libjpeg-turbo: Remove .func / .endfunc directives.

simd/jsimd_arm_neon.S

 /*
  * ARMv7 NEON optimizations for libjpeg-turbo
  *
  * Copyright (C) 2009-2011 Nokia Corporation and/or its subsidiary(-ies).
  * All rights reserved.
  * Author: Siarhei Siamashka <siarhei.siamashka@nokia.com>
  *
  * This software is provided 'as-is', without any express or implied
  * warranty.  In no event will the authors be held liable for any damages
  * arising from the use of this software.
@@ skipping 23 matching lines @@
 
 
 #define RESPECT_STRICT_ALIGNMENT 1
 
 
 /*****************************************************************************/
 
 /* Supplementary macro for setting function attributes */
 .macro asm_function fname
 #ifdef __APPLE__
-    .func _\fname
     .globl _\fname
 _\fname:
 #else
-    .func \fname
     .global \fname
 #ifdef __ELF__
     .hidden \fname
     .type \fname, %function
 #endif
 \fname:
 #endif
 .endm
 
 /* Transpose a block of 4x4 coefficients in four 64-bit registers */
@@ skipping 604 matching lines @@
     .unreq          ROW3L
     .unreq          ROW3R
     .unreq          ROW4L
     .unreq          ROW4R
     .unreq          ROW5L
     .unreq          ROW5R
     .unreq          ROW6L
     .unreq          ROW6R
     .unreq          ROW7L
     .unreq          ROW7R
-.endfunc
 
 
 /*****************************************************************************/
 
 /*
  * jsimd_idct_ifast_neon
  *
  * This function contains a fast, not so accurate integer implementation of
  * the inverse DCT (Discrete Cosine Transform). It uses the same calculations
  * and produces exactly the same output as IJG's original 'jpeg_idct_ifast'
@@ skipping 204 matching lines @@
     bx              lr
 
     .unreq          DCT_TABLE
     .unreq          COEF_BLOCK
     .unreq          OUTPUT_BUF
     .unreq          OUTPUT_COL
     .unreq          TMP1
     .unreq          TMP2
     .unreq          TMP3
     .unreq          TMP4
-.endfunc
 
 
 /*****************************************************************************/
 
 /*
  * jsimd_idct_4x4_neon
  *
  * This function contains inverse-DCT code for getting reduced-size
  * 4x4 pixels output from an 8x8 DCT block. It uses the same  calculations
  * and produces exactly the same output as IJG's original 'jpeg_idct_4x4'
@@ skipping 192 matching lines @@
     bx              lr
 
     .unreq          DCT_TABLE
     .unreq          COEF_BLOCK
     .unreq          OUTPUT_BUF
     .unreq          OUTPUT_COL
     .unreq          TMP1
     .unreq          TMP2
     .unreq          TMP3
     .unreq          TMP4
-.endfunc
 
 .purgem idct_helper
 
 
 /*****************************************************************************/
 
 /*
  * jsimd_idct_2x2_neon
  *
  * This function contains inverse-DCT code for getting reduced-size
@@ skipping 134 matching lines @@
 
     vpop            {d8-d15}
     bx              lr
 
     .unreq          DCT_TABLE
     .unreq          COEF_BLOCK
     .unreq          OUTPUT_BUF
     .unreq          OUTPUT_COL
     .unreq          TMP1
     .unreq          TMP2
-.endfunc
 
 .purgem idct_helper
 
 
 /*****************************************************************************/
 
 /*
  * jsimd_ycc_extrgb_convert_neon
  * jsimd_ycc_extbgr_convert_neon
  * jsimd_ycc_extrgbx_convert_neon
@@ skipping 263 matching lines @@
     .unreq          OUTPUT_BUF
     .unreq          NUM_ROWS
     .unreq          INPUT_BUF0
     .unreq          INPUT_BUF1
     .unreq          INPUT_BUF2
     .unreq          RGB
     .unreq          Y
     .unreq          U
     .unreq          V
     .unreq          N
-.endfunc
 
 .purgem do_yuv_to_rgb
 .purgem do_yuv_to_rgb_stage1
 .purgem do_yuv_to_rgb_stage2
 .purgem do_yuv_to_rgb_stage2_store_load_stage1
 
 .endm
 
 /*--------------------------------- id ----- bpp R  G  B */
 generate_jsimd_ycc_rgb_convert_neon extrgb,  24, 0, 1, 2
@@ skipping 290 matching lines @@
     .unreq          INPUT_BUF
     .unreq          NUM_ROWS
     .unreq          OUTPUT_BUF0
     .unreq          OUTPUT_BUF1
     .unreq          OUTPUT_BUF2
     .unreq          RGB
     .unreq          Y
     .unreq          U
     .unreq          V
     .unreq          N
-.endfunc
 
 .purgem do_rgb_to_yuv
 .purgem do_rgb_to_yuv_stage1
 .purgem do_rgb_to_yuv_stage2
 .purgem do_rgb_to_yuv_stage2_store_load_stage1
 
 .endm
 
 /*--------------------------------- id ----- bpp R  G  B */
 generate_jsimd_rgb_ycc_convert_neon extrgb,  24, 0, 1, 2
@@ skipping 61 matching lines @@
     pop             {r4, r5}
     bx              lr
 
     .unreq          SAMPLE_DATA
     .unreq          START_COL
     .unreq          WORKSPACE
     .unreq          TMP1
     .unreq          TMP2
     .unreq          TMP3
     .unreq          TMP4
-.endfunc
 
 
 /*****************************************************************************/
 
 /*
  * jsimd_fdct_ifast_neon
  *
  * This function contains a fast, not so accurate integer implementation of
  * the forward DCT (Discrete Cosine Transform). It uses the same calculations
  * and produces exactly the same output as IJG's original 'jpeg_fdct_ifast'
@@ skipping 103 matching lines @@
     vst1.16         {d16, d17, d18, d19}, [DATA, :128]!
     vst1.16         {d20, d21, d22, d23}, [DATA, :128]!
     vst1.16         {d24, d25, d26, d27}, [DATA, :128]!
     vst1.16         {d28, d29, d30, d31}, [DATA, :128]
 
     vpop            {d8-d15}
     bx              lr
 
     .unreq          DATA
     .unreq          TMP
-.endfunc
 
 
 /*****************************************************************************/
 
 /*
  * GLOBAL(void)
  * jsimd_quantize_neon (JCOEFPTR coef_block, DCTELEM * divisors,
  *                      DCTELEM * workspace);
  *
  * Note: the code uses 2 stage pipelining in order to improve instructions
@@ skipping 81 matching lines @@
 
     bx              lr /* return */
 
     .unreq          COEF_BLOCK
     .unreq          DIVISORS
     .unreq          WORKSPACE
     .unreq          RECIPROCAL
     .unreq          CORRECTION
     .unreq          SHIFT
     .unreq          LOOP_COUNT
-.endfunc
 
 
 /*****************************************************************************/
 
 /*
  * GLOBAL(void)
  * jsimd_h2v1_fancy_upsample_neon (int          max_v_samp_factor,
  *                                 JDIMENSION   downsampled_width,
  *                                 JSAMPARRAY   input_data,
  *                                 JSAMPARRAY * output_data_ptr);
@@ skipping 214 matching lines @@
     .unreq          DOWNSAMPLED_WIDTH
     .unreq          INPUT_DATA
     .unreq          OUTPUT_DATA_PTR
     .unreq          OUTPUT_DATA
 
     .unreq          OUTPTR
     .unreq          INPTR
     .unreq          WIDTH
     .unreq          TMP
 
-.endfunc
 
 .purgem upsample16
 .purgem upsample32
 .purgem upsample_row