Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Fix slr deser #9409

Merged
merged 3 commits into from
Mar 28, 2018
Merged

Fix slr deser #9409

Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
5877811
fix slr deser
typhoonzero Mar 27, 2018
094d509
fix slr deser
typhoonzero Mar 27, 2018
cc1c6af
fix slr serde
Mar 27, 2018
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
24 changes: 18 additions & 6 deletions paddle/fluid/operators/detail/variable_response.cc
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -48,6 +48,8 @@ bool ReadRaw(::google::protobuf::io::CodedInputStream* input,
void* dest, int size) {
const void* data = NULL;
int size_to_write = 0;
int length = size;
int total_written = 0;

if (platform::is_gpu_place(place)) {
#ifdef PADDLE_WITH_CUDA
Expand All @@ -56,16 +58,21 @@ bool ReadRaw(::google::protobuf::io::CodedInputStream* input,
platform::CPUPlace cpu;

char* p = reinterpret_cast<char*>(dest);
while (size > 0) {
while (total_written < length) {
if (!input->GetDirectBufferPointer(&data, &size_to_write)) {
return false;
}

// NOTE: if raw buffer is large and have two neighbor fields of raw
// buffers GetDirectBufferPointer can get all of them, use length to
// truncate it.
if (total_written + size_to_write > length) {
size_to_write = length - total_written;
}
memory::Copy(boost::get<platform::CUDAPlace>(place),
reinterpret_cast<void*>(p), cpu, data, size_to_write,
gpu_dev_ctx.stream());
p += size_to_write;
size -= size_to_write;
total_written += size_to_write;

input->Skip(size_to_write);
}
Expand All @@ -77,16 +84,21 @@ bool ReadRaw(::google::protobuf::io::CodedInputStream* input,
}

char* p = reinterpret_cast<char*>(dest);
while (size > 0) {
while (total_written < length) {
if (!input->GetDirectBufferPointer(&data, &size_to_write)) {
return false;
}
// NOTE: if raw buffer is large and have two neighbor fields of raw buffers
// GetDirectBufferPointer can get all of them, use length to truncate it.
if (total_written + size_to_write > length) {
size_to_write = length - total_written;
}
// TODO(gongwb): can we avoid copy?
platform::CPUPlace cpu;
memory::Copy(cpu, reinterpret_cast<void*>(p), cpu, data, size_to_write);

p += size_to_write;
size -= size_to_write;
total_written += size_to_write;

input->Skip(size_to_write);
}
Expand Down Expand Up @@ -153,6 +165,7 @@ bool VariableResponse::CopySelectRowsData(
const platform::DeviceContext& ctx, int length) {
auto var = scope_->FindVar(meta_.varname());
auto* slr = var->GetMutable<framework::SelectedRows>();
slr->mutable_rows()->resize(length / 8); // int64
int64_t* rows_data = slr->mutable_rows()->data();

// copy rows CPU data, GPU data will be copied lazily.
Expand Down Expand Up @@ -233,7 +246,6 @@ int VariableResponse::Parse(Source* source) {
if (tag != 0) {
return -1;
}

return 0;
}

Expand Down