FC/BC policy for libtorch stable ABI #163991
Conversation
[ghstack-poisoned]
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/163991
Note: Links to docs will display an error until the docs builds have been completed. ❗ 1 Active SEVsThere are 1 currently active SEVs. If your PR is affected, please view them below: ⏳ 2 Pending, 1 Unrelated FailureAs of commit a704132 with merge base e901866 ( FLAKY - The following job failed but was likely due to flakiness present on trunk:
UNSTABLE - The following jobs are marked as unstable, possibly due to flakiness on trunk:
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
[ghstack-poisoned]
| [MAJ ][ MIN ][PATCH ][ ABI TAG ] | ||
| ``` | ||
|
|
||
| In the present phase of development, APIs in the C-shim will be versioned based on major.minor.patch release that they are first introduced in, with 2.10 being the first release where this will be enforced. The ABI tag is reserved for future use. |
There was a problem hiding this comment.
| In the present phase of development, APIs in the C-shim will be versioned based on major.minor.patch release that they are first introduced in, with 2.10 being the first release where this will be enforced. The ABI tag is reserved for future use. | |
| In the present phase of development, APIs in the C-shim will be versioned based on the major.minor.patch release that they are first introduced in, with 2.10 being the first release where this will be enforced. The ABI tag is reserved for internal use. |
no promises
There was a problem hiding this comment.
I think this might indeed by future rather than internal (I'm thinking we will use it if we want to allow users to depend on nightlies and bump the version more frequently than each release when the ABI becomes more mature).
[ghstack-poisoned]
janeyx99
left a comment
janeyx99
left a comment
There was a problem hiding this comment.
For aoti_torch_call_dispatcher_v2, I'm okay with keeping the docs here still but warning that we don't make any guarantees on the underlying ops changing.
[ghstack-poisoned]
pytorch-bot
bot
added
ciflow/b200
ciflow/h100
ciflow/rocm
|
@pytorchbot merge |
|
This PR updates submodules third_party/fbgemm, third_party/kleidiai, third_party/cudnn_frontend, third_party/kineto If those updates are intentional, please add "submodule" keyword to PR title/description. |
[ghstack-poisoned]
|
@pytorchbot merge |
Merge startedYour change will be merged once all checks pass (ETA 0-4 Hours). Learn more about merging in the wiki. Questions? Feedback? Please reach out to the PyTorch DevX Team |
Merge failedReason: 1 jobs have failed, first few of them are: trunk / libtorch-linux-jammy-cuda12.8-py3.10-gcc11-debug / build Details for Dev Infra teamRaised by workflow job |
|
Build failure is unrelated to docs change |
|
@pytorchbot merge -i |
Merge startedYour change will be merged while ignoring the following 2 checks: pull / linux-jammy-py3.10-gcc11 / test (distributed, 2, 2, linux.2xlarge), trunk / libtorch-linux-jammy-cuda12.8-py3.10-gcc11-debug / build Learn more about merging in the wiki. Questions? Feedback? Please reach out to the PyTorch DevX Team |
Pull Request resolved: #163991 Approved by: https://github.com/janeyx99 ghstack dependencies: #163899
Stack from ghstack (oldest at bottom):