shithub: dav1d

Download patch

ref: 13fad75d96942a7c895d8f0f5e1577edbf7c0df4
parent: 3bfe8c7c8a553728e2d6556e4a95f5cd246d1c92
author: Martin Storsjö <martin@martin.st>
date: Sat Aug 29 09:29:07 EDT 2020 

arm64: mc: Use more descriptive element specifiers for loads/stores in 16 bpc put_neon

For loads of a half/full register, the actual size of the elements
doesn't matter, but it makes the code more readable and understandable.


--- a/src/arm/64/mc16.S
+++ b/src/arm/64/mc16.S
@@ -1004,11 +1004,11 @@
         b.gt            2b
         ret
 4:
-        ld1             {v0.8b}, [x2], x3
-        ld1             {v1.8b}, [x2], x3
+        ld1             {v0.4h}, [x2], x3
+        ld1             {v1.4h}, [x2], x3
         subs            w5,  w5,  #2
-        st1             {v0.8b}, [x0], x1
-        st1             {v1.8b}, [x0], x1
+        st1             {v0.4h}, [x0], x1
+        st1             {v1.4h}, [x0], x1
         b.gt            4b
         ret
 80:
@@ -1017,11 +1017,11 @@
         add             x9,  x2,  x3
         lsl             x3,  x3,  #1
 8:
-        ld1             {v0.16b}, [x2], x3
-        ld1             {v1.16b}, [x9], x3
+        ld1             {v0.8h}, [x2], x3
+        ld1             {v1.8h}, [x9], x3
         subs            w5,  w5,  #2
-        st1             {v0.16b}, [x0], x1
-        st1             {v1.16b}, [x8], x1
+        st1             {v0.8h}, [x0], x1
+        st1             {v1.8h}, [x8], x1
         b.gt            8b
         ret
 16: