Repository navigation
[executorch][native-ggml] Replay ggml graphs for repeated input shapes - #23836
Merged
Merged
Conversation
Pull Request resolved: #23761 Every execute on the ggml engine rebuilt, scheduled and allocated its graph, even when the input shapes matched the previous call, as in LLM decode. The ggml engine keeps a configurable number of built graphs per executable, keyed by input shapes, and replays them with fresh input bytes. Row indices that feed gathers and cache writes are rechecked on every replay. Authored with Claude and Codex assistance. ghstack-source-id: 447877040 @exported-using-ghexport Differential Revision: [D123796928](https://our.internmc.facebook.com/intern/diff/D123796928/)
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/23836
Note: Links to docs will display an error until the docs builds have been completed. ⏳ No Failures, 146 PendingAs of commit 0fdcd97 with merge base 0a4a05e ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
meta-cla
Bot
added
the
CLA Signed
digantdesai
added this pull request to stack #23862
October 14, 2026 18:20
digantdesai
approved these changes
Oct 14, 2026
This branch was successfully deployed
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
This PR was created by the merge bot to help merge the original PR into the main branch.
ghstack PR number: #23761 by @digantdesai
^ Please use this as the source of truth for the PR details, comments, and reviews
ghstack PR base: https://github.com/pytorch/executorch/tree/gh/digantdesai/152/base
ghstack PR head: https://github.com/pytorch/executorch/tree/gh/digantdesai/152/head
Merge bot PR base: https://github.com/pytorch/executorch/tree/gh/digantdesai/151/orig
Merge bot PR head: https://github.com/pytorch/executorch/tree/gh/digantdesai/152/orig
Differential Revision: D123796928
@diff-train-skip-merge