feat: Support for BLAS methods on AMD GPUs, CPU libraries, and improvements on CUDA APIs - #9

Open

sanjibansg wants to merge 28 commits into

mainfrom

Commits on Mar 2, 2026

Commits on Mar 3, 2026

fix: layout order configuration
sanjibansg
committed
8a27ffe
fix: remove look for requested workspace
sanjibansg
committed
54dce63

Commits on Apr 17, 2026

feat: matmul method for without bias matrix multiplication condition
sanjibansg
committed
c5128e6
feat: option to pass gpu pointers directly
sanjibansg
committed
9282ba9
fix: function signature for matmul method with void pointers
sanjibansg
committed
fc90737
fix: template types for arguments to function with pointer signatures
sanjibansg
committed
b591c4a
fix: use float signatures for pointer arguments for blascuda
sanjibansg
committed
19b5b3f
fix: use explicit data type for signatures with pointer arguments
sanjibansg
committed
c995464
fix: non transpose axis for matmul method
sanjibansg
committed
75f75a4
fix (experimental): layout shape order
sanjibansg
committed
ed303fb

Commits on May 4, 2026

feat: cuda cleanup and cpu blas api
sanjibansg
committed
54a9401
feat: matmul method on raw cuda pointers
sanjibansg
committed
066fdae
fix: remove extra template parameter in cuda matmul method on raw pointers
sanjibansg
committed
1f1c4c7
fix: method signature for matmul on cuda raw pointers
sanjibansg
committed
a867939
fix: dimension order for checkAddLayout method in cuda
sanjibansg
committed
ad41e99
fix: correct transpose values while adding layouts for cuda gemm methods
sanjibansg
committed
f86d299

Commits on May 11, 2026

feat: gemm apis for alpaka views
sanjibansg
committed
7b2133c
feat: gemm functions for cublas using raw pointers
sanjibansg
committed
6f2a9e1

Commits on May 25, 2026

feat: strided batch computation for gemm in cublaslt
sanjibansg
committed
fa108fb

Commits on Aug 11, 2026

feat: support for gemm functions on AMD hardware using hipBLASLt and cpu blas methods from different libraries
sanjibansg
committed
97c12bc

Commits on Aug 17, 2026