arm: Mangle external symbols properly in new vfp assembly files

[ffmpeg] / libavcodec / arm / simple_idct_neon.S

diff --git a/libavcodec/arm/simple_idct_neon.S b/libavcodec/arm/simple_idct_neon.S
index 08824812facddfad7df5ad959b015c2db40ad226..a1cde8d80a13a369ed72c5459006d748f9ffaa6a 100644 (file)
--- a/libavcodec/arm/simple_idct_neon.S
+++ b/libavcodec/arm/simple_idct_neon.S
@@ -6,24 +6,24 @@
  * Based on Simple IDCT
  * Copyright (c) 2001 Michael Niedermayer <michaelni@gmx.at>
  *
- * This file is part of FFmpeg.
+ * This file is part of Libav.
  *
- * FFmpeg is free software; you can redistribute it and/or
+ * 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.
  *
- * FFmpeg is distributed in the hope that it will be useful,
+ * 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 FFmpeg; if not, write to the Free Software
+ * License along with Libav; if not, write to the Free Software
  * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
  */
 
-#include "asm.S"
+#include "libavutil/arm/asm.S"
 
 #define W1  22725  //cos(i*M_PI/16)*sqrt(2)*(1<<14) + 0.5
 #define W2  21407  //cos(i*M_PI/16)*sqrt(2)*(1<<14) + 0.5
@@ -45,8 +45,6 @@
 #define w7 d1[2]
 #define w4c d1[3]
 
-        .fpu neon
-
         .macro idct_col4_top
         vmull.s16       q7,  d6,  w2    /* q9   = W2 * col[2] */
         vmull.s16       q8,  d6,  w6    /* q10  = W6 * col[2] */
@@ -73,13 +71,13 @@ function idct_row4_pld_neon
         add             r3,  r0,  r1,  lsl #2
         pld             [r0, r1]
         pld             [r0, r1, lsl #1]
-        pld             [r3, -r1]
+A       pld             [r3, -r1]
         pld             [r3]
         pld             [r3, r1]
         add             r3,  r3,  r1,  lsl #1
         pld             [r3]
         pld             [r3, r1]
-        .endfunc
+endfunc
 
 function idct_row4_neon
         vmov.i32        q15, #(1<<(ROW_SHIFT-1))
@@ -149,7 +147,7 @@ function idct_row4_neon
         vst1.64         {d6-d9},  [r2,:128]!
 
         bx              lr
-        .endfunc
+endfunc
 
 function idct_col4_neon
         mov             ip,  #16
@@ -161,11 +159,12 @@ function idct_col4_neon
         vmull.s16       q15, d30, w4 /* q15 = W4*(col[0]+(1<<COL_SHIFT-1)/W4)*/
         vld1.64         {d8}, [r2,:64], ip /* d5 = col[3] */
 
-        ldrd            r4,  [r2]
-        ldrd            r6,  [r2, #16]
+        ldrd            r4,  r5,  [r2]
+        ldrd            r6,  r7,  [r2, #16]
         orrs            r4,  r4,  r5
 
         idct_col4_top
+        it              eq
         addeq           r2,  r2,  #16
         beq             1f
 
@@ -177,7 +176,8 @@ function idct_col4_neon
         vadd.i32        q14, q14, q7
 
 1:      orrs            r6,  r6,  r7
-        ldrd            r4,  [r2, #16]
+        ldrd            r4,  r5,  [r2, #16]
+        it              eq
         addeq           r2,  r2,  #16
         beq             2f
 
@@ -188,7 +188,8 @@ function idct_col4_neon
         vmlal.s16       q6,  d5,  w3    /* q6  += W3 * col[5] */
 
 2:      orrs            r4,  r4,  r5
-        ldrd            r4,  [r2, #16]
+        ldrd            r4,  r5,  [r2, #16]
+        it              eq
         addeq           r2,  r2,  #16
         beq             3f
 
@@ -201,6 +202,7 @@ function idct_col4_neon
         vadd.i32        q13, q13, q8
 
 3:      orrs            r4,  r4,  r5
+        it              eq
         addeq           r2,  r2,  #16
         beq             4f
 
@@ -220,7 +222,7 @@ function idct_col4_neon
         vsubhn.i32      d6,  q14, q6
 
         bx              lr
-        .endfunc
+endfunc
 
         .align 6
 
@@ -239,13 +241,11 @@ function idct_col4_st8_neon
         vst1.32         {d5[1]}, [r0,:32], r1
 
         bx              lr
-        .endfunc
+endfunc
 
-        .section .rodata
-        .align 4
-idct_coeff_neon:
+const   idct_coeff_neon, align=4
         .short W1, W2, W3, W4, W5, W6, W7, W4c
-        .previous
+endconst
 
         .macro idct_start data
         push            {r4-r7, lr}
@@ -261,7 +261,7 @@ idct_coeff_neon:
         pop             {r4-r7, pc}
         .endm
 
-/* void ff_simple_idct_put_neon(uint8_t *dst, int line_size, DCTELEM *data); */
+/* void ff_simple_idct_put_neon(uint8_t *dst, int line_size, int16_t *data); */
 function ff_simple_idct_put_neon, export=1
         idct_start      r2
 
@@ -277,7 +277,7 @@ function ff_simple_idct_put_neon, export=1
         bl              idct_col4_st8_neon
 
         idct_end
-        .endfunc
+endfunc
 
         .align 6
 
@@ -314,9 +314,9 @@ function idct_col4_add8_neon
         vst1.32         {d5[1]},  [ip,:32], r1
 
         bx              lr
-        .endfunc
+endfunc
 
-/* void ff_simple_idct_add_neon(uint8_t *dst, int line_size, DCTELEM *data); */
+/* void ff_simple_idct_add_neon(uint8_t *dst, int line_size, int16_t *data); */
 function ff_simple_idct_add_neon, export=1
         idct_start      r2
 
@@ -332,7 +332,7 @@ function ff_simple_idct_add_neon, export=1
         bl              idct_col4_add8_neon
 
         idct_end
-        .endfunc
+endfunc
 
         .align 6
 
@@ -353,9 +353,9 @@ function idct_col4_st16_neon
         vst1.64         {d9}, [r2,:64], ip
 
         bx              lr
-        .endfunc
+endfunc
 
-/* void ff_simple_idct_neon(DCTELEM *data); */
+/* void ff_simple_idct_neon(int16_t *data); */
 function ff_simple_idct_neon, export=1
         idct_start      r0
 
@@ -372,4 +372,4 @@ function ff_simple_idct_neon, export=1
         bl              idct_col4_st16_neon
 
         idct_end
-        .endfunc
+endfunc