mirror of
https://github.com/zebrajr/pytorch.git
synced 2025-12-06 12:20:52 +01:00
889e3eeed3
Also partially fixes #122109 This PR: - We add a C++ flag (only_lift_cpu_tensors) to toggle the torch.tensor(1, device='cuda') ctor strategy. When false (default), it does the current PyTorch behavior of unconditionally constructing a concrete CUDA tensor then calling lift_fresh on it. When true, we instead construct a concrete CPU tensor, call lift_fresh, and then call Tensor.to(device) (under any ambient modes). - FakeTensorMode flips this flag depending on if CUDA is available or not. We don't unconditionally set the flag to True because that is likely BC-breaking. Test Plan: - existing tests Pull Request resolved: https://github.com/pytorch/pytorch/pull/124413 Approved by: https://github.com/eellison
63 lines
1.8 KiB
C++
63 lines
1.8 KiB
C++
#include <c10/core/impl/TorchDispatchModeTLS.h>
|
|
#include <torch/csrc/utils/device_lazy_init.h>
|
|
|
|
#include <torch/csrc/Exceptions.h>
|
|
#include <torch/csrc/python_headers.h>
|
|
#include <torch/csrc/utils/object_ptr.h>
|
|
#include <iostream>
|
|
namespace torch::utils {
|
|
namespace {
|
|
|
|
std::array<bool, at::COMPILE_TIME_MAX_DEVICE_TYPES> is_initialized{};
|
|
|
|
} // anonymous namespace
|
|
|
|
bool is_device_initialized(at::DeviceType device_type) {
|
|
pybind11::gil_scoped_acquire g;
|
|
return is_initialized[static_cast<int>(device_type)];
|
|
}
|
|
|
|
void device_lazy_init(at::DeviceType device_type) {
|
|
pybind11::gil_scoped_acquire g;
|
|
// Protected by the GIL. We don't use call_once because under ASAN it
|
|
// has a buggy implementation that deadlocks if an instance throws an
|
|
// exception. In any case, call_once isn't necessary, because we
|
|
// have taken a lock.
|
|
if (is_device_initialized(device_type)) {
|
|
return;
|
|
}
|
|
|
|
auto maybe_mode = c10::impl::TorchDispatchModeTLS::get_mode(
|
|
c10::impl::TorchDispatchModeKey::FAKE);
|
|
if (maybe_mode) {
|
|
return;
|
|
}
|
|
|
|
std::string module_name = "torch." + at::DeviceTypeName(device_type, true);
|
|
auto module = THPObjectPtr(PyImport_ImportModule(module_name.c_str()));
|
|
if (!module) {
|
|
throw python_error();
|
|
}
|
|
|
|
if (device_type == at::DeviceType::PrivateUse1) {
|
|
auto has_lazy_init_method =
|
|
PyObject_HasAttrString(module.get(), "_lazy_init") == 1;
|
|
if (!has_lazy_init_method) {
|
|
return;
|
|
}
|
|
}
|
|
|
|
auto res = THPObjectPtr(PyObject_CallMethod(module.get(), "_lazy_init", ""));
|
|
if (!res) {
|
|
throw python_error();
|
|
}
|
|
|
|
is_initialized[static_cast<int>(device_type)] = true;
|
|
}
|
|
|
|
void set_requires_device_init(at::DeviceType device_type, bool value) {
|
|
is_initialized[static_cast<int>(device_type)] = !value;
|
|
}
|
|
|
|
} // namespace torch::utils
|