m53297601
/
Paddle
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

fix errors

Browse Source
release/0.12.0
fengjiayi 7 years ago
parent 9f11da5931
commit c5e178f4fd
3 changed files with 9 additions and 9 deletions
Show Stats Download Patch File Download Diff File

3
paddle/fluid/framework/tensor_util.cc
Unescape View File

@ -20,8 +20,7 @@ namespace paddle {
namespace framework { namespace framework {
void TensorCopy(const Tensor& src, const platform::Place& dst_place, void TensorCopy(const Tensor& src, const platform::Place& dst_place,
const platform::DeviceContext& ctx, Tensor* dst, const platform::DeviceContext& ctx, Tensor* dst, bool sync) {
bool sync = false) {
VLOG(3) << "TensorCopy " << src.dims() << " from " << src.place() << " to " VLOG(3) << "TensorCopy " << src.dims() << " from " << src.place() << " to "
<< dst_place; << dst_place;
src.check_memory_size(); src.check_memory_size();

3
paddle/fluid/framework/tensor_util.h
Unescape View File

@ -24,7 +24,8 @@ namespace paddle {
namespace framework { namespace framework {
void TensorCopy(const Tensor& src, const platform::Place& dst_place, void TensorCopy(const Tensor& src, const platform::Place& dst_place,
const platform::DeviceContext& ctx, Tensor* dst); const platform::DeviceContext& ctx, Tensor* dst,
bool sync = false);
void TensorCopy(const Tensor& src, const platform::Place& dst_place, void TensorCopy(const Tensor& src, const platform::Place& dst_place,
Tensor* dst); Tensor* dst);

12
paddle/fluid/memory/memcpy.cc
Unescape View File

@ -30,7 +30,7 @@ void Copy<platform::CPUPlace, platform::CPUPlace>(platform::CPUPlace, void* dst,
template <> template <>
void Copy<platform::CPUPlace, platform::CUDAPlace>( void Copy<platform::CPUPlace, platform::CUDAPlace>(
platform::CPUPlace dst_place, void* dst, platform::CUDAPlace src_place, platform::CPUPlace dst_place, void* dst, platform::CUDAPlace src_place,
const void* src, size_t num, cudaStream_t stream = nullptr) { const void* src, size_t num, cudaStream_t stream) {
platform::SetDeviceId(src_place.device); platform::SetDeviceId(src_place.device);
if (stream) { if (stream) {
platform::GpuMemcpyAsync(dst, src, num, cudaMemcpyDeviceToHost, stream); platform::GpuMemcpyAsync(dst, src, num, cudaMemcpyDeviceToHost, stream);
@ -42,7 +42,7 @@ void Copy<platform::CPUPlace, platform::CUDAPlace>(
template <> template <>
void Copy<platform::CUDAPlace, platform::CPUPlace>( void Copy<platform::CUDAPlace, platform::CPUPlace>(
platform::CUDAPlace dst_place, void* dst, platform::CPUPlace src_place, platform::CUDAPlace dst_place, void* dst, platform::CPUPlace src_place,
const void* src, size_t num, cudaStream_t stream = nullptr) { const void* src, size_t num, cudaStream_t stream) {
platform::SetDeviceId(dst_place.device); platform::SetDeviceId(dst_place.device);
if (stream) { if (stream) {
platform::GpuMemcpyAsync(dst, src, num, cudaMemcpyHostToDevice, stream); platform::GpuMemcpyAsync(dst, src, num, cudaMemcpyHostToDevice, stream);
@ -54,7 +54,7 @@ void Copy<platform::CUDAPlace, platform::CPUPlace>(
template <> template <>
void Copy<platform::CUDAPlace, platform::CUDAPlace>( void Copy<platform::CUDAPlace, platform::CUDAPlace>(
platform::CUDAPlace dst_place, void* dst, platform::CUDAPlace src_place, platform::CUDAPlace dst_place, void* dst, platform::CUDAPlace src_place,
const void* src, size_t num, cudaStream_t stream = nullptr) { const void* src, size_t num, cudaStream_t stream) {
if (dst_place == src_place) { if (dst_place == src_place) {
platform::SetDeviceId(src_place.device); platform::SetDeviceId(src_place.device);
if (stream) { if (stream) {
@ -68,7 +68,7 @@ void Copy<platform::CUDAPlace, platform::CUDAPlace>(
num, stream); num, stream);
} else { } else {
platform::GpuMemcpyPeerSync(dst, dst_place.device, src, src_place.device, platform::GpuMemcpyPeerSync(dst, dst_place.device, src, src_place.device,
num, stream); num);
} }
} }
} }
@ -98,7 +98,7 @@ template <>
void Copy<platform::CUDAPinnedPlace, platform::CUDAPlace>( void Copy<platform::CUDAPinnedPlace, platform::CUDAPlace>(
platform::CUDAPinnedPlace dst_place, void* dst, platform::CUDAPinnedPlace dst_place, void* dst,
platform::CUDAPlace src_place, const void* src, size_t num, platform::CUDAPlace src_place, const void* src, size_t num,
cudaStream_t stream = nullptr) { cudaStream_t stream) {
platform::SetDeviceId(src_place.device); platform::SetDeviceId(src_place.device);
if (stream) { if (stream) {
platform::GpuMemcpyAsync(dst, src, num, cudaMemcpyDeviceToHost, stream); platform::GpuMemcpyAsync(dst, src, num, cudaMemcpyDeviceToHost, stream);
@ -111,7 +111,7 @@ template <>
void Copy<platform::CUDAPlace, platform::CUDAPinnedPlace>( void Copy<platform::CUDAPlace, platform::CUDAPinnedPlace>(
platform::CUDAPlace dst_place, void* dst, platform::CUDAPlace dst_place, void* dst,
platform::CUDAPinnedPlace src_place, const void* src, size_t num, platform::CUDAPinnedPlace src_place, const void* src, size_t num,
cudaStream_t stream = nullptr) { cudaStream_t stream) {
platform::SetDeviceId(dst_place.device); platform::SetDeviceId(dst_place.device);
if (stream) { if (stream) {
platform::GpuMemcpyAsync(dst, src, num, cudaMemcpyHostToDevice, stream); platform::GpuMemcpyAsync(dst, src, num, cudaMemcpyHostToDevice, stream);

Write Preview
Loading…
Cancel
Save