Skip to content

M
mindspore

magicwindyyd / mindspore
与 Fork 源项目一致

Fork自 MindSpore / mindspore

通知 1
Star 1
Fork 0
M
mindspore
提交 e48eba0d 编写于 作者: M mindspore-ci-bot 提交者： Gitee
浏览文件
操作

!956 optimize host trans output 

Merge pull request !956 from zjun/optimize_host_trans
上级 76a2f7c6 af690706
隐藏空白更改
内联 并排
Showing with 57 addition and 141 deletion
mindspore/ccsrc/common/trans.cc
浏览文件 @ e48eba0d
/**
* Copyright 2020 Huawei Technologies Co., Ltd
*
......@@ -63,6 +64,27 @@ const std::map<TypeId, size_t> type_map = {{kNumberTypeBool, 1}, {kNumberType
{kNumberTypeUInt32, 4}, {kNumberTypeUInt64, 8}, {kNumberTypeFloat, 4},
{kNumberTypeFloat16, 2}, {kNumberTypeFloat32, 4}, {kNumberTypeFloat64, 8}};
#define SetDataBysize(size, pad_zero) \
do { \
switch (size) { \
case 1: \
static_cast<uint8_t *>(result)[dst_idx] = pad_zero ? 0 : static_cast<const uint8_t *>(args.data)[src_idx]; \
break; \
case 2: \
static_cast<uint16_t *>(result)[dst_idx] = pad_zero ? 0 : static_cast<const uint16_t *>(args.data)[src_idx]; \
break; \
case 4: \
static_cast<uint32_t *>(result)[dst_idx] = pad_zero ? 0 : static_cast<const uint32_t *>(args.data)[src_idx]; \
break; \
case 8: \
static_cast<uint64_t *>(result)[dst_idx] = pad_zero ? 0 : static_cast<const uint64_t *>(args.data)[src_idx]; \
break; \
default: \
MS_LOG(ERROR) << "Trans data not support size " << size; \
return false; \
} \
} while (0)
template <typename T>
T Ceil(T n1, T n2) {
return (n2 != 0) ? (n1 - 1) / n2 + 1 : 0;
......@@ -372,10 +394,10 @@ std::vector<size_t> TransShapeToDevice(const std::vector<size_t> &shape, const s
temp_shape = PaddingShapeTo4dByDefault(shape);
}
auto iter = device_shape_map.find(format);
if (iter != device_shape_map.end()) {
return iter->second(temp_shape);
if (iter == device_shape_map.end()) {
MS_LOG(EXCEPTION) << "Unexpected format[" << format << "]";
}
MS_LOG(EXCEPTION) << "Unexpected format[" << format << "]";
return iter->second(temp_shape);
}
bool CheckArgs(const FormatArgs &args, size_t *size, size_t *total_size) {
......@@ -483,13 +505,13 @@ bool NchwToFracZ(const FormatArgs &args, void *result) {
size_t vf_cnt = c1 * hw;
size_t fractal_ele_cnt = c0 * kCubeSize;
size_t total_ele_cnt = hf_cnt * vf_cnt * fractal_ele_cnt;
size_t dst_size = total_ele_cnt * size;
if (dst_size != args.device_size) {
MS_LOG(ERROR) << "Illegal total data size."
<< "dst size is :" << dst_size << "device size is :" << args.device_size;
return false;
}
for (size_t vfi = 0; vfi < vf_cnt; vfi++) {
auto vf_base_i = vfi * hf_cnt; // vertical fractal matrix base index
for (size_t hfi = 0; hfi < hf_cnt; hfi++) {
......@@ -501,25 +523,10 @@ bool NchwToFracZ(const FormatArgs &args, void *result) {
auto src_row_offset = src_f_offset + row * hw;
for (size_t col = 0; col < kCubeSize; col++) {
auto src_ni = hfi * kCubeSize + col;
auto src_offset = src_row_offset + chw * col;
auto need_pad_zero = src_ni >= n || src_offset >= nchw || src_ci >= c;
auto idx = gfi * fractal_ele_cnt + col * c0 + row;
auto offset = idx * size;
auto protected_size = dst_size - offset < static_cast<size_t>(SECUREC_MEM_MAX_LEN)
? dst_size - offset
: static_cast<size_t>(SECUREC_MEM_MAX_LEN);
errno_t ret;
if (need_pad_zero) {
ret = memset_s(static_cast<uint8_t *>(result) + offset, protected_size, 0, size);
} else {
ret = memcpy_s(static_cast<uint8_t *>(result) + offset, protected_size,
static_cast<uint8_t const *>(args.data) + src_offset * size, size);
}
if (ret != 0) {
MS_LOG(ERROR) << "Failed to operate the dst memory error-code " << ret;
return false;
}
auto src_idx = src_row_offset + chw * col;
auto dst_idx = gfi * fractal_ele_cnt + col * c0 + row;
auto pad_zero = src_ni >= n || src_idx >= nchw || src_ci >= c;
SetDataBysize(size, pad_zero);
}
}
}
......@@ -573,17 +580,7 @@ bool FracZToNchw(const FormatArgs &args, void *result) {
size_t c0_idx = c_idx % c0;
size_t nc_idx = n_idx;
size_t src_idx = c1_idx * hwncc0 + h_idx * wncc0 + w_idx * ncc0 + nc_idx * c0 + c0_idx;
auto src_offset = src_idx * size;
auto dst_offset = dst_idx * size;
auto protected_size = total_size - dst_offset < static_cast<size_t>(SECUREC_MEM_MAX_LEN)
? total_size - dst_offset
: static_cast<size_t>(SECUREC_MEM_MAX_LEN);
auto ret = memcpy_s(static_cast<uint8_t *>(result) + dst_offset, protected_size,
static_cast<uint8_t const *>(args.data) + src_offset, size);
if (ret != EOK) {
MS_LOG(ERROR) << "Failed to operate the dst memory error-code " << ret;
return false;
}
SetDataBysize(size, 0);
}
}
}
......@@ -664,32 +661,16 @@ bool NchwToFracNz(const FormatArgs &args, void *result) {
auto h1h0_head = times_head + h1h0_idx * w0;
auto src_h_head = src_times_head + h1h0_idx * w;
for (size_t w1_idx = 0; w1_idx < num_w1; w1_idx++) {
size_t dst_offset = (h1h0_head + w1_idx * h1h0w0) * size;
size_t src_offset = (src_h_head + w1_idx * w0) * size;
auto protected_size = dst_size - dst_offset < static_cast<size_t>(SECUREC_MEM_MAX_LEN)
? dst_size - dst_offset
: static_cast<size_t>(SECUREC_MEM_MAX_LEN);
auto cp_ret = memcpy_s(static_cast<uint8_t *>(result) + dst_offset, protected_size,
static_cast<uint8_t const *>(args.data) + src_offset, size * w0);
if (cp_ret != EOK) {
MS_LOG(ERROR) << "Failed to operate the dst memory, error-code " << cp_ret;
return false;
}
size_t dst_idx = h1h0_head + w1_idx * h1h0w0;
size_t src_idx = src_h_head + w1_idx * w0;
SetDataBysize(size, 0);
}
auto w1_head = num_w1 * w0;
for (size_t w0_idx = 0; w1_head + w0_idx < w; w0_idx++) {
auto src_w_idx = w1_head + w0_idx;
size_t dst_offset = (h1h0_head + num_w1 * h1h0w0 + w0_idx) * size;
size_t src_offset = (src_h_head + src_w_idx) * size;
auto protected_size = dst_size - dst_offset < static_cast<size_t>(SECUREC_MEM_MAX_LEN)
? dst_size - dst_offset
: static_cast<size_t>(SECUREC_MEM_MAX_LEN);
auto cp_ret = memcpy_s(static_cast<uint8_t *>(result) + dst_offset, protected_size,
static_cast<uint8_t const *>(args.data) + src_offset, size);
if (cp_ret != EOK) {
MS_LOG(ERROR) << "Failed to operate the dst memory error-code " << cp_ret;
return false;
}
size_t dst_idx = h1h0_head + num_w1 * h1h0w0 + w0_idx;
size_t src_idx = src_h_head + src_w_idx;
SetDataBysize(size, 0);
}
}
}
......@@ -740,32 +721,16 @@ bool FracNzToNchw(const FormatArgs &args, void *result) {
auto h1h0_head = times_head + h1h0_idx * w0;
auto src_h_head = src_times_head + h1h0_idx * w;
for (size_t w1_idx = 0; w1_idx < num_w1; w1_idx++) {
size_t src_offset = (h1h0_head + w1_idx * h1h0w0) * size;
size_t dst_offset = (src_h_head + w1_idx * w0) * size;
auto protected_size = dst_size - dst_offset < static_cast<size_t>(SECUREC_MEM_MAX_LEN)
? dst_size - dst_offset
: static_cast<size_t>(SECUREC_MEM_MAX_LEN);
auto cp_ret = memcpy_s(static_cast<uint8_t *>(result) + dst_offset, protected_size,
static_cast<uint8_t const *>(args.data) + src_offset, size * w0);
if (cp_ret != EOK) {
MS_LOG(ERROR) << "Failed to operate the dst memory, error-code " << cp_ret;
return false;
}
size_t src_idx = h1h0_head + w1_idx * h1h0w0;
size_t dst_idx = src_h_head + w1_idx * w0;
SetDataBysize(size, 0);
}
auto w1_head = num_w1 * w0;
for (size_t w0_idx = 0; w1_head + w0_idx < w; w0_idx++) {
auto src_w_idx = w1_head + w0_idx;
size_t src_offset = (h1h0_head + num_w1 * h1h0w0 + w0_idx) * size;
size_t dst_offset = (src_h_head + src_w_idx) * size;
auto protected_size = dst_size - dst_offset < static_cast<size_t>(SECUREC_MEM_MAX_LEN)
? dst_size - dst_offset
: static_cast<size_t>(SECUREC_MEM_MAX_LEN);
auto cp_ret = memcpy_s(static_cast<uint8_t *>(result) + dst_offset, protected_size,
static_cast<uint8_t const *>(args.data) + src_offset, size);
if (cp_ret != EOK) {
MS_LOG(ERROR) << "Failed to operate the dst memory error-code " << cp_ret;
return false;
}
size_t src_idx = h1h0_head + num_w1 * h1h0w0 + w0_idx;
size_t dst_idx = src_h_head + src_w_idx;
SetDataBysize(size, 0);
}
}
}
......@@ -814,29 +779,11 @@ bool NchwToNc1hwc0(const FormatArgs &args, void *result) {
for (size_t w_idx = 0; w_idx < w; w_idx++) {
size_t w_head_addr = h_head_addr + w_idx * c0;
for (size_t c0_idx = 0; c0_idx < c0; c0_idx++) {
size_t dst_index = c0_idx + w_head_addr;
size_t dst_offset = dst_index * size;
auto protected_size = total_size - dst_offset < static_cast<size_t>(SECUREC_MEM_MAX_LEN)
? total_size - dst_offset
: static_cast<size_t>(SECUREC_MEM_MAX_LEN);
size_t dst_idx = c0_idx + w_head_addr;
size_t c_idx = c0_idx + c1_idx * c0;
size_t src_idx = n_idx * chw + c_idx * hw + h_idx * w + w_idx;
auto src_offset = src_idx * size;
if (c_idx < c) {
auto ret = memcpy_s(static_cast<uint8_t *>(result) + dst_offset, protected_size,
static_cast<uint8_t const *>(args.data) + src_offset, size);
if (ret != EOK) {
MS_LOG(ERROR) << "Failed to operate the dst memory error-code " << ret;
return false;
}
} else {
auto ret = memset_s(static_cast<uint8_t *>(result) + dst_offset, protected_size, 0, size);
if (ret != EOK) {
MS_LOG(ERROR) << "Failed to operate the dst memory error-code " << ret;
return false;
}
}
auto pad_zero = (c_idx < c) ? 0 : 1;
SetDataBysize(size, pad_zero);
}
}
}
......@@ -887,17 +834,7 @@ bool Nc1hwc0ToNchw(const FormatArgs &args, void *result) {
size_t c1_idx = c_idx / c0;
size_t c0_idx = c_idx % c0;
size_t src_idx = n_idx * c1hwc0 + c1_idx * hwc0 + h_idx * wc0 + w_idx * c0 + c0_idx;
auto src_offset = src_idx * size;
auto dst_offset = dst_idx * size;
auto protected_size = total_size - dst_offset < static_cast<size_t>(SECUREC_MEM_MAX_LEN)
? total_size - dst_offset
: static_cast<size_t>(SECUREC_MEM_MAX_LEN);
auto ret = memcpy_s(static_cast<uint8_t *>(result) + dst_offset, protected_size,
static_cast<uint8_t const *>(args.data) + src_offset, size);
if (ret != EOK) {
MS_LOG(ERROR) << "Failed to operate the dst memory error-code " << ret;
return false;
}
SetDataBysize(size, 0);
}
}
}
......@@ -922,31 +859,19 @@ bool NchwToC1hwncoc0(const FormatArgs &args, void *result) {
auto c1 = args.device_shape[0];
auto co = args.device_shape[4];
auto c0 = args.device_shape[5];
for (size_t c1_i = 0; c1_i < c1; c1_i++) {
for (size_t h_i = 0; h_i < h; h_i++) {
for (size_t w_i = 0; w_i < w; w_i++) {
for (size_t n_i = 0; n_i < n; n_i++) {
for (size_t co_i = 0; co_i < co; co_i++) {
for (size_t c0_i = 0; c0_i < c0; c0_i++) {
size_t dst_offset = (c1_i * h * w * n * co * c0 + h_i * w * n * co * c0 + w_i * n * co * c0 +
n_i * co * c0 + co_i * c0 + c0_i) *
size;
size_t protected_size = total_size - dst_offset < static_cast<size_t>(SECUREC_MEM_MAX_LEN)
? total_size - dst_offset
: static_cast<size_t>(SECUREC_MEM_MAX_LEN);
size_t dst_idx = c1_i * h * w * n * co * c0 + h_i * w * n * co * c0 + w_i * n * co * c0 + n_i * co * c0 +
co_i * c0 + c0_i;
size_t c_i = c0_i + c1_i * c0;
size_t src_offset = (n_i * c * h * w + c_i * h * w + h_i * w + w_i) * size;
errno_t ret;
if (c_i < c && c0_i == co_i) {
ret = memcpy_s(static_cast<uint8_t *>(result) + dst_offset, protected_size,
static_cast<uint8_t const *>(args.data) + src_offset, size);
} else {
ret = memset_s(static_cast<uint8_t *>(result) + dst_offset, protected_size, 0, size);
}
if (ret != EOK) {
MS_LOG(ERROR) << "Failed to operate the dst memory, error-code:" << ret;
return false;
}
size_t src_idx = n_i * c * h * w + c_i * h * w + h_i * w + w_i;
auto pad_zero = (c_i < c && c0_i == co_i) ? 0 : 1;
SetDataBysize(size, pad_zero);
}
}
}
......@@ -976,22 +901,13 @@ bool C1hwncoc0ToNchw(const FormatArgs &args, void *result) {
for (size_t c_i = 0; c_i < c; c_i++) {
for (size_t h_i = 0; h_i < h; h_i++) {
for (size_t w_i = 0; w_i < w; w_i++) {
size_t dst_offset = (n_i * c * h * w + c_i * h * w + h_i * w + w_i) * size;
size_t dst_idx = n_i * c * h * w + c_i * h * w + h_i * w + w_i;
size_t c1_i = c_i / kCubeSize;
size_t c0_i = c_i % kCubeSize;
size_t co_i = c0_i;
size_t src_offset = (c1_i * h * w * n * co * c0 + h_i * w * n * co * c0 + w_i * n * co * c0 + n_i * co * c0 +
co_i * c0 + c0_i) *
size;
size_t protected_size = total_size - dst_offset < static_cast<size_t>(SECUREC_MEM_MAX_LEN)
? total_size - dst_offset
: static_cast<size_t>(SECUREC_MEM_MAX_LEN);
auto ret = memcpy_s(static_cast<uint8_t *>(result) + dst_offset, protected_size,
static_cast<uint8_t const *>(args.data) + src_offset, size);
if (ret != EOK) {
MS_LOG(ERROR) << "Failed to operate the dst memory, error-code:" << ret;
return false;
}
size_t src_idx =
c1_i * h * w * n * co * c0 + h_i * w * n * co * c0 + w_i * n * co * c0 + n_i * co * c0 + co_i * c0 + c0_i;
SetDataBysize(size, 0);
}
}
}
......
Markdown is supported
0% 或 .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑！
想要评论请 注册 或