For the complete documentation index, see llms.txt. Markdown versions of all pages are available by appending .md to any URL (e.g. /get-started.md).
Mojo module
mma_nvidia_sm100
This module includes utilities for working with the SM100 MMA instructions.
Structsβ
- β
MMASmemDescriptor: Descriptor for shared memory operands tcgen05 mma instructions. - β
MMASmemDescriptorPair: Descriptor for shared memory operands tcgen05 mma instructions. - β
UMMAInsDescriptor: Descriptor for UMMA instructions. - β
UMMAKind: Struct for UMMA instruction types.
Functionsβ
- β
mma: Perform a matrix multiply-accumulate operation using the tcgen05.mma instruction. - β
mma_arrive: Arrive at the mbar pointer for the MMA instruction. - β
mma_arrive_multicast: Arrive at the mbar pointer for the MMA instruction for multiple ctas.
Was this page helpful?
Thank you! We'll create more content like this.
Thank you for helping us improve!