frozenleaves/pytorch
1
0
Fork 0
mirror of https://github.com/pytorch/pytorch.git synced 2025-10-20 21:14:14 +08:00
Code Issues Packages Projects Releases Wiki Activity
Files
0fd976b65c6daf3799a501d9202e4f50144446d1
pytorch/c10/core
History
Murray Steele 0fd976b65c Enable mimalloc on non-Windows platforms and make default for AArch64 builds (#164741) 
This change removes the Windows requirement for mimalloc builds, and makes mimalloc the default c10 system allocator for AArch64 builds. This significantly improves the performance of AArch64 builds of PyTorch as large allocations are better cached by mimalloc than glibc.

**Updated Results**

Torchbench FP32 eager Inference, 16 threads:
<img width="1510" height="733" alt="mimalloc-v2-fp32-diff" src="https://github.com/user-attachments/assets/7fe3ea0c-3b52-42e7-879b-612444479c90" />

Torchbench BF16 eager Inference, 16 threads:
<img width="1510" height="733" alt="mimalloc-v2-bf16-diff" src="https://github.com/user-attachments/assets/56469a72-9e06-4d57-ae2a-aeb139ca79a3" />

Pull Request resolved: https://github.com/pytorch/pytorch/pull/164741
Approved by: https://github.com/fadara01, https://github.com/aditew01, https://github.com/malfet
2025-10-09 20:49:46 +00:00
..
impl
Enable mimalloc on non-Windows platforms and make default for AArch64 builds (#164741)
2025-10-09 20:49:46 +00:00
alignment.h
[8/N] Fixes clang-tidy warnings in c10/{core,util}/*.h (#116082)
2023-12-20 12:22:21 +00:00
Allocator.cpp
Switch c10::string_view to std::string_view (#139635)
2024-11-27 01:41:18 +00:00
Allocator.h
[sigmoid] memory planner C10 deps (#151275)
2025-04-24 01:46:32 +00:00
AllocatorConfig.cpp
Revert "Generalize torch._C._set_allocator_settings to be generic (#156175)" (#161626)
2025-08-27 21:37:14 +00:00
AllocatorConfig.h
Revert "Generalize torch._C._set_allocator_settings to be generic (#156175)" (#161626)
2025-08-27 21:37:14 +00:00
AutogradState.cpp
[Typo] mulithreading_enabled => multithreading_enabled (#97054)
2023-03-21 20:11:59 +00:00
AutogradState.h
[13/N] Fix extra warnings brought by clang-tidy-17 (#140897)
2024-11-27 00:35:19 +00:00
Backend.h
[MPS] coalesce for sparse tensors (#159729)
2025-08-08 13:49:55 +00:00
BUILD.bazel
extract //c10/core:ScalarType to its own library (#70855)
2022-01-26 16:31:26 +00:00
build.bzl
Organize BUCK for torch/standalone (#156503)
2025-06-25 22:56:15 +00:00
CachingDeviceAllocator.cpp
Add DeviceAllocator as the base device allocator (#138222)
2025-08-08 17:41:10 +00:00
CachingDeviceAllocator.h
Add DeviceAllocator as the base device allocator (#138222)
2025-08-08 17:41:10 +00:00
CompileTimeFunctionPointer.h
[7/N] Fixes clang-tidy warnings in c10/{core,util}/*.h (#115495)
2023-12-19 02:14:30 +00:00
ConstantSymNodeImpl.cpp
Add is_nested_int() (#119975)
2024-02-21 21:10:02 +00:00
ConstantSymNodeImpl.h
[3/N] Fix Wunused-parameter warnings (#131271)
2024-07-20 23:31:03 +00:00
Contiguity.h
symbolic cpp channels_last_contiguous (#160402)
2025-09-05 21:40:32 +00:00
CopyBytes.cpp
[PyTorch] Autoformat c10 (#56830)
2021-04-30 21:23:28 -07:00
CopyBytes.h
[7/N] Fixes clang-tidy warnings in c10/{core,util}/*.h (#115495)
2023-12-19 02:14:30 +00:00
CPUAllocator.cpp
Mark unused parameters in C++ code (#164912)
2025-10-09 06:23:25 +00:00
CPUAllocator.h
Mark unused parameters in C++ code (#164912)
2025-10-09 06:23:25 +00:00
DefaultDtype.cpp
Disable avoid-non-const-global-variables lint check (#62008)
2021-07-22 18:04:40 -07:00
DefaultDtype.h
[T153220354] Fix header inclusions in c10 (#1541) (#101846)
2023-05-20 19:35:14 +00:00
DefaultTensorOptions.h
[5/N] Fixes clang-tidy warnings in c10/{core,util}/*.h (#115354)
2023-12-09 17:16:04 +00:00
Device.cpp
[Device] Add support for PrivateUse1 device type in parse_type function (#157609)
2025-07-17 01:27:44 +00:00
Device.h
[PyTorch Core] MTIA supports arbitrary strides (#157883)
2025-07-11 18:54:21 +00:00
DeviceArray.h
std::value/std::type -> std::_v/std::_t (#138746)
2024-10-26 20:59:24 +00:00
DeviceGuard.h
[2/N] Fix clang-tidy readability checks (#164652)
2025-10-06 01:06:01 +00:00
DeviceType.cpp
Fix: Use memory_order_relaxed instead of memory_order_relaxed (#159105)
2025-07-25 19:39:04 +00:00
DeviceType.h
Migrate DeviceType to torch/headeronly (#163999)
2025-09-30 23:13:27 +00:00
DispatchKey.cpp
[MPS] Add boilerplate sparse code support (#157238)
2025-06-30 01:53:45 +00:00
DispatchKey.h
Mark unused parameters in C++ code (#164912)
2025-10-09 06:23:25 +00:00
DispatchKeySet.cpp
Revert "Do not decompose in functionalization/proxy tensor if autograd wouldn't have decomposed (#164939)"
2025-10-09 09:50:59 +00:00
DispatchKeySet.h
Mark unused parameters in C++ code (#164912)
2025-10-09 06:23:25 +00:00
DynamicCast.h
Support unsigned int for randint, item, equality, fill, iinfo, tensor (#116805)
2024-01-10 02:17:23 +00:00
Event.h
Don't return values in void functions (#164809)
2025-10-08 01:04:14 +00:00
GeneratorImpl.cpp
Avoid std::chrono::system_clock (#153135)
2025-05-08 16:30:29 +00:00
GeneratorImpl.h
[1/N] Enable cppcoreguidelines-special-member-functions (#137405)
2024-10-23 00:16:53 +00:00
GradMode.cpp
cleanup unused include (#93359)
2023-02-04 02:15:50 +00:00
GradMode.h
[1/N] Enable cppcoreguidelines-special-member-functions (#137405)
2024-10-23 00:16:53 +00:00
InferenceMode.cpp
cleanup unused include (#93359)
2023-02-04 02:15:50 +00:00
InferenceMode.h
[1/N] Enable cppcoreguidelines-special-member-functions (#137405)
2024-10-23 00:16:53 +00:00
Layout.h
[MPS] coalesce for sparse tensors (#159729)
2025-08-08 13:49:55 +00:00
MemoryFormat.h
[7/N] Fixes clang-tidy warnings in c10/{core,util}/*.h (#115495)
2023-12-19 02:14:30 +00:00
OptionalRef.h
clamp: port to structured kernel (#61361)
2021-07-23 02:02:07 -07:00
PyHandleCache.h
[7/N] Fixes clang-tidy warnings in c10/{core,util}/*.h (#115495)
2023-12-19 02:14:30 +00:00
QEngine.h
[7/N] Fixes clang-tidy warnings in c10/{core,util}/*.h (#115495)
2023-12-19 02:14:30 +00:00
QScheme.h
[7/N] Fixes clang-tidy warnings in c10/{core,util}/*.h (#115495)
2023-12-19 02:14:30 +00:00
RefcountedDeleter.cpp
[18/N] Fix extra warnings brought by clang-tidy-17 (#144014)
2025-01-08 17:21:55 +00:00
RefcountedDeleter.h
[BE] fix typos in c10/ (#156078)
2025-06-18 10:24:44 +00:00
SafePyObject.cpp
cleanup unused include (#93359)
2023-02-04 02:15:50 +00:00
SafePyObject.h
Mark unused parameters in C++ code (#164912)
2025-10-09 06:23:25 +00:00
Scalar.cpp
Add SymInt to Scalar (#84958)
2022-09-25 23:51:06 +00:00
Scalar.h
Mark unused parameters in C++ code (#164912)
2025-10-09 06:23:25 +00:00
ScalarType.cpp
add torch.float4_e2m1fn_x2 to PyTorch (#148791)
2025-03-27 17:32:20 +00:00
ScalarType.h
[Reland] Migrate ScalarType to headeronly (#159911)
2025-08-06 07:36:37 +00:00
ScalarTypeToTypeMeta.h
[9/N] Replace c10::optional with std::optional (#130674)
2024-07-15 00:48:43 +00:00
Storage.cpp
Reland 2: Add PyObject preservation for UntypedStorage (#109039)
2023-09-12 22:26:05 +00:00
Storage.h
Mark unused parameters in C++ code (#164912)
2025-10-09 06:23:25 +00:00
StorageImpl.cpp
Turn some variables and functions into static (#136847)
2024-10-29 17:01:56 +00:00
StorageImpl.h
[sigmoid] memory planner C10 deps (#151275)
2025-04-24 01:46:32 +00:00
Stream.cpp
Add query and synchronize to c10::Stream (#59560)
2021-06-10 01:42:40 -07:00
Stream.h
Mark unused parameters in C++ code (#164912)
2025-10-09 06:23:25 +00:00
StreamGuard.h
[2/N] Fix clang-tidy readability checks (#164652)
2025-10-06 01:06:01 +00:00
SymbolicShapeMeta.cpp
symbolic cpp channels_last_contiguous (#160402)
2025-09-05 21:40:32 +00:00
SymbolicShapeMeta.h
[attempt 2] Compute contiguity symbolically to avoid dde, and introduce c++ sym_is_contiguous (#157472)
2025-07-02 23:12:29 +00:00
SymBool.cpp
Support C++ statically_known_true (#151346)
2025-04-18 06:42:12 +00:00
SymBool.h
Mark unused parameters in C++ code (#164912)
2025-10-09 06:23:25 +00:00
SymFloat.cpp
fix incorrect c10::SymFloat::sqrt (#141728)
2024-12-03 23:34:16 +00:00
SymFloat.h
Mark unused parameters in C++ code (#164912)
2025-10-09 06:23:25 +00:00
SymInt.cpp
Remove unused argument from DEFINE_BINARY macro. (#163868)
2025-09-28 06:32:41 +00:00
SymInt.h
Mark unused parameters in C++ code (#164912)
2025-10-09 06:23:25 +00:00
SymIntArrayRef.cpp
Inline asIntArrayRef (#86350)
2022-10-06 14:55:03 +00:00
SymIntArrayRef.h
[1/N] Change #include <c10/util/Optional.h> to #include <optional> (#128301)
2024-07-08 07:03:53 +00:00
SymNodeImpl.cpp
Unify SymIntNode and SymFloatNode into SymNode (#87817)
2022-10-27 20:56:02 +00:00
SymNodeImpl.h
[ez] Make SymNodeImpl comments less misleading (#154480)
2025-05-28 14:04:32 +00:00
TensorImpl.cpp
Remove guard_size_oblivious from default contiguity python check, and add aten.sym_is_contiguous. [attempt2] (#160869)
2025-09-08 22:59:13 +00:00
TensorImpl.h
Mark unused parameters in C++ code (#164912)
2025-10-09 06:23:25 +00:00
TensorOptions.cpp
[PyTorch] Redirect c10::optional to std::optional (#101995)
2023-11-30 02:46:41 +00:00
TensorOptions.h
Enable readability-redundant-declaration (#143982)
2024-12-31 00:20:10 +00:00
thread_pool.cpp
Fix extra semicolon warning (#148291)
2025-03-03 18:51:44 +00:00
thread_pool.h
[7/N] Fixes clang-tidy warnings in c10/{core,util}/*.h (#115495)
2023-12-19 02:14:30 +00:00
UndefinedTensorImpl.cpp
Mark unused parameters in C++ code (#164912)
2025-10-09 06:23:25 +00:00
UndefinedTensorImpl.h
[attempt 2] Compute contiguity symbolically to avoid dde, and introduce c++ sym_is_contiguous (#157472)
2025-07-02 23:12:29 +00:00
WrapDimMinimal.cpp
Bump Clang-tidy to 19.1.4 (#148648)
2025-03-10 17:32:30 +00:00
WrapDimMinimal.h
[7/N] Fixes clang-tidy warnings in c10/{core,util}/*.h (#115495)
2023-12-19 02:14:30 +00:00