libwebp-0.6.0-rc1

third_party/libwebp/dsp/dec_sse2.c

 // Copyright 2011 Google Inc. All Rights Reserved.
 //
 // Use of this source code is governed by a BSD-style license
 // that can be found in the COPYING 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.
 // -----------------------------------------------------------------------------
 //
 // SSE2 version of some decoding functions (idct, loop filtering).
 //
 // Author: somnath@google.com (Somnath Banerjee)
 //         cduvivier@google.com (Christian Duvivier)
 
 #include "./dsp.h"
 
 #if defined(WEBP_USE_SSE2)
 
 // The 3-coeff sparse transform in SSE2 is not really faster than the plain-C
 // one it seems => disable it by default. Uncomment the following to enable:
 // #define USE_TRANSFORM_AC3
 
 #include <emmintrin.h>
 #include "./common_sse2.h"
-#include "../dec/vp8i.h"
+#include "../dec/vp8i_dec.h"
 #include "../utils/utils.h"
 
 //------------------------------------------------------------------------------
 // Transforms (Paragraph 14.4)
 
 static void Transform(const int16_t* in, uint8_t* dst, int do_two) {
   // This implementation makes use of 16-bit fixed point versions of two
   // multiply constants:
   //    K1 = sqrt(2) * cos (pi/8) ~= 85627 / 2^16
   //    K2 = sqrt(2) * sin (pi/8) ~= 35468 / 2^16
@@ skipping 97 matching lines @@
     const __m128i tmp1 = _mm_add_epi16(b, c);
     const __m128i tmp2 = _mm_sub_epi16(b, c);
     const __m128i tmp3 = _mm_sub_epi16(a, d);
     const __m128i shifted0 = _mm_srai_epi16(tmp0, 3);
     const __m128i shifted1 = _mm_srai_epi16(tmp1, 3);
     const __m128i shifted2 = _mm_srai_epi16(tmp2, 3);
     const __m128i shifted3 = _mm_srai_epi16(tmp3, 3);
 
     // Transpose the two 4x4.
     VP8Transpose_2_4x4_16b(&shifted0, &shifted1, &shifted2, &shifted3, &T0, &T1,
-                        &T2, &T3);
+                           &T2, &T3);
   }
 
   // Add inverse transform to 'dst' and store.
   {
     const __m128i zero = _mm_setzero_si128();
     // Load the reference(s).
     __m128i dst0, dst1, dst2, dst3;
     if (do_two) {
       // Load eight bytes/pixels per line.
       dst0 = _mm_loadl_epi64((__m128i*)(dst + 0 * BPS));
@@ skipping 1068 matching lines @@
   VP8PredChroma8[4] = DC8uvNoTop;
   VP8PredChroma8[5] = DC8uvNoLeft;
   VP8PredChroma8[6] = DC8uvNoTopLeft;
 }
 
 #else  // !WEBP_USE_SSE2
 
 WEBP_DSP_INIT_STUB(VP8DspInitSSE2)
 
 #endif  // WEBP_USE_SSE2