Use none SSE functions when data is not 16 byte aligned

core/src/fxcodec/fx_libopenjpeg/libopenjpeg20/mct.c

 /*
  * The copyright in this software is being made available under the 2-clauses 
  * BSD License, included below. This software may be subject to other third 
  * party and contributor rights, including patent rights, and no such rights
  * are granted under this license.
  *
  * Copyright (c) 2002-2014, Universite catholique de Louvain (UCL), Belgium
  * Copyright (c) 2002-2014, Professor Benoit Macq
  * Copyright (c) 2001-2003, David Janssens
  * Copyright (c) 2002-2003, Yannick Verschueren
@@ skipping 171 matching lines @@
                         vg = _mm_sub_ps(_mm_sub_ps(vy, _mm_mul_ps(vu, vgu)), _mm_mul_ps(vv, vgv));
                         vb = _mm_add_ps(vy, _mm_mul_ps(vu, vbu));
                         _mm_store_ps(c0, vr);
                         _mm_store_ps(c1, vg);
                         _mm_store_ps(c2, vb);
                         c0 += 4;
                         c1 += 4;
                         c2 += 4;
                 }
                 n &= 7;
-        } else {

[jabdelmalek, 2014/06/06 17:25:31]

why remove this?

-                for (i = 0; i < n; ++i) {
-                        OPJ_FLOAT32 y = c0[i];
-                        OPJ_FLOAT32 u = c1[i];
-                        OPJ_FLOAT32 v = c2[i];
-                        OPJ_FLOAT32 r = y + (v * 1.402f);
-                        OPJ_FLOAT32 g = y - (u * 0.34413f) - (v * (0.71414f));
-                        OPJ_FLOAT32 b = y + (u * 1.772f);
-                        c0[i] = r;
-                        c1[i] = g;
-                        c2[i] = b;
-                }
         }
 #endif
         for(i = 0; i < n; ++i) {
                 OPJ_FLOAT32 y = c0[i];
                 OPJ_FLOAT32 u = c1[i];
                 OPJ_FLOAT32 v = c2[i];
                 OPJ_FLOAT32 r = y + (v * 1.402f);
                 OPJ_FLOAT32 g = y - (u * 0.34413f) - (v * (0.71414f));
                 OPJ_FLOAT32 b = y + (u * 1.772f);
                 c0[i] = r;
@@ skipping 118 matching lines @@
                 lIndex = i;
 
                 for     (j=0;j<pNbComps;++j) {
                         lCurrentValue = lMatrix[lIndex];
                         lIndex += pNbComps;
                         lNorms[i] += lCurrentValue * lCurrentValue;
                 }
                 lNorms[i] = sqrt(lNorms[i]);
         }
 }