Commit f6a6b224 authored by Martin Storsjö's avatar Martin Storsjö
Browse files

arm: looprestoration: Fix register shifting at the end of a vertical run

This uses the right registers, corresponding to the ones shifted
in the arm64 version.
parent e0538f04
Pipeline #4407 passed with stages
in 6 minutes and 46 seconds
......@@ -529,8 +529,8 @@ function wiener_filter_v_neon, export=1
8: // At this point, all registers up to q14-15,q1 are loaded with
// edge/padding (depending on how many rows are left).
filter 0 // This branches to 9f when done
vmov q13, q14
vmov q14, q1
vmov q14, q15
vmov q15, q1
b 8b
9: // End of one vertical slice.
Supports Markdown
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