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

Enable BFloat16 and TensorFloat32 conversions for cooperative matrices #2213

Merged
merged 3 commits into from
Nov 16, 2023
Merged

Enable BFloat16 and TensorFloat32 conversions for cooperative matrices #2213

merged 3 commits into from
Nov 16, 2023

Commits on Nov 16, 2023

  1. Enable BFloat16 and TensorFloat32 conversions for cooperative matrices 

    Previously added scalar/vector ConvertFToBF16INTEL, ConvertBF16ToFINTEL
and RoundFToTF32INTEL conversions are now enabled for cooperative matrix
type under SPV_INTEL_joint_matrix extension following the spec:
https://github.com/intel/llvm/blob/sycl/sycl/doc/design/spirv-extensions/SPV_INTEL_joint_matrix.asciidoc

Note, joint matrices are not allowed as input/output for these
conversions as it is being deprecated.

Signed-off-by: Sidorov, Dmitry <dmitry.sidorov@intel.com>
    @MrSidims
    MrSidims committed Nov 16, 2023
    Configuration menu
    Copy the full SHA
    3bd87ed View commit details
    Browse the repository at this point in the history

  2. format 

    Signed-off-by: Sidorov, Dmitry <dmitry.sidorov@intel.com>
    @MrSidims
    MrSidims committed Nov 16, 2023
    Configuration menu
    Copy the full SHA
    1df3121 View commit details
    Browse the repository at this point in the history

  3. apply suggestion 

    Signed-off-by: Sidorov, Dmitry <dmitry.sidorov@intel.com>
    @MrSidims
    MrSidims committed Nov 16, 2023
    Configuration menu
    Copy the full SHA
    638c4f1 View commit details
    Browse the repository at this point in the history