// Auto-generated file. Do not edit! // Template: src/qs8-gemm/rvv.c.in // Generator: tools/xngen // // Copyright 2024 SiFive, Inc. // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #include #include #include "xnnpack/gemm.h" #include "xnnpack/math.h" void xnn_qd8_f32_qc4w_gemm_minmax_ukernel_1x4v__rvv( size_t mr, size_t nc, size_t kc, const int8_t* restrict a, size_t a_stride, const void* restrict w, float* restrict c, size_t cm_stride, size_t cn_stride, const struct xnn_f32_qc4w_minmax_params params[restrict XNN_MIN_ELEMENTS(1)], const struct xnn_qd8_quantization_params quantization_params[restrict XNN_MIN_ELEMENTS(1)]) { assert(mr != 0); assert(mr <= 1); assert(nc != 0); assert(kc != 0); const int8_t* a0 = a; float* c0 = c; const size_t nr = __riscv_vsetvlmax_e32m4(); size_t vl = nr; kc = round_up_po2(kc, 2); do { if XNN_UNLIKELY(nc < nr) { vl = __riscv_vsetvl_e32m4(nc); } nc = nc - vl; vint32m4_t vksum = __riscv_vle32_v_i32m4((const int32_t*)w, vl); w = (const int32_t*) w + nr; const int32_t vinput_zero_point0 = quantization_params[0].zero_point; vint32m4_t vacc0 = __riscv_vmul_vx_i32m4(vksum, vinput_zero_point0, vl); size_t k = kc; for (; k >= 2 * sizeof(uint8_t); k -= 2 * sizeof(uint8_t)) { const int8_t va0c0 = a0[0]; const int8_t va0c1 = a0[1]; a0 += 2; const vint8m1_t vbi = __riscv_vle8_v_i8m1((const int8_t*) w, vl); w = (const uint8_t*) w + nr; const vint8m1_t vbc0 = __riscv_vsll_vx_i8m1(vbi, 4, vl); const vint8m1_t vbc1 = __riscv_vand_vx_i8m1(vbi, 0xF0, vl); vint16m2_t va0bc0 = __riscv_vwmul_vx_i16m2(vbc0, va0c0, vl); vacc0 = __riscv_vwadd_wv_i32m4(vacc0, va0bc0, vl); vint16m2_t va0bc1 = __riscv_vwmul_vx_i16m2(vbc1, va0c1, vl); vacc0 = __riscv_vwadd_wv_i32m4(vacc0, va0bc1, vl); } vacc0 = __riscv_vsra_vx_i32m4(vacc0, 4, vl); // i32 -> f32 vfloat32m4_t vout0 = __riscv_vfcvt_f_x_v_f32m4(vacc0, vl); // vout * input_scale const float vinput_scale0 = quantization_params[0].inv_scale; vout0 = __riscv_vfmul_vf_f32m4(vout0, vinput_scale0, vl); const vfloat32m4_t vfilter_output_scale = __riscv_vle32_v_f32m4((const float*) w, vl); w = (const float*) w + nr; vout0 = __riscv_vfmul_vv_f32m4(vout0, vfilter_output_scale, vl); const vfloat32m4_t vbias = __riscv_vle32_v_f32m4((const float*) w, vl); w = (const float*) w + nr; vout0 = __riscv_vfadd_vv_f32m4(vout0, vbias, vl); const float vmin = params->scalar.min; vout0 = __riscv_vfmax_vf_f32m4(vout0, vmin, vl); const float vmax = params->scalar.max; vout0 = __riscv_vfmin_vf_f32m4(vout0, vmax, vl); // store 1 x vl results to c __riscv_vse32_v_f32m4(c0, vout0, vl); c0 = (float*) ((uintptr_t) c0 + cn_stride); a0 = (const int8_t*) ((uintptr_t) a0 - kc); } while (nc != 0); }