New Feature: Support User-provided Mix Spatial Mapping, And spatial_mapping_hint Definition Can Be Provided in the Mapping File for Mix Spatial Mapping Auto-generation.

[JiacongSun]

New feature: Supporting Mixed Spatial Mapping

We have introduced a new feature that enables the inclusion of mixed spatial mapping in the mapping file, alongside the original single-layer-dimension spatial mapping. For examples, please refer to the files located under the /tests/main/test_with_mix_spatial_mapping folder.

New parameter in the input mapping file:

spatial_mapping_hint: Dict: {hardware_dimension: [supported_layer_dimensions]}. This parameter only works when spatial_mappingis not provided or partially provided in the mapping file. Now the generated spatial mappings will only be on the layer dimensions defined in spatial_mapping_hint. It will be assumed to include all layer dimensions if this parameter is not provided.

New parameters can be provided from the top:

enable_mix_spatial_mapping_generation:

• True --> SpatialMappingGeneratorStagewill auto-generate mix spatial mapping.
• False --> SpatialMappingGeneratorStagewill stick to spatial mapping of single layer dimension.

maximize_hardware_utilization:

• True --> SpatialMappingGeneratorStagewill only keep two spatial mapping candidates for each hardware dimension, which has the highest hardware utilization, for a faster simulation time.
• False --> SpatialMappingGeneratorStagewill keep all spatial mapping candidates it found.

Adds/Changes:

• SpatialMappingGeneratorStagenow can also generate mix spatial mappings.
• SpatialMappingConversionStagenow can also interpret mix spatial mappings.
• Pytest cases are added to the /tests/main/test_with_mix_spatial_mapping folder for testing scenarios involving mixed spatial mapping. Additionally, a dedicated test case with a spatial_mapping_hintdefinition can be found in test_tesla_npu_like.py.

Bug Fixes:

• The abortion in SpatialMappingGeneratorStagehas been fixed for cases where only parts of the spatial mapping are user-provided.
• An error in LomaStage, where the same layer dimension was mapped to different hardware dimensions, has been resolved. Now, these spatial mappings on different hardware dimensions are merged before being sent to LomaStage.

[JiacongSun]

New fixes:

• Corrected an erroneous API call in the original pytest cases.

• Updated pytest cases to include five tests, each addressing different scenarios related to mixed spatial mapping features:

1. When only spatial_mapping is provided while spatial_mapping_hint is not provided (Test: test_with_mix_spatial_mapping/test_ascend_like.py).
2. When both spatial_mapping and spatial_mapping_hint are provided (Test: test_with_mix_spatial_mapping/test_edge_tpu_like.py).
3. When an incomplete spatial_mapping is provided, and spatial_mapping_hint is provided (Test: test_with_mix_spatial_mapping/test_meta_prototype_like.py).
4. When only spatial_mapping_hint is provided (Test: test_with_mix_spatial_mapping/test_tesla_npu_like.py).
5. When the number of dimensions provided in the spatial_mapping_hint exceeds the dimensions available in the hardware (Test: test_with_mix_spatial_mapping/test_tpu_like.py).