Expand description
TMA-related runtime types
Structs§
- Im2col
Args - Args for im2col tensor maps
- Im2col
Wide Args - Args for im2col wide tensor maps
- Tiled
Args - Args for tiled tensor maps
Enums§
- OobFill
- What value to use when filling out of bounds values
- Tensor
MapFormat - Format of [
TensorMap] - Tensor
MapInterleave - Interleave setting for [
TensorMap] - Tensor
MapPrefetch - Additional prefetching to perform during load Specifies L2 fetch size which indicates the byte granularity at which L2 requests are filled from DRAM
- Tensor
MapSwizzle - Data are organized in a specific order in global memory; however, this may not match the order
in which the application accesses data in shared memory. This difference in data organization
may cause bank conflicts when shared memory is accessed. In order to avoid this problem, data
can be loaded to shared memory with shuffling across shared memory banks. When interleave is
TensorMapInterleave::B32, swizzle must beTensorMapSwizzle::B32. Other interleave modes can have any swizzling pattern.