onednn-src 0.1.13

Source of oneAPI Deep Neural Network Library (oneDNN)
Documentation 
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

/*******************************************************************************
* Copyright 2019 Intel Corporation
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
*     http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*******************************************************************************/

#include "gpu/intel/reorder/ref.hpp"
#include "common/utils.hpp"

namespace dnnl {
namespace impl {
namespace gpu {
namespace intel {
namespace reorder {

using namespace dnnl::impl::memory_tracking::names;

status_t ref_t::pd_t::init_conf(impl::engine_t *engine) {
    using namespace format_tag;
    using namespace data_type;

    const memory_desc_wrapper src_mdw(src_md());
    const memory_desc_wrapper dst_mdw(dst_md());

    conf.src_md_info = memory_desc_info_t::create(src_mdw);
    conf.dst_md_info = memory_desc_info_t::create(dst_mdw);

    conf.require_stateless_addressing = has_large_buffers();

    status_t status = status::success;

    const auto &padded_dims = dst_mdw.padded_dims();
    conf.src_quant = {attr(), src_mdw, DNNL_ARG_SRC};
    conf.dst_quant = {attr(), dst_mdw, DNNL_ARG_DST};
    conf.sum_quant = {attr()};
    conf.has_padding = !src_mdw.is_dense() || !dst_mdw.is_dense();
    conf.ndims = src_mdw.ndims();
    conf.nelems = utils::array_product(padded_dims, conf.ndims);

    auto *intel_engine = utils::downcast<intel::engine_t *>(engine);
    auto *device_info = intel_engine->device_info();
    int sub_group_size = 1;
    if (device_info->mayiuse_sub_group(16)) sub_group_size = 16;
    conf.sub_group_size = sub_group_size;

    if (conf.nelems == 0) return status::success;

    conf.dispatch = intel_engine->create_dispatch(dst_mdw.md_);
    conf.subbyte_pack
            = utils::one_of(dst_mdw.data_type(), u4, s4, f4_e2m1, f4_e3m0);

    dim_t blocks[MAX_NDIMS] = {1, 1, 0, 0, 0, 0};
    for (int i = 0; i < MAX_NDIMS; ++i) {
        auto dim_str = utils::format("D%d", i);
        if (i < dst_mdw.ndims()) {
            dim_t dim = padded_dims[i];
            // if needed to align vectorized dim with vector size, pad that dim again
            conf.dispatch.define_dim(dim_str, i, dim, blocks[i]);
        } else {
            conf.dispatch.define_dim(dim_str, 1);
        }
    }

    conf.dispatch.generate();
    return status;
}

status_t ref_t::pd_t::init_kernel_ctx(compute::kernel_ctx_t &kernel_ctx) const {
    using namespace format_tag;
    using namespace data_type;

    const memory_desc_wrapper src_mdw(src_md());
    const memory_desc_wrapper dst_mdw(dst_md());

    if (conf.nelems == 0) return status::success;

    kernel_ctx.define_int("NDIMS", conf.ndims);
    kernel_ctx.add_option("-cl-std=CL2.0");

    conf.src_quant.define_macros(kernel_ctx, "SRC");
    conf.dst_quant.define_macros(kernel_ctx, "DST");
    conf.sum_quant.define_macros(kernel_ctx, "SUM");

    def_dispatch(kernel_ctx, conf.dispatch);
    kernel_ctx.require_stateless_addressing(conf.require_stateless_addressing);

    kernel_ctx.define_int("REF_REORDER", 1);
    kernel_ctx.define_int("PAD_FILL_ZERO", conf.has_padding);
    kernel_ctx.define_int("SUB_GROUP_SIZE", conf.sub_group_size);

    auto dst_rnd_mode = attr()->rounding_mode_.get(DNNL_ARG_DST);
    kernel_ctx.define_int(
            "WITH_SROUND", dst_rnd_mode == rounding_mode::stochastic);
    kernel_ctx.define_int("DST_DT_DIGITS",
            dnnl::impl::types::digits<uint32_t>(conf.dst_md_info.data_type));

    def_memory_desc_info(kernel_ctx, conf.src_md_info, "SRC");
    def_memory_desc_info(kernel_ctx, conf.dst_md_info, "DST");

    return status::success;
}

void ref_t::pd_t::init_scratchpad() {
    auto scratchpad = scratchpad_registry().registrar();
    if (conf.subbyte_pack) {
        scratchpad.book(memory_tracking::names::key_reorder_space, conf.nelems,
                sizeof(char), OCL_BUFFER_ALIGNMENT);
    }
    if (conf.src_quant.with_scale()) {
        scratchpad.book(memory_tracking::names::key_reorder_src_scales,
                conf.src_quant.num_scales(), sizeof(float),
                OCL_BUFFER_ALIGNMENT);
    }
    if (conf.dst_quant.with_scale()) {
        scratchpad.book(memory_tracking::names::key_reorder_dst_scales,
                conf.dst_quant.num_scales(), sizeof(float),
                OCL_BUFFER_ALIGNMENT);
    }
}

status_t ref_t::execute(const exec_ctx_t &ctx) const {
    auto &src = CTX_IN_STORAGE(DNNL_ARG_FROM);
    auto &dst = CTX_OUT_STORAGE(DNNL_ARG_TO);
    auto tmp = ctx.get_scratchpad_grantor().get_memory_storage(
            memory_tracking::names::key_reorder_space);

    const auto &conf = pd()->conf;
    if (conf.nelems == 0) return status::success;

    compute::kernel_arg_list_t arg_list;
    arg_list.append(src);
    arg_list.append(conf.subbyte_pack ? *tmp : dst);

    arg_list.append(conf.src_quant.scales(ctx));
    arg_list.append(conf.src_quant.zero_points(ctx));
    arg_list.append(conf.dst_quant.scales(ctx));
    arg_list.append(conf.dst_quant.zero_points(ctx));

    arg_list.append(conf.sum_quant.scales());
    arg_list.append(conf.sum_quant.zero_points());

    if (!pd()->attr()->rounding_mode_.has_default_values())
        arg_list.append(CTX_IN_STORAGE(DNNL_ARG_ATTR_ROUNDING_SEED));

    auto nd_range = conf.dispatch.nd_range();
    CHECK(large_parallel_for(
            ctx, nd_range, kernels_[0], arg_list, arg_list.nargs()));

    if (conf.subbyte_pack) {
        compute::kernel_arg_list_t repack_arg_list;
        repack_arg_list.set(0, *tmp);
        repack_arg_list.set(1, dst);
        repack_arg_list.set(2, into<dim_t>(conf.nelems));
        repack_arg_list.set(3, 4);
        compute::range_t repack_gws((conf.nelems * 4 + 7) / 8);
        compute::nd_range_t repack_nd_range(repack_gws);
        CHECK(large_parallel_for(
                ctx, repack_nd_range, kernels_[1], repack_arg_list, 4));
    }
    CHECK(pd()->maybe_exec_zp_precompute_conv(ctx, zp_precomp_conv_));
    return status::success;
}

} // namespace reorder
} // namespace intel
} // namespace gpu
} // namespace impl
} // namespace dnnl