File size: 7,584 Bytes
b664585
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
#include "common.cuh"

struct mma_int_A_I16K4 {
    static constexpr int I  = 16;
    static constexpr int K  = 4;
    static constexpr int ne = 2;

    int x[ne] = {0};

    static __device__ __forceinline__ int get_i(const int l) {
        const int ret = (l%2) * (I/2) + threadIdx.x / K;
        GGML_CUDA_ASSUME(ret >= 0);
        GGML_CUDA_ASSUME(ret <  I);
        return ret;
    }

    static __device__ __forceinline__ int get_k(const int /* l */) {
        const int ret = threadIdx.x % K;
        GGML_CUDA_ASSUME(ret >= 0);
        GGML_CUDA_ASSUME(ret <  K);
        return ret;
    }

    __device__ __forceinline__ void load(const int * __restrict__ xs0, const int & stride) {
#if defined(INT8_MMA_AVAILABLE)
        const int * xs = xs0 + (threadIdx.x%I)*stride;
        asm("ldmatrix.sync.aligned.m8n8.x2.b16 {%0, %1}, [%2];"
            : "+r"(x[0]), "+r"(x[1])
            : "l"(xs));
#else
#pragma unroll
        for (int l = 0; l < ne; ++l) {
            x[l] = xs0[get_i(l)*stride + get_k(l)];
        }
#endif // defined(INT8_MMA_AVAILABLE)
    }
};

struct mma_int_A_I16K8 {
    static constexpr int I  = 16;
    static constexpr int K  = 8;
    static constexpr int ne = 4;

    int x[ne] = {0};

    static __device__ __forceinline__ int get_i(const int l) {
        const int ret = (l%2) * (I/2) + threadIdx.x / (K/2);
        GGML_CUDA_ASSUME(ret >= 0);
        GGML_CUDA_ASSUME(ret <  I);
        return ret;
    }

    static __device__ __forceinline__ int get_k(const int l) {
        const int ret = (l/2) * (K/2) + threadIdx.x % (K/2);
        GGML_CUDA_ASSUME(ret >= 0);
        GGML_CUDA_ASSUME(ret <  K);
        return ret;
    }

    __device__ __forceinline__ void load(const int * __restrict__ xs0, const int & stride) {
#if defined(INT8_MMA_AVAILABLE)
        const int * xs = xs0 + (threadIdx.x%I)*stride + (threadIdx.x/I)*(K/2);
        asm("ldmatrix.sync.aligned.m8n8.x4.b16 {%0, %1, %2, %3}, [%4];"
            : "+r"(x[0]), "+r"(x[1]), "+r"(x[2]), "+r"(x[3])
            : "l"(xs));
#else
#pragma unroll
        for (int l = 0; l < ne; ++l) {
            x[l] = xs0[get_i(l)*stride + get_k(l)];
        }
#endif // defined(INT8_MMA_AVAILABLE)
    }

    __device__ __forceinline__ void load_low(const int * __restrict__ xs0, const int & stride) {
        ((mma_int_A_I16K4 *) x)[0].load(xs0, stride);
    }
};

struct mma_int_B_J8K4 {
    static constexpr int J  = 8;
    static constexpr int K  = 4;
    static constexpr int ne = 1;

    int x[ne] = {0};

    static __device__ __forceinline__ int get_j(const int /* l */) {
        const int ret = threadIdx.x / K;
        GGML_CUDA_ASSUME(ret >= 0);
        GGML_CUDA_ASSUME(ret <  J);
        return ret;
    }

    static __device__ __forceinline__ int get_k(const int /* l */) {
        const int ret = threadIdx.x % K;
        GGML_CUDA_ASSUME(ret >= 0);
        GGML_CUDA_ASSUME(ret <  K);
        return ret;
    }

    __device__ __forceinline__ void load(const int * __restrict__ xs0, const int & stride) {
#if defined(INT8_MMA_AVAILABLE) && false // Loading as 4 byte values is faster
        const int * xs = xs0 + (threadIdx.x%J)*stride;
        asm("ldmatrix.sync.aligned.m8n8.x1.b16 {%0}, [%1];"
            : "+r"(x[0])
            : "l"(xs));
#else
#pragma unroll
        for (int l = 0; l < ne; ++l) {
            x[l] = xs0[get_j(l)*stride + get_k(l)];
        }
#endif // defined(INT8_MMA_AVAILABLE)
    }
};

struct mma_int_B_J8K8 {
    static constexpr int J  = 8;
    static constexpr int K  = 8;
    static constexpr int ne = 2;

    int x[ne] = {0};

    static __device__ __forceinline__ int get_j(const int /* l */) {
        const int ret = threadIdx.x / (K/2);
        GGML_CUDA_ASSUME(ret >= 0);
        GGML_CUDA_ASSUME(ret <  J);
        return ret;
    }

    static __device__ __forceinline__ int get_k(const int l) {
        const int ret = l * (K/2) + threadIdx.x % (K/2);
        GGML_CUDA_ASSUME(ret >= 0);
        GGML_CUDA_ASSUME(ret <  K);
        return ret;
    }

    __device__ __forceinline__ void load(const int * __restrict__ xs0, const int & stride) {
#if defined(INT8_MMA_AVAILABLE) && false // Loading as 4 byte values is faster
        const int * xs = xs0 + (threadIdx.x%J)*stride + ((threadIdx.x/J)*(K/2)) % K;
        asm("ldmatrix.sync.aligned.m8n8.x2.b16 {%0, %1}, [%2];"
            : "+r"(x[0]), "+r"(x[1])
            : "l"(xs));
#else
#pragma unroll
        for (int l = 0; l < ne; ++l) {
            x[l] = xs0[get_j(l)*stride + get_k(l)];
        }
#endif // defined(INT8_MMA_AVAILABLE)
    }
};

struct mma_int_C_I16J8 {
    static constexpr int I  = 16;
    static constexpr int J  = 8;
    static constexpr int ne = 4;

    int x[ne] = {0};

    static __device__ __forceinline__ int get_i(const int l) {
        const int ret = (l/2) * (I/2) + threadIdx.x / (J/2);
        GGML_CUDA_ASSUME(ret >= 0);
        GGML_CUDA_ASSUME(ret <  I);
        return ret;
    }

    static __device__ __forceinline__ int get_j(const int l) {
        const int ret = 2 * (threadIdx.x % (J/2)) + l%2;
        GGML_CUDA_ASSUME(ret >= 0);
        GGML_CUDA_ASSUME(ret <  J);
        return ret;
    }

    __device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) {
#ifdef INT8_MMA_AVAILABLE
#if __CUDA_ARCH__ >= CC_AMPERE
        asm("mma.sync.aligned.m16n8k16.row.col.s32.s8.s8.s32 {%0, %1, %2, %3}, {%4, %5}, {%6}, {%0, %1, %2, %3};"
            : "+r"(x[0]), "+r"(x[1]), "+r"(x[2]), "+r"(x[3])
            : "r"(mma_A.x[0]), "r"(mma_A.x[1]), "r"(mma_B.x[0]));
#else
        // On Turing m16n8k16 mma is not available, use 2x m8n8k16 mma instead:
        asm("mma.sync.aligned.m8n8k16.row.col.s32.s8.s8.s32 {%0, %1}, {%2}, {%3}, {%0, %1};"
            : "+r"(x[0]), "+r"(x[1])
            : "r"(mma_A.x[0]), "r"(mma_B.x[0]));
        asm("mma.sync.aligned.m8n8k16.row.col.s32.s8.s8.s32 {%0, %1}, {%2}, {%3}, {%0, %1};"
            : "+r"(x[2]), "+r"(x[3])
            : "r"(mma_A.x[1]), "r"(mma_B.x[0]));
#endif // __CUDA_ARCH__ >= CC_AMPERE
#else
        GGML_UNUSED(mma_A);
        GGML_UNUSED(mma_B);
        NO_DEVICE_CODE;
#endif // INT8_MMA_AVAILABLE
    }

    __device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) {
#ifdef INT8_MMA_AVAILABLE
#if __CUDA_ARCH__ >= CC_AMPERE
        asm("mma.sync.aligned.m16n8k32.row.col.s32.s8.s8.s32 {%0, %1, %2, %3}, {%4, %5, %6, %7}, {%8, %9}, {%0, %1, %2, %3};"
            : "+r"(x[0]), "+r"(x[1]), "+r"(x[2]), "+r"(x[3])
            : "r"(mma_A.x[0]), "r"(mma_A.x[1]), "r"(mma_A.x[2]), "r"(mma_A.x[3]), "r"(mma_B.x[0]), "r"(mma_B.x[1]));
#else
        // On Turing m16n8k32 mma is not available, use 4x m8n8k16 mma instead:
        asm("mma.sync.aligned.m8n8k16.row.col.s32.s8.s8.s32 {%0, %1}, {%2}, {%3}, {%0, %1};"
            : "+r"(x[0]), "+r"(x[1])
            : "r"(mma_A.x[0]), "r"(mma_B.x[0]));
        asm("mma.sync.aligned.m8n8k16.row.col.s32.s8.s8.s32 {%0, %1}, {%2}, {%3}, {%0, %1};"
            : "+r"(x[2]), "+r"(x[3])
            : "r"(mma_A.x[1]), "r"(mma_B.x[0]));
        asm("mma.sync.aligned.m8n8k16.row.col.s32.s8.s8.s32 {%0, %1}, {%2}, {%3}, {%0, %1};"
            : "+r"(x[0]), "+r"(x[1])
            : "r"(mma_A.x[2]), "r"(mma_B.x[1]));
        asm("mma.sync.aligned.m8n8k16.row.col.s32.s8.s8.s32 {%0, %1}, {%2}, {%3}, {%0, %1};"
            : "+r"(x[2]), "+r"(x[3])
            : "r"(mma_A.x[3]), "r"(mma_B.x[1]));
#endif // __CUDA_ARCH__ >= CC_AMPERE
#else
        GGML_UNUSED(mma_A);
        GGML_UNUSED(mma_B);
        NO_DEVICE_CODE;
#endif // INT8_MMA_AVAILABLE
    }
};