Vijay Thakkar

NVIDIA Remote Graduate student and mechanic @hpcgarage. Compute Architect @NVIDIA

Results 6 issues of


                                            Vijay Thakkar

cuASR 2.0

[WiP]: Cuasr 1.2

- Bump cutlass upstream dep to 2.7 - Include Ampere multi-stage SRGEMM using LDGSTS

Fix license headers in every file

1

comment

cuASR operators: combine Add and Multiply operator structs into one fused AddMultiply struct

ADD: Ampere multi-stage SRGEMM example

ADD: Ampere multi-stage SRGEMM (LDGSTS support)