Skip to content

Adding SSE instructions to ggml_vec_dot_q4_0_q8_0 #1413

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 9 commits into from
May 13, 2023
Merged

Adding SSE instructions to ggml_vec_dot_q4_0_q8_0 #1413

Changes from all commits
Commits
Show all changes
9 commits
Select commit Hold shift + click to select a range
524d6c9
- added sse instructions for ggml_vec_dot_q4_0_q8_0
May 12, 2023
607b9c7
- Split multiplication and addition to make it easier for the compile…
May 12, 2023
78bbb3c
- Use 4 accumulations instead of 2
May 12, 2023
7379dd2
- Added prefetch
May 12, 2023
8699fd0
- Cleanup
May 12, 2023
ca54314
- Improved prefetching
May 12, 2023
70c2b6c
Put __SSE3__ into defined()
May 12, 2023
fc26f54
- Put the whole line into defined()
May 12, 2023
25b448a
- rearranged defines, SSSE3 function only compiled if used
May 12, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
135 changes: 134 additions & 1 deletion ggml.c
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -472,7 +472,7 @@ static const size_t CACHE_LINE_SIZE_F32 = CACHE_LINE_SIZE/sizeof(float);
// quantization
//

#if __AVX__ || __AVX2__ || __AVX512F__
#if defined(__AVX__) || defined(__AVX2__) || defined(__AVX512F__) || defined(__SSSE3__)
// multiply int8_t, add results pairwise twice
static inline __m128i mul_sum_i8_pairs(const __m128i x, const __m128i y) {
// Get absolute values of x vectors
Expand All @@ -485,6 +485,7 @@ static inline __m128i mul_sum_i8_pairs(const __m128i x, const __m128i y) {
return _mm_madd_epi16(ones, dot);
}

#if __AVX__ || __AVX2__ || __AVX512F__
// horizontally add 8 floats
static inline float hsum_float_8(const __m256 x) {
__m128 res = _mm256_extractf128_ps(x, 1);
Expand Down Expand Up @@ -596,7 +597,19 @@ static inline __m128i packNibbles( __m128i bytes1, __m128i bytes2 )
return _mm_packus_epi16( bytes1, bytes2);
}
#endif
#elif defined(__SSSE3__)
// horizontally add 4x4 floats
static inline float hsum_float_4x4(const __m128 a, const __m128 b, const __m128 c, const __m128 d) {
__m128 res_0 =_mm_hadd_ps(a, b);
__m128 res_1 =_mm_hadd_ps(c, d);
__m128 res =_mm_hadd_ps(res_0, res_1);
res =_mm_hadd_ps(res, res);
res =_mm_hadd_ps(res, res);

return _mm_cvtss_f32(res);
}
#endif // __AVX__ || __AVX2__ || __AVX512F__
#endif // defined(__AVX__) || defined(__AVX2__) || defined(__AVX512F__) || defined(__SSSE3__)

#if __ARM_NEON

Expand Down Expand Up @@ -2129,6 +2142,126 @@ static void ggml_vec_dot_q4_0_q8_0(const int n, float * restrict s, const void *
}

*s = hsum_float_8(acc);
#elif defined(__SSSE3__)
// set constants
const __m128i lowMask = _mm_set1_epi8(0xF);
const __m128i off = _mm_set1_epi8(8);

// Initialize accumulator with zeros
__m128 acc_0 = _mm_setzero_ps();
__m128 acc_1 = _mm_setzero_ps();
__m128 acc_2 = _mm_setzero_ps();
__m128 acc_3 = _mm_setzero_ps();

// First round without accumulation
{
_mm_prefetch(&x[0] + sizeof(block_q4_0), _MM_HINT_T0);
_mm_prefetch(&y[0] + sizeof(block_q8_0), _MM_HINT_T0);

// Compute combined scale for the block 0 and 1
const __m128 d_0_1 = _mm_mul_ps( _mm_set1_ps( x[0].d ), _mm_set1_ps( y[0].d ) );

const __m128i tmp_0_1 = _mm_loadu_si128((const __m128i *)x[0].qs);

__m128i bx_0 = _mm_and_si128(lowMask, tmp_0_1);
__m128i by_0 = _mm_loadu_si128((const __m128i *)y[0].qs);
bx_0 = _mm_sub_epi8(bx_0, off);
const __m128i i32_0 = mul_sum_i8_pairs(bx_0, by_0);

__m128i bx_1 = _mm_and_si128(lowMask, _mm_srli_epi64(tmp_0_1, 4));
__m128i by_1 = _mm_loadu_si128((const __m128i *)(y[0].qs + 16));
bx_1 = _mm_sub_epi8(bx_1, off);
const __m128i i32_1 = mul_sum_i8_pairs(bx_1, by_1);

_mm_prefetch(&x[1] + sizeof(block_q4_0), _MM_HINT_T0);
_mm_prefetch(&y[1] + sizeof(block_q8_0), _MM_HINT_T0);

// Compute combined scale for the block 2 and 3
const __m128 d_2_3 = _mm_mul_ps( _mm_set1_ps( x[1].d ), _mm_set1_ps( y[1].d ) );

const __m128i tmp_2_3 = _mm_loadu_si128((const __m128i *)x[1].qs);

__m128i bx_2 = _mm_and_si128(lowMask, tmp_2_3);
__m128i by_2 = _mm_loadu_si128((const __m128i *)y[1].qs);
bx_2 = _mm_sub_epi8(bx_2, off);
const __m128i i32_2 = mul_sum_i8_pairs(bx_2, by_2);

__m128i bx_3 = _mm_and_si128(lowMask, _mm_srli_epi64(tmp_2_3, 4));
__m128i by_3 = _mm_loadu_si128((const __m128i *)(y[1].qs + 16));
bx_3 = _mm_sub_epi8(bx_3, off);
const __m128i i32_3 = mul_sum_i8_pairs(bx_3, by_3);

// Convert int32_t to float
__m128 p0 = _mm_cvtepi32_ps(i32_0);
__m128 p1 = _mm_cvtepi32_ps(i32_1);
__m128 p2 = _mm_cvtepi32_ps(i32_2);
__m128 p3 = _mm_cvtepi32_ps(i32_3);

// Apply the scale
acc_0 = _mm_mul_ps( d_0_1, p0 );
acc_1 = _mm_mul_ps( d_0_1, p1 );
acc_2 = _mm_mul_ps( d_2_3, p2 );
acc_3 = _mm_mul_ps( d_2_3, p3 );
}

// Main loop
for (int i = 2; i < nb; i+=2) {
_mm_prefetch(&x[i] + sizeof(block_q4_0), _MM_HINT_T0);
_mm_prefetch(&y[i] + sizeof(block_q8_0), _MM_HINT_T0);

// Compute combined scale for the block 0 and 1
const __m128 d_0_1 = _mm_mul_ps( _mm_set1_ps( x[i].d ), _mm_set1_ps( y[i].d ) );

const __m128i tmp_0_1 = _mm_loadu_si128((const __m128i *)x[i].qs);

__m128i bx_0 = _mm_and_si128(lowMask, tmp_0_1);
__m128i by_0 = _mm_loadu_si128((const __m128i *)y[i].qs);
bx_0 = _mm_sub_epi8(bx_0, off);
const __m128i i32_0 = mul_sum_i8_pairs(bx_0, by_0);

__m128i bx_1 = _mm_and_si128(lowMask, _mm_srli_epi64(tmp_0_1, 4));
__m128i by_1 = _mm_loadu_si128((const __m128i *)(y[i].qs + 16));
bx_1 = _mm_sub_epi8(bx_1, off);
const __m128i i32_1 = mul_sum_i8_pairs(bx_1, by_1);

_mm_prefetch(&x[i] + 2 * sizeof(block_q4_0), _MM_HINT_T0);
_mm_prefetch(&y[i] + 2 * sizeof(block_q8_0), _MM_HINT_T0);

// Compute combined scale for the block 2 and 3
const __m128 d_2_3 = _mm_mul_ps( _mm_set1_ps( x[i + 1].d ), _mm_set1_ps( y[i + 1].d ) );

const __m128i tmp_2_3 = _mm_loadu_si128((const __m128i *)x[i + 1].qs);

__m128i bx_2 = _mm_and_si128(lowMask, tmp_2_3);
__m128i by_2 = _mm_loadu_si128((const __m128i *)y[i + 1].qs);
bx_2 = _mm_sub_epi8(bx_2, off);
const __m128i i32_2 = mul_sum_i8_pairs(bx_2, by_2);

__m128i bx_3 = _mm_and_si128(lowMask, _mm_srli_epi64(tmp_2_3, 4));
__m128i by_3 = _mm_loadu_si128((const __m128i *)(y[i + 1].qs + 16));
bx_3 = _mm_sub_epi8(bx_3, off);
const __m128i i32_3 = mul_sum_i8_pairs(bx_3, by_3);

// Convert int32_t to float
__m128 p0 = _mm_cvtepi32_ps(i32_0);
__m128 p1 = _mm_cvtepi32_ps(i32_1);
__m128 p2 = _mm_cvtepi32_ps(i32_2);
__m128 p3 = _mm_cvtepi32_ps(i32_3);

// Apply the scale
__m128 p0_d = _mm_mul_ps( d_0_1, p0 );
__m128 p1_d = _mm_mul_ps( d_0_1, p1 );
__m128 p2_d = _mm_mul_ps( d_2_3, p2 );
__m128 p3_d = _mm_mul_ps( d_2_3, p3 );

// Acummulate
acc_0 = _mm_add_ps(p0_d, acc_0);
acc_1 = _mm_add_ps(p1_d, acc_1);
acc_2 = _mm_add_ps(p2_d, acc_2);
acc_3 = _mm_add_ps(p3_d, acc_3);
}

*s = hsum_float_4x4(acc_0, acc_1, acc_2, acc_3);
#else
// scalar
float sumf = 0.0;
Expand Down