aarch64: Use ret x<n> instead of br x<n> where possible

Change AArch64 assembly code to use:
  ret     x<n>
instead of:
  br      x<n>

"ret x<n>" is already used in a lot of places so this patch makes it
consistent across the code base. This does not change behavior or
performance.

In addition, this change reduces the number of landing pads needed in
a subsequent patch to support the Armv8.5-A Branch Target
Identification (BTI) security feature.

Signed-off-by: Jonathan Wright <jonathan.wright@arm.com>
Signed-off-by: Martin Storsjö <martin@martin.st>

1 files changed, 26 insertions, 26 deletions

diff --git a/libavcodec/aarch64/vp9lpf_neon.S b/libavcodec/aarch64/vp9lpf_neon.S
index 0878763020..694ff8956f 100644
--- a/libavcodec/aarch64/vp9lpf_neon.S
+++ b/libavcodec/aarch64/vp9lpf_neon.S
@@ -399,7 +399,7 @@
 .endif
         // If no pixels needed flat8in nor flat8out, jump to a
         // writeout of the inner 4 pixels
-        br              x14
+        ret             x14
 1:
 
         mov             x5,  v7.d[0]
@@ -411,7 +411,7 @@
         cbnz            x5,  1f
 .endif
         // If no pixels need flat8out, jump to a writeout of the inner 6 pixels
-        br              x15
+        ret             x15
 
 1:
         // flat8out
@@ -532,32 +532,32 @@ function vp9_loop_filter_4
         loop_filter     4,  .8b,  0,    v16, v17, v18, v19, v28, v29, v30, v31
         ret
 9:
-        br              x10
+        ret             x10
 endfunc
 
 function vp9_loop_filter_4_16b_mix_44
         loop_filter     4,  .16b, 44,   v16, v17, v18, v19, v28, v29, v30, v31
         ret
 9:
-        br              x10
+        ret             x10
 endfunc
 
 function vp9_loop_filter_8
         loop_filter     8,  .8b,  0,    v16, v17, v18, v19, v28, v29, v30, v31
         ret
 6:
-        br              x13
+        ret             x13
 9:
-        br              x10
+        ret             x10
 endfunc
 
 function vp9_loop_filter_8_16b_mix
         loop_filter     8,  .16b, 88,   v16, v17, v18, v19, v28, v29, v30, v31
         ret
 6:
-        br              x13
+        ret             x13
 9:
-        br              x10
+        ret             x10
 endfunc
 
 function vp9_loop_filter_16
@@ -568,7 +568,7 @@ function vp9_loop_filter_16
         ldp             d10, d11, [sp], 0x10
         ldp             d12, d13, [sp], 0x10
         ldp             d14, d15, [sp], 0x10
-        br              x10
+        ret             x10
 endfunc
 
 function vp9_loop_filter_16_16b
@@ -579,7 +579,7 @@ function vp9_loop_filter_16_16b
         ldp             d10, d11, [sp], 0x10
         ldp             d12, d13, [sp], 0x10
         ldp             d14, d15, [sp], 0x10
-        br              x10
+        ret             x10
 endfunc
 
 .macro loop_filter_4
@@ -648,7 +648,7 @@ function ff_vp9_loop_filter_v_4_8_neon, export=1
         st1             {v23.8b}, [x9], x1
         st1             {v25.8b}, [x0], x1
 
-        br              x10
+        ret             x10
 endfunc
 
 function ff_vp9_loop_filter_v_44_16_neon, export=1
@@ -672,7 +672,7 @@ function ff_vp9_loop_filter_v_44_16_neon, export=1
         st1             {v23.16b}, [x9], x1
         st1             {v25.16b}, [x0], x1
 
-        br              x10
+        ret             x10
 endfunc
 
 function ff_vp9_loop_filter_h_4_8_neon, export=1
@@ -714,7 +714,7 @@ function ff_vp9_loop_filter_h_4_8_neon, export=1
         st1             {v25.s}[0], [x9], x1
         st1             {v25.s}[1], [x0], x1
 
-        br              x10
+        ret             x10
 endfunc
 
 function ff_vp9_loop_filter_h_44_16_neon, export=1
@@ -766,7 +766,7 @@ function ff_vp9_loop_filter_h_44_16_neon, export=1
         st1             {v25.s}[1], [x9], x1
         st1             {v25.s}[3], [x0], x1
 
-        br              x10
+        ret             x10
 endfunc
 
 function ff_vp9_loop_filter_v_8_8_neon, export=1
@@ -793,14 +793,14 @@ function ff_vp9_loop_filter_v_8_8_neon, export=1
         st1             {v23.8b}, [x9], x1
         st1             {v26.8b}, [x0], x1
 
-        br              x10
+        ret             x10
 6:
         sub             x9,  x0,  x1, lsl #1
         st1             {v22.8b}, [x9], x1
         st1             {v24.8b}, [x0], x1
         st1             {v23.8b}, [x9], x1
         st1             {v25.8b}, [x0], x1
-        br              x10
+        ret             x10
 endfunc
 
 .macro mix_v_16 mix
@@ -828,14 +828,14 @@ function ff_vp9_loop_filter_v_\mix\()_16_neon, export=1
         st1             {v23.16b}, [x9], x1
         st1             {v26.16b}, [x0], x1
 
-        br              x10
+        ret             x10
 6:
         sub             x9,  x0,  x1, lsl #1
         st1             {v22.16b}, [x9], x1
         st1             {v24.16b}, [x0], x1
         st1             {v23.16b}, [x9], x1
         st1             {v25.16b}, [x0], x1
-        br              x10
+        ret             x10
 endfunc
 .endm
 
@@ -876,7 +876,7 @@ function ff_vp9_loop_filter_h_8_8_neon, export=1
         st1             {v23.8b}, [x9], x1
         st1             {v27.8b}, [x0], x1
 
-        br              x10
+        ret             x10
 6:
         // If we didn't need to do the flat8in part, we use the same writeback
         // as in loop_filter_h_4_8.
@@ -891,7 +891,7 @@ function ff_vp9_loop_filter_h_8_8_neon, export=1
         st1             {v24.s}[1], [x0], x1
         st1             {v25.s}[0], [x9], x1
         st1             {v25.s}[1], [x0], x1
-        br              x10
+        ret             x10
 endfunc
 
 .macro mix_h_16 mix
@@ -942,7 +942,7 @@ function ff_vp9_loop_filter_h_\mix\()_16_neon, export=1
         st1             {v27.8b},   [x9], x1
         st1             {v27.d}[1], [x0], x1
 
-        br              x10
+        ret             x10
 6:
         add             x9,  x9,  #2
         add             x0,  x0,  #2
@@ -963,7 +963,7 @@ function ff_vp9_loop_filter_h_\mix\()_16_neon, export=1
         st1             {v24.s}[3], [x0], x1
         st1             {v25.s}[1], [x9], x1
         st1             {v25.s}[3], [x0], x1
-        br              x10
+        ret             x10
 endfunc
 .endm
 
@@ -1022,7 +1022,7 @@ function ff_vp9_loop_filter_v_16_8_neon, export=1
         ldp             d10, d11, [sp], 0x10
         ldp             d12, d13, [sp], 0x10
         ldp             d14, d15, [sp], 0x10
-        br              x10
+        ret             x10
 8:
         add             x9,  x9,  x1, lsl #2
         // If we didn't do the flat8out part, the output is left in the
@@ -1091,7 +1091,7 @@ function ff_vp9_loop_filter_v_16_16_neon, export=1
         ldp             d10, d11, [sp], 0x10
         ldp             d12, d13, [sp], 0x10
         ldp             d14, d15, [sp], 0x10
-        br              x10
+        ret             x10
 8:
         add             x9,  x9,  x1, lsl #2
         st1             {v21.16b}, [x9], x1
@@ -1168,7 +1168,7 @@ function ff_vp9_loop_filter_h_16_8_neon, export=1
         ldp             d10, d11, [sp], 0x10
         ldp             d12, d13, [sp], 0x10
         ldp             d14, d15, [sp], 0x10
-        br              x10
+        ret             x10
 8:
         // The same writeback as in loop_filter_h_8_8
         sub             x9,  x0,  #4
@@ -1287,7 +1287,7 @@ function ff_vp9_loop_filter_h_16_16_neon, export=1
         ldp             d10, d11, [sp], 0x10
         ldp             d12, d13, [sp], 0x10
         ldp             d14, d15, [sp], 0x10
-        br              x10
+        ret             x10
 8:
         sub             x9,  x0,  #4
         add             x0,  x9,  x1, lsl #3