tree ac4341f7d57e313821af485d13a2ad7f396c3bae
parent 9dfc5b34d857de0003b888594fc7dc62c8504d0b
author Jonathan Wright <jonathan.wright@arm.com> 1654189762 +0100
committer James Zern <jzern@google.com> 1663380081 +0000

Optimize av1_convolve_2d_sr_neon using SDOT instruction

Add an alternative AArch64 implementation of
av1_convolve_2d_sr_horiz_neon - the horizontal portion of
av1_convolve_2d_sr_neon - using the Armv8.4-A SDOT (signed dot-
product) instruction.

The existing MLA-based implementation of is retained and used on
target CPUs that do not implement the SDOT instruction (or CPUs
executing in AArch32 mode.) The availability of the SDOT instruction
is indicated by the feature macro __ARM_FEATURE_DOTPROD.

Change-Id: I258e414e84c28f2c721855fbb3d8c4f4418fac63