Commit f32690a2 authored by Martin Storsjö's avatar Martin Storsjö

aarch64: vp9lpf: Use dup+rev16+uzp1 instead of dup+lsr+dup+trn1

This is one cycle faster in total, and three instructions fewer.

Before:
vp9_loop_filter_mix2_v_44_16_neon: 123.2
After:
vp9_loop_filter_mix2_v_44_16_neon: 122.2

This is cherrypicked from libav commit
3bf9c483.
Signed-off-by: 's avatarMartin Storsjö <martin@martin.st>
parent 3fbbad29
...@@ -162,18 +162,15 @@ ...@@ -162,18 +162,15 @@
dup v2\sz, w3 // I dup v2\sz, w3 // I
dup v3\sz, w4 // H dup v3\sz, w4 // H
.else .else
dup v0.8b, w2 // E dup v0.8h, w2 // E
dup v2.8b, w3 // I dup v2.8h, w3 // I
dup v3.8b, w4 // H dup v3.8h, w4 // H
lsr w5, w2, #8 rev16 v1.16b, v0.16b // E
lsr w6, w3, #8 rev16 v4.16b, v2.16b // I
lsr w7, w4, #8 rev16 v5.16b, v3.16b // H
dup v1.8b, w5 // E uzp1 v0.16b, v0.16b, v1.16b
dup v4.8b, w6 // I uzp1 v2.16b, v2.16b, v4.16b
dup v5.8b, w7 // H uzp1 v3.16b, v3.16b, v5.16b
trn1 v0.2d, v0.2d, v1.2d
trn1 v2.2d, v2.2d, v4.2d
trn1 v3.2d, v3.2d, v5.2d
.endif .endif
uabd v4\sz, v20\sz, v21\sz // abs(p3 - p2) uabd v4\sz, v20\sz, v21\sz // abs(p3 - p2)
......
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment