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

Add ARM aarch64 support to dispatcher #1

Closed
wants to merge 3 commits into from
Closed

Conversation

wd13384
Copy link

@wd13384 wd13384 commented Jan 28, 2021

Signed-off-by: William Dean wd13384@gmail.com

Tested on Apple M1 with oneVPL provided docker files (except Ubuntu 19.10, which does not appear to support aarch64 in docker hub)

Signed-off-by: William Dean <wd13384@gmail.com>
dispatcher/linux/mfxloader.cpp Outdated Show resolved Hide resolved
Signed-off-by: William Dean <wd13384@gmail.com>
@mav-intel
Copy link
Contributor

Thanks for the update! After some internal discussions we think the better direction is to remove the specific architectures completely and just check what we care about directly - the bit-ness.

So rather than checking for i386, perhaps something like?

#if INTPTR_MAX == INT32_MAX
    #define LIBMFXSW "libvplswref32.so.1"
    #define LIBMFXHW "libmfxhw32.so.1"
#elif INTPTR_MAX == INT64_MAX
    #define LIBMFXSW "libvplswref64.so.1"
    #define LIBMFXHW "libmfxhw64.so.1"
#else	
    #error Unsupported architecture	 
#endif

Signed-off-by: William Dean <wd13384@gmail.com>
@StefanBruens
Copy link
Contributor

@wd13384 - can you please rebase this?
@mav-intel - is there anything in here which is still unresolved and needs changing?

This not only fixes aarch64, but also RISC-V, and all the legacy archs out there.

StefanBruens referenced this pull request in StefanBruens/oneVPL Sep 13, 2022
Both oneVPL SW implementations and dGPUs are available on any platform,
so there is no reason to restrict it to x86_64 only.

Remove no longer used LIBMFXSW define.

Also see #1.
@StefanBruens
Copy link
Contributor

There is also another x86_64 check here, which is AFAICS completely unnecessary:
https://github.com/oneapi-src/oneVPL/blob/15b79165c30fc4a7bf6efdf261a39b128b63b896/tools/legacy/sample_common/src/sample_utils.cpp#L45-L49

@jonrecker
Copy link
Contributor

PR was out of sync with the HEAD so changes were pulled into the following commit: 6182082

@mav-intel mav-intel closed this Jan 30, 2023
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

4 participants