diff options
Diffstat (limited to 'src')
56 files changed, 457 insertions, 414 deletions
diff --git a/src/core/hle/kernel/handle_table.cpp b/src/core/hle/kernel/handle_table.cpp index e441a27fc..35448b576 100644 --- a/src/core/hle/kernel/handle_table.cpp +++ b/src/core/hle/kernel/handle_table.cpp | |||
| @@ -30,6 +30,7 @@ HandleTable::~HandleTable() = default; | |||
| 30 | 30 | ||
| 31 | ResultCode HandleTable::SetSize(s32 handle_table_size) { | 31 | ResultCode HandleTable::SetSize(s32 handle_table_size) { |
| 32 | if (static_cast<u32>(handle_table_size) > MAX_COUNT) { | 32 | if (static_cast<u32>(handle_table_size) > MAX_COUNT) { |
| 33 | LOG_ERROR(Kernel, "Handle table size {} is greater than {}", handle_table_size, MAX_COUNT); | ||
| 33 | return ERR_OUT_OF_MEMORY; | 34 | return ERR_OUT_OF_MEMORY; |
| 34 | } | 35 | } |
| 35 | 36 | ||
| @@ -80,6 +81,7 @@ ResultVal<Handle> HandleTable::Duplicate(Handle handle) { | |||
| 80 | 81 | ||
| 81 | ResultCode HandleTable::Close(Handle handle) { | 82 | ResultCode HandleTable::Close(Handle handle) { |
| 82 | if (!IsValid(handle)) { | 83 | if (!IsValid(handle)) { |
| 84 | LOG_ERROR(Kernel, "Handle is not valid! handle={:08X}", handle); | ||
| 83 | return ERR_INVALID_HANDLE; | 85 | return ERR_INVALID_HANDLE; |
| 84 | } | 86 | } |
| 85 | 87 | ||
diff --git a/src/core/hle/kernel/mutex.cpp b/src/core/hle/kernel/mutex.cpp index eff4e45b0..7869eb32b 100644 --- a/src/core/hle/kernel/mutex.cpp +++ b/src/core/hle/kernel/mutex.cpp | |||
| @@ -7,6 +7,7 @@ | |||
| 7 | #include <vector> | 7 | #include <vector> |
| 8 | 8 | ||
| 9 | #include "common/assert.h" | 9 | #include "common/assert.h" |
| 10 | #include "common/logging/log.h" | ||
| 10 | #include "core/core.h" | 11 | #include "core/core.h" |
| 11 | #include "core/hle/kernel/errors.h" | 12 | #include "core/hle/kernel/errors.h" |
| 12 | #include "core/hle/kernel/handle_table.h" | 13 | #include "core/hle/kernel/handle_table.h" |
| @@ -67,6 +68,7 @@ ResultCode Mutex::TryAcquire(VAddr address, Handle holding_thread_handle, | |||
| 67 | Handle requesting_thread_handle) { | 68 | Handle requesting_thread_handle) { |
| 68 | // The mutex address must be 4-byte aligned | 69 | // The mutex address must be 4-byte aligned |
| 69 | if ((address % sizeof(u32)) != 0) { | 70 | if ((address % sizeof(u32)) != 0) { |
| 71 | LOG_ERROR(Kernel, "Address is not 4-byte aligned! address={:016X}", address); | ||
| 70 | return ERR_INVALID_ADDRESS; | 72 | return ERR_INVALID_ADDRESS; |
| 71 | } | 73 | } |
| 72 | 74 | ||
| @@ -88,6 +90,8 @@ ResultCode Mutex::TryAcquire(VAddr address, Handle holding_thread_handle, | |||
| 88 | } | 90 | } |
| 89 | 91 | ||
| 90 | if (holding_thread == nullptr) { | 92 | if (holding_thread == nullptr) { |
| 93 | LOG_ERROR(Kernel, "Holding thread does not exist! thread_handle={:08X}", | ||
| 94 | holding_thread_handle); | ||
| 91 | return ERR_INVALID_HANDLE; | 95 | return ERR_INVALID_HANDLE; |
| 92 | } | 96 | } |
| 93 | 97 | ||
| @@ -109,6 +113,7 @@ ResultCode Mutex::TryAcquire(VAddr address, Handle holding_thread_handle, | |||
| 109 | ResultCode Mutex::Release(VAddr address) { | 113 | ResultCode Mutex::Release(VAddr address) { |
| 110 | // The mutex address must be 4-byte aligned | 114 | // The mutex address must be 4-byte aligned |
| 111 | if ((address % sizeof(u32)) != 0) { | 115 | if ((address % sizeof(u32)) != 0) { |
| 116 | LOG_ERROR(Kernel, "Address is not 4-byte aligned! address={:016X}", address); | ||
| 112 | return ERR_INVALID_ADDRESS; | 117 | return ERR_INVALID_ADDRESS; |
| 113 | } | 118 | } |
| 114 | 119 | ||
diff --git a/src/core/hle/kernel/process_capability.cpp b/src/core/hle/kernel/process_capability.cpp index 48e5ae682..63880f13d 100644 --- a/src/core/hle/kernel/process_capability.cpp +++ b/src/core/hle/kernel/process_capability.cpp | |||
| @@ -3,6 +3,7 @@ | |||
| 3 | // Refer to the license.txt file included. | 3 | // Refer to the license.txt file included. |
| 4 | 4 | ||
| 5 | #include "common/bit_util.h" | 5 | #include "common/bit_util.h" |
| 6 | #include "common/logging/log.h" | ||
| 6 | #include "core/hle/kernel/errors.h" | 7 | #include "core/hle/kernel/errors.h" |
| 7 | #include "core/hle/kernel/handle_table.h" | 8 | #include "core/hle/kernel/handle_table.h" |
| 8 | #include "core/hle/kernel/memory/page_table.h" | 9 | #include "core/hle/kernel/memory/page_table.h" |
| @@ -119,22 +120,30 @@ ResultCode ProcessCapabilities::ParseCapabilities(const u32* capabilities, | |||
| 119 | // The MapPhysical type uses two descriptor flags for its parameters. | 120 | // The MapPhysical type uses two descriptor flags for its parameters. |
| 120 | // If there's only one, then there's a problem. | 121 | // If there's only one, then there's a problem. |
| 121 | if (i >= num_capabilities) { | 122 | if (i >= num_capabilities) { |
| 123 | LOG_ERROR(Kernel, "Invalid combination! i={}", i); | ||
| 122 | return ERR_INVALID_COMBINATION; | 124 | return ERR_INVALID_COMBINATION; |
| 123 | } | 125 | } |
| 124 | 126 | ||
| 125 | const auto size_flags = capabilities[i]; | 127 | const auto size_flags = capabilities[i]; |
| 126 | if (GetCapabilityType(size_flags) != CapabilityType::MapPhysical) { | 128 | if (GetCapabilityType(size_flags) != CapabilityType::MapPhysical) { |
| 129 | LOG_ERROR(Kernel, "Invalid capability type! size_flags={}", size_flags); | ||
| 127 | return ERR_INVALID_COMBINATION; | 130 | return ERR_INVALID_COMBINATION; |
| 128 | } | 131 | } |
| 129 | 132 | ||
| 130 | const auto result = HandleMapPhysicalFlags(descriptor, size_flags, page_table); | 133 | const auto result = HandleMapPhysicalFlags(descriptor, size_flags, page_table); |
| 131 | if (result.IsError()) { | 134 | if (result.IsError()) { |
| 135 | LOG_ERROR(Kernel, "Failed to map physical flags! descriptor={}, size_flags={}", | ||
| 136 | descriptor, size_flags); | ||
| 132 | return result; | 137 | return result; |
| 133 | } | 138 | } |
| 134 | } else { | 139 | } else { |
| 135 | const auto result = | 140 | const auto result = |
| 136 | ParseSingleFlagCapability(set_flags, set_svc_bits, descriptor, page_table); | 141 | ParseSingleFlagCapability(set_flags, set_svc_bits, descriptor, page_table); |
| 137 | if (result.IsError()) { | 142 | if (result.IsError()) { |
| 143 | LOG_ERROR( | ||
| 144 | Kernel, | ||
| 145 | "Failed to parse capability flag! set_flags={}, set_svc_bits={}, descriptor={}", | ||
| 146 | set_flags, set_svc_bits, descriptor); | ||
| 138 | return result; | 147 | return result; |
| 139 | } | 148 | } |
| 140 | } | 149 | } |
| @@ -162,6 +171,9 @@ ResultCode ProcessCapabilities::ParseSingleFlagCapability(u32& set_flags, u32& s | |||
| 162 | const u32 flag_length = GetFlagBitOffset(type); | 171 | const u32 flag_length = GetFlagBitOffset(type); |
| 163 | const u32 set_flag = 1U << flag_length; | 172 | const u32 set_flag = 1U << flag_length; |
| 164 | if ((set_flag & set_flags & InitializeOnceMask) != 0) { | 173 | if ((set_flag & set_flags & InitializeOnceMask) != 0) { |
| 174 | LOG_ERROR(Kernel, | ||
| 175 | "Attempted to initialize flags that may only be initialized once. set_flags={}", | ||
| 176 | set_flags); | ||
| 165 | return ERR_INVALID_COMBINATION; | 177 | return ERR_INVALID_COMBINATION; |
| 166 | } | 178 | } |
| 167 | set_flags |= set_flag; | 179 | set_flags |= set_flag; |
| @@ -187,6 +199,7 @@ ResultCode ProcessCapabilities::ParseSingleFlagCapability(u32& set_flags, u32& s | |||
| 187 | break; | 199 | break; |
| 188 | } | 200 | } |
| 189 | 201 | ||
| 202 | LOG_ERROR(Kernel, "Invalid capability type! type={}", static_cast<u32>(type)); | ||
| 190 | return ERR_INVALID_CAPABILITY_DESCRIPTOR; | 203 | return ERR_INVALID_CAPABILITY_DESCRIPTOR; |
| 191 | } | 204 | } |
| 192 | 205 | ||
| @@ -208,23 +221,31 @@ void ProcessCapabilities::Clear() { | |||
| 208 | 221 | ||
| 209 | ResultCode ProcessCapabilities::HandlePriorityCoreNumFlags(u32 flags) { | 222 | ResultCode ProcessCapabilities::HandlePriorityCoreNumFlags(u32 flags) { |
| 210 | if (priority_mask != 0 || core_mask != 0) { | 223 | if (priority_mask != 0 || core_mask != 0) { |
| 224 | LOG_ERROR(Kernel, "Core or priority mask are not zero! priority_mask={}, core_mask={}", | ||
| 225 | priority_mask, core_mask); | ||
| 211 | return ERR_INVALID_CAPABILITY_DESCRIPTOR; | 226 | return ERR_INVALID_CAPABILITY_DESCRIPTOR; |
| 212 | } | 227 | } |
| 213 | 228 | ||
| 214 | const u32 core_num_min = (flags >> 16) & 0xFF; | 229 | const u32 core_num_min = (flags >> 16) & 0xFF; |
| 215 | const u32 core_num_max = (flags >> 24) & 0xFF; | 230 | const u32 core_num_max = (flags >> 24) & 0xFF; |
| 216 | if (core_num_min > core_num_max) { | 231 | if (core_num_min > core_num_max) { |
| 232 | LOG_ERROR(Kernel, "Core min is greater than core max! core_num_min={}, core_num_max={}", | ||
| 233 | core_num_min, core_num_max); | ||
| 217 | return ERR_INVALID_COMBINATION; | 234 | return ERR_INVALID_COMBINATION; |
| 218 | } | 235 | } |
| 219 | 236 | ||
| 220 | const u32 priority_min = (flags >> 10) & 0x3F; | 237 | const u32 priority_min = (flags >> 10) & 0x3F; |
| 221 | const u32 priority_max = (flags >> 4) & 0x3F; | 238 | const u32 priority_max = (flags >> 4) & 0x3F; |
| 222 | if (priority_min > priority_max) { | 239 | if (priority_min > priority_max) { |
| 240 | LOG_ERROR(Kernel, | ||
| 241 | "Priority min is greater than priority max! priority_min={}, priority_max={}", | ||
| 242 | core_num_min, priority_max); | ||
| 223 | return ERR_INVALID_COMBINATION; | 243 | return ERR_INVALID_COMBINATION; |
| 224 | } | 244 | } |
| 225 | 245 | ||
| 226 | // The switch only has 4 usable cores. | 246 | // The switch only has 4 usable cores. |
| 227 | if (core_num_max >= 4) { | 247 | if (core_num_max >= 4) { |
| 248 | LOG_ERROR(Kernel, "Invalid max cores specified! core_num_max={}", core_num_max); | ||
| 228 | return ERR_INVALID_PROCESSOR_ID; | 249 | return ERR_INVALID_PROCESSOR_ID; |
| 229 | } | 250 | } |
| 230 | 251 | ||
| @@ -259,6 +280,7 @@ ResultCode ProcessCapabilities::HandleSyscallFlags(u32& set_svc_bits, u32 flags) | |||
| 259 | } | 280 | } |
| 260 | 281 | ||
| 261 | if (svc_number >= svc_capabilities.size()) { | 282 | if (svc_number >= svc_capabilities.size()) { |
| 283 | LOG_ERROR(Kernel, "Process svc capability is out of range! svc_number={}", svc_number); | ||
| 262 | return ERR_OUT_OF_RANGE; | 284 | return ERR_OUT_OF_RANGE; |
| 263 | } | 285 | } |
| 264 | 286 | ||
| @@ -295,6 +317,8 @@ ResultCode ProcessCapabilities::HandleInterruptFlags(u32 flags) { | |||
| 295 | // emulate that, it's sufficient to mark every interrupt as defined. | 317 | // emulate that, it's sufficient to mark every interrupt as defined. |
| 296 | 318 | ||
| 297 | if (interrupt >= interrupt_capabilities.size()) { | 319 | if (interrupt >= interrupt_capabilities.size()) { |
| 320 | LOG_ERROR(Kernel, "Process interrupt capability is out of range! svc_number={}", | ||
| 321 | interrupt); | ||
| 298 | return ERR_OUT_OF_RANGE; | 322 | return ERR_OUT_OF_RANGE; |
| 299 | } | 323 | } |
| 300 | 324 | ||
| @@ -307,6 +331,7 @@ ResultCode ProcessCapabilities::HandleInterruptFlags(u32 flags) { | |||
| 307 | ResultCode ProcessCapabilities::HandleProgramTypeFlags(u32 flags) { | 331 | ResultCode ProcessCapabilities::HandleProgramTypeFlags(u32 flags) { |
| 308 | const u32 reserved = flags >> 17; | 332 | const u32 reserved = flags >> 17; |
| 309 | if (reserved != 0) { | 333 | if (reserved != 0) { |
| 334 | LOG_ERROR(Kernel, "Reserved value is non-zero! reserved={}", reserved); | ||
| 310 | return ERR_RESERVED_VALUE; | 335 | return ERR_RESERVED_VALUE; |
| 311 | } | 336 | } |
| 312 | 337 | ||
| @@ -324,6 +349,9 @@ ResultCode ProcessCapabilities::HandleKernelVersionFlags(u32 flags) { | |||
| 324 | const u32 major_version = kernel_version >> 19; | 349 | const u32 major_version = kernel_version >> 19; |
| 325 | 350 | ||
| 326 | if (major_version != 0 || flags < 0x80000) { | 351 | if (major_version != 0 || flags < 0x80000) { |
| 352 | LOG_ERROR(Kernel, | ||
| 353 | "Kernel version is non zero or flags are too small! major_version={}, flags={}", | ||
| 354 | major_version, flags); | ||
| 327 | return ERR_INVALID_CAPABILITY_DESCRIPTOR; | 355 | return ERR_INVALID_CAPABILITY_DESCRIPTOR; |
| 328 | } | 356 | } |
| 329 | 357 | ||
| @@ -334,6 +362,7 @@ ResultCode ProcessCapabilities::HandleKernelVersionFlags(u32 flags) { | |||
| 334 | ResultCode ProcessCapabilities::HandleHandleTableFlags(u32 flags) { | 362 | ResultCode ProcessCapabilities::HandleHandleTableFlags(u32 flags) { |
| 335 | const u32 reserved = flags >> 26; | 363 | const u32 reserved = flags >> 26; |
| 336 | if (reserved != 0) { | 364 | if (reserved != 0) { |
| 365 | LOG_ERROR(Kernel, "Reserved value is non-zero! reserved={}", reserved); | ||
| 337 | return ERR_RESERVED_VALUE; | 366 | return ERR_RESERVED_VALUE; |
| 338 | } | 367 | } |
| 339 | 368 | ||
| @@ -344,6 +373,7 @@ ResultCode ProcessCapabilities::HandleHandleTableFlags(u32 flags) { | |||
| 344 | ResultCode ProcessCapabilities::HandleDebugFlags(u32 flags) { | 373 | ResultCode ProcessCapabilities::HandleDebugFlags(u32 flags) { |
| 345 | const u32 reserved = flags >> 19; | 374 | const u32 reserved = flags >> 19; |
| 346 | if (reserved != 0) { | 375 | if (reserved != 0) { |
| 376 | LOG_ERROR(Kernel, "Reserved value is non-zero! reserved={}", reserved); | ||
| 347 | return ERR_RESERVED_VALUE; | 377 | return ERR_RESERVED_VALUE; |
| 348 | } | 378 | } |
| 349 | 379 | ||
diff --git a/src/core/hle/kernel/readable_event.cpp b/src/core/hle/kernel/readable_event.cpp index 9d3d3a81b..00860fcbd 100644 --- a/src/core/hle/kernel/readable_event.cpp +++ b/src/core/hle/kernel/readable_event.cpp | |||
| @@ -4,6 +4,7 @@ | |||
| 4 | 4 | ||
| 5 | #include <algorithm> | 5 | #include <algorithm> |
| 6 | #include "common/assert.h" | 6 | #include "common/assert.h" |
| 7 | #include "common/logging/log.h" | ||
| 7 | #include "core/hle/kernel/errors.h" | 8 | #include "core/hle/kernel/errors.h" |
| 8 | #include "core/hle/kernel/object.h" | 9 | #include "core/hle/kernel/object.h" |
| 9 | #include "core/hle/kernel/readable_event.h" | 10 | #include "core/hle/kernel/readable_event.h" |
| @@ -23,10 +24,12 @@ void ReadableEvent::Acquire(Thread* thread) { | |||
| 23 | } | 24 | } |
| 24 | 25 | ||
| 25 | void ReadableEvent::Signal() { | 26 | void ReadableEvent::Signal() { |
| 26 | if (!is_signaled) { | 27 | if (is_signaled) { |
| 27 | is_signaled = true; | 28 | return; |
| 28 | SynchronizationObject::Signal(); | 29 | } |
| 29 | }; | 30 | |
| 31 | is_signaled = true; | ||
| 32 | SynchronizationObject::Signal(); | ||
| 30 | } | 33 | } |
| 31 | 34 | ||
| 32 | void ReadableEvent::Clear() { | 35 | void ReadableEvent::Clear() { |
| @@ -35,6 +38,8 @@ void ReadableEvent::Clear() { | |||
| 35 | 38 | ||
| 36 | ResultCode ReadableEvent::Reset() { | 39 | ResultCode ReadableEvent::Reset() { |
| 37 | if (!is_signaled) { | 40 | if (!is_signaled) { |
| 41 | LOG_ERROR(Kernel, "Handle is not signaled! object_id={}, object_type={}, object_name={}", | ||
| 42 | GetObjectId(), GetTypeName(), GetName()); | ||
| 38 | return ERR_INVALID_STATE; | 43 | return ERR_INVALID_STATE; |
| 39 | } | 44 | } |
| 40 | 45 | ||
diff --git a/src/core/hle/kernel/resource_limit.cpp b/src/core/hle/kernel/resource_limit.cpp index 96e5b9892..d9beaa3a4 100644 --- a/src/core/hle/kernel/resource_limit.cpp +++ b/src/core/hle/kernel/resource_limit.cpp | |||
| @@ -69,6 +69,8 @@ ResultCode ResourceLimit::SetLimitValue(ResourceType resource, s64 value) { | |||
| 69 | limit[index] = value; | 69 | limit[index] = value; |
| 70 | return RESULT_SUCCESS; | 70 | return RESULT_SUCCESS; |
| 71 | } else { | 71 | } else { |
| 72 | LOG_ERROR(Kernel, "Limit value is too large! resource={}, value={}, index={}", | ||
| 73 | static_cast<u32>(resource), value, index); | ||
| 72 | return ERR_INVALID_STATE; | 74 | return ERR_INVALID_STATE; |
| 73 | } | 75 | } |
| 74 | } | 76 | } |
diff --git a/src/core/hle/kernel/svc.cpp b/src/core/hle/kernel/svc.cpp index 41ef2caf6..4ae4529f5 100644 --- a/src/core/hle/kernel/svc.cpp +++ b/src/core/hle/kernel/svc.cpp | |||
| @@ -685,6 +685,8 @@ static ResultCode GetInfo(Core::System& system, u64* result, u64 info_id, u64 ha | |||
| 685 | case GetInfoType::TotalPhysicalMemoryAvailableWithoutSystemResource: | 685 | case GetInfoType::TotalPhysicalMemoryAvailableWithoutSystemResource: |
| 686 | case GetInfoType::TotalPhysicalMemoryUsedWithoutSystemResource: { | 686 | case GetInfoType::TotalPhysicalMemoryUsedWithoutSystemResource: { |
| 687 | if (info_sub_id != 0) { | 687 | if (info_sub_id != 0) { |
| 688 | LOG_ERROR(Kernel_SVC, "Info sub id is non zero! info_id={}, info_sub_id={}", info_id, | ||
| 689 | info_sub_id); | ||
| 688 | return ERR_INVALID_ENUM_VALUE; | 690 | return ERR_INVALID_ENUM_VALUE; |
| 689 | } | 691 | } |
| 690 | 692 | ||
| @@ -692,6 +694,8 @@ static ResultCode GetInfo(Core::System& system, u64* result, u64 info_id, u64 ha | |||
| 692 | system.Kernel().CurrentProcess()->GetHandleTable(); | 694 | system.Kernel().CurrentProcess()->GetHandleTable(); |
| 693 | const auto process = current_process_handle_table.Get<Process>(static_cast<Handle>(handle)); | 695 | const auto process = current_process_handle_table.Get<Process>(static_cast<Handle>(handle)); |
| 694 | if (!process) { | 696 | if (!process) { |
| 697 | LOG_ERROR(Kernel_SVC, "Process is not valid! info_id={}, info_sub_id={}, handle={:08X}", | ||
| 698 | info_id, info_sub_id, handle); | ||
| 695 | return ERR_INVALID_HANDLE; | 699 | return ERR_INVALID_HANDLE; |
| 696 | } | 700 | } |
| 697 | 701 | ||
| @@ -783,10 +787,13 @@ static ResultCode GetInfo(Core::System& system, u64* result, u64 info_id, u64 ha | |||
| 783 | 787 | ||
| 784 | case GetInfoType::RegisterResourceLimit: { | 788 | case GetInfoType::RegisterResourceLimit: { |
| 785 | if (handle != 0) { | 789 | if (handle != 0) { |
| 790 | LOG_ERROR(Kernel, "Handle is non zero! handle={:08X}", handle); | ||
| 786 | return ERR_INVALID_HANDLE; | 791 | return ERR_INVALID_HANDLE; |
| 787 | } | 792 | } |
| 788 | 793 | ||
| 789 | if (info_sub_id != 0) { | 794 | if (info_sub_id != 0) { |
| 795 | LOG_ERROR(Kernel, "Info sub id is non zero! info_id={}, info_sub_id={}", info_id, | ||
| 796 | info_sub_id); | ||
| 790 | return ERR_INVALID_COMBINATION; | 797 | return ERR_INVALID_COMBINATION; |
| 791 | } | 798 | } |
| 792 | 799 | ||
diff --git a/src/core/hle/kernel/thread.cpp b/src/core/hle/kernel/thread.cpp index a919750a6..db7f379ac 100644 --- a/src/core/hle/kernel/thread.cpp +++ b/src/core/hle/kernel/thread.cpp | |||
| @@ -423,6 +423,8 @@ ResultCode Thread::SetCoreAndAffinityMask(s32 new_core, u64 new_affinity_mask) { | |||
| 423 | if (new_core == THREADPROCESSORID_DONT_UPDATE) { | 423 | if (new_core == THREADPROCESSORID_DONT_UPDATE) { |
| 424 | new_core = use_override ? ideal_core_override : ideal_core; | 424 | new_core = use_override ? ideal_core_override : ideal_core; |
| 425 | if ((new_affinity_mask & (1ULL << new_core)) == 0) { | 425 | if ((new_affinity_mask & (1ULL << new_core)) == 0) { |
| 426 | LOG_ERROR(Kernel, "New affinity mask is incorrect! new_core={}, new_affinity_mask={}", | ||
| 427 | new_core, new_affinity_mask); | ||
| 426 | return ERR_INVALID_COMBINATION; | 428 | return ERR_INVALID_COMBINATION; |
| 427 | } | 429 | } |
| 428 | } | 430 | } |
diff --git a/src/core/hle/service/am/am.cpp b/src/core/hle/service/am/am.cpp index 9450de6e9..a967e6ef7 100644 --- a/src/core/hle/service/am/am.cpp +++ b/src/core/hle/service/am/am.cpp | |||
| @@ -1335,12 +1335,23 @@ void IApplicationFunctions::SetTerminateResult(Kernel::HLERequestContext& ctx) { | |||
| 1335 | } | 1335 | } |
| 1336 | 1336 | ||
| 1337 | void IApplicationFunctions::GetDisplayVersion(Kernel::HLERequestContext& ctx) { | 1337 | void IApplicationFunctions::GetDisplayVersion(Kernel::HLERequestContext& ctx) { |
| 1338 | LOG_WARNING(Service_AM, "(STUBBED) called"); | 1338 | LOG_DEBUG(Service_AM, "called"); |
| 1339 | |||
| 1340 | std::array<u8, 0x10> version_string{}; | ||
| 1341 | |||
| 1342 | FileSys::PatchManager pm{system.CurrentProcess()->GetTitleID()}; | ||
| 1343 | const auto res = pm.GetControlMetadata(); | ||
| 1344 | if (res.first != nullptr) { | ||
| 1345 | const auto& version = res.first->GetVersionString(); | ||
| 1346 | std::copy(version.begin(), version.end(), version_string.begin()); | ||
| 1347 | } else { | ||
| 1348 | constexpr u128 default_version = {1, 0}; | ||
| 1349 | std::memcpy(version_string.data(), default_version.data(), sizeof(u128)); | ||
| 1350 | } | ||
| 1339 | 1351 | ||
| 1340 | IPC::ResponseBuilder rb{ctx, 6}; | 1352 | IPC::ResponseBuilder rb{ctx, 6}; |
| 1341 | rb.Push(RESULT_SUCCESS); | 1353 | rb.Push(RESULT_SUCCESS); |
| 1342 | rb.Push<u64>(1); | 1354 | rb.PushRaw(version_string); |
| 1343 | rb.Push<u64>(0); | ||
| 1344 | } | 1355 | } |
| 1345 | 1356 | ||
| 1346 | void IApplicationFunctions::GetDesiredLanguage(Kernel::HLERequestContext& ctx) { | 1357 | void IApplicationFunctions::GetDesiredLanguage(Kernel::HLERequestContext& ctx) { |
diff --git a/src/core/hle/service/audio/audin_u.cpp b/src/core/hle/service/audio/audin_u.cpp index d7f1d348d..3e2299426 100644 --- a/src/core/hle/service/audio/audin_u.cpp +++ b/src/core/hle/service/audio/audin_u.cpp | |||
| @@ -2,6 +2,9 @@ | |||
| 2 | // Licensed under GPLv2 or any later version | 2 | // Licensed under GPLv2 or any later version |
| 3 | // Refer to the license.txt file included. | 3 | // Refer to the license.txt file included. |
| 4 | 4 | ||
| 5 | #include "common/logging/log.h" | ||
| 6 | #include "core/hle/ipc_helpers.h" | ||
| 7 | #include "core/hle/kernel/hle_ipc.h" | ||
| 5 | #include "core/hle/service/audio/audin_u.h" | 8 | #include "core/hle/service/audio/audin_u.h" |
| 6 | 9 | ||
| 7 | namespace Service::Audio { | 10 | namespace Service::Audio { |
| @@ -36,11 +39,12 @@ public: | |||
| 36 | AudInU::AudInU() : ServiceFramework("audin:u") { | 39 | AudInU::AudInU() : ServiceFramework("audin:u") { |
| 37 | // clang-format off | 40 | // clang-format off |
| 38 | static const FunctionInfo functions[] = { | 41 | static const FunctionInfo functions[] = { |
| 39 | {0, nullptr, "ListAudioIns"}, | 42 | {0, &AudInU::ListAudioIns, "ListAudioIns"}, |
| 40 | {1, nullptr, "OpenAudioIn"}, | 43 | {1, &AudInU::OpenAudioIn, "OpenAudioIn"}, |
| 41 | {2, nullptr, "Unknown"}, | 44 | {2, &AudInU::ListAudioIns, "ListAudioInsAuto"}, |
| 42 | {3, nullptr, "OpenAudioInAuto"}, | 45 | {3, &AudInU::OpenAudioIn, "OpenAudioInAuto"}, |
| 43 | {4, nullptr, "ListAudioInsAuto"}, | 46 | {4, &AudInU::ListAudioInsAutoFiltered, "ListAudioInsAutoFiltered"}, |
| 47 | {5, &AudInU::OpenAudioInProtocolSpecified, "OpenAudioInProtocolSpecified"}, | ||
| 44 | }; | 48 | }; |
| 45 | // clang-format on | 49 | // clang-format on |
| 46 | 50 | ||
| @@ -49,4 +53,60 @@ AudInU::AudInU() : ServiceFramework("audin:u") { | |||
| 49 | 53 | ||
| 50 | AudInU::~AudInU() = default; | 54 | AudInU::~AudInU() = default; |
| 51 | 55 | ||
| 56 | void AudInU::ListAudioIns(Kernel::HLERequestContext& ctx) { | ||
| 57 | LOG_DEBUG(Service_Audio, "called"); | ||
| 58 | const std::size_t count = ctx.GetWriteBufferSize() / sizeof(AudioInDeviceName); | ||
| 59 | |||
| 60 | const std::size_t device_count = std::min(count, audio_device_names.size()); | ||
| 61 | std::vector<AudioInDeviceName> device_names; | ||
| 62 | device_names.reserve(device_count); | ||
| 63 | |||
| 64 | for (std::size_t i = 0; i < device_count; i++) { | ||
| 65 | const auto& device_name = audio_device_names[i]; | ||
| 66 | auto& entry = device_names.emplace_back(); | ||
| 67 | device_name.copy(entry.data(), device_name.size()); | ||
| 68 | } | ||
| 69 | |||
| 70 | ctx.WriteBuffer(device_names); | ||
| 71 | |||
| 72 | IPC::ResponseBuilder rb{ctx, 3}; | ||
| 73 | rb.Push(RESULT_SUCCESS); | ||
| 74 | rb.Push(static_cast<u32>(device_names.size())); | ||
| 75 | } | ||
| 76 | |||
| 77 | void AudInU::ListAudioInsAutoFiltered(Kernel::HLERequestContext& ctx) { | ||
| 78 | LOG_DEBUG(Service_Audio, "called"); | ||
| 79 | constexpr u32 device_count = 0; | ||
| 80 | |||
| 81 | // Since we don't actually use any other audio input devices, we return 0 devices. Filtered | ||
| 82 | // device listing just omits the default input device | ||
| 83 | |||
| 84 | IPC::ResponseBuilder rb{ctx, 3}; | ||
| 85 | rb.Push(RESULT_SUCCESS); | ||
| 86 | rb.Push(static_cast<u32>(device_count)); | ||
| 87 | } | ||
| 88 | |||
| 89 | void AudInU::OpenInOutImpl(Kernel::HLERequestContext& ctx) { | ||
| 90 | AudInOutParams params{}; | ||
| 91 | params.channel_count = 2; | ||
| 92 | params.sample_format = SampleFormat::PCM16; | ||
| 93 | params.sample_rate = 48000; | ||
| 94 | params.state = State::Started; | ||
| 95 | |||
| 96 | IPC::ResponseBuilder rb{ctx, 6, 0, 1}; | ||
| 97 | rb.Push(RESULT_SUCCESS); | ||
| 98 | rb.PushRaw<AudInOutParams>(params); | ||
| 99 | rb.PushIpcInterface<IAudioIn>(); | ||
| 100 | } | ||
| 101 | |||
| 102 | void AudInU::OpenAudioIn(Kernel::HLERequestContext& ctx) { | ||
| 103 | LOG_WARNING(Service_Audio, "(STUBBED) called"); | ||
| 104 | OpenInOutImpl(ctx); | ||
| 105 | } | ||
| 106 | |||
| 107 | void AudInU::OpenAudioInProtocolSpecified(Kernel::HLERequestContext& ctx) { | ||
| 108 | LOG_WARNING(Service_Audio, "(STUBBED) called"); | ||
| 109 | OpenInOutImpl(ctx); | ||
| 110 | } | ||
| 111 | |||
| 52 | } // namespace Service::Audio | 112 | } // namespace Service::Audio |
diff --git a/src/core/hle/service/audio/audin_u.h b/src/core/hle/service/audio/audin_u.h index 0538b9560..a599f4a64 100644 --- a/src/core/hle/service/audio/audin_u.h +++ b/src/core/hle/service/audio/audin_u.h | |||
| @@ -16,6 +16,35 @@ class AudInU final : public ServiceFramework<AudInU> { | |||
| 16 | public: | 16 | public: |
| 17 | explicit AudInU(); | 17 | explicit AudInU(); |
| 18 | ~AudInU() override; | 18 | ~AudInU() override; |
| 19 | |||
| 20 | private: | ||
| 21 | enum class SampleFormat : u32_le { | ||
| 22 | PCM16 = 2, | ||
| 23 | }; | ||
| 24 | |||
| 25 | enum class State : u32_le { | ||
| 26 | Started = 0, | ||
| 27 | Stopped = 1, | ||
| 28 | }; | ||
| 29 | |||
| 30 | struct AudInOutParams { | ||
| 31 | u32_le sample_rate{}; | ||
| 32 | u32_le channel_count{}; | ||
| 33 | SampleFormat sample_format{}; | ||
| 34 | State state{}; | ||
| 35 | }; | ||
| 36 | static_assert(sizeof(AudInOutParams) == 0x10, "AudInOutParams is an invalid size"); | ||
| 37 | |||
| 38 | using AudioInDeviceName = std::array<char, 256>; | ||
| 39 | static constexpr std::array<std::string_view, 1> audio_device_names{{ | ||
| 40 | "BuiltInHeadset", | ||
| 41 | }}; | ||
| 42 | |||
| 43 | void ListAudioIns(Kernel::HLERequestContext& ctx); | ||
| 44 | void ListAudioInsAutoFiltered(Kernel::HLERequestContext& ctx); | ||
| 45 | void OpenInOutImpl(Kernel::HLERequestContext& ctx); | ||
| 46 | void OpenAudioIn(Kernel::HLERequestContext& ctx); | ||
| 47 | void OpenAudioInProtocolSpecified(Kernel::HLERequestContext& ctx); | ||
| 19 | }; | 48 | }; |
| 20 | 49 | ||
| 21 | } // namespace Service::Audio | 50 | } // namespace Service::Audio |
diff --git a/src/core/hle/service/caps/caps_su.cpp b/src/core/hle/service/caps/caps_su.cpp index b4d9355ef..e8b0698e8 100644 --- a/src/core/hle/service/caps/caps_su.cpp +++ b/src/core/hle/service/caps/caps_su.cpp | |||
| @@ -2,6 +2,8 @@ | |||
| 2 | // Licensed under GPLv2 or any later version | 2 | // Licensed under GPLv2 or any later version |
| 3 | // Refer to the license.txt file included. | 3 | // Refer to the license.txt file included. |
| 4 | 4 | ||
| 5 | #include "common/logging/log.h" | ||
| 6 | #include "core/hle/ipc_helpers.h" | ||
| 5 | #include "core/hle/service/caps/caps_su.h" | 7 | #include "core/hle/service/caps/caps_su.h" |
| 6 | 8 | ||
| 7 | namespace Service::Capture { | 9 | namespace Service::Capture { |
| @@ -9,7 +11,7 @@ namespace Service::Capture { | |||
| 9 | CAPS_SU::CAPS_SU() : ServiceFramework("caps:su") { | 11 | CAPS_SU::CAPS_SU() : ServiceFramework("caps:su") { |
| 10 | // clang-format off | 12 | // clang-format off |
| 11 | static const FunctionInfo functions[] = { | 13 | static const FunctionInfo functions[] = { |
| 12 | {32, nullptr, "SetShimLibraryVersion"}, | 14 | {32, &CAPS_SU::SetShimLibraryVersion, "SetShimLibraryVersion"}, |
| 13 | {201, nullptr, "SaveScreenShot"}, | 15 | {201, nullptr, "SaveScreenShot"}, |
| 14 | {203, nullptr, "SaveScreenShotEx0"}, | 16 | {203, nullptr, "SaveScreenShotEx0"}, |
| 15 | {205, nullptr, "SaveScreenShotEx1"}, | 17 | {205, nullptr, "SaveScreenShotEx1"}, |
| @@ -22,4 +24,11 @@ CAPS_SU::CAPS_SU() : ServiceFramework("caps:su") { | |||
| 22 | 24 | ||
| 23 | CAPS_SU::~CAPS_SU() = default; | 25 | CAPS_SU::~CAPS_SU() = default; |
| 24 | 26 | ||
| 27 | void CAPS_SU::SetShimLibraryVersion(Kernel::HLERequestContext& ctx) { | ||
| 28 | LOG_WARNING(Service_Capture, "(STUBBED) called"); | ||
| 29 | |||
| 30 | IPC::ResponseBuilder rb{ctx, 2}; | ||
| 31 | rb.Push(RESULT_SUCCESS); | ||
| 32 | } | ||
| 33 | |||
| 25 | } // namespace Service::Capture | 34 | } // namespace Service::Capture |
diff --git a/src/core/hle/service/caps/caps_su.h b/src/core/hle/service/caps/caps_su.h index cb11f7c9a..c494d7c84 100644 --- a/src/core/hle/service/caps/caps_su.h +++ b/src/core/hle/service/caps/caps_su.h | |||
| @@ -16,6 +16,9 @@ class CAPS_SU final : public ServiceFramework<CAPS_SU> { | |||
| 16 | public: | 16 | public: |
| 17 | explicit CAPS_SU(); | 17 | explicit CAPS_SU(); |
| 18 | ~CAPS_SU() override; | 18 | ~CAPS_SU() override; |
| 19 | |||
| 20 | private: | ||
| 21 | void SetShimLibraryVersion(Kernel::HLERequestContext& ctx); | ||
| 19 | }; | 22 | }; |
| 20 | 23 | ||
| 21 | } // namespace Service::Capture | 24 | } // namespace Service::Capture |
diff --git a/src/input_common/main.cpp b/src/input_common/main.cpp index c98c848cf..95e351e24 100644 --- a/src/input_common/main.cpp +++ b/src/input_common/main.cpp | |||
| @@ -18,7 +18,9 @@ namespace InputCommon { | |||
| 18 | 18 | ||
| 19 | static std::shared_ptr<Keyboard> keyboard; | 19 | static std::shared_ptr<Keyboard> keyboard; |
| 20 | static std::shared_ptr<MotionEmu> motion_emu; | 20 | static std::shared_ptr<MotionEmu> motion_emu; |
| 21 | #ifdef HAVE_SDL2 | ||
| 21 | static std::unique_ptr<SDL::State> sdl; | 22 | static std::unique_ptr<SDL::State> sdl; |
| 23 | #endif | ||
| 22 | static std::unique_ptr<CemuhookUDP::State> udp; | 24 | static std::unique_ptr<CemuhookUDP::State> udp; |
| 23 | 25 | ||
| 24 | void Init() { | 26 | void Init() { |
| @@ -29,7 +31,9 @@ void Init() { | |||
| 29 | motion_emu = std::make_shared<MotionEmu>(); | 31 | motion_emu = std::make_shared<MotionEmu>(); |
| 30 | Input::RegisterFactory<Input::MotionDevice>("motion_emu", motion_emu); | 32 | Input::RegisterFactory<Input::MotionDevice>("motion_emu", motion_emu); |
| 31 | 33 | ||
| 34 | #ifdef HAVE_SDL2 | ||
| 32 | sdl = SDL::Init(); | 35 | sdl = SDL::Init(); |
| 36 | #endif | ||
| 33 | 37 | ||
| 34 | udp = CemuhookUDP::Init(); | 38 | udp = CemuhookUDP::Init(); |
| 35 | } | 39 | } |
| @@ -40,7 +44,9 @@ void Shutdown() { | |||
| 40 | Input::UnregisterFactory<Input::AnalogDevice>("analog_from_button"); | 44 | Input::UnregisterFactory<Input::AnalogDevice>("analog_from_button"); |
| 41 | Input::UnregisterFactory<Input::MotionDevice>("motion_emu"); | 45 | Input::UnregisterFactory<Input::MotionDevice>("motion_emu"); |
| 42 | motion_emu.reset(); | 46 | motion_emu.reset(); |
| 47 | #ifdef HAVE_SDL2 | ||
| 43 | sdl.reset(); | 48 | sdl.reset(); |
| 49 | #endif | ||
| 44 | udp.reset(); | 50 | udp.reset(); |
| 45 | } | 51 | } |
| 46 | 52 | ||
diff --git a/src/video_core/buffer_cache/buffer_cache.h b/src/video_core/buffer_cache/buffer_cache.h index 510f11089..398f16181 100644 --- a/src/video_core/buffer_cache/buffer_cache.h +++ b/src/video_core/buffer_cache/buffer_cache.h | |||
| @@ -4,7 +4,6 @@ | |||
| 4 | 4 | ||
| 5 | #pragma once | 5 | #pragma once |
| 6 | 6 | ||
| 7 | #include <array> | ||
| 8 | #include <list> | 7 | #include <list> |
| 9 | #include <memory> | 8 | #include <memory> |
| 10 | #include <mutex> | 9 | #include <mutex> |
diff --git a/src/video_core/fence_manager.h b/src/video_core/fence_manager.h index dabd1588c..8b2a6a42c 100644 --- a/src/video_core/fence_manager.h +++ b/src/video_core/fence_manager.h | |||
| @@ -88,7 +88,8 @@ public: | |||
| 88 | } | 88 | } |
| 89 | PopAsyncFlushes(); | 89 | PopAsyncFlushes(); |
| 90 | if (current_fence->IsSemaphore()) { | 90 | if (current_fence->IsSemaphore()) { |
| 91 | memory_manager.Write<u32>(current_fence->GetAddress(), current_fence->GetPayload()); | 91 | memory_manager.template Write<u32>(current_fence->GetAddress(), |
| 92 | current_fence->GetPayload()); | ||
| 92 | } else { | 93 | } else { |
| 93 | gpu.IncrementSyncPoint(current_fence->GetPayload()); | 94 | gpu.IncrementSyncPoint(current_fence->GetPayload()); |
| 94 | } | 95 | } |
| @@ -134,7 +135,8 @@ private: | |||
| 134 | } | 135 | } |
| 135 | PopAsyncFlushes(); | 136 | PopAsyncFlushes(); |
| 136 | if (current_fence->IsSemaphore()) { | 137 | if (current_fence->IsSemaphore()) { |
| 137 | memory_manager.Write<u32>(current_fence->GetAddress(), current_fence->GetPayload()); | 138 | memory_manager.template Write<u32>(current_fence->GetAddress(), |
| 139 | current_fence->GetPayload()); | ||
| 138 | } else { | 140 | } else { |
| 139 | gpu.IncrementSyncPoint(current_fence->GetPayload()); | 141 | gpu.IncrementSyncPoint(current_fence->GetPayload()); |
| 140 | } | 142 | } |
diff --git a/src/video_core/renderer_opengl/gl_rasterizer.cpp b/src/video_core/renderer_opengl/gl_rasterizer.cpp index f33c4a8f9..725b4c32d 100644 --- a/src/video_core/renderer_opengl/gl_rasterizer.cpp +++ b/src/video_core/renderer_opengl/gl_rasterizer.cpp | |||
| @@ -59,14 +59,12 @@ constexpr std::size_t NumSupportedVertexAttributes = 16; | |||
| 59 | template <typename Engine, typename Entry> | 59 | template <typename Engine, typename Entry> |
| 60 | Tegra::Texture::FullTextureInfo GetTextureInfo(const Engine& engine, const Entry& entry, | 60 | Tegra::Texture::FullTextureInfo GetTextureInfo(const Engine& engine, const Entry& entry, |
| 61 | ShaderType shader_type, std::size_t index = 0) { | 61 | ShaderType shader_type, std::size_t index = 0) { |
| 62 | if (entry.IsBindless()) { | 62 | if (entry.is_bindless) { |
| 63 | const Tegra::Texture::TextureHandle tex_handle = | 63 | const auto tex_handle = engine.AccessConstBuffer32(shader_type, entry.buffer, entry.offset); |
| 64 | engine.AccessConstBuffer32(shader_type, entry.GetBuffer(), entry.GetOffset()); | ||
| 65 | return engine.GetTextureInfo(tex_handle); | 64 | return engine.GetTextureInfo(tex_handle); |
| 66 | } | 65 | } |
| 67 | const auto& gpu_profile = engine.AccessGuestDriverProfile(); | 66 | const auto& gpu_profile = engine.AccessGuestDriverProfile(); |
| 68 | const u32 offset = | 67 | const u32 offset = entry.offset + static_cast<u32>(index * gpu_profile.GetTextureHandlerSize()); |
| 69 | entry.GetOffset() + static_cast<u32>(index * gpu_profile.GetTextureHandlerSize()); | ||
| 70 | if constexpr (std::is_same_v<Engine, Tegra::Engines::Maxwell3D>) { | 68 | if constexpr (std::is_same_v<Engine, Tegra::Engines::Maxwell3D>) { |
| 71 | return engine.GetStageTexture(shader_type, offset); | 69 | return engine.GetStageTexture(shader_type, offset); |
| 72 | } else { | 70 | } else { |
| @@ -856,9 +854,9 @@ void RasterizerOpenGL::SetupDrawGlobalMemory(std::size_t stage_index, const Shad | |||
| 856 | 854 | ||
| 857 | u32 binding = device.GetBaseBindings(stage_index).shader_storage_buffer; | 855 | u32 binding = device.GetBaseBindings(stage_index).shader_storage_buffer; |
| 858 | for (const auto& entry : shader->GetEntries().global_memory_entries) { | 856 | for (const auto& entry : shader->GetEntries().global_memory_entries) { |
| 859 | const auto addr{cbufs.const_buffers[entry.GetCbufIndex()].address + entry.GetCbufOffset()}; | 857 | const GPUVAddr addr{cbufs.const_buffers[entry.cbuf_index].address + entry.cbuf_offset}; |
| 860 | const auto gpu_addr{memory_manager.Read<u64>(addr)}; | 858 | const GPUVAddr gpu_addr{memory_manager.Read<u64>(addr)}; |
| 861 | const auto size{memory_manager.Read<u32>(addr + 8)}; | 859 | const u32 size{memory_manager.Read<u32>(addr + 8)}; |
| 862 | SetupGlobalMemory(binding++, entry, gpu_addr, size); | 860 | SetupGlobalMemory(binding++, entry, gpu_addr, size); |
| 863 | } | 861 | } |
| 864 | } | 862 | } |
| @@ -870,7 +868,7 @@ void RasterizerOpenGL::SetupComputeGlobalMemory(const Shader& kernel) { | |||
| 870 | 868 | ||
| 871 | u32 binding = 0; | 869 | u32 binding = 0; |
| 872 | for (const auto& entry : kernel->GetEntries().global_memory_entries) { | 870 | for (const auto& entry : kernel->GetEntries().global_memory_entries) { |
| 873 | const auto addr{cbufs[entry.GetCbufIndex()].Address() + entry.GetCbufOffset()}; | 871 | const auto addr{cbufs[entry.cbuf_index].Address() + entry.cbuf_offset}; |
| 874 | const auto gpu_addr{memory_manager.Read<u64>(addr)}; | 872 | const auto gpu_addr{memory_manager.Read<u64>(addr)}; |
| 875 | const auto size{memory_manager.Read<u32>(addr + 8)}; | 873 | const auto size{memory_manager.Read<u32>(addr + 8)}; |
| 876 | SetupGlobalMemory(binding++, entry, gpu_addr, size); | 874 | SetupGlobalMemory(binding++, entry, gpu_addr, size); |
| @@ -881,7 +879,7 @@ void RasterizerOpenGL::SetupGlobalMemory(u32 binding, const GlobalMemoryEntry& e | |||
| 881 | GPUVAddr gpu_addr, std::size_t size) { | 879 | GPUVAddr gpu_addr, std::size_t size) { |
| 882 | const auto alignment{device.GetShaderStorageBufferAlignment()}; | 880 | const auto alignment{device.GetShaderStorageBufferAlignment()}; |
| 883 | const auto [ssbo, buffer_offset] = | 881 | const auto [ssbo, buffer_offset] = |
| 884 | buffer_cache.UploadMemory(gpu_addr, size, alignment, entry.IsWritten()); | 882 | buffer_cache.UploadMemory(gpu_addr, size, alignment, entry.is_written); |
| 885 | glBindBufferRange(GL_SHADER_STORAGE_BUFFER, binding, ssbo, buffer_offset, | 883 | glBindBufferRange(GL_SHADER_STORAGE_BUFFER, binding, ssbo, buffer_offset, |
| 886 | static_cast<GLsizeiptr>(size)); | 884 | static_cast<GLsizeiptr>(size)); |
| 887 | } | 885 | } |
| @@ -892,7 +890,7 @@ void RasterizerOpenGL::SetupDrawTextures(std::size_t stage_index, const Shader& | |||
| 892 | u32 binding = device.GetBaseBindings(stage_index).sampler; | 890 | u32 binding = device.GetBaseBindings(stage_index).sampler; |
| 893 | for (const auto& entry : shader->GetEntries().samplers) { | 891 | for (const auto& entry : shader->GetEntries().samplers) { |
| 894 | const auto shader_type = static_cast<ShaderType>(stage_index); | 892 | const auto shader_type = static_cast<ShaderType>(stage_index); |
| 895 | for (std::size_t i = 0; i < entry.Size(); ++i) { | 893 | for (std::size_t i = 0; i < entry.size; ++i) { |
| 896 | const auto texture = GetTextureInfo(maxwell3d, entry, shader_type, i); | 894 | const auto texture = GetTextureInfo(maxwell3d, entry, shader_type, i); |
| 897 | SetupTexture(binding++, texture, entry); | 895 | SetupTexture(binding++, texture, entry); |
| 898 | } | 896 | } |
| @@ -904,7 +902,7 @@ void RasterizerOpenGL::SetupComputeTextures(const Shader& kernel) { | |||
| 904 | const auto& compute = system.GPU().KeplerCompute(); | 902 | const auto& compute = system.GPU().KeplerCompute(); |
| 905 | u32 binding = 0; | 903 | u32 binding = 0; |
| 906 | for (const auto& entry : kernel->GetEntries().samplers) { | 904 | for (const auto& entry : kernel->GetEntries().samplers) { |
| 907 | for (std::size_t i = 0; i < entry.Size(); ++i) { | 905 | for (std::size_t i = 0; i < entry.size; ++i) { |
| 908 | const auto texture = GetTextureInfo(compute, entry, ShaderType::Compute, i); | 906 | const auto texture = GetTextureInfo(compute, entry, ShaderType::Compute, i); |
| 909 | SetupTexture(binding++, texture, entry); | 907 | SetupTexture(binding++, texture, entry); |
| 910 | } | 908 | } |
| @@ -961,7 +959,7 @@ void RasterizerOpenGL::SetupImage(u32 binding, const Tegra::Texture::TICEntry& t | |||
| 961 | if (!tic.IsBuffer()) { | 959 | if (!tic.IsBuffer()) { |
| 962 | view->ApplySwizzle(tic.x_source, tic.y_source, tic.z_source, tic.w_source); | 960 | view->ApplySwizzle(tic.x_source, tic.y_source, tic.z_source, tic.w_source); |
| 963 | } | 961 | } |
| 964 | if (entry.IsWritten()) { | 962 | if (entry.is_written) { |
| 965 | view->MarkAsModified(texture_cache.Tick()); | 963 | view->MarkAsModified(texture_cache.Tick()); |
| 966 | } | 964 | } |
| 967 | glBindImageTexture(binding, view->GetTexture(), 0, GL_TRUE, 0, GL_READ_WRITE, | 965 | glBindImageTexture(binding, view->GetTexture(), 0, GL_TRUE, 0, GL_READ_WRITE, |
diff --git a/src/video_core/renderer_opengl/gl_shader_decompiler.cpp b/src/video_core/renderer_opengl/gl_shader_decompiler.cpp index 3803a6f3a..99fd4ae2c 100644 --- a/src/video_core/renderer_opengl/gl_shader_decompiler.cpp +++ b/src/video_core/renderer_opengl/gl_shader_decompiler.cpp | |||
| @@ -870,13 +870,13 @@ private: | |||
| 870 | for (const auto& sampler : ir.GetSamplers()) { | 870 | for (const auto& sampler : ir.GetSamplers()) { |
| 871 | const std::string name = GetSampler(sampler); | 871 | const std::string name = GetSampler(sampler); |
| 872 | const std::string description = fmt::format("layout (binding = {}) uniform", binding); | 872 | const std::string description = fmt::format("layout (binding = {}) uniform", binding); |
| 873 | binding += sampler.IsIndexed() ? sampler.Size() : 1; | 873 | binding += sampler.is_indexed ? sampler.size : 1; |
| 874 | 874 | ||
| 875 | std::string sampler_type = [&]() { | 875 | std::string sampler_type = [&]() { |
| 876 | if (sampler.IsBuffer()) { | 876 | if (sampler.is_buffer) { |
| 877 | return "samplerBuffer"; | 877 | return "samplerBuffer"; |
| 878 | } | 878 | } |
| 879 | switch (sampler.GetType()) { | 879 | switch (sampler.type) { |
| 880 | case Tegra::Shader::TextureType::Texture1D: | 880 | case Tegra::Shader::TextureType::Texture1D: |
| 881 | return "sampler1D"; | 881 | return "sampler1D"; |
| 882 | case Tegra::Shader::TextureType::Texture2D: | 882 | case Tegra::Shader::TextureType::Texture2D: |
| @@ -890,17 +890,17 @@ private: | |||
| 890 | return "sampler2D"; | 890 | return "sampler2D"; |
| 891 | } | 891 | } |
| 892 | }(); | 892 | }(); |
| 893 | if (sampler.IsArray()) { | 893 | if (sampler.is_array) { |
| 894 | sampler_type += "Array"; | 894 | sampler_type += "Array"; |
| 895 | } | 895 | } |
| 896 | if (sampler.IsShadow()) { | 896 | if (sampler.is_shadow) { |
| 897 | sampler_type += "Shadow"; | 897 | sampler_type += "Shadow"; |
| 898 | } | 898 | } |
| 899 | 899 | ||
| 900 | if (!sampler.IsIndexed()) { | 900 | if (!sampler.is_indexed) { |
| 901 | code.AddLine("{} {} {};", description, sampler_type, name); | 901 | code.AddLine("{} {} {};", description, sampler_type, name); |
| 902 | } else { | 902 | } else { |
| 903 | code.AddLine("{} {} {}[{}];", description, sampler_type, name, sampler.Size()); | 903 | code.AddLine("{} {} {}[{}];", description, sampler_type, name, sampler.size); |
| 904 | } | 904 | } |
| 905 | } | 905 | } |
| 906 | if (!ir.GetSamplers().empty()) { | 906 | if (!ir.GetSamplers().empty()) { |
| @@ -946,14 +946,14 @@ private: | |||
| 946 | u32 binding = device.GetBaseBindings(stage).image; | 946 | u32 binding = device.GetBaseBindings(stage).image; |
| 947 | for (const auto& image : ir.GetImages()) { | 947 | for (const auto& image : ir.GetImages()) { |
| 948 | std::string qualifier = "coherent volatile"; | 948 | std::string qualifier = "coherent volatile"; |
| 949 | if (image.IsRead() && !image.IsWritten()) { | 949 | if (image.is_read && !image.is_written) { |
| 950 | qualifier += " readonly"; | 950 | qualifier += " readonly"; |
| 951 | } else if (image.IsWritten() && !image.IsRead()) { | 951 | } else if (image.is_written && !image.is_read) { |
| 952 | qualifier += " writeonly"; | 952 | qualifier += " writeonly"; |
| 953 | } | 953 | } |
| 954 | 954 | ||
| 955 | const char* format = image.IsAtomic() ? "r32ui, " : ""; | 955 | const char* format = image.is_atomic ? "r32ui, " : ""; |
| 956 | const char* type_declaration = GetImageTypeDeclaration(image.GetType()); | 956 | const char* type_declaration = GetImageTypeDeclaration(image.type); |
| 957 | code.AddLine("layout ({}binding = {}) {} uniform uimage{} {};", format, binding++, | 957 | code.AddLine("layout ({}binding = {}) {} uniform uimage{} {};", format, binding++, |
| 958 | qualifier, type_declaration, GetImage(image)); | 958 | qualifier, type_declaration, GetImage(image)); |
| 959 | } | 959 | } |
| @@ -1337,8 +1337,8 @@ private: | |||
| 1337 | ASSERT(meta); | 1337 | ASSERT(meta); |
| 1338 | 1338 | ||
| 1339 | const std::size_t count = operation.GetOperandsCount(); | 1339 | const std::size_t count = operation.GetOperandsCount(); |
| 1340 | const bool has_array = meta->sampler.IsArray(); | 1340 | const bool has_array = meta->sampler.is_array; |
| 1341 | const bool has_shadow = meta->sampler.IsShadow(); | 1341 | const bool has_shadow = meta->sampler.is_shadow; |
| 1342 | 1342 | ||
| 1343 | std::string expr = "texture" + function_suffix; | 1343 | std::string expr = "texture" + function_suffix; |
| 1344 | if (!meta->aoffi.empty()) { | 1344 | if (!meta->aoffi.empty()) { |
| @@ -1346,7 +1346,7 @@ private: | |||
| 1346 | } else if (!meta->ptp.empty()) { | 1346 | } else if (!meta->ptp.empty()) { |
| 1347 | expr += "Offsets"; | 1347 | expr += "Offsets"; |
| 1348 | } | 1348 | } |
| 1349 | if (!meta->sampler.IsIndexed()) { | 1349 | if (!meta->sampler.is_indexed) { |
| 1350 | expr += '(' + GetSampler(meta->sampler) + ", "; | 1350 | expr += '(' + GetSampler(meta->sampler) + ", "; |
| 1351 | } else { | 1351 | } else { |
| 1352 | expr += '(' + GetSampler(meta->sampler) + '[' + Visit(meta->index).AsUint() + "], "; | 1352 | expr += '(' + GetSampler(meta->sampler) + '[' + Visit(meta->index).AsUint() + "], "; |
| @@ -1982,7 +1982,7 @@ private: | |||
| 1982 | 1982 | ||
| 1983 | std::string expr = GenerateTexture( | 1983 | std::string expr = GenerateTexture( |
| 1984 | operation, "", {TextureOffset{}, TextureArgument{Type::Float, meta->bias}}); | 1984 | operation, "", {TextureOffset{}, TextureArgument{Type::Float, meta->bias}}); |
| 1985 | if (meta->sampler.IsShadow()) { | 1985 | if (meta->sampler.is_shadow) { |
| 1986 | expr = "vec4(" + expr + ')'; | 1986 | expr = "vec4(" + expr + ')'; |
| 1987 | } | 1987 | } |
| 1988 | return {expr + GetSwizzle(meta->element), Type::Float}; | 1988 | return {expr + GetSwizzle(meta->element), Type::Float}; |
| @@ -1994,7 +1994,7 @@ private: | |||
| 1994 | 1994 | ||
| 1995 | std::string expr = GenerateTexture( | 1995 | std::string expr = GenerateTexture( |
| 1996 | operation, "Lod", {TextureArgument{Type::Float, meta->lod}, TextureOffset{}}); | 1996 | operation, "Lod", {TextureArgument{Type::Float, meta->lod}, TextureOffset{}}); |
| 1997 | if (meta->sampler.IsShadow()) { | 1997 | if (meta->sampler.is_shadow) { |
| 1998 | expr = "vec4(" + expr + ')'; | 1998 | expr = "vec4(" + expr + ')'; |
| 1999 | } | 1999 | } |
| 2000 | return {expr + GetSwizzle(meta->element), Type::Float}; | 2000 | return {expr + GetSwizzle(meta->element), Type::Float}; |
| @@ -2003,11 +2003,11 @@ private: | |||
| 2003 | Expression TextureGather(Operation operation) { | 2003 | Expression TextureGather(Operation operation) { |
| 2004 | const auto& meta = std::get<MetaTexture>(operation.GetMeta()); | 2004 | const auto& meta = std::get<MetaTexture>(operation.GetMeta()); |
| 2005 | 2005 | ||
| 2006 | const auto type = meta.sampler.IsShadow() ? Type::Float : Type::Int; | 2006 | const auto type = meta.sampler.is_shadow ? Type::Float : Type::Int; |
| 2007 | const bool separate_dc = meta.sampler.IsShadow(); | 2007 | const bool separate_dc = meta.sampler.is_shadow; |
| 2008 | 2008 | ||
| 2009 | std::vector<TextureIR> ir; | 2009 | std::vector<TextureIR> ir; |
| 2010 | if (meta.sampler.IsShadow()) { | 2010 | if (meta.sampler.is_shadow) { |
| 2011 | ir = {TextureOffset{}}; | 2011 | ir = {TextureOffset{}}; |
| 2012 | } else { | 2012 | } else { |
| 2013 | ir = {TextureOffset{}, TextureArgument{type, meta.component}}; | 2013 | ir = {TextureOffset{}, TextureArgument{type, meta.component}}; |
| @@ -2052,7 +2052,7 @@ private: | |||
| 2052 | constexpr std::array constructors = {"int", "ivec2", "ivec3", "ivec4"}; | 2052 | constexpr std::array constructors = {"int", "ivec2", "ivec3", "ivec4"}; |
| 2053 | const auto meta = std::get_if<MetaTexture>(&operation.GetMeta()); | 2053 | const auto meta = std::get_if<MetaTexture>(&operation.GetMeta()); |
| 2054 | ASSERT(meta); | 2054 | ASSERT(meta); |
| 2055 | UNIMPLEMENTED_IF(meta->sampler.IsArray()); | 2055 | UNIMPLEMENTED_IF(meta->sampler.is_array); |
| 2056 | const std::size_t count = operation.GetOperandsCount(); | 2056 | const std::size_t count = operation.GetOperandsCount(); |
| 2057 | 2057 | ||
| 2058 | std::string expr = "texelFetch("; | 2058 | std::string expr = "texelFetch("; |
| @@ -2073,7 +2073,7 @@ private: | |||
| 2073 | } | 2073 | } |
| 2074 | expr += ')'; | 2074 | expr += ')'; |
| 2075 | 2075 | ||
| 2076 | if (meta->lod && !meta->sampler.IsBuffer()) { | 2076 | if (meta->lod && !meta->sampler.is_buffer) { |
| 2077 | expr += ", "; | 2077 | expr += ", "; |
| 2078 | expr += Visit(meta->lod).AsInt(); | 2078 | expr += Visit(meta->lod).AsInt(); |
| 2079 | } | 2079 | } |
| @@ -2084,12 +2084,10 @@ private: | |||
| 2084 | } | 2084 | } |
| 2085 | 2085 | ||
| 2086 | Expression TextureGradient(Operation operation) { | 2086 | Expression TextureGradient(Operation operation) { |
| 2087 | const auto meta = std::get_if<MetaTexture>(&operation.GetMeta()); | 2087 | const auto& meta = std::get<MetaTexture>(operation.GetMeta()); |
| 2088 | ASSERT(meta); | ||
| 2089 | |||
| 2090 | std::string expr = | 2088 | std::string expr = |
| 2091 | GenerateTexture(operation, "Grad", {TextureDerivates{}, TextureOffset{}}); | 2089 | GenerateTexture(operation, "Grad", {TextureDerivates{}, TextureOffset{}}); |
| 2092 | return {std::move(expr) + GetSwizzle(meta->element), Type::Float}; | 2090 | return {std::move(expr) + GetSwizzle(meta.element), Type::Float}; |
| 2093 | } | 2091 | } |
| 2094 | 2092 | ||
| 2095 | Expression ImageLoad(Operation operation) { | 2093 | Expression ImageLoad(Operation operation) { |
| @@ -2608,11 +2606,11 @@ private: | |||
| 2608 | } | 2606 | } |
| 2609 | 2607 | ||
| 2610 | std::string GetSampler(const Sampler& sampler) const { | 2608 | std::string GetSampler(const Sampler& sampler) const { |
| 2611 | return AppendSuffix(static_cast<u32>(sampler.GetIndex()), "sampler"); | 2609 | return AppendSuffix(sampler.index, "sampler"); |
| 2612 | } | 2610 | } |
| 2613 | 2611 | ||
| 2614 | std::string GetImage(const Image& image) const { | 2612 | std::string GetImage(const Image& image) const { |
| 2615 | return AppendSuffix(static_cast<u32>(image.GetIndex()), "image"); | 2613 | return AppendSuffix(image.index, "image"); |
| 2616 | } | 2614 | } |
| 2617 | 2615 | ||
| 2618 | std::string AppendSuffix(u32 index, std::string_view name) const { | 2616 | std::string AppendSuffix(u32 index, std::string_view name) const { |
diff --git a/src/video_core/renderer_opengl/gl_shader_decompiler.h b/src/video_core/renderer_opengl/gl_shader_decompiler.h index e7dbd810c..e8a178764 100644 --- a/src/video_core/renderer_opengl/gl_shader_decompiler.h +++ b/src/video_core/renderer_opengl/gl_shader_decompiler.h | |||
| @@ -33,36 +33,19 @@ public: | |||
| 33 | } | 33 | } |
| 34 | 34 | ||
| 35 | private: | 35 | private: |
| 36 | u32 index{}; | 36 | u32 index = 0; |
| 37 | }; | 37 | }; |
| 38 | 38 | ||
| 39 | class GlobalMemoryEntry { | 39 | struct GlobalMemoryEntry { |
| 40 | public: | 40 | constexpr explicit GlobalMemoryEntry(u32 cbuf_index, u32 cbuf_offset, bool is_read, |
| 41 | explicit GlobalMemoryEntry(u32 cbuf_index, u32 cbuf_offset, bool is_read, bool is_written) | 41 | bool is_written) |
| 42 | : cbuf_index{cbuf_index}, cbuf_offset{cbuf_offset}, is_read{is_read}, is_written{ | 42 | : cbuf_index{cbuf_index}, cbuf_offset{cbuf_offset}, is_read{is_read}, is_written{ |
| 43 | is_written} {} | 43 | is_written} {} |
| 44 | 44 | ||
| 45 | u32 GetCbufIndex() const { | 45 | u32 cbuf_index = 0; |
| 46 | return cbuf_index; | 46 | u32 cbuf_offset = 0; |
| 47 | } | 47 | bool is_read = false; |
| 48 | 48 | bool is_written = false; | |
| 49 | u32 GetCbufOffset() const { | ||
| 50 | return cbuf_offset; | ||
| 51 | } | ||
| 52 | |||
| 53 | bool IsRead() const { | ||
| 54 | return is_read; | ||
| 55 | } | ||
| 56 | |||
| 57 | bool IsWritten() const { | ||
| 58 | return is_written; | ||
| 59 | } | ||
| 60 | |||
| 61 | private: | ||
| 62 | u32 cbuf_index{}; | ||
| 63 | u32 cbuf_offset{}; | ||
| 64 | bool is_read{}; | ||
| 65 | bool is_written{}; | ||
| 66 | }; | 49 | }; |
| 67 | 50 | ||
| 68 | struct ShaderEntries { | 51 | struct ShaderEntries { |
diff --git a/src/video_core/renderer_vulkan/renderer_vulkan.cpp b/src/video_core/renderer_vulkan/renderer_vulkan.cpp index 04532f8f8..59b441943 100644 --- a/src/video_core/renderer_vulkan/renderer_vulkan.cpp +++ b/src/video_core/renderer_vulkan/renderer_vulkan.cpp | |||
| @@ -12,15 +12,12 @@ | |||
| 12 | 12 | ||
| 13 | #include <fmt/format.h> | 13 | #include <fmt/format.h> |
| 14 | 14 | ||
| 15 | #include "common/assert.h" | ||
| 16 | #include "common/dynamic_library.h" | 15 | #include "common/dynamic_library.h" |
| 17 | #include "common/logging/log.h" | 16 | #include "common/logging/log.h" |
| 18 | #include "common/telemetry.h" | 17 | #include "common/telemetry.h" |
| 19 | #include "core/core.h" | 18 | #include "core/core.h" |
| 20 | #include "core/core_timing.h" | 19 | #include "core/core_timing.h" |
| 21 | #include "core/frontend/emu_window.h" | 20 | #include "core/frontend/emu_window.h" |
| 22 | #include "core/memory.h" | ||
| 23 | #include "core/perf_stats.h" | ||
| 24 | #include "core/settings.h" | 21 | #include "core/settings.h" |
| 25 | #include "core/telemetry_session.h" | 22 | #include "core/telemetry_session.h" |
| 26 | #include "video_core/gpu.h" | 23 | #include "video_core/gpu.h" |
diff --git a/src/video_core/renderer_vulkan/renderer_vulkan.h b/src/video_core/renderer_vulkan/renderer_vulkan.h index 18270909b..522b5bff8 100644 --- a/src/video_core/renderer_vulkan/renderer_vulkan.h +++ b/src/video_core/renderer_vulkan/renderer_vulkan.h | |||
| @@ -5,7 +5,6 @@ | |||
| 5 | #pragma once | 5 | #pragma once |
| 6 | 6 | ||
| 7 | #include <memory> | 7 | #include <memory> |
| 8 | #include <optional> | ||
| 9 | #include <string> | 8 | #include <string> |
| 10 | #include <vector> | 9 | #include <vector> |
| 11 | 10 | ||
diff --git a/src/video_core/renderer_vulkan/vk_blit_screen.h b/src/video_core/renderer_vulkan/vk_blit_screen.h index 5eb544aea..243640fab 100644 --- a/src/video_core/renderer_vulkan/vk_blit_screen.h +++ b/src/video_core/renderer_vulkan/vk_blit_screen.h | |||
| @@ -4,7 +4,6 @@ | |||
| 4 | 4 | ||
| 5 | #pragma once | 5 | #pragma once |
| 6 | 6 | ||
| 7 | #include <array> | ||
| 8 | #include <memory> | 7 | #include <memory> |
| 9 | #include <tuple> | 8 | #include <tuple> |
| 10 | 9 | ||
diff --git a/src/video_core/renderer_vulkan/vk_buffer_cache.cpp b/src/video_core/renderer_vulkan/vk_buffer_cache.cpp index 81e1de2be..5b494da8c 100644 --- a/src/video_core/renderer_vulkan/vk_buffer_cache.cpp +++ b/src/video_core/renderer_vulkan/vk_buffer_cache.cpp | |||
| @@ -5,11 +5,7 @@ | |||
| 5 | #include <algorithm> | 5 | #include <algorithm> |
| 6 | #include <cstring> | 6 | #include <cstring> |
| 7 | #include <memory> | 7 | #include <memory> |
| 8 | #include <optional> | ||
| 9 | #include <tuple> | ||
| 10 | 8 | ||
| 11 | #include "common/assert.h" | ||
| 12 | #include "common/bit_util.h" | ||
| 13 | #include "core/core.h" | 9 | #include "core/core.h" |
| 14 | #include "video_core/renderer_vulkan/vk_buffer_cache.h" | 10 | #include "video_core/renderer_vulkan/vk_buffer_cache.h" |
| 15 | #include "video_core/renderer_vulkan/vk_device.h" | 11 | #include "video_core/renderer_vulkan/vk_device.h" |
diff --git a/src/video_core/renderer_vulkan/vk_buffer_cache.h b/src/video_core/renderer_vulkan/vk_buffer_cache.h index 3cd2e2774..a2d0b42b1 100644 --- a/src/video_core/renderer_vulkan/vk_buffer_cache.h +++ b/src/video_core/renderer_vulkan/vk_buffer_cache.h | |||
| @@ -5,14 +5,11 @@ | |||
| 5 | #pragma once | 5 | #pragma once |
| 6 | 6 | ||
| 7 | #include <memory> | 7 | #include <memory> |
| 8 | #include <unordered_map> | ||
| 9 | #include <vector> | ||
| 10 | 8 | ||
| 11 | #include "common/common_types.h" | 9 | #include "common/common_types.h" |
| 12 | #include "video_core/buffer_cache/buffer_cache.h" | 10 | #include "video_core/buffer_cache/buffer_cache.h" |
| 13 | #include "video_core/rasterizer_cache.h" | 11 | #include "video_core/rasterizer_cache.h" |
| 14 | #include "video_core/renderer_vulkan/vk_memory_manager.h" | 12 | #include "video_core/renderer_vulkan/vk_memory_manager.h" |
| 15 | #include "video_core/renderer_vulkan/vk_resource_manager.h" | ||
| 16 | #include "video_core/renderer_vulkan/vk_staging_buffer_pool.h" | 13 | #include "video_core/renderer_vulkan/vk_staging_buffer_pool.h" |
| 17 | #include "video_core/renderer_vulkan/vk_stream_buffer.h" | 14 | #include "video_core/renderer_vulkan/vk_stream_buffer.h" |
| 18 | #include "video_core/renderer_vulkan/wrapper.h" | 15 | #include "video_core/renderer_vulkan/wrapper.h" |
diff --git a/src/video_core/renderer_vulkan/vk_compute_pass.cpp b/src/video_core/renderer_vulkan/vk_compute_pass.cpp index 7b0268033..da71e710c 100644 --- a/src/video_core/renderer_vulkan/vk_compute_pass.cpp +++ b/src/video_core/renderer_vulkan/vk_compute_pass.cpp | |||
| @@ -6,7 +6,7 @@ | |||
| 6 | #include <memory> | 6 | #include <memory> |
| 7 | #include <optional> | 7 | #include <optional> |
| 8 | #include <utility> | 8 | #include <utility> |
| 9 | #include <vector> | 9 | |
| 10 | #include "common/alignment.h" | 10 | #include "common/alignment.h" |
| 11 | #include "common/assert.h" | 11 | #include "common/assert.h" |
| 12 | #include "common/common_types.h" | 12 | #include "common/common_types.h" |
diff --git a/src/video_core/renderer_vulkan/vk_compute_pass.h b/src/video_core/renderer_vulkan/vk_compute_pass.h index 26bf834de..230b526bc 100644 --- a/src/video_core/renderer_vulkan/vk_compute_pass.h +++ b/src/video_core/renderer_vulkan/vk_compute_pass.h | |||
| @@ -6,7 +6,7 @@ | |||
| 6 | 6 | ||
| 7 | #include <optional> | 7 | #include <optional> |
| 8 | #include <utility> | 8 | #include <utility> |
| 9 | #include <vector> | 9 | |
| 10 | #include "common/common_types.h" | 10 | #include "common/common_types.h" |
| 11 | #include "video_core/engines/maxwell_3d.h" | 11 | #include "video_core/engines/maxwell_3d.h" |
| 12 | #include "video_core/renderer_vulkan/vk_descriptor_pool.h" | 12 | #include "video_core/renderer_vulkan/vk_descriptor_pool.h" |
diff --git a/src/video_core/renderer_vulkan/vk_compute_pipeline.cpp b/src/video_core/renderer_vulkan/vk_compute_pipeline.cpp index 52566bb79..8e1b46277 100644 --- a/src/video_core/renderer_vulkan/vk_compute_pipeline.cpp +++ b/src/video_core/renderer_vulkan/vk_compute_pipeline.cpp | |||
| @@ -2,14 +2,12 @@ | |||
| 2 | // Licensed under GPLv2 or any later version | 2 | // Licensed under GPLv2 or any later version |
| 3 | // Refer to the license.txt file included. | 3 | // Refer to the license.txt file included. |
| 4 | 4 | ||
| 5 | #include <memory> | ||
| 6 | #include <vector> | 5 | #include <vector> |
| 7 | 6 | ||
| 8 | #include "video_core/renderer_vulkan/vk_compute_pipeline.h" | 7 | #include "video_core/renderer_vulkan/vk_compute_pipeline.h" |
| 9 | #include "video_core/renderer_vulkan/vk_descriptor_pool.h" | 8 | #include "video_core/renderer_vulkan/vk_descriptor_pool.h" |
| 10 | #include "video_core/renderer_vulkan/vk_device.h" | 9 | #include "video_core/renderer_vulkan/vk_device.h" |
| 11 | #include "video_core/renderer_vulkan/vk_pipeline_cache.h" | 10 | #include "video_core/renderer_vulkan/vk_pipeline_cache.h" |
| 12 | #include "video_core/renderer_vulkan/vk_resource_manager.h" | ||
| 13 | #include "video_core/renderer_vulkan/vk_scheduler.h" | 11 | #include "video_core/renderer_vulkan/vk_scheduler.h" |
| 14 | #include "video_core/renderer_vulkan/vk_shader_decompiler.h" | 12 | #include "video_core/renderer_vulkan/vk_shader_decompiler.h" |
| 15 | #include "video_core/renderer_vulkan/vk_update_descriptor.h" | 13 | #include "video_core/renderer_vulkan/vk_update_descriptor.h" |
diff --git a/src/video_core/renderer_vulkan/vk_compute_pipeline.h b/src/video_core/renderer_vulkan/vk_compute_pipeline.h index 33b9af29e..6e2f22a4a 100644 --- a/src/video_core/renderer_vulkan/vk_compute_pipeline.h +++ b/src/video_core/renderer_vulkan/vk_compute_pipeline.h | |||
| @@ -4,8 +4,6 @@ | |||
| 4 | 4 | ||
| 5 | #pragma once | 5 | #pragma once |
| 6 | 6 | ||
| 7 | #include <memory> | ||
| 8 | |||
| 9 | #include "common/common_types.h" | 7 | #include "common/common_types.h" |
| 10 | #include "video_core/renderer_vulkan/vk_descriptor_pool.h" | 8 | #include "video_core/renderer_vulkan/vk_descriptor_pool.h" |
| 11 | #include "video_core/renderer_vulkan/vk_shader_decompiler.h" | 9 | #include "video_core/renderer_vulkan/vk_shader_decompiler.h" |
diff --git a/src/video_core/renderer_vulkan/vk_descriptor_pool.cpp b/src/video_core/renderer_vulkan/vk_descriptor_pool.cpp index e9d528aa6..890fd52cf 100644 --- a/src/video_core/renderer_vulkan/vk_descriptor_pool.cpp +++ b/src/video_core/renderer_vulkan/vk_descriptor_pool.cpp | |||
| @@ -2,7 +2,6 @@ | |||
| 2 | // Licensed under GPLv2 or any later version | 2 | // Licensed under GPLv2 or any later version |
| 3 | // Refer to the license.txt file included. | 3 | // Refer to the license.txt file included. |
| 4 | 4 | ||
| 5 | #include <memory> | ||
| 6 | #include <vector> | 5 | #include <vector> |
| 7 | 6 | ||
| 8 | #include "common/common_types.h" | 7 | #include "common/common_types.h" |
diff --git a/src/video_core/renderer_vulkan/vk_descriptor_pool.h b/src/video_core/renderer_vulkan/vk_descriptor_pool.h index ab40c70f0..9efa66bef 100644 --- a/src/video_core/renderer_vulkan/vk_descriptor_pool.h +++ b/src/video_core/renderer_vulkan/vk_descriptor_pool.h | |||
| @@ -4,10 +4,8 @@ | |||
| 4 | 4 | ||
| 5 | #pragma once | 5 | #pragma once |
| 6 | 6 | ||
| 7 | #include <memory> | ||
| 8 | #include <vector> | 7 | #include <vector> |
| 9 | 8 | ||
| 10 | #include "common/common_types.h" | ||
| 11 | #include "video_core/renderer_vulkan/vk_resource_manager.h" | 9 | #include "video_core/renderer_vulkan/vk_resource_manager.h" |
| 12 | #include "video_core/renderer_vulkan/wrapper.h" | 10 | #include "video_core/renderer_vulkan/wrapper.h" |
| 13 | 11 | ||
diff --git a/src/video_core/renderer_vulkan/vk_device.cpp b/src/video_core/renderer_vulkan/vk_device.cpp index e90c76492..0e4bbca97 100644 --- a/src/video_core/renderer_vulkan/vk_device.cpp +++ b/src/video_core/renderer_vulkan/vk_device.cpp | |||
| @@ -4,7 +4,6 @@ | |||
| 4 | 4 | ||
| 5 | #include <bitset> | 5 | #include <bitset> |
| 6 | #include <chrono> | 6 | #include <chrono> |
| 7 | #include <cstdlib> | ||
| 8 | #include <optional> | 7 | #include <optional> |
| 9 | #include <string_view> | 8 | #include <string_view> |
| 10 | #include <thread> | 9 | #include <thread> |
diff --git a/src/video_core/renderer_vulkan/vk_graphics_pipeline.cpp b/src/video_core/renderer_vulkan/vk_graphics_pipeline.cpp index 852a17a70..1ac981974 100644 --- a/src/video_core/renderer_vulkan/vk_graphics_pipeline.cpp +++ b/src/video_core/renderer_vulkan/vk_graphics_pipeline.cpp | |||
| @@ -6,7 +6,6 @@ | |||
| 6 | #include <cstring> | 6 | #include <cstring> |
| 7 | #include <vector> | 7 | #include <vector> |
| 8 | 8 | ||
| 9 | #include "common/assert.h" | ||
| 10 | #include "common/common_types.h" | 9 | #include "common/common_types.h" |
| 11 | #include "common/microprofile.h" | 10 | #include "common/microprofile.h" |
| 12 | #include "video_core/renderer_vulkan/fixed_pipeline_state.h" | 11 | #include "video_core/renderer_vulkan/fixed_pipeline_state.h" |
diff --git a/src/video_core/renderer_vulkan/vk_graphics_pipeline.h b/src/video_core/renderer_vulkan/vk_graphics_pipeline.h index 7aba70960..a1d699a6c 100644 --- a/src/video_core/renderer_vulkan/vk_graphics_pipeline.h +++ b/src/video_core/renderer_vulkan/vk_graphics_pipeline.h | |||
| @@ -5,16 +5,13 @@ | |||
| 5 | #pragma once | 5 | #pragma once |
| 6 | 6 | ||
| 7 | #include <array> | 7 | #include <array> |
| 8 | #include <memory> | ||
| 9 | #include <optional> | 8 | #include <optional> |
| 10 | #include <unordered_map> | ||
| 11 | #include <vector> | 9 | #include <vector> |
| 12 | 10 | ||
| 13 | #include "video_core/engines/maxwell_3d.h" | 11 | #include "video_core/engines/maxwell_3d.h" |
| 14 | #include "video_core/renderer_vulkan/fixed_pipeline_state.h" | 12 | #include "video_core/renderer_vulkan/fixed_pipeline_state.h" |
| 15 | #include "video_core/renderer_vulkan/vk_descriptor_pool.h" | 13 | #include "video_core/renderer_vulkan/vk_descriptor_pool.h" |
| 16 | #include "video_core/renderer_vulkan/vk_renderpass_cache.h" | 14 | #include "video_core/renderer_vulkan/vk_renderpass_cache.h" |
| 17 | #include "video_core/renderer_vulkan/vk_resource_manager.h" | ||
| 18 | #include "video_core/renderer_vulkan/vk_shader_decompiler.h" | 15 | #include "video_core/renderer_vulkan/vk_shader_decompiler.h" |
| 19 | #include "video_core/renderer_vulkan/wrapper.h" | 16 | #include "video_core/renderer_vulkan/wrapper.h" |
| 20 | 17 | ||
diff --git a/src/video_core/renderer_vulkan/vk_pipeline_cache.cpp b/src/video_core/renderer_vulkan/vk_pipeline_cache.cpp index 9b703a2f0..fe45ed269 100644 --- a/src/video_core/renderer_vulkan/vk_pipeline_cache.cpp +++ b/src/video_core/renderer_vulkan/vk_pipeline_cache.cpp | |||
| @@ -22,7 +22,6 @@ | |||
| 22 | #include "video_core/renderer_vulkan/vk_pipeline_cache.h" | 22 | #include "video_core/renderer_vulkan/vk_pipeline_cache.h" |
| 23 | #include "video_core/renderer_vulkan/vk_rasterizer.h" | 23 | #include "video_core/renderer_vulkan/vk_rasterizer.h" |
| 24 | #include "video_core/renderer_vulkan/vk_renderpass_cache.h" | 24 | #include "video_core/renderer_vulkan/vk_renderpass_cache.h" |
| 25 | #include "video_core/renderer_vulkan/vk_resource_manager.h" | ||
| 26 | #include "video_core/renderer_vulkan/vk_scheduler.h" | 25 | #include "video_core/renderer_vulkan/vk_scheduler.h" |
| 27 | #include "video_core/renderer_vulkan/vk_update_descriptor.h" | 26 | #include "video_core/renderer_vulkan/vk_update_descriptor.h" |
| 28 | #include "video_core/renderer_vulkan/wrapper.h" | 27 | #include "video_core/renderer_vulkan/wrapper.h" |
| @@ -85,7 +84,7 @@ void AddBindings(std::vector<VkDescriptorSetLayoutBinding>& bindings, u32& bindi | |||
| 85 | u32 count = 1; | 84 | u32 count = 1; |
| 86 | if constexpr (descriptor_type == VK_DESCRIPTOR_TYPE_COMBINED_IMAGE_SAMPLER) { | 85 | if constexpr (descriptor_type == VK_DESCRIPTOR_TYPE_COMBINED_IMAGE_SAMPLER) { |
| 87 | // Combined image samplers can be arrayed. | 86 | // Combined image samplers can be arrayed. |
| 88 | count = container[i].Size(); | 87 | count = container[i].size; |
| 89 | } | 88 | } |
| 90 | VkDescriptorSetLayoutBinding& entry = bindings.emplace_back(); | 89 | VkDescriptorSetLayoutBinding& entry = bindings.emplace_back(); |
| 91 | entry.binding = binding++; | 90 | entry.binding = binding++; |
| @@ -362,7 +361,7 @@ void AddEntry(std::vector<VkDescriptorUpdateTemplateEntry>& template_entries, u3 | |||
| 362 | 361 | ||
| 363 | if constexpr (descriptor_type == COMBINED_IMAGE_SAMPLER) { | 362 | if constexpr (descriptor_type == COMBINED_IMAGE_SAMPLER) { |
| 364 | for (u32 i = 0; i < count; ++i) { | 363 | for (u32 i = 0; i < count; ++i) { |
| 365 | const u32 num_samplers = container[i].Size(); | 364 | const u32 num_samplers = container[i].size; |
| 366 | VkDescriptorUpdateTemplateEntry& entry = template_entries.emplace_back(); | 365 | VkDescriptorUpdateTemplateEntry& entry = template_entries.emplace_back(); |
| 367 | entry.dstBinding = binding; | 366 | entry.dstBinding = binding; |
| 368 | entry.dstArrayElement = 0; | 367 | entry.dstArrayElement = 0; |
diff --git a/src/video_core/renderer_vulkan/vk_pipeline_cache.h b/src/video_core/renderer_vulkan/vk_pipeline_cache.h index ebddafb73..0b5796fef 100644 --- a/src/video_core/renderer_vulkan/vk_pipeline_cache.h +++ b/src/video_core/renderer_vulkan/vk_pipeline_cache.h | |||
| @@ -21,13 +21,11 @@ | |||
| 21 | #include "video_core/renderer_vulkan/fixed_pipeline_state.h" | 21 | #include "video_core/renderer_vulkan/fixed_pipeline_state.h" |
| 22 | #include "video_core/renderer_vulkan/vk_graphics_pipeline.h" | 22 | #include "video_core/renderer_vulkan/vk_graphics_pipeline.h" |
| 23 | #include "video_core/renderer_vulkan/vk_renderpass_cache.h" | 23 | #include "video_core/renderer_vulkan/vk_renderpass_cache.h" |
| 24 | #include "video_core/renderer_vulkan/vk_resource_manager.h" | ||
| 25 | #include "video_core/renderer_vulkan/vk_shader_decompiler.h" | 24 | #include "video_core/renderer_vulkan/vk_shader_decompiler.h" |
| 26 | #include "video_core/renderer_vulkan/wrapper.h" | 25 | #include "video_core/renderer_vulkan/wrapper.h" |
| 27 | #include "video_core/shader/memory_util.h" | 26 | #include "video_core/shader/memory_util.h" |
| 28 | #include "video_core/shader/registry.h" | 27 | #include "video_core/shader/registry.h" |
| 29 | #include "video_core/shader/shader_ir.h" | 28 | #include "video_core/shader/shader_ir.h" |
| 30 | #include "video_core/surface.h" | ||
| 31 | 29 | ||
| 32 | namespace Core { | 30 | namespace Core { |
| 33 | class System; | 31 | class System; |
diff --git a/src/video_core/renderer_vulkan/vk_query_cache.cpp b/src/video_core/renderer_vulkan/vk_query_cache.cpp index 813f7c162..bc91c48cc 100644 --- a/src/video_core/renderer_vulkan/vk_query_cache.cpp +++ b/src/video_core/renderer_vulkan/vk_query_cache.cpp | |||
| @@ -4,7 +4,6 @@ | |||
| 4 | 4 | ||
| 5 | #include <algorithm> | 5 | #include <algorithm> |
| 6 | #include <cstddef> | 6 | #include <cstddef> |
| 7 | #include <cstdint> | ||
| 8 | #include <utility> | 7 | #include <utility> |
| 9 | #include <vector> | 8 | #include <vector> |
| 10 | 9 | ||
diff --git a/src/video_core/renderer_vulkan/vk_query_cache.h b/src/video_core/renderer_vulkan/vk_query_cache.h index b63784f4b..40119e6d3 100644 --- a/src/video_core/renderer_vulkan/vk_query_cache.h +++ b/src/video_core/renderer_vulkan/vk_query_cache.h | |||
| @@ -5,7 +5,6 @@ | |||
| 5 | #pragma once | 5 | #pragma once |
| 6 | 6 | ||
| 7 | #include <cstddef> | 7 | #include <cstddef> |
| 8 | #include <cstdint> | ||
| 9 | #include <memory> | 8 | #include <memory> |
| 10 | #include <utility> | 9 | #include <utility> |
| 11 | #include <vector> | 10 | #include <vector> |
diff --git a/src/video_core/renderer_vulkan/vk_rasterizer.cpp b/src/video_core/renderer_vulkan/vk_rasterizer.cpp index 776053de5..722fde384 100644 --- a/src/video_core/renderer_vulkan/vk_rasterizer.cpp +++ b/src/video_core/renderer_vulkan/vk_rasterizer.cpp | |||
| @@ -9,14 +9,12 @@ | |||
| 9 | #include <vector> | 9 | #include <vector> |
| 10 | 10 | ||
| 11 | #include <boost/container/static_vector.hpp> | 11 | #include <boost/container/static_vector.hpp> |
| 12 | #include <boost/functional/hash.hpp> | ||
| 13 | 12 | ||
| 14 | #include "common/alignment.h" | 13 | #include "common/alignment.h" |
| 15 | #include "common/assert.h" | 14 | #include "common/assert.h" |
| 16 | #include "common/logging/log.h" | 15 | #include "common/logging/log.h" |
| 17 | #include "common/microprofile.h" | 16 | #include "common/microprofile.h" |
| 18 | #include "core/core.h" | 17 | #include "core/core.h" |
| 19 | #include "core/memory.h" | ||
| 20 | #include "core/settings.h" | 18 | #include "core/settings.h" |
| 21 | #include "video_core/engines/kepler_compute.h" | 19 | #include "video_core/engines/kepler_compute.h" |
| 22 | #include "video_core/engines/maxwell_3d.h" | 20 | #include "video_core/engines/maxwell_3d.h" |
| @@ -119,14 +117,13 @@ template <typename Engine, typename Entry> | |||
| 119 | Tegra::Texture::FullTextureInfo GetTextureInfo(const Engine& engine, const Entry& entry, | 117 | Tegra::Texture::FullTextureInfo GetTextureInfo(const Engine& engine, const Entry& entry, |
| 120 | std::size_t stage, std::size_t index = 0) { | 118 | std::size_t stage, std::size_t index = 0) { |
| 121 | const auto stage_type = static_cast<Tegra::Engines::ShaderType>(stage); | 119 | const auto stage_type = static_cast<Tegra::Engines::ShaderType>(stage); |
| 122 | if (entry.IsBindless()) { | 120 | if (entry.is_bindless) { |
| 123 | const Tegra::Texture::TextureHandle tex_handle = | 121 | const auto tex_handle = engine.AccessConstBuffer32(stage_type, entry.buffer, entry.offset); |
| 124 | engine.AccessConstBuffer32(stage_type, entry.GetBuffer(), entry.GetOffset()); | ||
| 125 | return engine.GetTextureInfo(tex_handle); | 122 | return engine.GetTextureInfo(tex_handle); |
| 126 | } | 123 | } |
| 127 | const auto& gpu_profile = engine.AccessGuestDriverProfile(); | 124 | const auto& gpu_profile = engine.AccessGuestDriverProfile(); |
| 128 | const u32 entry_offset = static_cast<u32>(index * gpu_profile.GetTextureHandlerSize()); | 125 | const u32 entry_offset = static_cast<u32>(index * gpu_profile.GetTextureHandlerSize()); |
| 129 | const u32 offset = entry.GetOffset() + entry_offset; | 126 | const u32 offset = entry.offset + entry_offset; |
| 130 | if constexpr (std::is_same_v<Engine, Tegra::Engines::Maxwell3D>) { | 127 | if constexpr (std::is_same_v<Engine, Tegra::Engines::Maxwell3D>) { |
| 131 | return engine.GetStageTexture(stage_type, offset); | 128 | return engine.GetStageTexture(stage_type, offset); |
| 132 | } else { | 129 | } else { |
| @@ -896,6 +893,9 @@ void RasterizerVulkan::SetupVertexArrays(FixedPipelineState::VertexInput& vertex | |||
| 896 | 893 | ||
| 897 | void RasterizerVulkan::SetupIndexBuffer(BufferBindings& buffer_bindings, DrawParameters& params, | 894 | void RasterizerVulkan::SetupIndexBuffer(BufferBindings& buffer_bindings, DrawParameters& params, |
| 898 | bool is_indexed) { | 895 | bool is_indexed) { |
| 896 | if (params.num_vertices == 0) { | ||
| 897 | return; | ||
| 898 | } | ||
| 899 | const auto& regs = system.GPU().Maxwell3D().regs; | 899 | const auto& regs = system.GPU().Maxwell3D().regs; |
| 900 | switch (regs.draw.topology) { | 900 | switch (regs.draw.topology) { |
| 901 | case Maxwell::PrimitiveTopology::Quads: { | 901 | case Maxwell::PrimitiveTopology::Quads: { |
| @@ -971,7 +971,7 @@ void RasterizerVulkan::SetupGraphicsTextures(const ShaderEntries& entries, std:: | |||
| 971 | MICROPROFILE_SCOPE(Vulkan_Textures); | 971 | MICROPROFILE_SCOPE(Vulkan_Textures); |
| 972 | const auto& gpu = system.GPU().Maxwell3D(); | 972 | const auto& gpu = system.GPU().Maxwell3D(); |
| 973 | for (const auto& entry : entries.samplers) { | 973 | for (const auto& entry : entries.samplers) { |
| 974 | for (std::size_t i = 0; i < entry.Size(); ++i) { | 974 | for (std::size_t i = 0; i < entry.size; ++i) { |
| 975 | const auto texture = GetTextureInfo(gpu, entry, stage, i); | 975 | const auto texture = GetTextureInfo(gpu, entry, stage, i); |
| 976 | SetupTexture(texture, entry); | 976 | SetupTexture(texture, entry); |
| 977 | } | 977 | } |
| @@ -1023,7 +1023,7 @@ void RasterizerVulkan::SetupComputeTextures(const ShaderEntries& entries) { | |||
| 1023 | MICROPROFILE_SCOPE(Vulkan_Textures); | 1023 | MICROPROFILE_SCOPE(Vulkan_Textures); |
| 1024 | const auto& gpu = system.GPU().KeplerCompute(); | 1024 | const auto& gpu = system.GPU().KeplerCompute(); |
| 1025 | for (const auto& entry : entries.samplers) { | 1025 | for (const auto& entry : entries.samplers) { |
| 1026 | for (std::size_t i = 0; i < entry.Size(); ++i) { | 1026 | for (std::size_t i = 0; i < entry.size; ++i) { |
| 1027 | const auto texture = GetTextureInfo(gpu, entry, ComputeShaderIndex, i); | 1027 | const auto texture = GetTextureInfo(gpu, entry, ComputeShaderIndex, i); |
| 1028 | SetupTexture(texture, entry); | 1028 | SetupTexture(texture, entry); |
| 1029 | } | 1029 | } |
| @@ -1105,7 +1105,7 @@ void RasterizerVulkan::SetupTexture(const Tegra::Texture::FullTextureInfo& textu | |||
| 1105 | void RasterizerVulkan::SetupImage(const Tegra::Texture::TICEntry& tic, const ImageEntry& entry) { | 1105 | void RasterizerVulkan::SetupImage(const Tegra::Texture::TICEntry& tic, const ImageEntry& entry) { |
| 1106 | auto view = texture_cache.GetImageSurface(tic, entry); | 1106 | auto view = texture_cache.GetImageSurface(tic, entry); |
| 1107 | 1107 | ||
| 1108 | if (entry.IsWritten()) { | 1108 | if (entry.is_written) { |
| 1109 | view->MarkAsModified(texture_cache.Tick()); | 1109 | view->MarkAsModified(texture_cache.Tick()); |
| 1110 | } | 1110 | } |
| 1111 | 1111 | ||
diff --git a/src/video_core/renderer_vulkan/vk_rasterizer.h b/src/video_core/renderer_vulkan/vk_rasterizer.h index d41a7929e..703a094c3 100644 --- a/src/video_core/renderer_vulkan/vk_rasterizer.h +++ b/src/video_core/renderer_vulkan/vk_rasterizer.h | |||
| @@ -14,7 +14,6 @@ | |||
| 14 | #include <boost/functional/hash.hpp> | 14 | #include <boost/functional/hash.hpp> |
| 15 | 15 | ||
| 16 | #include "common/common_types.h" | 16 | #include "common/common_types.h" |
| 17 | #include "video_core/memory_manager.h" | ||
| 18 | #include "video_core/rasterizer_accelerated.h" | 17 | #include "video_core/rasterizer_accelerated.h" |
| 19 | #include "video_core/rasterizer_interface.h" | 18 | #include "video_core/rasterizer_interface.h" |
| 20 | #include "video_core/renderer_vulkan/fixed_pipeline_state.h" | 19 | #include "video_core/renderer_vulkan/fixed_pipeline_state.h" |
diff --git a/src/video_core/renderer_vulkan/vk_sampler_cache.cpp b/src/video_core/renderer_vulkan/vk_sampler_cache.cpp index 07bbcf520..2687d8d95 100644 --- a/src/video_core/renderer_vulkan/vk_sampler_cache.cpp +++ b/src/video_core/renderer_vulkan/vk_sampler_cache.cpp | |||
| @@ -2,11 +2,8 @@ | |||
| 2 | // Licensed under GPLv2 or any later version | 2 | // Licensed under GPLv2 or any later version |
| 3 | // Refer to the license.txt file included. | 3 | // Refer to the license.txt file included. |
| 4 | 4 | ||
| 5 | #include <cstring> | ||
| 6 | #include <optional> | ||
| 7 | #include <unordered_map> | 5 | #include <unordered_map> |
| 8 | 6 | ||
| 9 | #include "common/assert.h" | ||
| 10 | #include "video_core/renderer_vulkan/maxwell_to_vk.h" | 7 | #include "video_core/renderer_vulkan/maxwell_to_vk.h" |
| 11 | #include "video_core/renderer_vulkan/vk_sampler_cache.h" | 8 | #include "video_core/renderer_vulkan/vk_sampler_cache.h" |
| 12 | #include "video_core/renderer_vulkan/wrapper.h" | 9 | #include "video_core/renderer_vulkan/wrapper.h" |
diff --git a/src/video_core/renderer_vulkan/vk_scheduler.cpp b/src/video_core/renderer_vulkan/vk_scheduler.cpp index ae7ba3eb5..82ec9180e 100644 --- a/src/video_core/renderer_vulkan/vk_scheduler.cpp +++ b/src/video_core/renderer_vulkan/vk_scheduler.cpp | |||
| @@ -8,7 +8,6 @@ | |||
| 8 | #include <thread> | 8 | #include <thread> |
| 9 | #include <utility> | 9 | #include <utility> |
| 10 | 10 | ||
| 11 | #include "common/assert.h" | ||
| 12 | #include "common/microprofile.h" | 11 | #include "common/microprofile.h" |
| 13 | #include "video_core/renderer_vulkan/vk_device.h" | 12 | #include "video_core/renderer_vulkan/vk_device.h" |
| 14 | #include "video_core/renderer_vulkan/vk_query_cache.h" | 13 | #include "video_core/renderer_vulkan/vk_query_cache.h" |
diff --git a/src/video_core/renderer_vulkan/vk_scheduler.h b/src/video_core/renderer_vulkan/vk_scheduler.h index 82a8adc69..970a65566 100644 --- a/src/video_core/renderer_vulkan/vk_scheduler.h +++ b/src/video_core/renderer_vulkan/vk_scheduler.h | |||
| @@ -7,7 +7,6 @@ | |||
| 7 | #include <atomic> | 7 | #include <atomic> |
| 8 | #include <condition_variable> | 8 | #include <condition_variable> |
| 9 | #include <memory> | 9 | #include <memory> |
| 10 | #include <optional> | ||
| 11 | #include <stack> | 10 | #include <stack> |
| 12 | #include <thread> | 11 | #include <thread> |
| 13 | #include <utility> | 12 | #include <utility> |
diff --git a/src/video_core/renderer_vulkan/vk_shader_decompiler.cpp b/src/video_core/renderer_vulkan/vk_shader_decompiler.cpp index 20b6ca0ad..18678968c 100644 --- a/src/video_core/renderer_vulkan/vk_shader_decompiler.cpp +++ b/src/video_core/renderer_vulkan/vk_shader_decompiler.cpp | |||
| @@ -103,8 +103,8 @@ struct GenericVaryingDescription { | |||
| 103 | }; | 103 | }; |
| 104 | 104 | ||
| 105 | spv::Dim GetSamplerDim(const Sampler& sampler) { | 105 | spv::Dim GetSamplerDim(const Sampler& sampler) { |
| 106 | ASSERT(!sampler.IsBuffer()); | 106 | ASSERT(!sampler.is_buffer); |
| 107 | switch (sampler.GetType()) { | 107 | switch (sampler.type) { |
| 108 | case Tegra::Shader::TextureType::Texture1D: | 108 | case Tegra::Shader::TextureType::Texture1D: |
| 109 | return spv::Dim::Dim1D; | 109 | return spv::Dim::Dim1D; |
| 110 | case Tegra::Shader::TextureType::Texture2D: | 110 | case Tegra::Shader::TextureType::Texture2D: |
| @@ -114,13 +114,13 @@ spv::Dim GetSamplerDim(const Sampler& sampler) { | |||
| 114 | case Tegra::Shader::TextureType::TextureCube: | 114 | case Tegra::Shader::TextureType::TextureCube: |
| 115 | return spv::Dim::Cube; | 115 | return spv::Dim::Cube; |
| 116 | default: | 116 | default: |
| 117 | UNIMPLEMENTED_MSG("Unimplemented sampler type={}", static_cast<u32>(sampler.GetType())); | 117 | UNIMPLEMENTED_MSG("Unimplemented sampler type={}", static_cast<int>(sampler.type)); |
| 118 | return spv::Dim::Dim2D; | 118 | return spv::Dim::Dim2D; |
| 119 | } | 119 | } |
| 120 | } | 120 | } |
| 121 | 121 | ||
| 122 | std::pair<spv::Dim, bool> GetImageDim(const Image& image) { | 122 | std::pair<spv::Dim, bool> GetImageDim(const Image& image) { |
| 123 | switch (image.GetType()) { | 123 | switch (image.type) { |
| 124 | case Tegra::Shader::ImageType::Texture1D: | 124 | case Tegra::Shader::ImageType::Texture1D: |
| 125 | return {spv::Dim::Dim1D, false}; | 125 | return {spv::Dim::Dim1D, false}; |
| 126 | case Tegra::Shader::ImageType::TextureBuffer: | 126 | case Tegra::Shader::ImageType::TextureBuffer: |
| @@ -134,7 +134,7 @@ std::pair<spv::Dim, bool> GetImageDim(const Image& image) { | |||
| 134 | case Tegra::Shader::ImageType::Texture3D: | 134 | case Tegra::Shader::ImageType::Texture3D: |
| 135 | return {spv::Dim::Dim3D, false}; | 135 | return {spv::Dim::Dim3D, false}; |
| 136 | default: | 136 | default: |
| 137 | UNIMPLEMENTED_MSG("Unimplemented image type={}", static_cast<u32>(image.GetType())); | 137 | UNIMPLEMENTED_MSG("Unimplemented image type={}", static_cast<int>(image.type)); |
| 138 | return {spv::Dim::Dim2D, false}; | 138 | return {spv::Dim::Dim2D, false}; |
| 139 | } | 139 | } |
| 140 | } | 140 | } |
| @@ -879,11 +879,11 @@ private: | |||
| 879 | 879 | ||
| 880 | u32 DeclareTexelBuffers(u32 binding) { | 880 | u32 DeclareTexelBuffers(u32 binding) { |
| 881 | for (const auto& sampler : ir.GetSamplers()) { | 881 | for (const auto& sampler : ir.GetSamplers()) { |
| 882 | if (!sampler.IsBuffer()) { | 882 | if (!sampler.is_buffer) { |
| 883 | continue; | 883 | continue; |
| 884 | } | 884 | } |
| 885 | ASSERT(!sampler.IsArray()); | 885 | ASSERT(!sampler.is_array); |
| 886 | ASSERT(!sampler.IsShadow()); | 886 | ASSERT(!sampler.is_shadow); |
| 887 | 887 | ||
| 888 | constexpr auto dim = spv::Dim::Buffer; | 888 | constexpr auto dim = spv::Dim::Buffer; |
| 889 | constexpr int depth = 0; | 889 | constexpr int depth = 0; |
| @@ -894,23 +894,23 @@ private: | |||
| 894 | const Id image_type = TypeImage(t_float, dim, depth, arrayed, ms, sampled, format); | 894 | const Id image_type = TypeImage(t_float, dim, depth, arrayed, ms, sampled, format); |
| 895 | const Id pointer_type = TypePointer(spv::StorageClass::UniformConstant, image_type); | 895 | const Id pointer_type = TypePointer(spv::StorageClass::UniformConstant, image_type); |
| 896 | const Id id = OpVariable(pointer_type, spv::StorageClass::UniformConstant); | 896 | const Id id = OpVariable(pointer_type, spv::StorageClass::UniformConstant); |
| 897 | AddGlobalVariable(Name(id, fmt::format("sampler_{}", sampler.GetIndex()))); | 897 | AddGlobalVariable(Name(id, fmt::format("sampler_{}", sampler.index))); |
| 898 | Decorate(id, spv::Decoration::Binding, binding++); | 898 | Decorate(id, spv::Decoration::Binding, binding++); |
| 899 | Decorate(id, spv::Decoration::DescriptorSet, DESCRIPTOR_SET); | 899 | Decorate(id, spv::Decoration::DescriptorSet, DESCRIPTOR_SET); |
| 900 | 900 | ||
| 901 | texel_buffers.emplace(sampler.GetIndex(), TexelBuffer{image_type, id}); | 901 | texel_buffers.emplace(sampler.index, TexelBuffer{image_type, id}); |
| 902 | } | 902 | } |
| 903 | return binding; | 903 | return binding; |
| 904 | } | 904 | } |
| 905 | 905 | ||
| 906 | u32 DeclareSamplers(u32 binding) { | 906 | u32 DeclareSamplers(u32 binding) { |
| 907 | for (const auto& sampler : ir.GetSamplers()) { | 907 | for (const auto& sampler : ir.GetSamplers()) { |
| 908 | if (sampler.IsBuffer()) { | 908 | if (sampler.is_buffer) { |
| 909 | continue; | 909 | continue; |
| 910 | } | 910 | } |
| 911 | const auto dim = GetSamplerDim(sampler); | 911 | const auto dim = GetSamplerDim(sampler); |
| 912 | const int depth = sampler.IsShadow() ? 1 : 0; | 912 | const int depth = sampler.is_shadow ? 1 : 0; |
| 913 | const int arrayed = sampler.IsArray() ? 1 : 0; | 913 | const int arrayed = sampler.is_array ? 1 : 0; |
| 914 | constexpr bool ms = false; | 914 | constexpr bool ms = false; |
| 915 | constexpr int sampled = 1; | 915 | constexpr int sampled = 1; |
| 916 | constexpr auto format = spv::ImageFormat::Unknown; | 916 | constexpr auto format = spv::ImageFormat::Unknown; |
| @@ -918,17 +918,17 @@ private: | |||
| 918 | const Id sampler_type = TypeSampledImage(image_type); | 918 | const Id sampler_type = TypeSampledImage(image_type); |
| 919 | const Id sampler_pointer_type = | 919 | const Id sampler_pointer_type = |
| 920 | TypePointer(spv::StorageClass::UniformConstant, sampler_type); | 920 | TypePointer(spv::StorageClass::UniformConstant, sampler_type); |
| 921 | const Id type = sampler.IsIndexed() | 921 | const Id type = sampler.is_indexed |
| 922 | ? TypeArray(sampler_type, Constant(t_uint, sampler.Size())) | 922 | ? TypeArray(sampler_type, Constant(t_uint, sampler.size)) |
| 923 | : sampler_type; | 923 | : sampler_type; |
| 924 | const Id pointer_type = TypePointer(spv::StorageClass::UniformConstant, type); | 924 | const Id pointer_type = TypePointer(spv::StorageClass::UniformConstant, type); |
| 925 | const Id id = OpVariable(pointer_type, spv::StorageClass::UniformConstant); | 925 | const Id id = OpVariable(pointer_type, spv::StorageClass::UniformConstant); |
| 926 | AddGlobalVariable(Name(id, fmt::format("sampler_{}", sampler.GetIndex()))); | 926 | AddGlobalVariable(Name(id, fmt::format("sampler_{}", sampler.index))); |
| 927 | Decorate(id, spv::Decoration::Binding, binding++); | 927 | Decorate(id, spv::Decoration::Binding, binding++); |
| 928 | Decorate(id, spv::Decoration::DescriptorSet, DESCRIPTOR_SET); | 928 | Decorate(id, spv::Decoration::DescriptorSet, DESCRIPTOR_SET); |
| 929 | 929 | ||
| 930 | sampled_images.emplace(sampler.GetIndex(), SampledImage{image_type, sampler_type, | 930 | sampled_images.emplace( |
| 931 | sampler_pointer_type, id}); | 931 | sampler.index, SampledImage{image_type, sampler_type, sampler_pointer_type, id}); |
| 932 | } | 932 | } |
| 933 | return binding; | 933 | return binding; |
| 934 | } | 934 | } |
| @@ -943,17 +943,17 @@ private: | |||
| 943 | const Id image_type = TypeImage(t_uint, dim, depth, arrayed, ms, sampled, format, {}); | 943 | const Id image_type = TypeImage(t_uint, dim, depth, arrayed, ms, sampled, format, {}); |
| 944 | const Id pointer_type = TypePointer(spv::StorageClass::UniformConstant, image_type); | 944 | const Id pointer_type = TypePointer(spv::StorageClass::UniformConstant, image_type); |
| 945 | const Id id = OpVariable(pointer_type, spv::StorageClass::UniformConstant); | 945 | const Id id = OpVariable(pointer_type, spv::StorageClass::UniformConstant); |
| 946 | AddGlobalVariable(Name(id, fmt::format("image_{}", image.GetIndex()))); | 946 | AddGlobalVariable(Name(id, fmt::format("image_{}", image.index))); |
| 947 | 947 | ||
| 948 | Decorate(id, spv::Decoration::Binding, binding++); | 948 | Decorate(id, spv::Decoration::Binding, binding++); |
| 949 | Decorate(id, spv::Decoration::DescriptorSet, DESCRIPTOR_SET); | 949 | Decorate(id, spv::Decoration::DescriptorSet, DESCRIPTOR_SET); |
| 950 | if (image.IsRead() && !image.IsWritten()) { | 950 | if (image.is_read && !image.is_written) { |
| 951 | Decorate(id, spv::Decoration::NonWritable); | 951 | Decorate(id, spv::Decoration::NonWritable); |
| 952 | } else if (image.IsWritten() && !image.IsRead()) { | 952 | } else if (image.is_written && !image.is_read) { |
| 953 | Decorate(id, spv::Decoration::NonReadable); | 953 | Decorate(id, spv::Decoration::NonReadable); |
| 954 | } | 954 | } |
| 955 | 955 | ||
| 956 | images.emplace(static_cast<u32>(image.GetIndex()), StorageImage{image_type, id}); | 956 | images.emplace(image.index, StorageImage{image_type, id}); |
| 957 | } | 957 | } |
| 958 | return binding; | 958 | return binding; |
| 959 | } | 959 | } |
| @@ -1620,11 +1620,11 @@ private: | |||
| 1620 | 1620 | ||
| 1621 | Id GetTextureSampler(Operation operation) { | 1621 | Id GetTextureSampler(Operation operation) { |
| 1622 | const auto& meta = std::get<MetaTexture>(operation.GetMeta()); | 1622 | const auto& meta = std::get<MetaTexture>(operation.GetMeta()); |
| 1623 | ASSERT(!meta.sampler.IsBuffer()); | 1623 | ASSERT(!meta.sampler.is_buffer); |
| 1624 | 1624 | ||
| 1625 | const auto& entry = sampled_images.at(meta.sampler.GetIndex()); | 1625 | const auto& entry = sampled_images.at(meta.sampler.index); |
| 1626 | Id sampler = entry.variable; | 1626 | Id sampler = entry.variable; |
| 1627 | if (meta.sampler.IsIndexed()) { | 1627 | if (meta.sampler.is_indexed) { |
| 1628 | const Id index = AsInt(Visit(meta.index)); | 1628 | const Id index = AsInt(Visit(meta.index)); |
| 1629 | sampler = OpAccessChain(entry.sampler_pointer_type, sampler, index); | 1629 | sampler = OpAccessChain(entry.sampler_pointer_type, sampler, index); |
| 1630 | } | 1630 | } |
| @@ -1633,8 +1633,8 @@ private: | |||
| 1633 | 1633 | ||
| 1634 | Id GetTextureImage(Operation operation) { | 1634 | Id GetTextureImage(Operation operation) { |
| 1635 | const auto& meta = std::get<MetaTexture>(operation.GetMeta()); | 1635 | const auto& meta = std::get<MetaTexture>(operation.GetMeta()); |
| 1636 | const u32 index = meta.sampler.GetIndex(); | 1636 | const u32 index = meta.sampler.index; |
| 1637 | if (meta.sampler.IsBuffer()) { | 1637 | if (meta.sampler.is_buffer) { |
| 1638 | const auto& entry = texel_buffers.at(index); | 1638 | const auto& entry = texel_buffers.at(index); |
| 1639 | return OpLoad(entry.image_type, entry.image); | 1639 | return OpLoad(entry.image_type, entry.image); |
| 1640 | } else { | 1640 | } else { |
| @@ -1645,7 +1645,7 @@ private: | |||
| 1645 | 1645 | ||
| 1646 | Id GetImage(Operation operation) { | 1646 | Id GetImage(Operation operation) { |
| 1647 | const auto& meta = std::get<MetaImage>(operation.GetMeta()); | 1647 | const auto& meta = std::get<MetaImage>(operation.GetMeta()); |
| 1648 | const auto entry = images.at(meta.image.GetIndex()); | 1648 | const auto entry = images.at(meta.image.index); |
| 1649 | return OpLoad(entry.image_type, entry.image); | 1649 | return OpLoad(entry.image_type, entry.image); |
| 1650 | } | 1650 | } |
| 1651 | 1651 | ||
| @@ -1661,7 +1661,7 @@ private: | |||
| 1661 | } | 1661 | } |
| 1662 | if (const auto meta = std::get_if<MetaTexture>(&operation.GetMeta())) { | 1662 | if (const auto meta = std::get_if<MetaTexture>(&operation.GetMeta())) { |
| 1663 | // Add array coordinate for textures | 1663 | // Add array coordinate for textures |
| 1664 | if (meta->sampler.IsArray()) { | 1664 | if (meta->sampler.is_array) { |
| 1665 | Id array = AsInt(Visit(meta->array)); | 1665 | Id array = AsInt(Visit(meta->array)); |
| 1666 | if (type == Type::Float) { | 1666 | if (type == Type::Float) { |
| 1667 | array = OpConvertSToF(t_float, array); | 1667 | array = OpConvertSToF(t_float, array); |
| @@ -1767,7 +1767,7 @@ private: | |||
| 1767 | operands.push_back(GetOffsetCoordinates(operation)); | 1767 | operands.push_back(GetOffsetCoordinates(operation)); |
| 1768 | } | 1768 | } |
| 1769 | 1769 | ||
| 1770 | if (meta.sampler.IsShadow()) { | 1770 | if (meta.sampler.is_shadow) { |
| 1771 | const Id dref = AsFloat(Visit(meta.depth_compare)); | 1771 | const Id dref = AsFloat(Visit(meta.depth_compare)); |
| 1772 | return {OpImageSampleDrefExplicitLod(t_float, sampler, coords, dref, mask, operands), | 1772 | return {OpImageSampleDrefExplicitLod(t_float, sampler, coords, dref, mask, operands), |
| 1773 | Type::Float}; | 1773 | Type::Float}; |
| @@ -1782,7 +1782,7 @@ private: | |||
| 1782 | 1782 | ||
| 1783 | const Id coords = GetCoordinates(operation, Type::Float); | 1783 | const Id coords = GetCoordinates(operation, Type::Float); |
| 1784 | Id texture{}; | 1784 | Id texture{}; |
| 1785 | if (meta.sampler.IsShadow()) { | 1785 | if (meta.sampler.is_shadow) { |
| 1786 | texture = OpImageDrefGather(t_float4, GetTextureSampler(operation), coords, | 1786 | texture = OpImageDrefGather(t_float4, GetTextureSampler(operation), coords, |
| 1787 | AsFloat(Visit(meta.depth_compare))); | 1787 | AsFloat(Visit(meta.depth_compare))); |
| 1788 | } else { | 1788 | } else { |
| @@ -1809,8 +1809,8 @@ private: | |||
| 1809 | } | 1809 | } |
| 1810 | 1810 | ||
| 1811 | const Id lod = AsUint(Visit(operation[0])); | 1811 | const Id lod = AsUint(Visit(operation[0])); |
| 1812 | const std::size_t coords_count = [&]() { | 1812 | const std::size_t coords_count = [&meta] { |
| 1813 | switch (const auto type = meta.sampler.GetType(); type) { | 1813 | switch (const auto type = meta.sampler.type) { |
| 1814 | case Tegra::Shader::TextureType::Texture1D: | 1814 | case Tegra::Shader::TextureType::Texture1D: |
| 1815 | return 1; | 1815 | return 1; |
| 1816 | case Tegra::Shader::TextureType::Texture2D: | 1816 | case Tegra::Shader::TextureType::Texture2D: |
| @@ -1819,7 +1819,7 @@ private: | |||
| 1819 | case Tegra::Shader::TextureType::Texture3D: | 1819 | case Tegra::Shader::TextureType::Texture3D: |
| 1820 | return 3; | 1820 | return 3; |
| 1821 | default: | 1821 | default: |
| 1822 | UNREACHABLE_MSG("Invalid texture type={}", static_cast<u32>(type)); | 1822 | UNREACHABLE_MSG("Invalid texture type={}", static_cast<int>(type)); |
| 1823 | return 2; | 1823 | return 2; |
| 1824 | } | 1824 | } |
| 1825 | }(); | 1825 | }(); |
| @@ -1862,7 +1862,7 @@ private: | |||
| 1862 | const Id image = GetTextureImage(operation); | 1862 | const Id image = GetTextureImage(operation); |
| 1863 | const Id coords = GetCoordinates(operation, Type::Int); | 1863 | const Id coords = GetCoordinates(operation, Type::Int); |
| 1864 | Id fetch; | 1864 | Id fetch; |
| 1865 | if (meta.lod && !meta.sampler.IsBuffer()) { | 1865 | if (meta.lod && !meta.sampler.is_buffer) { |
| 1866 | fetch = OpImageFetch(t_float4, image, coords, spv::ImageOperandsMask::Lod, | 1866 | fetch = OpImageFetch(t_float4, image, coords, spv::ImageOperandsMask::Lod, |
| 1867 | AsInt(Visit(meta.lod))); | 1867 | AsInt(Visit(meta.lod))); |
| 1868 | } else { | 1868 | } else { |
| @@ -2980,7 +2980,7 @@ ShaderEntries GenerateShaderEntries(const VideoCommon::Shader::ShaderIR& ir) { | |||
| 2980 | entries.global_buffers.emplace_back(base.cbuf_index, base.cbuf_offset, usage.is_written); | 2980 | entries.global_buffers.emplace_back(base.cbuf_index, base.cbuf_offset, usage.is_written); |
| 2981 | } | 2981 | } |
| 2982 | for (const auto& sampler : ir.GetSamplers()) { | 2982 | for (const auto& sampler : ir.GetSamplers()) { |
| 2983 | if (sampler.IsBuffer()) { | 2983 | if (sampler.is_buffer) { |
| 2984 | entries.texel_buffers.emplace_back(sampler); | 2984 | entries.texel_buffers.emplace_back(sampler); |
| 2985 | } else { | 2985 | } else { |
| 2986 | entries.samplers.emplace_back(sampler); | 2986 | entries.samplers.emplace_back(sampler); |
diff --git a/src/video_core/renderer_vulkan/vk_shader_decompiler.h b/src/video_core/renderer_vulkan/vk_shader_decompiler.h index ffea4709e..f4c05ac3c 100644 --- a/src/video_core/renderer_vulkan/vk_shader_decompiler.h +++ b/src/video_core/renderer_vulkan/vk_shader_decompiler.h | |||
| @@ -5,11 +5,7 @@ | |||
| 5 | #pragma once | 5 | #pragma once |
| 6 | 6 | ||
| 7 | #include <array> | 7 | #include <array> |
| 8 | #include <bitset> | ||
| 9 | #include <memory> | ||
| 10 | #include <set> | 8 | #include <set> |
| 11 | #include <type_traits> | ||
| 12 | #include <utility> | ||
| 13 | #include <vector> | 9 | #include <vector> |
| 14 | 10 | ||
| 15 | #include "common/common_types.h" | 11 | #include "common/common_types.h" |
diff --git a/src/video_core/renderer_vulkan/vk_shader_util.cpp b/src/video_core/renderer_vulkan/vk_shader_util.cpp index 784839327..112df9c71 100644 --- a/src/video_core/renderer_vulkan/vk_shader_util.cpp +++ b/src/video_core/renderer_vulkan/vk_shader_util.cpp | |||
| @@ -4,8 +4,7 @@ | |||
| 4 | 4 | ||
| 5 | #include <cstring> | 5 | #include <cstring> |
| 6 | #include <memory> | 6 | #include <memory> |
| 7 | #include <vector> | 7 | |
| 8 | #include "common/alignment.h" | ||
| 9 | #include "common/assert.h" | 8 | #include "common/assert.h" |
| 10 | #include "common/common_types.h" | 9 | #include "common/common_types.h" |
| 11 | #include "video_core/renderer_vulkan/vk_device.h" | 10 | #include "video_core/renderer_vulkan/vk_device.h" |
diff --git a/src/video_core/renderer_vulkan/vk_shader_util.h b/src/video_core/renderer_vulkan/vk_shader_util.h index be38d6697..d1d3f3cae 100644 --- a/src/video_core/renderer_vulkan/vk_shader_util.h +++ b/src/video_core/renderer_vulkan/vk_shader_util.h | |||
| @@ -4,7 +4,6 @@ | |||
| 4 | 4 | ||
| 5 | #pragma once | 5 | #pragma once |
| 6 | 6 | ||
| 7 | #include <vector> | ||
| 8 | #include "common/common_types.h" | 7 | #include "common/common_types.h" |
| 9 | #include "video_core/renderer_vulkan/wrapper.h" | 8 | #include "video_core/renderer_vulkan/wrapper.h" |
| 10 | 9 | ||
diff --git a/src/video_core/renderer_vulkan/vk_staging_buffer_pool.h b/src/video_core/renderer_vulkan/vk_staging_buffer_pool.h index faf6418fd..3c4901437 100644 --- a/src/video_core/renderer_vulkan/vk_staging_buffer_pool.h +++ b/src/video_core/renderer_vulkan/vk_staging_buffer_pool.h | |||
| @@ -5,8 +5,6 @@ | |||
| 5 | #pragma once | 5 | #pragma once |
| 6 | 6 | ||
| 7 | #include <climits> | 7 | #include <climits> |
| 8 | #include <unordered_map> | ||
| 9 | #include <utility> | ||
| 10 | #include <vector> | 8 | #include <vector> |
| 11 | 9 | ||
| 12 | #include "common/common_types.h" | 10 | #include "common/common_types.h" |
diff --git a/src/video_core/renderer_vulkan/vk_texture_cache.cpp b/src/video_core/renderer_vulkan/vk_texture_cache.cpp index de4c23120..55f43e61b 100644 --- a/src/video_core/renderer_vulkan/vk_texture_cache.cpp +++ b/src/video_core/renderer_vulkan/vk_texture_cache.cpp | |||
| @@ -10,11 +10,9 @@ | |||
| 10 | #include <variant> | 10 | #include <variant> |
| 11 | #include <vector> | 11 | #include <vector> |
| 12 | 12 | ||
| 13 | #include "common/alignment.h" | ||
| 14 | #include "common/assert.h" | 13 | #include "common/assert.h" |
| 15 | #include "common/common_types.h" | 14 | #include "common/common_types.h" |
| 16 | #include "core/core.h" | 15 | #include "core/core.h" |
| 17 | #include "core/memory.h" | ||
| 18 | #include "video_core/engines/maxwell_3d.h" | 16 | #include "video_core/engines/maxwell_3d.h" |
| 19 | #include "video_core/morton.h" | 17 | #include "video_core/morton.h" |
| 20 | #include "video_core/renderer_vulkan/maxwell_to_vk.h" | 18 | #include "video_core/renderer_vulkan/maxwell_to_vk.h" |
| @@ -26,7 +24,6 @@ | |||
| 26 | #include "video_core/renderer_vulkan/vk_texture_cache.h" | 24 | #include "video_core/renderer_vulkan/vk_texture_cache.h" |
| 27 | #include "video_core/renderer_vulkan/wrapper.h" | 25 | #include "video_core/renderer_vulkan/wrapper.h" |
| 28 | #include "video_core/surface.h" | 26 | #include "video_core/surface.h" |
| 29 | #include "video_core/textures/convert.h" | ||
| 30 | 27 | ||
| 31 | namespace Vulkan { | 28 | namespace Vulkan { |
| 32 | 29 | ||
diff --git a/src/video_core/renderer_vulkan/vk_texture_cache.h b/src/video_core/renderer_vulkan/vk_texture_cache.h index 115595f28..f211ccb1e 100644 --- a/src/video_core/renderer_vulkan/vk_texture_cache.h +++ b/src/video_core/renderer_vulkan/vk_texture_cache.h | |||
| @@ -7,19 +7,13 @@ | |||
| 7 | #include <memory> | 7 | #include <memory> |
| 8 | #include <unordered_map> | 8 | #include <unordered_map> |
| 9 | 9 | ||
| 10 | #include "common/assert.h" | ||
| 11 | #include "common/common_types.h" | 10 | #include "common/common_types.h" |
| 12 | #include "common/logging/log.h" | ||
| 13 | #include "common/math_util.h" | ||
| 14 | #include "video_core/gpu.h" | ||
| 15 | #include "video_core/rasterizer_cache.h" | ||
| 16 | #include "video_core/renderer_vulkan/vk_image.h" | 11 | #include "video_core/renderer_vulkan/vk_image.h" |
| 17 | #include "video_core/renderer_vulkan/vk_memory_manager.h" | 12 | #include "video_core/renderer_vulkan/vk_memory_manager.h" |
| 18 | #include "video_core/renderer_vulkan/vk_scheduler.h" | 13 | #include "video_core/renderer_vulkan/vk_scheduler.h" |
| 19 | #include "video_core/renderer_vulkan/wrapper.h" | 14 | #include "video_core/renderer_vulkan/wrapper.h" |
| 20 | #include "video_core/texture_cache/surface_base.h" | 15 | #include "video_core/texture_cache/surface_base.h" |
| 21 | #include "video_core/texture_cache/texture_cache.h" | 16 | #include "video_core/texture_cache/texture_cache.h" |
| 22 | #include "video_core/textures/decoders.h" | ||
| 23 | 17 | ||
| 24 | namespace Core { | 18 | namespace Core { |
| 25 | class System; | 19 | class System; |
diff --git a/src/video_core/renderer_vulkan/vk_update_descriptor.h b/src/video_core/renderer_vulkan/vk_update_descriptor.h index 6ba2c9997..cc7e3dff4 100644 --- a/src/video_core/renderer_vulkan/vk_update_descriptor.h +++ b/src/video_core/renderer_vulkan/vk_update_descriptor.h | |||
| @@ -4,7 +4,6 @@ | |||
| 4 | 4 | ||
| 5 | #pragma once | 5 | #pragma once |
| 6 | 6 | ||
| 7 | #include <type_traits> | ||
| 8 | #include <variant> | 7 | #include <variant> |
| 9 | #include <boost/container/static_vector.hpp> | 8 | #include <boost/container/static_vector.hpp> |
| 10 | 9 | ||
diff --git a/src/video_core/shader/decode.cpp b/src/video_core/shader/decode.cpp index 1167ff4ec..a75a5cc63 100644 --- a/src/video_core/shader/decode.cpp +++ b/src/video_core/shader/decode.cpp | |||
| @@ -32,11 +32,11 @@ void DeduceTextureHandlerSize(VideoCore::GuestDriverProfile& gpu_driver, | |||
| 32 | u32 count{}; | 32 | u32 count{}; |
| 33 | std::vector<u32> bound_offsets; | 33 | std::vector<u32> bound_offsets; |
| 34 | for (const auto& sampler : used_samplers) { | 34 | for (const auto& sampler : used_samplers) { |
| 35 | if (sampler.IsBindless()) { | 35 | if (sampler.is_bindless) { |
| 36 | continue; | 36 | continue; |
| 37 | } | 37 | } |
| 38 | ++count; | 38 | ++count; |
| 39 | bound_offsets.emplace_back(sampler.GetOffset()); | 39 | bound_offsets.emplace_back(sampler.offset); |
| 40 | } | 40 | } |
| 41 | if (count > 1) { | 41 | if (count > 1) { |
| 42 | gpu_driver.DeduceTextureHandlerSize(std::move(bound_offsets)); | 42 | gpu_driver.DeduceTextureHandlerSize(std::move(bound_offsets)); |
| @@ -46,14 +46,14 @@ void DeduceTextureHandlerSize(VideoCore::GuestDriverProfile& gpu_driver, | |||
| 46 | std::optional<u32> TryDeduceSamplerSize(const Sampler& sampler_to_deduce, | 46 | std::optional<u32> TryDeduceSamplerSize(const Sampler& sampler_to_deduce, |
| 47 | VideoCore::GuestDriverProfile& gpu_driver, | 47 | VideoCore::GuestDriverProfile& gpu_driver, |
| 48 | const std::list<Sampler>& used_samplers) { | 48 | const std::list<Sampler>& used_samplers) { |
| 49 | const u32 base_offset = sampler_to_deduce.GetOffset(); | 49 | const u32 base_offset = sampler_to_deduce.offset; |
| 50 | u32 max_offset{std::numeric_limits<u32>::max()}; | 50 | u32 max_offset{std::numeric_limits<u32>::max()}; |
| 51 | for (const auto& sampler : used_samplers) { | 51 | for (const auto& sampler : used_samplers) { |
| 52 | if (sampler.IsBindless()) { | 52 | if (sampler.is_bindless) { |
| 53 | continue; | 53 | continue; |
| 54 | } | 54 | } |
| 55 | if (sampler.GetOffset() > base_offset) { | 55 | if (sampler.offset > base_offset) { |
| 56 | max_offset = std::min(sampler.GetOffset(), max_offset); | 56 | max_offset = std::min(sampler.offset, max_offset); |
| 57 | } | 57 | } |
| 58 | } | 58 | } |
| 59 | if (max_offset == std::numeric_limits<u32>::max()) { | 59 | if (max_offset == std::numeric_limits<u32>::max()) { |
| @@ -353,14 +353,14 @@ void ShaderIR::PostDecode() { | |||
| 353 | return; | 353 | return; |
| 354 | } | 354 | } |
| 355 | for (auto& sampler : used_samplers) { | 355 | for (auto& sampler : used_samplers) { |
| 356 | if (!sampler.IsIndexed()) { | 356 | if (!sampler.is_indexed) { |
| 357 | continue; | 357 | continue; |
| 358 | } | 358 | } |
| 359 | if (const auto size = TryDeduceSamplerSize(sampler, gpu_driver, used_samplers)) { | 359 | if (const auto size = TryDeduceSamplerSize(sampler, gpu_driver, used_samplers)) { |
| 360 | sampler.SetSize(*size); | 360 | sampler.size = *size; |
| 361 | } else { | 361 | } else { |
| 362 | LOG_CRITICAL(HW_GPU, "Failed to deduce size of indexed sampler"); | 362 | LOG_CRITICAL(HW_GPU, "Failed to deduce size of indexed sampler"); |
| 363 | sampler.SetSize(1); | 363 | sampler.size = 1; |
| 364 | } | 364 | } |
| 365 | } | 365 | } |
| 366 | } | 366 | } |
diff --git a/src/video_core/shader/decode/image.cpp b/src/video_core/shader/decode/image.cpp index 85ee9aa5e..60b6ad72a 100644 --- a/src/video_core/shader/decode/image.cpp +++ b/src/video_core/shader/decode/image.cpp | |||
| @@ -485,11 +485,10 @@ u32 ShaderIR::DecodeImage(NodeBlock& bb, u32 pc) { | |||
| 485 | Image& ShaderIR::GetImage(Tegra::Shader::Image image, Tegra::Shader::ImageType type) { | 485 | Image& ShaderIR::GetImage(Tegra::Shader::Image image, Tegra::Shader::ImageType type) { |
| 486 | const auto offset = static_cast<u32>(image.index.Value()); | 486 | const auto offset = static_cast<u32>(image.index.Value()); |
| 487 | 487 | ||
| 488 | const auto it = | 488 | const auto it = std::find_if(std::begin(used_images), std::end(used_images), |
| 489 | std::find_if(std::begin(used_images), std::end(used_images), | 489 | [offset](const Image& entry) { return entry.offset == offset; }); |
| 490 | [offset](const Image& entry) { return entry.GetOffset() == offset; }); | ||
| 491 | if (it != std::end(used_images)) { | 490 | if (it != std::end(used_images)) { |
| 492 | ASSERT(!it->IsBindless() && it->GetType() == it->GetType()); | 491 | ASSERT(!it->is_bindless && it->type == type); |
| 493 | return *it; | 492 | return *it; |
| 494 | } | 493 | } |
| 495 | 494 | ||
| @@ -505,13 +504,12 @@ Image& ShaderIR::GetBindlessImage(Tegra::Shader::Register reg, Tegra::Shader::Im | |||
| 505 | const auto buffer = std::get<1>(result); | 504 | const auto buffer = std::get<1>(result); |
| 506 | const auto offset = std::get<2>(result); | 505 | const auto offset = std::get<2>(result); |
| 507 | 506 | ||
| 508 | const auto it = | 507 | const auto it = std::find_if(std::begin(used_images), std::end(used_images), |
| 509 | std::find_if(std::begin(used_images), std::end(used_images), | 508 | [buffer, offset](const Image& entry) { |
| 510 | [buffer = buffer, offset = offset](const Image& entry) { | 509 | return entry.buffer == buffer && entry.offset == offset; |
| 511 | return entry.GetBuffer() == buffer && entry.GetOffset() == offset; | 510 | }); |
| 512 | }); | ||
| 513 | if (it != std::end(used_images)) { | 511 | if (it != std::end(used_images)) { |
| 514 | ASSERT(it->IsBindless() && it->GetType() == it->GetType()); | 512 | ASSERT(it->is_bindless && it->type == type); |
| 515 | return *it; | 513 | return *it; |
| 516 | } | 514 | } |
| 517 | 515 | ||
diff --git a/src/video_core/shader/decode/texture.cpp b/src/video_core/shader/decode/texture.cpp index e68f1d305..8f0bb996e 100644 --- a/src/video_core/shader/decode/texture.cpp +++ b/src/video_core/shader/decode/texture.cpp | |||
| @@ -139,15 +139,15 @@ u32 ShaderIR::DecodeTexture(NodeBlock& bb, u32 pc) { | |||
| 139 | } | 139 | } |
| 140 | const Node component = Immediate(static_cast<u32>(instr.tld4s.component)); | 140 | const Node component = Immediate(static_cast<u32>(instr.tld4s.component)); |
| 141 | 141 | ||
| 142 | const SamplerInfo info{TextureType::Texture2D, false, is_depth_compare, false}; | 142 | SamplerInfo info; |
| 143 | const Sampler& sampler = *GetSampler(instr.sampler, info); | 143 | info.is_shadow = is_depth_compare; |
| 144 | const std::optional<Sampler> sampler = GetSampler(instr.sampler, info); | ||
| 144 | 145 | ||
| 145 | Node4 values; | 146 | Node4 values; |
| 146 | for (u32 element = 0; element < values.size(); ++element) { | 147 | for (u32 element = 0; element < values.size(); ++element) { |
| 147 | auto coords_copy = coords; | 148 | MetaTexture meta{*sampler, {}, depth_compare, aoffi, {}, {}, |
| 148 | MetaTexture meta{sampler, {}, depth_compare, aoffi, {}, {}, | 149 | {}, {}, component, element, {}}; |
| 149 | {}, {}, component, element, {}}; | 150 | values[element] = Operation(OperationCode::TextureGather, meta, coords); |
| 150 | values[element] = Operation(OperationCode::TextureGather, meta, std::move(coords_copy)); | ||
| 151 | } | 151 | } |
| 152 | 152 | ||
| 153 | if (instr.tld4s.fp16_flag) { | 153 | if (instr.tld4s.fp16_flag) { |
| @@ -165,18 +165,20 @@ u32 ShaderIR::DecodeTexture(NodeBlock& bb, u32 pc) { | |||
| 165 | "AOFFI is not implemented"); | 165 | "AOFFI is not implemented"); |
| 166 | 166 | ||
| 167 | const bool is_array = instr.txd.is_array != 0; | 167 | const bool is_array = instr.txd.is_array != 0; |
| 168 | u64 base_reg = instr.gpr8.Value(); | ||
| 169 | const auto derivate_reg = instr.gpr20.Value(); | 168 | const auto derivate_reg = instr.gpr20.Value(); |
| 170 | const auto texture_type = instr.txd.texture_type.Value(); | 169 | const auto texture_type = instr.txd.texture_type.Value(); |
| 171 | const auto coord_count = GetCoordCount(texture_type); | 170 | const auto coord_count = GetCoordCount(texture_type); |
| 172 | Node index_var{}; | 171 | u64 base_reg = instr.gpr8.Value(); |
| 173 | const Sampler* sampler = | 172 | Node index_var; |
| 174 | is_bindless | 173 | SamplerInfo info; |
| 175 | ? GetBindlessSampler(base_reg, index_var, {{texture_type, is_array, false, false}}) | 174 | info.type = texture_type; |
| 176 | : GetSampler(instr.sampler, {{texture_type, is_array, false, false}}); | 175 | info.is_array = is_array; |
| 176 | const std::optional<Sampler> sampler = is_bindless | ||
| 177 | ? GetBindlessSampler(base_reg, info, index_var) | ||
| 178 | : GetSampler(instr.sampler, info); | ||
| 177 | Node4 values; | 179 | Node4 values; |
| 178 | if (sampler == nullptr) { | 180 | if (!sampler) { |
| 179 | std::generate(values.begin(), values.end(), [] { return Immediate(0); }); | 181 | std::generate(values.begin(), values.end(), [this] { return Immediate(0); }); |
| 180 | WriteTexInstructionFloat(bb, instr, values); | 182 | WriteTexInstructionFloat(bb, instr, values); |
| 181 | break; | 183 | break; |
| 182 | } | 184 | } |
| @@ -214,14 +216,12 @@ u32 ShaderIR::DecodeTexture(NodeBlock& bb, u32 pc) { | |||
| 214 | is_bindless = true; | 216 | is_bindless = true; |
| 215 | [[fallthrough]]; | 217 | [[fallthrough]]; |
| 216 | case OpCode::Id::TXQ: { | 218 | case OpCode::Id::TXQ: { |
| 217 | // TODO: The new commits on the texture refactor, change the way samplers work. | 219 | Node index_var; |
| 218 | // Sadly, not all texture instructions specify the type of texture their sampler | 220 | const std::optional<Sampler> sampler = is_bindless |
| 219 | // uses. This must be fixed at a later instance. | 221 | ? GetBindlessSampler(instr.gpr8, {}, index_var) |
| 220 | Node index_var{}; | 222 | : GetSampler(instr.sampler, {}); |
| 221 | const Sampler* sampler = | 223 | |
| 222 | is_bindless ? GetBindlessSampler(instr.gpr8, index_var) : GetSampler(instr.sampler); | 224 | if (!sampler) { |
| 223 | |||
| 224 | if (sampler == nullptr) { | ||
| 225 | u32 indexer = 0; | 225 | u32 indexer = 0; |
| 226 | for (u32 element = 0; element < 4; ++element) { | 226 | for (u32 element = 0; element < 4; ++element) { |
| 227 | if (!instr.txq.IsComponentEnabled(element)) { | 227 | if (!instr.txq.IsComponentEnabled(element)) { |
| @@ -267,12 +267,17 @@ u32 ShaderIR::DecodeTexture(NodeBlock& bb, u32 pc) { | |||
| 267 | UNIMPLEMENTED_IF_MSG(instr.tmml.UsesMiscMode(Tegra::Shader::TextureMiscMode::NDV), | 267 | UNIMPLEMENTED_IF_MSG(instr.tmml.UsesMiscMode(Tegra::Shader::TextureMiscMode::NDV), |
| 268 | "NDV is not implemented"); | 268 | "NDV is not implemented"); |
| 269 | 269 | ||
| 270 | auto texture_type = instr.tmml.texture_type.Value(); | 270 | const auto texture_type = instr.tmml.texture_type.Value(); |
| 271 | Node index_var{}; | 271 | const bool is_array = instr.tmml.array != 0; |
| 272 | const Sampler* sampler = | 272 | SamplerInfo info; |
| 273 | is_bindless ? GetBindlessSampler(instr.gpr20, index_var) : GetSampler(instr.sampler); | 273 | info.type = texture_type; |
| 274 | 274 | info.is_array = is_array; | |
| 275 | if (sampler == nullptr) { | 275 | Node index_var; |
| 276 | const std::optional<Sampler> sampler = | ||
| 277 | is_bindless ? GetBindlessSampler(instr.gpr20, info, index_var) | ||
| 278 | : GetSampler(instr.sampler, info); | ||
| 279 | |||
| 280 | if (!sampler) { | ||
| 276 | u32 indexer = 0; | 281 | u32 indexer = 0; |
| 277 | for (u32 element = 0; element < 2; ++element) { | 282 | for (u32 element = 0; element < 2; ++element) { |
| 278 | if (!instr.tmml.IsComponentEnabled(element)) { | 283 | if (!instr.tmml.IsComponentEnabled(element)) { |
| @@ -299,12 +304,11 @@ u32 ShaderIR::DecodeTexture(NodeBlock& bb, u32 pc) { | |||
| 299 | coords.push_back(GetRegister(instr.gpr8.Value() + 1)); | 304 | coords.push_back(GetRegister(instr.gpr8.Value() + 1)); |
| 300 | break; | 305 | break; |
| 301 | default: | 306 | default: |
| 302 | UNIMPLEMENTED_MSG("Unhandled texture type {}", static_cast<u32>(texture_type)); | 307 | UNIMPLEMENTED_MSG("Unhandled texture type {}", static_cast<int>(texture_type)); |
| 303 | 308 | ||
| 304 | // Fallback to interpreting as a 2D texture for now | 309 | // Fallback to interpreting as a 2D texture for now |
| 305 | coords.push_back(GetRegister(instr.gpr8.Value() + 0)); | 310 | coords.push_back(GetRegister(instr.gpr8.Value() + 0)); |
| 306 | coords.push_back(GetRegister(instr.gpr8.Value() + 1)); | 311 | coords.push_back(GetRegister(instr.gpr8.Value() + 1)); |
| 307 | texture_type = TextureType::Texture2D; | ||
| 308 | } | 312 | } |
| 309 | u32 indexer = 0; | 313 | u32 indexer = 0; |
| 310 | for (u32 element = 0; element < 2; ++element) { | 314 | for (u32 element = 0; element < 2; ++element) { |
| @@ -353,98 +357,103 @@ u32 ShaderIR::DecodeTexture(NodeBlock& bb, u32 pc) { | |||
| 353 | return pc; | 357 | return pc; |
| 354 | } | 358 | } |
| 355 | 359 | ||
| 356 | ShaderIR::SamplerInfo ShaderIR::GetSamplerInfo(std::optional<SamplerInfo> sampler_info, u32 offset, | 360 | ShaderIR::SamplerInfo ShaderIR::GetSamplerInfo(SamplerInfo info, u32 offset, |
| 357 | std::optional<u32> buffer) { | 361 | std::optional<u32> buffer) { |
| 358 | if (sampler_info) { | 362 | if (info.IsComplete()) { |
| 359 | return *sampler_info; | 363 | return info; |
| 360 | } | 364 | } |
| 361 | const auto sampler = buffer ? registry.ObtainBindlessSampler(*buffer, offset) | 365 | const auto sampler = buffer ? registry.ObtainBindlessSampler(*buffer, offset) |
| 362 | : registry.ObtainBoundSampler(offset); | 366 | : registry.ObtainBoundSampler(offset); |
| 363 | if (!sampler) { | 367 | if (!sampler) { |
| 364 | LOG_WARNING(HW_GPU, "Unknown sampler info"); | 368 | LOG_WARNING(HW_GPU, "Unknown sampler info"); |
| 365 | return SamplerInfo{TextureType::Texture2D, false, false, false}; | 369 | info.type = info.type.value_or(Tegra::Shader::TextureType::Texture2D); |
| 366 | } | 370 | info.is_array = info.is_array.value_or(false); |
| 367 | return SamplerInfo{sampler->texture_type, sampler->is_array != 0, sampler->is_shadow != 0, | 371 | info.is_shadow = info.is_shadow.value_or(false); |
| 368 | sampler->is_buffer != 0}; | 372 | info.is_buffer = info.is_buffer.value_or(false); |
| 373 | return info; | ||
| 374 | } | ||
| 375 | info.type = info.type.value_or(sampler->texture_type); | ||
| 376 | info.is_array = info.is_array.value_or(sampler->is_array != 0); | ||
| 377 | info.is_shadow = info.is_shadow.value_or(sampler->is_shadow != 0); | ||
| 378 | info.is_buffer = info.is_buffer.value_or(sampler->is_buffer != 0); | ||
| 379 | return info; | ||
| 369 | } | 380 | } |
| 370 | 381 | ||
| 371 | const Sampler* ShaderIR::GetSampler(const Tegra::Shader::Sampler& sampler, | 382 | std::optional<Sampler> ShaderIR::GetSampler(Tegra::Shader::Sampler sampler, |
| 372 | std::optional<SamplerInfo> sampler_info) { | 383 | SamplerInfo sampler_info) { |
| 373 | const auto offset = static_cast<u32>(sampler.index.Value()); | 384 | const auto offset = static_cast<u32>(sampler.index.Value()); |
| 374 | const auto info = GetSamplerInfo(sampler_info, offset); | 385 | const auto info = GetSamplerInfo(sampler_info, offset); |
| 375 | 386 | ||
| 376 | // If this sampler has already been used, return the existing mapping. | 387 | // If this sampler has already been used, return the existing mapping. |
| 377 | const auto it = | 388 | const auto it = std::find_if(used_samplers.begin(), used_samplers.end(), |
| 378 | std::find_if(used_samplers.begin(), used_samplers.end(), | 389 | [offset](const Sampler& entry) { return entry.offset == offset; }); |
| 379 | [offset](const Sampler& entry) { return entry.GetOffset() == offset; }); | ||
| 380 | if (it != used_samplers.end()) { | 390 | if (it != used_samplers.end()) { |
| 381 | ASSERT(!it->IsBindless() && it->GetType() == info.type && it->IsArray() == info.is_array && | 391 | ASSERT(!it->is_bindless && it->type == info.type && it->is_array == info.is_array && |
| 382 | it->IsShadow() == info.is_shadow && it->IsBuffer() == info.is_buffer); | 392 | it->is_shadow == info.is_shadow && it->is_buffer == info.is_buffer); |
| 383 | return &*it; | 393 | return *it; |
| 384 | } | 394 | } |
| 385 | 395 | ||
| 386 | // Otherwise create a new mapping for this sampler | 396 | // Otherwise create a new mapping for this sampler |
| 387 | const auto next_index = static_cast<u32>(used_samplers.size()); | 397 | const auto next_index = static_cast<u32>(used_samplers.size()); |
| 388 | return &used_samplers.emplace_back(next_index, offset, info.type, info.is_array, info.is_shadow, | 398 | return used_samplers.emplace_back(next_index, offset, *info.type, *info.is_array, |
| 389 | info.is_buffer, false); | 399 | *info.is_shadow, *info.is_buffer, false); |
| 390 | } | 400 | } |
| 391 | 401 | ||
| 392 | const Sampler* ShaderIR::GetBindlessSampler(Tegra::Shader::Register reg, Node& index_var, | 402 | std::optional<Sampler> ShaderIR::GetBindlessSampler(Tegra::Shader::Register reg, SamplerInfo info, |
| 393 | std::optional<SamplerInfo> sampler_info) { | 403 | Node& index_var) { |
| 394 | const Node sampler_register = GetRegister(reg); | 404 | const Node sampler_register = GetRegister(reg); |
| 395 | const auto [base_node, tracked_sampler_info] = | 405 | const auto [base_node, tracked_sampler_info] = |
| 396 | TrackBindlessSampler(sampler_register, global_code, static_cast<s64>(global_code.size())); | 406 | TrackBindlessSampler(sampler_register, global_code, static_cast<s64>(global_code.size())); |
| 397 | ASSERT(base_node != nullptr); | 407 | ASSERT(base_node != nullptr); |
| 398 | if (base_node == nullptr) { | 408 | if (base_node == nullptr) { |
| 399 | return nullptr; | 409 | return std::nullopt; |
| 400 | } | 410 | } |
| 401 | 411 | ||
| 402 | if (const auto bindless_sampler_info = | 412 | if (const auto bindless_sampler_info = |
| 403 | std::get_if<BindlessSamplerNode>(&*tracked_sampler_info)) { | 413 | std::get_if<BindlessSamplerNode>(&*tracked_sampler_info)) { |
| 404 | const u32 buffer = bindless_sampler_info->GetIndex(); | 414 | const u32 buffer = bindless_sampler_info->GetIndex(); |
| 405 | const u32 offset = bindless_sampler_info->GetOffset(); | 415 | const u32 offset = bindless_sampler_info->GetOffset(); |
| 406 | const auto info = GetSamplerInfo(sampler_info, offset, buffer); | 416 | info = GetSamplerInfo(info, offset, buffer); |
| 407 | 417 | ||
| 408 | // If this sampler has already been used, return the existing mapping. | 418 | // If this sampler has already been used, return the existing mapping. |
| 409 | const auto it = | 419 | const auto it = std::find_if(used_samplers.begin(), used_samplers.end(), |
| 410 | std::find_if(used_samplers.begin(), used_samplers.end(), | 420 | [buffer = buffer, offset = offset](const Sampler& entry) { |
| 411 | [buffer = buffer, offset = offset](const Sampler& entry) { | 421 | return entry.buffer == buffer && entry.offset == offset; |
| 412 | return entry.GetBuffer() == buffer && entry.GetOffset() == offset; | 422 | }); |
| 413 | }); | ||
| 414 | if (it != used_samplers.end()) { | 423 | if (it != used_samplers.end()) { |
| 415 | ASSERT(it->IsBindless() && it->GetType() == info.type && | 424 | ASSERT(it->is_bindless && it->type == info.type && it->is_array == info.is_array && |
| 416 | it->IsArray() == info.is_array && it->IsShadow() == info.is_shadow); | 425 | it->is_shadow == info.is_shadow); |
| 417 | return &*it; | 426 | return *it; |
| 418 | } | 427 | } |
| 419 | 428 | ||
| 420 | // Otherwise create a new mapping for this sampler | 429 | // Otherwise create a new mapping for this sampler |
| 421 | const auto next_index = static_cast<u32>(used_samplers.size()); | 430 | const auto next_index = static_cast<u32>(used_samplers.size()); |
| 422 | return &used_samplers.emplace_back(next_index, offset, buffer, info.type, info.is_array, | 431 | return used_samplers.emplace_back(next_index, offset, buffer, *info.type, *info.is_array, |
| 423 | info.is_shadow, info.is_buffer, false); | 432 | *info.is_shadow, *info.is_buffer, false); |
| 424 | } else if (const auto array_sampler_info = | 433 | } |
| 425 | std::get_if<ArraySamplerNode>(&*tracked_sampler_info)) { | 434 | if (const auto array_sampler_info = std::get_if<ArraySamplerNode>(&*tracked_sampler_info)) { |
| 426 | const u32 base_offset = array_sampler_info->GetBaseOffset() / 4; | 435 | const u32 base_offset = array_sampler_info->GetBaseOffset() / 4; |
| 427 | index_var = GetCustomVariable(array_sampler_info->GetIndexVar()); | 436 | index_var = GetCustomVariable(array_sampler_info->GetIndexVar()); |
| 428 | const auto info = GetSamplerInfo(sampler_info, base_offset); | 437 | info = GetSamplerInfo(info, base_offset); |
| 429 | 438 | ||
| 430 | // If this sampler has already been used, return the existing mapping. | 439 | // If this sampler has already been used, return the existing mapping. |
| 431 | const auto it = std::find_if( | 440 | const auto it = std::find_if( |
| 432 | used_samplers.begin(), used_samplers.end(), | 441 | used_samplers.begin(), used_samplers.end(), |
| 433 | [base_offset](const Sampler& entry) { return entry.GetOffset() == base_offset; }); | 442 | [base_offset](const Sampler& entry) { return entry.offset == base_offset; }); |
| 434 | if (it != used_samplers.end()) { | 443 | if (it != used_samplers.end()) { |
| 435 | ASSERT(!it->IsBindless() && it->GetType() == info.type && | 444 | ASSERT(!it->is_bindless && it->type == info.type && it->is_array == info.is_array && |
| 436 | it->IsArray() == info.is_array && it->IsShadow() == info.is_shadow && | 445 | it->is_shadow == info.is_shadow && it->is_buffer == info.is_buffer && |
| 437 | it->IsBuffer() == info.is_buffer && it->IsIndexed()); | 446 | it->is_indexed); |
| 438 | return &*it; | 447 | return *it; |
| 439 | } | 448 | } |
| 440 | 449 | ||
| 441 | uses_indexed_samplers = true; | 450 | uses_indexed_samplers = true; |
| 442 | // Otherwise create a new mapping for this sampler | 451 | // Otherwise create a new mapping for this sampler |
| 443 | const auto next_index = static_cast<u32>(used_samplers.size()); | 452 | const auto next_index = static_cast<u32>(used_samplers.size()); |
| 444 | return &used_samplers.emplace_back(next_index, base_offset, info.type, info.is_array, | 453 | return used_samplers.emplace_back(next_index, base_offset, *info.type, *info.is_array, |
| 445 | info.is_shadow, info.is_buffer, true); | 454 | *info.is_shadow, *info.is_buffer, true); |
| 446 | } | 455 | } |
| 447 | return nullptr; | 456 | return std::nullopt; |
| 448 | } | 457 | } |
| 449 | 458 | ||
| 450 | void ShaderIR::WriteTexInstructionFloat(NodeBlock& bb, Instruction instr, const Node4& components) { | 459 | void ShaderIR::WriteTexInstructionFloat(NodeBlock& bb, Instruction instr, const Node4& components) { |
| @@ -529,10 +538,16 @@ Node4 ShaderIR::GetTextureCode(Instruction instr, TextureType texture_type, | |||
| 529 | ASSERT_MSG(texture_type != TextureType::Texture3D || !is_array || !is_shadow, | 538 | ASSERT_MSG(texture_type != TextureType::Texture3D || !is_array || !is_shadow, |
| 530 | "Illegal texture type"); | 539 | "Illegal texture type"); |
| 531 | 540 | ||
| 532 | const SamplerInfo info{texture_type, is_array, is_shadow, false}; | 541 | SamplerInfo info; |
| 542 | info.type = texture_type; | ||
| 543 | info.is_array = is_array; | ||
| 544 | info.is_shadow = is_shadow; | ||
| 545 | info.is_buffer = false; | ||
| 546 | |||
| 533 | Node index_var; | 547 | Node index_var; |
| 534 | const Sampler* sampler = is_bindless ? GetBindlessSampler(*bindless_reg, index_var, info) | 548 | const std::optional<Sampler> sampler = is_bindless |
| 535 | : GetSampler(instr.sampler, info); | 549 | ? GetBindlessSampler(*bindless_reg, info, index_var) |
| 550 | : GetSampler(instr.sampler, info); | ||
| 536 | if (!sampler) { | 551 | if (!sampler) { |
| 537 | return {Immediate(0), Immediate(0), Immediate(0), Immediate(0)}; | 552 | return {Immediate(0), Immediate(0), Immediate(0), Immediate(0)}; |
| 538 | } | 553 | } |
| @@ -683,12 +698,17 @@ Node4 ShaderIR::GetTld4Code(Instruction instr, TextureType texture_type, bool de | |||
| 683 | 698 | ||
| 684 | u64 parameter_register = instr.gpr20.Value(); | 699 | u64 parameter_register = instr.gpr20.Value(); |
| 685 | 700 | ||
| 686 | const SamplerInfo info{texture_type, is_array, depth_compare, false}; | 701 | SamplerInfo info; |
| 687 | Node index_var{}; | 702 | info.type = texture_type; |
| 688 | const Sampler* sampler = is_bindless ? GetBindlessSampler(parameter_register++, index_var, info) | 703 | info.is_array = is_array; |
| 689 | : GetSampler(instr.sampler, info); | 704 | info.is_shadow = depth_compare; |
| 705 | |||
| 706 | Node index_var; | ||
| 707 | const std::optional<Sampler> sampler = | ||
| 708 | is_bindless ? GetBindlessSampler(parameter_register++, info, index_var) | ||
| 709 | : GetSampler(instr.sampler, info); | ||
| 690 | Node4 values; | 710 | Node4 values; |
| 691 | if (sampler == nullptr) { | 711 | if (!sampler) { |
| 692 | for (u32 element = 0; element < values.size(); ++element) { | 712 | for (u32 element = 0; element < values.size(); ++element) { |
| 693 | values[element] = Immediate(0); | 713 | values[element] = Immediate(0); |
| 694 | } | 714 | } |
| @@ -743,12 +763,12 @@ Node4 ShaderIR::GetTldCode(Tegra::Shader::Instruction instr) { | |||
| 743 | // const Node aoffi_register{is_aoffi ? GetRegister(gpr20_cursor++) : nullptr}; | 763 | // const Node aoffi_register{is_aoffi ? GetRegister(gpr20_cursor++) : nullptr}; |
| 744 | // const Node multisample{is_multisample ? GetRegister(gpr20_cursor++) : nullptr}; | 764 | // const Node multisample{is_multisample ? GetRegister(gpr20_cursor++) : nullptr}; |
| 745 | 765 | ||
| 746 | const auto& sampler = *GetSampler(instr.sampler); | 766 | const std::optional<Sampler> sampler = GetSampler(instr.sampler, {}); |
| 747 | 767 | ||
| 748 | Node4 values; | 768 | Node4 values; |
| 749 | for (u32 element = 0; element < values.size(); ++element) { | 769 | for (u32 element = 0; element < values.size(); ++element) { |
| 750 | auto coords_copy = coords; | 770 | auto coords_copy = coords; |
| 751 | MetaTexture meta{sampler, array_register, {}, {}, {}, {}, {}, lod, {}, element, {}}; | 771 | MetaTexture meta{*sampler, array_register, {}, {}, {}, {}, {}, lod, {}, element, {}}; |
| 752 | values[element] = Operation(OperationCode::TexelFetch, meta, std::move(coords_copy)); | 772 | values[element] = Operation(OperationCode::TexelFetch, meta, std::move(coords_copy)); |
| 753 | } | 773 | } |
| 754 | 774 | ||
| @@ -756,7 +776,11 @@ Node4 ShaderIR::GetTldCode(Tegra::Shader::Instruction instr) { | |||
| 756 | } | 776 | } |
| 757 | 777 | ||
| 758 | Node4 ShaderIR::GetTldsCode(Instruction instr, TextureType texture_type, bool is_array) { | 778 | Node4 ShaderIR::GetTldsCode(Instruction instr, TextureType texture_type, bool is_array) { |
| 759 | const Sampler& sampler = *GetSampler(instr.sampler); | 779 | SamplerInfo info; |
| 780 | info.type = texture_type; | ||
| 781 | info.is_array = is_array; | ||
| 782 | info.is_shadow = false; | ||
| 783 | const std::optional<Sampler> sampler = GetSampler(instr.sampler, info); | ||
| 760 | 784 | ||
| 761 | const std::size_t type_coord_count = GetCoordCount(texture_type); | 785 | const std::size_t type_coord_count = GetCoordCount(texture_type); |
| 762 | const bool lod_enabled = instr.tlds.GetTextureProcessMode() == TextureProcessMode::LL; | 786 | const bool lod_enabled = instr.tlds.GetTextureProcessMode() == TextureProcessMode::LL; |
| @@ -784,7 +808,7 @@ Node4 ShaderIR::GetTldsCode(Instruction instr, TextureType texture_type, bool is | |||
| 784 | Node4 values; | 808 | Node4 values; |
| 785 | for (u32 element = 0; element < values.size(); ++element) { | 809 | for (u32 element = 0; element < values.size(); ++element) { |
| 786 | auto coords_copy = coords; | 810 | auto coords_copy = coords; |
| 787 | MetaTexture meta{sampler, array, {}, {}, {}, {}, {}, lod, {}, element, {}}; | 811 | MetaTexture meta{*sampler, array, {}, {}, {}, {}, {}, lod, {}, element, {}}; |
| 788 | values[element] = Operation(OperationCode::TexelFetch, meta, std::move(coords_copy)); | 812 | values[element] = Operation(OperationCode::TexelFetch, meta, std::move(coords_copy)); |
| 789 | } | 813 | } |
| 790 | return values; | 814 | return values; |
diff --git a/src/video_core/shader/node.h b/src/video_core/shader/node.h index 3f5a7bc7a..601c822d2 100644 --- a/src/video_core/shader/node.h +++ b/src/video_core/shader/node.h | |||
| @@ -267,76 +267,30 @@ class ArraySamplerNode; | |||
| 267 | using TrackSamplerData = std::variant<BindlessSamplerNode, ArraySamplerNode>; | 267 | using TrackSamplerData = std::variant<BindlessSamplerNode, ArraySamplerNode>; |
| 268 | using TrackSampler = std::shared_ptr<TrackSamplerData>; | 268 | using TrackSampler = std::shared_ptr<TrackSamplerData>; |
| 269 | 269 | ||
| 270 | class Sampler { | 270 | struct Sampler { |
| 271 | public: | 271 | /// Bound samplers constructor |
| 272 | /// This constructor is for bound samplers | ||
| 273 | constexpr explicit Sampler(u32 index, u32 offset, Tegra::Shader::TextureType type, | 272 | constexpr explicit Sampler(u32 index, u32 offset, Tegra::Shader::TextureType type, |
| 274 | bool is_array, bool is_shadow, bool is_buffer, bool is_indexed) | 273 | bool is_array, bool is_shadow, bool is_buffer, bool is_indexed) |
| 275 | : index{index}, offset{offset}, type{type}, is_array{is_array}, is_shadow{is_shadow}, | 274 | : index{index}, offset{offset}, type{type}, is_array{is_array}, is_shadow{is_shadow}, |
| 276 | is_buffer{is_buffer}, is_indexed{is_indexed} {} | 275 | is_buffer{is_buffer}, is_indexed{is_indexed} {} |
| 277 | 276 | ||
| 278 | /// This constructor is for bindless samplers | 277 | /// Bindless samplers constructor |
| 279 | constexpr explicit Sampler(u32 index, u32 offset, u32 buffer, Tegra::Shader::TextureType type, | 278 | constexpr explicit Sampler(u32 index, u32 offset, u32 buffer, Tegra::Shader::TextureType type, |
| 280 | bool is_array, bool is_shadow, bool is_buffer, bool is_indexed) | 279 | bool is_array, bool is_shadow, bool is_buffer, bool is_indexed) |
| 281 | : index{index}, offset{offset}, buffer{buffer}, type{type}, is_array{is_array}, | 280 | : index{index}, offset{offset}, buffer{buffer}, type{type}, is_array{is_array}, |
| 282 | is_shadow{is_shadow}, is_buffer{is_buffer}, is_bindless{true}, is_indexed{is_indexed} {} | 281 | is_shadow{is_shadow}, is_buffer{is_buffer}, is_bindless{true}, is_indexed{is_indexed} {} |
| 283 | 282 | ||
| 284 | constexpr u32 GetIndex() const { | 283 | u32 index = 0; ///< Emulated index given for the this sampler. |
| 285 | return index; | 284 | u32 offset = 0; ///< Offset in the const buffer from where the sampler is being read. |
| 286 | } | 285 | u32 buffer = 0; ///< Buffer where the bindless sampler is being read (unused on bound samplers). |
| 287 | 286 | u32 size = 1; ///< Size of the sampler. | |
| 288 | constexpr u32 GetOffset() const { | ||
| 289 | return offset; | ||
| 290 | } | ||
| 291 | |||
| 292 | constexpr u32 GetBuffer() const { | ||
| 293 | return buffer; | ||
| 294 | } | ||
| 295 | |||
| 296 | constexpr Tegra::Shader::TextureType GetType() const { | ||
| 297 | return type; | ||
| 298 | } | ||
| 299 | |||
| 300 | constexpr bool IsArray() const { | ||
| 301 | return is_array; | ||
| 302 | } | ||
| 303 | |||
| 304 | constexpr bool IsShadow() const { | ||
| 305 | return is_shadow; | ||
| 306 | } | ||
| 307 | |||
| 308 | constexpr bool IsBuffer() const { | ||
| 309 | return is_buffer; | ||
| 310 | } | ||
| 311 | |||
| 312 | constexpr bool IsBindless() const { | ||
| 313 | return is_bindless; | ||
| 314 | } | ||
| 315 | |||
| 316 | constexpr bool IsIndexed() const { | ||
| 317 | return is_indexed; | ||
| 318 | } | ||
| 319 | |||
| 320 | constexpr u32 Size() const { | ||
| 321 | return size; | ||
| 322 | } | ||
| 323 | |||
| 324 | constexpr void SetSize(u32 new_size) { | ||
| 325 | size = new_size; | ||
| 326 | } | ||
| 327 | |||
| 328 | private: | ||
| 329 | u32 index{}; ///< Emulated index given for the this sampler. | ||
| 330 | u32 offset{}; ///< Offset in the const buffer from where the sampler is being read. | ||
| 331 | u32 buffer{}; ///< Buffer where the bindless sampler is being read (unused on bound samplers). | ||
| 332 | u32 size{1}; ///< Size of the sampler. | ||
| 333 | 287 | ||
| 334 | Tegra::Shader::TextureType type{}; ///< The type used to sample this texture (Texture2D, etc) | 288 | Tegra::Shader::TextureType type{}; ///< The type used to sample this texture (Texture2D, etc) |
| 335 | bool is_array{}; ///< Whether the texture is being sampled as an array texture or not. | 289 | bool is_array = false; ///< Whether the texture is being sampled as an array texture or not. |
| 336 | bool is_shadow{}; ///< Whether the texture is being sampled as a depth texture or not. | 290 | bool is_shadow = false; ///< Whether the texture is being sampled as a depth texture or not. |
| 337 | bool is_buffer{}; ///< Whether the texture is a texture buffer without sampler. | 291 | bool is_buffer = false; ///< Whether the texture is a texture buffer without sampler. |
| 338 | bool is_bindless{}; ///< Whether this sampler belongs to a bindless texture or not. | 292 | bool is_bindless = false; ///< Whether this sampler belongs to a bindless texture or not. |
| 339 | bool is_indexed{}; ///< Whether this sampler is an indexed array of textures. | 293 | bool is_indexed = false; ///< Whether this sampler is an indexed array of textures. |
| 340 | }; | 294 | }; |
| 341 | 295 | ||
| 342 | /// Represents a tracked bindless sampler into a direct const buffer | 296 | /// Represents a tracked bindless sampler into a direct const buffer |
| @@ -381,13 +335,13 @@ private: | |||
| 381 | u32 offset; | 335 | u32 offset; |
| 382 | }; | 336 | }; |
| 383 | 337 | ||
| 384 | class Image final { | 338 | struct Image { |
| 385 | public: | 339 | public: |
| 386 | /// This constructor is for bound images | 340 | /// Bound images constructor |
| 387 | constexpr explicit Image(u32 index, u32 offset, Tegra::Shader::ImageType type) | 341 | constexpr explicit Image(u32 index, u32 offset, Tegra::Shader::ImageType type) |
| 388 | : index{index}, offset{offset}, type{type} {} | 342 | : index{index}, offset{offset}, type{type} {} |
| 389 | 343 | ||
| 390 | /// This constructor is for bindless samplers | 344 | /// Bindless samplers constructor |
| 391 | constexpr explicit Image(u32 index, u32 offset, u32 buffer, Tegra::Shader::ImageType type) | 345 | constexpr explicit Image(u32 index, u32 offset, u32 buffer, Tegra::Shader::ImageType type) |
| 392 | : index{index}, offset{offset}, buffer{buffer}, type{type}, is_bindless{true} {} | 346 | : index{index}, offset{offset}, buffer{buffer}, type{type}, is_bindless{true} {} |
| 393 | 347 | ||
| @@ -405,53 +359,20 @@ public: | |||
| 405 | is_atomic = true; | 359 | is_atomic = true; |
| 406 | } | 360 | } |
| 407 | 361 | ||
| 408 | constexpr u32 GetIndex() const { | 362 | u32 index = 0; |
| 409 | return index; | 363 | u32 offset = 0; |
| 410 | } | 364 | u32 buffer = 0; |
| 411 | |||
| 412 | constexpr u32 GetOffset() const { | ||
| 413 | return offset; | ||
| 414 | } | ||
| 415 | |||
| 416 | constexpr u32 GetBuffer() const { | ||
| 417 | return buffer; | ||
| 418 | } | ||
| 419 | |||
| 420 | constexpr Tegra::Shader::ImageType GetType() const { | ||
| 421 | return type; | ||
| 422 | } | ||
| 423 | |||
| 424 | constexpr bool IsBindless() const { | ||
| 425 | return is_bindless; | ||
| 426 | } | ||
| 427 | |||
| 428 | constexpr bool IsWritten() const { | ||
| 429 | return is_written; | ||
| 430 | } | ||
| 431 | |||
| 432 | constexpr bool IsRead() const { | ||
| 433 | return is_read; | ||
| 434 | } | ||
| 435 | |||
| 436 | constexpr bool IsAtomic() const { | ||
| 437 | return is_atomic; | ||
| 438 | } | ||
| 439 | |||
| 440 | private: | ||
| 441 | u32 index{}; | ||
| 442 | u32 offset{}; | ||
| 443 | u32 buffer{}; | ||
| 444 | 365 | ||
| 445 | Tegra::Shader::ImageType type{}; | 366 | Tegra::Shader::ImageType type{}; |
| 446 | bool is_bindless{}; | 367 | bool is_bindless = false; |
| 447 | bool is_written{}; | 368 | bool is_written = false; |
| 448 | bool is_read{}; | 369 | bool is_read = false; |
| 449 | bool is_atomic{}; | 370 | bool is_atomic = false; |
| 450 | }; | 371 | }; |
| 451 | 372 | ||
| 452 | struct GlobalMemoryBase { | 373 | struct GlobalMemoryBase { |
| 453 | u32 cbuf_index{}; | 374 | u32 cbuf_index = 0; |
| 454 | u32 cbuf_offset{}; | 375 | u32 cbuf_offset = 0; |
| 455 | 376 | ||
| 456 | bool operator<(const GlobalMemoryBase& rhs) const { | 377 | bool operator<(const GlobalMemoryBase& rhs) const { |
| 457 | return std::tie(cbuf_index, cbuf_offset) < std::tie(rhs.cbuf_index, rhs.cbuf_offset); | 378 | return std::tie(cbuf_index, cbuf_offset) < std::tie(rhs.cbuf_index, rhs.cbuf_offset); |
| @@ -465,7 +386,7 @@ struct MetaArithmetic { | |||
| 465 | 386 | ||
| 466 | /// Parameters describing a texture sampler | 387 | /// Parameters describing a texture sampler |
| 467 | struct MetaTexture { | 388 | struct MetaTexture { |
| 468 | const Sampler& sampler; | 389 | Sampler sampler; |
| 469 | Node array; | 390 | Node array; |
| 470 | Node depth_compare; | 391 | Node depth_compare; |
| 471 | std::vector<Node> aoffi; | 392 | std::vector<Node> aoffi; |
diff --git a/src/video_core/shader/shader_ir.h b/src/video_core/shader/shader_ir.h index 69de5e68b..15ae152f2 100644 --- a/src/video_core/shader/shader_ir.h +++ b/src/video_core/shader/shader_ir.h | |||
| @@ -28,12 +28,11 @@ struct ShaderBlock; | |||
| 28 | 28 | ||
| 29 | constexpr u32 MAX_PROGRAM_LENGTH = 0x1000; | 29 | constexpr u32 MAX_PROGRAM_LENGTH = 0x1000; |
| 30 | 30 | ||
| 31 | class ConstBuffer { | 31 | struct ConstBuffer { |
| 32 | public: | 32 | constexpr explicit ConstBuffer(u32 max_offset, bool is_indirect) |
| 33 | explicit ConstBuffer(u32 max_offset, bool is_indirect) | ||
| 34 | : max_offset{max_offset}, is_indirect{is_indirect} {} | 33 | : max_offset{max_offset}, is_indirect{is_indirect} {} |
| 35 | 34 | ||
| 36 | ConstBuffer() = default; | 35 | constexpr ConstBuffer() = default; |
| 37 | 36 | ||
| 38 | void MarkAsUsed(u64 offset) { | 37 | void MarkAsUsed(u64 offset) { |
| 39 | max_offset = std::max(max_offset, static_cast<u32>(offset)); | 38 | max_offset = std::max(max_offset, static_cast<u32>(offset)); |
| @@ -56,8 +55,8 @@ public: | |||
| 56 | } | 55 | } |
| 57 | 56 | ||
| 58 | private: | 57 | private: |
| 59 | u32 max_offset{}; | 58 | u32 max_offset = 0; |
| 60 | bool is_indirect{}; | 59 | bool is_indirect = false; |
| 61 | }; | 60 | }; |
| 62 | 61 | ||
| 63 | struct GlobalMemoryUsage { | 62 | struct GlobalMemoryUsage { |
| @@ -191,10 +190,14 @@ private: | |||
| 191 | friend class ASTDecoder; | 190 | friend class ASTDecoder; |
| 192 | 191 | ||
| 193 | struct SamplerInfo { | 192 | struct SamplerInfo { |
| 194 | Tegra::Shader::TextureType type; | 193 | std::optional<Tegra::Shader::TextureType> type; |
| 195 | bool is_array; | 194 | std::optional<bool> is_array; |
| 196 | bool is_shadow; | 195 | std::optional<bool> is_shadow; |
| 197 | bool is_buffer; | 196 | std::optional<bool> is_buffer; |
| 197 | |||
| 198 | constexpr bool IsComplete() const noexcept { | ||
| 199 | return type && is_array && is_shadow && is_buffer; | ||
| 200 | } | ||
| 198 | }; | 201 | }; |
| 199 | 202 | ||
| 200 | void Decode(); | 203 | void Decode(); |
| @@ -327,16 +330,15 @@ private: | |||
| 327 | OperationCode GetPredicateCombiner(Tegra::Shader::PredOperation operation); | 330 | OperationCode GetPredicateCombiner(Tegra::Shader::PredOperation operation); |
| 328 | 331 | ||
| 329 | /// Queries the missing sampler info from the execution context. | 332 | /// Queries the missing sampler info from the execution context. |
| 330 | SamplerInfo GetSamplerInfo(std::optional<SamplerInfo> sampler_info, u32 offset, | 333 | SamplerInfo GetSamplerInfo(SamplerInfo info, u32 offset, |
| 331 | std::optional<u32> buffer = std::nullopt); | 334 | std::optional<u32> buffer = std::nullopt); |
| 332 | 335 | ||
| 333 | /// Accesses a texture sampler | 336 | /// Accesses a texture sampler. |
| 334 | const Sampler* GetSampler(const Tegra::Shader::Sampler& sampler, | 337 | std::optional<Sampler> GetSampler(Tegra::Shader::Sampler sampler, SamplerInfo info); |
| 335 | std::optional<SamplerInfo> sampler_info = std::nullopt); | ||
| 336 | 338 | ||
| 337 | /// Accesses a texture sampler for a bindless texture. | 339 | /// Accesses a texture sampler for a bindless texture. |
| 338 | const Sampler* GetBindlessSampler(Tegra::Shader::Register reg, Node& index_var, | 340 | std::optional<Sampler> GetBindlessSampler(Tegra::Shader::Register reg, SamplerInfo info, |
| 339 | std::optional<SamplerInfo> sampler_info = std::nullopt); | 341 | Node& index_var); |
| 340 | 342 | ||
| 341 | /// Accesses an image. | 343 | /// Accesses an image. |
| 342 | Image& GetImage(Tegra::Shader::Image image, Tegra::Shader::ImageType type); | 344 | Image& GetImage(Tegra::Shader::Image image, Tegra::Shader::ImageType type); |
diff --git a/src/video_core/texture_cache/surface_params.cpp b/src/video_core/texture_cache/surface_params.cpp index 0de499946..884fabffe 100644 --- a/src/video_core/texture_cache/surface_params.cpp +++ b/src/video_core/texture_cache/surface_params.cpp | |||
| @@ -81,7 +81,7 @@ SurfaceParams SurfaceParams::CreateForTexture(const FormatLookupTable& lookup_ta | |||
| 81 | params.pixel_format = lookup_table.GetPixelFormat( | 81 | params.pixel_format = lookup_table.GetPixelFormat( |
| 82 | tic.format, params.srgb_conversion, tic.r_type, tic.g_type, tic.b_type, tic.a_type); | 82 | tic.format, params.srgb_conversion, tic.r_type, tic.g_type, tic.b_type, tic.a_type); |
| 83 | params.type = GetFormatType(params.pixel_format); | 83 | params.type = GetFormatType(params.pixel_format); |
| 84 | if (entry.IsShadow() && params.type == SurfaceType::ColorTexture) { | 84 | if (entry.is_shadow && params.type == SurfaceType::ColorTexture) { |
| 85 | switch (params.pixel_format) { | 85 | switch (params.pixel_format) { |
| 86 | case PixelFormat::R16U: | 86 | case PixelFormat::R16U: |
| 87 | case PixelFormat::R16F: | 87 | case PixelFormat::R16F: |
| @@ -108,7 +108,7 @@ SurfaceParams SurfaceParams::CreateForTexture(const FormatLookupTable& lookup_ta | |||
| 108 | params.emulated_levels = 1; | 108 | params.emulated_levels = 1; |
| 109 | params.is_layered = false; | 109 | params.is_layered = false; |
| 110 | } else { | 110 | } else { |
| 111 | params.target = TextureTypeToSurfaceTarget(entry.GetType(), entry.IsArray()); | 111 | params.target = TextureTypeToSurfaceTarget(entry.type, entry.is_array); |
| 112 | params.width = tic.Width(); | 112 | params.width = tic.Width(); |
| 113 | params.height = tic.Height(); | 113 | params.height = tic.Height(); |
| 114 | params.depth = tic.Depth(); | 114 | params.depth = tic.Depth(); |
| @@ -138,7 +138,7 @@ SurfaceParams SurfaceParams::CreateForImage(const FormatLookupTable& lookup_tabl | |||
| 138 | tic.format, params.srgb_conversion, tic.r_type, tic.g_type, tic.b_type, tic.a_type); | 138 | tic.format, params.srgb_conversion, tic.r_type, tic.g_type, tic.b_type, tic.a_type); |
| 139 | params.type = GetFormatType(params.pixel_format); | 139 | params.type = GetFormatType(params.pixel_format); |
| 140 | params.type = GetFormatType(params.pixel_format); | 140 | params.type = GetFormatType(params.pixel_format); |
| 141 | params.target = ImageTypeToSurfaceTarget(entry.GetType()); | 141 | params.target = ImageTypeToSurfaceTarget(entry.type); |
| 142 | // TODO: on 1DBuffer we should use the tic info. | 142 | // TODO: on 1DBuffer we should use the tic info. |
| 143 | if (tic.IsBuffer()) { | 143 | if (tic.IsBuffer()) { |
| 144 | params.target = SurfaceTarget::TextureBuffer; | 144 | params.target = SurfaceTarget::TextureBuffer; |
| @@ -248,12 +248,12 @@ SurfaceParams SurfaceParams::CreateForFermiCopySurface( | |||
| 248 | 248 | ||
| 249 | VideoCore::Surface::SurfaceTarget SurfaceParams::ExpectedTarget( | 249 | VideoCore::Surface::SurfaceTarget SurfaceParams::ExpectedTarget( |
| 250 | const VideoCommon::Shader::Sampler& entry) { | 250 | const VideoCommon::Shader::Sampler& entry) { |
| 251 | return TextureTypeToSurfaceTarget(entry.GetType(), entry.IsArray()); | 251 | return TextureTypeToSurfaceTarget(entry.type, entry.is_array); |
| 252 | } | 252 | } |
| 253 | 253 | ||
| 254 | VideoCore::Surface::SurfaceTarget SurfaceParams::ExpectedTarget( | 254 | VideoCore::Surface::SurfaceTarget SurfaceParams::ExpectedTarget( |
| 255 | const VideoCommon::Shader::Image& entry) { | 255 | const VideoCommon::Shader::Image& entry) { |
| 256 | return ImageTypeToSurfaceTarget(entry.GetType()); | 256 | return ImageTypeToSurfaceTarget(entry.type); |
| 257 | } | 257 | } |
| 258 | 258 | ||
| 259 | bool SurfaceParams::IsLayered() const { | 259 | bool SurfaceParams::IsLayered() const { |
diff --git a/src/video_core/texture_cache/texture_cache.h b/src/video_core/texture_cache/texture_cache.h index d2d2846e6..d6efc34b2 100644 --- a/src/video_core/texture_cache/texture_cache.h +++ b/src/video_core/texture_cache/texture_cache.h | |||
| @@ -1172,7 +1172,7 @@ private: | |||
| 1172 | /// Returns true the shader sampler entry is compatible with the TIC texture type. | 1172 | /// Returns true the shader sampler entry is compatible with the TIC texture type. |
| 1173 | static bool IsTypeCompatible(Tegra::Texture::TextureType tic_type, | 1173 | static bool IsTypeCompatible(Tegra::Texture::TextureType tic_type, |
| 1174 | const VideoCommon::Shader::Sampler& entry) { | 1174 | const VideoCommon::Shader::Sampler& entry) { |
| 1175 | const auto shader_type = entry.GetType(); | 1175 | const auto shader_type = entry.type; |
| 1176 | switch (tic_type) { | 1176 | switch (tic_type) { |
| 1177 | case Tegra::Texture::TextureType::Texture1D: | 1177 | case Tegra::Texture::TextureType::Texture1D: |
| 1178 | case Tegra::Texture::TextureType::Texture1DArray: | 1178 | case Tegra::Texture::TextureType::Texture1DArray: |
| @@ -1193,7 +1193,7 @@ private: | |||
| 1193 | if (shader_type == Tegra::Shader::TextureType::TextureCube) { | 1193 | if (shader_type == Tegra::Shader::TextureType::TextureCube) { |
| 1194 | return true; | 1194 | return true; |
| 1195 | } | 1195 | } |
| 1196 | return shader_type == Tegra::Shader::TextureType::Texture2D && entry.IsArray(); | 1196 | return shader_type == Tegra::Shader::TextureType::Texture2D && entry.is_array; |
| 1197 | } | 1197 | } |
| 1198 | UNREACHABLE(); | 1198 | UNREACHABLE(); |
| 1199 | return true; | 1199 | return true; |