Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Resnet Host Side Optimizations #8801

Merged
merged 8 commits into from
May 25, 2024
Merged

Resnet Host Side Optimizations #8801

merged 8 commits into from
May 25, 2024

Commits on May 25, 2024

  1. #8837: Bypass buffer page mapping generation for height sharding. Opt… 

    …imize 1D MM cache hit for rt args
    @tt-aho
    tt-aho committed May 25, 2024
    Configuration menu
    Copy the full SHA
    56a9730 View commit details
    Browse the repository at this point in the history

  2. #8837: Try 256B load stores for writing to hugepage 

    Fallback to 128B load stores before 32B
Get rid of conditional at start of memcpy_to_device and fallthrough instead
    @tt-aho
    tt-aho committed May 25, 2024
    Configuration menu
    Copy the full SHA
    5626c54 View commit details
    Browse the repository at this point in the history

  3. #8837: Minor cleanup/fix of pushing to issue queue for EP

    @tt-aho
    tt-aho committed May 25, 2024
    Configuration menu
    Copy the full SHA
    0decc44 View commit details
    Browse the repository at this point in the history

  4. #8837: Remove no-op cores from convs and refactor setting RTAs on cac… 

    …he hit
    @tt-aho
    tt-aho committed May 25, 2024
    Configuration menu
    Copy the full SHA
    2558c95 View commit details
    Browse the repository at this point in the history

  5. #8837: Optimize cache hit updating RTAs for some ops

    @tt-aho
    tt-aho committed May 25, 2024
    Configuration menu
    Copy the full SHA
    e027ca2 View commit details
    Browse the repository at this point in the history

  6. #8837: Remove some unneeded / duplicated assertions

    @tt-aho
    tt-aho committed May 25, 2024
    Configuration menu
    Copy the full SHA
    5d9c540 View commit details
    Browse the repository at this point in the history

  7. #8837: Skip generating buffer page mapping for linear sharded reads

    @tt-aho
    tt-aho committed May 25, 2024
    Configuration menu
    Copy the full SHA
    497bb81 View commit details
    Browse the repository at this point in the history

  8. #8837: Add 256B nt memcpy tests to test_pull_from_pcie

    @tt-aho
    tt-aho committed May 25, 2024
    Configuration menu
    Copy the full SHA
    d75cd0c View commit details
    Browse the repository at this point in the history