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.

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

[EXPERIMENTNAL][DO NOT MEREG] Add a prototype of F16/BF16xINT4 GEMM #1572

Draft
wants to merge 39 commits into
base: develop
Choose a base branch
from

Commits on Oct 11, 2024

  1. add a prototype of int4

    Jing Zhang committed Oct 11, 2024
    Configuration menu
    Copy the full SHA
    0f3b88b View commit details
    Browse the repository at this point in the history

Commits on Oct 13, 2024

  1. clean

    Jing Zhang committed Oct 13, 2024
    Configuration menu
    Copy the full SHA
    3ef4d2c View commit details
    Browse the repository at this point in the history

Commits on Oct 14, 2024

  1. debug

    Jing Zhang committed Oct 14, 2024
    Configuration menu
    Copy the full SHA
    c3d05c0 View commit details
    Browse the repository at this point in the history

Commits on Oct 15, 2024

  1. clean

    Jing Zhang committed Oct 15, 2024
    Configuration menu
    Copy the full SHA
    40d038e View commit details
    Browse the repository at this point in the history
  2. clean

    Jing Zhang committed Oct 15, 2024
    Configuration menu
    Copy the full SHA
    77ad000 View commit details
    Browse the repository at this point in the history
  3. move packed into dynamic_buffer

    Jing Zhang committed Oct 15, 2024
    Configuration menu
    Copy the full SHA
    c3a4652 View commit details
    Browse the repository at this point in the history

Commits on Oct 16, 2024

  1. fixed coord reset

    Jing Zhang committed Oct 16, 2024
    Configuration menu
    Copy the full SHA
    24e18ae View commit details
    Browse the repository at this point in the history

Commits on Oct 18, 2024

  1. add fast pki4 to half conversion

    Jing Zhang committed Oct 18, 2024
    Configuration menu
    Copy the full SHA
    c13366a View commit details
    Browse the repository at this point in the history
  2. fix

    Jing Zhang committed Oct 18, 2024
    Configuration menu
    Copy the full SHA
    205e036 View commit details
    Browse the repository at this point in the history

Commits on Oct 20, 2024

  1. fixed reference and host_tensor

    Jing Zhang committed Oct 20, 2024
    Configuration menu
    Copy the full SHA
    05ab910 View commit details
    Browse the repository at this point in the history
  2. fixed tensor init

    Jing Zhang committed Oct 20, 2024
    Configuration menu
    Copy the full SHA
    2807c69 View commit details
    Browse the repository at this point in the history
  3. format

    Jing Zhang committed Oct 20, 2024
    Configuration menu
    Copy the full SHA
    222e968 View commit details
    Browse the repository at this point in the history

Commits on Oct 21, 2024

  1. debug i4_to_f16_convert

    Jing Zhang committed Oct 21, 2024
    Configuration menu
    Copy the full SHA
    398f885 View commit details
    Browse the repository at this point in the history
  2. format

    Jing Zhang committed Oct 21, 2024
    Configuration menu
    Copy the full SHA
    65cfb2a View commit details
    Browse the repository at this point in the history
  3. fixed splitk

    Jing Zhang committed Oct 21, 2024
    Configuration menu
    Copy the full SHA
    82bb8dd View commit details
    Browse the repository at this point in the history

Commits on Oct 22, 2024

  1. weight permute

    Jing Zhang committed Oct 22, 2024
    Configuration menu
    Copy the full SHA
    e053e94 View commit details
    Browse the repository at this point in the history
  2. add b tile permute

    Jing Zhang committed Oct 22, 2024
    Configuration menu
    Copy the full SHA
    be98313 View commit details
    Browse the repository at this point in the history
  3. clean

    Jing Zhang committed Oct 22, 2024
    Configuration menu
    Copy the full SHA
    35d8627 View commit details
    Browse the repository at this point in the history
  4. weight permute with splitki

    Jing Zhang committed Oct 22, 2024
    Configuration menu
    Copy the full SHA
    9fed0ad View commit details
    Browse the repository at this point in the history
  5. format

    Jing Zhang committed Oct 22, 2024
    Configuration menu
    Copy the full SHA
    5d42067 View commit details
    Browse the repository at this point in the history

Commits on Oct 23, 2024

  1. improve weight layout

    Jing Zhang committed Oct 23, 2024
    Configuration menu
    Copy the full SHA
    6d0e78b View commit details
    Browse the repository at this point in the history
  2. add and_or_b32

    Jing Zhang committed Oct 23, 2024
    Configuration menu
    Copy the full SHA
    af2c016 View commit details
    Browse the repository at this point in the history
  3. fixed splitk crush

    Jing Zhang committed Oct 23, 2024
    Configuration menu
    Copy the full SHA
    6a2521e View commit details
    Browse the repository at this point in the history
  4. add permute switch as a template

    Jing Zhang committed Oct 23, 2024
    Configuration menu
    Copy the full SHA
    786a0fa View commit details
    Browse the repository at this point in the history
  5. recover v3r1

    Jing Zhang committed Oct 23, 2024
    Configuration menu
    Copy the full SHA
    7cb3d6f View commit details
    Browse the repository at this point in the history
  6. clean

    Jing Zhang committed Oct 23, 2024
    Configuration menu
    Copy the full SHA
    e9b7f26 View commit details
    Browse the repository at this point in the history
  7. failure with intrawave v2

    Jing Zhang committed Oct 23, 2024
    Configuration menu
    Copy the full SHA
    f16f55a View commit details
    Browse the repository at this point in the history

Commits on Oct 24, 2024

  1. fixed

    Jing Zhang committed Oct 24, 2024
    Configuration menu
    Copy the full SHA
    9e15aa3 View commit details
    Browse the repository at this point in the history
  2. fixed

    Jing Zhang committed Oct 24, 2024
    Configuration menu
    Copy the full SHA
    e463256 View commit details
    Browse the repository at this point in the history

Commits on Oct 27, 2024

  1. add ckProfiler

    Jing Zhang committed Oct 27, 2024
    Configuration menu
    Copy the full SHA
    f03dda4 View commit details
    Browse the repository at this point in the history

Commits on Oct 28, 2024

  1. add bfp16 support

    Jing Zhang committed Oct 28, 2024
    Configuration menu
    Copy the full SHA
    1d82d46 View commit details
    Browse the repository at this point in the history
  2. add bf16 example

    Jing Zhang committed Oct 28, 2024
    Configuration menu
    Copy the full SHA
    d642ce4 View commit details
    Browse the repository at this point in the history

Commits on Oct 29, 2024

  1. fixed int4 to bhalf_t conversion

    Jing Zhang committed Oct 29, 2024
    Configuration menu
    Copy the full SHA
    a6ccd2e View commit details
    Browse the repository at this point in the history
  2. format

    Jing Zhang committed Oct 29, 2024
    Configuration menu
    Copy the full SHA
    9de3a08 View commit details
    Browse the repository at this point in the history
  3. fixed int4 to bf16 conversion

    Jing Zhang committed Oct 29, 2024
    Configuration menu
    Copy the full SHA
    d6a7acf View commit details
    Browse the repository at this point in the history
  4. clean

    Jing Zhang committed Oct 29, 2024
    Configuration menu
    Copy the full SHA
    8441cbe View commit details
    Browse the repository at this point in the history

Commits on Nov 11, 2024

  1. add instances for mem

    Jing Zhang committed Nov 11, 2024
    Configuration menu
    Copy the full SHA
    23bdf72 View commit details
    Browse the repository at this point in the history

Commits on Nov 14, 2024

  1. clean

    Jing Zhang committed Nov 14, 2024
    Configuration menu
    Copy the full SHA
    66873f3 View commit details
    Browse the repository at this point in the history
  2. fixed host tensor size

    Jing Zhang committed Nov 14, 2024
    Configuration menu
    Copy the full SHA
    b0b6766 View commit details
    Browse the repository at this point in the history