For the complete documentation index, see llms.txt. Markdown versions of all pages are available by appending .md to any URL (e.g. /max/get-started.md).
Mojo module
structuring
Provides structured sparsity utilities for scatter-gather operations on GPU.
comptime valuesβ
NVIDIASharedMemoryManagerβ
comptime NVIDIASharedMemoryManager = SharedMemoryManager[NVIDIASharedMemoryBasePtr]
Structsβ
- β
IteratorScatterGatherAmd: Iterator-based AMD scatter-gather for DRAM-register data movement. - β
NVIDIASharedMemoryBasePtr: NVIDIA implementation ofSharedMemoryBasePtrusing external memory. - β
ScatterGatherAmd: AMD tile-based scatter-gather for DRAM-register data movement. - β
SharedMemoryManager: Manages bump allocation of tiles and arrays from a shared memory base pointer.
Traitsβ
- β
SharedMemoryBasePtr: Defines a base pointer into GPU shared memory with a fixed alignment.
Was this page helpful?
Thank you! We'll create more content like this.
Thank you for helping us improve!