Clang SME ACLE intrinsics

In our downstream compiler, we have previously implemented some custom C intrinsics for ARM Scalable Matrix Extension. Since ARM posted the ACLE draft for SME (, we have started refactoring our implementation to support ACLE instead. The first patch has been posted on Phabricator (⚙ D127910 [Clang][AArch64] Add SME C intrinsics for load and store). I would appreciate it if you could review and give us some feedback. We would love to contribute more to the effort if this is the right direction.


I have added another patch for the read and write (mova) SME intrinsics (⚙ D128648 [Clang][AArch64] Add SME C intrinsics for read and write (mova)) . We have other intrinsics implemented and will upstream them as well.

I will edit and rebase these patches on the ACLE attributes patch.