[x264-devel] aarch64: Simplify the decimate_score functions

[Martin Storsjö]

x264 | branch: master | Martin Storsjö <martin at martin.st> | Thu Aug 13 23:59:26 2015 +0300| [ef6034812162fc8b51bfd5e87387f405d1cc30cb] | committer: Henrik Gramner

aarch64: Simplify the decimate_score functions

After doing a left shift by the number of bits returned by clz,
only bits set to zero can be shifted out, so if the register
was nonzero to start with (which is checked), it can't become
zero here.

> http://git.videolan.org/gitweb.cgi/x264.git/?a=commit;h=ef6034812162fc8b51bfd5e87387f405d1cc30cb
---

 common/aarch64/quant-a.S |   12 ++----------
 1 file changed, 2 insertions(+), 10 deletions(-)

diff --git a/common/aarch64/quant-a.S b/common/aarch64/quant-a.S
index 2e1558b..443a91d 100644
--- a/common/aarch64/quant-a.S
+++ b/common/aarch64/quant-a.S
@@ -328,17 +328,13 @@ function x264_decimate_score\size\()_neon, export=1
     lsr         x6,  x3,  #2
     lsl         x1,  x1,  x3
     ldrb        w7,  [x5, x6]
-    cbz         x1,  2f
     lsl         x1,  x1,  #4
     add         w0,  w0,  w7
     cbnz        x1,  1b
     ret
-2:
-    add         w0,  w0,  w7
-0:
-    ret
 9:
     mov         w0,  #9
+0:
     ret
 endfunc
 .endm
@@ -399,17 +395,13 @@ function x264_decimate_score64_neon, export=1
     clz         x3,  x1
     lsl         x1,  x1,  x3
     ldrb        w7,  [x5, x3]
-    cbz         x1,  2f
     lsl         x1,  x1,  #1
     add         w0,  w0,  w7
     cbnz        x1,  1b
     ret
-2:
-    add         w0,  w0,  w7
-0:
-    ret
 9:
     mov         w0,  #9
+0:
     ret
 endfunc