Commit Graph

9 Commits

Author SHA1 Message Date
Michael Lazos
0d44a8aea1 [Hierarchical Compile] Apply deduplication after output node creation (#150306)
Pull Request resolved: https://github.com/pytorch/pytorch/pull/150306
Approved by: https://github.com/anijain2305
ghstack dependencies: #150303, #150304, #150305
2025-04-01 20:54:18 +00:00
Michael Lazos
8740ffa760 [Hierarchical Compile] Add cycle detection to graph region expansion (#150305)
Pull Request resolved: https://github.com/pytorch/pytorch/pull/150305
Approved by: https://github.com/anijain2305
ghstack dependencies: #150303, #150304
2025-04-01 20:54:18 +00:00
Michael Lazos
d2c0c65ea1 [Dynamo] Add debug linting option for graph dedupe (#150053)
As title

Pull Request resolved: https://github.com/pytorch/pytorch/pull/150053
Approved by: https://github.com/StrongerXi, https://github.com/anijain2305
2025-03-28 14:27:09 +00:00
Raymond Li
21c2565f35 Document dynamo (#146736)
Many files in dynamo are currently lacking file/module-level documentation, which makes it hard to know what they do at a glance and without digging into the code. This fixes that.

Note: documentation was AI-generated and could be incorrect, please review carefully.

Pull Request resolved: https://github.com/pytorch/pytorch/pull/146736
Approved by: https://github.com/jansel, https://github.com/StrongerXi, https://github.com/anijain2305, https://github.com/zou3519
2025-02-13 00:02:21 +00:00
Aaron Orenstein
a79100ab11 PEP585 update - torch/_dynamo (#145105)
See #145101 for details.

Pull Request resolved: https://github.com/pytorch/pytorch/pull/145105
Approved by: https://github.com/bobrenjc93
2025-01-18 20:47:11 +00:00
bobrenjc93
1fe3af2c68 Migrate from Tuple -> tuple in torch/_dynamo (#144261)
Pull Request resolved: https://github.com/pytorch/pytorch/pull/144261
Approved by: https://github.com/aorenste, https://github.com/zou3519
2025-01-10 07:45:57 +00:00
Michael Lazos
fd23cf5848 [Dynamo] check node class first for graph dedup (#143609)
as title

Pull Request resolved: https://github.com/pytorch/pytorch/pull/143609
Approved by: https://github.com/williamwen42
2024-12-20 04:09:46 +00:00
Michael Lazos
4eafbe5288 [Dynamo] Flatten slices during graph deduplication (#143522)
I encountered this issue while debugging torchtune - overall we need to make sure to not miss nodes that are slice arguments.

Pull Request resolved: https://github.com/pytorch/pytorch/pull/143522
Approved by: https://github.com/williamwen42
2024-12-18 23:12:34 +00:00
Michael Lazos
c31543c7ae [Dynamo] Initial deduplication pass impl (#141383)
This PR implements the deduplication pass (blocked by config currently) for dynamo where identical regions from https://github.com/pytorch/pytorch/pull/141381 are replaced with a common subgraph.

The two phases of deduplication are explained below.

**Subgraph creation**:
Subgraph creation works by taking one representative region from each region group and creating a subgraph from it, which will then be used to replace all regions in the group. This is implemented by first copying all nodes of the region to the new subgraph and then finding all inputs which are not within the region and creating placeholders for them. For the outputs, all regions in a region group need to be scanned to ensure the largest set of outputs is found, and then an output node is created which returns a tuple of all outputs.

**Graph replacement**:
To replace each region with the extracted subgraph, the node index in the region and argument index within the node's flattened args and kwargs are recorded once during subgraph creation. This allows us to determine which (external to the region) nodes and in which order these nodes are passed as inputs. For the outputs, getitem nodes are created for each output, and all nodes in the region with external outputs are replaced by the proper getitem node. Finally, all original nodes are erased (there should be no uses of these left in the graph).

Pull Request resolved: https://github.com/pytorch/pytorch/pull/141383
Approved by: https://github.com/zou3519
ghstack dependencies: #141381, #141382
2024-12-11 02:22:21 +00:00