# Reverse Fusion of Node Pairs in Scheduler

**URL:** https://dev-discuss.pytorch.org/t/reverse-fusion-of-node-pairs-in-scheduler/2153
**Category:** compiler
**Created:** [June 14, 2024, 10:15am UTC](https://dev-discuss.pytorch.org/t/reverse-fusion-of-node-pairs-in-scheduler/2153 "2024-06-14T10:15:59Z")
**Posts on this page:** 1
**Page:** 1

<div class="post-metadata">

### Author: ![NiklasZ](https://avatars.discourse-cdn.com/v4/letter/n/8e8cbc/32.png) [@NiklasZ](https://dev-discuss.pytorch.org/u/NiklasZ)
#### Post date: [June 14, 2024, 10:15am UTC](https://dev-discuss.pytorch.org/t/reverse-fusion-of-node-pairs-in-scheduler/2153/1 "2024-06-14T10:15:59Z")

</div>

Hi,

I’m interested in PyTorch’s operator fusion. I have been reading the Scheduler’s [source code](https://github.com/pytorch/pytorch/blob/main/torch/_inductor/scheduler.py) and noticed a strange case where it seems to consider the fusability of a pair of nodes `node1, node2` in the normal order: `self.can_fuse(node1, node2)` but also the reverse order: `self.can_fuse(node2, node1)` . Specifically, I’m referring to this code here:

```python
def check_all_pairs(nodes):
    for node1_index, node1 in enumerate(nodes):
        for node2 in nodes[node1_index + 1 :]:
            key = (node1, node2)
            if key in seen:
                continue
            seen.add(key)

            if self.can_fuse(node1, node2):
                possible_fusions.append(key)
            elif (node2.is_template() or node2.is_foreach()) and self.can_fuse(
                node2, node1
            ):
                # foreach fusions and epilogue fusions are order dependent
                possible_fusions.append((node2, node1))

```

I inspected this function on some trivial functions via `torch.compile()` and it seems that `nodes` is typically a set containing a graph node `n` and all the nodes it outputs to `m_1,...m_k` (it’s probably not that simple, so feel free to correct me there). So we have a set (`n`, `m_1`, … `m_k`). So in the `self.can_fuse(node1, node2)` case we have:

1. Check whether we can fuse `n` with some `m_i+1` (propbably vertical fusion)
2. Check whether we can fuse `m_i` with `m_i+1` (probably horizontal fusion)  
for some `i` in `1...k`.

Now my question is, when is `self.can_fuse(node2, node1)` meaningful? Is this specifically for cases where the order of operations doesn’t affect the correctness? Although in that case `self.can_fuse(node1, node2)` should also be sufficient.

Would appreciate someone’s insight here. Many thanks!

FYI: this is a re-post of [this question](https://discuss.pytorch.org/t/reverse-fusion-of-node-pairs-in-scheduler/202350) on the pytorch forum. I’m hoping that’s ok given it didn’t see any response there.