Exported AVX512 DGEMV kernels (#131)

- Exported DGEMV AVX512 kernels so that they can be directly called by libflame to avoid blis and omp overhead.
This commit is contained in:
Sharma, Shubham
2025-08-12 17:17:41 +05:30
committed by GitHub
parent 021f6bc960
commit b7b9d3ec53

View File

@@ -146,7 +146,25 @@ GEMV_KER_PROT( double, d, gemv_n_zen_int_8x1n_avx512 )
GEMV_KER_PROT( double, d, gemv_n_zen_int_m_leftx1n_avx512 )
// dgemv_t kernels for handling op(A) = 't', i.e., transa = 't' cases.
GEMV_KER_PROT( double, d, gemv_t_zen4_int )
// export gemv kernel so that it can be directly called avoiding blis overhead.
BLIS_EXPORT void bli_dgemv_t_zen4_int
(
conj_t conja,
conj_t conjx,
dim_t m,
dim_t n,
double* restrict alpha,
double* restrict a,
inc_t rs,
inc_t cs,
double* restrict x,
inc_t incx,
double* restrict beta,
double* restrict y,
inc_t incy,
cntx_t* restrict cntx
);
GEMV_KER_PROT( double, d, gemv_t_zen4_int_32x7m )
GEMV_KER_PROT( double, d, gemv_t_zen4_int_32x6m )
GEMV_KER_PROT( double, d, gemv_t_zen4_int_32x5m )
@@ -499,7 +517,7 @@ void bli_dgemv_n_zen4_int
cntx_t* cntx
);
void bli_dgemv_n_zen4_40x2_int_st
BLIS_EXPORT void bli_dgemv_n_zen4_40x2_int_st
(
trans_t transa,
conj_t conjx,