mirror of
https://github.com/zebrajr/pytorch.git
synced 2025-12-07 12:21:27 +01:00
d4e05f4e1e
Summary: This eliminates the need for any heuristics regarding stack size limits. Pull Request resolved: https://github.com/pytorch/pytorch/pull/11534 Differential Revision: D9779866 Pulled By: resistor fbshipit-source-id: 96753eead7904bbdc2869fb01f7bd42141032347
90 lines
2.8 KiB
C++
90 lines
2.8 KiB
C++
#include "torch/csrc/autograd/function.h"
|
|
|
|
#include "torch/csrc/autograd/engine.h"
|
|
#include "torch/csrc/autograd/variable.h"
|
|
#include "torch/csrc/jit/ir.h"
|
|
|
|
#include <ATen/ATen.h>
|
|
|
|
#include <algorithm>
|
|
#include <cstdint>
|
|
#include <memory>
|
|
#include <stdexcept>
|
|
#include <string>
|
|
#include <utility>
|
|
#include <vector>
|
|
#include <deque>
|
|
|
|
namespace torch { namespace autograd {
|
|
|
|
/// Monotonically incrementing (thread local!) counter to supply sequence
|
|
/// numbers.
|
|
thread_local uint64_t Function_next_sequence_nr_ = 0;
|
|
|
|
uint64_t& Function::get_next_sequence_nr() {
|
|
return Function_next_sequence_nr_;
|
|
}
|
|
|
|
auto Function::name() const -> std::string {
|
|
return at::demangle(typeid(*this).name());
|
|
}
|
|
|
|
AnomalyMetadata* Function::metadata() noexcept {
|
|
if (!anomaly_metadata_) {
|
|
anomaly_metadata_ = Engine::get_default_engine().make_anomaly_metadata();
|
|
}
|
|
return anomaly_metadata_.get();
|
|
}
|
|
|
|
static void gatherFunctions(Function* func,
|
|
std::vector<std::shared_ptr<Function>>& stack) {
|
|
for (auto& edge : func->next_edges()) {
|
|
if (edge.function.use_count() == 1) {
|
|
stack.emplace_back(std::move(edge.function));
|
|
}
|
|
}
|
|
}
|
|
|
|
/*
|
|
* Fix for #5534: prevent stack overflow on deletion of deep computation graph
|
|
*
|
|
* Sometimes one can end up with a very big computation graph of Functions
|
|
* and Edges. Each std::shared_ptr<Function> contains a list of Edge, and
|
|
* each Edge contains a std::shared_ptr<Function>. Deleting a
|
|
* std::shared_ptr<Function> can trigger the recursive deletion of other
|
|
* std::shared_ptr<Function>'s: this can stack overflow if the graph
|
|
* is deep enough. Here is an example of such a graph:
|
|
*
|
|
* shared_ptr<Function> -> Edge -> shared_ptr<Function> -> Edge -> ... -> shared_ptr<Function>
|
|
*
|
|
* The solution here is to detect when we are decrementing away the last
|
|
* reference to a Function, and when doing so to buffer up the Function's
|
|
* that will be recursively decremented. We can then decrement (and free)
|
|
* the original Function without causing a recursive cascade, before
|
|
* draining the buffer applying the same behavior. This is, in effect,
|
|
* converting recursion to a loop, using a heap buffer in place of the
|
|
* recursive call stack.
|
|
*/
|
|
void deleteFunction(Function* function) {
|
|
// To avoid stack overflow on large computational graphs,
|
|
// we need to track reference decrementing and freeing
|
|
// on the heap.
|
|
std::vector<std::shared_ptr<Function>> stack;
|
|
gatherFunctions(function, stack);
|
|
delete function;
|
|
|
|
while (!stack.empty()) {
|
|
auto& curr_func = stack.back();
|
|
|
|
if (curr_func.use_count() == 1) {
|
|
// If this is the last reference, gather function references
|
|
// that will be recursively decremented.
|
|
gatherFunctions(curr_func.get(), stack);
|
|
}
|
|
|
|
stack.pop_back();
|
|
}
|
|
}
|
|
|
|
}} // namespace torch::autograd
|