void THTensor_(copyAsyncCuda)(THCState *state, THTensor *self, struct THCTensor *src)
{
THArgCheck(THTensor_(nElement)(self) == THCTensor_(nElement)(state, src), 2, "sizes do not match");
THArgCheck(THTensor_(isContiguous)(self), 2, "Target tensor must be contiguous");
THArgCheck(THCTensor_(isContiguous)(state, src), 3, "Source tensor must be contiguous");
if (THTensor_(nElement)(self) == 0) return;
// Perform the copy wrt the current stream on the CudaTensor's device.
int tensorDevice = THCTensor_(getDevice)(state, src);
int currentDevice;
THCudaCheck(cudaGetDevice(¤tDevice));
if (currentDevice != tensorDevice) {
THCudaCheck(cudaSetDevice(tensorDevice));
}
THCStream *stream = THCState_getStream(state);
THCudaCheck(cudaMemcpyAsync(THTensor_(data)(self),
THCTensor_(data)(state, src),
THCTensor_(nElement)(state, src) * sizeof(real),
cudaMemcpyDeviceToHost,
stream->stream));
THCudaCheck(THCCachingHostAllocator_recordEvent(THCStorage_(data)(state, src->storage), stream));
if (currentDevice != tensorDevice) {
THCudaCheck(cudaSetDevice(currentDevice));
}
}
void THCStorage_(set)(THCState *state, THCStorage *self, ptrdiff_t index, scalar_t value)
{
THArgCheck((index >= 0) && (index < self->numel()), 2, "index out of bounds");
cudaStream_t stream = THCState_getCurrentStream(state);
THCudaCheck(cudaMemcpyAsync(THCStorage_(data)(state, self) + index, &value, sizeof(scalar_t),
cudaMemcpyHostToDevice,
stream));
THCudaCheck(cudaStreamSynchronize(stream));
}
scalar_t THCStorage_(get)(THCState *state, const THCStorage *self, ptrdiff_t index)
{
THArgCheck((index >= 0) && (index < self->numel()), 2, "index out of bounds");
scalar_t value;
cudaStream_t stream = THCState_getCurrentStream(state);
THCudaCheck(cudaMemcpyAsync(&value, THCStorage_(data)(state, self) + index, sizeof(scalar_t),
cudaMemcpyDeviceToHost, stream));
THCudaCheck(cudaStreamSynchronize(stream));
return value;
}
static int torch_Tensor_(storage)(lua_State *L)
{
THCTensor *tensor = luaT_checkudata(L, 1, torch_Tensor);
if(tensor->storage)
{
THCStorage_(retain)(cutorch_getstate(L), tensor->storage);
luaT_pushudata(L, tensor->storage, torch_Storage);
}
else
lua_pushnil(L);
return 1;
}
static PyObject * THPStorage_(pynew)(PyTypeObject *type, PyObject *args, PyObject *kwargs)
{
HANDLE_TH_ERRORS
Py_ssize_t num_args = args ? PyTuple_Size(args) : 0;
THPStoragePtr self((THPStorage *)type->tp_alloc(type, 0));
THPUtils_assert(self, "failed to allocate a " THPStorageStr " object");
THAllocator* allocator = NULL;
// Internally we allow constructing with a keywoard only argument cdata
if (kwargs != NULL) {
PyObject *allocator_ptr = PyDict_GetItemString(kwargs, "allocator");
if (allocator_ptr) {
THPUtils_assert(THPUtils_checkLong(allocator_ptr), "invalid allocator");
allocator = (THAllocator*) PyLong_AsVoidPtr(allocator_ptr);
PyDict_DelItemString(kwargs, "allocator");
}
Py_ssize_t num_kwargs = PyDict_Size(kwargs);
if (num_args == 0) {
PyObject *cdata_ptr = PyDict_GetItemString(kwargs, "cdata");
if (num_kwargs == 1 && cdata_ptr && THPUtils_checkLong(cdata_ptr)) {
THStorage *ptr = (THStorage*)PyLong_AsVoidPtr(cdata_ptr);
self->cdata = ptr;
return (PyObject*)self.release();
}
}
THPUtils_assert(num_kwargs == 0, THPStorageStr "(): invalid keyword arguments");
}
// torch.Storage()
if (num_args == 0) {
if (allocator) {
self->cdata = THPStorage_(newWithAllocator)(0, allocator);
} else {
self->cdata = THStorage_(new)(LIBRARY_STATE_NOARGS);
}
return (PyObject*)self.release();
}
PyObject *first_arg = PyTuple_GET_ITEM(args, 0);
// torch.Storage(size)
if (num_args == 1 && THPUtils_checkLong(first_arg)) {
int64_t size = THPUtils_unpackLong(first_arg);
if (allocator) {
self->cdata = THPStorage_(newWithAllocator)(size, allocator);
} else {
self->cdata = THStorage_(newWithSize)(LIBRARY_STATE size);
}
return (PyObject*)self.release();
}
// torch.Storage(view_source, [offset, [size]])
if (num_args < 4 && THPStorage_(Check)(first_arg)) {
#ifdef THD_GENERIC_FILE
THPUtils_setError("distributed storages don't support storage views");
return NULL;
#else
THPStorage *storage_arg = (THPStorage *)first_arg;
int64_t numel = storage_arg->cdata->size;
int64_t offset = 0;
if (num_args >= 2) {
PyObject *second_arg = PyTuple_GET_ITEM(args, 1);
if (!THPUtils_checkLong(second_arg))
goto invalid_arguments;
offset = THPUtils_unpackLong(second_arg);
}
int64_t size = numel - offset;
if (num_args >= 3) {
PyObject *third_arg = PyTuple_GET_ITEM(args, 2);
if (!THPUtils_checkLong(third_arg))
goto invalid_arguments;
size = THPUtils_unpackLong(third_arg);
}
THPUtils_assert(offset >= 0 && offset <= numel, "specified an offset of "
"%" PRId64 ", but the viewed storage has only %" PRId64 " element(s)", offset, numel);
THPUtils_assert(size >= 1 && size <= numel - offset, "specified a size of "
"%" PRId64 ", but the viewed storage has only %" PRId64 " element(s) after offset %" PRId64,
size, numel - offset, offset);
real *data_ptr = THStorage_(data)(LIBRARY_STATE storage_arg->cdata) + offset;
THStoragePtr storage(THStorage_(newWithData)(LIBRARY_STATE data_ptr, size));
storage->flag = TH_STORAGE_REFCOUNTED | TH_STORAGE_VIEW;
storage->view = storage_arg->cdata;
THStorage_(retain)(LIBRARY_STATE storage_arg->cdata);
self->cdata = storage.release();
return (PyObject*)self.release();
#endif
}
// torch.Storage(sequence)
if (num_args == 1 && PySequence_Check(first_arg)) {
#ifdef THD_GENERIC_FILE
THPUtils_setError("distributed storages don't support construction from a sequence");
#else
Py_ssize_t length = PySequence_Length(first_arg);
THPUtils_assert(length >= 0, "couldn't obtain the length of %s",
THPUtils_typename(first_arg));
self->cdata = THStorage_(newWithSize)(LIBRARY_STATE length);
THPObjectPtr item;
try {
for (Py_ssize_t i = 0; i < length; i++) {
item = PySequence_GetItem(first_arg, i);
real value = THPUtils_(unpackReal)(item.get());
#if !defined(THC_GENERIC_FILE)
self->cdata->unsafe_data<real>()[i] = value;
#else
// TODO: this might be slow - consider batched updates?
THCStorage_(set)(LIBRARY_STATE self->cdata, i, value);
#endif
}
} catch (std::runtime_error &e) {
THPUtils_setError("tried to construct a storage from a sequence (%s), "
"but one of the items was of type %s instead of %s",
THPUtils_typename(first_arg),
THPUtils_typename(item.get()),
THPUtils_typeTraits<real>::python_type_str);
return NULL;
}
return (PyObject*)self.release();
#endif
}
#ifndef THD_GENERIC_FILE
invalid_arguments:
#endif
THPUtils_invalidArguments(args, kwargs, THPStorageStr " constructor", 6,
"no arguments",
"(int size)",
"(Sequence data)",
"(" THPStorageStr " view_source)",
"(" THPStorageStr " view_source, int offset)",
"(" THPStorageStr " view_source, int offset, int size)");
return NULL;
END_HANDLE_TH_ERRORS
}
static int torch_Tensor_(elementSize)(lua_State *L)
{
lua_pushnumber(L, THCStorage_(elementSize)(cutorch_getstate(L)));
return 1;
}
static int cutorch_Storage_(copy)(lua_State *L)
{
THCState *state = cutorch_getstate(L);
THCStorage *storage = luaT_checkudata(L, 1, torch_Storage);
void *src;
if( (src = luaT_toudata(L, 2, "torch.CudaByteStorage")) )
THCStorage_(copyCudaByte)(state, storage, src);
else if( (src = luaT_toudata(L, 2, "torch.CudaCharStorage")) )
THCStorage_(copyCudaChar)(state, storage, src);
else if( (src = luaT_toudata(L, 2, "torch.CudaShortStorage")) )
THCStorage_(copyCudaShort)(state, storage, src);
else if( (src = luaT_toudata(L, 2, "torch.CudaIntStorage")) )
THCStorage_(copyCudaInt)(state, storage, src);
else if( (src = luaT_toudata(L, 2, "torch.CudaLongStorage")) )
THCStorage_(copyCudaLong)(state, storage, src);
else if( (src = luaT_toudata(L, 2, "torch.CudaStorage")) )
THCStorage_(copyCudaFloat)(state, storage, src);
else if( (src = luaT_toudata(L, 2, "torch.CudaDoubleStorage")) )
THCStorage_(copyCudaDouble)(state, storage, src);
#if CUDA_VERSION >= 7050
else if( (src = luaT_toudata(L, 2, "torch.CudaHalfStorage")) )
THCStorage_(copyCudaHalf)(state, storage, src);
#endif
else if( (src = luaT_toudata(L, 2, "torch.ByteStorage")) )
THCStorage_(copyByte)(state, storage, src);
else if( (src = luaT_toudata(L, 2, "torch.CharStorage")) )
THCStorage_(copyChar)(state, storage, src);
else if( (src = luaT_toudata(L, 2, "torch.ShortStorage")) )
THCStorage_(copyShort)(state, storage, src);
else if( (src = luaT_toudata(L, 2, "torch.IntStorage")) )
THCStorage_(copyInt)(state, storage, src);
else if( (src = luaT_toudata(L, 2, "torch.LongStorage")) )
THCStorage_(copyLong)(state, storage, src);
else if( (src = luaT_toudata(L, 2, "torch.FloatStorage")) )
THCStorage_(copyFloat)(state, storage, src);
else if( (src = luaT_toudata(L, 2, "torch.DoubleStorage")) )
THCStorage_(copyDouble)(state, storage, src);
else
luaL_typerror(L, 2, "torch.*Storage");
lua_settop(L, 1);
return 1;
}
