tree d075e2570c1b7963bd8bfcacab5fc2b736203e2f
parent add4f6479ddd4cb8f3b10326a25716af10627fe3
author Jonathan Wright <jonathan.wright@arm.com> 1659708338 +0100
committer James Zern <jzern@google.com> 1663724169 +0000

Optimize av1_convolve_x_sr_neon using SDOT instruction

Add an alternative AArch64 implementation of av1_convolve_x_sr_neon
for targets that implement the Armv8.4-A SDOT (signed dot-product)
instruction.

The existing MLA implementation of av1_convolve_x_sr_neon is retained
for use 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: I5bbd56af03ce0c0a27c7ea1f221a7e61e4249e1a