Removed SVE Implementation of Dequantized row

This commit is contained in:
vithulep 2025-06-17 14:50:06 +05:30
parent eca058e225
commit 2add787715
2 changed files with 7 additions and 23 deletions

View File

@ -43,7 +43,8 @@ void quantize_row_q8_0(const float * GGML_RESTRICT x, void * GGML_RESTRICT vy, i
const int nb = k / QK8_0;
block_q8_0 * GGML_RESTRICT y = vy;
// printf("Here");
#if defined(__ARM_FEATURE_SVE)
const int sve_register_length = ggml_cpu_get_sve_cnt() * 8;
const int ggml_f32_epr = sve_register_length / 32;

View File

@ -340,7 +340,6 @@ void dequantize_row_q5_1(const block_q5_1 * GGML_RESTRICT x, float * GGML_RESTRI
}
}
// SVE Support added for Scaler Implementation
void dequantize_row_q8_0(const block_q8_0 * GGML_RESTRICT x, float * GGML_RESTRICT y, int64_t k) {
static const int qk = QK8_0;
@ -348,29 +347,13 @@ void dequantize_row_q8_0(const block_q8_0 * GGML_RESTRICT x, float * GGML_RESTRI
const int nb = k / qk;
#if defined(__ARM_FEATURE_SVE)
svbool_t pg = svptrue_b32();
const svfloat32_t inactive1 = svdup_n_f32(0.0f);
const int ggml_f32_epr = svcntw();
for (int i = 0; i < nb; i+=1) {
const float d1 = GGML_FP16_TO_FP32(x[i].d); // d:0
const int8_t *x_data1 = x[i].qs;
float *y_base = y + i * qk;
for (int j = 0; j < qk; j+=ggml_f32_epr) {
svint32_t vec0 = svld1sb_s32(pg, x_data1 + j);
svfloat32_t fvec0 = svmul_n_f32_m(pg, svcvt_f32_s32_m(inactive1, pg, vec0), d1); // Convert to float and scale
svst1_f32(pg, y_base + j, fvec0);
}
}
#else
for (int i = 0; i < nb; i++) {
const float d = GGML_FP16_TO_FP32(x[i].d);
for (int i = 0; i < nb; i++) {
const float d = GGML_FP16_TO_FP32(x[i].d);
for (int j = 0; j < qk; ++j) {
y[i*qk + j] = x[i].qs[j]*d;
}
for (int j = 0; j < qk; ++j) {
y[i*qk + j] = x[i].qs[j]*d;
}
#endif
}
}
//