Boyuan Feng eecc8e362c [Inductor] Inplacing with Donated Buffer (#140113) ...

Currently, inductor does not inplace update a buffer if it is an input buffer. Because we don't know if an input will be used by other functions.

Donated buffer provides additional information that an input buffer will not be used by other functions. So we can inplace update donated buffer when possible.

[Dashboard](https://hud.pytorch.org/benchmark/torchbench/inductor_dynamic?dashboard=torchinductor&startTime=Mon,%2011%20Nov%202024%2018:14:36%20GMT&stopTime=Mon,%2018%20Nov%202024%2018:14:36%20GMT&granularity=hour&mode=training&dtype=amp&deviceName=cuda%20(a100)&lBranch=bf/donated-buffer-inplace&lCommit=5df0769c00e6f9000caeb10fd5cbf0b165f69c2a&rBranch=main&rCommit=2b39a8db7741b816b03677a9c6fec1af05640dee)

![image](https://github.com/user-attachments/assets/f19d961f-7973-418e-9de8-5c2a97950478)
![image](https://github.com/user-attachments/assets/df3bd6a9-58b8-4e8a-8397-9e3b1de9adfe)

Pull Request resolved: https://github.com/pytorch/pytorch/pull/140113
Approved by: https://github.com/eellison

2024-11-26 17:19:50 +00:00

..

aoti_runtime

[AOTI][refactor] Separate header codegen (#138882)

2024-10-27 14:14:27 +00:00

cuda

[CUTLASS] Lift shape & stride information as kernel args (#138611)

2024-11-25 17:52:33 +00:00

rocm

[ROCm][Inductor][CK] Enable scaled mm with bias in gemm max autotune with CK backend (#140674)

2024-11-15 22:08:38 +00:00

xpu

[AOTI XPU] Enable Cpp wraper for Intel GPU. (#135318)

2024-11-26 11:51:32 +00:00

__init__.py

Move TorchDynamo into PyTorch core (#86461)

2022-10-13 23:18:06 +00:00

aoti_hipify_utils.py

[BE][Easy][16/19] enforce style for empty lines in import segments in torch/_i*/ (#129768)

2024-07-20 16:20:58 +00:00

common.py

[AOTI XPU] Enable Cpp wraper for Intel GPU. (#135318)

2024-11-26 11:51:32 +00:00

cpp_gemm_template.py

[Inductor][CPP] Modularize the CPP GEMM Template (#141006)

2024-11-26 14:32:40 +00:00

cpp_micro_gemm.py

Simplify & rectify dequantized B buffer loading for AMX GEMM micro-kernel for WoQ int8 case (#140258)

2024-11-22 01:34:06 +00:00

cpp_prefix.h

std::value/std::type -> std::_v/std::_t (#138746)

2024-10-26 20:59:24 +00:00

cpp_template_kernel.py

[inductor] Add typing to ir.py 2 (#140915)

2024-11-22 04:56:54 +00:00

cpp_template.py

[AOTI] Remove the non-ABI-compatible mode (part 1) (#138009)

2024-10-17 02:48:26 +00:00

cpp_utils.py

[AOTI XPU] Enable Cpp wraper for Intel GPU. (#135318)

2024-11-26 11:51:32 +00:00

cpp_wrapper_cpu_array_ref.py

[inductor] Refactor ir.Layout into ir.OutputSpec (#140910)

2024-11-21 20:01:57 +00:00

cpp_wrapper_cpu.py

[AOTI XPU] Enable Cpp wraper for Intel GPU. (#135318)

2024-11-26 11:51:32 +00:00

cpp_wrapper_gpu.py

[inductor] Refactor MutableBox to make IRNode typing easier (#140895)

2024-11-20 19:50:46 +00:00

cpp.py

[inductor] modify the heuristic for loop split optimization (#137550)

2024-11-25 09:16:30 +00:00

cpu_device_op_overrides.py

Add Triton CPU as an Inductor backend (#133408)

2024-09-30 20:24:52 +00:00

cuda_combined_scheduling.py

[BE]: Update mypy to 1.11.2 (#133816)

2024-09-16 19:44:11 +00:00

debug_utils.py

[BE]: Apply PERF401 autofixes from ruff (#140980)

2024-11-20 17:52:07 +00:00

halide.py

Revert "Move Sympy printers to torch/utils/_sympy/printers.py (#140597)"

2024-11-22 15:43:39 +00:00

memory_planning.py

[inductor] Generalize WorkspaceArg for graph-level semaphores (#138170)

2024-10-18 23:05:54 +00:00

multi_kernel.py

[BE]: Apply PERF401 autofixes from ruff (#140980)

2024-11-20 17:52:07 +00:00

simd_kernel_features.py

[inductor] Refactor reduction type choices into V.choices (#139585)

2024-11-17 16:10:37 +00:00

simd.py

[inductor] Refactor ir.Layout into ir.OutputSpec (#140910)

2024-11-21 20:01:57 +00:00

triton_combo_kernel.py

[BE]: Apply PERF401 autofixes from ruff (#140980)

2024-11-20 17:52:07 +00:00

triton_split_scan.py

[inductor] Support fixed triton configs defined at compile time (#140217)

2024-11-17 16:10:37 +00:00

triton_utils.py

[inductor] Move V.graph.scheduler.current_device to V.graph.current_device (#138252)

2024-10-18 23:05:54 +00:00

triton.py

[inductor] Fix for "Failed to find static RBLOCK" (#141434)

2024-11-23 22:08:56 +00:00

wrapper.py

[Inductor] Inplacing with Donated Buffer (#140113)

2024-11-26 17:19:50 +00:00