ikawrakow/ik_llama.cpp
1
0
Fork 0
mirror of https://github.com/ikawrakow/ik_llama.cpp.git synced 2026-04-29 02:41:47 +00:00
Code Issues Packages Projects Releases Wiki Activity

iq6_k: NEON

Browse Source 
Respectable performance, only slightly slower than Q6_K.
This commit is contained in:
Iwan Kawrakow
2024-08-08 15:51:10 +02:00
parent 0d55d5f078
commit b8abe10b6d
1 changed files with 38 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

38
ggml/src/iqk/iqk_mul_mat.cpp
View File

@@ -4132,6 +4132,41 @@ struct DequantizerIQ5K final : public BaseDequantizer<block_iq5_k> {
float d; float d;
}; };
struct DequantizerIQ6K final : public BaseDequantizer<block_iq6_k> {
DequantizerIQ6K(const void * vx, size_t bx, int nrc) : BaseDequantizer(vx, bx, nrc), values(vld1q_s8_x4(iq6nl_values)) {}
constexpr static int num_blocks() { return 16; }
constexpr static bool should_scale_quants() { return false; }
template <typename Q8>
inline int32x4x4_t new_block(int i, const Q8& q8, float32x4_t * acc) {
d = GGML_FP16_TO_FP32(x[i].d);
return Scale16Extra::new_block(i, d, x[i].extra, 1, vld1q_s8(x[i].scales), q8, acc);
}
inline void prepare(int i, int j) {
bits.prepare(x[i].qs+64*j);
auto hbits = vld1q_u8_x2(x[i].qh + 32*j);
bits.b1.val[0] = vorrq_u8(bits.b1.val[0], vandq_u8(vshlq_n_u8(hbits.val[0], 4), hm));
bits.b1.val[1] = vorrq_u8(bits.b1.val[1], vandq_u8(vshlq_n_u8(hbits.val[1], 4), hm));
bits.b1.val[2] = vorrq_u8(bits.b1.val[2], vandq_u8(vshlq_n_u8(hbits.val[0], 2), hm));
bits.b1.val[3] = vorrq_u8(bits.b1.val[3], vandq_u8(vshlq_n_u8(hbits.val[1], 2), hm));
bits.b2.val[0] = vorrq_u8(bits.b2.val[0], vandq_u8(hbits.val[0], hm));
bits.b2.val[1] = vorrq_u8(bits.b2.val[1], vandq_u8(hbits.val[1], hm));
bits.b2.val[2] = vorrq_u8(bits.b2.val[2], vandq_u8(vshrq_n_u8(hbits.val[0], 2), hm));
bits.b2.val[3] = vorrq_u8(bits.b2.val[3], vandq_u8(vshrq_n_u8(hbits.val[1], 2), hm));
for (int k = 0; k < 4; ++k) {
bits.b1.val[k] = vqtbl4q_s8(values, bits.b1.val[k]);
bits.b2.val[k] = vqtbl4q_s8(values, bits.b2.val[k]);
}
}
Q4bits bits;
const int8x16x4_t values;
const uint8x16_t hm = vdupq_n_u8(0x30);
float d;
};
struct DequantizerIQ2K final : public BaseDequantizer<block_iq2_k> { struct DequantizerIQ2K final : public BaseDequantizer<block_iq2_k> {
DequantizerIQ2K(const void * vx, size_t bx, int nrc) : BaseDequantizer(vx, bx, nrc) {} DequantizerIQ2K(const void * vx, size_t bx, int nrc) : BaseDequantizer(vx, bx, nrc) {}
@@ -5720,6 +5755,9 @@ bool MulMat::prepare(int typeA, int typeB, int ne00, MulMat& m, int /*Ny*/) {
case GGML_TYPE_IQ5_K: case GGML_TYPE_IQ5_K:
MulMat::set_functions<DequantizerIQ5K>(m); MulMat::set_functions<DequantizerIQ5K>(m);
break; break;
case GGML_TYPE_IQ6_K:
MulMat::set_functions<DequantizerIQ6K>(m);
break;
case GGML_TYPE_IQ2_K: case GGML_TYPE_IQ2_K:
MulMat::set_functions<DequantizerIQ2K>(m); MulMat::set_functions<DequantizerIQ2K>(m);
break; break;