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

Rework setSIMDExtension to compile with partial SIMD enabled #455

Merged
merged 2 commits into from
Oct 29, 2024
Merged

Rework setSIMDExtension to compile with partial SIMD enabled #455

merged 2 commits into from
Oct 29, 2024

Commits on Oct 29, 2024

  1. Rework setSIMDExtension to compile with partial SIMD enabled 

    There are existing options to enable both Arm native and SIMDe-based
intrinsics implementations, however if only one of these is enabled then
compilation for Arm targets currently fails.

Fix this by adjusting the existing #ifdef guards and adding new ones to
cover the previously failing option combinations. Also stop checking
REAL_TARGET_ARM since it does not distinguish between which SIMD targets
are enabled.
    @georges-arm
    georges-arm committed Oct 29, 2024
    Configuration menu
    Copy the full SHA
    ffe6ea8 View commit details
    Browse the repository at this point in the history

  2. Disable RdCost Neon code unless TARGET_SIMD_X86 is enabled 

    The existing Neon code in RdCostARM.h depends on being able to call into
the existing SIMDe-based x86 kernels. If VVENC_ENABLE_X86_SIMD is
disabled in the CMake then these kernels are unavailable and compilation
fails.

Until native Neon kernels are added for the missing functions, simply
disable the optimised RdCost Neon code unless the SIMDe kernels are also
available.
    @georges-arm
    georges-arm committed Oct 29, 2024
    Configuration menu
    Copy the full SHA
    b183393 View commit details
    Browse the repository at this point in the history