::Dual);
 
         IPC::ResponseBuilder rb{ctx, 2};
         rb.Push(RESULT_SUCCESS);
-        LOG_DEBUG(Service_HID, "called");
     }
 
     void MergeSingleJoyAsDualJoy(Kernel::HLERequestContext& ctx) {
+        LOG_WARNING(Service_HID, "(STUBBED) called");
+
         IPC::ResponseBuilder rb{ctx, 2};
         rb.Push(RESULT_SUCCESS);
-        LOG_WARNING(Service_HID, "(STUBBED) called");
     }
 
     void SetNpadHandheldActivationMode(Kernel::HLERequestContext& ctx) {
         IPC::RequestParser rp{ctx};
         auto mode = rp.PopRaw<u32>();
+        LOG_WARNING(Service_HID, "(STUBBED) called with mode={}", mode);
+
         IPC::ResponseBuilder rb{ctx, 2};
         rb.Push(RESULT_SUCCESS);
-        LOG_WARNING(Service_HID, "(STUBBED) called");
     }
 
     void GetVibrationDeviceInfo(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_HID, "called");
+
         IPC::ResponseBuilder rb{ctx, 4};
         rb.Push(RESULT_SUCCESS);
         rb.Push<u32>(1);
         rb.Push<u32>(0);
-        LOG_DEBUG(Service_HID, "called");
     }
 
     void CreateActiveVibrationDeviceList(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_HID, "called");
+
         IPC::ResponseBuilder rb{ctx, 2, 0, 1};
         rb.Push(RESULT_SUCCESS);
         rb.PushIpcInterface<IActiveVibrationDeviceList>();
-        LOG_DEBUG(Service_HID, "called");
     }
 
     void ActivateConsoleSixAxisSensor(Kernel::HLERequestContext& ctx) {
+        LOG_WARNING(Service_HID, "(STUBBED) called");
+
         IPC::ResponseBuilder rb{ctx, 2};
         rb.Push(RESULT_SUCCESS);
-        LOG_WARNING(Service_HID, "(STUBBED) called");
     }
 
     void StartConsoleSixAxisSensor(Kernel::HLERequestContext& ctx) {
+        LOG_WARNING(Service_HID, "(STUBBED) called");
+
         IPC::ResponseBuilder rb{ctx, 2};
         rb.Push(RESULT_SUCCESS);
-        LOG_WARNING(Service_HID, "(STUBBED) called");
     }
 
     void StopSixAxisSensor(Kernel::HLERequestContext& ctx) {
+        LOG_WARNING(Service_HID, "(STUBBED) called");
+
         IPC::ResponseBuilder rb{ctx, 2};
         rb.Push(RESULT_SUCCESS);
-        LOG_WARNING(Service_HID, "(STUBBED) called");
     }
 
     void SetIsPalmaAllConnectable(Kernel::HLERequestContext& ctx) {
+        LOG_WARNING(Service_HID, "(STUBBED) called");
+
         IPC::ResponseBuilder rb{ctx, 2};
         rb.Push(RESULT_SUCCESS);
-        LOG_WARNING(Service_HID, "(STUBBED) called");
     }
 
     void SetPalmaBoostMode(Kernel::HLERequestContext& ctx) {
+        LOG_WARNING(Service_HID, "(STUBBED) called");
+
         IPC::ResponseBuilder rb{ctx, 2};
         rb.Push(RESULT_SUCCESS);
-        LOG_WARNING(Service_HID, "(STUBBED) called");
     }
 };
 

src/core/hle/service/hid/irs.cpp

@@ -44,115 +44,133 @@ IRS::IRS() : ServiceFramework{"irs"} {
 }
 
 void IRS::ActivateIrsensor(Kernel::HLERequestContext& ctx) {
+    LOG_WARNING(Service_IRS, "(STUBBED) called");
+
     IPC::ResponseBuilder rb{ctx, 2};
     rb.Push(RESULT_SUCCESS);
-    LOG_WARNING(Service_IRS, "(STUBBED) called");
 }
 
 void IRS::DeactivateIrsensor(Kernel::HLERequestContext& ctx) {
+    LOG_WARNING(Service_IRS, "(STUBBED) called");
+
     IPC::ResponseBuilder rb{ctx, 2};
     rb.Push(RESULT_SUCCESS);
-    LOG_WARNING(Service_IRS, "(STUBBED) called");
 }
 
 void IRS::GetIrsensorSharedMemoryHandle(Kernel::HLERequestContext& ctx) {
+    LOG_DEBUG(Service_IRS, "called");
+
     IPC::ResponseBuilder rb{ctx, 2, 1};
     rb.Push(RESULT_SUCCESS);
     rb.PushCopyObjects(shared_mem);
-    LOG_DEBUG(Service_IRS, "called");
 }
 
 void IRS::StopImageProcessor(Kernel::HLERequestContext& ctx) {
+    LOG_WARNING(Service_IRS, "(STUBBED) called");
+
     IPC::ResponseBuilder rb{ctx, 2};
     rb.Push(RESULT_SUCCESS);
-    LOG_WARNING(Service_IRS, "(STUBBED) called");
 }
 
 void IRS::RunMomentProcessor(Kernel::HLERequestContext& ctx) {
+    LOG_WARNING(Service_IRS, "(STUBBED) called");
+
     IPC::ResponseBuilder rb{ctx, 2};
     rb.Push(RESULT_SUCCESS);
-    LOG_WARNING(Service_IRS, "(STUBBED) called");
 }
 
 void IRS::RunClusteringProcessor(Kernel::HLERequestContext& ctx) {
+    LOG_WARNING(Service_IRS, "(STUBBED) called");
+
     IPC::ResponseBuilder rb{ctx, 2};
     rb.Push(RESULT_SUCCESS);
-    LOG_WARNING(Service_IRS, "(STUBBED) called");
 }
 
 void IRS::RunImageTransferProcessor(Kernel::HLERequestContext& ctx) {
+    LOG_WARNING(Service_IRS, "(STUBBED) called");
+
     IPC::ResponseBuilder rb{ctx, 2};
     rb.Push(RESULT_SUCCESS);
-    LOG_WARNING(Service_IRS, "(STUBBED) called");
 }
 
 void IRS::GetImageTransferProcessorState(Kernel::HLERequestContext& ctx) {
+    LOG_WARNING(Service_IRS, "(STUBBED) called");
+
     IPC::ResponseBuilder rb{ctx, 5};
     rb.Push(RESULT_SUCCESS);
     rb.PushRaw<u64>(CoreTiming::GetTicks());
     rb.PushRaw<u32>(0);
-    LOG_WARNING(Service_IRS, "(STUBBED) called");
 }
 
 void IRS::RunTeraPluginProcessor(Kernel::HLERequestContext& ctx) {
+    LOG_WARNING(Service_IRS, "(STUBBED) called");
+
     IPC::ResponseBuilder rb{ctx, 2};
     rb.Push(RESULT_SUCCESS);
-    LOG_WARNING(Service_IRS, "(STUBBED) called");
 }
 
 void IRS::GetNpadIrCameraHandle(Kernel::HLERequestContext& ctx) {
+    LOG_WARNING(Service_IRS, "(STUBBED) called");
+
     IPC::ResponseBuilder rb{ctx, 3};
     rb.Push(RESULT_SUCCESS);
     rb.PushRaw<u32>(device_handle);
-    LOG_WARNING(Service_IRS, "(STUBBED) called");
 }
 
 void IRS::RunPointingProcessor(Kernel::HLERequestContext& ctx) {
+    LOG_WARNING(Service_IRS, "(STUBBED) called");
+
     IPC::ResponseBuilder rb{ctx, 2};
     rb.Push(RESULT_SUCCESS);
-    LOG_WARNING(Service_IRS, "(STUBBED) called");
 }
 
 void IRS::SuspendImageProcessor(Kernel::HLERequestContext& ctx) {
+    LOG_WARNING(Service_IRS, "(STUBBED) called");
+
     IPC::ResponseBuilder rb{ctx, 2};
     rb.Push(RESULT_SUCCESS);
-    LOG_WARNING(Service_IRS, "(STUBBED) called");
 }
 
 void IRS::CheckFirmwareVersion(Kernel::HLERequestContext& ctx) {
+    LOG_WARNING(Service_IRS, "(STUBBED) called");
+
     IPC::ResponseBuilder rb{ctx, 2};
     rb.Push(RESULT_SUCCESS);
-    LOG_WARNING(Service_IRS, "(STUBBED) called");
 }
 
 void IRS::SetFunctionLevel(Kernel::HLERequestContext& ctx) {
+    LOG_WARNING(Service_IRS, "(STUBBED) called");
+
     IPC::ResponseBuilder rb{ctx, 2};
     rb.Push(RESULT_SUCCESS);
-    LOG_WARNING(Service_IRS, "(STUBBED) called");
 }
 
 void IRS::RunImageTransferExProcessor(Kernel::HLERequestContext& ctx) {
+    LOG_WARNING(Service_IRS, "(STUBBED) called");
+
     IPC::ResponseBuilder rb{ctx, 2};
     rb.Push(RESULT_SUCCESS);
-    LOG_WARNING(Service_IRS, "(STUBBED) called");
 }
 
 void IRS::RunIrLedProcessor(Kernel::HLERequestContext& ctx) {
+    LOG_WARNING(Service_IRS, "(STUBBED) called");
+
     IPC::ResponseBuilder rb{ctx, 2};
     rb.Push(RESULT_SUCCESS);
-    LOG_WARNING(Service_IRS, "(STUBBED) called");
 }
 
 void IRS::StopImageProcessorAsync(Kernel::HLERequestContext& ctx) {
+    LOG_WARNING(Service_IRS, "(STUBBED) called");
+
     IPC::ResponseBuilder rb{ctx, 2};
     rb.Push(RESULT_SUCCESS);
-    LOG_WARNING(Service_IRS, "(STUBBED) called");
 }
 
 void IRS::ActivateIrsensorWithFunctionLevel(Kernel::HLERequestContext& ctx) {
+    LOG_WARNING(Service_IRS, "(STUBBED) called");
+
     IPC::ResponseBuilder rb{ctx, 2};
     rb.Push(RESULT_SUCCESS);
-    LOG_WARNING(Service_IRS, "(STUBBED) called");
 }
 
 IRS::~IRS() = default;

src/core/hle/service/lbl/lbl.cpp

@@ -55,29 +55,29 @@ public:
 
 private:
     void EnableVrMode(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_LBL, "called");
+
         IPC::ResponseBuilder rb{ctx, 2};
         rb.Push(RESULT_SUCCESS);
 
         vr_mode_enabled = true;
-
-        LOG_DEBUG(Service_LBL, "called");
     }
 
     void DisableVrMode(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_LBL, "called");
+
         IPC::ResponseBuilder rb{ctx, 2};
         rb.Push(RESULT_SUCCESS);
 
         vr_mode_enabled = false;
-
-        LOG_DEBUG(Service_LBL, "called");
     }
 
     void IsVrModeEnabled(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_LBL, "called");
+
         IPC::ResponseBuilder rb{ctx, 3};
         rb.Push(RESULT_SUCCESS);
         rb.Push(vr_mode_enabled);
-
-        LOG_DEBUG(Service_LBL, "called");
     }
 
     bool vr_mode_enabled = false;

src/core/hle/service/ldn/ldn.cpp

@@ -44,11 +44,11 @@ public:
     }
 
     void CreateMonitorService(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_LDN, "called");
+
         IPC::ResponseBuilder rb{ctx, 2, 0, 1};
         rb.Push(RESULT_SUCCESS);
         rb.PushIpcInterface<IMonitorService>();
-
-        LOG_DEBUG(Service_LDN, "called");
     }
 };
 
@@ -104,11 +104,11 @@ public:
     }
 
     void CreateSystemLocalCommunicationService(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_LDN, "called");
+
         IPC::ResponseBuilder rb{ctx, 2, 0, 1};
         rb.Push(RESULT_SUCCESS);
         rb.PushIpcInterface<ILocalCommunicationService>("ISystemLocalCommunicationService");
-
-        LOG_DEBUG(Service_LDN, "called");
     }
 };
 
@@ -125,11 +125,11 @@ public:
     }
 
     void CreateUserLocalCommunicationService(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_LDN, "called");
+
         IPC::ResponseBuilder rb{ctx, 2, 0, 1};
         rb.Push(RESULT_SUCCESS);
         rb.PushIpcInterface<ILocalCommunicationService>("IUserLocalCommunicationService");
-
-        LOG_DEBUG(Service_LDN, "called");
     }
 };
 

src/core/hle/service/ldr/ldr.cpp

@@ -97,6 +97,8 @@ public:
         rp.Skip(2, false);
         const VAddr nrr_addr{rp.Pop<VAddr>()};
         const u64 nrr_size{rp.Pop<u64>()};
+        LOG_DEBUG(Service_LDR, "called with nrr_addr={:016X}, nrr_size={:016X}", nrr_addr,
+                  nrr_size);
 
         if (!initialized) {
             LOG_ERROR(Service_LDR, "LDR:RO not initialized before use!");
@@ -189,6 +191,7 @@ public:
         IPC::RequestParser rp{ctx};
         rp.Skip(2, false);
         const auto nrr_addr{rp.Pop<VAddr>()};
+        LOG_DEBUG(Service_LDR, "called with nrr_addr={:016X}", nrr_addr);
 
         if (!Common::Is4KBAligned(nrr_addr)) {
             LOG_ERROR(Service_LDR, "NRR Address has invalid alignment (actual {:016X})!", nrr_addr);
@@ -219,6 +222,10 @@ public:
         const u64 nro_size{rp.Pop<u64>()};
         const VAddr bss_addr{rp.Pop<VAddr>()};
         const u64 bss_size{rp.Pop<u64>()};
+        LOG_DEBUG(
+            Service_LDR,
+            "called with nro_addr={:016X}, nro_size={:016X}, bss_addr={:016X}, bss_size={:016X}",
+            nro_addr, nro_size, bss_addr, bss_size);
 
         if (!initialized) {
             LOG_ERROR(Service_LDR, "LDR:RO not initialized before use!");
@@ -345,6 +352,8 @@ public:
         rp.Skip(2, false);
         const VAddr mapped_addr{rp.PopRaw<VAddr>()};
         const VAddr heap_addr{rp.PopRaw<VAddr>()};
+        LOG_DEBUG(Service_LDR, "called with mapped_addr={:016X}, heap_addr={:016X}", mapped_addr,
+                  heap_addr);
 
         if (!initialized) {
             LOG_ERROR(Service_LDR, "LDR:RO not initialized before use!");
@@ -393,11 +402,12 @@ public:
     }
 
     void Initialize(Kernel::HLERequestContext& ctx) {
+        LOG_WARNING(Service_LDR, "(STUBBED) called");
+
         initialized = true;
 
         IPC::ResponseBuilder rb{ctx, 2};
         rb.Push(RESULT_SUCCESS);
-        LOG_WARNING(Service_LDR, "(STUBBED) called");
     }
 
 private:

src/core/hle/service/lm/lm.cpp

@@ -209,11 +209,11 @@ public:
      *      0: ResultCode
      */
     void OpenLogger(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_LM, "called");
+
         IPC::ResponseBuilder rb{ctx, 2, 0, 1};
         rb.Push(RESULT_SUCCESS);
         rb.PushIpcInterface<ILogger>();
-
-        LOG_DEBUG(Service_LM, "called");
     }
 };
 

src/core/hle/service/mm/mm_u.cpp

@@ -31,12 +31,14 @@ public:
 private:
     void Initialize(Kernel::HLERequestContext& ctx) {
         LOG_WARNING(Service_MM, "(STUBBED) called");
+
         IPC::ResponseBuilder rb{ctx, 2};
         rb.Push(RESULT_SUCCESS);
     }
 
     void Finalize(Kernel::HLERequestContext& ctx) {
         LOG_WARNING(Service_MM, "(STUBBED) called");
+
         IPC::ResponseBuilder rb{ctx, 2};
         rb.Push(RESULT_SUCCESS);
     }
@@ -45,15 +47,16 @@ private:
         IPC::RequestParser rp{ctx};
         min = rp.Pop<u32>();
         max = rp.Pop<u32>();
-        current = min;
-
         LOG_WARNING(Service_MM, "(STUBBED) called, min=0x{:X}, max=0x{:X}", min, max);
+
+        current = min;
         IPC::ResponseBuilder rb{ctx, 2};
         rb.Push(RESULT_SUCCESS);
     }
 
     void Get(Kernel::HLERequestContext& ctx) {
         LOG_WARNING(Service_MM, "(STUBBED) called");
+
         IPC::ResponseBuilder rb{ctx, 3};
         rb.Push(RESULT_SUCCESS);
         rb.Push(current);
@@ -61,6 +64,7 @@ private:
 
     void InitializeWithId(Kernel::HLERequestContext& ctx) {
         LOG_WARNING(Service_MM, "(STUBBED) called");
+
         IPC::ResponseBuilder rb{ctx, 3};
         rb.Push(RESULT_SUCCESS);
         rb.Push<u32>(id); // Any non zero value
@@ -68,6 +72,7 @@ private:
 
     void FinalizeWithId(Kernel::HLERequestContext& ctx) {
         LOG_WARNING(Service_MM, "(STUBBED) called");
+
         IPC::ResponseBuilder rb{ctx, 2};
         rb.Push(RESULT_SUCCESS);
     }
@@ -77,16 +82,17 @@ private:
         u32 input_id = rp.Pop<u32>();
         min = rp.Pop<u32>();
         max = rp.Pop<u32>();
-        current = min;
-
         LOG_WARNING(Service_MM, "(STUBBED) called, input_id=0x{:X}, min=0x{:X}, max=0x{:X}",
                     input_id, min, max);
+
+        current = min;
         IPC::ResponseBuilder rb{ctx, 2};
         rb.Push(RESULT_SUCCESS);
     }
 
     void GetWithId(Kernel::HLERequestContext& ctx) {
         LOG_WARNING(Service_MM, "(STUBBED) called");
+
         IPC::ResponseBuilder rb{ctx, 3};
         rb.Push(RESULT_SUCCESS);
         rb.Push(current);

src/core/hle/service/nfc/nfc.cpp

@@ -43,11 +43,11 @@ public:
 
 private:
     void CreateAmInterface(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_NFC, "called");
+
         IPC::ResponseBuilder rb{ctx, 2, 0, 1};
         rb.Push(RESULT_SUCCESS);
         rb.PushIpcInterface<IAm>();
-
-        LOG_DEBUG(Service_NFC, "called");
     }
 };
 
@@ -91,11 +91,11 @@ public:
 
 private:
     void CreateUserInterface(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_NFC, "called");
+
         IPC::ResponseBuilder rb{ctx, 2, 0, 1};
         rb.Push(RESULT_SUCCESS);
         rb.PushIpcInterface<MFIUser>();
-
-        LOG_DEBUG(Service_NFC, "called");
     }
 };
 
@@ -138,19 +138,19 @@ private:
     };
 
     void InitializeOld(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_NFC, "called");
+
         IPC::ResponseBuilder rb{ctx, 2, 0};
         rb.Push(RESULT_SUCCESS);
-
         // We don't deal with hardware initialization so we can just stub this.
-        LOG_DEBUG(Service_NFC, "called");
     }
 
     void IsNfcEnabledOld(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_NFC, "IsNfcEnabledOld");
+
         IPC::ResponseBuilder rb{ctx, 3};
         rb.Push(RESULT_SUCCESS);
         rb.PushRaw<u8>(Settings::values.enable_nfc);
-
-        LOG_DEBUG(Service_NFC, "IsNfcEnabledOld");
     }
 
     void GetStateOld(Kernel::HLERequestContext& ctx) {
@@ -183,11 +183,11 @@ public:
 
 private:
     void CreateUserInterface(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_NFC, "called");
+
         IPC::ResponseBuilder rb{ctx, 2, 0, 1};
         rb.Push(RESULT_SUCCESS);
         rb.PushIpcInterface<IUser>();
-
-        LOG_DEBUG(Service_NFC, "called");
     }
 };
 
@@ -241,11 +241,11 @@ public:
 
 private:
     void CreateSystemInterface(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_NFC, "called");
+
         IPC::ResponseBuilder rb{ctx, 2, 0, 1};
         rb.Push(RESULT_SUCCESS);
         rb.PushIpcInterface<ISystem>();
-
-        LOG_DEBUG(Service_NFC, "called");
     }
 };
 

src/core/hle/service/nfp/nfp.cpp

@@ -108,30 +108,29 @@ private:
     static_assert(sizeof(CommonInfo) == 0x40, "CommonInfo is an invalid size");
 
     void Initialize(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_NFC, "called");
+
         IPC::ResponseBuilder rb{ctx, 2, 0};
         rb.Push(RESULT_SUCCESS);
 
         state = State::Initialized;
-
-        LOG_DEBUG(Service_NFC, "called");
     }
 
     void GetState(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_NFC, "called");
+
         IPC::ResponseBuilder rb{ctx, 3, 0};
         rb.Push(RESULT_SUCCESS);
         rb.PushRaw<u32>(static_cast<u32>(state));
-
-        LOG_DEBUG(Service_NFC, "called");
     }
 
     void ListDevices(Kernel::HLERequestContext& ctx) {
         IPC::RequestParser rp{ctx};
         const u32 array_size = rp.Pop<u32>();
+        LOG_DEBUG(Service_NFP, "called, array_size={}", array_size);
 
         ctx.WriteBuffer(&device_handle, sizeof(device_handle));
 
-        LOG_DEBUG(Service_NFP, "called, array_size={}", array_size);
-
         IPC::ResponseBuilder rb{ctx, 3};
         rb.Push(RESULT_SUCCESS);
         rb.Push<u32>(1);
@@ -141,6 +140,7 @@ private:
         IPC::RequestParser rp{ctx};
         const u64 dev_handle = rp.Pop<u64>();
         LOG_DEBUG(Service_NFP, "called, dev_handle=0x{:X}", dev_handle);
+
         IPC::ResponseBuilder rb{ctx, 3};
         rb.Push(RESULT_SUCCESS);
         rb.Push<u32>(npad_id);
@@ -150,6 +150,7 @@ private:
         IPC::RequestParser rp{ctx};
         const u64 dev_handle = rp.Pop<u64>();
         LOG_DEBUG(Service_NFP, "called, dev_handle=0x{:X}", dev_handle);
+
         IPC::ResponseBuilder rb{ctx, 2, 1};
         rb.Push(RESULT_SUCCESS);
         rb.PushCopyObjects(nfp_interface.GetNFCEvent());
@@ -168,6 +169,7 @@ private:
 
     void StopDetection(Kernel::HLERequestContext& ctx) {
         LOG_DEBUG(Service_NFP, "called");
+
         switch (device_state) {
         case DeviceState::TagFound:
         case DeviceState::TagNearby:
@@ -185,6 +187,7 @@ private:
 
     void GetDeviceState(Kernel::HLERequestContext& ctx) {
         LOG_DEBUG(Service_NFP, "called");
+
         auto nfc_event = nfp_interface.GetNFCEvent();
         if (!nfc_event->ShouldWait(Kernel::GetCurrentThread()) && !has_attached_handle) {
             device_state = DeviceState::TagFound;
@@ -323,6 +326,7 @@ private:
 
 void Module::Interface::CreateUserInterface(Kernel::HLERequestContext& ctx) {
     LOG_DEBUG(Service_NFP, "called");
+
     IPC::ResponseBuilder rb{ctx, 2, 0, 1};
     rb.Push(RESULT_SUCCESS);
     rb.PushIpcInterface<IUser>(*this);

src/core/hle/service/nifm/nifm.cpp

@@ -63,12 +63,14 @@ public:
 private:
     void Submit(Kernel::HLERequestContext& ctx) {
         LOG_WARNING(Service_NIFM, "(STUBBED) called");
+
         IPC::ResponseBuilder rb{ctx, 2};
         rb.Push(RESULT_SUCCESS);
     }
 
     void GetRequestState(Kernel::HLERequestContext& ctx) {
         LOG_WARNING(Service_NIFM, "(STUBBED) called");
+
         IPC::ResponseBuilder rb{ctx, 3};
         rb.Push(RESULT_SUCCESS);
         rb.Push<u32>(0);
@@ -76,12 +78,14 @@ private:
 
     void GetResult(Kernel::HLERequestContext& ctx) {
         LOG_WARNING(Service_NIFM, "(STUBBED) called");
+
         IPC::ResponseBuilder rb{ctx, 2};
         rb.Push(RESULT_SUCCESS);
     }
 
     void GetSystemEventReadableHandles(Kernel::HLERequestContext& ctx) {
         LOG_WARNING(Service_NIFM, "(STUBBED) called");
+
         IPC::ResponseBuilder rb{ctx, 2, 2};
         rb.Push(RESULT_SUCCESS);
         rb.PushCopyObjects(event1, event2);
@@ -89,12 +93,14 @@ private:
 
     void Cancel(Kernel::HLERequestContext& ctx) {
         LOG_WARNING(Service_NIFM, "(STUBBED) called");
+
         IPC::ResponseBuilder rb{ctx, 2};
         rb.Push(RESULT_SUCCESS);
     }
 
     void SetConnectionConfirmationOption(Kernel::HLERequestContext& ctx) {
         LOG_WARNING(Service_NIFM, "(STUBBED) called");
+
         IPC::ResponseBuilder rb{ctx, 2};
         rb.Push(RESULT_SUCCESS);
     }
@@ -122,32 +128,36 @@ private:
     void GetClientId(Kernel::HLERequestContext& ctx) {
         static constexpr u32 client_id = 1;
         LOG_WARNING(Service_NIFM, "(STUBBED) called");
+
         IPC::ResponseBuilder rb{ctx, 4};
         rb.Push(RESULT_SUCCESS);
         rb.Push<u64>(client_id); // Client ID needs to be non zero otherwise it's considered invalid
     }
     void CreateScanRequest(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_NIFM, "called");
+
         IPC::ResponseBuilder rb{ctx, 2, 0, 1};
 
         rb.Push(RESULT_SUCCESS);
         rb.PushIpcInterface<IScanRequest>();
-
-        LOG_DEBUG(Service_NIFM, "called");
     }
     void CreateRequest(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_NIFM, "called");
+
         IPC::ResponseBuilder rb{ctx, 2, 0, 1};
 
         rb.Push(RESULT_SUCCESS);
         rb.PushIpcInterface<IRequest>();
-
-        LOG_DEBUG(Service_NIFM, "called");
     }
     void RemoveNetworkProfile(Kernel::HLERequestContext& ctx) {
         LOG_WARNING(Service_NIFM, "(STUBBED) called");
+
         IPC::ResponseBuilder rb{ctx, 2};
         rb.Push(RESULT_SUCCESS);
     }
     void CreateTemporaryNetworkProfile(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_NIFM, "called");
+
         ASSERT_MSG(ctx.GetReadBufferSize() == 0x17c, "NetworkProfileData is not the correct size");
         u128 uuid{};
         auto buffer = ctx.ReadBuffer();
@@ -158,23 +168,24 @@ private:
         rb.Push(RESULT_SUCCESS);
         rb.PushIpcInterface<INetworkProfile>();
         rb.PushRaw<u128>(uuid);
-
-        LOG_DEBUG(Service_NIFM, "called");
     }
     void IsWirelessCommunicationEnabled(Kernel::HLERequestContext& ctx) {
         LOG_WARNING(Service_NIFM, "(STUBBED) called");
+
         IPC::ResponseBuilder rb{ctx, 3};
         rb.Push(RESULT_SUCCESS);
         rb.Push<u8>(0);
     }
     void IsEthernetCommunicationEnabled(Kernel::HLERequestContext& ctx) {
         LOG_WARNING(Service_NIFM, "(STUBBED) called");
+
         IPC::ResponseBuilder rb{ctx, 3};
         rb.Push(RESULT_SUCCESS);
         rb.Push<u8>(0);
     }
     void IsAnyInternetRequestAccepted(Kernel::HLERequestContext& ctx) {
         LOG_WARNING(Service_NIFM, "(STUBBED) called");
+
         IPC::ResponseBuilder rb{ctx, 3};
         rb.Push(RESULT_SUCCESS);
         rb.Push<u8>(0);
@@ -235,17 +246,19 @@ public:
     }
 
     void CreateGeneralServiceOld(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_NIFM, "called");
+
         IPC::ResponseBuilder rb{ctx, 2, 0, 1};
         rb.Push(RESULT_SUCCESS);
         rb.PushIpcInterface<IGeneralService>();
-        LOG_DEBUG(Service_NIFM, "called");
     }
 
     void CreateGeneralService(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_NIFM, "called");
+
         IPC::ResponseBuilder rb{ctx, 2, 0, 1};
         rb.Push(RESULT_SUCCESS);
         rb.PushIpcInterface<IGeneralService>();
-        LOG_DEBUG(Service_NIFM, "called");
     }
 };
 

src/core/hle/service/nim/nim.cpp

@@ -148,47 +148,53 @@ private:
 
     void StartTask(Kernel::HLERequestContext& ctx) {
         // No need to connect to the internet, just finish the task straight away.
+        LOG_DEBUG(Service_NIM, "called");
+
         finished_event->Signal();
         IPC::ResponseBuilder rb{ctx, 2};
         rb.Push(RESULT_SUCCESS);
-        LOG_DEBUG(Service_NIM, "called");
     }
 
     void GetFinishNotificationEvent(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_NIM, "called");
+
         IPC::ResponseBuilder rb{ctx, 2, 1};
         rb.Push(RESULT_SUCCESS);
         rb.PushCopyObjects(finished_event);
-        LOG_DEBUG(Service_NIM, "called");
     }
 
     void GetResult(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_NIM, "called");
+
         IPC::ResponseBuilder rb{ctx, 2};
         rb.Push(RESULT_SUCCESS);
-        LOG_DEBUG(Service_NIM, "called");
     }
 
     void Cancel(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_NIM, "called");
+
         finished_event->Clear();
         IPC::ResponseBuilder rb{ctx, 2};
         rb.Push(RESULT_SUCCESS);
-        LOG_DEBUG(Service_NIM, "called");
     }
 
     void IsProcessing(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_NIM, "called");
+
         IPC::ResponseBuilder rb{ctx, 3};
         rb.Push(RESULT_SUCCESS);
         rb.PushRaw<u32>(0); // We instantly process the request
-        LOG_DEBUG(Service_NIM, "called");
     }
 
     void GetServerTime(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_NIM, "called");
+
         const s64 server_time{std::chrono::duration_cast<std::chrono::seconds>(
                                   std::chrono::system_clock::now().time_since_epoch())
                                   .count()};
         IPC::ResponseBuilder rb{ctx, 4};
         rb.Push(RESULT_SUCCESS);
         rb.PushRaw<s64>(server_time);
-        LOG_DEBUG(Service_NIM, "called");
     }
 };
 
@@ -208,23 +214,26 @@ public:
 
 private:
     void OpenEnsureNetworkClockAvailabilityService(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_NIM, "called");
+
         IPC::ResponseBuilder rb{ctx, 2, 0, 1};
         rb.Push(RESULT_SUCCESS);
         rb.PushIpcInterface<IEnsureNetworkClockAvailabilityService>();
-        LOG_DEBUG(Service_NIM, "called");
     }
 
     // TODO(ogniK): Do we need these?
     void SuspendAutonomicTimeCorrection(Kernel::HLERequestContext& ctx) {
+        LOG_WARNING(Service_NIM, "(STUBBED) called");
+
         IPC::ResponseBuilder rb{ctx, 2};
         rb.Push(RESULT_SUCCESS);
-        LOG_WARNING(Service_NIM, "(STUBBED) called");
     }
 
     void ResumeAutonomicTimeCorrection(Kernel::HLERequestContext& ctx) {
+        LOG_WARNING(Service_NIM, "(STUBBED) called");
+
         IPC::ResponseBuilder rb{ctx, 2};
         rb.Push(RESULT_SUCCESS);
-        LOG_WARNING(Service_NIM, "(STUBBED) called");
     }
 };
 

src/core/hle/service/ns/ns.cpp

@@ -433,11 +433,11 @@ public:
 private:
     template <typename T>
     void PushInterface(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_NS, "called");
+
         IPC::ResponseBuilder rb{ctx, 2, 0, 1};
         rb.Push(RESULT_SUCCESS);
         rb.PushIpcInterface<T>();
-
-        LOG_DEBUG(Service_NS, "called");
     }
 };
 
@@ -526,11 +526,11 @@ public:
 
 private:
     void OpenSystemUpdateControl(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_NS, "called");
+
         IPC::ResponseBuilder rb{ctx, 2, 0, 1};
         rb.Push(RESULT_SUCCESS);
         rb.PushIpcInterface<ISystemUpdateControl>();
-
-        LOG_DEBUG(Service_NS, "called");
     }
 };
 

src/core/hle/service/ns/pl_u.cpp

@@ -281,6 +281,7 @@ void PL_U::RequestLoad(Kernel::HLERequestContext& ctx) {
     const u32 shared_font_type{rp.Pop<u32>()};
     // Games don't call this so all fonts should be loaded
     LOG_DEBUG(Service_NS, "called, shared_font_type={}", shared_font_type);
+
     IPC::ResponseBuilder rb{ctx, 2};
     rb.Push(RESULT_SUCCESS);
 }
@@ -288,8 +289,8 @@ void PL_U::RequestLoad(Kernel::HLERequestContext& ctx) {
 void PL_U::GetLoadState(Kernel::HLERequestContext& ctx) {
     IPC::RequestParser rp{ctx};
     const u32 font_id{rp.Pop<u32>()};
-
     LOG_DEBUG(Service_NS, "called, font_id={}", font_id);
+
     IPC::ResponseBuilder rb{ctx, 3};
     rb.Push(RESULT_SUCCESS);
     rb.Push<u32>(static_cast<u32>(LoadState::Done));
@@ -298,8 +299,8 @@ void PL_U::GetLoadState(Kernel::HLERequestContext& ctx) {
 void PL_U::GetSize(Kernel::HLERequestContext& ctx) {
     IPC::RequestParser rp{ctx};
     const u32 font_id{rp.Pop<u32>()};
-
     LOG_DEBUG(Service_NS, "called, font_id={}", font_id);
+
     IPC::ResponseBuilder rb{ctx, 3};
     rb.Push(RESULT_SUCCESS);
     rb.Push<u32>(impl->GetSharedFontRegion(font_id).size);
@@ -308,8 +309,8 @@ void PL_U::GetSize(Kernel::HLERequestContext& ctx) {
 void PL_U::GetSharedMemoryAddressOffset(Kernel::HLERequestContext& ctx) {
     IPC::RequestParser rp{ctx};
     const u32 font_id{rp.Pop<u32>()};
-
     LOG_DEBUG(Service_NS, "called, font_id={}", font_id);
+
     IPC::ResponseBuilder rb{ctx, 3};
     rb.Push(RESULT_SUCCESS);
     rb.Push<u32>(impl->GetSharedFontRegion(font_id).offset);
@@ -317,6 +318,7 @@ void PL_U::GetSharedMemoryAddressOffset(Kernel::HLERequestContext& ctx) {
 
 void PL_U::GetSharedMemoryNativeHandle(Kernel::HLERequestContext& ctx) {
     // Map backing memory for the font data
+    LOG_DEBUG(Service_NS, "called");
     Core::CurrentProcess()->VMManager().MapMemoryBlock(SHARED_FONT_MEM_VADDR, impl->shared_font, 0,
                                                        SHARED_FONT_MEM_SIZE,
                                                        Kernel::MemoryState::Shared);
@@ -328,7 +330,6 @@ void PL_U::GetSharedMemoryNativeHandle(Kernel::HLERequestContext& ctx) {
         Kernel::MemoryPermission::Read, SHARED_FONT_MEM_VADDR, Kernel::MemoryRegion::BASE,
         "PL_U:shared_font_mem");
 
-    LOG_DEBUG(Service_NS, "called");
     IPC::ResponseBuilder rb{ctx, 2, 1};
     rb.Push(RESULT_SUCCESS);
     rb.PushCopyObjects(impl->shared_font_mem);
@@ -338,6 +339,7 @@ void PL_U::GetSharedFontInOrderOfPriority(Kernel::HLERequestContext& ctx) {
     IPC::RequestParser rp{ctx};
     const u64 language_code{rp.Pop<u64>()}; // TODO(ogniK): Find out what this is used for
     LOG_DEBUG(Service_NS, "called, language_code={:X}", language_code);
+
     IPC::ResponseBuilder rb{ctx, 4};
     std::vector<u32> font_codes;
     std::vector<u32> font_offsets;

src/core/hle/service/nvdrv/devices/nvhost_as_gpu.cpp

@@ -54,6 +54,7 @@ u32 nvhost_as_gpu::InitalizeEx(const std::vector<u8>& input, std::vector<u8>& ou
     IoctlInitalizeEx params{};
     std::memcpy(&params, input.data(), input.size());
     LOG_WARNING(Service_NVDRV, "(STUBBED) called, big_page_size=0x{:X}", params.big_page_size);
+
     return 0;
 }
 
@@ -191,6 +192,7 @@ u32 nvhost_as_gpu::BindChannel(const std::vector<u8>& input, std::vector<u8>& ou
     IoctlBindChannel params{};
     std::memcpy(&params, input.data(), input.size());
     LOG_DEBUG(Service_NVDRV, "called, fd={:X}", params.fd);
+
     channel = params.fd;
     return 0;
 }

src/core/hle/service/nvdrv/devices/nvhost_ctrl_gpu.cpp

@@ -103,6 +103,7 @@ u32 nvhost_ctrl_gpu::GetTPCMasks(const std::vector<u8>& input, std::vector<u8>&
 
 u32 nvhost_ctrl_gpu::GetActiveSlotMask(const std::vector<u8>& input, std::vector<u8>& output) {
     LOG_DEBUG(Service_NVDRV, "called");
+
     IoctlActiveSlotMask params{};
     if (input.size() > 0) {
         std::memcpy(&params, input.data(), input.size());
@@ -115,6 +116,7 @@ u32 nvhost_ctrl_gpu::GetActiveSlotMask(const std::vector<u8>& input, std::vector
 
 u32 nvhost_ctrl_gpu::ZCullGetCtxSize(const std::vector<u8>& input, std::vector<u8>& output) {
     LOG_DEBUG(Service_NVDRV, "called");
+
     IoctlZcullGetCtxSize params{};
     if (input.size() > 0) {
         std::memcpy(&params, input.data(), input.size());
@@ -126,6 +128,7 @@ u32 nvhost_ctrl_gpu::ZCullGetCtxSize(const std::vector<u8>& input, std::vector<u
 
 u32 nvhost_ctrl_gpu::ZCullGetInfo(const std::vector<u8>& input, std::vector<u8>& output) {
     LOG_DEBUG(Service_NVDRV, "called");
+
     IoctlNvgpuGpuZcullGetInfoArgs params{};
 
     if (input.size() > 0) {
@@ -148,6 +151,7 @@ u32 nvhost_ctrl_gpu::ZCullGetInfo(const std::vector<u8>& input, std::vector<u8>&
 
 u32 nvhost_ctrl_gpu::ZBCSetTable(const std::vector<u8>& input, std::vector<u8>& output) {
     LOG_WARNING(Service_NVDRV, "(STUBBED) called");
+
     IoctlZbcSetTable params{};
     std::memcpy(&params, input.data(), input.size());
     // TODO(ogniK): What does this even actually do?
@@ -157,6 +161,7 @@ u32 nvhost_ctrl_gpu::ZBCSetTable(const std::vector<u8>& input, std::vector<u8>&
 
 u32 nvhost_ctrl_gpu::ZBCQueryTable(const std::vector<u8>& input, std::vector<u8>& output) {
     LOG_WARNING(Service_NVDRV, "(STUBBED) called");
+
     IoctlZbcQueryTable params{};
     std::memcpy(&params, input.data(), input.size());
     // TODO : To implement properly
@@ -166,6 +171,7 @@ u32 nvhost_ctrl_gpu::ZBCQueryTable(const std::vector<u8>& input, std::vector<u8>
 
 u32 nvhost_ctrl_gpu::FlushL2(const std::vector<u8>& input, std::vector<u8>& output) {
     LOG_WARNING(Service_NVDRV, "(STUBBED) called");
+
     IoctlFlushL2 params{};
     std::memcpy(&params, input.data(), input.size());
     // TODO : To implement properly
@@ -175,6 +181,7 @@ u32 nvhost_ctrl_gpu::FlushL2(const std::vector<u8>& input, std::vector<u8>& outp
 
 u32 nvhost_ctrl_gpu::GetGpuTime(const std::vector<u8>& input, std::vector<u8>& output) {
     LOG_DEBUG(Service_NVDRV, "called");
+
     IoctlGetGpuTime params{};
     std::memcpy(&params, input.data(), input.size());
     params.gpu_time = CoreTiming::cyclesToNs(CoreTiming::GetTicks());

src/core/hle/service/nvdrv/devices/nvhost_gpu.cpp

@@ -8,7 +8,6 @@
 #include "core/core.h"
 #include "core/hle/service/nvdrv/devices/nvhost_gpu.h"
 #include "core/memory.h"
-#include "video_core/command_processor.h"
 #include "video_core/gpu.h"
 #include "video_core/memory_manager.h"
 
@@ -61,12 +60,14 @@ u32 nvhost_gpu::SetNVMAPfd(const std::vector<u8>& input, std::vector<u8>& output
     IoctlSetNvmapFD params{};
     std::memcpy(&params, input.data(), input.size());
     LOG_DEBUG(Service_NVDRV, "called, fd={}", params.nvmap_fd);
+
     nvmap_fd = params.nvmap_fd;
     return 0;
 }
 
 u32 nvhost_gpu::SetClientData(const std::vector<u8>& input, std::vector<u8>& output) {
     LOG_DEBUG(Service_NVDRV, "called");
+
     IoctlClientData params{};
     std::memcpy(&params, input.data(), input.size());
     user_data = params.data;
@@ -75,6 +76,7 @@ u32 nvhost_gpu::SetClientData(const std::vector<u8>& input, std::vector<u8>& out
 
 u32 nvhost_gpu::GetClientData(const std::vector<u8>& input, std::vector<u8>& output) {
     LOG_DEBUG(Service_NVDRV, "called");
+
     IoctlClientData params{};
     std::memcpy(&params, input.data(), input.size());
     params.data = user_data;
@@ -86,6 +88,7 @@ u32 nvhost_gpu::ZCullBind(const std::vector<u8>& input, std::vector<u8>& output)
     std::memcpy(&zcull_params, input.data(), input.size());
     LOG_DEBUG(Service_NVDRV, "called, gpu_va={:X}, mode={:X}", zcull_params.gpu_va,
               zcull_params.mode);
+
     std::memcpy(output.data(), &zcull_params, output.size());
     return 0;
 }
@@ -95,6 +98,7 @@ u32 nvhost_gpu::SetErrorNotifier(const std::vector<u8>& input, std::vector<u8>&
     std::memcpy(&params, input.data(), input.size());
     LOG_WARNING(Service_NVDRV, "(STUBBED) called, offset={:X}, size={:X}, mem={:X}", params.offset,
                 params.size, params.mem);
+
     std::memcpy(output.data(), &params, output.size());
     return 0;
 }
@@ -102,6 +106,7 @@ u32 nvhost_gpu::SetErrorNotifier(const std::vector<u8>& input, std::vector<u8>&
 u32 nvhost_gpu::SetChannelPriority(const std::vector<u8>& input, std::vector<u8>& output) {
     std::memcpy(&channel_priority, input.data(), input.size());
     LOG_DEBUG(Service_NVDRV, "(STUBBED) called, priority={:X}", channel_priority);
+
     return 0;
 }
 
@@ -113,6 +118,7 @@ u32 nvhost_gpu::AllocGPFIFOEx2(const std::vector<u8>& input, std::vector<u8>& ou
                 "unk1={:X}, unk2={:X}, unk3={:X}",
                 params.num_entries, params.flags, params.unk0, params.unk1, params.unk2,
                 params.unk3);
+
     params.fence_out.id = 0;
     params.fence_out.value = 0;
     std::memcpy(output.data(), &params, output.size());
@@ -124,11 +130,18 @@ u32 nvhost_gpu::AllocateObjectContext(const std::vector<u8>& input, std::vector<
     std::memcpy(&params, input.data(), input.size());
     LOG_WARNING(Service_NVDRV, "(STUBBED) called, class_num={:X}, flags={:X}", params.class_num,
                 params.flags);
+
     params.obj_id = 0x0;
     std::memcpy(output.data(), &params, output.size());
     return 0;
 }
 
+static void PushGPUEntries(Tegra::CommandList&& entries) {
+    auto& dma_pusher{Core::System::GetInstance().GPU().DmaPusher()};
+    dma_pusher.Push(std::move(entries));
+    dma_pusher.DispatchCalls();
+}
+
 u32 nvhost_gpu::SubmitGPFIFO(const std::vector<u8>& input, std::vector<u8>& output) {
     if (input.size() < sizeof(IoctlSubmitGpfifo)) {
         UNIMPLEMENTED();
@@ -142,11 +155,11 @@ u32 nvhost_gpu::SubmitGPFIFO(const std::vector<u8>& input, std::vector<u8>& outp
                                    params.num_entries * sizeof(Tegra::CommandListHeader),
                "Incorrect input size");
 
-    std::vector<Tegra::CommandListHeader> entries(params.num_entries);
+    Tegra::CommandList entries(params.num_entries);
     std::memcpy(entries.data(), &input[sizeof(IoctlSubmitGpfifo)],
                 params.num_entries * sizeof(Tegra::CommandListHeader));
 
-    Core::System::GetInstance().GPU().ProcessCommandLists(entries);
+    PushGPUEntries(std::move(entries));
 
     params.fence_out.id = 0;
     params.fence_out.value = 0;
@@ -163,11 +176,11 @@ u32 nvhost_gpu::KickoffPB(const std::vector<u8>& input, std::vector<u8>& output)
     LOG_WARNING(Service_NVDRV, "(STUBBED) called, gpfifo={:X}, num_entries={:X}, flags={:X}",
                 params.address, params.num_entries, params.flags);
 
-    std::vector<Tegra::CommandListHeader> entries(params.num_entries);
+    Tegra::CommandList entries(params.num_entries);
     Memory::ReadBlock(params.address, entries.data(),
                       params.num_entries * sizeof(Tegra::CommandListHeader));
 
-    Core::System::GetInstance().GPU().ProcessCommandLists(entries);
+    PushGPUEntries(std::move(entries));
 
     params.fence_out.id = 0;
     params.fence_out.value = 0;
@@ -179,6 +192,7 @@ u32 nvhost_gpu::GetWaitbase(const std::vector<u8>& input, std::vector<u8>& outpu
     IoctlGetWaitbase params{};
     std::memcpy(&params, input.data(), sizeof(IoctlGetWaitbase));
     LOG_INFO(Service_NVDRV, "called, unknown=0x{:X}", params.unknown);
+
     params.value = 0; // Seems to be hard coded at 0
     std::memcpy(output.data(), &params, output.size());
     return 0;
@@ -188,6 +202,7 @@ u32 nvhost_gpu::ChannelSetTimeout(const std::vector<u8>& input, std::vector<u8>&
     IoctlChannelSetTimeout params{};
     std::memcpy(&params, input.data(), sizeof(IoctlChannelSetTimeout));
     LOG_INFO(Service_NVDRV, "called, timeout=0x{:X}", params.timeout);
+
     return 0;
 }
 

src/core/hle/service/nvdrv/devices/nvhost_nvdec.cpp

@@ -30,6 +30,7 @@ u32 nvhost_nvdec::SetNVMAPfd(const std::vector<u8>& input, std::vector<u8>& outp
     IoctlSetNvmapFD params{};
     std::memcpy(&params, input.data(), input.size());
     LOG_DEBUG(Service_NVDRV, "called, fd={}", params.nvmap_fd);
+
     nvmap_fd = params.nvmap_fd;
     return 0;
 }

src/core/hle/service/nvdrv/devices/nvhost_nvjpg.cpp

@@ -30,6 +30,7 @@ u32 nvhost_nvjpg::SetNVMAPfd(const std::vector<u8>& input, std::vector<u8>& outp
     IoctlSetNvmapFD params{};
     std::memcpy(&params, input.data(), input.size());
     LOG_DEBUG(Service_NVDRV, "called, fd={}", params.nvmap_fd);
+
     nvmap_fd = params.nvmap_fd;
     return 0;
 }

src/core/hle/service/nvdrv/devices/nvhost_vic.cpp

@@ -30,6 +30,7 @@ u32 nvhost_vic::SetNVMAPfd(const std::vector<u8>& input, std::vector<u8>& output
     IoctlSetNvmapFD params{};
     std::memcpy(&params, input.data(), input.size());
     LOG_DEBUG(Service_NVDRV, "called, fd={}", params.nvmap_fd);
+
     nvmap_fd = params.nvmap_fd;
     return 0;
 }

src/core/hle/service/nvdrv/devices/nvmap.cpp

@@ -54,6 +54,7 @@ u32 nvmap::IocCreate(const std::vector<u8>& input, std::vector<u8>& output) {
     LOG_DEBUG(Service_NVDRV, "size=0x{:08X}", params.size);
 
     if (!params.size) {
+        LOG_ERROR(Service_NVDRV, "Size is 0");
         return static_cast<u32>(NvErrCodes::InvalidValue);
     }
     // Create a new nvmap object and obtain a handle to it.
@@ -78,10 +79,12 @@ u32 nvmap::IocAlloc(const std::vector<u8>& input, std::vector<u8>& output) {
     LOG_DEBUG(Service_NVDRV, "called, addr={:X}", params.addr);
 
     if (!params.handle) {
+        LOG_ERROR(Service_NVDRV, "Handle is 0");
         return static_cast<u32>(NvErrCodes::InvalidValue);
     }
 
     if ((params.align - 1) & params.align) {
+        LOG_ERROR(Service_NVDRV, "Incorrect alignment used, alignment={:08X}", params.align);
         return static_cast<u32>(NvErrCodes::InvalidValue);
     }
 
@@ -92,10 +95,12 @@ u32 nvmap::IocAlloc(const std::vector<u8>& input, std::vector<u8>& output) {
 
     auto object = GetObject(params.handle);
     if (!object) {
+        LOG_ERROR(Service_NVDRV, "Object does not exist, handle={:08X}", params.handle);
         return static_cast<u32>(NvErrCodes::InvalidValue);
     }
 
     if (object->status == Object::Status::Allocated) {
+        LOG_ERROR(Service_NVDRV, "Object is already allocated, handle={:08X}", params.handle);
         return static_cast<u32>(NvErrCodes::OperationNotPermitted);
     }
 
@@ -116,11 +121,13 @@ u32 nvmap::IocGetId(const std::vector<u8>& input, std::vector<u8>& output) {
     LOG_WARNING(Service_NVDRV, "called");
 
     if (!params.handle) {
+        LOG_ERROR(Service_NVDRV, "Handle is zero");
         return static_cast<u32>(NvErrCodes::InvalidValue);
     }
 
     auto object = GetObject(params.handle);
     if (!object) {
+        LOG_ERROR(Service_NVDRV, "Object does not exist, handle={:08X}", params.handle);
         return static_cast<u32>(NvErrCodes::OperationNotPermitted);
     }
 
@@ -139,11 +146,13 @@ u32 nvmap::IocFromId(const std::vector<u8>& input, std::vector<u8>& output) {
     auto itr = std::find_if(handles.begin(), handles.end(),
                             [&](const auto& entry) { return entry.second->id == params.id; });
     if (itr == handles.end()) {
+        LOG_ERROR(Service_NVDRV, "Object does not exist, handle={:08X}", params.handle);
         return static_cast<u32>(NvErrCodes::InvalidValue);
     }
 
     auto& object = itr->second;
     if (object->status != Object::Status::Allocated) {
+        LOG_ERROR(Service_NVDRV, "Object is not allocated, handle={:08X}", params.handle);
         return static_cast<u32>(NvErrCodes::InvalidValue);
     }
 
@@ -166,10 +175,12 @@ u32 nvmap::IocParam(const std::vector<u8>& input, std::vector<u8>& output) {
 
     auto object = GetObject(params.handle);
     if (!object) {
+        LOG_ERROR(Service_NVDRV, "Object does not exist, handle={:08X}", params.handle);
         return static_cast<u32>(NvErrCodes::InvalidValue);
     }
 
     if (object->status != Object::Status::Allocated) {
+        LOG_ERROR(Service_NVDRV, "Object is not allocated, handle={:08X}", params.handle);
         return static_cast<u32>(NvErrCodes::OperationNotPermitted);
     }
 
@@ -209,9 +220,14 @@ u32 nvmap::IocFree(const std::vector<u8>& input, std::vector<u8>& output) {
 
     auto itr = handles.find(params.handle);
     if (itr == handles.end()) {
+        LOG_ERROR(Service_NVDRV, "Object does not exist, handle={:08X}", params.handle);
         return static_cast<u32>(NvErrCodes::InvalidValue);
     }
     if (!itr->second->refcount) {
+        LOG_ERROR(
+            Service_NVDRV,
+            "There is no references to this object. The object is already freed. handle={:08X}",
+            params.handle);
         return static_cast<u32>(NvErrCodes::InvalidValue);
     }
 

src/core/hle/service/nvdrv/interface.cpp

@@ -55,6 +55,7 @@ void NVDRV::Close(Kernel::HLERequestContext& ctx) {
 
 void NVDRV::Initialize(Kernel::HLERequestContext& ctx) {
     LOG_WARNING(Service_NVDRV, "(STUBBED) called");
+
     IPC::ResponseBuilder rb{ctx, 3};
     rb.Push(RESULT_SUCCESS);
     rb.Push<u32>(0);
@@ -75,8 +76,8 @@ void NVDRV::QueryEvent(Kernel::HLERequestContext& ctx) {
 void NVDRV::SetClientPID(Kernel::HLERequestContext& ctx) {
     IPC::RequestParser rp{ctx};
     pid = rp.Pop<u64>();
-
     LOG_WARNING(Service_NVDRV, "(STUBBED) called, pid=0x{:X}", pid);
+
     IPC::ResponseBuilder rb{ctx, 3};
     rb.Push(RESULT_SUCCESS);
     rb.Push<u32>(0);
@@ -84,6 +85,23 @@ void NVDRV::SetClientPID(Kernel::HLERequestContext& ctx) {
 
 void NVDRV::FinishInitialize(Kernel::HLERequestContext& ctx) {
     LOG_WARNING(Service_NVDRV, "(STUBBED) called");
+
+    IPC::ResponseBuilder rb{ctx, 2};
+    rb.Push(RESULT_SUCCESS);
+}
+
+void NVDRV::GetStatus(Kernel::HLERequestContext& ctx) {
+    LOG_WARNING(Service_NVDRV, "(STUBBED) called");
+
+    IPC::ResponseBuilder rb{ctx, 2};
+    rb.Push(RESULT_SUCCESS);
+}
+
+void NVDRV::DumpGraphicsMemoryInfo(Kernel::HLERequestContext& ctx) {
+    // According to SwitchBrew, this has no inputs and no outputs, so effectively does nothing on
+    // retail hardware.
+    LOG_DEBUG(Service_NVDRV, "called");
+
     IPC::ResponseBuilder rb{ctx, 2};
     rb.Push(RESULT_SUCCESS);
 }
@@ -97,10 +115,10 @@ NVDRV::NVDRV(std::shared_ptr<Module> nvdrv, const char* name)
         {3, &NVDRV::Initialize, "Initialize"},
         {4, &NVDRV::QueryEvent, "QueryEvent"},
         {5, nullptr, "MapSharedMem"},
-        {6, nullptr, "GetStatus"},
+        {6, &NVDRV::GetStatus, "GetStatus"},
         {7, nullptr, "ForceSetClientPID"},
         {8, &NVDRV::SetClientPID, "SetClientPID"},
-        {9, nullptr, "DumpGraphicsMemoryInfo"},
+        {9, &NVDRV::DumpGraphicsMemoryInfo, "DumpGraphicsMemoryInfo"},
         {10, nullptr, "InitializeDevtools"},
         {11, &NVDRV::Ioctl, "Ioctl2"},
         {12, nullptr, "Ioctl3"},

src/core/hle/service/nvdrv/interface.h

@@ -24,6 +24,8 @@ private:
     void QueryEvent(Kernel::HLERequestContext& ctx);
     void SetClientPID(Kernel::HLERequestContext& ctx);
     void FinishInitialize(Kernel::HLERequestContext& ctx);
+    void GetStatus(Kernel::HLERequestContext& ctx);
+    void DumpGraphicsMemoryInfo(Kernel::HLERequestContext& ctx);
 
     std::shared_ptr<Module> nvdrv;
 

src/core/hle/service/nvflinger/buffer_queue.cpp

@@ -20,13 +20,13 @@ BufferQueue::BufferQueue(u32 id, u64 layer_id) : id(id), layer_id(layer_id) {
 BufferQueue::~BufferQueue() = default;
 
 void BufferQueue::SetPreallocatedBuffer(u32 slot, const IGBPBuffer& igbp_buffer) {
+    LOG_WARNING(Service, "Adding graphics buffer {}", slot);
+
     Buffer buffer{};
     buffer.slot = slot;
     buffer.igbp_buffer = igbp_buffer;
     buffer.status = Buffer::Status::Free;
 
-    LOG_WARNING(Service, "Adding graphics buffer {}", slot);
-
     queue.emplace_back(buffer);
     buffer_wait_event->Signal();
 }
@@ -92,6 +92,7 @@ void BufferQueue::ReleaseBuffer(u32 slot) {
 
 u32 BufferQueue::Query(QueryType type) {
     LOG_WARNING(Service, "(STUBBED) called type={}", static_cast<u32>(type));
+
     switch (type) {
     case QueryType::NativeWindowFormat:
         // TODO(Subv): Use an enum for this

src/core/hle/service/pctl/module.cpp

@@ -114,29 +114,33 @@ public:
 private:
     void Initialize(Kernel::HLERequestContext& ctx) {
         LOG_WARNING(Service_PCTL, "(STUBBED) called");
+
         IPC::ResponseBuilder rb{ctx, 2, 0, 0};
         rb.Push(RESULT_SUCCESS);
     }
 
     void CheckFreeCommunicationPermission(Kernel::HLERequestContext& ctx) {
         LOG_WARNING(Service_PCTL, "(STUBBED) called");
+
         IPC::ResponseBuilder rb{ctx, 2};
         rb.Push(RESULT_SUCCESS);
     }
 };
 
 void Module::Interface::CreateService(Kernel::HLERequestContext& ctx) {
+    LOG_DEBUG(Service_PCTL, "called");
+
     IPC::ResponseBuilder rb{ctx, 2, 0, 1};
     rb.Push(RESULT_SUCCESS);
     rb.PushIpcInterface<IParentalControlService>();
-    LOG_DEBUG(Service_PCTL, "called");
 }
 
 void Module::Interface::CreateServiceWithoutInitialize(Kernel::HLERequestContext& ctx) {
+    LOG_DEBUG(Service_PCTL, "called");
+
     IPC::ResponseBuilder rb{ctx, 2, 0, 1};
     rb.Push(RESULT_SUCCESS);
     rb.PushIpcInterface<IParentalControlService>();
-    LOG_DEBUG(Service_PCTL, "called");
 }
 
 Module::Interface::Interface(std::shared_ptr<Module> module, const char* name)

src/core/hle/service/pm/pm.cpp

@@ -20,11 +20,11 @@ public:
 
 private:
     void GetBootMode(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_PM, "called");
+
         IPC::ResponseBuilder rb{ctx, 3};
         rb.Push(RESULT_SUCCESS);
         rb.Push<u32>(static_cast<u32>(SystemBootMode::Normal)); // Normal boot mode
-
-        LOG_DEBUG(Service_PM, "called");
     }
 };
 

src/core/hle/service/psc/psc.cpp

@@ -61,11 +61,11 @@ public:
 
 private:
     void GetPmModule(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_PSC, "called");
+
         IPC::ResponseBuilder rb{ctx, 2, 0, 1};
         rb.Push(RESULT_SUCCESS);
         rb.PushIpcInterface<IPmModule>();
-
-        LOG_DEBUG(Service_PSC, "called");
     }
 };
 

src/core/hle/service/set/set.cpp

@@ -35,6 +35,8 @@ constexpr std::array<LanguageCode, 17> available_language_codes = {{
 constexpr std::size_t pre4_0_0_max_entries = 0xF;
 constexpr std::size_t post4_0_0_max_entries = 0x40;
 
+constexpr ResultCode ERR_INVALID_LANGUAGE{ErrorModule::Settings, 625};
+
 LanguageCode GetLanguageCodeFromIndex(std::size_t index) {
     return available_language_codes.at(index);
 }
@@ -49,38 +51,54 @@ static std::array<LanguageCode, size> MakeLanguageCodeSubset() {
 static void PushResponseLanguageCode(Kernel::HLERequestContext& ctx, std::size_t max_size) {
     IPC::ResponseBuilder rb{ctx, 3};
     rb.Push(RESULT_SUCCESS);
-    if (available_language_codes.size() > max_size)
+    if (available_language_codes.size() > max_size) {
         rb.Push(static_cast<u32>(max_size));
-    else
+    } else {
         rb.Push(static_cast<u32>(available_language_codes.size()));
+    }
 }
 
 void SET::GetAvailableLanguageCodes(Kernel::HLERequestContext& ctx) {
-    if (available_language_codes.size() > pre4_0_0_max_entries)
-        ctx.WriteBuffer(MakeLanguageCodeSubset<pre4_0_0_max_entries>());
-    else
-        ctx.WriteBuffer(available_language_codes);
-
-    PushResponseLanguageCode(ctx, pre4_0_0_max_entries);
-
     LOG_DEBUG(Service_SET, "called");
+
+    if (available_language_codes.size() > pre4_0_0_max_entries) {
+        ctx.WriteBuffer(MakeLanguageCodeSubset<pre4_0_0_max_entries>());
+    } else {
+        ctx.WriteBuffer(available_language_codes);
+    }
+    PushResponseLanguageCode(ctx, pre4_0_0_max_entries);
+}
+
+void SET::MakeLanguageCode(Kernel::HLERequestContext& ctx) {
+    IPC::RequestParser rp{ctx};
+    const auto index = rp.Pop<u32>();
+
+    if (index >= available_language_codes.size()) {
+        IPC::ResponseBuilder rb{ctx, 2};
+        rb.Push(ERR_INVALID_LANGUAGE);
+        return;
+    }
+
+    IPC::ResponseBuilder rb{ctx, 4};
+    rb.Push(RESULT_SUCCESS);
+    rb.PushEnum(available_language_codes[index]);
 }
 
 void SET::GetAvailableLanguageCodes2(Kernel::HLERequestContext& ctx) {
-    if (available_language_codes.size() > post4_0_0_max_entries)
-        ctx.WriteBuffer(MakeLanguageCodeSubset<post4_0_0_max_entries>());
-    else
-        ctx.WriteBuffer(available_language_codes);
-
-    PushResponseLanguageCode(ctx, post4_0_0_max_entries);
-
     LOG_DEBUG(Service_SET, "called");
+
+    if (available_language_codes.size() > post4_0_0_max_entries) {
+        ctx.WriteBuffer(MakeLanguageCodeSubset<post4_0_0_max_entries>());
+    } else {
+        ctx.WriteBuffer(available_language_codes);
+    }
+    PushResponseLanguageCode(ctx, post4_0_0_max_entries);
 }
 
 void SET::GetAvailableLanguageCodeCount(Kernel::HLERequestContext& ctx) {
-    PushResponseLanguageCode(ctx, pre4_0_0_max_entries);
-
     LOG_DEBUG(Service_SET, "called");
+
+    PushResponseLanguageCode(ctx, pre4_0_0_max_entries);
 }
 
 void SET::GetAvailableLanguageCodeCount2(Kernel::HLERequestContext& ctx) {
@@ -90,18 +108,18 @@ void SET::GetAvailableLanguageCodeCount2(Kernel::HLERequestContext& ctx) {
 }
 
 void SET::GetLanguageCode(Kernel::HLERequestContext& ctx) {
+    LOG_DEBUG(Service_SET, "called {}", Settings::values.language_index);
+
     IPC::ResponseBuilder rb{ctx, 4};
     rb.Push(RESULT_SUCCESS);
-    rb.Push(static_cast<u64>(available_language_codes[Settings::values.language_index]));
-
-    LOG_DEBUG(Service_SET, "called {}", Settings::values.language_index);
+    rb.PushEnum(available_language_codes[Settings::values.language_index]);
 }
 
 SET::SET() : ServiceFramework("set") {
     static const FunctionInfo functions[] = {
         {0, &SET::GetLanguageCode, "GetLanguageCode"},
         {1, &SET::GetAvailableLanguageCodes, "GetAvailableLanguageCodes"},
-        {2, nullptr, "MakeLanguageCode"},
+        {2, &SET::MakeLanguageCode, "MakeLanguageCode"},
         {3, &SET::GetAvailableLanguageCodeCount, "GetAvailableLanguageCodeCount"},
         {4, nullptr, "GetRegionCode"},
         {5, &SET::GetAvailableLanguageCodes2, "GetAvailableLanguageCodes2"},

src/core/hle/service/set/set.h

@@ -38,6 +38,7 @@ public:
 private:
     void GetLanguageCode(Kernel::HLERequestContext& ctx);
     void GetAvailableLanguageCodes(Kernel::HLERequestContext& ctx);
+    void MakeLanguageCode(Kernel::HLERequestContext& ctx);
     void GetAvailableLanguageCodes2(Kernel::HLERequestContext& ctx);
     void GetAvailableLanguageCodeCount(Kernel::HLERequestContext& ctx);
     void GetAvailableLanguageCodeCount2(Kernel::HLERequestContext& ctx);

src/core/hle/service/set/set_sys.cpp

@@ -10,22 +10,22 @@
 namespace Service::Set {
 
 void SET_SYS::GetColorSetId(Kernel::HLERequestContext& ctx) {
+    LOG_DEBUG(Service_SET, "called");
+
     IPC::ResponseBuilder rb{ctx, 3};
 
     rb.Push(RESULT_SUCCESS);
     rb.PushEnum(color_set);
-
-    LOG_DEBUG(Service_SET, "called");
 }
 
 void SET_SYS::SetColorSetId(Kernel::HLERequestContext& ctx) {
+    LOG_DEBUG(Service_SET, "called");
+
     IPC::RequestParser rp{ctx};
     color_set = rp.PopEnum<ColorSet>();
 
     IPC::ResponseBuilder rb{ctx, 2};
     rb.Push(RESULT_SUCCESS);
-
-    LOG_DEBUG(Service_SET, "called");
 }
 
 SET_SYS::SET_SYS() : ServiceFramework("set:sys") {

src/core/hle/service/sm/controller.cpp

@@ -14,25 +14,26 @@ namespace Service::SM {
 
 void Controller::ConvertSessionToDomain(Kernel::HLERequestContext& ctx) {
     ASSERT_MSG(ctx.Session()->IsSession(), "Session is already a domain");
+    LOG_DEBUG(Service, "called, server_session={}", ctx.Session()->GetObjectId());
     ctx.Session()->ConvertToDomain();
 
     IPC::ResponseBuilder rb{ctx, 3};
     rb.Push(RESULT_SUCCESS);
     rb.Push<u32>(1); // Converted sessions start with 1 request handler
-
-    LOG_DEBUG(Service, "called, server_session={}", ctx.Session()->GetObjectId());
 }
 
 void Controller::DuplicateSession(Kernel::HLERequestContext& ctx) {
     // TODO(bunnei): This is just creating a new handle to the same Session. I assume this is wrong
     // and that we probably want to actually make an entirely new Session, but we still need to
     // verify this on hardware.
+    LOG_DEBUG(Service, "called");
+
     IPC::ResponseBuilder rb{ctx, 2, 0, 1, IPC::ResponseBuilder::Flags::AlwaysMoveHandles};
     rb.Push(RESULT_SUCCESS);
     Kernel::SharedPtr<Kernel::ClientSession> session{ctx.Session()->parent->client};
     rb.PushMoveObjects(session);
 
-    LOG_DEBUG(Service, "called, session={}", session->GetObjectId());
+    LOG_DEBUG(Service, "session={}", session->GetObjectId());
 }
 
 void Controller::DuplicateSessionEx(Kernel::HLERequestContext& ctx) {
@@ -42,11 +43,11 @@ void Controller::DuplicateSessionEx(Kernel::HLERequestContext& ctx) {
 }
 
 void Controller::QueryPointerBufferSize(Kernel::HLERequestContext& ctx) {
+    LOG_WARNING(Service, "(STUBBED) called");
+
     IPC::ResponseBuilder rb{ctx, 3};
     rb.Push(RESULT_SUCCESS);
     rb.Push<u16>(0x500);
-
-    LOG_WARNING(Service, "(STUBBED) called");
 }
 
 Controller::Controller() : ServiceFramework("IpcController") {

src/core/hle/service/sm/sm.cpp

@@ -103,9 +103,10 @@ SM::~SM() = default;
  *      0: ResultCode
  */
 void SM::Initialize(Kernel::HLERequestContext& ctx) {
+    LOG_DEBUG(Service_SM, "called");
+
     IPC::ResponseBuilder rb{ctx, 2};
     rb.Push(RESULT_SUCCESS);
-    LOG_DEBUG(Service_SM, "called");
 }
 
 void SM::GetService(Kernel::HLERequestContext& ctx) {
@@ -172,6 +173,7 @@ void SM::UnregisterService(Kernel::HLERequestContext& ctx) {
     const auto end = std::find(name_buf.begin(), name_buf.end(), '\0');
 
     const std::string name(name_buf.begin(), end);
+    LOG_DEBUG(Service_SM, "called with name={}", name);
 
     IPC::ResponseBuilder rb{ctx, 2};
     rb.Push(service_manager->UnregisterService(name));

src/core/hle/service/spl/module.cpp

@@ -24,6 +24,8 @@ Module::Interface::Interface(std::shared_ptr<Module> module, const char* name)
 Module::Interface::~Interface() = default;
 
 void Module::Interface::GetRandomBytes(Kernel::HLERequestContext& ctx) {
+    LOG_DEBUG(Service_SPL, "called");
+
     IPC::RequestParser rp{ctx};
 
     std::size_t size = ctx.GetWriteBufferSize();
@@ -36,7 +38,6 @@ void Module::Interface::GetRandomBytes(Kernel::HLERequestContext& ctx) {
 
     IPC::ResponseBuilder rb{ctx, 2};
     rb.Push(RESULT_SUCCESS);
-    LOG_DEBUG(Service_SPL, "called");
 }
 
 void InstallInterfaces(SM::ServiceManager& service_manager) {

src/core/hle/service/ssl/ssl.cpp

@@ -69,6 +69,7 @@ public:
 private:
     void SetOption(Kernel::HLERequestContext& ctx) {
         LOG_WARNING(Service_SSL, "(STUBBED) called");
+
         IPC::RequestParser rp{ctx};
 
         IPC::ResponseBuilder rb{ctx, 2};
@@ -114,6 +115,7 @@ private:
 
     void SetInterfaceVersion(Kernel::HLERequestContext& ctx) {
         LOG_DEBUG(Service_SSL, "called");
+
         IPC::RequestParser rp{ctx};
         ssl_version = rp.Pop<u32>();
 

src/core/hle/service/time/time.cpp

@@ -72,6 +72,7 @@ private:
                                        std::chrono::system_clock::now().time_since_epoch())
                                        .count()};
         LOG_DEBUG(Service_Time, "called");
+
         IPC::ResponseBuilder rb{ctx, 4};
         rb.Push(RESULT_SUCCESS);
         rb.Push<u64>(time_since_epoch);
@@ -79,6 +80,7 @@ private:
 
     void GetSystemClockContext(Kernel::HLERequestContext& ctx) {
         LOG_WARNING(Service_Time, "(STUBBED) called");
+
         SystemClockContext system_clock_ontext{};
         IPC::ResponseBuilder rb{ctx, (sizeof(SystemClockContext) / 4) + 2};
         rb.Push(RESULT_SUCCESS);
@@ -98,6 +100,7 @@ public:
 private:
     void GetCurrentTimePoint(Kernel::HLERequestContext& ctx) {
         LOG_DEBUG(Service_Time, "called");
+
         SteadyClockTimePoint steady_clock_time_point{
             CoreTiming::cyclesToMs(CoreTiming::GetTicks()) / 1000};
         IPC::ResponseBuilder rb{ctx, (sizeof(SteadyClockTimePoint) / 4) + 2};
@@ -130,6 +133,7 @@ private:
 
     void GetDeviceLocationName(Kernel::HLERequestContext& ctx) {
         LOG_DEBUG(Service_Time, "called");
+
         IPC::ResponseBuilder rb{ctx, (sizeof(LocationName) / 4) + 2};
         rb.Push(RESULT_SUCCESS);
         rb.PushRaw(location_name);
@@ -137,6 +141,7 @@ private:
 
     void GetTotalLocationNameCount(Kernel::HLERequestContext& ctx) {
         LOG_WARNING(Service_Time, "(STUBBED) called");
+
         IPC::ResponseBuilder rb{ctx, 3};
         rb.Push(RESULT_SUCCESS);
         rb.Push<u32>(0);
@@ -154,7 +159,6 @@ private:
     void ToCalendarTime(Kernel::HLERequestContext& ctx) {
         IPC::RequestParser rp{ctx};
         const u64 posix_time = rp.Pop<u64>();
-
         LOG_WARNING(Service_Time, "(STUBBED) called, posix_time=0x{:016X}", posix_time);
 
         TimeZoneRule time_zone_rule{};
@@ -175,7 +179,6 @@ private:
     void ToCalendarTimeWithMyRule(Kernel::HLERequestContext& ctx) {
         IPC::RequestParser rp{ctx};
         const u64 posix_time = rp.Pop<u64>();
-
         LOG_WARNING(Service_Time, "(STUBBED) called, posix_time=0x{:016X}", posix_time);
 
         CalendarTime calendar_time{2018, 1, 1, 0, 0, 0};
@@ -192,6 +195,7 @@ private:
     void ToPosixTime(Kernel::HLERequestContext& ctx) {
         // TODO(ogniK): Figure out how to handle multiple times
         LOG_WARNING(Service_Time, "(STUBBED) called");
+
         IPC::RequestParser rp{ctx};
         auto calendar_time = rp.PopRaw<CalendarTime>();
         auto posix_time = CalendarToPosix(calendar_time, {});
@@ -204,6 +208,7 @@ private:
 
     void ToPosixTimeWithMyRule(Kernel::HLERequestContext& ctx) {
         LOG_WARNING(Service_Time, "(STUBBED) called");
+
         IPC::RequestParser rp{ctx};
         auto calendar_time = rp.PopRaw<CalendarTime>();
         auto posix_time = CalendarToPosix(calendar_time, {});
@@ -216,38 +221,43 @@ private:
 };
 
 void Module::Interface::GetStandardUserSystemClock(Kernel::HLERequestContext& ctx) {
+    LOG_DEBUG(Service_Time, "called");
+
     IPC::ResponseBuilder rb{ctx, 2, 0, 1};
     rb.Push(RESULT_SUCCESS);
     rb.PushIpcInterface<ISystemClock>();
-    LOG_DEBUG(Service_Time, "called");
 }
 
 void Module::Interface::GetStandardNetworkSystemClock(Kernel::HLERequestContext& ctx) {
+    LOG_DEBUG(Service_Time, "called");
+
     IPC::ResponseBuilder rb{ctx, 2, 0, 1};
     rb.Push(RESULT_SUCCESS);
     rb.PushIpcInterface<ISystemClock>();
-    LOG_DEBUG(Service_Time, "called");
 }
 
 void Module::Interface::GetStandardSteadyClock(Kernel::HLERequestContext& ctx) {
+    LOG_DEBUG(Service_Time, "called");
+
     IPC::ResponseBuilder rb{ctx, 2, 0, 1};
     rb.Push(RESULT_SUCCESS);
     rb.PushIpcInterface<ISteadyClock>();
-    LOG_DEBUG(Service_Time, "called");
 }
 
 void Module::Interface::GetTimeZoneService(Kernel::HLERequestContext& ctx) {
+    LOG_DEBUG(Service_Time, "called");
+
     IPC::ResponseBuilder rb{ctx, 2, 0, 1};
     rb.Push(RESULT_SUCCESS);
     rb.PushIpcInterface<ITimeZoneService>();
-    LOG_DEBUG(Service_Time, "called");
 }
 
 void Module::Interface::GetStandardLocalSystemClock(Kernel::HLERequestContext& ctx) {
+    LOG_DEBUG(Service_Time, "called");
+
     IPC::ResponseBuilder rb{ctx, 2, 0, 1};
     rb.Push(RESULT_SUCCESS);
     rb.PushIpcInterface<ISystemClock>();
-    LOG_DEBUG(Service_Time, "called");
 }
 
 void Module::Interface::GetClockSnapshot(Kernel::HLERequestContext& ctx) {
@@ -265,6 +275,7 @@ void Module::Interface::GetClockSnapshot(Kernel::HLERequestContext& ctx) {
     const std::time_t time(time_since_epoch);
     const std::tm* tm = std::localtime(&time);
     if (tm == nullptr) {
+        LOG_ERROR(Service_Time, "tm is a nullptr");
         IPC::ResponseBuilder rb{ctx, 2};
         rb.Push(ResultCode(-1)); // TODO(ogniK): Find appropriate error code
         return;

src/core/hle/service/usb/usb.cpp

@@ -159,11 +159,11 @@ public:
 
 private:
     void GetPdSession(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_USB, "called");
+
         IPC::ResponseBuilder rb{ctx, 2, 0, 1};
         rb.Push(RESULT_SUCCESS);
         rb.PushIpcInterface<IPdSession>();
-
-        LOG_DEBUG(Service_USB, "called");
     }
 };
 

src/core/hle/service/vi/vi.cpp

@@ -504,10 +504,10 @@ private:
         u32 id = rp.Pop<u32>();
         auto transaction = static_cast<TransactionId>(rp.Pop<u32>());
         u32 flags = rp.Pop<u32>();
-        auto buffer_queue = nv_flinger->GetBufferQueue(id);
-
         LOG_DEBUG(Service_VI, "called, transaction={:X}", static_cast<u32>(transaction));
 
+        auto buffer_queue = nv_flinger->GetBufferQueue(id);
+
         if (transaction == TransactionId::Connect) {
             IGBPConnectRequestParcel request{ctx.ReadBuffer()};
             IGBPConnectResponseParcel response{
@@ -593,9 +593,9 @@ private:
         u32 id = rp.Pop<u32>();
         s32 addval = rp.PopRaw<s32>();
         u32 type = rp.Pop<u32>();
-
         LOG_WARNING(Service_VI, "(STUBBED) called id={}, addval={:08X}, type={:08X}", id, addval,
                     type);
+
         IPC::ResponseBuilder rb{ctx, 2};
         rb.Push(RESULT_SUCCESS);
     }
@@ -604,12 +604,11 @@ private:
         IPC::RequestParser rp{ctx};
         u32 id = rp.Pop<u32>();
         u32 unknown = rp.Pop<u32>();
+        LOG_WARNING(Service_VI, "(STUBBED) called id={}, unknown={:08X}", id, unknown);
 
         auto buffer_queue = nv_flinger->GetBufferQueue(id);
 
         // TODO(Subv): Find out what this actually is.
-
-        LOG_WARNING(Service_VI, "(STUBBED) called id={}, unknown={:08X}", id, unknown);
         IPC::ResponseBuilder rb{ctx, 2, 1};
         rb.Push(RESULT_SUCCESS);
         rb.PushCopyObjects(buffer_queue->GetBufferWaitEvent());
@@ -673,6 +672,7 @@ public:
 private:
     void SetLayerZ(Kernel::HLERequestContext& ctx) {
         LOG_WARNING(Service_VI, "(STUBBED) called");
+
         IPC::RequestParser rp{ctx};
         u64 layer_id = rp.Pop<u64>();
         u64 z_value = rp.Pop<u64>();
@@ -685,13 +685,16 @@ private:
         IPC::RequestParser rp{ctx};
         u64 layer_id = rp.Pop<u64>();
         bool visibility = rp.Pop<bool>();
-        IPC::ResponseBuilder rb{ctx, 2};
-        rb.Push(RESULT_SUCCESS);
         LOG_WARNING(Service_VI, "(STUBBED) called, layer_id=0x{:08X}, visibility={}", layer_id,
                     visibility);
+
+        IPC::ResponseBuilder rb{ctx, 2};
+        rb.Push(RESULT_SUCCESS);
     }
 
     void GetDisplayMode(Kernel::HLERequestContext& ctx) {
+        LOG_WARNING(Service_VI, "(STUBBED) called");
+
         IPC::ResponseBuilder rb{ctx, 6};
         rb.Push(RESULT_SUCCESS);
 
@@ -707,10 +710,8 @@ private:
                     static_cast<u32>(Settings::values.resolution_factor));
         }
 
-        rb.PushRaw<float>(60.0f);
+        rb.PushRaw<float>(60.0f); // This wouldn't seem to be correct for 30 fps games.
         rb.Push<u32>(0);
-
-        LOG_DEBUG(Service_VI, "called");
     }
 };
 
@@ -793,6 +794,7 @@ public:
 private:
     void CloseDisplay(Kernel::HLERequestContext& ctx) {
         LOG_WARNING(Service_VI, "(STUBBED) called");
+
         IPC::RequestParser rp{ctx};
         u64 display = rp.Pop<u64>();
 
@@ -802,6 +804,7 @@ private:
 
     void CreateManagedLayer(Kernel::HLERequestContext& ctx) {
         LOG_WARNING(Service_VI, "(STUBBED) called");
+
         IPC::RequestParser rp{ctx};
         u32 unknown = rp.Pop<u32>();
         rp.Skip(1, false);
@@ -817,6 +820,7 @@ private:
 
     void AddToLayerStack(Kernel::HLERequestContext& ctx) {
         LOG_WARNING(Service_VI, "(STUBBED) called");
+
         IPC::RequestParser rp{ctx};
         u32 stack = rp.Pop<u32>();
         u64 layer_id = rp.Pop<u64>();
@@ -829,10 +833,11 @@ private:
         IPC::RequestParser rp{ctx};
         u64 layer_id = rp.Pop<u64>();
         bool visibility = rp.Pop<bool>();
-        IPC::ResponseBuilder rb{ctx, 2};
-        rb.Push(RESULT_SUCCESS);
         LOG_WARNING(Service_VI, "(STUBBED) called, layer_id=0x{:X}, visibility={}", layer_id,
                     visibility);
+
+        IPC::ResponseBuilder rb{ctx, 2};
+        rb.Push(RESULT_SUCCESS);
     }
 
     std::shared_ptr<NVFlinger::NVFlinger> nv_flinger;
@@ -878,6 +883,7 @@ private:
 
     void OpenDisplay(Kernel::HLERequestContext& ctx) {
         LOG_WARNING(Service_VI, "(STUBBED) called");
+
         IPC::RequestParser rp{ctx};
         auto name_buf = rp.PopRaw<std::array<u8, 0x40>>();
         auto end = std::find(name_buf.begin(), name_buf.end(), '\0');
@@ -893,6 +899,7 @@ private:
 
     void CloseDisplay(Kernel::HLERequestContext& ctx) {
         LOG_WARNING(Service_VI, "(STUBBED) called");
+
         IPC::RequestParser rp{ctx};
         u64 display_id = rp.Pop<u64>();
 
@@ -902,6 +909,7 @@ private:
 
     void GetDisplayResolution(Kernel::HLERequestContext& ctx) {
         LOG_WARNING(Service_VI, "(STUBBED) called");
+
         IPC::RequestParser rp{ctx};
         u64 display_id = rp.Pop<u64>();
 
@@ -923,6 +931,7 @@ private:
 
     void SetLayerScalingMode(Kernel::HLERequestContext& ctx) {
         LOG_WARNING(Service_VI, "(STUBBED) called");
+
         IPC::RequestParser rp{ctx};
         u32 scaling_mode = rp.Pop<u32>();
         u64 unknown = rp.Pop<u64>();
@@ -932,6 +941,8 @@ private:
     }
 
     void ListDisplays(Kernel::HLERequestContext& ctx) {
+        LOG_WARNING(Service_VI, "(STUBBED) called");
+
         IPC::RequestParser rp{ctx};
         DisplayInfo display_info;
         display_info.width *= static_cast<u64>(Settings::values.resolution_factor);
@@ -940,11 +951,11 @@ private:
         IPC::ResponseBuilder rb{ctx, 4};
         rb.Push(RESULT_SUCCESS);
         rb.Push<u64>(1);
-        LOG_WARNING(Service_VI, "(STUBBED) called");
     }
 
     void OpenLayer(Kernel::HLERequestContext& ctx) {
         LOG_DEBUG(Service_VI, "called");
+
         IPC::RequestParser rp{ctx};
         auto name_buf = rp.PopRaw<std::array<u8, 0x40>>();
         auto end = std::find(name_buf.begin(), name_buf.end(), '\0');
@@ -995,6 +1006,7 @@ private:
 
     void GetDisplayVsyncEvent(Kernel::HLERequestContext& ctx) {
         LOG_WARNING(Service_VI, "(STUBBED) called");
+
         IPC::RequestParser rp{ctx};
         u64 display_id = rp.Pop<u64>();
 

src/video_core/CMakeLists.txt

@@ -1,6 +1,6 @@
 add_library(video_core STATIC
-    command_processor.cpp
-    command_processor.h
+    dma_pusher.cpp
+    dma_pusher.h
     debug_utils/debug_utils.cpp
     debug_utils/debug_utils.h
     engines/fermi_2d.cpp
@@ -21,6 +21,8 @@ add_library(video_core STATIC
     macro_interpreter.h
     memory_manager.cpp
     memory_manager.h
+    morton.cpp
+    morton.h
     rasterizer_cache.cpp
     rasterizer_cache.h
     rasterizer_interface.h
@@ -62,7 +64,6 @@ add_library(video_core STATIC
     textures/decoders.cpp
     textures/decoders.h
     textures/texture.h
-    utils.h
     video_core.cpp
     video_core.h
 )

src/video_core/command_processor.cpp

@@ -34,6 +34,9 @@ MICROPROFILE_DEFINE(ProcessCommandLists, "GPU", "Execute command buffer", MP_RGB
 void GPU::ProcessCommandLists(const std::vector<CommandListHeader>& commands) {
     MICROPROFILE_SCOPE(ProcessCommandLists);
 
+    // On entering GPU code, assume all memory may be touched by the ARM core.
+    maxwell_3d->dirty_flags.OnMemoryWrite();
+
     auto WriteReg = [this](u32 method, u32 subchannel, u32 value, u32 remaining_params) {
         LOG_TRACE(HW_GPU,
                   "Processing method {:08X} on subchannel {} value "

src/video_core/command_processor.h

@@ -1,53 +0,0 @@
-// Copyright 2018 yuzu Emulator Project
-// Licensed under GPLv2 or any later version
-// Refer to the license.txt file included.
-
-#pragma once
-
-#include <type_traits>
-#include "common/bit_field.h"
-#include "common/common_types.h"
-#include "video_core/memory_manager.h"
-
-namespace Tegra {
-
-enum class SubmissionMode : u32 {
-    IncreasingOld = 0,
-    Increasing = 1,
-    NonIncreasingOld = 2,
-    NonIncreasing = 3,
-    Inline = 4,
-    IncreaseOnce = 5
-};
-
-struct CommandListHeader {
-    u32 entry0; // gpu_va_lo
-    union {
-        u32 entry1; // gpu_va_hi | (unk_0x02 << 0x08) | (size << 0x0A) | (unk_0x01 << 0x1F)
-        BitField<0, 8, u32> gpu_va_hi;
-        BitField<8, 2, u32> unk1;
-        BitField<10, 21, u32> sz;
-        BitField<31, 1, u32> unk2;
-    };
-
-    GPUVAddr Address() const {
-        return (static_cast<GPUVAddr>(gpu_va_hi) << 32) | entry0;
-    }
-};
-static_assert(sizeof(CommandListHeader) == 8, "CommandListHeader is incorrect size");
-
-union CommandHeader {
-    u32 hex;
-
-    BitField<0, 13, u32> method;
-    BitField<13, 3, u32> subchannel;
-
-    BitField<16, 13, u32> arg_count;
-    BitField<16, 13, u32> inline_data;
-
-    BitField<29, 3, SubmissionMode> mode;
-};
-static_assert(std::is_standard_layout_v<CommandHeader>, "CommandHeader is not standard layout");
-static_assert(sizeof(CommandHeader) == sizeof(u32), "CommandHeader has incorrect size!");
-
-} // namespace Tegra

src/video_core/dma_pusher.cpp

@@ -0,0 +1,123 @@
+// Copyright 2018 yuzu Emulator Project
+// Licensed under GPLv2 or any later version
+// Refer to the license.txt file included.
+
+#include "common/microprofile.h"
+#include "core/core.h"
+#include "core/memory.h"
+#include "video_core/dma_pusher.h"
+#include "video_core/engines/maxwell_3d.h"
+#include "video_core/gpu.h"
+
+namespace Tegra {
+
+DmaPusher::DmaPusher(GPU& gpu) : gpu(gpu) {}
+
+DmaPusher::~DmaPusher() = default;
+
+MICROPROFILE_DEFINE(DispatchCalls, "GPU", "Execute command buffer", MP_RGB(128, 128, 192));
+
+void DmaPusher::DispatchCalls() {
+    MICROPROFILE_SCOPE(DispatchCalls);
+
+    // On entering GPU code, assume all memory may be touched by the ARM core.
+    gpu.Maxwell3D().dirty_flags.OnMemoryWrite();
+
+    dma_pushbuffer_subindex = 0;
+
+    while (Core::System::GetInstance().IsPoweredOn()) {
+        if (!Step()) {
+            break;
+        }
+    }
+}
+
+bool DmaPusher::Step() {
+    if (dma_get != dma_put) {
+        // Push buffer non-empty, read a word
+        const CommandHeader command_header{
+            Memory::Read32(*gpu.MemoryManager().GpuToCpuAddress(dma_get))};
+
+        dma_get += sizeof(u32);
+
+        if (!non_main) {
+            dma_mget = dma_get;
+        }
+
+        // now, see if we're in the middle of a command
+        if (dma_state.length_pending) {
+            // Second word of long non-inc methods command - method count
+            dma_state.length_pending = 0;
+            dma_state.method_count = command_header.method_count_;
+        } else if (dma_state.method_count) {
+            // Data word of methods command
+            CallMethod(command_header.argument);
+
+            if (!dma_state.non_incrementing) {
+                dma_state.method++;
+            }
+
+            if (dma_increment_once) {
+                dma_state.non_incrementing = true;
+            }
+
+            dma_state.method_count--;
+        } else {
+            // No command active - this is the first word of a new one
+            switch (command_header.mode) {
+            case SubmissionMode::Increasing:
+                SetState(command_header);
+                dma_state.non_incrementing = false;
+                dma_increment_once = false;
+                break;
+            case SubmissionMode::NonIncreasing:
+                SetState(command_header);
+                dma_state.non_incrementing = true;
+                dma_increment_once = false;
+                break;
+            case SubmissionMode::Inline:
+                dma_state.method = command_header.method;
+                dma_state.subchannel = command_header.subchannel;
+                CallMethod(command_header.arg_count);
+                dma_state.non_incrementing = true;
+                dma_increment_once = false;
+                break;
+            case SubmissionMode::IncreaseOnce:
+                SetState(command_header);
+                dma_state.non_incrementing = false;
+                dma_increment_once = true;
+                break;
+            }
+        }
+    } else if (ib_enable && !dma_pushbuffer.empty()) {
+        // Current pushbuffer empty, but we have more IB entries to read
+        const CommandList& command_list{dma_pushbuffer.front()};
+        const CommandListHeader& command_list_header{command_list[dma_pushbuffer_subindex++]};
+        dma_get = command_list_header.addr;
+        dma_put = dma_get + command_list_header.size * sizeof(u32);
+        non_main = command_list_header.is_non_main;
+
+        if (dma_pushbuffer_subindex >= command_list.size()) {
+            // We've gone through the current list, remove it from the queue
+            dma_pushbuffer.pop();
+            dma_pushbuffer_subindex = 0;
+        }
+    } else {
+        // Otherwise, pushbuffer empty and IB empty or nonexistent - nothing to do
+        return {};
+    }
+
+    return true;
+}
+
+void DmaPusher::SetState(const CommandHeader& command_header) {
+    dma_state.method = command_header.method;
+    dma_state.subchannel = command_header.subchannel;
+    dma_state.method_count = command_header.method_count;
+}
+
+void DmaPusher::CallMethod(u32 argument) const {
+    gpu.CallMethod({dma_state.method, argument, dma_state.subchannel, dma_state.method_count});
+}
+
+} // namespace Tegra

src/video_core/dma_pusher.h

@@ -0,0 +1,99 @@
+// Copyright 2018 yuzu Emulator Project
+// Licensed under GPLv2 or any later version
+// Refer to the license.txt file included.
+
+#pragma once
+
+#include <vector>
+#include <queue>
+
+#include "common/bit_field.h"
+#include "common/common_types.h"
+#include "video_core/memory_manager.h"
+
+namespace Tegra {
+
+enum class SubmissionMode : u32 {
+    IncreasingOld = 0,
+    Increasing = 1,
+    NonIncreasingOld = 2,
+    NonIncreasing = 3,
+    Inline = 4,
+    IncreaseOnce = 5
+};
+
+struct CommandListHeader {
+    union {
+        u64 raw;
+        BitField<0, 40, GPUVAddr> addr;
+        BitField<41, 1, u64> is_non_main;
+        BitField<42, 21, u64> size;
+    };
+};
+static_assert(sizeof(CommandListHeader) == sizeof(u64), "CommandListHeader is incorrect size");
+
+union CommandHeader {
+    u32 argument;
+    BitField<0, 13, u32> method;
+    BitField<0, 24, u32> method_count_;
+    BitField<13, 3, u32> subchannel;
+    BitField<16, 13, u32> arg_count;
+    BitField<16, 13, u32> method_count;
+    BitField<29, 3, SubmissionMode> mode;
+};
+static_assert(std::is_standard_layout_v<CommandHeader>, "CommandHeader is not standard layout");
+static_assert(sizeof(CommandHeader) == sizeof(u32), "CommandHeader has incorrect size!");
+
+class GPU;
+
+using CommandList = std::vector<Tegra::CommandListHeader>;
+
+/**
+ * The DmaPusher class implements DMA submission to FIFOs, providing an area of memory that the
+ * emulated app fills with commands and tells PFIFO to process. The pushbuffers are then assembled
+ * into a "command stream" consisting of 32-bit words that make up "commands".
+ * See https://envytools.readthedocs.io/en/latest/hw/fifo/dma-pusher.html#fifo-dma-pusher for
+ * details on this implementation.
+ */
+class DmaPusher {
+public:
+    explicit DmaPusher(GPU& gpu);
+    ~DmaPusher();
+
+    void Push(CommandList&& entries) {
+        dma_pushbuffer.push(std::move(entries));
+    }
+
+    void DispatchCalls();
+
+private:
+    bool Step();
+
+    void SetState(const CommandHeader& command_header);
+
+    void CallMethod(u32 argument) const;
+
+    GPU& gpu;
+
+    std::queue<CommandList> dma_pushbuffer; ///< Queue of command lists to be processed
+    std::size_t dma_pushbuffer_subindex{};  ///< Index within a command list within the pushbuffer
+
+    struct DmaState {
+        u32 method;            ///< Current method
+        u32 subchannel;        ///< Current subchannel
+        u32 method_count;      ///< Current method count
+        u32 length_pending;    ///< Large NI command length pending
+        bool non_incrementing; ///< Current command’s NI flag
+    };
+
+    DmaState dma_state{};
+    bool dma_increment_once{};
+
+    GPUVAddr dma_put{};   ///< pushbuffer current end address
+    GPUVAddr dma_get{};   ///< pushbuffer current read address
+    GPUVAddr dma_mget{};  ///< main pushbuffer last read address
+    bool ib_enable{true}; ///< IB mode enabled
+    bool non_main{};      ///< non-main pushbuffer active
+};
+
+} // namespace Tegra

src/video_core/engines/fermi_2d.cpp

@@ -2,8 +2,10 @@
 // Licensed under GPLv2 or any later version
 // Refer to the license.txt file included.
 
+#include "core/core.h"
 #include "core/memory.h"
 #include "video_core/engines/fermi_2d.h"
+#include "video_core/engines/maxwell_3d.h"
 #include "video_core/rasterizer_interface.h"
 #include "video_core/textures/decoders.h"
 
@@ -12,13 +14,13 @@ namespace Tegra::Engines {
 Fermi2D::Fermi2D(VideoCore::RasterizerInterface& rasterizer, MemoryManager& memory_manager)
     : memory_manager(memory_manager), rasterizer{rasterizer} {}
 
-void Fermi2D::WriteReg(u32 method, u32 value) {
-    ASSERT_MSG(method < Regs::NUM_REGS,
+void Fermi2D::CallMethod(const GPU::MethodCall& method_call) {
+    ASSERT_MSG(method_call.method < Regs::NUM_REGS,
                "Invalid Fermi2D register, increase the size of the Regs structure");
 
-    regs.reg_array[method] = value;
+    regs.reg_array[method_call.method] = method_call.argument;
 
-    switch (method) {
+    switch (method_call.method) {
     case FERMI2D_REG_INDEX(trigger): {
         HandleSurfaceCopy();
         break;
@@ -47,6 +49,9 @@ void Fermi2D::HandleSurfaceCopy() {
     u32 dst_bytes_per_pixel = RenderTargetBytesPerPixel(regs.dst.format);
 
     if (!rasterizer.AccelerateSurfaceCopy(regs.src, regs.dst)) {
+        // All copies here update the main memory, so mark all rasterizer states as invalid.
+        Core::System::GetInstance().GPU().Maxwell3D().dirty_flags.OnMemoryWrite();
+
         rasterizer.FlushRegion(source_cpu, src_bytes_per_pixel * regs.src.width * regs.src.height);
         // We have to invalidate the destination region to evict any outdated surfaces from the
         // cache. We do this before actually writing the new data because the destination address
@@ -68,13 +73,13 @@ void Fermi2D::HandleSurfaceCopy() {
             Texture::CopySwizzledData(regs.src.width, regs.src.height, regs.src.depth,
                                       src_bytes_per_pixel, dst_bytes_per_pixel, src_buffer,
                                       dst_buffer, true, regs.src.BlockHeight(),
-                                      regs.src.BlockDepth());
+                                      regs.src.BlockDepth(), 0);
         } else {
             // If the input is linear and the output is tiled, swizzle the input and copy it over.
             Texture::CopySwizzledData(regs.src.width, regs.src.height, regs.src.depth,
                                       src_bytes_per_pixel, dst_bytes_per_pixel, dst_buffer,
                                       src_buffer, false, regs.dst.BlockHeight(),
-                                      regs.dst.BlockDepth());
+                                      regs.dst.BlockDepth(), 0);
         }
     }
 }

src/video_core/engines/fermi_2d.h

@@ -27,7 +27,7 @@ public:
     ~Fermi2D() = default;
 
     /// Write the value to the register identified by method.
-    void WriteReg(u32 method, u32 value);
+    void CallMethod(const GPU::MethodCall& method_call);
 
     struct Regs {
         static constexpr std::size_t NUM_REGS = 0x258;

src/video_core/engines/kepler_memory.cpp

@@ -3,8 +3,10 @@
 // Refer to the license.txt file included.
 
 #include "common/logging/log.h"
+#include "core/core.h"
 #include "core/memory.h"
 #include "video_core/engines/kepler_memory.h"
+#include "video_core/engines/maxwell_3d.h"
 #include "video_core/rasterizer_interface.h"
 
 namespace Tegra::Engines {
@@ -15,19 +17,19 @@ KeplerMemory::KeplerMemory(VideoCore::RasterizerInterface& rasterizer,
 
 KeplerMemory::~KeplerMemory() = default;
 
-void KeplerMemory::WriteReg(u32 method, u32 value) {
-    ASSERT_MSG(method < Regs::NUM_REGS,
+void KeplerMemory::CallMethod(const GPU::MethodCall& method_call) {
+    ASSERT_MSG(method_call.method < Regs::NUM_REGS,
                "Invalid KeplerMemory register, increase the size of the Regs structure");
 
-    regs.reg_array[method] = value;
+    regs.reg_array[method_call.method] = method_call.argument;
 
-    switch (method) {
+    switch (method_call.method) {
     case KEPLERMEMORY_REG_INDEX(exec): {
         state.write_offset = 0;
         break;
     }
     case KEPLERMEMORY_REG_INDEX(data): {
-        ProcessData(value);
+        ProcessData(method_call.argument);
         break;
     }
     }
@@ -47,6 +49,7 @@ void KeplerMemory::ProcessData(u32 data) {
     rasterizer.InvalidateRegion(dest_address, sizeof(u32));
 
     Memory::Write32(dest_address, data);
+    Core::System::GetInstance().GPU().Maxwell3D().dirty_flags.OnMemoryWrite();
 
     state.write_offset++;
 }

src/video_core/engines/kepler_memory.h

@@ -9,6 +9,7 @@
 #include "common/bit_field.h"
 #include "common/common_funcs.h"
 #include "common/common_types.h"
+#include "video_core/gpu.h"
 #include "video_core/memory_manager.h"
 
 namespace VideoCore {
@@ -26,7 +27,7 @@ public:
     ~KeplerMemory();
 
     /// Write the value to the register identified by method.
-    void WriteReg(u32 method, u32 value);
+    void CallMethod(const GPU::MethodCall& method_call);
 
     struct Regs {
         static constexpr size_t NUM_REGS = 0x7F;

src/video_core/engines/maxwell_3d.cpp

@@ -97,57 +97,74 @@ void Maxwell3D::CallMacroMethod(u32 method, std::vector<u32> parameters) {
     macro_interpreter.Execute(search->second, std::move(parameters));
 }
 
-void Maxwell3D::WriteReg(u32 method, u32 value, u32 remaining_params) {
+void Maxwell3D::CallMethod(const GPU::MethodCall& method_call) {
     auto debug_context = Core::System::GetInstance().GetGPUDebugContext();
 
     // It is an error to write to a register other than the current macro's ARG register before it
     // has finished execution.
     if (executing_macro != 0) {
-        ASSERT(method == executing_macro + 1);
+        ASSERT(method_call.method == executing_macro + 1);
     }
 
     // Methods after 0xE00 are special, they're actually triggers for some microcode that was
     // uploaded to the GPU during initialization.
-    if (method >= MacroRegistersStart) {
+    if (method_call.method >= MacroRegistersStart) {
         // We're trying to execute a macro
         if (executing_macro == 0) {
             // A macro call must begin by writing the macro method's register, not its argument.
-            ASSERT_MSG((method % 2) == 0,
+            ASSERT_MSG((method_call.method % 2) == 0,
                        "Can't start macro execution by writing to the ARGS register");
-            executing_macro = method;
+            executing_macro = method_call.method;
         }
 
-        macro_params.push_back(value);
+        macro_params.push_back(method_call.argument);
 
         // Call the macro when there are no more parameters in the command buffer
-        if (remaining_params == 0) {
+        if (method_call.IsLastCall()) {
             CallMacroMethod(executing_macro, std::move(macro_params));
         }
         return;
     }
 
-    ASSERT_MSG(method < Regs::NUM_REGS,
+    ASSERT_MSG(method_call.method < Regs::NUM_REGS,
                "Invalid Maxwell3D register, increase the size of the Regs structure");
 
     if (debug_context) {
         debug_context->OnEvent(Tegra::DebugContext::Event::MaxwellCommandLoaded, nullptr);
     }
 
-    if (regs.reg_array[method] != value) {
-        regs.reg_array[method] = value;
-        if (method >= MAXWELL3D_REG_INDEX(vertex_attrib_format) &&
-            method < MAXWELL3D_REG_INDEX(vertex_attrib_format) + regs.vertex_attrib_format.size()) {
+    if (regs.reg_array[method_call.method] != method_call.argument) {
+        regs.reg_array[method_call.method] = method_call.argument;
+        // Vertex format
+        if (method_call.method >= MAXWELL3D_REG_INDEX(vertex_attrib_format) &&
+            method_call.method <
+                MAXWELL3D_REG_INDEX(vertex_attrib_format) + regs.vertex_attrib_format.size()) {
             dirty_flags.vertex_attrib_format = true;
         }
+
+        // Vertex buffer
+        if (method_call.method >= MAXWELL3D_REG_INDEX(vertex_array) &&
+            method_call.method < MAXWELL3D_REG_INDEX(vertex_array) + 4 * 32) {
+            dirty_flags.vertex_array |=
+                1u << ((method_call.method - MAXWELL3D_REG_INDEX(vertex_array)) >> 2);
+        } else if (method_call.method >= MAXWELL3D_REG_INDEX(vertex_array_limit) &&
+                   method_call.method < MAXWELL3D_REG_INDEX(vertex_array_limit) + 2 * 32) {
+            dirty_flags.vertex_array |=
+                1u << ((method_call.method - MAXWELL3D_REG_INDEX(vertex_array_limit)) >> 1);
+        } else if (method_call.method >= MAXWELL3D_REG_INDEX(instanced_arrays) &&
+                   method_call.method < MAXWELL3D_REG_INDEX(instanced_arrays) + 32) {
+            dirty_flags.vertex_array |=
+                1u << (method_call.method - MAXWELL3D_REG_INDEX(instanced_arrays));
+        }
     }
 
-    switch (method) {
+    switch (method_call.method) {
     case MAXWELL3D_REG_INDEX(macros.data): {
-        ProcessMacroUpload(value);
+        ProcessMacroUpload(method_call.argument);
         break;
     }
     case MAXWELL3D_REG_INDEX(macros.bind): {
-        ProcessMacroBind(value);
+        ProcessMacroBind(method_call.argument);
         break;
     }
     case MAXWELL3D_REG_INDEX(const_buffer.cb_data[0]):
@@ -166,7 +183,7 @@ void Maxwell3D::WriteReg(u32 method, u32 value, u32 remaining_params) {
     case MAXWELL3D_REG_INDEX(const_buffer.cb_data[13]):
     case MAXWELL3D_REG_INDEX(const_buffer.cb_data[14]):
     case MAXWELL3D_REG_INDEX(const_buffer.cb_data[15]): {
-        ProcessCBData(value);
+        ProcessCBData(method_call.argument);
         break;
     }
     case MAXWELL3D_REG_INDEX(cb_bind[0].raw_config): {
@@ -270,6 +287,7 @@ void Maxwell3D::ProcessQueryGet() {
             query_result.timestamp = CoreTiming::GetTicks();
             Memory::WriteBlock(*address, &query_result, sizeof(query_result));
         }
+        dirty_flags.OnMemoryWrite();
         break;
     }
     default:
@@ -346,6 +364,7 @@ void Maxwell3D::ProcessCBData(u32 value) {
         memory_manager.GpuToCpuAddress(buffer_address + regs.const_buffer.cb_pos);
 
     Memory::Write32(*address, value);
+    dirty_flags.OnMemoryWrite();
 
     // Increment the current buffer position.
     regs.const_buffer.cb_pos = regs.const_buffer.cb_pos + 4;

src/video_core/engines/maxwell_3d.h

@@ -590,10 +590,18 @@ public:
 
                 float clear_color[4];
                 float clear_depth;
+
                 INSERT_PADDING_WORDS(0x3);
+
                 s32 clear_stencil;
 
-                INSERT_PADDING_WORDS(0x17);
+                INSERT_PADDING_WORDS(0x7);
+
+                u32 polygon_offset_point_enable;
+                u32 polygon_offset_line_enable;
+                u32 polygon_offset_fill_enable;
+
+                INSERT_PADDING_WORDS(0xD);
 
                 std::array<ScissorTest, NumViewports> scissor_test;
 
@@ -728,6 +736,7 @@ public:
                 u32 frag_color_clamp;
 
                 union {
+                    BitField<0, 1, u32> y_negate;
                     BitField<4, 1, u32> triangle_rast_flip;
                 } screen_y_control;
 
@@ -735,7 +744,20 @@ public:
 
                 u32 vb_element_base;
 
-                INSERT_PADDING_WORDS(0x38);
+                INSERT_PADDING_WORDS(0x36);
+
+                union {
+                    BitField<0, 1, u32> c0;
+                    BitField<1, 1, u32> c1;
+                    BitField<2, 1, u32> c2;
+                    BitField<3, 1, u32> c3;
+                    BitField<4, 1, u32> c4;
+                    BitField<5, 1, u32> c5;
+                    BitField<6, 1, u32> c6;
+                    BitField<7, 1, u32> c7;
+                } clip_distance_enabled;
+
+                INSERT_PADDING_WORDS(0x1);
 
                 float point_size;
 
@@ -761,7 +783,11 @@ public:
                     }
                 } tsc;
 
-                INSERT_PADDING_WORDS(0x3);
+                INSERT_PADDING_WORDS(0x1);
+
+                float polygon_offset_factor;
+
+                INSERT_PADDING_WORDS(0x1);
 
                 struct {
                     u32 tic_address_high;
@@ -786,7 +812,9 @@ public:
 
                 u32 framebuffer_srgb;
 
-                INSERT_PADDING_WORDS(0x12);
+                float polygon_offset_units;
+
+                INSERT_PADDING_WORDS(0x11);
 
                 union {
                     BitField<2, 1, u32> coord_origin;
@@ -863,7 +891,9 @@ public:
 
                 INSERT_PADDING_WORDS(0x7);
 
-                INSERT_PADDING_WORDS(0x20);
+                INSERT_PADDING_WORDS(0x1F);
+
+                float polygon_offset_clamp;
 
                 struct {
                     u32 is_instanced[NumVertexArrays];
@@ -879,8 +909,21 @@ public:
 
                 Cull cull;
 
-                INSERT_PADDING_WORDS(0x28);
+                u32 pixel_center_integer;
 
+                INSERT_PADDING_WORDS(0x1);
+
+                u32 viewport_transform_enabled;
+
+                INSERT_PADDING_WORDS(0x3);
+
+                union {
+                    BitField<0, 1, u32> depth_range_0_1;
+                    BitField<3, 1, u32> depth_clamp_near;
+                    BitField<4, 1, u32> depth_clamp_far;
+                } view_volume_clip_control;
+
+                INSERT_PADDING_WORDS(0x21);
                 struct {
                     u32 enable;
                     LogicOperation operation;
@@ -1044,6 +1087,11 @@ public:
 
     struct DirtyFlags {
         bool vertex_attrib_format = true;
+        u32 vertex_array = 0xFFFFFFFF;
+
+        void OnMemoryWrite() {
+            vertex_array = 0xFFFFFFFF;
+        }
     };
 
     DirtyFlags dirty_flags;
@@ -1052,7 +1100,7 @@ public:
     u32 GetRegisterValue(u32 method) const;
 
     /// Write the value to the register identified by method.
-    void WriteReg(u32 method, u32 value, u32 remaining_params);
+    void CallMethod(const GPU::MethodCall& method_call);
 
     /// Returns a list of enabled textures for the specified shader stage.
     std::vector<Texture::FullTextureInfo> GetStageTextures(Regs::ShaderStage stage) const;
@@ -1136,6 +1184,9 @@ ASSERT_REG_POSITION(vertex_buffer, 0x35D);
 ASSERT_REG_POSITION(clear_color[0], 0x360);
 ASSERT_REG_POSITION(clear_depth, 0x364);
 ASSERT_REG_POSITION(clear_stencil, 0x368);
+ASSERT_REG_POSITION(polygon_offset_point_enable, 0x370);
+ASSERT_REG_POSITION(polygon_offset_line_enable, 0x371);
+ASSERT_REG_POSITION(polygon_offset_fill_enable, 0x372);
 ASSERT_REG_POSITION(scissor_test, 0x380);
 ASSERT_REG_POSITION(stencil_back_func_ref, 0x3D5);
 ASSERT_REG_POSITION(stencil_back_mask, 0x3D6);
@@ -1170,10 +1221,12 @@ ASSERT_REG_POSITION(stencil_front_mask, 0x4E7);
 ASSERT_REG_POSITION(frag_color_clamp, 0x4EA);
 ASSERT_REG_POSITION(screen_y_control, 0x4EB);
 ASSERT_REG_POSITION(vb_element_base, 0x50D);
+ASSERT_REG_POSITION(clip_distance_enabled, 0x544);
 ASSERT_REG_POSITION(point_size, 0x546);
 ASSERT_REG_POSITION(zeta_enable, 0x54E);
 ASSERT_REG_POSITION(multisample_control, 0x54F);
 ASSERT_REG_POSITION(tsc, 0x557);
+ASSERT_REG_POSITION(polygon_offset_factor, 0x55b);
 ASSERT_REG_POSITION(tic, 0x55D);
 ASSERT_REG_POSITION(stencil_two_side_enable, 0x565);
 ASSERT_REG_POSITION(stencil_back_op_fail, 0x566);
@@ -1181,13 +1234,18 @@ ASSERT_REG_POSITION(stencil_back_op_zfail, 0x567);
 ASSERT_REG_POSITION(stencil_back_op_zpass, 0x568);
 ASSERT_REG_POSITION(stencil_back_func_func, 0x569);
 ASSERT_REG_POSITION(framebuffer_srgb, 0x56E);
+ASSERT_REG_POSITION(polygon_offset_units, 0x56F);
 ASSERT_REG_POSITION(point_coord_replace, 0x581);
 ASSERT_REG_POSITION(code_address, 0x582);
 ASSERT_REG_POSITION(draw, 0x585);
 ASSERT_REG_POSITION(primitive_restart, 0x591);
 ASSERT_REG_POSITION(index_array, 0x5F2);
+ASSERT_REG_POSITION(polygon_offset_clamp, 0x61F);
 ASSERT_REG_POSITION(instanced_arrays, 0x620);
 ASSERT_REG_POSITION(cull, 0x646);
+ASSERT_REG_POSITION(pixel_center_integer, 0x649);
+ASSERT_REG_POSITION(viewport_transform_enabled, 0x64B);
+ASSERT_REG_POSITION(view_volume_clip_control, 0x64F);
 ASSERT_REG_POSITION(logic_op, 0x671);
 ASSERT_REG_POSITION(clear_buffers, 0x674);
 ASSERT_REG_POSITION(color_mask, 0x680);

src/video_core/engines/maxwell_compute.cpp

@@ -8,13 +8,13 @@
 
 namespace Tegra::Engines {
 
-void MaxwellCompute::WriteReg(u32 method, u32 value) {
-    ASSERT_MSG(method < Regs::NUM_REGS,
+void MaxwellCompute::CallMethod(const GPU::MethodCall& method_call) {
+    ASSERT_MSG(method_call.method < Regs::NUM_REGS,
                "Invalid MaxwellCompute register, increase the size of the Regs structure");
 
-    regs.reg_array[method] = value;
+    regs.reg_array[method_call.method] = method_call.argument;
 
-    switch (method) {
+    switch (method_call.method) {
     case MAXWELL_COMPUTE_REG_INDEX(compute): {
         LOG_CRITICAL(HW_GPU, "Compute shaders are not implemented");
         UNREACHABLE();

src/video_core/engines/maxwell_compute.h

@@ -9,6 +9,7 @@
 #include "common/bit_field.h"
 #include "common/common_funcs.h"
 #include "common/common_types.h"
+#include "video_core/gpu.h"
 
 namespace Tegra::Engines {
 
@@ -42,7 +43,7 @@ public:
                   "MaxwellCompute Regs has wrong size");
 
     /// Write the value to the register identified by method.
-    void WriteReg(u32 method, u32 value);
+    void CallMethod(const GPU::MethodCall& method_call);
 };
 
 #define ASSERT_REG_POSITION(field_name, position)                                                  \

src/video_core/engines/maxwell_dma.cpp

@@ -2,7 +2,9 @@
 // Licensed under GPLv2 or any later version
 // Refer to the license.txt file included.
 
+#include "core/core.h"
 #include "core/memory.h"
+#include "video_core/engines/maxwell_3d.h"
 #include "video_core/engines/maxwell_dma.h"
 #include "video_core/rasterizer_interface.h"
 #include "video_core/textures/decoders.h"
@@ -12,16 +14,16 @@ namespace Tegra::Engines {
 MaxwellDMA::MaxwellDMA(VideoCore::RasterizerInterface& rasterizer, MemoryManager& memory_manager)
     : memory_manager(memory_manager), rasterizer{rasterizer} {}
 
-void MaxwellDMA::WriteReg(u32 method, u32 value) {
-    ASSERT_MSG(method < Regs::NUM_REGS,
+void MaxwellDMA::CallMethod(const GPU::MethodCall& method_call) {
+    ASSERT_MSG(method_call.method < Regs::NUM_REGS,
                "Invalid MaxwellDMA register, increase the size of the Regs structure");
 
-    regs.reg_array[method] = value;
+    regs.reg_array[method_call.method] = method_call.argument;
 
 #define MAXWELLDMA_REG_INDEX(field_name)                                                           \
     (offsetof(Tegra::Engines::MaxwellDMA::Regs, field_name) / sizeof(u32))
 
-    switch (method) {
+    switch (method_call.method) {
     case MAXWELLDMA_REG_INDEX(exec): {
         HandleCopy();
         break;
@@ -54,6 +56,9 @@ void MaxwellDMA::HandleCopy() {
         return;
     }
 
+    // All copies here update the main memory, so mark all rasterizer states as invalid.
+    Core::System::GetInstance().GPU().Maxwell3D().dirty_flags.OnMemoryWrite();
+
     if (regs.exec.is_dst_linear && regs.exec.is_src_linear) {
         // When the enable_2d bit is disabled, the copy is performed as if we were copying a 1D
         // buffer of length `x_count`, otherwise we copy a 2D image of dimensions (x_count,

src/video_core/engines/maxwell_dma.h

@@ -24,7 +24,7 @@ public:
     ~MaxwellDMA() = default;
 
     /// Write the value to the register identified by method.
-    void WriteReg(u32 method, u32 value);
+    void CallMethod(const GPU::MethodCall& method_call);
 
     struct Regs {
         static constexpr std::size_t NUM_REGS = 0x1D6;

src/video_core/engines/shader_bytecode.h

@@ -82,6 +82,8 @@ union Attribute {
         Position = 7,
         Attribute_0 = 8,
         Attribute_31 = 39,
+        ClipDistances0123 = 44,
+        ClipDistances4567 = 45,
         PointCoord = 46,
         // This attribute contains a tuple of (~, ~, InstanceId, VertexId) when inside a vertex
         // shader, and a tuple of (TessCoord.x, TessCoord.y, TessCoord.z, ~) when inside a Tess Eval
@@ -366,6 +368,11 @@ enum class HalfPrecision : u64 {
     FMZ = 2,
 };
 
+enum class R2pMode : u64 {
+    Pr = 0,
+    Cc = 1,
+};
+
 enum class IpaInterpMode : u64 {
     Linear = 0,
     Perspective = 1,
@@ -854,6 +861,12 @@ union Instruction {
         BitField<39, 3, u64> pred39;
     } hsetp2;
 
+    union {
+        BitField<40, 1, R2pMode> mode;
+        BitField<41, 2, u64> byte;
+        BitField<20, 7, u64> immediate_mask;
+    } r2p;
+
     union {
         BitField<39, 3, u64> pred39;
         BitField<42, 1, u64> neg_pred;
@@ -1256,6 +1269,7 @@ public:
         BFE_C,
         BFE_R,
         BFE_IMM,
+        BFI_IMM_R,
         BRA,
         PBK,
         LD_A,
@@ -1381,6 +1395,7 @@ public:
         PSETP,
         PSET,
         CSETP,
+        R2P_IMM,
         XMAD_IMM,
         XMAD_CR,
         XMAD_RC,
@@ -1396,6 +1411,7 @@ public:
         ArithmeticHalf,
         ArithmeticHalfImmediate,
         Bfe,
+        Bfi,
         Shift,
         Ffma,
         Hfma2,
@@ -1410,6 +1426,7 @@ public:
         HalfSetPredicate,
         PredicateSetPredicate,
         PredicateSetRegister,
+        RegisterSetPredicate,
         Conversion,
         Xmad,
         Unknown,
@@ -1613,6 +1630,7 @@ private:
             INST("0100110000000---", Id::BFE_C, Type::Bfe, "BFE_C"),
             INST("0101110000000---", Id::BFE_R, Type::Bfe, "BFE_R"),
             INST("0011100-00000---", Id::BFE_IMM, Type::Bfe, "BFE_IMM"),
+            INST("0011011-11110---", Id::BFI_IMM_R, Type::Bfi, "BFI_IMM_R"),
             INST("0100110001000---", Id::LOP_C, Type::ArithmeticInteger, "LOP_C"),
             INST("0101110001000---", Id::LOP_R, Type::ArithmeticInteger, "LOP_R"),
             INST("0011100001000---", Id::LOP_IMM, Type::ArithmeticInteger, "LOP_IMM"),
@@ -1647,6 +1665,7 @@ private:
             INST("0101000010001---", Id::PSET, Type::PredicateSetRegister, "PSET"),
             INST("0101000010010---", Id::PSETP, Type::PredicateSetPredicate, "PSETP"),
             INST("010100001010----", Id::CSETP, Type::PredicateSetPredicate, "CSETP"),
+            INST("0011100-11110---", Id::R2P_IMM, Type::RegisterSetPredicate, "R2P_IMM"),
             INST("0011011-00------", Id::XMAD_IMM, Type::Xmad, "XMAD_IMM"),
             INST("0100111---------", Id::XMAD_CR, Type::Xmad, "XMAD_CR"),
             INST("010100010-------", Id::XMAD_RC, Type::Xmad, "XMAD_RC"),

src/video_core/engines/shader_header.h

@@ -62,7 +62,16 @@ struct Header {
             INSERT_PADDING_BYTES(1);  // ImapSystemValuesB
             INSERT_PADDING_BYTES(16); // ImapGenericVector[32]
             INSERT_PADDING_BYTES(2);  // ImapColor
-            INSERT_PADDING_BYTES(2);  // ImapSystemValuesC
+            union {
+                BitField<0, 8, u16> clip_distances;
+                BitField<8, 1, u16> point_sprite_s;
+                BitField<9, 1, u16> point_sprite_t;
+                BitField<10, 1, u16> fog_coordinate;
+                BitField<12, 1, u16> tessellation_eval_point_u;
+                BitField<13, 1, u16> tessellation_eval_point_v;
+                BitField<14, 1, u16> instance_id;
+                BitField<15, 1, u16> vertex_id;
+            };
             INSERT_PADDING_BYTES(5);  // ImapFixedFncTexture[10]
             INSERT_PADDING_BYTES(1);  // ImapReserved
             INSERT_PADDING_BYTES(3);  // OmapSystemValuesA

src/video_core/gpu.cpp

@@ -26,6 +26,7 @@ u32 FramebufferConfig::BytesPerPixel(PixelFormat format) {
 
 GPU::GPU(VideoCore::RasterizerInterface& rasterizer) {
     memory_manager = std::make_unique<Tegra::MemoryManager>();
+    dma_pusher = std::make_unique<Tegra::DmaPusher>(*this);
     maxwell_3d = std::make_unique<Engines::Maxwell3D>(rasterizer, *memory_manager);
     fermi_2d = std::make_unique<Engines::Fermi2D>(rasterizer, *memory_manager);
     maxwell_compute = std::make_unique<Engines::MaxwellCompute>();
@@ -51,6 +52,14 @@ const MemoryManager& GPU::MemoryManager() const {
     return *memory_manager;
 }
 
+DmaPusher& GPU::DmaPusher() {
+    return *dma_pusher;
+}
+
+const DmaPusher& GPU::DmaPusher() const {
+    return *dma_pusher;
+}
+
 u32 RenderTargetBytesPerPixel(RenderTargetFormat format) {
     ASSERT(format != RenderTargetFormat::NONE);
 
@@ -113,4 +122,48 @@ u32 DepthFormatBytesPerPixel(DepthFormat format) {
     }
 }
 
+enum class BufferMethods {
+    BindObject = 0,
+    CountBufferMethods = 0x40,
+};
+
+void GPU::CallMethod(const MethodCall& method_call) {
+    LOG_TRACE(HW_GPU,
+              "Processing method {:08X} on subchannel {} value "
+              "{:08X} remaining params {}",
+              MethCall.method, MethCall.subchannel, value, remaining_params);
+
+    ASSERT(method_call.subchannel < bound_engines.size());
+
+    if (method_call.method == static_cast<u32>(BufferMethods::BindObject)) {
+        // Bind the current subchannel to the desired engine id.
+        LOG_DEBUG(HW_GPU, "Binding subchannel {} to engine {}", method_call.subchannel,
+                  method_call.argument);
+        bound_engines[method_call.subchannel] = static_cast<EngineID>(method_call.argument);
+        return;
+    }
+
+    const EngineID engine = bound_engines[method_call.subchannel];
+
+    switch (engine) {
+    case EngineID::FERMI_TWOD_A:
+        fermi_2d->CallMethod(method_call);
+        break;
+    case EngineID::MAXWELL_B:
+        maxwell_3d->CallMethod(method_call);
+        break;
+    case EngineID::MAXWELL_COMPUTE_B:
+        maxwell_compute->CallMethod(method_call);
+        break;
+    case EngineID::MAXWELL_DMA_COPY_A:
+        maxwell_dma->CallMethod(method_call);
+        break;
+    case EngineID::KEPLER_INLINE_TO_MEMORY_B:
+        kepler_memory->CallMethod(method_call);
+        break;
+    default:
+        UNIMPLEMENTED_MSG("Unimplemented engine");
+    }
+}
+
 } // namespace Tegra

src/video_core/gpu.h

@@ -9,6 +9,7 @@
 #include <vector>
 #include "common/common_types.h"
 #include "core/hle/service/nvflinger/buffer_queue.h"
+#include "video_core/dma_pusher.h"
 #include "video_core/memory_manager.h"
 
 namespace VideoCore {
@@ -119,8 +120,23 @@ public:
     explicit GPU(VideoCore::RasterizerInterface& rasterizer);
     ~GPU();
 
-    /// Processes a command list stored at the specified address in GPU memory.
-    void ProcessCommandLists(const std::vector<CommandListHeader>& commands);
+    struct MethodCall {
+        u32 method{};
+        u32 argument{};
+        u32 subchannel{};
+        u32 method_count{};
+
+        bool IsLastCall() const {
+            return method_count <= 1;
+        }
+
+        MethodCall(u32 method, u32 argument, u32 subchannel = 0, u32 method_count = 0)
+            : method(method), argument(argument), subchannel(subchannel),
+              method_count(method_count) {}
+    };
+
+    /// Calls a GPU method.
+    void CallMethod(const MethodCall& method_call);
 
     /// Returns a reference to the Maxwell3D GPU engine.
     Engines::Maxwell3D& Maxwell3D();
@@ -134,7 +150,14 @@ public:
     /// Returns a const reference to the GPU memory manager.
     const Tegra::MemoryManager& MemoryManager() const;
 
+    /// Returns a reference to the GPU DMA pusher.
+    Tegra::DmaPusher& DmaPusher();
+
+    /// Returns a const reference to the GPU DMA pusher.
+    const Tegra::DmaPusher& DmaPusher() const;
+
 private:
+    std::unique_ptr<Tegra::DmaPusher> dma_pusher;
     std::unique_ptr<Tegra::MemoryManager> memory_manager;
 
     /// Mapping of command subchannels to their bound engine ids.

src/video_core/macro_interpreter.cpp

@@ -250,7 +250,7 @@ void MacroInterpreter::SetMethodAddress(u32 address) {
 }
 
 void MacroInterpreter::Send(u32 value) {
-    maxwell3d.WriteReg(method_address.address, value, 0);
+    maxwell3d.CallMethod({method_address.address, value});
     // Increment the method address by the method increment.
     method_address.address.Assign(method_address.address.Value() +
                                   method_address.increment.Value());

src/video_core/memory_manager.cpp

@@ -9,6 +9,13 @@
 
 namespace Tegra {
 
+MemoryManager::MemoryManager() {
+    // Mark the first page as reserved, so that 0 is not a valid GPUVAddr. Otherwise, games might
+    // try to use 0 as a valid address, which is also used to mean nullptr. This fixes a bug with
+    // Undertale using 0 for a render target.
+    PageSlot(0) = static_cast<u64>(PageStatus::Reserved);
+}
+
 GPUVAddr MemoryManager::AllocateSpace(u64 size, u64 align) {
     const std::optional<GPUVAddr> gpu_addr{FindFreeBlock(0, size, align, PageStatus::Unmapped)};
 

src/video_core/memory_manager.h

@@ -18,7 +18,7 @@ using GPUVAddr = u64;
 
 class MemoryManager final {
 public:
-    MemoryManager() = default;
+    MemoryManager();
 
     GPUVAddr AllocateSpace(u64 size, u64 align);
     GPUVAddr AllocateSpace(GPUVAddr gpu_addr, u64 size, u64 align);
@@ -37,6 +37,7 @@ private:
     enum class PageStatus : u64 {
         Unmapped = 0xFFFFFFFFFFFFFFFFULL,
         Allocated = 0xFFFFFFFFFFFFFFFEULL,
+        Reserved = 0xFFFFFFFFFFFFFFFDULL,
     };
 
     std::optional<GPUVAddr> FindFreeBlock(GPUVAddr region_start, u64 size, u64 align,

src/video_core/morton.cpp

@@ -0,0 +1,355 @@
+// Copyright 2018 yuzu Emulator Project
+// Licensed under GPLv2 or any later version
+// Refer to the license.txt file included.
+
+#include <array>
+#include <cstring>
+#include "common/assert.h"
+#include "common/common_types.h"
+#include "core/memory.h"
+#include "video_core/morton.h"
+#include "video_core/surface.h"
+#include "video_core/textures/decoders.h"
+
+namespace VideoCore {
+
+using Surface::GetBytesPerPixel;
+using Surface::PixelFormat;
+
+using MortonCopyFn = void (*)(u32, u32, u32, u32, u32, u32, u8*, std::size_t, VAddr);
+using ConversionArray = std::array<MortonCopyFn, Surface::MaxPixelFormat>;
+
+template <bool morton_to_linear, PixelFormat format>
+static void MortonCopy(u32 stride, u32 block_height, u32 height, u32 block_depth, u32 depth,
+                       u32 tile_width_spacing, u8* buffer, std::size_t buffer_size, VAddr addr) {
+    constexpr u32 bytes_per_pixel = GetBytesPerPixel(format);
+
+    // With the BCn formats (DXT and DXN), each 4x4 tile is swizzled instead of just individual
+    // pixel values.
+    const u32 tile_size_x{GetDefaultBlockWidth(format)};
+    const u32 tile_size_y{GetDefaultBlockHeight(format)};
+
+    if constexpr (morton_to_linear) {
+        Tegra::Texture::UnswizzleTexture(buffer, addr, tile_size_x, tile_size_y, bytes_per_pixel,
+                                         stride, height, depth, block_height, block_depth,
+                                         tile_width_spacing);
+    } else {
+        Tegra::Texture::CopySwizzledData(
+            (stride + tile_size_x - 1) / tile_size_x, (height + tile_size_y - 1) / tile_size_y,
+            depth, bytes_per_pixel, bytes_per_pixel, Memory::GetPointer(addr), buffer, false,
+            block_height, block_depth, tile_width_spacing);
+    }
+}
+
+static constexpr ConversionArray morton_to_linear_fns = {
+    // clang-format off
+        MortonCopy<true, PixelFormat::ABGR8U>,
+        MortonCopy<true, PixelFormat::ABGR8S>,
+        MortonCopy<true, PixelFormat::ABGR8UI>,
+        MortonCopy<true, PixelFormat::B5G6R5U>,
+        MortonCopy<true, PixelFormat::A2B10G10R10U>,
+        MortonCopy<true, PixelFormat::A1B5G5R5U>,
+        MortonCopy<true, PixelFormat::R8U>,
+        MortonCopy<true, PixelFormat::R8UI>,
+        MortonCopy<true, PixelFormat::RGBA16F>,
+        MortonCopy<true, PixelFormat::RGBA16U>,
+        MortonCopy<true, PixelFormat::RGBA16UI>,
+        MortonCopy<true, PixelFormat::R11FG11FB10F>,
+        MortonCopy<true, PixelFormat::RGBA32UI>,
+        MortonCopy<true, PixelFormat::DXT1>,
+        MortonCopy<true, PixelFormat::DXT23>,
+        MortonCopy<true, PixelFormat::DXT45>,
+        MortonCopy<true, PixelFormat::DXN1>,
+        MortonCopy<true, PixelFormat::DXN2UNORM>,
+        MortonCopy<true, PixelFormat::DXN2SNORM>,
+        MortonCopy<true, PixelFormat::BC7U>,
+        MortonCopy<true, PixelFormat::BC6H_UF16>,
+        MortonCopy<true, PixelFormat::BC6H_SF16>,
+        MortonCopy<true, PixelFormat::ASTC_2D_4X4>,
+        MortonCopy<true, PixelFormat::G8R8U>,
+        MortonCopy<true, PixelFormat::G8R8S>,
+        MortonCopy<true, PixelFormat::BGRA8>,
+        MortonCopy<true, PixelFormat::RGBA32F>,
+        MortonCopy<true, PixelFormat::RG32F>,
+        MortonCopy<true, PixelFormat::R32F>,
+        MortonCopy<true, PixelFormat::R16F>,
+        MortonCopy<true, PixelFormat::R16U>,
+        MortonCopy<true, PixelFormat::R16S>,
+        MortonCopy<true, PixelFormat::R16UI>,
+        MortonCopy<true, PixelFormat::R16I>,
+        MortonCopy<true, PixelFormat::RG16>,
+        MortonCopy<true, PixelFormat::RG16F>,
+        MortonCopy<true, PixelFormat::RG16UI>,
+        MortonCopy<true, PixelFormat::RG16I>,
+        MortonCopy<true, PixelFormat::RG16S>,
+        MortonCopy<true, PixelFormat::RGB32F>,
+        MortonCopy<true, PixelFormat::RGBA8_SRGB>,
+        MortonCopy<true, PixelFormat::RG8U>,
+        MortonCopy<true, PixelFormat::RG8S>,
+        MortonCopy<true, PixelFormat::RG32UI>,
+        MortonCopy<true, PixelFormat::R32UI>,
+        MortonCopy<true, PixelFormat::ASTC_2D_8X8>,
+        MortonCopy<true, PixelFormat::ASTC_2D_8X5>,
+        MortonCopy<true, PixelFormat::ASTC_2D_5X4>,
+        MortonCopy<true, PixelFormat::BGRA8_SRGB>,
+        MortonCopy<true, PixelFormat::DXT1_SRGB>,
+        MortonCopy<true, PixelFormat::DXT23_SRGB>,
+        MortonCopy<true, PixelFormat::DXT45_SRGB>,
+        MortonCopy<true, PixelFormat::BC7U_SRGB>,
+        MortonCopy<true, PixelFormat::ASTC_2D_4X4_SRGB>,
+        MortonCopy<true, PixelFormat::ASTC_2D_8X8_SRGB>,
+        MortonCopy<true, PixelFormat::ASTC_2D_8X5_SRGB>,
+        MortonCopy<true, PixelFormat::ASTC_2D_5X4_SRGB>,
+        MortonCopy<true, PixelFormat::ASTC_2D_5X5>,
+        MortonCopy<true, PixelFormat::ASTC_2D_5X5_SRGB>,
+        MortonCopy<true, PixelFormat::ASTC_2D_10X8>,
+        MortonCopy<true, PixelFormat::ASTC_2D_10X8_SRGB>,
+        MortonCopy<true, PixelFormat::Z32F>,
+        MortonCopy<true, PixelFormat::Z16>,
+        MortonCopy<true, PixelFormat::Z24S8>,
+        MortonCopy<true, PixelFormat::S8Z24>,
+        MortonCopy<true, PixelFormat::Z32FS8>,
+    // clang-format on
+};
+
+static constexpr ConversionArray linear_to_morton_fns = {
+    // clang-format off
+        MortonCopy<false, PixelFormat::ABGR8U>,
+        MortonCopy<false, PixelFormat::ABGR8S>,
+        MortonCopy<false, PixelFormat::ABGR8UI>,
+        MortonCopy<false, PixelFormat::B5G6R5U>,
+        MortonCopy<false, PixelFormat::A2B10G10R10U>,
+        MortonCopy<false, PixelFormat::A1B5G5R5U>,
+        MortonCopy<false, PixelFormat::R8U>,
+        MortonCopy<false, PixelFormat::R8UI>,
+        MortonCopy<false, PixelFormat::RGBA16F>,
+        MortonCopy<false, PixelFormat::RGBA16U>,
+        MortonCopy<false, PixelFormat::RGBA16UI>,
+        MortonCopy<false, PixelFormat::R11FG11FB10F>,
+        MortonCopy<false, PixelFormat::RGBA32UI>,
+        MortonCopy<false, PixelFormat::DXT1>,
+        MortonCopy<false, PixelFormat::DXT23>,
+        MortonCopy<false, PixelFormat::DXT45>,
+        MortonCopy<false, PixelFormat::DXN1>,
+        MortonCopy<false, PixelFormat::DXN2UNORM>,
+        MortonCopy<false, PixelFormat::DXN2SNORM>,
+        MortonCopy<false, PixelFormat::BC7U>,
+        MortonCopy<false, PixelFormat::BC6H_UF16>,
+        MortonCopy<false, PixelFormat::BC6H_SF16>,
+        // TODO(Subv): Swizzling ASTC formats are not supported
+        nullptr,
+        MortonCopy<false, PixelFormat::G8R8U>,
+        MortonCopy<false, PixelFormat::G8R8S>,
+        MortonCopy<false, PixelFormat::BGRA8>,
+        MortonCopy<false, PixelFormat::RGBA32F>,
+        MortonCopy<false, PixelFormat::RG32F>,
+        MortonCopy<false, PixelFormat::R32F>,
+        MortonCopy<false, PixelFormat::R16F>,
+        MortonCopy<false, PixelFormat::R16U>,
+        MortonCopy<false, PixelFormat::R16S>,
+        MortonCopy<false, PixelFormat::R16UI>,
+        MortonCopy<false, PixelFormat::R16I>,
+        MortonCopy<false, PixelFormat::RG16>,
+        MortonCopy<false, PixelFormat::RG16F>,
+        MortonCopy<false, PixelFormat::RG16UI>,
+        MortonCopy<false, PixelFormat::RG16I>,
+        MortonCopy<false, PixelFormat::RG16S>,
+        MortonCopy<false, PixelFormat::RGB32F>,
+        MortonCopy<false, PixelFormat::RGBA8_SRGB>,
+        MortonCopy<false, PixelFormat::RG8U>,
+        MortonCopy<false, PixelFormat::RG8S>,
+        MortonCopy<false, PixelFormat::RG32UI>,
+        MortonCopy<false, PixelFormat::R32UI>,
+        nullptr,
+        nullptr,
+        nullptr,
+        MortonCopy<false, PixelFormat::BGRA8_SRGB>,
+        MortonCopy<false, PixelFormat::DXT1_SRGB>,
+        MortonCopy<false, PixelFormat::DXT23_SRGB>,
+        MortonCopy<false, PixelFormat::DXT45_SRGB>,
+        MortonCopy<false, PixelFormat::BC7U_SRGB>,
+        nullptr,
+        nullptr,
+        nullptr,
+        nullptr,
+        nullptr,
+        nullptr,
+        nullptr,
+        nullptr,
+        MortonCopy<false, PixelFormat::Z32F>,
+        MortonCopy<false, PixelFormat::Z16>,
+        MortonCopy<false, PixelFormat::Z24S8>,
+        MortonCopy<false, PixelFormat::S8Z24>,
+        MortonCopy<false, PixelFormat::Z32FS8>,
+    // clang-format on
+};
+
+static MortonCopyFn GetSwizzleFunction(MortonSwizzleMode mode, Surface::PixelFormat format) {
+    switch (mode) {
+    case MortonSwizzleMode::MortonToLinear:
+        return morton_to_linear_fns[static_cast<std::size_t>(format)];
+    case MortonSwizzleMode::LinearToMorton:
+        return linear_to_morton_fns[static_cast<std::size_t>(format)];
+    }
+    UNREACHABLE();
+}
+
+/// 8x8 Z-Order coordinate from 2D coordinates
+static u32 MortonInterleave(u32 x, u32 y) {
+    static const u32 xlut[] = {0x00, 0x01, 0x04, 0x05, 0x10, 0x11, 0x14, 0x15};
+    static const u32 ylut[] = {0x00, 0x02, 0x08, 0x0a, 0x20, 0x22, 0x28, 0x2a};
+    return xlut[x % 8] + ylut[y % 8];
+}
+
+/// Calculates the offset of the position of the pixel in Morton order
+static u32 GetMortonOffset(u32 x, u32 y, u32 bytes_per_pixel) {
+    // Images are split into 8x8 tiles. Each tile is composed of four 4x4 subtiles each
+    // of which is composed of four 2x2 subtiles each of which is composed of four texels.
+    // Each structure is embedded into the next-bigger one in a diagonal pattern, e.g.
+    // texels are laid out in a 2x2 subtile like this:
+    // 2 3
+    // 0 1
+    //
+    // The full 8x8 tile has the texels arranged like this:
+    //
+    // 42 43 46 47 58 59 62 63
+    // 40 41 44 45 56 57 60 61
+    // 34 35 38 39 50 51 54 55
+    // 32 33 36 37 48 49 52 53
+    // 10 11 14 15 26 27 30 31
+    // 08 09 12 13 24 25 28 29
+    // 02 03 06 07 18 19 22 23
+    // 00 01 04 05 16 17 20 21
+    //
+    // This pattern is what's called Z-order curve, or Morton order.
+
+    const unsigned int block_height = 8;
+    const unsigned int coarse_x = x & ~7;
+
+    u32 i = MortonInterleave(x, y);
+
+    const unsigned int offset = coarse_x * block_height;
+
+    return (i + offset) * bytes_per_pixel;
+}
+
+static u32 MortonInterleave128(u32 x, u32 y) {
+    // 128x128 Z-Order coordinate from 2D coordinates
+    static constexpr u32 xlut[] = {
+        0x0000, 0x0001, 0x0002, 0x0003, 0x0008, 0x0009, 0x000a, 0x000b, 0x0040, 0x0041, 0x0042,
+        0x0043, 0x0048, 0x0049, 0x004a, 0x004b, 0x0800, 0x0801, 0x0802, 0x0803, 0x0808, 0x0809,
+        0x080a, 0x080b, 0x0840, 0x0841, 0x0842, 0x0843, 0x0848, 0x0849, 0x084a, 0x084b, 0x1000,
+        0x1001, 0x1002, 0x1003, 0x1008, 0x1009, 0x100a, 0x100b, 0x1040, 0x1041, 0x1042, 0x1043,
+        0x1048, 0x1049, 0x104a, 0x104b, 0x1800, 0x1801, 0x1802, 0x1803, 0x1808, 0x1809, 0x180a,
+        0x180b, 0x1840, 0x1841, 0x1842, 0x1843, 0x1848, 0x1849, 0x184a, 0x184b, 0x2000, 0x2001,
+        0x2002, 0x2003, 0x2008, 0x2009, 0x200a, 0x200b, 0x2040, 0x2041, 0x2042, 0x2043, 0x2048,
+        0x2049, 0x204a, 0x204b, 0x2800, 0x2801, 0x2802, 0x2803, 0x2808, 0x2809, 0x280a, 0x280b,
+        0x2840, 0x2841, 0x2842, 0x2843, 0x2848, 0x2849, 0x284a, 0x284b, 0x3000, 0x3001, 0x3002,
+        0x3003, 0x3008, 0x3009, 0x300a, 0x300b, 0x3040, 0x3041, 0x3042, 0x3043, 0x3048, 0x3049,
+        0x304a, 0x304b, 0x3800, 0x3801, 0x3802, 0x3803, 0x3808, 0x3809, 0x380a, 0x380b, 0x3840,
+        0x3841, 0x3842, 0x3843, 0x3848, 0x3849, 0x384a, 0x384b, 0x0000, 0x0001, 0x0002, 0x0003,
+        0x0008, 0x0009, 0x000a, 0x000b, 0x0040, 0x0041, 0x0042, 0x0043, 0x0048, 0x0049, 0x004a,
+        0x004b, 0x0800, 0x0801, 0x0802, 0x0803, 0x0808, 0x0809, 0x080a, 0x080b, 0x0840, 0x0841,
+        0x0842, 0x0843, 0x0848, 0x0849, 0x084a, 0x084b, 0x1000, 0x1001, 0x1002, 0x1003, 0x1008,
+        0x1009, 0x100a, 0x100b, 0x1040, 0x1041, 0x1042, 0x1043, 0x1048, 0x1049, 0x104a, 0x104b,
+        0x1800, 0x1801, 0x1802, 0x1803, 0x1808, 0x1809, 0x180a, 0x180b, 0x1840, 0x1841, 0x1842,
+        0x1843, 0x1848, 0x1849, 0x184a, 0x184b, 0x2000, 0x2001, 0x2002, 0x2003, 0x2008, 0x2009,
+        0x200a, 0x200b, 0x2040, 0x2041, 0x2042, 0x2043, 0x2048, 0x2049, 0x204a, 0x204b, 0x2800,
+        0x2801, 0x2802, 0x2803, 0x2808, 0x2809, 0x280a, 0x280b, 0x2840, 0x2841, 0x2842, 0x2843,
+        0x2848, 0x2849, 0x284a, 0x284b, 0x3000, 0x3001, 0x3002, 0x3003, 0x3008, 0x3009, 0x300a,
+        0x300b, 0x3040, 0x3041, 0x3042, 0x3043, 0x3048, 0x3049, 0x304a, 0x304b, 0x3800, 0x3801,
+        0x3802, 0x3803, 0x3808, 0x3809, 0x380a, 0x380b, 0x3840, 0x3841, 0x3842, 0x3843, 0x3848,
+        0x3849, 0x384a, 0x384b, 0x0000, 0x0001, 0x0002, 0x0003, 0x0008, 0x0009, 0x000a, 0x000b,
+        0x0040, 0x0041, 0x0042, 0x0043, 0x0048, 0x0049, 0x004a, 0x004b, 0x0800, 0x0801, 0x0802,
+        0x0803, 0x0808, 0x0809, 0x080a, 0x080b, 0x0840, 0x0841, 0x0842, 0x0843, 0x0848, 0x0849,
+        0x084a, 0x084b, 0x1000, 0x1001, 0x1002, 0x1003, 0x1008, 0x1009, 0x100a, 0x100b, 0x1040,
+        0x1041, 0x1042, 0x1043, 0x1048, 0x1049, 0x104a, 0x104b, 0x1800, 0x1801, 0x1802, 0x1803,
+        0x1808, 0x1809, 0x180a, 0x180b, 0x1840, 0x1841, 0x1842, 0x1843, 0x1848, 0x1849, 0x184a,
+        0x184b, 0x2000, 0x2001, 0x2002, 0x2003, 0x2008, 0x2009, 0x200a, 0x200b, 0x2040, 0x2041,
+        0x2042, 0x2043, 0x2048, 0x2049, 0x204a, 0x204b, 0x2800, 0x2801, 0x2802, 0x2803, 0x2808,
+        0x2809, 0x280a, 0x280b, 0x2840, 0x2841, 0x2842, 0x2843, 0x2848, 0x2849, 0x284a, 0x284b,
+        0x3000, 0x3001, 0x3002, 0x3003, 0x3008, 0x3009, 0x300a, 0x300b, 0x3040, 0x3041, 0x3042,
+        0x3043, 0x3048, 0x3049, 0x304a, 0x304b, 0x3800, 0x3801, 0x3802, 0x3803, 0x3808, 0x3809,
+        0x380a, 0x380b, 0x3840, 0x3841, 0x3842, 0x3843, 0x3848, 0x3849, 0x384a, 0x384b,
+    };
+    static constexpr u32 ylut[] = {
+        0x0000, 0x0004, 0x0010, 0x0014, 0x0020, 0x0024, 0x0030, 0x0034, 0x0080, 0x0084, 0x0090,
+        0x0094, 0x00a0, 0x00a4, 0x00b0, 0x00b4, 0x0100, 0x0104, 0x0110, 0x0114, 0x0120, 0x0124,
+        0x0130, 0x0134, 0x0180, 0x0184, 0x0190, 0x0194, 0x01a0, 0x01a4, 0x01b0, 0x01b4, 0x0200,
+        0x0204, 0x0210, 0x0214, 0x0220, 0x0224, 0x0230, 0x0234, 0x0280, 0x0284, 0x0290, 0x0294,
+        0x02a0, 0x02a4, 0x02b0, 0x02b4, 0x0300, 0x0304, 0x0310, 0x0314, 0x0320, 0x0324, 0x0330,
+        0x0334, 0x0380, 0x0384, 0x0390, 0x0394, 0x03a0, 0x03a4, 0x03b0, 0x03b4, 0x0400, 0x0404,
+        0x0410, 0x0414, 0x0420, 0x0424, 0x0430, 0x0434, 0x0480, 0x0484, 0x0490, 0x0494, 0x04a0,
+        0x04a4, 0x04b0, 0x04b4, 0x0500, 0x0504, 0x0510, 0x0514, 0x0520, 0x0524, 0x0530, 0x0534,
+        0x0580, 0x0584, 0x0590, 0x0594, 0x05a0, 0x05a4, 0x05b0, 0x05b4, 0x0600, 0x0604, 0x0610,
+        0x0614, 0x0620, 0x0624, 0x0630, 0x0634, 0x0680, 0x0684, 0x0690, 0x0694, 0x06a0, 0x06a4,
+        0x06b0, 0x06b4, 0x0700, 0x0704, 0x0710, 0x0714, 0x0720, 0x0724, 0x0730, 0x0734, 0x0780,
+        0x0784, 0x0790, 0x0794, 0x07a0, 0x07a4, 0x07b0, 0x07b4, 0x0000, 0x0004, 0x0010, 0x0014,
+        0x0020, 0x0024, 0x0030, 0x0034, 0x0080, 0x0084, 0x0090, 0x0094, 0x00a0, 0x00a4, 0x00b0,
+        0x00b4, 0x0100, 0x0104, 0x0110, 0x0114, 0x0120, 0x0124, 0x0130, 0x0134, 0x0180, 0x0184,
+        0x0190, 0x0194, 0x01a0, 0x01a4, 0x01b0, 0x01b4, 0x0200, 0x0204, 0x0210, 0x0214, 0x0220,
+        0x0224, 0x0230, 0x0234, 0x0280, 0x0284, 0x0290, 0x0294, 0x02a0, 0x02a4, 0x02b0, 0x02b4,
+        0x0300, 0x0304, 0x0310, 0x0314, 0x0320, 0x0324, 0x0330, 0x0334, 0x0380, 0x0384, 0x0390,
+        0x0394, 0x03a0, 0x03a4, 0x03b0, 0x03b4, 0x0400, 0x0404, 0x0410, 0x0414, 0x0420, 0x0424,
+        0x0430, 0x0434, 0x0480, 0x0484, 0x0490, 0x0494, 0x04a0, 0x04a4, 0x04b0, 0x04b4, 0x0500,
+        0x0504, 0x0510, 0x0514, 0x0520, 0x0524, 0x0530, 0x0534, 0x0580, 0x0584, 0x0590, 0x0594,
+        0x05a0, 0x05a4, 0x05b0, 0x05b4, 0x0600, 0x0604, 0x0610, 0x0614, 0x0620, 0x0624, 0x0630,
+        0x0634, 0x0680, 0x0684, 0x0690, 0x0694, 0x06a0, 0x06a4, 0x06b0, 0x06b4, 0x0700, 0x0704,
+        0x0710, 0x0714, 0x0720, 0x0724, 0x0730, 0x0734, 0x0780, 0x0784, 0x0790, 0x0794, 0x07a0,
+        0x07a4, 0x07b0, 0x07b4, 0x0000, 0x0004, 0x0010, 0x0014, 0x0020, 0x0024, 0x0030, 0x0034,
+        0x0080, 0x0084, 0x0090, 0x0094, 0x00a0, 0x00a4, 0x00b0, 0x00b4, 0x0100, 0x0104, 0x0110,
+        0x0114, 0x0120, 0x0124, 0x0130, 0x0134, 0x0180, 0x0184, 0x0190, 0x0194, 0x01a0, 0x01a4,
+        0x01b0, 0x01b4, 0x0200, 0x0204, 0x0210, 0x0214, 0x0220, 0x0224, 0x0230, 0x0234, 0x0280,
+        0x0284, 0x0290, 0x0294, 0x02a0, 0x02a4, 0x02b0, 0x02b4, 0x0300, 0x0304, 0x0310, 0x0314,
+        0x0320, 0x0324, 0x0330, 0x0334, 0x0380, 0x0384, 0x0390, 0x0394, 0x03a0, 0x03a4, 0x03b0,
+        0x03b4, 0x0400, 0x0404, 0x0410, 0x0414, 0x0420, 0x0424, 0x0430, 0x0434, 0x0480, 0x0484,
+        0x0490, 0x0494, 0x04a0, 0x04a4, 0x04b0, 0x04b4, 0x0500, 0x0504, 0x0510, 0x0514, 0x0520,
+        0x0524, 0x0530, 0x0534, 0x0580, 0x0584, 0x0590, 0x0594, 0x05a0, 0x05a4, 0x05b0, 0x05b4,
+        0x0600, 0x0604, 0x0610, 0x0614, 0x0620, 0x0624, 0x0630, 0x0634, 0x0680, 0x0684, 0x0690,
+        0x0694, 0x06a0, 0x06a4, 0x06b0, 0x06b4, 0x0700, 0x0704, 0x0710, 0x0714, 0x0720, 0x0724,
+        0x0730, 0x0734, 0x0780, 0x0784, 0x0790, 0x0794, 0x07a0, 0x07a4, 0x07b0, 0x07b4,
+    };
+    return xlut[x % 128] + ylut[y % 128];
+}
+
+static u32 GetMortonOffset128(u32 x, u32 y, u32 bytes_per_pixel) {
+    // Calculates the offset of the position of the pixel in Morton order
+    // Framebuffer images are split into 128x128 tiles.
+
+    constexpr u32 block_height = 128;
+    const u32 coarse_x = x & ~127;
+
+    const u32 i = MortonInterleave128(x, y);
+
+    const u32 offset = coarse_x * block_height;
+
+    return (i + offset) * bytes_per_pixel;
+}
+
+void MortonSwizzle(MortonSwizzleMode mode, Surface::PixelFormat format, u32 stride,
+                   u32 block_height, u32 height, u32 block_depth, u32 depth, u32 tile_width_spacing,
+                   u8* buffer, std::size_t buffer_size, VAddr addr) {
+
+    GetSwizzleFunction(mode, format)(stride, block_height, height, block_depth, depth,
+                                     tile_width_spacing, buffer, buffer_size, addr);
+}
+
+void MortonCopyPixels128(u32 width, u32 height, u32 bytes_per_pixel, u32 linear_bytes_per_pixel,
+                         u8* morton_data, u8* linear_data, bool morton_to_linear) {
+    u8* data_ptrs[2];
+    for (u32 y = 0; y < height; ++y) {
+        for (u32 x = 0; x < width; ++x) {
+            const u32 coarse_y = y & ~127;
+            const u32 morton_offset =
+                GetMortonOffset128(x, y, bytes_per_pixel) + coarse_y * width * bytes_per_pixel;
+            const u32 linear_pixel_index = (x + y * width) * linear_bytes_per_pixel;
+
+            data_ptrs[morton_to_linear ? 1 : 0] = morton_data + morton_offset;
+            data_ptrs[morton_to_linear ? 0 : 1] = &linear_data[linear_pixel_index];
+
+            std::memcpy(data_ptrs[0], data_ptrs[1], bytes_per_pixel);
+        }
+    }
+}
+
+} // namespace VideoCore

src/video_core/morton.h

@@ -0,0 +1,21 @@
+// Copyright 2018 yuzu Emulator Project
+// Licensed under GPLv2 or any later version
+// Refer to the license.txt file included.
+
+#pragma once
+
+#include "common/common_types.h"
+#include "video_core/surface.h"
+
+namespace VideoCore {
+
+enum class MortonSwizzleMode { MortonToLinear, LinearToMorton };
+
+void MortonSwizzle(MortonSwizzleMode mode, VideoCore::Surface::PixelFormat format, u32 stride,
+                   u32 block_height, u32 height, u32 block_depth, u32 depth, u32 tile_width_spacing,
+                   u8* buffer, std::size_t buffer_size, VAddr addr);
+
+void MortonCopyPixels128(u32 width, u32 height, u32 bytes_per_pixel, u32 linear_bytes_per_pixel,
+                         u8* morton_data, u8* linear_data, bool morton_to_linear);
+
+} // namespace VideoCore

src/video_core/renderer_opengl/gl_buffer_cache.cpp

@@ -76,7 +76,7 @@ std::tuple<u8*, GLintptr> OGLBufferCache::ReserveMemory(std::size_t size, std::s
     return std::make_tuple(uploaded_ptr, uploaded_offset);
 }
 
-void OGLBufferCache::Map(std::size_t max_size) {
+bool OGLBufferCache::Map(std::size_t max_size) {
     bool invalidate;
     std::tie(buffer_ptr, buffer_offset_base, invalidate) =
         stream_buffer.Map(static_cast<GLsizeiptr>(max_size), 4);
@@ -85,6 +85,7 @@ void OGLBufferCache::Map(std::size_t max_size) {
     if (invalidate) {
         InvalidateAll();
     }
+    return invalidate;
 }
 
 void OGLBufferCache::Unmap() {

src/video_core/renderer_opengl/gl_buffer_cache.h

@@ -50,7 +50,7 @@ public:
     /// Reserves memory to be used by host's CPU. Returns mapped address and offset.
     std::tuple<u8*, GLintptr> ReserveMemory(std::size_t size, std::size_t alignment = 4);
 
-    void Map(std::size_t max_size);
+    bool Map(std::size_t max_size);
     void Unmap();
 
     GLuint GetHandle() const;

src/video_core/renderer_opengl/gl_rasterizer.cpp

@@ -113,10 +113,24 @@ RasterizerOpenGL::RasterizerOpenGL(Core::Frontend::EmuWindow& window, ScreenInfo
     glGetIntegerv(GL_UNIFORM_BUFFER_OFFSET_ALIGNMENT, &uniform_buffer_alignment);
 
     LOG_CRITICAL(Render_OpenGL, "Sync fixed function OpenGL state here!");
+    CheckExtensions();
 }
 
 RasterizerOpenGL::~RasterizerOpenGL() {}
 
+void RasterizerOpenGL::CheckExtensions() {
+    if (!GLAD_GL_ARB_texture_filter_anisotropic && !GLAD_GL_EXT_texture_filter_anisotropic) {
+        LOG_WARNING(
+            Render_OpenGL,
+            "Anisotropic filter is not supported! This can cause graphical issues in some games.");
+    }
+    if (!GLAD_GL_ARB_buffer_storage) {
+        LOG_WARNING(
+            Render_OpenGL,
+            "Buffer storage control is not supported! This can cause performance degradation.");
+    }
+}
+
 void RasterizerOpenGL::SetupVertexFormat() {
     auto& gpu = Core::System::GetInstance().GPU().Maxwell3D();
     const auto& regs = gpu.regs;
@@ -176,15 +190,25 @@ void RasterizerOpenGL::SetupVertexFormat() {
     }
     state.draw.vertex_array = VAO.handle;
     state.ApplyVertexBufferState();
+
+    // Rebinding the VAO invalidates the vertex buffer bindings.
+    gpu.dirty_flags.vertex_array = 0xFFFFFFFF;
 }
 
 void RasterizerOpenGL::SetupVertexBuffer() {
-    MICROPROFILE_SCOPE(OpenGL_VB);
-    const auto& gpu = Core::System::GetInstance().GPU().Maxwell3D();
+    auto& gpu = Core::System::GetInstance().GPU().Maxwell3D();
     const auto& regs = gpu.regs;
 
+    if (!gpu.dirty_flags.vertex_array)
+        return;
+
+    MICROPROFILE_SCOPE(OpenGL_VB);
+
     // Upload all guest vertex arrays sequentially to our buffer
     for (u32 index = 0; index < Maxwell::NumVertexArrays; ++index) {
+        if (~gpu.dirty_flags.vertex_array & (1u << index))
+            continue;
+
         const auto& vertex_array = regs.vertex_array[index];
         if (!vertex_array.IsEnabled())
             continue;
@@ -211,6 +235,8 @@ void RasterizerOpenGL::SetupVertexBuffer() {
 
     // Implicit set by glBindVertexBuffer. Stupid glstate handling...
     state.draw.vertex_buffer = buffer_cache.GetHandle();
+
+    gpu.dirty_flags.vertex_array = 0;
 }
 
 DrawParameters RasterizerOpenGL::SetupDraw() {
@@ -600,7 +626,7 @@ void RasterizerOpenGL::DrawArrays() {
         return;
 
     MICROPROFILE_SCOPE(OpenGL_Drawing);
-    const auto& gpu = Core::System::GetInstance().GPU().Maxwell3D();
+    auto& gpu = Core::System::GetInstance().GPU().Maxwell3D();
     const auto& regs = gpu.regs;
 
     ScopeAcquireGLContext acquire_context{emu_window};
@@ -616,11 +642,12 @@ void RasterizerOpenGL::DrawArrays() {
     SyncCullMode();
     SyncPrimitiveRestart();
     SyncScissorTest(state);
+    SyncClipEnabled();
     // Alpha Testing is synced on shaders.
     SyncTransformFeedback();
     SyncPointState();
     CheckAlphaTests();
-
+    SyncPolygonOffset();
     // TODO(bunnei): Sync framebuffer_scale uniform here
     // TODO(bunnei): Sync scissorbox uniform(s) here
 
@@ -653,7 +680,11 @@ void RasterizerOpenGL::DrawArrays() {
     // Add space for at least 18 constant buffers
     buffer_size += Maxwell::MaxConstBuffers * (MaxConstbufferSize + uniform_buffer_alignment);
 
-    buffer_cache.Map(buffer_size);
+    bool invalidate = buffer_cache.Map(buffer_size);
+    if (invalidate) {
+        // As all cached buffers are invalidated, we need to recheck their state.
+        gpu.dirty_flags.vertex_array = 0xFFFFFFFF;
+    }
 
     SetupVertexFormat();
     SetupVertexBuffer();
@@ -969,9 +1000,14 @@ u32 RasterizerOpenGL::SetupTextures(Maxwell::ShaderStage stage, Shader& shader,
 
 void RasterizerOpenGL::SyncViewport(OpenGLState& current_state) {
     const auto& regs = Core::System::GetInstance().GPU().Maxwell3D().regs;
-    for (std::size_t i = 0; i < Tegra::Engines::Maxwell3D::Regs::NumViewports; i++) {
-        const MathUtil::Rectangle<s32> viewport_rect{regs.viewport_transform[i].GetRect()};
+    const bool geometry_shaders_enabled =
+        regs.IsShaderConfigEnabled(static_cast<size_t>(Maxwell::ShaderProgram::Geometry));
+    const std::size_t viewport_count =
+        geometry_shaders_enabled ? Tegra::Engines::Maxwell3D::Regs::NumViewports : 1;
+    for (std::size_t i = 0; i < viewport_count; i++) {
         auto& viewport = current_state.viewports[i];
+        const auto& src = regs.viewports[i];
+        const MathUtil::Rectangle<s32> viewport_rect{regs.viewport_transform[i].GetRect()};
         viewport.x = viewport_rect.left;
         viewport.y = viewport_rect.bottom;
         viewport.width = viewport_rect.GetWidth();
@@ -979,10 +1015,20 @@ void RasterizerOpenGL::SyncViewport(OpenGLState& current_state) {
         viewport.depth_range_far = regs.viewports[i].depth_range_far;
         viewport.depth_range_near = regs.viewports[i].depth_range_near;
     }
+    state.depth_clamp.far_plane = regs.view_volume_clip_control.depth_clamp_far != 0;
+    state.depth_clamp.near_plane = regs.view_volume_clip_control.depth_clamp_near != 0;
 }
 
 void RasterizerOpenGL::SyncClipEnabled() {
-    UNREACHABLE();
+    const auto& regs = Core::System::GetInstance().GPU().Maxwell3D().regs;
+    state.clip_distance[0] = regs.clip_distance_enabled.c0 != 0;
+    state.clip_distance[1] = regs.clip_distance_enabled.c1 != 0;
+    state.clip_distance[2] = regs.clip_distance_enabled.c2 != 0;
+    state.clip_distance[3] = regs.clip_distance_enabled.c3 != 0;
+    state.clip_distance[4] = regs.clip_distance_enabled.c4 != 0;
+    state.clip_distance[5] = regs.clip_distance_enabled.c5 != 0;
+    state.clip_distance[6] = regs.clip_distance_enabled.c6 != 0;
+    state.clip_distance[7] = regs.clip_distance_enabled.c7 != 0;
 }
 
 void RasterizerOpenGL::SyncClipCoef() {
@@ -1149,7 +1195,11 @@ void RasterizerOpenGL::SyncLogicOpState() {
 
 void RasterizerOpenGL::SyncScissorTest(OpenGLState& current_state) {
     const auto& regs = Core::System::GetInstance().GPU().Maxwell3D().regs;
-    for (std::size_t i = 0; i < Tegra::Engines::Maxwell3D::Regs::NumViewports; i++) {
+    const bool geometry_shaders_enabled =
+        regs.IsShaderConfigEnabled(static_cast<size_t>(Maxwell::ShaderProgram::Geometry));
+    const std::size_t viewport_count =
+        geometry_shaders_enabled ? Tegra::Engines::Maxwell3D::Regs::NumViewports : 1;
+    for (std::size_t i = 0; i < viewport_count; i++) {
         const auto& src = regs.scissor_test[i];
         auto& dst = current_state.viewports[i].scissor;
         dst.enabled = (src.enable != 0);
@@ -1179,6 +1229,16 @@ void RasterizerOpenGL::SyncPointState() {
     state.point.size = regs.point_size;
 }
 
+void RasterizerOpenGL::SyncPolygonOffset() {
+    const auto& regs = Core::System::GetInstance().GPU().Maxwell3D().regs;
+    state.polygon_offset.fill_enable = regs.polygon_offset_fill_enable != 0;
+    state.polygon_offset.line_enable = regs.polygon_offset_line_enable != 0;
+    state.polygon_offset.point_enable = regs.polygon_offset_point_enable != 0;
+    state.polygon_offset.units = regs.polygon_offset_units;
+    state.polygon_offset.factor = regs.polygon_offset_factor;
+    state.polygon_offset.clamp = regs.polygon_offset_clamp;
+}
+
 void RasterizerOpenGL::CheckAlphaTests() {
     const auto& regs = Core::System::GetInstance().GPU().Maxwell3D().regs;
 

src/video_core/renderer_opengl/gl_rasterizer.h

@@ -183,9 +183,16 @@ private:
     /// Syncs Color Mask
     void SyncColorMask();
 
+    /// Syncs the polygon offsets
+    void SyncPolygonOffset();
+
     /// Check asserts for alpha testing.
     void CheckAlphaTests();
 
+    /// Check for extension that are not strictly required
+    /// but are needed for correct emulation
+    void CheckExtensions();
+
     bool has_ARB_direct_state_access = false;
     bool has_ARB_multi_bind = false;
 

src/video_core/renderer_opengl/gl_rasterizer_cache.cpp

@@ -15,6 +15,7 @@
 #include "core/memory.h"
 #include "core/settings.h"
 #include "video_core/engines/maxwell_3d.h"
+#include "video_core/morton.h"
 #include "video_core/renderer_opengl/gl_rasterizer.h"
 #include "video_core/renderer_opengl/gl_rasterizer_cache.h"
 #include "video_core/renderer_opengl/gl_state.h"
@@ -22,10 +23,11 @@
 #include "video_core/surface.h"
 #include "video_core/textures/astc.h"
 #include "video_core/textures/decoders.h"
-#include "video_core/utils.h"
 
 namespace OpenGL {
 
+using VideoCore::MortonSwizzle;
+using VideoCore::MortonSwizzleMode;
 using VideoCore::Surface::ComponentTypeFromDepthFormat;
 using VideoCore::Surface::ComponentTypeFromRenderTarget;
 using VideoCore::Surface::ComponentTypeFromTexture;
@@ -95,6 +97,7 @@ std::size_t SurfaceParams::InnerMemorySize(bool force_gl, bool layer_only,
     params.block_width = params.is_tiled ? config.tic.BlockWidth() : 0,
     params.block_height = params.is_tiled ? config.tic.BlockHeight() : 0,
     params.block_depth = params.is_tiled ? config.tic.BlockDepth() : 0,
+    params.tile_width_spacing = params.is_tiled ? (1 << config.tic.tile_width_spacing.Value()) : 1;
     params.srgb_conversion = config.tic.IsSrgbConversionEnabled();
     params.pixel_format = PixelFormatFromTextureFormat(config.tic.format, config.tic.r_type.Value(),
                                                        params.srgb_conversion);
@@ -160,6 +163,7 @@ std::size_t SurfaceParams::InnerMemorySize(bool force_gl, bool layer_only,
     params.block_width = 1 << config.memory_layout.block_width;
     params.block_height = 1 << config.memory_layout.block_height;
     params.block_depth = 1 << config.memory_layout.block_depth;
+    params.tile_width_spacing = 1;
     params.pixel_format = PixelFormatFromRenderTargetFormat(config.format);
     params.srgb_conversion = config.format == Tegra::RenderTargetFormat::BGRA8_SRGB ||
                              config.format == Tegra::RenderTargetFormat::RGBA8_SRGB;
@@ -195,6 +199,7 @@ std::size_t SurfaceParams::InnerMemorySize(bool force_gl, bool layer_only,
     params.block_width = 1 << std::min(block_width, 5U);
     params.block_height = 1 << std::min(block_height, 5U);
     params.block_depth = 1 << std::min(block_depth, 5U);
+    params.tile_width_spacing = 1;
     params.pixel_format = PixelFormatFromDepthFormat(format);
     params.component_type = ComponentTypeFromDepthFormat(format);
     params.type = GetFormatType(params.pixel_format);
@@ -221,6 +226,7 @@ std::size_t SurfaceParams::InnerMemorySize(bool force_gl, bool layer_only,
     params.block_width = params.is_tiled ? std::min(config.BlockWidth(), 32U) : 0,
     params.block_height = params.is_tiled ? std::min(config.BlockHeight(), 32U) : 0,
     params.block_depth = params.is_tiled ? std::min(config.BlockDepth(), 32U) : 0,
+    params.tile_width_spacing = 1;
     params.pixel_format = PixelFormatFromRenderTargetFormat(config.format);
     params.srgb_conversion = config.format == Tegra::RenderTargetFormat::BGRA8_SRGB ||
                              config.format == Tegra::RenderTargetFormat::RGBA8_SRGB;
@@ -370,174 +376,7 @@ MathUtil::Rectangle<u32> SurfaceParams::GetRect(u32 mip_level) const {
     return {0, actual_height, MipWidth(mip_level), 0};
 }
 
-template <bool morton_to_gl, PixelFormat format>
-void MortonCopy(u32 stride, u32 block_height, u32 height, u32 block_depth, u32 depth, u8* gl_buffer,
-                std::size_t gl_buffer_size, VAddr addr) {
-    constexpr u32 bytes_per_pixel = GetBytesPerPixel(format);
-
-    // With the BCn formats (DXT and DXN), each 4x4 tile is swizzled instead of just individual
-    // pixel values.
-    const u32 tile_size_x{GetDefaultBlockWidth(format)};
-    const u32 tile_size_y{GetDefaultBlockHeight(format)};
-
-    if (morton_to_gl) {
-        Tegra::Texture::UnswizzleTexture(gl_buffer, addr, tile_size_x, tile_size_y, bytes_per_pixel,
-                                         stride, height, depth, block_height, block_depth);
-    } else {
-        Tegra::Texture::CopySwizzledData((stride + tile_size_x - 1) / tile_size_x,
-                                         (height + tile_size_y - 1) / tile_size_y, depth,
-                                         bytes_per_pixel, bytes_per_pixel, Memory::GetPointer(addr),
-                                         gl_buffer, false, block_height, block_depth);
-    }
-}
-
-using GLConversionArray = std::array<void (*)(u32, u32, u32, u32, u32, u8*, std::size_t, VAddr),
-                                     VideoCore::Surface::MaxPixelFormat>;
-
-static constexpr GLConversionArray morton_to_gl_fns = {
-    // clang-format off
-        MortonCopy<true, PixelFormat::ABGR8U>,
-        MortonCopy<true, PixelFormat::ABGR8S>,
-        MortonCopy<true, PixelFormat::ABGR8UI>,
-        MortonCopy<true, PixelFormat::B5G6R5U>,
-        MortonCopy<true, PixelFormat::A2B10G10R10U>,
-        MortonCopy<true, PixelFormat::A1B5G5R5U>,
-        MortonCopy<true, PixelFormat::R8U>,
-        MortonCopy<true, PixelFormat::R8UI>,
-        MortonCopy<true, PixelFormat::RGBA16F>,
-        MortonCopy<true, PixelFormat::RGBA16U>,
-        MortonCopy<true, PixelFormat::RGBA16UI>,
-        MortonCopy<true, PixelFormat::R11FG11FB10F>,
-        MortonCopy<true, PixelFormat::RGBA32UI>,
-        MortonCopy<true, PixelFormat::DXT1>,
-        MortonCopy<true, PixelFormat::DXT23>,
-        MortonCopy<true, PixelFormat::DXT45>,
-        MortonCopy<true, PixelFormat::DXN1>,
-        MortonCopy<true, PixelFormat::DXN2UNORM>,
-        MortonCopy<true, PixelFormat::DXN2SNORM>,
-        MortonCopy<true, PixelFormat::BC7U>,
-        MortonCopy<true, PixelFormat::BC6H_UF16>,
-        MortonCopy<true, PixelFormat::BC6H_SF16>,
-        MortonCopy<true, PixelFormat::ASTC_2D_4X4>,
-        MortonCopy<true, PixelFormat::G8R8U>,
-        MortonCopy<true, PixelFormat::G8R8S>,
-        MortonCopy<true, PixelFormat::BGRA8>,
-        MortonCopy<true, PixelFormat::RGBA32F>,
-        MortonCopy<true, PixelFormat::RG32F>,
-        MortonCopy<true, PixelFormat::R32F>,
-        MortonCopy<true, PixelFormat::R16F>,
-        MortonCopy<true, PixelFormat::R16U>,
-        MortonCopy<true, PixelFormat::R16S>,
-        MortonCopy<true, PixelFormat::R16UI>,
-        MortonCopy<true, PixelFormat::R16I>,
-        MortonCopy<true, PixelFormat::RG16>,
-        MortonCopy<true, PixelFormat::RG16F>,
-        MortonCopy<true, PixelFormat::RG16UI>,
-        MortonCopy<true, PixelFormat::RG16I>,
-        MortonCopy<true, PixelFormat::RG16S>,
-        MortonCopy<true, PixelFormat::RGB32F>,
-        MortonCopy<true, PixelFormat::RGBA8_SRGB>,
-        MortonCopy<true, PixelFormat::RG8U>,
-        MortonCopy<true, PixelFormat::RG8S>,
-        MortonCopy<true, PixelFormat::RG32UI>,
-        MortonCopy<true, PixelFormat::R32UI>,
-        MortonCopy<true, PixelFormat::ASTC_2D_8X8>,
-        MortonCopy<true, PixelFormat::ASTC_2D_8X5>,
-        MortonCopy<true, PixelFormat::ASTC_2D_5X4>,
-        MortonCopy<true, PixelFormat::BGRA8_SRGB>,
-        MortonCopy<true, PixelFormat::DXT1_SRGB>,
-        MortonCopy<true, PixelFormat::DXT23_SRGB>,
-        MortonCopy<true, PixelFormat::DXT45_SRGB>,
-        MortonCopy<true, PixelFormat::BC7U_SRGB>,
-        MortonCopy<true, PixelFormat::ASTC_2D_4X4_SRGB>,
-        MortonCopy<true, PixelFormat::ASTC_2D_8X8_SRGB>,
-        MortonCopy<true, PixelFormat::ASTC_2D_8X5_SRGB>,
-        MortonCopy<true, PixelFormat::ASTC_2D_5X4_SRGB>,
-        MortonCopy<true, PixelFormat::ASTC_2D_5X5>,
-        MortonCopy<true, PixelFormat::ASTC_2D_5X5_SRGB>,
-        MortonCopy<true, PixelFormat::ASTC_2D_10X8>,
-        MortonCopy<true, PixelFormat::ASTC_2D_10X8_SRGB>,
-        MortonCopy<true, PixelFormat::Z32F>,
-        MortonCopy<true, PixelFormat::Z16>,
-        MortonCopy<true, PixelFormat::Z24S8>,
-        MortonCopy<true, PixelFormat::S8Z24>,
-        MortonCopy<true, PixelFormat::Z32FS8>,
-    // clang-format on
-};
-
-static constexpr GLConversionArray gl_to_morton_fns = {
-    // clang-format off
-        MortonCopy<false, PixelFormat::ABGR8U>,
-        MortonCopy<false, PixelFormat::ABGR8S>,
-        MortonCopy<false, PixelFormat::ABGR8UI>,
-        MortonCopy<false, PixelFormat::B5G6R5U>,
-        MortonCopy<false, PixelFormat::A2B10G10R10U>,
-        MortonCopy<false, PixelFormat::A1B5G5R5U>,
-        MortonCopy<false, PixelFormat::R8U>,
-        MortonCopy<false, PixelFormat::R8UI>,
-        MortonCopy<false, PixelFormat::RGBA16F>,
-        MortonCopy<false, PixelFormat::RGBA16U>,
-        MortonCopy<false, PixelFormat::RGBA16UI>,
-        MortonCopy<false, PixelFormat::R11FG11FB10F>,
-        MortonCopy<false, PixelFormat::RGBA32UI>,
-        MortonCopy<false, PixelFormat::DXT1>,
-        MortonCopy<false, PixelFormat::DXT23>,
-        MortonCopy<false, PixelFormat::DXT45>,
-        MortonCopy<false, PixelFormat::DXN1>,
-        MortonCopy<false, PixelFormat::DXN2UNORM>,
-        MortonCopy<false, PixelFormat::DXN2SNORM>,
-        MortonCopy<false, PixelFormat::BC7U>,
-        MortonCopy<false, PixelFormat::BC6H_UF16>,
-        MortonCopy<false, PixelFormat::BC6H_SF16>,
-        // TODO(Subv): Swizzling ASTC formats are not supported
-        nullptr,
-        MortonCopy<false, PixelFormat::G8R8U>,
-        MortonCopy<false, PixelFormat::G8R8S>,
-        MortonCopy<false, PixelFormat::BGRA8>,
-        MortonCopy<false, PixelFormat::RGBA32F>,
-        MortonCopy<false, PixelFormat::RG32F>,
-        MortonCopy<false, PixelFormat::R32F>,
-        MortonCopy<false, PixelFormat::R16F>,
-        MortonCopy<false, PixelFormat::R16U>,
-        MortonCopy<false, PixelFormat::R16S>,
-        MortonCopy<false, PixelFormat::R16UI>,
-        MortonCopy<false, PixelFormat::R16I>,
-        MortonCopy<false, PixelFormat::RG16>,
-        MortonCopy<false, PixelFormat::RG16F>,
-        MortonCopy<false, PixelFormat::RG16UI>,
-        MortonCopy<false, PixelFormat::RG16I>,
-        MortonCopy<false, PixelFormat::RG16S>,
-        MortonCopy<false, PixelFormat::RGB32F>,
-        MortonCopy<false, PixelFormat::RGBA8_SRGB>,
-        MortonCopy<false, PixelFormat::RG8U>,
-        MortonCopy<false, PixelFormat::RG8S>,
-        MortonCopy<false, PixelFormat::RG32UI>,
-        MortonCopy<false, PixelFormat::R32UI>,
-        nullptr,
-        nullptr,
-        nullptr,
-        MortonCopy<false, PixelFormat::BGRA8_SRGB>,
-        MortonCopy<false, PixelFormat::DXT1_SRGB>,
-        MortonCopy<false, PixelFormat::DXT23_SRGB>,
-        MortonCopy<false, PixelFormat::DXT45_SRGB>,
-        MortonCopy<false, PixelFormat::BC7U_SRGB>,
-        nullptr,
-        nullptr,
-        nullptr,
-        nullptr,
-        nullptr,
-        nullptr,
-        nullptr,
-        nullptr,
-        MortonCopy<false, PixelFormat::Z32F>,
-        MortonCopy<false, PixelFormat::Z16>,
-        MortonCopy<false, PixelFormat::Z24S8>,
-        MortonCopy<false, PixelFormat::S8Z24>,
-        MortonCopy<false, PixelFormat::Z32FS8>,
-    // clang-format on
-};
-
-void SwizzleFunc(const GLConversionArray& functions, const SurfaceParams& params,
+void SwizzleFunc(const MortonSwizzleMode& mode, const SurfaceParams& params,
                  std::vector<u8>& gl_buffer, u32 mip_level) {
     u32 depth = params.MipDepth(mip_level);
     if (params.target == SurfaceTarget::Texture2D) {
@@ -550,19 +389,19 @@ void SwizzleFunc(const GLConversionArray& functions, const SurfaceParams& params
         const u64 layer_size = params.LayerMemorySize();
         const u64 gl_size = params.LayerSizeGL(mip_level);
         for (u32 i = 0; i < params.depth; i++) {
-            functions[static_cast<std::size_t>(params.pixel_format)](
-                params.MipWidth(mip_level), params.MipBlockHeight(mip_level),
-                params.MipHeight(mip_level), params.MipBlockDepth(mip_level), 1,
-                gl_buffer.data() + offset_gl, gl_size, params.addr + offset);
+            MortonSwizzle(mode, params.pixel_format, params.MipWidth(mip_level),
+                          params.MipBlockHeight(mip_level), params.MipHeight(mip_level),
+                          params.MipBlockDepth(mip_level), params.tile_width_spacing, 1,
+                          gl_buffer.data() + offset_gl, gl_size, params.addr + offset);
             offset += layer_size;
             offset_gl += gl_size;
         }
     } else {
         const u64 offset = params.GetMipmapLevelOffset(mip_level);
-        functions[static_cast<std::size_t>(params.pixel_format)](
-            params.MipWidth(mip_level), params.MipBlockHeight(mip_level),
-            params.MipHeight(mip_level), params.MipBlockDepth(mip_level), depth, gl_buffer.data(),
-            gl_buffer.size(), params.addr + offset);
+        MortonSwizzle(mode, params.pixel_format, params.MipWidth(mip_level),
+                      params.MipBlockHeight(mip_level), params.MipHeight(mip_level),
+                      params.MipBlockDepth(mip_level), depth, params.tile_width_spacing,
+                      gl_buffer.data(), gl_buffer.size(), params.addr + offset);
     }
 }
 
@@ -996,7 +835,7 @@ void CachedSurface::LoadGLBuffer() {
         ASSERT_MSG(params.block_width == 1, "Block width is defined as {} on texture type {}",
                    params.block_width, static_cast<u32>(params.target));
         for (u32 i = 0; i < params.max_mip_level; i++)
-            SwizzleFunc(morton_to_gl_fns, params, gl_buffer[i], i);
+            SwizzleFunc(MortonSwizzleMode::MortonToLinear, params, gl_buffer[i], i);
     } else {
         const auto texture_src_data{Memory::GetPointer(params.addr)};
         const auto texture_src_data_end{texture_src_data + params.size_in_bytes_gl};
@@ -1035,7 +874,7 @@ void CachedSurface::FlushGLBuffer() {
         ASSERT_MSG(params.block_width == 1, "Block width is defined as {} on texture type {}",
                    params.block_width, static_cast<u32>(params.target));
 
-        SwizzleFunc(gl_to_morton_fns, params, gl_buffer[0], 0);
+        SwizzleFunc(MortonSwizzleMode::LinearToMorton, params, gl_buffer[0], 0);
     } else {
         std::memcpy(Memory::GetPointer(GetAddr()), gl_buffer[0].data(), GetSizeInBytes());
     }

src/video_core/renderer_opengl/gl_rasterizer_cache.h

@@ -196,9 +196,15 @@ struct SurfaceParams {
 
     /// Checks if surfaces are compatible for caching
     bool IsCompatibleSurface(const SurfaceParams& other) const {
-        return std::tie(pixel_format, type, width, height, target, depth) ==
-               std::tie(other.pixel_format, other.type, other.width, other.height, other.target,
-                        other.depth);
+        if (std::tie(pixel_format, type, width, height, target, depth, is_tiled) ==
+            std::tie(other.pixel_format, other.type, other.width, other.height, other.target,
+                     other.depth, other.is_tiled)) {
+            if (!is_tiled)
+                return true;
+            return std::tie(block_height, block_depth, tile_width_spacing) ==
+                   std::tie(other.block_height, other.block_depth, other.tile_width_spacing);
+        }
+        return false;
     }
 
     /// Initializes parameters for caching, should be called after everything has been initialized
@@ -208,6 +214,7 @@ struct SurfaceParams {
     u32 block_width;
     u32 block_height;
     u32 block_depth;
+    u32 tile_width_spacing;
     PixelFormat pixel_format;
     ComponentType component_type;
     SurfaceType type;

src/video_core/renderer_opengl/gl_shader_decompiler.cpp

@@ -500,27 +500,42 @@ public:
                                       const Register& buf_reg) {
         const std::string dest = GetOutputAttribute(attribute);
         const std::string src = GetRegisterAsFloat(val_reg);
+        if (dest.empty())
+            return;
 
-        if (!dest.empty()) {
-            // Can happen with unknown/unimplemented output attributes, in which case we ignore the
-            // instruction for now.
-            if (stage == Maxwell3D::Regs::ShaderStage::Geometry) {
-                // TODO(Rodrigo): nouveau sets some attributes after setting emitting a geometry
-                // shader. These instructions use a dirty register as buffer index, to avoid some
-                // drivers from complaining about out of boundary writes, guard them.
-                const std::string buf_index{"((" + GetRegisterAsInteger(buf_reg) + ") % " +
-                                            std::to_string(MAX_GEOMETRY_BUFFERS) + ')'};
-                shader.AddLine("amem[" + buf_index + "][" +
-                               std::to_string(static_cast<u32>(attribute)) + ']' +
-                               GetSwizzle(elem) + " = " + src + ';');
-            } else {
-                if (attribute == Attribute::Index::PointSize) {
-                    fixed_pipeline_output_attributes_used.insert(attribute);
-                    shader.AddLine(dest + " = " + src + ';');
-                } else {
-                    shader.AddLine(dest + GetSwizzle(elem) + " = " + src + ';');
-                }
-            }
+        // Can happen with unknown/unimplemented output attributes, in which case we ignore the
+        // instruction for now.
+        if (stage == Maxwell3D::Regs::ShaderStage::Geometry) {
+            // TODO(Rodrigo): nouveau sets some attributes after setting emitting a geometry
+            // shader. These instructions use a dirty register as buffer index, to avoid some
+            // drivers from complaining about out of boundary writes, guard them.
+            const std::string buf_index{"((" + GetRegisterAsInteger(buf_reg) + ") % " +
+                                        std::to_string(MAX_GEOMETRY_BUFFERS) + ')'};
+            shader.AddLine("amem[" + buf_index + "][" +
+                           std::to_string(static_cast<u32>(attribute)) + ']' + GetSwizzle(elem) +
+                           " = " + src + ';');
+            return;
+        }
+
+        switch (attribute) {
+        case Attribute::Index::ClipDistances0123:
+        case Attribute::Index::ClipDistances4567: {
+            const u64 index = (attribute == Attribute::Index::ClipDistances4567 ? 4 : 0) + elem;
+            UNIMPLEMENTED_IF_MSG(
+                ((header.vtg.clip_distances >> index) & 1) == 0,
+                "Shader is setting gl_ClipDistance{} without enabling it in the header", index);
+
+            fixed_pipeline_output_attributes_used.insert(attribute);
+            shader.AddLine(dest + '[' + std::to_string(index) + "] = " + src + ';');
+            break;
+        }
+        case Attribute::Index::PointSize:
+            fixed_pipeline_output_attributes_used.insert(attribute);
+            shader.AddLine(dest + " = " + src + ';');
+            break;
+        default:
+            shader.AddLine(dest + GetSwizzle(elem) + " = " + src + ';');
+            break;
         }
     }
 
@@ -740,12 +755,19 @@ private:
     void GenerateVertex() {
         if (stage != Maxwell3D::Regs::ShaderStage::Vertex)
             return;
+        bool clip_distances_declared = false;
+
         declarations.AddLine("out gl_PerVertex {");
         ++declarations.scope;
         declarations.AddLine("vec4 gl_Position;");
         for (auto& o : fixed_pipeline_output_attributes_used) {
             if (o == Attribute::Index::PointSize)
                 declarations.AddLine("float gl_PointSize;");
+            if (!clip_distances_declared && (o == Attribute::Index::ClipDistances0123 ||
+                                             o == Attribute::Index::ClipDistances4567)) {
+                declarations.AddLine("float gl_ClipDistance[];");
+                clip_distances_declared = true;
+            }
         }
         --declarations.scope;
         declarations.AddLine("};");
@@ -845,7 +867,8 @@ private:
             // vertex shader, and what's the value of the fourth element when inside a Tess Eval
             // shader.
             ASSERT(stage == Maxwell3D::Regs::ShaderStage::Vertex);
-            return "vec4(0, 0, uintBitsToFloat(instance_id.x), uintBitsToFloat(gl_VertexID))";
+            // Config pack's first value is instance_id.
+            return "vec4(0, 0, uintBitsToFloat(config_pack[0]), uintBitsToFloat(gl_VertexID))";
         case Attribute::Index::FrontFacing:
             // TODO(Subv): Find out what the values are for the other elements.
             ASSERT(stage == Maxwell3D::Regs::ShaderStage::Fragment);
@@ -916,6 +939,10 @@ private:
             return "gl_PointSize";
         case Attribute::Index::Position:
             return "position";
+        case Attribute::Index::ClipDistances0123:
+        case Attribute::Index::ClipDistances4567: {
+            return "gl_ClipDistance";
+        }
         default:
             const u32 index{static_cast<u32>(attribute) -
                             static_cast<u32>(Attribute::Index::Attribute_0)};
@@ -1266,7 +1293,15 @@ private:
         regs.SetRegisterToInteger(dest, true, 0, result, 1, 1);
     }
 
-    void WriteTexsInstruction(const Instruction& instr, const std::string& texture) {
+    void WriteTexsInstruction(const Instruction& instr, const std::string& coord,
+                              const std::string& texture) {
+        // Add an extra scope and declare the texture coords inside to prevent
+        // overwriting them in case they are used as outputs of the texs instruction.
+        shader.AddLine('{');
+        ++shader.scope;
+        shader.AddLine(coord);
+        shader.AddLine("vec4 texture_tmp = " + texture + ';');
+
         // TEXS has two destination registers and a swizzle. The first two elements in the swizzle
         // go into gpr0+0 and gpr0+1, and the rest goes into gpr28+0 and gpr28+1
 
@@ -1278,17 +1313,19 @@ private:
 
             if (written_components < 2) {
                 // Write the first two swizzle components to gpr0 and gpr0+1
-                regs.SetRegisterToFloat(instr.gpr0, component, texture, 1, 4, false,
+                regs.SetRegisterToFloat(instr.gpr0, component, "texture_tmp", 1, 4, false,
                                         written_components % 2);
             } else {
                 ASSERT(instr.texs.HasTwoDestinations());
                 // Write the rest of the swizzle components to gpr28 and gpr28+1
-                regs.SetRegisterToFloat(instr.gpr28, component, texture, 1, 4, false,
+                regs.SetRegisterToFloat(instr.gpr28, component, "texture_tmp", 1, 4, false,
                                         written_components % 2);
             }
 
             ++written_components;
         }
+        --shader.scope;
+        shader.AddLine('}');
     }
 
     static u32 TextureCoordinates(Tegra::Shader::TextureType texture_type) {
@@ -1685,6 +1722,26 @@ private:
 
             break;
         }
+        case OpCode::Type::Bfi: {
+            UNIMPLEMENTED_IF(instr.generates_cc);
+
+            const auto [base, packed_shift] = [&]() -> std::tuple<std::string, std::string> {
+                switch (opcode->get().GetId()) {
+                case OpCode::Id::BFI_IMM_R:
+                    return {regs.GetRegisterAsInteger(instr.gpr39, 0, false),
+                            std::to_string(instr.alu.GetSignedImm20_20())};
+                default:
+                    UNREACHABLE();
+                }
+            }();
+            const std::string offset = '(' + packed_shift + " & 0xff)";
+            const std::string bits = "((" + packed_shift + " >> 8) & 0xff)";
+            const std::string insert = regs.GetRegisterAsInteger(instr.gpr8, 0, false);
+            regs.SetRegisterToInteger(
+                instr.gpr0, false, 0,
+                "bitfieldInsert(" + base + ", " + insert + ", " + offset + ", " + bits + ')', 1, 1);
+            break;
+        }
         case OpCode::Type::Shift: {
             std::string op_a = regs.GetRegisterAsInteger(instr.gpr8, 0, true);
             std::string op_b;
@@ -2510,61 +2567,83 @@ private:
                 const bool depth_compare =
                     instr.tex.UsesMiscMode(Tegra::Shader::TextureMiscMode::DC);
                 u32 num_coordinates = TextureCoordinates(texture_type);
-                if (depth_compare)
-                    num_coordinates += 1;
+                u32 start_index = 0;
+                std::string array_elem;
+                if (is_array) {
+                    array_elem = regs.GetRegisterAsInteger(instr.gpr8);
+                    start_index = 1;
+                }
+                const auto process_mode = instr.tex.GetTextureProcessMode();
+                u32 start_index_b = 0;
+                std::string lod_value;
+                if (process_mode != Tegra::Shader::TextureProcessMode::LZ &&
+                    process_mode != Tegra::Shader::TextureProcessMode::None) {
+                    start_index_b = 1;
+                    lod_value = regs.GetRegisterAsFloat(instr.gpr20);
+                }
+
+                std::string depth_value;
+                if (depth_compare) {
+                    depth_value = regs.GetRegisterAsFloat(instr.gpr20.Value() + start_index_b);
+                }
+
+                bool depth_compare_extra = false;
 
                 switch (num_coordinates) {
                 case 1: {
+                    const std::string x = regs.GetRegisterAsFloat(instr.gpr8.Value() + start_index);
                     if (is_array) {
-                        const std::string index = regs.GetRegisterAsInteger(instr.gpr8);
-                        const std::string x = regs.GetRegisterAsFloat(instr.gpr8.Value() + 1);
-                        coord = "vec2 coords = vec2(" + x + ", " + index + ");";
+                        if (depth_compare) {
+                            coord = "vec3 coords = vec3(" + x + ", " + depth_value + ", " +
+                                    array_elem + ");";
+                        } else {
+                            coord = "vec2 coords = vec2(" + x + ", " + array_elem + ");";
+                        }
                     } else {
-                        const std::string x = regs.GetRegisterAsFloat(instr.gpr8);
-                        coord = "float coords = " + x + ';';
+                        if (depth_compare) {
+                            coord = "vec2 coords = vec2(" + x + ", " + depth_value + ");";
+                        } else {
+                            coord = "float coords = " + x + ';';
+                        }
                     }
                     break;
                 }
                 case 2: {
+                    const std::string x = regs.GetRegisterAsFloat(instr.gpr8.Value() + start_index);
+                    const std::string y =
+                        regs.GetRegisterAsFloat(instr.gpr8.Value() + start_index + 1);
                     if (is_array) {
-                        const std::string index = regs.GetRegisterAsInteger(instr.gpr8);
-                        const std::string x = regs.GetRegisterAsFloat(instr.gpr8.Value() + 1);
-                        const std::string y = regs.GetRegisterAsFloat(instr.gpr8.Value() + 2);
-                        coord = "vec3 coords = vec3(" + x + ", " + y + ", " + index + ");";
+                        if (depth_compare) {
+                            coord = "vec4 coords = vec4(" + x + ", " + y + ", " + depth_value +
+                                    ", " + array_elem + ");";
+                        } else {
+                            coord = "vec3 coords = vec3(" + x + ", " + y + ", " + array_elem + ");";
+                        }
                     } else {
-                        const std::string x = regs.GetRegisterAsFloat(instr.gpr8);
-                        const std::string y = regs.GetRegisterAsFloat(instr.gpr8.Value() + 1);
-                        coord = "vec2 coords = vec2(" + x + ", " + y + ");";
+                        if (depth_compare) {
+                            coord =
+                                "vec3 coords = vec3(" + x + ", " + y + ", " + depth_value + ");";
+                        } else {
+                            coord = "vec2 coords = vec2(" + x + ", " + y + ");";
+                        }
                     }
                     break;
                 }
                 case 3: {
-                    if (depth_compare) {
-                        if (is_array) {
-                            const std::string index = regs.GetRegisterAsInteger(instr.gpr8);
-                            const std::string x = regs.GetRegisterAsFloat(instr.gpr8.Value() + 1);
-                            const std::string y = regs.GetRegisterAsFloat(instr.gpr20);
-                            const std::string z = regs.GetRegisterAsFloat(instr.gpr20.Value() + 1);
-                            coord = "vec4 coords = vec4(" + x + ", " + y + ", " + z + ", " + index +
-                                    ");";
-                        } else {
-                            const std::string x = regs.GetRegisterAsFloat(instr.gpr8);
-                            const std::string y = regs.GetRegisterAsFloat(instr.gpr8.Value() + 1);
-                            const std::string z = regs.GetRegisterAsFloat(instr.gpr20);
-                            coord = "vec3 coords = vec3(" + x + ", " + y + ", " + z + ");";
-                        }
+                    const std::string x = regs.GetRegisterAsFloat(instr.gpr8.Value() + start_index);
+                    const std::string y =
+                        regs.GetRegisterAsFloat(instr.gpr8.Value() + start_index + 1);
+                    const std::string z =
+                        regs.GetRegisterAsFloat(instr.gpr8.Value() + start_index + 2);
+                    if (is_array) {
+                        depth_compare_extra = depth_compare;
+                        coord = "vec4 coords = vec4(" + x + ", " + y + ", " + z + ", " +
+                                array_elem + ");";
                     } else {
-                        if (is_array) {
-                            const std::string index = regs.GetRegisterAsInteger(instr.gpr8);
-                            const std::string x = regs.GetRegisterAsFloat(instr.gpr8.Value() + 1);
-                            const std::string y = regs.GetRegisterAsFloat(instr.gpr8.Value() + 2);
-                            const std::string z = regs.GetRegisterAsFloat(instr.gpr8.Value() + 3);
-                            coord = "vec4 coords = vec4(" + x + ", " + y + ", " + z + ", " + index +
-                                    ");";
+                        if (depth_compare) {
+                            coord = "vec4 coords = vec4(" + x + ", " + y + ", " + z + ", " +
+                                    depth_value + ");";
                         } else {
-                            const std::string x = regs.GetRegisterAsFloat(instr.gpr8);
-                            const std::string y = regs.GetRegisterAsFloat(instr.gpr8.Value() + 1);
-                            const std::string z = regs.GetRegisterAsFloat(instr.gpr8.Value() + 2);
                             coord = "vec3 coords = vec3(" + x + ", " + y + ", " + z + ");";
                         }
                     }
@@ -2580,82 +2659,85 @@ private:
                     coord = "vec2 coords = vec2(" + x + ", " + y + ");";
                     texture_type = Tegra::Shader::TextureType::Texture2D;
                 }
-                // TODO: make sure coordinates are always indexed to gpr8 and gpr20 is always bias
-                // or lod.
 
                 const std::string sampler =
                     GetSampler(instr.sampler, texture_type, is_array, depth_compare);
                 // Add an extra scope and declare the texture coords inside to prevent
                 // overwriting them in case they are used as outputs of the texs instruction.
 
-                shader.AddLine("{");
+                shader.AddLine('{');
                 ++shader.scope;
                 shader.AddLine(coord);
                 std::string texture;
 
                 switch (instr.tex.GetTextureProcessMode()) {
                 case Tegra::Shader::TextureProcessMode::None: {
-                    texture = "texture(" + sampler + ", coords)";
+                    if (!depth_compare_extra) {
+                        texture = "texture(" + sampler + ", coords)";
+                    } else {
+                        texture = "texture(" + sampler + ", coords, " + depth_value + ')';
+                    }
                     break;
                 }
                 case Tegra::Shader::TextureProcessMode::LZ: {
-                    texture = "textureLod(" + sampler + ", coords, 0.0)";
+                    if (!depth_compare_extra) {
+                        texture = "textureLod(" + sampler + ", coords, 0.0)";
+                    } else {
+                        texture = "texture(" + sampler + ", coords, " + depth_value + ')';
+                    }
                     break;
                 }
                 case Tegra::Shader::TextureProcessMode::LB:
                 case Tegra::Shader::TextureProcessMode::LBA: {
-                    const std::string bias = [&]() {
-                        if (depth_compare) {
-                            if (is_array)
-                                return regs.GetRegisterAsFloat(instr.gpr20.Value() + 2);
-                            else
-                                return regs.GetRegisterAsFloat(instr.gpr20.Value() + 1);
-                        } else {
-                            return regs.GetRegisterAsFloat(instr.gpr20);
-                        }
-                    }();
-                    shader.AddLine("float bias = " + bias + ';');
-
                     // TODO: Figure if A suffix changes the equation at all.
-                    texture = "texture(" + sampler + ", coords, bias)";
+                    if (!depth_compare_extra) {
+                        texture = "texture(" + sampler + ", coords, " + lod_value + ')';
+                    } else {
+                        texture = "texture(" + sampler + ", coords, " + depth_value + ')';
+                        LOG_WARNING(HW_GPU,
+                                    "OpenGL Limitation: can't set bias value along depth compare");
+                    }
                     break;
                 }
                 case Tegra::Shader::TextureProcessMode::LL:
                 case Tegra::Shader::TextureProcessMode::LLA: {
-                    const std::string lod = [&]() {
-                        if (num_coordinates <= 2) {
-                            return regs.GetRegisterAsFloat(instr.gpr20);
-                        } else {
-                            return regs.GetRegisterAsFloat(instr.gpr20.Value() + 1);
-                        }
-                    }();
-                    shader.AddLine("float lod = " + lod + ';');
-
                     // TODO: Figure if A suffix changes the equation at all.
-                    texture = "textureLod(" + sampler + ", coords, lod)";
+                    if (!depth_compare_extra) {
+                        texture = "textureLod(" + sampler + ", coords, " + lod_value + ')';
+                    } else {
+                        texture = "texture(" + sampler + ", coords, " + depth_value + ')';
+                        LOG_WARNING(HW_GPU,
+                                    "OpenGL Limitation: can't set lod value along depth compare");
+                    }
                     break;
                 }
                 default: {
-                    texture = "texture(" + sampler + ", coords)";
+                    if (!depth_compare_extra) {
+                        texture = "texture(" + sampler + ", coords)";
+                    } else {
+                        texture = "texture(" + sampler + ", coords, " + depth_value + ')';
+                    }
                     UNIMPLEMENTED_MSG("Unhandled texture process mode {}",
                                       static_cast<u32>(instr.tex.GetTextureProcessMode()));
                 }
                 }
                 if (!depth_compare) {
+                    shader.AddLine("vec4 texture_tmp = " + texture + ';');
                     std::size_t dest_elem{};
                     for (std::size_t elem = 0; elem < 4; ++elem) {
                         if (!instr.tex.IsComponentEnabled(elem)) {
                             // Skip disabled components
                             continue;
                         }
-                        regs.SetRegisterToFloat(instr.gpr0, elem, texture, 1, 4, false, dest_elem);
+                        regs.SetRegisterToFloat(instr.gpr0, elem, "texture_tmp", 1, 4, false,
+                                                dest_elem);
                         ++dest_elem;
                     }
                 } else {
                     regs.SetRegisterToFloat(instr.gpr0, 0, texture, 1, 1, false);
                 }
                 --shader.scope;
-                shader.AddLine("}");
+                shader.AddLine('}');
                 break;
             }
             case OpCode::Id::TEXS: {
@@ -2668,41 +2750,76 @@ private:
                 const bool depth_compare =
                     instr.texs.UsesMiscMode(Tegra::Shader::TextureMiscMode::DC);
                 u32 num_coordinates = TextureCoordinates(texture_type);
-                if (depth_compare)
-                    num_coordinates += 1;
-
-                // Scope to avoid variable name overlaps.
-                shader.AddLine('{');
-                ++shader.scope;
+                const auto process_mode = instr.texs.GetTextureProcessMode();
+                std::string lod_value;
+                std::string coord;
+                u32 lod_offset = 0;
+                if (process_mode == Tegra::Shader::TextureProcessMode::LL) {
+                    if (num_coordinates > 2) {
+                        lod_value = regs.GetRegisterAsFloat(instr.gpr20.Value() + 1);
+                        lod_offset = 2;
+                    } else {
+                        lod_value = regs.GetRegisterAsFloat(instr.gpr20);
+                        lod_offset = 1;
+                    }
+                }
 
                 switch (num_coordinates) {
+                case 1: {
+                    coord = "float coords = " + regs.GetRegisterAsFloat(instr.gpr8) + ';';
+                    break;
+                }
                 case 2: {
                     if (is_array) {
-                        const std::string index = regs.GetRegisterAsInteger(instr.gpr8);
-                        const std::string x = regs.GetRegisterAsFloat(instr.gpr8.Value() + 1);
-                        const std::string y = regs.GetRegisterAsFloat(instr.gpr20);
-                        shader.AddLine("vec3 coords = vec3(" + x + ", " + y + ", " + index + ");");
+                        if (depth_compare) {
+                            const std::string index = regs.GetRegisterAsInteger(instr.gpr8);
+                            const std::string x = regs.GetRegisterAsFloat(instr.gpr8.Value() + 1);
+                            const std::string y = regs.GetRegisterAsFloat(instr.gpr20);
+                            const std::string z = regs.GetRegisterAsFloat(instr.gpr20.Value() + 1);
+                            coord = "vec4 coords = vec4(" + x + ", " + y + ", " + z + ", " + index +
+                                    ");";
+                        } else {
+                            const std::string index = regs.GetRegisterAsInteger(instr.gpr8);
+                            const std::string x = regs.GetRegisterAsFloat(instr.gpr8.Value() + 1);
+                            const std::string y = regs.GetRegisterAsFloat(instr.gpr20);
+                            coord = "vec3 coords = vec3(" + x + ", " + y + ", " + index + ");";
+                        }
                     } else {
-                        const std::string x = regs.GetRegisterAsFloat(instr.gpr8);
-                        const std::string y = regs.GetRegisterAsFloat(instr.gpr20);
-                        shader.AddLine("vec2 coords = vec2(" + x + ", " + y + ");");
+                        if (lod_offset != 0) {
+                            if (depth_compare) {
+                                const std::string x = regs.GetRegisterAsFloat(instr.gpr8);
+                                const std::string y =
+                                    regs.GetRegisterAsFloat(instr.gpr8.Value() + 1);
+                                const std::string z =
+                                    regs.GetRegisterAsFloat(instr.gpr20.Value() + lod_offset);
+                                coord = "vec3 coords = vec3(" + x + ", " + y + ", " + z + ");";
+                            } else {
+                                const std::string x = regs.GetRegisterAsFloat(instr.gpr8);
+                                const std::string y =
+                                    regs.GetRegisterAsFloat(instr.gpr8.Value() + 1);
+                                coord = "vec2 coords = vec2(" + x + ", " + y + ");";
+                            }
+                        } else {
+                            if (depth_compare) {
+                                const std::string x = regs.GetRegisterAsFloat(instr.gpr8);
+                                const std::string y =
+                                    regs.GetRegisterAsFloat(instr.gpr8.Value() + 1);
+                                const std::string z = regs.GetRegisterAsFloat(instr.gpr20);
+                                coord = "vec3 coords = vec3(" + x + ", " + y + ", " + z + ");";
+                            } else {
+                                const std::string x = regs.GetRegisterAsFloat(instr.gpr8);
+                                const std::string y = regs.GetRegisterAsFloat(instr.gpr20);
+                                coord = "vec2 coords = vec2(" + x + ", " + y + ");";
+                            }
+                        }
                     }
                     break;
                 }
                 case 3: {
-                    if (is_array) {
-                        const std::string index = regs.GetRegisterAsInteger(instr.gpr8);
-                        const std::string x = regs.GetRegisterAsFloat(instr.gpr8.Value() + 1);
-                        const std::string y = regs.GetRegisterAsFloat(instr.gpr8.Value() + 2);
-                        const std::string z = regs.GetRegisterAsFloat(instr.gpr20);
-                        shader.AddLine("vec4 coords = vec4(" + x + ", " + y + ", " + z + ", " +
-                                       index + ");");
-                    } else {
-                        const std::string x = regs.GetRegisterAsFloat(instr.gpr8);
-                        const std::string y = regs.GetRegisterAsFloat(instr.gpr8.Value() + 1);
-                        const std::string z = regs.GetRegisterAsFloat(instr.gpr20);
-                        shader.AddLine("vec3 coords = vec3(" + x + ", " + y + ", " + z + ");");
-                    }
+                    const std::string x = regs.GetRegisterAsFloat(instr.gpr8);
+                    const std::string y = regs.GetRegisterAsFloat(instr.gpr8.Value() + 1);
+                    const std::string z = regs.GetRegisterAsFloat(instr.gpr20);
+                    coord = "vec3 coords = vec3(" + x + ", " + y + ", " + z + ");";
                     break;
                 }
                 default:
@@ -2712,14 +2829,14 @@ private:
                     // Fallback to interpreting as a 2D texture for now
                     const std::string x = regs.GetRegisterAsFloat(instr.gpr8);
                     const std::string y = regs.GetRegisterAsFloat(instr.gpr20);
-                    shader.AddLine("vec2 coords = vec2(" + x + ", " + y + ");");
+                    coord = "vec2 coords = vec2(" + x + ", " + y + ");";
                     texture_type = Tegra::Shader::TextureType::Texture2D;
                     is_array = false;
                 }
                 const std::string sampler =
                     GetSampler(instr.sampler, texture_type, is_array, depth_compare);
                 std::string texture;
-                switch (instr.texs.GetTextureProcessMode()) {
+                switch (process_mode) {
                 case Tegra::Shader::TextureProcessMode::None: {
                     texture = "texture(" + sampler + ", coords)";
                     break;
@@ -2733,8 +2850,7 @@ private:
                     break;
                 }
                 case Tegra::Shader::TextureProcessMode::LL: {
-                    const std::string op_c = regs.GetRegisterAsFloat(instr.gpr20.Value() + 1);
-                    texture = "textureLod(" + sampler + ", coords, " + op_c + ')';
+                    texture = "textureLod(" + sampler + ", coords, " + lod_value + ')';
                     break;
                 }
                 default: {
@@ -2744,13 +2860,11 @@ private:
                 }
                 }
                 if (!depth_compare) {
-                    WriteTexsInstruction(instr, texture);
+                    WriteTexsInstruction(instr, coord, texture);
                 } else {
-                    WriteTexsInstruction(instr, "vec4(" + texture + ')');
+                    WriteTexsInstruction(instr, coord, "vec4(" + texture + ')');
                 }
 
-                shader.AddLine('}');
-                --shader.scope;
                 break;
             }
             case OpCode::Id::TLDS: {
@@ -2772,11 +2886,12 @@ private:
                 // Scope to avoid variable name overlaps.
                 shader.AddLine('{');
                 ++shader.scope;
+                std::string coords;
 
                 switch (texture_type) {
                 case Tegra::Shader::TextureType::Texture1D: {
                     const std::string x = regs.GetRegisterAsInteger(instr.gpr8);
-                    shader.AddLine("int coords = " + x + ';');
+                    coords = "float coords = " + x + ';';
                     break;
                 }
                 case Tegra::Shader::TextureType::Texture2D: {
@@ -2784,7 +2899,8 @@ private:
 
                     const std::string x = regs.GetRegisterAsInteger(instr.gpr8);
                     const std::string y = regs.GetRegisterAsInteger(instr.gpr20);
-                    shader.AddLine("ivec2 coords = ivec2(" + x + ", " + y + ");");
+                    // shader.AddLine("ivec2 coords = ivec2(" + x + ", " + y + ");");
+                    coords = "ivec2 coords = ivec2(" + x + ", " + y + ");";
                     extra_op_offset = 1;
                     break;
                 }
@@ -2812,7 +2928,7 @@ private:
                                       static_cast<u32>(instr.tlds.GetTextureProcessMode()));
                 }
                 }
-                WriteTexsInstruction(instr, texture);
+                WriteTexsInstruction(instr, coords, texture);
 
                 --shader.scope;
                 shader.AddLine('}');
@@ -2871,14 +2987,17 @@ private:
 
                 const std::string texture = "textureGather(" + sampler + ", coords, " +
                                             std::to_string(instr.tld4.component) + ')';
+
                 if (!depth_compare) {
+                    shader.AddLine("vec4 texture_tmp = " + texture + ';');
                     std::size_t dest_elem{};
                     for (std::size_t elem = 0; elem < 4; ++elem) {
                         if (!instr.tex.IsComponentEnabled(elem)) {
                             // Skip disabled components
                             continue;
                         }
-                        regs.SetRegisterToFloat(instr.gpr0, elem, texture, 1, 4, false, dest_elem);
+                        regs.SetRegisterToFloat(instr.gpr0, elem, "texture_tmp", 1, 4, false,
+                                                dest_elem);
                         ++dest_elem;
                     }
                 } else {
@@ -2899,6 +3018,7 @@ private:
                 // Scope to avoid variable name overlaps.
                 shader.AddLine('{');
                 ++shader.scope;
+                std::string coords;
 
                 const bool depth_compare =
                     instr.tld4s.UsesMiscMode(Tegra::Shader::TextureMiscMode::DC);
@@ -2908,20 +3028,19 @@ private:
                 const std::string sampler = GetSampler(
                     instr.sampler, Tegra::Shader::TextureType::Texture2D, false, depth_compare);
                 if (!depth_compare) {
-                    shader.AddLine("vec2 coords = vec2(" + op_a + ", " + op_b + ");");
+                    coords = "vec2 coords = vec2(" + op_a + ", " + op_b + ");";
                 } else {
                     // Note: TLD4S coordinate encoding works just like TEXS's
-                    shader.AddLine(
-                        "float op_y = " + regs.GetRegisterAsFloat(instr.gpr8.Value() + 1) + ';');
-                    shader.AddLine("vec3 coords = vec3(" + op_a + ", op_y, " + op_b + ");");
+                    const std::string op_y = regs.GetRegisterAsFloat(instr.gpr8.Value() + 1);
+                    coords = "vec3 coords = vec3(" + op_a + ", " + op_y + ", " + op_b + ");";
                 }
                 const std::string texture = "textureGather(" + sampler + ", coords, " +
                                             std::to_string(instr.tld4s.component) + ')';
 
                 if (!depth_compare) {
-                    WriteTexsInstruction(instr, texture);
+                    WriteTexsInstruction(instr, coords, texture);
                 } else {
-                    WriteTexsInstruction(instr, "vec4(" + texture + ')');
+                    WriteTexsInstruction(instr, coords, "vec4(" + texture + ')');
                 }
 
                 --shader.scope;
@@ -3217,6 +3336,34 @@ private:
             }
             break;
         }
+        case OpCode::Type::RegisterSetPredicate: {
+            UNIMPLEMENTED_IF(instr.r2p.mode != Tegra::Shader::R2pMode::Pr);
+
+            const std::string apply_mask = [&]() {
+                switch (opcode->get().GetId()) {
+                case OpCode::Id::R2P_IMM:
+                    return std::to_string(instr.r2p.immediate_mask);
+                default:
+                    UNREACHABLE();
+                }
+            }();
+            const std::string mask = '(' + regs.GetRegisterAsInteger(instr.gpr8, 0, false) +
+                                     " >> " + std::to_string(instr.r2p.byte) + ')';
+
+            constexpr u64 programmable_preds = 7;
+            for (u64 pred = 0; pred < programmable_preds; ++pred) {
+                const auto shift = std::to_string(1 << pred);
+
+                shader.AddLine("if ((" + apply_mask + " & " + shift + ") != 0) {");
+                ++shader.scope;
+
+                SetPredicate(pred, '(' + mask + " & " + shift + ") != 0");
+
+                --shader.scope;
+                shader.AddLine('}');
+            }
+            break;
+        }
         case OpCode::Type::FloatSet: {
             const std::string op_a = GetOperandAbsNeg(regs.GetRegisterAsFloat(instr.gpr8),
                                                       instr.fset.abs_a != 0, instr.fset.neg_a != 0);
@@ -3254,6 +3401,10 @@ private:
                 regs.SetRegisterToInteger(instr.gpr0, false, 0, predicate + " ? 0xFFFFFFFF : 0", 1,
                                           1);
             }
+            if (instr.generates_cc.Value() != 0) {
+                regs.SetInternalFlag(InternalFlag::ZeroFlag, predicate);
+                LOG_WARNING(HW_GPU, "FSET Condition Code is incomplete");
+            }
             break;
         }
         case OpCode::Type::IntegerSet: {
@@ -3507,6 +3658,11 @@ private:
                     regs.SetRegisterToInteger(instr.gpr0, false, 0, "0u", 1, 1);
                     break;
                 }
+                case Tegra::Shader::SystemVariable::Ydirection: {
+                    // Config pack's third value is Y_NEGATE's state.
+                    regs.SetRegisterToFloat(instr.gpr0, 0, "uintBitsToFloat(config_pack[2])", 1, 1);
+                    break;
+                }
                 default: {
                     UNIMPLEMENTED_MSG("Unhandled system move: {}",
                                       static_cast<u32>(instr.sys20.Value()));
@@ -3530,11 +3686,17 @@ private:
                                      "BRA with constant buffers are not implemented");
 
                 const Tegra::Shader::ConditionCode cc = instr.flow_condition_code;
-                UNIMPLEMENTED_IF_MSG(cc != Tegra::Shader::ConditionCode::T,
-                                     "BRA condition code used: {}", static_cast<u32>(cc));
-
                 const u32 target = offset + instr.bra.GetBranchTarget();
-                shader.AddLine("{ jmp_to = " + std::to_string(target) + "u; break; }");
+                if (cc != Tegra::Shader::ConditionCode::T) {
+                    const std::string condition_code = regs.GetConditionCode(cc);
+                    shader.AddLine("if (" + condition_code + "){");
+                    shader.scope++;
+                    shader.AddLine("{ jmp_to = " + std::to_string(target) + "u; break; }");
+                    shader.scope--;
+                    shader.AddLine('}');
+                } else {
+                    shader.AddLine("{ jmp_to = " + std::to_string(target) + "u; break; }");
+                }
                 break;
             }
             case OpCode::Id::IPA: {

src/video_core/renderer_opengl/gl_shader_gen.cpp

@@ -24,8 +24,7 @@ layout (location = 0) out vec4 position;
 
 layout(std140) uniform vs_config {
     vec4 viewport_flip;
-    uvec4 instance_id;
-    uvec4 flip_stage;
+    uvec4 config_pack; // instance_id, flip_stage, y_direction, padding
     uvec4 alpha_test;
 };
 )";
@@ -63,7 +62,8 @@ void main() {
     out += R"(
 
     // Check if the flip stage is VertexB
-    if (flip_stage[0] == 1) {
+    // Config pack's second value is flip_stage
+    if (config_pack[1] == 1) {
         // Viewport can be flipped, which is unsupported by glViewport
         position.xy *= viewport_flip.xy;
     }
@@ -71,7 +71,7 @@ void main() {
 
     // TODO(bunnei): This is likely a hack, position.w should be interpolated as 1.0
     // For now, this is here to bring order in lieu of proper emulation
-    if (flip_stage[0] == 1) {
+    if (config_pack[1] == 1) {
         position.w = 1.0;
     }
 }
@@ -101,8 +101,7 @@ layout (location = 0) out vec4 position;
 
 layout (std140) uniform gs_config {
     vec4 viewport_flip;
-    uvec4 instance_id;
-    uvec4 flip_stage;
+    uvec4 config_pack; // instance_id, flip_stage, y_direction, padding
     uvec4 alpha_test;
 };
 
@@ -139,8 +138,7 @@ layout (location = 0) in vec4 position;
 
 layout (std140) uniform fs_config {
     vec4 viewport_flip;
-    uvec4 instance_id;
-    uvec4 flip_stage;
+    uvec4 config_pack; // instance_id, flip_stage, y_direction, padding
     uvec4 alpha_test;
 };
 

src/video_core/renderer_opengl/gl_shader_manager.cpp

@@ -27,16 +27,18 @@ void MaxwellUniformData::SetFromRegs(const Maxwell3D::State::ShaderStageInfo& sh
     alpha_test.func = func;
     alpha_test.ref = regs.alpha_test_ref;
 
-    // We only assign the instance to the first component of the vector, the rest is just padding.
-    instance_id[0] = state.current_instance;
+    instance_id = state.current_instance;
 
     // Assign in which stage the position has to be flipped
     // (the last stage before the fragment shader).
     if (gpu.regs.shader_config[static_cast<u32>(Maxwell3D::Regs::ShaderProgram::Geometry)].enable) {
-        flip_stage[0] = static_cast<u32>(Maxwell3D::Regs::ShaderProgram::Geometry);
+        flip_stage = static_cast<u32>(Maxwell3D::Regs::ShaderProgram::Geometry);
     } else {
-        flip_stage[0] = static_cast<u32>(Maxwell3D::Regs::ShaderProgram::VertexB);
+        flip_stage = static_cast<u32>(Maxwell3D::Regs::ShaderProgram::VertexB);
     }
+
+    // Y_NEGATE controls what value S2R returns for the Y_DIRECTION system value.
+    y_direction = regs.screen_y_control.y_negate == 0 ? 1.f : -1.f;
 }
 
 } // namespace OpenGL::GLShader

src/video_core/renderer_opengl/gl_shader_manager.h

@@ -21,8 +21,11 @@ using Tegra::Engines::Maxwell3D;
 struct MaxwellUniformData {
     void SetFromRegs(const Maxwell3D::State::ShaderStageInfo& shader_stage);
     alignas(16) GLvec4 viewport_flip;
-    alignas(16) GLuvec4 instance_id;
-    alignas(16) GLuvec4 flip_stage;
+    struct alignas(16) {
+        GLuint instance_id;
+        GLuint flip_stage;
+        GLfloat y_direction;
+    };
     struct alignas(16) {
         GLuint enabled;
         GLuint func;
@@ -30,7 +33,7 @@ struct MaxwellUniformData {
         GLuint padding;
     } alpha_test;
 };
-static_assert(sizeof(MaxwellUniformData) == 64, "MaxwellUniformData structure size is incorrect");
+static_assert(sizeof(MaxwellUniformData) == 48, "MaxwellUniformData structure size is incorrect");
 static_assert(sizeof(MaxwellUniformData) < 16384,
               "MaxwellUniformData structure must be less than 16kb as per the OpenGL spec");