Port ARGBToRGB565 from aarch64 neon to 32 bit

source/row_neon.cc

 /*
  *  Copyright 2011 The LibYuv Project Authors. All rights reserved.
  *
  *  Use of this source code is governed by a BSD-style license
  *  that can be found in the LICENSE file in the root of the source
  *  tree. An additional intellectual property rights grant can be found
  *  in the file PATENTS. All contributing project authors may
  *  be found in the AUTHORS file in the root of the source tree.
  */
 
@@ skipping 299 matching lines @@
     : [kUVToRB]"r"(&yuvconstants->kUVToRB),
       [kUVToG]"r"(&yuvconstants->kUVToG),
       [kUVBiasBGR]"r"(&yuvconstants->kUVBiasBGR),
       [kYToRgb]"r"(&yuvconstants->kYToRgb)
     : "cc", "memory", "q0", "q1", "q2", "q3", "q4",
       "q8", "q9", "q10", "q11", "q12", "q13", "q14", "q15"
   );
 }
 
 #define ARGBTORGB565                                                           \
-    "vshr.u8    d20, d20, #3                   \n"  /* B                    */ \
-    "vshr.u8    d21, d21, #2                   \n"  /* G                    */ \
-    "vshr.u8    d22, d22, #3                   \n"  /* R                    */ \
-    "vmovl.u8   q8, d20                        \n"  /* B                    */ \
-    "vmovl.u8   q9, d21                        \n"  /* G                    */ \
-    "vmovl.u8   q10, d22                       \n"  /* R                    */ \
-    "vshl.u16   q9, q9, #5                     \n"  /* G                    */ \
-    "vshl.u16   q10, q10, #11                  \n"  /* R                    */ \
-    "vorr       q0, q8, q9                     \n"  /* BG                   */ \
-    "vorr       q0, q0, q10                    \n"  /* BGR                  */
+    "vshll.u8    q0, d22, #8                   \n"  /* R                    */ \
+    "vshll.u8    q8, d21, #8                   \n"  /* G                    */ \
+    "vshll.u8    q9, d20, #8                   \n"  /* B                    */ \
+    "vsri.16     q0, q8, #5                    \n"  /* RG                   */ \
+    "vsri.16     q0, q9, #11                   \n"  /* RGB                  */
 
 void I422ToRGB565Row_NEON(const uint8* src_y,
                           const uint8* src_u,
                           const uint8* src_v,
                           uint8* dst_rgb565,
                           const struct YuvConstants* yuvconstants,
                           int width) {
   asm volatile (
     YUVTORGB_SETUP
   "1:                                          \n"
@@ skipping 12 matching lines @@
     : [kUVToRB]"r"(&yuvconstants->kUVToRB),
       [kUVToG]"r"(&yuvconstants->kUVToG),
       [kUVBiasBGR]"r"(&yuvconstants->kUVBiasBGR),
       [kYToRgb]"r"(&yuvconstants->kYToRgb)
     : "cc", "memory", "q0", "q1", "q2", "q3", "q4",
       "q8", "q9", "q10", "q11", "q12", "q13", "q14", "q15"
   );
 }
 
 #define ARGBTOARGB1555                                                         \
-    "vshr.u8    q10, q10, #3                   \n"  /* B                    */ \
-    "vshr.u8    d22, d22, #3                   \n"  /* R                    */ \
-    "vshr.u8    d23, d23, #7                   \n"  /* A                    */ \
-    "vmovl.u8   q8, d20                        \n"  /* B                    */ \
-    "vmovl.u8   q9, d21                        \n"  /* G                    */ \
-    "vmovl.u8   q10, d22                       \n"  /* R                    */ \
-    "vmovl.u8   q11, d23                       \n"  /* A                    */ \
-    "vshl.u16   q9, q9, #5                     \n"  /* G                    */ \
-    "vshl.u16   q10, q10, #10                  \n"  /* R                    */ \
-    "vshl.u16   q11, q11, #15                  \n"  /* A                    */ \
-    "vorr       q0, q8, q9                     \n"  /* BG                   */ \
-    "vorr       q1, q10, q11                   \n"  /* RA                   */ \
-    "vorr       q0, q0, q1                     \n"  /* BGRA                 */
+    "vshll.u8    q0, d23, #8                   \n"  /* A                    */ \
+    "vshll.u8    q8, d22, #8                   \n"  /* R                    */ \
+    "vshll.u8    q9, d21, #8                   \n"  /* G                    */ \
+    "vshll.u8    q10, d20, #8                  \n"  /* B                    */ \
+    "vsri.16     q0, q8, #1                    \n"  /* AR                   */ \
+    "vsri.16     q0, q9, #5                    \n"  /* ARG                  */ \

[fbarchard1, 2016/02/24 18:49:18]

This shift should be 6.  1 for alpha + 5 for R.

[fbarchard1, 2016/02/24 18:57:01]

Done.

+    "vsri.16     q0, q10, #11                  \n"  /* ARGB                 */
 
 void I422ToARGB1555Row_NEON(const uint8* src_y,
                             const uint8* src_u,
                             const uint8* src_v,
                             uint8* dst_argb1555,
                             const struct YuvConstants* yuvconstants,
                             int width) {
   asm volatile (
     YUVTORGB_SETUP
   "1:                                          \n"
@@ skipping 2445 matching lines @@
     "r"(6)             // %5
   : "cc", "memory", "q0", "q1"  // Clobber List
   );
 }
 #endif  // defined(__ARM_NEON__) && !defined(__aarch64__)
 
 #ifdef __cplusplus
 }  // extern "C"
 }  // namespace libyuv
 #endif