Page MenuHomePhabricator

[AArch64][SVE] Add VScale Intrinsic
Changes PlannedPublic

Authored by huntergr on Jun 5 2018, 5:34 AM.

Details

Summary

Adds a new experimental vscale intrinsic to represent the runtime
multiple for scalable vectors.

Implements lowering/legalization for this in the AArch64 backend,
and a unit test for the SVE 'rdvl' instruction over scalable integer
vector types.

Part of the initial SVE codegen series, rfc will be posted soon. Note: this patch is currently just to support the scalable vector codegen rfc, and does not need a full review yet.

Diff Detail

Event Timeline

huntergr created this revision.Jun 5 2018, 5:34 AM
huntergr planned changes to this revision.Mar 22 2019, 3:00 AM

Needs to be updated with better isel patterns, instead of just the bare minimum required for demonstration.