Skip to content

Conversation

@hsharma35
Copy link
Contributor

Summary:
Add idma copy and wait op definitions + fake implementation. These can be used for asynchronous copy/wait.

idma_load: Used for Any -> DTCM data transfers. (Note: DTCM to DTCM is ok)
idma_store: Used for DTCM -> Any (similar to idma_load).

Since the above ops explicitly specify the direction of data transfer, we can use these to plan memory in specific dtcm banks.

Reviewed By: nitish2112

Differential Revision: D77164673

@pytorch-bot
Copy link

pytorch-bot bot commented Jun 24, 2025

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/11867

Note: Links to docs will display an error until the docs builds have been completed.

✅ No Failures

As of commit 423f218 with merge base a627be6 (image):
💚 Looks good so far! There are no failures yet. 💚

This comment was automatically generated by Dr. CI and updates every 15 minutes.

@facebook-github-bot facebook-github-bot added the CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. label Jun 24, 2025
@facebook-github-bot
Copy link
Contributor

This pull request was exported from Phabricator. Differential Revision: D77164673

hsharma35 added a commit to hsharma35/executorch that referenced this pull request Jun 24, 2025
Summary:

Add idma copy and wait op definitions + fake implementation. These can be used for asynchronous copy/wait.

idma_load: Used for `Any -> DTCM` data transfers. (Note: DTCM to DTCM is ok)
idma_store: Used for `DTCM -> Any` (similar to idma_load).

Since the above ops explicitly specify the direction of data transfer, we can use these to plan memory in specific dtcm banks.

Reviewed By: nitish2112

Differential Revision: D77164673
@facebook-github-bot
Copy link
Contributor

This pull request was exported from Phabricator. Differential Revision: D77164673

hsharma35 added a commit to hsharma35/executorch that referenced this pull request Jun 24, 2025
Summary:
Pull Request resolved: pytorch#11867

Add idma copy and wait op definitions + fake implementation. These can be used for asynchronous copy/wait.

idma_load: Used for `Any -> DTCM` data transfers. (Note: DTCM to DTCM is ok)
idma_store: Used for `DTCM -> Any` (similar to idma_load).

Since the above ops explicitly specify the direction of data transfer, we can use these to plan memory in specific dtcm banks.

Reviewed By: nitish2112

Differential Revision: D77164673
@hsharma35 hsharma35 added the release notes: none Do not include this in the release notes label Jun 24, 2025
hsharma35 added a commit to hsharma35/executorch that referenced this pull request Jun 24, 2025
Summary:

Add idma copy and wait op definitions + fake implementation. These can be used for asynchronous copy/wait.

idma_load: Used for `Any -> DTCM` data transfers. (Note: DTCM to DTCM is ok)
idma_store: Used for `DTCM -> Any` (similar to idma_load).

Since the above ops explicitly specify the direction of data transfer, we can use these to plan memory in specific dtcm banks.

Reviewed By: nitish2112

Differential Revision: D77164673
@facebook-github-bot
Copy link
Contributor

This pull request was exported from Phabricator. Differential Revision: D77164673

hsharma35 added a commit to hsharma35/executorch that referenced this pull request Jun 24, 2025
Summary:

Add idma copy and wait op definitions + fake implementation. These can be used for asynchronous copy/wait.

idma_load: Used for `Any -> DTCM` data transfers. (Note: DTCM to DTCM is ok)
idma_store: Used for `DTCM -> Any` (similar to idma_load).

Since the above ops explicitly specify the direction of data transfer, we can use these to plan memory in specific dtcm banks.

Reviewed By: zonglinpeng, nitish2112

Differential Revision: D77164673
@facebook-github-bot
Copy link
Contributor

This pull request was exported from Phabricator. Differential Revision: D77164673

hsharma35 added a commit to hsharma35/executorch that referenced this pull request Jun 24, 2025
Summary:

Add idma copy and wait op definitions + fake implementation. These can be used for asynchronous copy/wait.

idma_load: Used for `Any -> DTCM` data transfers. (Note: DTCM to DTCM is ok)
idma_store: Used for `DTCM -> Any` (similar to idma_load).

Since the above ops explicitly specify the direction of data transfer, we can use these to plan memory in specific dtcm banks.

Reviewed By: zonglinpeng, nitish2112

Differential Revision: D77164673
@facebook-github-bot
Copy link
Contributor

This pull request was exported from Phabricator. Differential Revision: D77164673

hsharma35 added a commit to hsharma35/executorch that referenced this pull request Jun 25, 2025
Summary:

Add idma copy and wait op definitions + fake implementation. These can be used for asynchronous copy/wait.

idma_load: Used for `Any -> DTCM` data transfers. (Note: DTCM to DTCM is ok)
idma_store: Used for `DTCM -> Any` (similar to idma_load).

Since the above ops explicitly specify the direction of data transfer, we can use these to plan memory in specific dtcm banks.

Reviewed By: zonglinpeng, nitish2112

Differential Revision: D77164673
@facebook-github-bot
Copy link
Contributor

This pull request was exported from Phabricator. Differential Revision: D77164673

hsharma35 added a commit to hsharma35/executorch that referenced this pull request Jun 25, 2025
Summary:
Pull Request resolved: pytorch#11867

Add idma copy and wait op definitions + fake implementation. These can be used for asynchronous copy/wait.

idma_load: Used for `Any -> DTCM` data transfers. (Note: DTCM to DTCM is ok)
idma_store: Used for `DTCM -> Any` (similar to idma_load).

Since the above ops explicitly specify the direction of data transfer, we can use these to plan memory in specific dtcm banks.

Reviewed By: zonglinpeng, nitish2112

Differential Revision: D77164673
hsharma35 added a commit to hsharma35/executorch that referenced this pull request Jun 25, 2025
Summary:

Add idma copy and wait op definitions + fake implementation. These can be used for asynchronous copy/wait.

idma_load: Used for `Any -> DTCM` data transfers. (Note: DTCM to DTCM is ok)
idma_store: Used for `DTCM -> Any` (similar to idma_load).

Since the above ops explicitly specify the direction of data transfer, we can use these to plan memory in specific dtcm banks.

Reviewed By: zonglinpeng, nitish2112

Differential Revision: D77164673
@facebook-github-bot
Copy link
Contributor

This pull request was exported from Phabricator. Differential Revision: D77164673

hsharma35 added a commit to hsharma35/executorch that referenced this pull request Jun 26, 2025
Summary:

Add idma copy and wait op definitions + fake implementation. These can be used for asynchronous copy/wait.

idma_load: Used for `Any -> DTCM` data transfers. (Note: DTCM to DTCM is ok)
idma_store: Used for `DTCM -> Any` (similar to idma_load).

Since the above ops explicitly specify the direction of data transfer, we can use these to plan memory in specific dtcm banks.

Reviewed By: zonglinpeng, nitish2112

Differential Revision: D77164673
Summary:
Pull Request resolved: pytorch#11867

Add idma copy and wait op definitions + fake implementation. These can be used for asynchronous copy/wait.

idma_load: Used for `Any -> DTCM` data transfers. (Note: DTCM to DTCM is ok)
idma_store: Used for `DTCM -> Any` (similar to idma_load).

Since the above ops explicitly specify the direction of data transfer, we can use these to plan memory in specific dtcm banks.

Reviewed By: zonglinpeng, nitish2112

Differential Revision: D77164673
@facebook-github-bot
Copy link
Contributor

This pull request was exported from Phabricator. Differential Revision: D77164673

@facebook-github-bot facebook-github-bot merged commit 2e56274 into pytorch:main Jun 26, 2025
97 of 98 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. fb-exported release notes: none Do not include this in the release notes

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants