mirror of
https://github.com/zebrajr/pytorch.git
synced 2025-12-06 12:20:52 +01:00
deb776319b
This PR refactors the bfloat16_support_literal constant in the PyTorch build logic to eliminate duplicated ROCm-specific code. Previously, there were two nearly identical branches for ROCM_VERSION < 70000 and ROCM_VERSION >= 70000, differing only by a single typedef. These have been unified into one conditional block with a minimal version guard inside. (https://github.com/ROCm/pytorch/pull/2502) Pull Request resolved: https://github.com/pytorch/pytorch/pull/166147 Approved by: https://github.com/jerrymannil, https://github.com/jeffdaily |
||
|---|---|---|
| .. | ||
| cpu | ||
| cuda | ||
| arg_spec.h | ||
| codegen.cpp | ||
| codegen.h | ||
| compiler.cpp | ||
| compiler.h | ||
| executor.cpp | ||
| executor.h | ||
| fallback.cpp | ||
| fallback.h | ||
| fused_kernel.h | ||
| interface.cpp | ||
| interface.h | ||
| kernel_cache.cpp | ||
| kernel_cache.h | ||
| kernel_spec.h | ||
| partition_desc.h | ||
| README.md | ||
| tensor_desc.h | ||
| tensor_info.h | ||
PyTorch Fuser
The fuser accepts subgraphs wrapped in "fusion nodes" and tries to execute them by just-in-time (JIT) compiling kernels that run all the graph operations.
Code Organization
The fuser is designed hierarchically with device-independent logic eventually deferring to device-specific logic and implementation. The device-specific code is (mostly) found in each devices' subdirectory. The device-independent logic has six components:
- The Interface (interface.h/cpp) has functions to register and run fusions, interrogate fusion functionality, and perform debugging.
- The Compiler (compiler.h/cpp) performs "upfront" and "runtime" compilation. When fusions are registered, upfront compilation produces fallback code and performs some shape inference. When a fusion is run, runtime compilation invokes code generation and the device-specific compilation logic.
- The Code Generator (codegen.h/cpp) produces the string to be compiled on the device.
- The Executor (executor.h/cpp) runs requested fusions. It performs shape inference, expands tensors as necessary, determines the device to run on, acquires a cached compiled kernel or requests the Compiler produce a new one, invokes device-specific code to launch the kernel and updates the stack.
- The Fallback (fallback.h/cpp) runs subgraphs that can't be fused because shape inference didn't determine a common tensor size or the device the tensors are on doesn't support fusion.
- The Kernel Specification Cache (kernel_cache.h/cpp) is a thread-safe cache holding the device-independent specifications produced during upfront compilation. These specifications each have their own thread-safe stores of compiled kernels that the Executor checks before requesting runtime compilation.
The device-specific components have logic for compiling and running code in FusedKernelCPU (cpu/fused_kernel.h/cpp) and FusedKernelCUDA (cuda/fused_kernel.h/cpp).