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

xpu/ocl: Minor generic-vendor fixes #2084

Open
wants to merge 4 commits into
base: main
Choose a base branch
from
Open

xpu/ocl: Minor generic-vendor fixes #2084

wants to merge 4 commits into from
+10 −14

Commits on Sep 5, 2024

  1. xpu: fix engine_factory_t::count() 

    If status != success then it is a non-zero value, but in that case we
have zero devices. Return zero instead so we don't index off the end of
anything.
    @nwnk
    nwnk committed Sep 5, 2024
    Configuration menu
    Copy the full SHA
    cedbd91 View commit details
    Browse the repository at this point in the history

  2. xpu/ocl: Always consider 0 devices to be a runtime error

    @nwnk
    nwnk committed Sep 5, 2024
    Configuration menu
    Copy the full SHA
    c65af51 View commit details
    Browse the repository at this point in the history

  3. xpu/ocl: Don't filter by vendor ID 

    We can't do that at this level because it means device indices change
based on how we were called.
    @nwnk
    nwnk committed Sep 5, 2024
    Configuration menu
    Copy the full SHA
    36f25ab View commit details
    Browse the repository at this point in the history

  4. xpu/ocl: Permit initialization on any capable OpenCL platform 

    This isn't the place for it, this is generic code and doesn't rely on
Intel's OpenCL platform. There's another check later, conditional on
DNNL_VENDOR_INTEL at build time, that has the same effect. On a
CometLake system, with recent Mesa using rusticl and iris:

    [----------] 2 tests from AllEngineKinds/engine_test_t
    [ RUN      ] AllEngineKinds/engine_test_t.TestMultithreading/0
    onednn_verbose,v1,info,oneDNN v3.6.0 (commit d36af6e408cd77b84531b5bbe4b6f3874df625a3)
    onednn_verbose,v1,info,cpu,runtime:OpenMP,nthr:20
    onednn_verbose,v1,info,cpu,isa:Intel AVX2
    onednn_verbose,v1,info,gpu,runtime:OpenCL
    WARNING: OpenCL support via iris driver is incomplete.
    For a complete and conformant OpenCL implementation, use
    https://github.com/intel/compute-runtime instead
    onednn_verbose,v1,common,error,runtime,unsupported ocl platform (expected intel got rusticl)
    @nwnk
    nwnk committed Sep 5, 2024
    Configuration menu
    Copy the full SHA
    ae7ee94 View commit details
    Browse the repository at this point in the history