Skip to content

feat: add MPS device support to auto_dataloader pin_memory logic#3764

Closed
Aanushka001 wants to merge 1 commit into
pytorch:masterfrom
Aanushka001:mps-pin-memory
Closed

feat: add MPS device support to auto_dataloader pin_memory logic#3764
Aanushka001 wants to merge 1 commit into
pytorch:masterfrom
Aanushka001:mps-pin-memory

Conversation

@Aanushka001
Copy link
Copy Markdown

Description

This PR extends the auto_dataloader() function to also enable pin_memory for MPS (Apple Silicon) devices, not just CUDA devices. This ensures data loading performance is optimized when using Apple Silicon GPUs.

Changes

  • Updated auto_dataloader() docstring to mention MPS
  • Updated pin_memory default logic to check for MPS devices

Related

Based on original PR #3763

@github-actions github-actions Bot added the module: distributed Distributed module label May 31, 2026
@vfdev-5
Copy link
Copy Markdown
Collaborator

vfdev-5 commented May 31, 2026

Same, this PR contains irrelevant and inappropriate changes unrelated to the topic of the PR. Closing.

@vfdev-5 vfdev-5 closed this May 31, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

module: distributed Distributed module

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants