ARM: make some NEON macros reusable

Signed-off-by: Mans Rullgard <mans@mansr.com>

libavcodec/arm/h264dsp_neon.S

@@ -19,46 +19,7 @@
  */
 
 #include "asm.S"
-
-        .macro transpose_8x8 r0 r1 r2 r3 r4 r5 r6 r7
-        vtrn.32         \r0, \r4
-        vtrn.32         \r1, \r5
-        vtrn.32         \r2, \r6
-        vtrn.32         \r3, \r7
-        vtrn.16         \r0, \r2
-        vtrn.16         \r1, \r3
-        vtrn.16         \r4, \r6
-        vtrn.16         \r5, \r7
-        vtrn.8          \r0, \r1
-        vtrn.8          \r2, \r3
-        vtrn.8          \r4, \r5
-        vtrn.8          \r6, \r7
-        .endm
-
-        .macro transpose_4x4 r0 r1 r2 r3
-        vtrn.16         \r0, \r2
-        vtrn.16         \r1, \r3
-        vtrn.8          \r0, \r1
-        vtrn.8          \r2, \r3
-        .endm
-
-        .macro swap4 r0 r1 r2 r3 r4 r5 r6 r7
-        vswp            \r0, \r4
-        vswp            \r1, \r5
-        vswp            \r2, \r6
-        vswp            \r3, \r7
-        .endm
-
-        .macro transpose16_4x4 r0 r1 r2 r3 r4 r5 r6 r7
-        vtrn.32         \r0, \r2
-        vtrn.32         \r1, \r3
-        vtrn.32         \r4, \r6
-        vtrn.32         \r5, \r7
-        vtrn.16         \r0, \r1
-        vtrn.16         \r2, \r3
-        vtrn.16         \r4, \r5
-        vtrn.16         \r6, \r7
-        .endm
+#include "neon.S"
 
 /* chroma_mc8(uint8_t *dst, uint8_t *src, int stride, int h, int x, int y) */
         .macro  h264_chroma_mc8 type

libavcodec/arm/neon.S

@@ -0,0 +1,59 @@
+/*
+ * Copyright (c) 2008 Mans Rullgard <mans@mansr.com>
+ *
+ * This file is part of Libav.
+ *
+ * Libav is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation; either
+ * version 2.1 of the License, or (at your option) any later version.
+ *
+ * Libav is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with Libav; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
+ */
+
+.macro  transpose_8x8   r0, r1, r2, r3, r4, r5, r6, r7
+        vtrn.32         \r0, \r4
+        vtrn.32         \r1, \r5
+        vtrn.32         \r2, \r6
+        vtrn.32         \r3, \r7
+        vtrn.16         \r0, \r2
+        vtrn.16         \r1, \r3
+        vtrn.16         \r4, \r6
+        vtrn.16         \r5, \r7
+        vtrn.8          \r0, \r1
+        vtrn.8          \r2, \r3
+        vtrn.8          \r4, \r5
+        vtrn.8          \r6, \r7
+.endm
+
+.macro  transpose_4x4   r0, r1, r2, r3
+        vtrn.16         \r0, \r2
+        vtrn.16         \r1, \r3
+        vtrn.8          \r0, \r1
+        vtrn.8          \r2, \r3
+.endm
+
+.macro  swap4           r0, r1, r2, r3, r4, r5, r6, r7
+        vswp            \r0, \r4
+        vswp            \r1, \r5
+        vswp            \r2, \r6
+        vswp            \r3, \r7
+.endm
+
+.macro  transpose16_4x4 r0, r1, r2, r3, r4, r5, r6, r7
+        vtrn.32         \r0, \r2
+        vtrn.32         \r1, \r3
+        vtrn.32         \r4, \r6
+        vtrn.32         \r5, \r7
+        vtrn.16         \r0, \r1
+        vtrn.16         \r2, \r3
+        vtrn.16         \r4, \r5
+        vtrn.16         \r6, \r7
+.endm

libavcodec/arm/vp8dsp_neon.S

@@ -22,6 +22,7 @@
  */
 
 #include "asm.S"
+#include "neon.S"
 
 function ff_vp8_luma_dc_wht_neon, export=1
         vld1.16         {q0-q1},  [r1,:128]
@@ -454,23 +455,6 @@ endfunc
     .endif
 .endm
 
-.macro transpose8x16matrix
-        vtrn.32         q0,   q4
-        vtrn.32         q1,   q5
-        vtrn.32         q2,   q6
-        vtrn.32         q3,   q7
-
-        vtrn.16         q0,   q2
-        vtrn.16         q1,   q3
-        vtrn.16         q4,   q6
-        vtrn.16         q5,   q7
-
-        vtrn.8          q0,   q1
-        vtrn.8          q2,   q3
-        vtrn.8          q4,   q5
-        vtrn.8          q6,   q7
-.endm
-
 .macro  vp8_v_loop_filter16 name, inner=0, simple=0
 function ff_vp8_v_loop_filter16\name\()_neon, export=1
         vpush           {q4-q7}
@@ -605,7 +589,7 @@ function ff_vp8_h_loop_filter16\name\()_neon, export=1
         vld1.8          {d13},    [r0], r1
         vld1.8          {d15},    [r0], r1
 
-        transpose8x16matrix
+        transpose_8x8   q0,  q1,  q2,  q3,  q4,  q5,  q6,  q7
 
         vdup.8          q14, r2                 @ flim_E
     .if !\simple
@@ -616,7 +600,7 @@ function ff_vp8_h_loop_filter16\name\()_neon, export=1
 
         sub             r0,  r0,  r1, lsl #4    @ backup 16 rows
 
-        transpose8x16matrix
+        transpose_8x8   q0,  q1,  q2,  q3,  q4,  q5,  q6,  q7
 
         @ Store pixels:
         vst1.8          {d0},     [r0],     r1
@@ -670,7 +654,7 @@ function ff_vp8_h_loop_filter8uv\name\()_neon, export=1
         vld1.8          {d14},    [r0], r2
         vld1.8          {d15},    [r1], r2
 
-        transpose8x16matrix
+        transpose_8x8   q0,  q1,  q2,  q3,  q4,  q5,  q6,  q7
 
         vdup.8          q14, r3                 @ flim_E
         vdup.8          q15, r12                @ flim_I
@@ -681,7 +665,7 @@ function ff_vp8_h_loop_filter8uv\name\()_neon, export=1
         sub             r0,  r0,  r2, lsl #3    @ backup u 8 rows
         sub             r1,  r1,  r2, lsl #3    @ backup v 8 rows
 
-        transpose8x16matrix
+        transpose_8x8   q0,  q1,  q2,  q3,  q4,  q5,  q6,  q7
 
         @ Store pixels:
         vst1.8          {d0},     [r0], r2