ggml : add Q8_0 quantization for intermediate results
This commit is contained in:
Georgi Gerganov
parent
0e07e6a839
commit
a520b33b3a
3 changed files with 273 additions and 18 deletions
4
Makefile
4
Makefile
|
|
@ -142,10 +142,10 @@ default: main quantize perplexity embedding
|
|||
ggml.o: ggml.c ggml.h
|
||||
$(CC) $(CFLAGS) -c ggml.c -o ggml.o
|
||||
|
||||
llama.o: llama.cpp llama.h llama_util.h
|
||||
llama.o: llama.cpp llama.h llama_util.h ggml.h
|
||||
$(CXX) $(CXXFLAGS) -c llama.cpp -o llama.o
|
||||
|
||||
common.o: examples/common.cpp examples/common.h
|
||||
common.o: examples/common.cpp examples/common.h llama.h
|
||||
$(CXX) $(CXXFLAGS) -c examples/common.cpp -o common.o
|
||||
|
||||
clean:
|
||||
|
|
|
|||
286
ggml.c
286
ggml.c
|
|
@ -566,7 +566,7 @@ inline uint8x8_t vzip2_u8(uint8x8_t a, uint8x8_t b) {
|
|||
// blocks of QK elements
|
||||
// represented with a single float (delta) and QK/2 8-bit ints (i.e QK 4-bit signed integer factors)
|
||||
typedef struct {
|
||||
float d; // delta
|
||||
float d; // delta
|
||||
uint8_t qs[QK / 2]; // nibbles / quants
|
||||
} block_q4_0;
|
||||
static_assert(sizeof(block_q4_0) == sizeof(float) + QK / 2, "wrong q4_0 block size/padding");
|
||||
|
|
@ -575,12 +575,19 @@ static_assert(sizeof(block_q4_0) == sizeof(float) + QK / 2, "wrong q4_0 block si
|
|||
// blocks of QK elements
|
||||
// represented with 2 floats (delta + min) and QK/2 8-bit ints (i.e QK 4-bit unsigned integer factors)
|
||||
typedef struct {
|
||||
float d;
|
||||
float m;
|
||||
float d; // delta
|
||||
float m; // min
|
||||
uint8_t qs[QK / 2]; // nibbles / quants
|
||||
} block_q4_1;
|
||||
static_assert(sizeof(block_q4_1) == sizeof(float) * 2 + QK / 2, "wrong q4_1 block size/padding");
|
||||
|
||||
typedef struct {
|
||||
float d; // delta
|
||||
uint8_t qs[QK]; // nibbles / quants
|
||||
} block_q8_0;
|
||||
static_assert(sizeof(block_q8_0) == sizeof(float) + QK, "wrong q8_0 block size/padding");
|
||||
|
||||
|
||||
// reference implementation for deterministic creation of model files
|
||||
static void quantize_row_q4_0_reference(const float * restrict x, block_q4_0 * restrict y, int k) {
|
||||
assert(k % QK == 0);
|
||||
|
|
@ -1033,6 +1040,76 @@ static void quantize_row_q4_1(const float * restrict x, void * restrict vy, int
|
|||
#endif
|
||||
}
|
||||
|
||||
// reference implementation for deterministic creation of model files
|
||||
static void quantize_row_q8_0_reference(const float * restrict x, block_q8_0 * restrict y, int k) {
|
||||
assert(k % QK == 0);
|
||||
const int nb = k / QK;
|
||||
|
||||
for (int i = 0; i < nb; i++) {
|
||||
float amax = 0.0f; // absolute max
|
||||
|
||||
for (int l = 0; l < QK; l++) {
|
||||
const float v = x[i*QK + l];
|
||||
amax = MAX(amax, fabsf(v));
|
||||
}
|
||||
|
||||
const float d = amax / ((1 << 7) - 1);
|
||||
const float id = d ? 1.0f/d : 0.0f;
|
||||
|
||||
y[i].d = d;
|
||||
|
||||
for (int l = 0; l < QK; ++l) {
|
||||
const float v = x[i*QK + l]*id;
|
||||
const uint8_t vi = (int8_t)roundf(v) + 128;
|
||||
|
||||
y[i].qs[l] = vi;
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
static void quantize_row_q8_0(const float * restrict x, void * restrict vy, int k) {
|
||||
assert(k % QK == 0);
|
||||
const int nb = k / QK;
|
||||
|
||||
block_q8_0 * restrict y = vy;
|
||||
|
||||
#if defined(__ARM_NEON)
|
||||
for (int i = 0; i < nb; i++) {
|
||||
float32x4_t srcv [8];
|
||||
float32x4_t asrcv[8];
|
||||
float32x4_t amaxv[8];
|
||||
|
||||
for (int l = 0; l < 8; l++) srcv[l] = vld1q_f32(x + i*32 + 4*l);
|
||||
for (int l = 0; l < 8; l++) asrcv[l] = vabsq_f32(srcv[l]);
|
||||
|
||||
for (int l = 0; l < 4; l++) amaxv[2*l] = vmaxq_f32(asrcv[2*l], asrcv[2*l+1]);
|
||||
for (int l = 0; l < 2; l++) amaxv[4*l] = vmaxq_f32(amaxv[4*l], amaxv[4*l+2]);
|
||||
for (int l = 0; l < 1; l++) amaxv[8*l] = vmaxq_f32(amaxv[8*l], amaxv[8*l+4]);
|
||||
|
||||
const float amax = vmaxvq_f32(amaxv[0]);
|
||||
|
||||
const float d = amax / ((1 << 7) - 1);
|
||||
const float id = d ? 1.0f/d : 0.0f;
|
||||
|
||||
y[i].d = d;
|
||||
|
||||
for (int l = 0; l < 8; l++) {
|
||||
const float32x4_t v = vmulq_n_f32(srcv[l], id);
|
||||
const float32x4_t vf = vaddq_f32(v, vdupq_n_f32(128.5f));
|
||||
const int32x4_t vi = vcvtq_s32_f32(vf);
|
||||
|
||||
y[i].qs[4*l + 0] = vgetq_lane_s32(vi, 0);
|
||||
y[i].qs[4*l + 1] = vgetq_lane_s32(vi, 1);
|
||||
y[i].qs[4*l + 2] = vgetq_lane_s32(vi, 2);
|
||||
y[i].qs[4*l + 3] = vgetq_lane_s32(vi, 3);
|
||||
}
|
||||
}
|
||||
#else
|
||||
// scalar
|
||||
quantize_row_q8_0_reference(x, y, k);
|
||||
#endif
|
||||
}
|
||||
|
||||
static void dequantize_row_q4_0(const void * restrict vx, float * restrict y, int k) {
|
||||
assert(k % QK == 0);
|
||||
const int nb = k / QK;
|
||||
|
|
@ -2335,12 +2412,12 @@ static void ggml_vec_dot_q4_1(const int n, float * restrict s, const void * rest
|
|||
const uint8x16_t v1_1h = vshrq_n_u8(v1_1, 4);
|
||||
|
||||
sum00 += x0->m*y0->m;
|
||||
sum01 += y0->m*x0->d*(vaddvq_u8(v0_0l) + vaddvq_u8(v0_0h));
|
||||
sum10 += x0->m*y0->d*(vaddvq_u8(v1_0l) + vaddvq_u8(v1_0h));
|
||||
sum01 += y0->m*x0->d*((uint16_t)vaddvq_u8(v0_0l) + (uint16_t)vaddvq_u8(v0_0h));
|
||||
sum10 += x0->m*y0->d*((uint16_t)vaddvq_u8(v1_0l) + (uint16_t)vaddvq_u8(v1_0h));
|
||||
|
||||
sum00 += x1->m*y1->m;
|
||||
sum01 += y1->m*x1->d*(vaddvq_u8(v0_1l) + vaddvq_u8(v0_1h));
|
||||
sum10 += x1->m*y1->d*(vaddvq_u8(v1_1l) + vaddvq_u8(v1_1h));
|
||||
sum01 += y1->m*x1->d*((uint16_t)vaddvq_u8(v0_1l) + (uint16_t)vaddvq_u8(v0_1h));
|
||||
sum10 += x1->m*y1->d*((uint16_t)vaddvq_u8(v1_1l) + (uint16_t)vaddvq_u8(v1_1h));
|
||||
|
||||
#if defined(__ARM_FEATURE_DOTPROD)
|
||||
// dot product into int32x4_t
|
||||
|
|
@ -2408,6 +2485,129 @@ static void ggml_vec_dot_q4_1(const int n, float * restrict s, const void * rest
|
|||
*s = sumf;
|
||||
}
|
||||
|
||||
static void ggml_vec_dot_q4_0_q8_0(const int n, float * restrict s, const void * restrict vx, const void * restrict vy) {
|
||||
const int nb = n / QK;
|
||||
|
||||
assert(n % QK == 0);
|
||||
assert(nb % 2 == 0);
|
||||
|
||||
const block_q4_0 * restrict x = vx;
|
||||
const block_q8_0 * restrict y = vy;
|
||||
|
||||
float sumf = 0.0;
|
||||
|
||||
#if defined(__ARM_NEON)
|
||||
float sum0 = 0.0f;
|
||||
float sum1 = 0.0f;
|
||||
|
||||
for (int i = 0; i < nb; i += 2) {
|
||||
const block_q4_0 * restrict x0 = &x[i + 0];
|
||||
const block_q4_0 * restrict x1 = &x[i + 1];
|
||||
const block_q8_0 * restrict y0 = &y[i + 0];
|
||||
const block_q8_0 * restrict y1 = &y[i + 1];
|
||||
|
||||
const uint8x16_t m4b = vdupq_n_u8(0xf);
|
||||
const int8x16_t s8b = vdupq_n_s8(0x8);
|
||||
const uint8x16_t u128b = vdupq_n_u8(128);
|
||||
|
||||
const uint8x16_t v0_0 = vld1q_u8(x0->qs);
|
||||
const uint8x16_t v0_1 = vld1q_u8(x1->qs);
|
||||
|
||||
// 4-bit -> 8-bit
|
||||
const int8x16_t v0_0l = vreinterpretq_s8_u8(vandq_u8 (v0_0, m4b));
|
||||
const int8x16_t v0_0h = vreinterpretq_s8_u8(vshrq_n_u8(v0_0, 4));
|
||||
const int8x16_t v0_1l = vreinterpretq_s8_u8(vandq_u8 (v0_1, m4b));
|
||||
const int8x16_t v0_1h = vreinterpretq_s8_u8(vshrq_n_u8(v0_1, 4));
|
||||
|
||||
// sub 8
|
||||
const int8x16_t v0_0ls = vsubq_s8(v0_0l, s8b);
|
||||
const int8x16_t v0_0hs = vsubq_s8(v0_0h, s8b);
|
||||
const int8x16_t v0_1ls = vsubq_s8(v0_1l, s8b);
|
||||
const int8x16_t v0_1hs = vsubq_s8(v0_1h, s8b);
|
||||
|
||||
// load y
|
||||
const uint8x16_t v1_0l = vld1q_u8(y0->qs);
|
||||
const uint8x16_t v1_0h = vld1q_u8(y0->qs + 16);
|
||||
const uint8x16_t v1_1l = vld1q_u8(y1->qs);
|
||||
const uint8x16_t v1_1h = vld1q_u8(y1->qs + 16);
|
||||
|
||||
// interleave
|
||||
const uint8x16_t v1_0lz = vuzp1q_u8(v1_0l, v1_0h);
|
||||
const uint8x16_t v1_0hz = vuzp2q_u8(v1_0l, v1_0h);
|
||||
const uint8x16_t v1_1lz = vuzp1q_u8(v1_1l, v1_1h);
|
||||
const uint8x16_t v1_1hz = vuzp2q_u8(v1_1l, v1_1h);
|
||||
|
||||
const int8x16_t v1_0ls = vreinterpretq_s8_u8(vsubq_u8(v1_0lz, u128b));
|
||||
const int8x16_t v1_0hs = vreinterpretq_s8_u8(vsubq_u8(v1_0hz, u128b));
|
||||
const int8x16_t v1_1ls = vreinterpretq_s8_u8(vsubq_u8(v1_1lz, u128b));
|
||||
const int8x16_t v1_1hs = vreinterpretq_s8_u8(vsubq_u8(v1_1hz, u128b));
|
||||
|
||||
#if defined(__ARM_FEATURE_DOTPROD)
|
||||
// dot product into int32x4_t
|
||||
int32x4_t p_0 = vdotq_s32(vdupq_n_s32(0), v0_0ls, v1_0ls);
|
||||
int32x4_t p_1 = vdotq_s32(vdupq_n_s32(0), v0_1ls, v1_1ls);
|
||||
|
||||
p_0 = vdotq_s32(p_0, v0_0hs, v1_0hs);
|
||||
p_1 = vdotq_s32(p_1, v0_1hs, v1_1hs);
|
||||
|
||||
sum0 += x0->d*y0->d*vaddvq_s32(p_0);
|
||||
sum1 += x1->d*y1->d*vaddvq_s32(p_1);
|
||||
#else
|
||||
const int16x8_t pl0l = vmull_s8(vget_low_s8 (v0_0ls), vget_low_s8 (v1_0ls));
|
||||
const int16x8_t pl0h = vmull_s8(vget_high_s8(v0_0ls), vget_high_s8(v1_0ls));
|
||||
const int16x8_t ph0l = vmull_s8(vget_low_s8 (v0_0hs), vget_low_s8 (v1_0hs));
|
||||
const int16x8_t ph0h = vmull_s8(vget_high_s8(v0_0hs), vget_high_s8(v1_0hs));
|
||||
|
||||
const int16x8_t pl1l = vmull_s8(vget_low_s8 (v0_1ls), vget_low_s8 (v1_1ls));
|
||||
const int16x8_t pl1h = vmull_s8(vget_high_s8(v0_1ls), vget_high_s8(v1_1ls));
|
||||
const int16x8_t ph1l = vmull_s8(vget_low_s8 (v0_1hs), vget_low_s8 (v1_1hs));
|
||||
const int16x8_t ph1h = vmull_s8(vget_high_s8(v0_1hs), vget_high_s8(v1_1hs));
|
||||
|
||||
const int16x8_t pl_0 = vaddq_s16(pl0l, pl0h);
|
||||
const int16x8_t ph_0 = vaddq_s16(ph0l, ph0h);
|
||||
|
||||
const int16x8_t pl_1 = vaddq_s16(pl1l, pl1h);
|
||||
const int16x8_t ph_1 = vaddq_s16(ph1l, ph1h);
|
||||
|
||||
const int16x8_t p_0 = vaddq_s16(pl_0, ph_0);
|
||||
const int16x8_t p_1 = vaddq_s16(pl_1, ph_1);
|
||||
|
||||
sum0 += x0->d*y0->d*vaddvq_s16(p_0);
|
||||
sum1 += x1->d*y1->d*vaddvq_s16(p_1);
|
||||
#endif
|
||||
}
|
||||
|
||||
sumf = sum0 + sum1;
|
||||
#else
|
||||
// scalar
|
||||
for (int i = 0; i < nb; i++) {
|
||||
const float d0 = x[i].d;
|
||||
const float d1 = y[i].d;
|
||||
|
||||
const uint8_t * restrict p0 = x[i].qs;
|
||||
const uint8_t * restrict p1 = y[i].qs;
|
||||
|
||||
int sumi = 0;
|
||||
for (int j = 0; j < QK/2; j++) {
|
||||
const uint8_t v0 = p0[j];
|
||||
|
||||
const int i0 = (int8_t) (v0 & 0xf) - 8;
|
||||
const int i1 = (int8_t) (v0 >> 4) - 8;
|
||||
|
||||
const int i2 = (int) p1[2*j + 0] - 128;
|
||||
const int i3 = (int) p1[2*j + 1] - 128;
|
||||
|
||||
/*printf("dot product: i0=%4d i1=%4d i2=%4d i3=%4d\n", i0, i1, i2, i3);*/
|
||||
|
||||
sumi += i0*i2 + i1*i3;
|
||||
}
|
||||
sumf += d0*d1*sumi;
|
||||
}
|
||||
#endif
|
||||
|
||||
*s = sumf;
|
||||
}
|
||||
|
||||
// compute GGML_VEC_DOT_UNROLL dot products at once
|
||||
// xs - x row stride in bytes
|
||||
inline static void ggml_vec_dot_f16_unroll(const int n, const int xs, float * restrict s, void * restrict xv, ggml_fp16_t * restrict y) {
|
||||
|
|
@ -2654,22 +2854,24 @@ static const int GGML_BLCK_SIZE[GGML_TYPE_COUNT] = {
|
|||
[GGML_TYPE_F16] = 1,
|
||||
[GGML_TYPE_Q4_0] = QK,
|
||||
[GGML_TYPE_Q4_1] = QK,
|
||||
[GGML_TYPE_Q8_0] = QK,
|
||||
[GGML_TYPE_I8] = 1,
|
||||
[GGML_TYPE_I16] = 1,
|
||||
[GGML_TYPE_I32] = 1,
|
||||
};
|
||||
static_assert(GGML_TYPE_COUNT == 7, "GGML_BLCK_SIZE is outdated");
|
||||
static_assert(GGML_TYPE_COUNT == 8, "GGML_BLCK_SIZE is outdated");
|
||||
|
||||
static const size_t GGML_TYPE_SIZE[GGML_TYPE_COUNT] = {
|
||||
[GGML_TYPE_F32] = sizeof(float),
|
||||
[GGML_TYPE_F16] = sizeof(ggml_fp16_t),
|
||||
[GGML_TYPE_Q4_0] = sizeof(block_q4_0),
|
||||
[GGML_TYPE_Q4_1] = sizeof(block_q4_1),
|
||||
[GGML_TYPE_Q8_0] = sizeof(block_q8_0),
|
||||
[GGML_TYPE_I8] = sizeof(int8_t),
|
||||
[GGML_TYPE_I16] = sizeof(int16_t),
|
||||
[GGML_TYPE_I32] = sizeof(int32_t),
|
||||
};
|
||||
static_assert(GGML_TYPE_COUNT == 7, "GGML_TYPE_SIZE is outdated");
|
||||
static_assert(GGML_TYPE_COUNT == 8, "GGML_TYPE_SIZE is outdated");
|
||||
|
||||
static const char * GGML_OP_LABEL[GGML_OP_COUNT] = {
|
||||
"NONE",
|
||||
|
|
@ -3337,6 +3539,10 @@ struct ggml_tensor * ggml_set_i32 (struct ggml_tensor * tensor, int32_t value) {
|
|||
{
|
||||
GGML_ASSERT(false);
|
||||
} break;
|
||||
case GGML_TYPE_Q8_0:
|
||||
{
|
||||
GGML_ASSERT(false);
|
||||
} break;
|
||||
case GGML_TYPE_I8:
|
||||
{
|
||||
assert(tensor->nb[0] == sizeof(int8_t));
|
||||
|
|
@ -3397,6 +3603,10 @@ struct ggml_tensor * ggml_set_f32(struct ggml_tensor * tensor, float value) {
|
|||
{
|
||||
GGML_ASSERT(false);
|
||||
} break;
|
||||
case GGML_TYPE_Q8_0:
|
||||
{
|
||||
GGML_ASSERT(false);
|
||||
} break;
|
||||
case GGML_TYPE_I8:
|
||||
{
|
||||
assert(tensor->nb[0] == sizeof(int8_t));
|
||||
|
|
@ -3451,6 +3661,10 @@ int32_t ggml_get_i32_1d(const struct ggml_tensor * tensor, int i) {
|
|||
{
|
||||
GGML_ASSERT(false);
|
||||
} break;
|
||||
case GGML_TYPE_Q8_0:
|
||||
{
|
||||
GGML_ASSERT(false);
|
||||
} break;
|
||||
case GGML_TYPE_I8:
|
||||
{
|
||||
GGML_ASSERT(tensor->nb[0] == sizeof(int8_t));
|
||||
|
|
@ -3495,6 +3709,10 @@ void ggml_set_i32_1d(const struct ggml_tensor * tensor, int i, int32_t value) {
|
|||
{
|
||||
GGML_ASSERT(false);
|
||||
} break;
|
||||
case GGML_TYPE_Q8_0:
|
||||
{
|
||||
GGML_ASSERT(false);
|
||||
} break;
|
||||
case GGML_TYPE_I8:
|
||||
{
|
||||
GGML_ASSERT(tensor->nb[0] == sizeof(int8_t));
|
||||
|
|
@ -3537,6 +3755,10 @@ float ggml_get_f32_1d(const struct ggml_tensor * tensor, int i) {
|
|||
{
|
||||
GGML_ASSERT(false);
|
||||
} break;
|
||||
case GGML_TYPE_Q8_0:
|
||||
{
|
||||
GGML_ASSERT(false);
|
||||
} break;
|
||||
case GGML_TYPE_I8:
|
||||
{
|
||||
GGML_ASSERT(tensor->nb[0] == sizeof(int8_t));
|
||||
|
|
@ -3581,6 +3803,10 @@ void ggml_set_f32_1d(const struct ggml_tensor * tensor, int i, float value) {
|
|||
{
|
||||
GGML_ASSERT(false);
|
||||
} break;
|
||||
case GGML_TYPE_Q8_0:
|
||||
{
|
||||
GGML_ASSERT(false);
|
||||
} break;
|
||||
case GGML_TYPE_I8:
|
||||
{
|
||||
GGML_ASSERT(tensor->nb[0] == sizeof(int8_t));
|
||||
|
|
@ -5319,6 +5545,7 @@ static void ggml_compute_forward_dup(
|
|||
} break;
|
||||
case GGML_TYPE_Q4_0:
|
||||
case GGML_TYPE_Q4_1:
|
||||
case GGML_TYPE_Q8_0:
|
||||
case GGML_TYPE_I8:
|
||||
case GGML_TYPE_I16:
|
||||
case GGML_TYPE_I32:
|
||||
|
|
@ -5400,6 +5627,7 @@ static void ggml_compute_forward_add(
|
|||
} break;
|
||||
case GGML_TYPE_Q4_0:
|
||||
case GGML_TYPE_Q4_1:
|
||||
case GGML_TYPE_Q8_0:
|
||||
case GGML_TYPE_I8:
|
||||
case GGML_TYPE_I16:
|
||||
case GGML_TYPE_I32:
|
||||
|
|
@ -5452,6 +5680,7 @@ static void ggml_compute_forward_sub(
|
|||
} break;
|
||||
case GGML_TYPE_Q4_0:
|
||||
case GGML_TYPE_Q4_1:
|
||||
case GGML_TYPE_Q8_0:
|
||||
case GGML_TYPE_I8:
|
||||
case GGML_TYPE_I16:
|
||||
case GGML_TYPE_I32:
|
||||
|
|
@ -5504,6 +5733,7 @@ static void ggml_compute_forward_mul(
|
|||
} break;
|
||||
case GGML_TYPE_Q4_0:
|
||||
case GGML_TYPE_Q4_1:
|
||||
case GGML_TYPE_Q8_0:
|
||||
case GGML_TYPE_I8:
|
||||
case GGML_TYPE_I16:
|
||||
case GGML_TYPE_I32:
|
||||
|
|
@ -5556,6 +5786,7 @@ static void ggml_compute_forward_div(
|
|||
} break;
|
||||
case GGML_TYPE_Q4_0:
|
||||
case GGML_TYPE_Q4_1:
|
||||
case GGML_TYPE_Q8_0:
|
||||
case GGML_TYPE_I8:
|
||||
case GGML_TYPE_I16:
|
||||
case GGML_TYPE_I32:
|
||||
|
|
@ -5604,6 +5835,7 @@ static void ggml_compute_forward_sqr(
|
|||
} break;
|
||||
case GGML_TYPE_Q4_0:
|
||||
case GGML_TYPE_Q4_1:
|
||||
case GGML_TYPE_Q8_0:
|
||||
case GGML_TYPE_I8:
|
||||
case GGML_TYPE_I16:
|
||||
case GGML_TYPE_I32:
|
||||
|
|
@ -5652,6 +5884,7 @@ static void ggml_compute_forward_sqrt(
|
|||
} break;
|
||||
case GGML_TYPE_Q4_0:
|
||||
case GGML_TYPE_Q4_1:
|
||||
case GGML_TYPE_Q8_0:
|
||||
case GGML_TYPE_I8:
|
||||
case GGML_TYPE_I16:
|
||||
case GGML_TYPE_I32:
|
||||
|
|
@ -5710,6 +5943,7 @@ static void ggml_compute_forward_sum(
|
|||
} break;
|
||||
case GGML_TYPE_Q4_0:
|
||||
case GGML_TYPE_Q4_1:
|
||||
case GGML_TYPE_Q8_0:
|
||||
case GGML_TYPE_I8:
|
||||
case GGML_TYPE_I16:
|
||||
case GGML_TYPE_I32:
|
||||
|
|
@ -5787,6 +6021,7 @@ static void ggml_compute_forward_mean(
|
|||
} break;
|
||||
case GGML_TYPE_Q4_0:
|
||||
case GGML_TYPE_Q4_1:
|
||||
case GGML_TYPE_Q8_0:
|
||||
case GGML_TYPE_I8:
|
||||
case GGML_TYPE_I16:
|
||||
case GGML_TYPE_I32:
|
||||
|
|
@ -5851,6 +6086,7 @@ static void ggml_compute_forward_repeat(
|
|||
} break;
|
||||
case GGML_TYPE_Q4_0:
|
||||
case GGML_TYPE_Q4_1:
|
||||
case GGML_TYPE_Q8_0:
|
||||
case GGML_TYPE_I8:
|
||||
case GGML_TYPE_I16:
|
||||
case GGML_TYPE_I32:
|
||||
|
|
@ -5899,6 +6135,7 @@ static void ggml_compute_forward_abs(
|
|||
} break;
|
||||
case GGML_TYPE_Q4_0:
|
||||
case GGML_TYPE_Q4_1:
|
||||
case GGML_TYPE_Q8_0:
|
||||
case GGML_TYPE_I8:
|
||||
case GGML_TYPE_I16:
|
||||
case GGML_TYPE_I32:
|
||||
|
|
@ -5947,6 +6184,7 @@ static void ggml_compute_forward_sgn(
|
|||
} break;
|
||||
case GGML_TYPE_Q4_0:
|
||||
case GGML_TYPE_Q4_1:
|
||||
case GGML_TYPE_Q8_0:
|
||||
case GGML_TYPE_I8:
|
||||
case GGML_TYPE_I16:
|
||||
case GGML_TYPE_I32:
|
||||
|
|
@ -5995,6 +6233,7 @@ static void ggml_compute_forward_neg(
|
|||
} break;
|
||||
case GGML_TYPE_Q4_0:
|
||||
case GGML_TYPE_Q4_1:
|
||||
case GGML_TYPE_Q8_0:
|
||||
case GGML_TYPE_I8:
|
||||
case GGML_TYPE_I16:
|
||||
case GGML_TYPE_I32:
|
||||
|
|
@ -6043,6 +6282,7 @@ static void ggml_compute_forward_step(
|
|||
} break;
|
||||
case GGML_TYPE_Q4_0:
|
||||
case GGML_TYPE_Q4_1:
|
||||
case GGML_TYPE_Q8_0:
|
||||
case GGML_TYPE_I8:
|
||||
case GGML_TYPE_I16:
|
||||
case GGML_TYPE_I32:
|
||||
|
|
@ -6091,6 +6331,7 @@ static void ggml_compute_forward_relu(
|
|||
} break;
|
||||
case GGML_TYPE_Q4_0:
|
||||
case GGML_TYPE_Q4_1:
|
||||
case GGML_TYPE_Q8_0:
|
||||
case GGML_TYPE_I8:
|
||||
case GGML_TYPE_I16:
|
||||
case GGML_TYPE_I32:
|
||||
|
|
@ -6156,6 +6397,7 @@ static void ggml_compute_forward_gelu(
|
|||
} break;
|
||||
case GGML_TYPE_Q4_0:
|
||||
case GGML_TYPE_Q4_1:
|
||||
case GGML_TYPE_Q8_0:
|
||||
case GGML_TYPE_I8:
|
||||
case GGML_TYPE_I16:
|
||||
case GGML_TYPE_I32:
|
||||
|
|
@ -6223,6 +6465,7 @@ static void ggml_compute_forward_silu(
|
|||
} break;
|
||||
case GGML_TYPE_Q4_0:
|
||||
case GGML_TYPE_Q4_1:
|
||||
case GGML_TYPE_Q8_0:
|
||||
case GGML_TYPE_I8:
|
||||
case GGML_TYPE_I16:
|
||||
case GGML_TYPE_I32:
|
||||
|
|
@ -6309,6 +6552,7 @@ static void ggml_compute_forward_norm(
|
|||
} break;
|
||||
case GGML_TYPE_Q4_0:
|
||||
case GGML_TYPE_Q4_1:
|
||||
case GGML_TYPE_Q8_0:
|
||||
case GGML_TYPE_I8:
|
||||
case GGML_TYPE_I16:
|
||||
case GGML_TYPE_I32:
|
||||
|
|
@ -6389,6 +6633,7 @@ static void ggml_compute_forward_rms_norm(
|
|||
} break;
|
||||
case GGML_TYPE_Q4_0:
|
||||
case GGML_TYPE_Q4_1:
|
||||
case GGML_TYPE_Q8_0:
|
||||
case GGML_TYPE_I8:
|
||||
case GGML_TYPE_I16:
|
||||
case GGML_TYPE_I32:
|
||||
|
|
@ -6788,9 +7033,9 @@ static void ggml_compute_forward_mul_mat_f16_f32(
|
|||
static const quantize_fns_t quantize_fns[GGML_TYPE_COUNT] = {
|
||||
[GGML_TYPE_Q4_0] = {
|
||||
.dequantize_row_q = dequantize_row_q4_0,
|
||||
.quantize_row_q = quantize_row_q4_0,
|
||||
.quantize_row_q = quantize_row_q8_0,
|
||||
.quantize_row_q_reference = (quantize_row_q_t) quantize_row_q4_0_reference,
|
||||
.vec_dot_q = ggml_vec_dot_q4_0,
|
||||
.vec_dot_q = ggml_vec_dot_q4_0_q8_0,
|
||||
},
|
||||
[GGML_TYPE_Q4_1] = {
|
||||
.dequantize_row_q = dequantize_row_q4_1,
|
||||
|
|
@ -6798,6 +7043,7 @@ static const quantize_fns_t quantize_fns[GGML_TYPE_COUNT] = {
|
|||
.quantize_row_q_reference = (quantize_row_q_t) quantize_row_q4_1_reference,
|
||||
.vec_dot_q = ggml_vec_dot_q4_1,
|
||||
},
|
||||
// TODO: GGML_TYPE_Q8_0
|
||||
};
|
||||
|
||||
// For internal test use
|
||||
|
|
@ -6923,7 +7169,7 @@ static void ggml_compute_forward_mul_mat_q_f32(
|
|||
|
||||
if (params->type == GGML_TASK_INIT) {
|
||||
char * wdata = params->wdata;
|
||||
const size_t row_size = ne10*GGML_TYPE_SIZE[type]/GGML_BLCK_SIZE[type];
|
||||
const size_t row_size = ne10*GGML_TYPE_SIZE[GGML_TYPE_Q8_0]/GGML_BLCK_SIZE[GGML_TYPE_Q8_0];
|
||||
|
||||
for (int64_t i13 = 0; i13 < ne13; ++i13) {
|
||||
for (int64_t i12 = 0; i12 < ne12; ++i12) {
|
||||
|
|
@ -6954,7 +7200,7 @@ static void ggml_compute_forward_mul_mat_q_f32(
|
|||
const int ir1 = MIN(ir0 + dr, nr);
|
||||
|
||||
void * wdata = params->wdata;
|
||||
const size_t row_size = ne00*GGML_TYPE_SIZE[type]/GGML_BLCK_SIZE[type];
|
||||
const size_t row_size = ne00*GGML_TYPE_SIZE[GGML_TYPE_Q8_0]/GGML_BLCK_SIZE[GGML_TYPE_Q8_0];
|
||||
|
||||
for (int ir = ir0; ir < ir1; ++ir) {
|
||||
// src0 indices
|
||||
|
|
@ -7002,6 +7248,7 @@ static void ggml_compute_forward_mul_mat(
|
|||
switch (src0->type) {
|
||||
case GGML_TYPE_Q4_0:
|
||||
case GGML_TYPE_Q4_1:
|
||||
case GGML_TYPE_Q8_0:
|
||||
{
|
||||
ggml_compute_forward_mul_mat_q_f32(params, src0, src1, dst);
|
||||
} break;
|
||||
|
|
@ -7100,6 +7347,7 @@ static void ggml_compute_forward_scale(
|
|||
} break;
|
||||
case GGML_TYPE_Q4_0:
|
||||
case GGML_TYPE_Q4_1:
|
||||
case GGML_TYPE_Q8_0:
|
||||
case GGML_TYPE_I8:
|
||||
case GGML_TYPE_I16:
|
||||
case GGML_TYPE_I32:
|
||||
|
|
@ -7265,6 +7513,7 @@ static void ggml_compute_forward_get_rows(
|
|||
switch (src0->type) {
|
||||
case GGML_TYPE_Q4_0:
|
||||
case GGML_TYPE_Q4_1:
|
||||
case GGML_TYPE_Q8_0:
|
||||
{
|
||||
ggml_compute_forward_get_rows_q(params, src0, src1, dst);
|
||||
} break;
|
||||
|
|
@ -7354,6 +7603,7 @@ static void ggml_compute_forward_diag_mask_inf(
|
|||
} break;
|
||||
case GGML_TYPE_Q4_0:
|
||||
case GGML_TYPE_Q4_1:
|
||||
case GGML_TYPE_Q8_0:
|
||||
case GGML_TYPE_I8:
|
||||
case GGML_TYPE_I16:
|
||||
case GGML_TYPE_I32:
|
||||
|
|
@ -7448,6 +7698,7 @@ static void ggml_compute_forward_soft_max(
|
|||
} break;
|
||||
case GGML_TYPE_Q4_0:
|
||||
case GGML_TYPE_Q4_1:
|
||||
case GGML_TYPE_Q8_0:
|
||||
case GGML_TYPE_I8:
|
||||
case GGML_TYPE_I16:
|
||||
case GGML_TYPE_I32:
|
||||
|
|
@ -7623,6 +7874,7 @@ static void ggml_compute_forward_rope(
|
|||
} break;
|
||||
case GGML_TYPE_Q4_0:
|
||||
case GGML_TYPE_Q4_1:
|
||||
case GGML_TYPE_Q8_0:
|
||||
case GGML_TYPE_I8:
|
||||
case GGML_TYPE_I16:
|
||||
case GGML_TYPE_I32:
|
||||
|
|
@ -7891,6 +8143,7 @@ static void ggml_compute_forward_conv_1d_1s(
|
|||
} break;
|
||||
case GGML_TYPE_Q4_0:
|
||||
case GGML_TYPE_Q4_1:
|
||||
case GGML_TYPE_Q8_0:
|
||||
case GGML_TYPE_I8:
|
||||
case GGML_TYPE_I16:
|
||||
case GGML_TYPE_I32:
|
||||
|
|
@ -8159,6 +8412,7 @@ static void ggml_compute_forward_conv_1d_2s(
|
|||
} break;
|
||||
case GGML_TYPE_Q4_0:
|
||||
case GGML_TYPE_Q4_1:
|
||||
case GGML_TYPE_Q8_0:
|
||||
case GGML_TYPE_I8:
|
||||
case GGML_TYPE_I16:
|
||||
case GGML_TYPE_I32:
|
||||
|
|
@ -8644,6 +8898,7 @@ static void ggml_compute_forward_flash_attn(
|
|||
} break;
|
||||
case GGML_TYPE_Q4_0:
|
||||
case GGML_TYPE_Q4_1:
|
||||
case GGML_TYPE_Q8_0:
|
||||
case GGML_TYPE_I8:
|
||||
case GGML_TYPE_I16:
|
||||
case GGML_TYPE_I32:
|
||||
|
|
@ -8855,6 +9110,7 @@ static void ggml_compute_forward_flash_ff(
|
|||
} break;
|
||||
case GGML_TYPE_Q4_0:
|
||||
case GGML_TYPE_Q4_1:
|
||||
case GGML_TYPE_Q8_0:
|
||||
case GGML_TYPE_I8:
|
||||
case GGML_TYPE_I16:
|
||||
case GGML_TYPE_I32:
|
||||
|
|
@ -9665,9 +9921,7 @@ void ggml_graph_compute(struct ggml_context * ctx, struct ggml_cgraph * cgraph)
|
|||
cur = GGML_TYPE_SIZE[GGML_TYPE_F32]*(node->src0->ne[0]*node->src0->ne[1]);
|
||||
} else
|
||||
#endif
|
||||
{
|
||||
cur = GGML_TYPE_SIZE[node->src0->type]*ggml_nelements(node->src1)/GGML_BLCK_SIZE[node->src0->type];
|
||||
}
|
||||
cur = GGML_TYPE_SIZE[GGML_TYPE_Q8_0]*ggml_nelements(node->src1)/GGML_BLCK_SIZE[GGML_TYPE_Q8_0];
|
||||
} else {
|
||||
GGML_ASSERT(false);
|
||||
}
|
||||
|
|
|
|||
1
ggml.h
1
ggml.h
|
|
@ -204,6 +204,7 @@ enum ggml_type {
|
|||
GGML_TYPE_F16 = 1,
|
||||
GGML_TYPE_Q4_0 = 2,
|
||||
GGML_TYPE_Q4_1 = 3,
|
||||
GGML_TYPE_Q8_0 = 4,
|
||||
GGML_TYPE_I8,
|
||||
GGML_TYPE_I16,
|
||||
GGML_TYPE_I32,
|
||||
|
|
|
|||
Loading…
Add table
Add a link
Reference in a new issue