char-fork/llama.cpp

git clone https://git.t4t.associates/char-fork/llama.cpp

Neo Zhang Jianyurefactor pad_reflect_1d to make the UT case pass (#17204)efaaccdd6

char/aliceai-t5-moe
brancheschar/aliceai-t5-moeall refs →
3.5 KiB100 linesraw
1#include "pad_reflect_1d.hpp"
2
3static void pad_reflect_1d_kernel_f32(
4    const void *__restrict__ src0, void *__restrict__ dst, const int64_t ne0,
5    const int64_t ne00, const sycl::uint3 ne01, const int64_t ne02,
6    const int64_t ne03, const int64_t nb00, const int64_t nb01,
7    const int64_t nb02, const int64_t nb03, const int64_t nb0,
8    const int64_t nb1, const int64_t nb2, const int64_t nb3, const int p0,
9    const int p1, sycl::nd_item<3> item_ct1) {
10
11    const int64_t i3 = item_ct1.get_group(0);
12    const int64_t i2 = item_ct1.get_group(1);
13
14    const sycl::uint2 div_mod_packed =
15        fast_div_modulo(item_ct1.get_group(2), ne01);
16    const int64_t tile1 = div_mod_packed.y();
17    const int64_t tile0 = div_mod_packed.x();
18    const int64_t i1 = tile1;
19    const int64_t i0 =
20        item_ct1.get_local_id(2) + tile0 * item_ct1.get_local_range(2);
21
22    if (i0 >= ne0 || i1 >= ne01.z() || i2 >= ne02 || i3 >= ne03) {
23        return;
24    }
25
26    const char *src0_ptr =
27        (const char *)src0 + i3 * nb03 + i2 * nb02 + i1 * nb01;
28    char *dst_ptr = (char *)dst + i3 * nb3 + i2 * nb2 + i1 * nb1;
29
30    const int64_t rel_i0 = i0 - p0; // relative i0 in src0
31    int64_t src_idx;
32
33    if (rel_i0 < 0) {
34        // Left padding - reflect
35        src_idx = -rel_i0;
36    } else if (rel_i0 < ne00) {
37        // Middle - copy
38        src_idx = rel_i0;
39    } else {
40        // Right padding - reflect
41        src_idx = 2 * ne00 - 2 - rel_i0;
42    }
43    const float value = *(const float *)(src0_ptr + src_idx * nb00);
44    *(float *)(dst_ptr + i0 * nb0) = value;
45
46    GGML_UNUSED(p1);
47}
48
49void ggml_sycl_op_pad_reflect_1d(ggml_backend_sycl_context &ctx,
50                                 ggml_tensor *dst) {
51
52    const ggml_tensor *src0 = dst->src[0];
53    dpct::queue_ptr stream = ctx.stream();
54
55    GGML_ASSERT(src0->type == GGML_TYPE_F32);
56    GGML_ASSERT(dst->type == GGML_TYPE_F32);
57
58    const int32_t *opts = (const int32_t *)dst->op_params;
59    const int p0 = opts[0];
60    const int p1 = opts[1];
61
62    const int64_t ne00 = src0->ne[0];
63    const int64_t ne01 = src0->ne[1];
64    const sycl::uint3 ne01_packed = init_fastdiv_values(ne01);
65    const int64_t ne02 = src0->ne[2];
66    const int64_t ne03 = src0->ne[3];
67
68    const int64_t ne0 = dst->ne[0];
69
70    GGML_ASSERT(ne0 == ne00 + p0 + p1);
71
72    constexpr int64_t bx = SYCL_PAD_REFLECT_1D_BLOCK_SIZE;
73    const int64_t tiles0 = (ne0 + bx - 1) / bx;
74    const dpct::dim3 grid_dims((unsigned)(ne01 * tiles0), (unsigned)ne02,
75                               (unsigned)ne03);
76    const dpct::dim3 block_dims((unsigned)bx, 1, 1);
77
78    stream->submit([&](sycl::handler &cgh) {
79        auto src0_data_ct0 = src0->data;
80        auto dst_data_ct1 = dst->data;
81        auto src0_nb_ct7 = src0->nb[0];
82        auto src0_nb_ct8 = src0->nb[1];
83        auto src0_nb_ct9 = src0->nb[2];
84        auto src0_nb_ct10 = src0->nb[3];
85        auto dst_nb_ct11 = dst->nb[0];
86        auto dst_nb_ct12 = dst->nb[1];
87        auto dst_nb_ct13 = dst->nb[2];
88        auto dst_nb_ct14 = dst->nb[3];
89
90        cgh.parallel_for(sycl::nd_range<3>(grid_dims * block_dims, block_dims),
91                         [=](sycl::nd_item<3> item_ct1) {
92                             pad_reflect_1d_kernel_f32(
93                                 src0_data_ct0, dst_data_ct1, ne0, ne00,
94                                 ne01_packed, ne02, ne03, src0_nb_ct7,
95                                 src0_nb_ct8, src0_nb_ct9, src0_nb_ct10,
96                                 dst_nb_ct11, dst_nb_ct12, dst_nb_ct13,
97                                 dst_nb_ct14, p0, p1, item_ct1);
98                         });
99    });
100}