Compile Options

bishengir-compile Options

BiShengIR Feature Control Options

Option	Description	Type	Default	Status
–enable-triton-kernel-compile	Enable Triton kernel compilation.	bool	false	In use
–enable-torch-compile	Enable Torch-MLIR compilation.	bool	false (when BISHENGIR_ENABLE_TORCH_CONVERSIONS)	In use
–enable-hivm-compile	Enable BiShengHIR HIVM compilation.	bool	true	In use
–enable-hfusion-compile	Enable BiShengHIR HFusion compilation.	bool	false	In use
–enable-symbol-analysis	Enable symbol analysis.	bool	false	In use
–enable-multi-kernel	When disabled, graph must fuse as single kernel; when enabled, outline multiple kernels.	bool	false	In use
–enable-manage-host-resources	Enable managing resource for Host functions.	bool	false	In use
–ensure-no-implicit-broadcast	Whether to ensure that there is no implicit broadcast semantics. If there is a dynamic to dynamic dim broadcast, raise a runtime error.	bool	false (when BISHENGIR_ENABLE_TORCH_CONVERSIONS)	In use
–disable-auto-inject-block-sync	Disable generating blocksync wait/set by injectBlockSync pass.	bool	false	In use
–enable-hivm-graph-sync-solver	Use hivm graph-sync-solver instead of inject-sync.	bool	false	In use
–disable-auto-cv-work-space-manage	In combination with the disableAutoInjectBlockSync option.	bool	false	In use
–disable-hivm-auto-inject-sync	Disable auto inject sync intra core.	bool	false	In use
–disable-hivm-tensor-compile	Disable BiShengHIR HIVM Tensor compilation.	bool	false	In use

BiShengIR General Optimization Options

Option	Description	Type	Default	Status
–enable-auto-multi-buffer	Enable auto multi buffer.	bool	false	In use
–limit-auto-multi-buffer-only-for-local-buffer	When enable-auto-multi-buffer = true, limit it only to work for local buffer	bool	true	In use
–enable-tuning-mode	Enable tuning mode and will not try compile multi times in case of plan memory failure	bool	false	In use
–block-dim=<uint>	Number of blocks to use	unsigned	1	In use

BiShengIR HFusion Optimization Options

Option	Description	Type	Default	Status
–enable-deterministic-computing	If enabled, the computation result is deterministic. If disabled, we will enable extra optimizations that might boost performance, e.g. bind reduce to multiple cores. However, the result will be non-deterministic.	bool	true	In use
–enable-ops-reorder	Enable ops reorder to opt pipeline.	bool	true	In use
–hfusion-max-horizontal-fusion-size=<int>	Number of horizontal fusion attempt (Default: unlimited).	int32_t	-1	In use
–hfusion-max-buffer-count-tuning=<long>	Max buffer count tuning in HFusion auto schedule.	int64_t	0	In use
–cube-tiling-tuning=<long>	Cube block size tuning in HFusion auto schedule	list int64_t	“”	In use
–enable-hfusion-count-buffer-dma-opt	If enabled, the buffer used by DMA operations will not be reused by Vector operations.	bool	false	In use

BiShengIR Target Options

The compilation option --target=Ascend<Name> is used to specify the target platform for MLIR compilation. <Name> is a placeholder, and its specific content varies by AI processor model, which can be acquired through dedicated query commands.

The AI processor models and corresponding query methods are described below:

Method 1: Query with the command npu-smi info

Applicable Products

• Atlas A2 training products/Atlas A2 inference products

• Atlas 200I/500 A2 inference products

• Atlas inference products

• Atlas training products

Run this command on the server equipped with the AI processor to get the value of <Name>. The full configuration value is Ascend<Name>. Example: If <Name> is xxx, the configuration value is Ascendxxx.

Method 2: Query with the command npu-smi info -t board -i id -c chip_id

Applicable Products

• Atlas 350 accelerator card

• Atlas A3 training products and Atlas A3 inference products

Execute this command on the server with the AI processor to obtain Chip Name and NPU Name. The actual configuration value is formatted as <Chip Name>_<NPU Name>. Example: If Chip Name is Ascendxxx and the NPU Name is yyy, the configuration value is Ascendxxx_yyy.

Command parameter description:

• id: Device ID, namely the NPU ID queried by the npu-smi info -l command.

• chip_id: Chip ID, namely the Chip ID queried by the npu-smi info -m command.