Skip to content

riscv64/mc: Add w_mask_420_8bpc functions

K230:

checkasm: VLEN=128 bits, using random seed 42
RVV:
 - mc_8bpc.w_mask               [OK]
checkasm: all 6 tests passed
w_mask_420_w4_8bpc_c:        844.6 ( 1.00x)
w_mask_420_w4_8bpc_rvv:      314.1 ( 2.69x)
w_mask_420_w8_8bpc_c:       2592.1 ( 1.00x)
w_mask_420_w8_8bpc_rvv:      550.6 ( 4.71x)
w_mask_420_w16_8bpc_c:      8428.0 ( 1.00x)
w_mask_420_w16_8bpc_rvv:    1253.7 ( 6.72x)
w_mask_420_w32_8bpc_c:     33510.0 ( 1.00x)
w_mask_420_w32_8bpc_rvv:    4282.9 ( 7.82x)
w_mask_420_w64_8bpc_c:     81881.3 ( 1.00x)
w_mask_420_w64_8bpc_rvv:   11112.1 ( 7.37x)
w_mask_420_w128_8bpc_c:   206257.0 ( 1.00x)
w_mask_420_w128_8bpc_rvv:  28101.4 ( 7.34x)

Spacemit K1:

checkasm: VLEN=256 bits, using random seed 42
RVV:
 - mc_8bpc.w_mask               [OK]
checkasm: all 6 tests passed
w_mask_420_w4_8bpc_c:        754.7 ( 1.00x)
w_mask_420_w4_8bpc_rvv:      291.3 ( 2.59x)
w_mask_420_w8_8bpc_c:       2336.8 ( 1.00x)
w_mask_420_w8_8bpc_rvv:      479.6 ( 4.87x)
w_mask_420_w16_8bpc_c:      7582.8 ( 1.00x)
w_mask_420_w16_8bpc_rvv:     885.8 ( 8.56x)
w_mask_420_w32_8bpc_c:     30409.7 ( 1.00x)
w_mask_420_w32_8bpc_rvv:    2594.5 (11.72x)
w_mask_420_w64_8bpc_c:     73000.9 ( 1.00x)
w_mask_420_w64_8bpc_rvv:    6018.6 (12.13x)
w_mask_420_w128_8bpc_c:   180952.4 ( 1.00x)
w_mask_420_w128_8bpc_rvv:  15731.5 (11.50x)

Merge request reports

Loading