COSMETICS: Remove all trailing whitespace.

Originally committed as revision 4749 to svn://svn.ffmpeg.org/ffmpeg/trunk

14 files changed, 397 insertions, 397 deletions

diff --git a/libavcodec/i386/cputest.c b/libavcodec/i386/cputest.c
index 593e0550db..f02c63d449 100644
--- a/libavcodec/i386/cputest.c
+++ b/libavcodec/i386/cputest.c
@@ -29,28 +29,28 @@ int mm_support(void)
     int eax, ebx, ecx, edx;
     int max_std_level, max_ext_level, std_caps=0, ext_caps=0;
     long a, c;
-    
+
     __asm__ __volatile__ (
                           /* See if CPUID instruction is supported ... */
                           /* ... Get copies of EFLAGS into eax and ecx */
                           "pushf\n\t"
                           "pop %0\n\t"
                           "mov %0, %1\n\t"
-                          
+
                           /* ... Toggle the ID bit in one copy and store */
                           /*     to the EFLAGS reg */
                           "xor $0x200000, %0\n\t"
                           "push %0\n\t"
                           "popf\n\t"
-                          
+
                           /* ... Get the (hopefully modified) EFLAGS */
                           "pushf\n\t"
                           "pop %0\n\t"
                           : "=a" (a), "=c" (c)
                           :
-                          : "cc" 
+                          : "cc"
                           );
-    
+
     if (a == c)
         return 0; /* CPUID not supported */
 
@@ -60,9 +60,9 @@ int mm_support(void)
         cpuid(1, eax, ebx, ecx, std_caps);
         if (std_caps & (1<<23))
             rval |= MM_MMX;
-        if (std_caps & (1<<25)) 
+        if (std_caps & (1<<25))
             rval |= MM_MMXEXT | MM_SSE;
-        if (std_caps & (1<<26)) 
+        if (std_caps & (1<<26))
             rval |= MM_SSE2;
     }
 
@@ -103,18 +103,18 @@ int mm_support(void)
            According to the table, the only CPU which supports level
            2 is also the only one which supports extended CPUID levels.
         */
-        if (eax < 2) 
+        if (eax < 2)
             return rval;
         if (ext_caps & (1<<24))
             rval |= MM_MMXEXT;
     }
 #if 0
-    av_log(NULL, AV_LOG_DEBUG, "%s%s%s%s%s%s\n", 
-        (rval&MM_MMX) ? "MMX ":"", 
-        (rval&MM_MMXEXT) ? "MMX2 ":"", 
-        (rval&MM_SSE) ? "SSE ":"", 
-        (rval&MM_SSE2) ? "SSE2 ":"", 
-        (rval&MM_3DNOW) ? "3DNow ":"", 
+    av_log(NULL, AV_LOG_DEBUG, "%s%s%s%s%s%s\n",
+        (rval&MM_MMX) ? "MMX ":"",
+        (rval&MM_MMXEXT) ? "MMX2 ":"",
+        (rval&MM_SSE) ? "SSE ":"",
+        (rval&MM_SSE2) ? "SSE2 ":"",
+        (rval&MM_3DNOW) ? "3DNow ":"",
         (rval&MM_3DNOWEXT) ? "3DNowExt ":"");
 #endif
     return rval;
diff --git a/libavcodec/i386/dsputil_mmx.c b/libavcodec/i386/dsputil_mmx.c
index d8e655269b..7566b5d16a 100644
--- a/libavcodec/i386/dsputil_mmx.c
+++ b/libavcodec/i386/dsputil_mmx.c
@@ -602,9 +602,9 @@ static void h263_v_loop_filter_mmx(uint8_t *src, int stride, int qscale){
     const int strength= ff_h263_loop_filter_strength[qscale];
 
     asm volatile(
-    
+
         H263_LOOP_FILTER
-        
+
         "movq %%mm3, %1			\n\t"
         "movq %%mm4, %2			\n\t"
         "movq %%mm5, %0			\n\t"
@@ -634,7 +634,7 @@ static inline void transpose4x4(uint8_t *dst, uint8_t *src, int dst_stride, int
         "movd  %%mm1, %2		\n\t"
         "punpckhdq %%mm1, %%mm1		\n\t"
         "movd  %%mm1, %3		\n\t"
-        
+
         : "=m" (*(uint32_t*)(dst + 0*dst_stride)),
           "=m" (*(uint32_t*)(dst + 1*dst_stride)),
           "=m" (*(uint32_t*)(dst + 2*dst_stride)),
@@ -650,14 +650,14 @@ static void h263_h_loop_filter_mmx(uint8_t *src, int stride, int qscale){
     const int strength= ff_h263_loop_filter_strength[qscale];
     uint64_t temp[4] __attribute__ ((aligned(8)));
     uint8_t *btemp= (uint8_t*)temp;
-    
+
     src -= 2;
 
     transpose4x4(btemp  , src           , 8, stride);
     transpose4x4(btemp+4, src + 4*stride, 8, stride);
     asm volatile(
         H263_LOOP_FILTER // 5 3 4 6
-        
+
         : "+m" (temp[0]),
           "+m" (temp[1]),
           "+m" (temp[2]),
@@ -796,7 +796,7 @@ static int sse8_mmx(void *v, uint8_t * pix1, uint8_t * pix2, int line_size, int
       "psrlq $32, %%mm7\n"	/* shift hi dword to lo */
       "paddd %%mm7,%%mm1\n"
       "movd %%mm1,%2\n"
-      : "+r" (pix1), "+r" (pix2), "=r"(tmp) 
+      : "+r" (pix1), "+r" (pix2), "=r"(tmp)
       : "r" ((long)line_size) , "m" (h)
       : "%ecx");
     return tmp;
@@ -856,7 +856,7 @@ static int sse16_mmx(void *v, uint8_t * pix1, uint8_t * pix2, int line_size, int
       "psrlq $32, %%mm7\n"	/* shift hi dword to lo */
       "paddd %%mm7,%%mm1\n"
       "movd %%mm1,%2\n"
-      : "+r" (pix1), "+r" (pix2), "=r"(tmp) 
+      : "+r" (pix1), "+r" (pix2), "=r"(tmp)
       : "r" ((long)line_size) , "m" (h)
       : "%ecx");
     return tmp;
@@ -919,7 +919,7 @@ static int sse16_sse2(void *v, uint8_t * pix1, uint8_t * pix2, int line_size, in
       "psrldq $4, %%xmm7\n"	/* shift hi dword to lo */
       "paddd %%xmm1,%%xmm7\n"
       "movd %%xmm7,%3\n"
-      : "+r" (pix1), "+r" (pix2), "+r"(h), "=r"(tmp) 
+      : "+r" (pix1), "+r" (pix2), "+r"(h), "=r"(tmp)
       : "r" ((long)line_size));
     return tmp;
 }
@@ -930,7 +930,7 @@ static int hf_noise8_mmx(uint8_t * pix1, int line_size, int h) {
       "movl %3,%%ecx\n"
       "pxor %%mm7,%%mm7\n"
       "pxor %%mm6,%%mm6\n"
-      
+
       "movq (%0),%%mm0\n"
       "movq %%mm0, %%mm1\n"
       "psllq $8, %%mm0\n"
@@ -944,9 +944,9 @@ static int hf_noise8_mmx(uint8_t * pix1, int line_size, int h) {
       "punpckhbw %%mm7,%%mm3\n"
       "psubw %%mm1, %%mm0\n"
       "psubw %%mm3, %%mm2\n"
-      
+
       "add %2,%0\n"
-      
+
       "movq (%0),%%mm4\n"
       "movq %%mm4, %%mm1\n"
       "psllq $8, %%mm4\n"
@@ -968,14 +968,14 @@ static int hf_noise8_mmx(uint8_t * pix1, int line_size, int h) {
       "pcmpgtw %%mm2, %%mm1\n\t"
       "pxor %%mm3, %%mm0\n"
       "pxor %%mm1, %%mm2\n"
-      "psubw %%mm3, %%mm0\n" 
+      "psubw %%mm3, %%mm0\n"
       "psubw %%mm1, %%mm2\n"
       "paddw %%mm0, %%mm2\n"
       "paddw %%mm2, %%mm6\n"
 
       "add %2,%0\n"
       "1:\n"
-  
+
       "movq (%0),%%mm0\n"
       "movq %%mm0, %%mm1\n"
       "psllq $8, %%mm0\n"
@@ -997,13 +997,13 @@ static int hf_noise8_mmx(uint8_t * pix1, int line_size, int h) {
       "pcmpgtw %%mm5, %%mm1\n\t"
       "pxor %%mm3, %%mm4\n"
       "pxor %%mm1, %%mm5\n"
-      "psubw %%mm3, %%mm4\n" 
+      "psubw %%mm3, %%mm4\n"
       "psubw %%mm1, %%mm5\n"
       "paddw %%mm4, %%mm5\n"
       "paddw %%mm5, %%mm6\n"
-      
+
       "add %2,%0\n"
-      
+
       "movq (%0),%%mm4\n"
       "movq %%mm4, %%mm1\n"
       "psllq $8, %%mm4\n"
@@ -1025,7 +1025,7 @@ static int hf_noise8_mmx(uint8_t * pix1, int line_size, int h) {
       "pcmpgtw %%mm2, %%mm1\n\t"
       "pxor %%mm3, %%mm0\n"
       "pxor %%mm1, %%mm2\n"
-      "psubw %%mm3, %%mm0\n" 
+      "psubw %%mm3, %%mm0\n"
       "psubw %%mm1, %%mm2\n"
       "paddw %%mm0, %%mm2\n"
       "paddw %%mm2, %%mm6\n"
@@ -1038,12 +1038,12 @@ static int hf_noise8_mmx(uint8_t * pix1, int line_size, int h) {
       "punpcklwd %%mm7,%%mm0\n"
       "punpckhwd %%mm7,%%mm6\n"
       "paddd %%mm0, %%mm6\n"
-      
+
       "movq %%mm6,%%mm0\n"
       "psrlq $32, %%mm6\n"
       "paddd %%mm6,%%mm0\n"
       "movd %%mm0,%1\n"
-      : "+r" (pix1), "=r"(tmp) 
+      : "+r" (pix1), "=r"(tmp)
       : "r" ((long)line_size) , "g" (h-2)
       : "%ecx");
       return tmp;
@@ -1056,7 +1056,7 @@ static int hf_noise16_mmx(uint8_t * pix1, int line_size, int h) {
       "movl %3,%%ecx\n"
       "pxor %%mm7,%%mm7\n"
       "pxor %%mm6,%%mm6\n"
-      
+
       "movq (%0),%%mm0\n"
       "movq 1(%0),%%mm1\n"
       "movq %%mm0, %%mm2\n"
@@ -1067,9 +1067,9 @@ static int hf_noise16_mmx(uint8_t * pix1, int line_size, int h) {
       "punpckhbw %%mm7,%%mm3\n"
       "psubw %%mm1, %%mm0\n"
       "psubw %%mm3, %%mm2\n"
-      
+
       "add %2,%0\n"
-      
+
       "movq (%0),%%mm4\n"
       "movq 1(%0),%%mm1\n"
       "movq %%mm4, %%mm5\n"
@@ -1088,14 +1088,14 @@ static int hf_noise16_mmx(uint8_t * pix1, int line_size, int h) {
       "pcmpgtw %%mm2, %%mm1\n\t"
       "pxor %%mm3, %%mm0\n"
       "pxor %%mm1, %%mm2\n"
-      "psubw %%mm3, %%mm0\n" 
+      "psubw %%mm3, %%mm0\n"
       "psubw %%mm1, %%mm2\n"
       "paddw %%mm0, %%mm2\n"
       "paddw %%mm2, %%mm6\n"
 
       "add %2,%0\n"
       "1:\n"
-  
+
       "movq (%0),%%mm0\n"
       "movq 1(%0),%%mm1\n"
       "movq %%mm0, %%mm2\n"
@@ -1118,9 +1118,9 @@ static int hf_noise16_mmx(uint8_t * pix1, int line_size, int h) {
       "psubw %%mm1, %%mm5\n"
       "paddw %%mm4, %%mm5\n"
       "paddw %%mm5, %%mm6\n"
-      
+
       "add %2,%0\n"
-      
+
       "movq (%0),%%mm4\n"
       "movq 1(%0),%%mm1\n"
       "movq %%mm4, %%mm5\n"
@@ -1139,7 +1139,7 @@ static int hf_noise16_mmx(uint8_t * pix1, int line_size, int h) {
       "pcmpgtw %%mm2, %%mm1\n\t"
       "pxor %%mm3, %%mm0\n"
       "pxor %%mm1, %%mm2\n"
-      "psubw %%mm3, %%mm0\n" 
+      "psubw %%mm3, %%mm0\n"
       "psubw %%mm1, %%mm2\n"
       "paddw %%mm0, %%mm2\n"
       "paddw %%mm2, %%mm6\n"
@@ -1152,12 +1152,12 @@ static int hf_noise16_mmx(uint8_t * pix1, int line_size, int h) {
       "punpcklwd %%mm7,%%mm0\n"
       "punpckhwd %%mm7,%%mm6\n"
       "paddd %%mm0, %%mm6\n"
-      
+
       "movq %%mm6,%%mm0\n"
       "psrlq $32, %%mm6\n"
       "paddd %%mm6,%%mm0\n"
       "movd %%mm0,%1\n"
-      : "+r" (pix1), "=r"(tmp) 
+      : "+r" (pix1), "=r"(tmp)
       : "r" ((long)line_size) , "g" (h-2)
       : "%ecx");
       return tmp + hf_noise8_mmx(pix+8, line_size, h);
@@ -1186,10 +1186,10 @@ static int nsse8_mmx(void *p, uint8_t * pix1, uint8_t * pix2, int line_size, int
 
 static int vsad_intra16_mmx(void *v, uint8_t * pix, uint8_t * dummy, int line_size, int h) {
     int tmp;
-    
+
     assert( (((int)pix) & 7) == 0);
     assert((line_size &7) ==0);
-    
+
 #define SUM(in0, in1, out0, out1) \
       "movq (%0), %%mm2\n"\
       "movq 8(%0), %%mm3\n"\
@@ -1213,7 +1213,7 @@ static int vsad_intra16_mmx(void *v, uint8_t * pix, uint8_t * dummy, int line_si
       "paddw %%mm2, " #in0 "\n"\
       "paddw " #in0 ", %%mm6\n"
 
-    
+
   asm volatile (
       "movl %3,%%ecx\n"
       "pxor %%mm6,%%mm6\n"
@@ -1224,11 +1224,11 @@ static int vsad_intra16_mmx(void *v, uint8_t * pix, uint8_t * dummy, int line_si
       "subl $2, %%ecx\n"
       SUM(%%mm0, %%mm1, %%mm4, %%mm5)
       "1:\n"
-      
+
       SUM(%%mm4, %%mm5, %%mm0, %%mm1)
-      
+
       SUM(%%mm0, %%mm1, %%mm4, %%mm5)
-      
+
       "subl $2, %%ecx\n"
       "jnz 1b\n"
 
@@ -1239,7 +1239,7 @@ static int vsad_intra16_mmx(void *v, uint8_t * pix, uint8_t * dummy, int line_si
       "psrlq $16, %%mm0\n"
       "paddw %%mm6,%%mm0\n"
       "movd %%mm0,%1\n"
-      : "+r" (pix), "=r"(tmp) 
+      : "+r" (pix), "=r"(tmp)
       : "r" ((long)line_size) , "m" (h)
       : "%ecx");
     return tmp & 0xFFFF;
@@ -1248,10 +1248,10 @@ static int vsad_intra16_mmx(void *v, uint8_t * pix, uint8_t * dummy, int line_si
 
 static int vsad_intra16_mmx2(void *v, uint8_t * pix, uint8_t * dummy, int line_size, int h) {
     int tmp;
-    
+
     assert( (((int)pix) & 7) == 0);
     assert((line_size &7) ==0);
-    
+
 #define SUM(in0, in1, out0, out1) \
       "movq (%0), " #out0 "\n"\
       "movq 8(%0), " #out1 "\n"\
@@ -1271,16 +1271,16 @@ static int vsad_intra16_mmx2(void *v, uint8_t * pix, uint8_t * dummy, int line_s
       "subl $2, %%ecx\n"
       SUM(%%mm0, %%mm1, %%mm4, %%mm5)
       "1:\n"
-      
+
       SUM(%%mm4, %%mm5, %%mm0, %%mm1)
-      
+
       SUM(%%mm0, %%mm1, %%mm4, %%mm5)
-      
+
       "subl $2, %%ecx\n"
       "jnz 1b\n"
 
       "movd %%mm6,%1\n"
-      : "+r" (pix), "=r"(tmp) 
+      : "+r" (pix), "=r"(tmp)
       : "r" ((long)line_size) , "m" (h)
       : "%ecx");
     return tmp;
@@ -1289,11 +1289,11 @@ static int vsad_intra16_mmx2(void *v, uint8_t * pix, uint8_t * dummy, int line_s
 
 static int vsad16_mmx(void *v, uint8_t * pix1, uint8_t * pix2, int line_size, int h) {
     int tmp;
-    
+
     assert( (((int)pix1) & 7) == 0);
     assert( (((int)pix2) & 7) == 0);
     assert((line_size &7) ==0);
-    
+
 #define SUM(in0, in1, out0, out1) \
       "movq (%0),%%mm2\n"\
       "movq (%1)," #out0 "\n"\
@@ -1324,7 +1324,7 @@ static int vsad16_mmx(void *v, uint8_t * pix1, uint8_t * pix2, int line_size, in
       "paddw %%mm2, " #in0 "\n"\
       "paddw " #in0 ", %%mm6\n"
 
-    
+
   asm volatile (
       "movl %4,%%ecx\n"
       "pxor %%mm6,%%mm6\n"
@@ -1344,11 +1344,11 @@ static int vsad16_mmx(void *v, uint8_t * pix1, uint8_t * pix2, int line_size, in
       "pxor %%mm7, %%mm1\n"
       SUM(%%mm0, %%mm1, %%mm4, %%mm5)
       "1:\n"
-      
+
       SUM(%%mm4, %%mm5, %%mm0, %%mm1)
-      
+
       SUM(%%mm0, %%mm1, %%mm4, %%mm5)
-      
+
       "subl $2, %%ecx\n"
       "jnz 1b\n"
 
@@ -1359,7 +1359,7 @@ static int vsad16_mmx(void *v, uint8_t * pix1, uint8_t * pix2, int line_size, in
       "psrlq $16, %%mm0\n"
       "paddw %%mm6,%%mm0\n"
       "movd %%mm0,%2\n"
-      : "+r" (pix1), "+r" (pix2), "=r"(tmp) 
+      : "+r" (pix1), "+r" (pix2), "=r"(tmp)
       : "r" ((long)line_size) , "m" (h)
       : "%ecx");
     return tmp & 0x7FFF;
@@ -1368,11 +1368,11 @@ static int vsad16_mmx(void *v, uint8_t * pix1, uint8_t * pix2, int line_size, in
 
 static int vsad16_mmx2(void *v, uint8_t * pix1, uint8_t * pix2, int line_size, int h) {
     int tmp;
-    
+
     assert( (((int)pix1) & 7) == 0);
     assert( (((int)pix2) & 7) == 0);
     assert((line_size &7) ==0);
-    
+
 #define SUM(in0, in1, out0, out1) \
       "movq (%0)," #out0 "\n"\
       "movq (%1),%%mm2\n"\
@@ -1408,16 +1408,16 @@ static int vsad16_mmx2(void *v, uint8_t * pix1, uint8_t * pix2, int line_size, i
       "pxor %%mm7, %%mm1\n"
       SUM(%%mm0, %%mm1, %%mm4, %%mm5)
       "1:\n"
-      
+
       SUM(%%mm4, %%mm5, %%mm0, %%mm1)
-      
+
       SUM(%%mm0, %%mm1, %%mm4, %%mm5)
-      
+
       "subl $2, %%ecx\n"
       "jnz 1b\n"
 
       "movd %%mm6,%2\n"
-      : "+r" (pix1), "+r" (pix2), "=r"(tmp) 
+      : "+r" (pix1), "+r" (pix2), "=r"(tmp)
       : "r" ((long)line_size) , "m" (h)
       : "%ecx");
     return tmp;
@@ -1449,7 +1449,7 @@ static void diff_bytes_mmx(uint8_t *dst, uint8_t *src1, uint8_t *src2, int w){
 static void sub_hfyu_median_prediction_mmx2(uint8_t *dst, uint8_t *src1, uint8_t *src2, int w, int *left, int *left_top){
     long i=0;
     uint8_t l, lt;
-    
+
     asm volatile(
         "1:				\n\t"
         "movq  -1(%1, %0), %%mm0	\n\t" // LT
@@ -1462,7 +1462,7 @@ static void sub_hfyu_median_prediction_mmx2(uint8_t *dst, uint8_t *src1, uint8_t
         "movq %%mm4, %%mm5		\n\t" // L
         "pmaxub %%mm1, %%mm4		\n\t" // max(T, L)
         "pminub %%mm5, %%mm1		\n\t" // min(T, L)
-        "pminub %%mm2, %%mm4		\n\t" 
+        "pminub %%mm2, %%mm4		\n\t"
         "pmaxub %%mm1, %%mm4		\n\t"
         "psubb %%mm4, %%mm3		\n\t" // dst - pred
         "movq %%mm3, (%3, %0)		\n\t"
@@ -1475,9 +1475,9 @@ static void sub_hfyu_median_prediction_mmx2(uint8_t *dst, uint8_t *src1, uint8_t
 
     l= *left;
     lt= *left_top;
-    
+
     dst[0]= src2[0] - mid_pred(l, src1[0], (l + src1[0] - lt)&0xFF);
-    
+
     *left_top= src1[w-1];
     *left    = src2[w-1];
 }
@@ -1521,7 +1521,7 @@ static void sub_hfyu_median_prediction_mmx2(uint8_t *dst, uint8_t *src1, uint8_t
     "psubw " #a ", " #z "		\n\t"\
     "pmaxsw " #z ", " #a "		\n\t"\
     "paddusw " #a ", " #sum "		\n\t"
-        
+
 #define SBUTTERFLY(a,b,t,n)\
     "movq " #a ", " #t "		\n\t" /* abcd */\
     "punpckl" #n " " #b ", " #a "	\n\t" /* aebf */\
@@ -1548,7 +1548,7 @@ static void sub_hfyu_median_prediction_mmx2(uint8_t *dst, uint8_t *src1, uint8_t
 static int hadamard8_diff_mmx(void *s, uint8_t *src1, uint8_t *src2, int stride, int h){
     uint64_t temp[16] __align8;
     int sum=0;
-    
+
     assert(h==8);
 
     diff_pixels_mmx((DCTELEM*)temp, src1, src2, stride);
@@ -1556,38 +1556,38 @@ static int hadamard8_diff_mmx(void *s, uint8_t *src1, uint8_t *src2, int stride,
     asm volatile(
         LOAD4(0 , %%mm0, %%mm1, %%mm2, %%mm3)
         LOAD4(64, %%mm4, %%mm5, %%mm6, %%mm7)
-        
+
         HADAMARD48
-        
+
         "movq %%mm7, 112(%1)		\n\t"
-        
+
         TRANSPOSE4(%%mm0, %%mm1, %%mm2, %%mm3, %%mm7)
         STORE4(0 , %%mm0, %%mm3, %%mm7, %%mm2)
-        
+
         "movq 112(%1), %%mm7 		\n\t"
         TRANSPOSE4(%%mm4, %%mm5, %%mm6, %%mm7, %%mm0)
         STORE4(64, %%mm4, %%mm7, %%mm0, %%mm6)
 
         LOAD4(8 , %%mm0, %%mm1, %%mm2, %%mm3)
         LOAD4(72, %%mm4, %%mm5, %%mm6, %%mm7)
-        
+
         HADAMARD48
-        
+
         "movq %%mm7, 120(%1)		\n\t"
-        
+
         TRANSPOSE4(%%mm0, %%mm1, %%mm2, %%mm3, %%mm7)
         STORE4(8 , %%mm0, %%mm3, %%mm7, %%mm2)
-        
+
         "movq 120(%1), %%mm7 		\n\t"
         TRANSPOSE4(%%mm4, %%mm5, %%mm6, %%mm7, %%mm0)
         "movq %%mm7, %%mm5		\n\t"//FIXME remove
         "movq %%mm6, %%mm7		\n\t"
         "movq %%mm0, %%mm6		\n\t"
 //        STORE4(72, %%mm4, %%mm7, %%mm0, %%mm6) //FIXME remove
-        
+
         LOAD4(64, %%mm0, %%mm1, %%mm2, %%mm3)
 //        LOAD4(72, %%mm4, %%mm5, %%mm6, %%mm7)
-        
+
         HADAMARD48
         "movq %%mm7, 64(%1)		\n\t"
         MMABS(%%mm0, %%mm7)
@@ -1600,10 +1600,10 @@ static int hadamard8_diff_mmx(void *s, uint8_t *src1, uint8_t *src2, int stride,
         "movq 64(%1), %%mm1		\n\t"
         MMABS_SUM(%%mm1, %%mm7, %%mm0)
         "movq %%mm0, 64(%1)		\n\t"
-        
+
         LOAD4(0 , %%mm0, %%mm1, %%mm2, %%mm3)
         LOAD4(8 , %%mm4, %%mm5, %%mm6, %%mm7)
-        
+
         HADAMARD48
         "movq %%mm7, (%1)		\n\t"
         MMABS(%%mm0, %%mm7)
@@ -1617,7 +1617,7 @@ static int hadamard8_diff_mmx(void *s, uint8_t *src1, uint8_t *src2, int stride,
         MMABS_SUM(%%mm1, %%mm7, %%mm0)
         "movq 64(%1), %%mm1		\n\t"
         MMABS_SUM(%%mm1, %%mm7, %%mm0)
-        
+
         "movq %%mm0, %%mm1		\n\t"
         "psrlq $32, %%mm0		\n\t"
         "paddusw %%mm1, %%mm0		\n\t"
@@ -1625,7 +1625,7 @@ static int hadamard8_diff_mmx(void *s, uint8_t *src1, uint8_t *src2, int stride,
         "psrlq $16, %%mm0		\n\t"
         "paddusw %%mm1, %%mm0		\n\t"
         "movd %%mm0, %0			\n\t"
-                
+
         : "=r" (sum)
         : "r"(temp)
     );
@@ -1635,7 +1635,7 @@ static int hadamard8_diff_mmx(void *s, uint8_t *src1, uint8_t *src2, int stride,
 static int hadamard8_diff_mmx2(void *s, uint8_t *src1, uint8_t *src2, int stride, int h){
     uint64_t temp[16] __align8;
     int sum=0;
-    
+
     assert(h==8);
 
     diff_pixels_mmx((DCTELEM*)temp, src1, src2, stride);
@@ -1643,38 +1643,38 @@ static int hadamard8_diff_mmx2(void *s, uint8_t *src1, uint8_t *src2, int stride
     asm volatile(
         LOAD4(0 , %%mm0, %%mm1, %%mm2, %%mm3)
         LOAD4(64, %%mm4, %%mm5, %%mm6, %%mm7)
-        
+
         HADAMARD48
-        
+
         "movq %%mm7, 112(%1)		\n\t"
-        
+
         TRANSPOSE4(%%mm0, %%mm1, %%mm2, %%mm3, %%mm7)
         STORE4(0 , %%mm0, %%mm3, %%mm7, %%mm2)
-        
+
         "movq 112(%1), %%mm7 		\n\t"
         TRANSPOSE4(%%mm4, %%mm5, %%mm6, %%mm7, %%mm0)
         STORE4(64, %%mm4, %%mm7, %%mm0, %%mm6)
 
         LOAD4(8 , %%mm0, %%mm1, %%mm2, %%mm3)
         LOAD4(72, %%mm4, %%mm5, %%mm6, %%mm7)
-        
+
         HADAMARD48
-        
+
         "movq %%mm7, 120(%1)		\n\t"
-        
+
         TRANSPOSE4(%%mm0, %%mm1, %%mm2, %%mm3, %%mm7)
         STORE4(8 , %%mm0, %%mm3, %%mm7, %%mm2)
-        
+
         "movq 120(%1), %%mm7 		\n\t"
         TRANSPOSE4(%%mm4, %%mm5, %%mm6, %%mm7, %%mm0)
         "movq %%mm7, %%mm5		\n\t"//FIXME remove
         "movq %%mm6, %%mm7		\n\t"
         "movq %%mm0, %%mm6		\n\t"
 //        STORE4(72, %%mm4, %%mm7, %%mm0, %%mm6) //FIXME remove
-        
+
         LOAD4(64, %%mm0, %%mm1, %%mm2, %%mm3)
 //        LOAD4(72, %%mm4, %%mm5, %%mm6, %%mm7)
-        
+
         HADAMARD48
         "movq %%mm7, 64(%1)		\n\t"
         MMABS_MMX2(%%mm0, %%mm7)
@@ -1687,10 +1687,10 @@ static int hadamard8_diff_mmx2(void *s, uint8_t *src1, uint8_t *src2, int stride
         "movq 64(%1), %%mm1		\n\t"
         MMABS_SUM_MMX2(%%mm1, %%mm7, %%mm0)
         "movq %%mm0, 64(%1)		\n\t"
-        
+
         LOAD4(0 , %%mm0, %%mm1, %%mm2, %%mm3)
         LOAD4(8 , %%mm4, %%mm5, %%mm6, %%mm7)
-        
+
         HADAMARD48
         "movq %%mm7, (%1)		\n\t"
         MMABS_MMX2(%%mm0, %%mm7)
@@ -1704,13 +1704,13 @@ static int hadamard8_diff_mmx2(void *s, uint8_t *src1, uint8_t *src2, int stride
         MMABS_SUM_MMX2(%%mm1, %%mm7, %%mm0)
         "movq 64(%1), %%mm1		\n\t"
         MMABS_SUM_MMX2(%%mm1, %%mm7, %%mm0)
-        
+
         "pshufw $0x0E, %%mm0, %%mm1     \n\t"
         "paddusw %%mm1, %%mm0		\n\t"
         "pshufw $0x01, %%mm0, %%mm1     \n\t"
         "paddusw %%mm1, %%mm0		\n\t"
         "movd %%mm0, %0			\n\t"
-                
+
         : "=r" (sum)
         : "r"(temp)
     );
@@ -2405,7 +2405,7 @@ static void just_return() { return; }
 
 static int try_8x8basis_mmx(int16_t rem[64], int16_t weight[64], int16_t basis[64], int scale){
     long i=0;
-    
+
     assert(ABS(scale) < 256);
     scale<<= 16 + 1 - BASIS_SHIFT + RECON_SHIFT;
 
@@ -2413,11 +2413,11 @@ static int try_8x8basis_mmx(int16_t rem[64], int16_t weight[64], int16_t basis[6
         "pcmpeqw %%mm6, %%mm6		\n\t" // -1w
         "psrlw $15, %%mm6		\n\t" //  1w
         "pxor %%mm7, %%mm7		\n\t"
-        "movd  %4, %%mm5		\n\t" 
-        "punpcklwd %%mm5, %%mm5		\n\t" 
-        "punpcklwd %%mm5, %%mm5		\n\t" 
+        "movd  %4, %%mm5		\n\t"
+        "punpcklwd %%mm5, %%mm5		\n\t"
+        "punpcklwd %%mm5, %%mm5		\n\t"
         "1:				\n\t"
-        "movq  (%1, %0), %%mm0		\n\t" 
+        "movq  (%1, %0), %%mm0		\n\t"
         "movq  8(%1, %0), %%mm1		\n\t"
         "pmulhw %%mm5, %%mm0		\n\t"
         "pmulhw %%mm5, %%mm1		\n\t"
@@ -2444,7 +2444,7 @@ static int try_8x8basis_mmx(int16_t rem[64], int16_t weight[64], int16_t basis[6
         "paddd %%mm6, %%mm7		\n\t"
         "psrld $2, %%mm7		\n\t"
         "movd %%mm7, %0			\n\t"
-        
+
         : "+r" (i)
         : "r"(basis), "r"(rem), "r"(weight), "g"(scale)
     );
@@ -2453,21 +2453,21 @@ static int try_8x8basis_mmx(int16_t rem[64], int16_t weight[64], int16_t basis[6
 
 static void add_8x8basis_mmx(int16_t rem[64], int16_t basis[64], int scale){
     long i=0;
-    
+
     if(ABS(scale) < 256){
         scale<<= 16 + 1 - BASIS_SHIFT + RECON_SHIFT;
         asm volatile(
                 "pcmpeqw %%mm6, %%mm6		\n\t" // -1w
                 "psrlw $15, %%mm6		\n\t" //  1w
-                "movd  %3, %%mm5		\n\t" 
-                "punpcklwd %%mm5, %%mm5		\n\t" 
-                "punpcklwd %%mm5, %%mm5		\n\t" 
+                "movd  %3, %%mm5		\n\t"
+                "punpcklwd %%mm5, %%mm5		\n\t"
+                "punpcklwd %%mm5, %%mm5		\n\t"
                 "1:				\n\t"
-                "movq  (%1, %0), %%mm0		\n\t" 
+                "movq  (%1, %0), %%mm0		\n\t"
                 "movq  8(%1, %0), %%mm1		\n\t"
                 "pmulhw %%mm5, %%mm0		\n\t"
                 "pmulhw %%mm5, %%mm1		\n\t"
-                "paddw %%mm6, %%mm0		\n\t" 
+                "paddw %%mm6, %%mm0		\n\t"
                 "paddw %%mm6, %%mm1		\n\t"
                 "psraw $1, %%mm0		\n\t"
                 "psraw $1, %%mm1		\n\t"
@@ -2478,19 +2478,19 @@ static void add_8x8basis_mmx(int16_t rem[64], int16_t basis[64], int scale){
                 "add $16, %0			\n\t"
                 "cmp $128, %0			\n\t" //FIXME optimize & bench
                 " jb 1b				\n\t"
-                
+
                 : "+r" (i)
                 : "r"(basis), "r"(rem), "g"(scale)
         );
     }else{
         for(i=0; i<8*8; i++){
             rem[i] += (basis[i]*scale + (1<<(BASIS_SHIFT - RECON_SHIFT-1)))>>(BASIS_SHIFT - RECON_SHIFT);
-        }    
+        }
     }
 }
 
 #include "h264dsp_mmx.c"
-    
+
 /* external functions, from idct_mmx.c */
 void ff_mmx_idct(DCTELEM *block);
 void ff_mmxext_idct(DCTELEM *block);
@@ -2563,7 +2563,7 @@ static void ff_idct_xvid_mmx2_add(uint8_t *dest, int line_size, DCTELEM *block)
     add_pixels_clamped_mmx(block, dest, line_size);
 }
 #endif
-    
+
 void dsputil_init_mmx(DSPContext* c, AVCodecContext *avctx)
 {
     mm_flags = mm_support();
@@ -2701,14 +2701,14 @@ void dsputil_init_mmx(DSPContext* c, AVCodecContext *avctx)
         c->avg_no_rnd_pixels_tab[1][1] = avg_no_rnd_pixels8_x2_mmx;
         c->avg_no_rnd_pixels_tab[1][2] = avg_no_rnd_pixels8_y2_mmx;
         c->avg_no_rnd_pixels_tab[1][3] = avg_no_rnd_pixels8_xy2_mmx;
-                
+
         c->add_bytes= add_bytes_mmx;
 #ifdef CONFIG_ENCODERS
         c->diff_bytes= diff_bytes_mmx;
-        
+
         c->hadamard8_diff[0]= hadamard8_diff16_mmx;
         c->hadamard8_diff[1]= hadamard8_diff_mmx;
-        
+
 	c->pix_norm1 = pix_norm1_mmx;
 	c->sse[0] = (mm_flags & MM_SSE2) ? sse16_sse2 : sse16_mmx;
   	c->sse[1] = sse8_mmx;
@@ -2719,19 +2719,19 @@ void dsputil_init_mmx(DSPContext* c, AVCodecContext *avctx)
         if(!(avctx->flags & CODEC_FLAG_BITEXACT)){
             c->vsad[0] = vsad16_mmx;
         }
-        
+
         if(!(avctx->flags & CODEC_FLAG_BITEXACT)){
             c->try_8x8basis= try_8x8basis_mmx;
         }
         c->add_8x8basis= add_8x8basis_mmx;
-        
+
 #endif //CONFIG_ENCODERS
 
         c->h263_v_loop_filter= h263_v_loop_filter_mmx;
-        c->h263_h_loop_filter= h263_h_loop_filter_mmx;        
+        c->h263_h_loop_filter= h263_h_loop_filter_mmx;
 	c->put_h264_chroma_pixels_tab[0]= put_h264_chroma_mc8_mmx;
         c->put_h264_chroma_pixels_tab[1]= put_h264_chroma_mc4_mmx;
-        
+
         if (mm_flags & MM_MMXEXT) {
             c->put_pixels_tab[0][1] = put_pixels16_x2_mmx2;
             c->put_pixels_tab[0][2] = put_pixels16_y2_mmx2;
@@ -2945,7 +2945,7 @@ void dsputil_init_mmx(DSPContext* c, AVCodecContext *avctx)
             c->avg_h264_chroma_pixels_tab[1]= avg_h264_chroma_mc4_3dnow;
         }
     }
-        
+
 #ifdef CONFIG_ENCODERS
     dsputil_init_pix_mmx(c, avctx);
 #endif //CONFIG_ENCODERS
diff --git a/libavcodec/i386/dsputil_mmx_avg.h b/libavcodec/i386/dsputil_mmx_avg.h
index c708913048..434bc3a0e8 100644
--- a/libavcodec/i386/dsputil_mmx_avg.h
+++ b/libavcodec/i386/dsputil_mmx_avg.h
@@ -21,7 +21,7 @@
  * mostly rewritten by Michael Niedermayer <michaelni@gmx.at>
  * and improved by Zdenek Kabelac <kabi@users.sf.net>
  */
- 
+
 /* XXX: we use explicit registers to avoid a gcc 2.95.2 register asm
    clobber bug - now it will work with 2.95.2 and also with -fPIC
  */
@@ -100,7 +100,7 @@ static void DEF(put_pixels4_l2)(uint8_t *dst, uint8_t *src1, uint8_t *src2, int
 	:"+b"(h), "+a"(src1), "+c"(src2), "+d"(dst)
 #endif
 	:"S"((long)src1Stride), "D"((long)dstStride)
-	:"memory"); 
+	:"memory");
 }
 
 
@@ -147,7 +147,7 @@ static void DEF(put_pixels8_l2)(uint8_t *dst, uint8_t *src1, uint8_t *src2, int
 	:"+b"(h), "+a"(src1), "+c"(src2), "+d"(dst)
 #endif
 	:"S"((long)src1Stride), "D"((long)dstStride)
-	:"memory"); 
+	:"memory");
 //the following should be used, though better not with gcc ...
 /*	:"+g"(h), "+r"(src1), "+r"(src2), "+r"(dst)
 	:"r"(src1Stride), "r"(dstStride)
@@ -217,7 +217,7 @@ static void DEF(put_no_rnd_pixels8_l2)(uint8_t *dst, uint8_t *src1, uint8_t *src
 	:"+b"(h), "+a"(src1), "+c"(src2), "+d"(dst)
 #endif
 	:"S"((long)src1Stride), "D"((long)dstStride)
-	:"memory"); 
+	:"memory");
 //the following should be used, though better not with gcc ...
 /*	:"+g"(h), "+r"(src1), "+r"(src2), "+r"(dst)
 	:"r"(src1Stride), "r"(dstStride)
@@ -272,7 +272,7 @@ static void DEF(avg_pixels4_l2)(uint8_t *dst, uint8_t *src1, uint8_t *src2, int
 	:"+b"(h), "+a"(src1), "+c"(src2), "+d"(dst)
 #endif
 	:"S"((long)src1Stride), "D"((long)dstStride)
-	:"memory"); 
+	:"memory");
 }
 
 
@@ -324,7 +324,7 @@ static void DEF(avg_pixels8_l2)(uint8_t *dst, uint8_t *src1, uint8_t *src2, int
 	:"+b"(h), "+a"(src1), "+c"(src2), "+d"(dst)
 #endif
 	:"S"((long)src1Stride), "D"((long)dstStride)
-	:"memory"); 
+	:"memory");
 //the following should be used, though better not with gcc ...
 /*	:"+g"(h), "+r"(src1), "+r"(src2), "+r"(dst)
 	:"r"(src1Stride), "r"(dstStride)
@@ -412,7 +412,7 @@ static void DEF(put_pixels16_l2)(uint8_t *dst, uint8_t *src1, uint8_t *src2, int
 	:"+b"(h), "+a"(src1), "+c"(src2), "+d"(dst)
 #endif
 	:"S"((long)src1Stride), "D"((long)dstStride)
-	:"memory"); 
+	:"memory");
 //the following should be used, though better not with gcc ...
 /*	:"+g"(h), "+r"(src1), "+r"(src2), "+r"(dst)
 	:"r"(src1Stride), "r"(dstStride)
@@ -466,7 +466,7 @@ static void DEF(avg_pixels16_l2)(uint8_t *dst, uint8_t *src1, uint8_t *src2, int
 	:"+b"(h), "+a"(src1), "+c"(src2), "+d"(dst)
 #endif
 	:"S"((long)src1Stride), "D"((long)dstStride)
-	:"memory"); 
+	:"memory");
 //the following should be used, though better not with gcc ...
 /*	:"+g"(h), "+r"(src1), "+r"(src2), "+r"(dst)
 	:"r"(src1Stride), "r"(dstStride)
@@ -539,13 +539,13 @@ static void DEF(put_no_rnd_pixels16_l2)(uint8_t *dst, uint8_t *src1, uint8_t *sr
 	:"+b"(h), "+a"(src1), "+c"(src2), "+d"(dst)
 #endif
 	:"S"((long)src1Stride), "D"((long)dstStride)
-	:"memory"); 
+	:"memory");
 //the following should be used, though better not with gcc ...
 /*	:"+g"(h), "+r"(src1), "+r"(src2), "+r"(dst)
 	:"r"(src1Stride), "r"(dstStride)
 	:"memory");*/
 }
- 
+
 /* GL: this function does incorrect rounding if overflow */
 static void DEF(put_no_rnd_pixels8_x2)(uint8_t *block, const uint8_t *pixels, int line_size, int h)
 {
@@ -746,7 +746,7 @@ static void DEF(avg_pixels8_y2)(uint8_t *block, const uint8_t *pixels, int line_
 	:"%"REG_a, "memory");
 }
 
-// Note this is not correctly rounded, but this function is only used for b frames so it doesnt matter 
+// Note this is not correctly rounded, but this function is only used for b frames so it doesnt matter
 static void DEF(avg_pixels8_xy2)(uint8_t *block, const uint8_t *pixels, int line_size, int h)
 {
     MOVQ_BONE(mm6);
diff --git a/libavcodec/i386/dsputil_mmx_rnd.h b/libavcodec/i386/dsputil_mmx_rnd.h
index a56374b63e..6d93f9d55f 100644
--- a/libavcodec/i386/dsputil_mmx_rnd.h
+++ b/libavcodec/i386/dsputil_mmx_rnd.h
@@ -197,7 +197,7 @@ static void attribute_unused DEF(put, pixels16_l2)(uint8_t *dst, uint8_t *src1,
 	:"+b"(h), "+a"(src1), "+c"(src2), "+d"(dst)
 #endif
 	:"S"((long)src1Stride), "D"((long)dstStride)
-	:"memory"); 
+	:"memory");
 }
 
 static void DEF(put, pixels8_y2)(uint8_t *block, const uint8_t *pixels, int line_size, int h)
diff --git a/libavcodec/i386/fdct_mmx.c b/libavcodec/i386/fdct_mmx.c
index 6a13090a13..f3023549a2 100644
--- a/libavcodec/i386/fdct_mmx.c
+++ b/libavcodec/i386/fdct_mmx.c
@@ -5,7 +5,7 @@
  * SSE2 optimization is Copyright (c) 2004 Denes Balatoni.
  *
  * from  fdctam32.c - AP922 MMX(3D-Now) forward-DCT
- * 
+ *
  *  Intel Application Note AP-922 - fast, precise implementation of DCT
  *        http://developer.intel.com/vtune/cbts/appnotes.htm
  *
@@ -51,7 +51,7 @@ static const int64_t fdct_one_corr ATTR_ALIGN(8) = 0x0001000100010001LL;
 
 static const int32_t fdct_r_row[2] ATTR_ALIGN(8) = {RND_FRW_ROW, RND_FRW_ROW };
 
-struct 
+struct
 {
  const int32_t fdct_r_row_sse2[4] ATTR_ALIGN(16);
 } fdct_r_row_sse2 ATTR_ALIGN(16)=
@@ -61,90 +61,90 @@ struct
 //static const long fdct_r_row_sse2[4] ATTR_ALIGN(16) = {RND_FRW_ROW, RND_FRW_ROW, RND_FRW_ROW, RND_FRW_ROW};
 
 static const int16_t tab_frw_01234567[] ATTR_ALIGN(8) = {  // forward_dct coeff table
-  16384,   16384,   22725,   19266, 
-  16384,   16384,   12873,    4520, 
-  21407,    8867,   19266,   -4520, 
-  -8867,  -21407,  -22725,  -12873, 
-  16384,  -16384,   12873,  -22725, 
- -16384,   16384,    4520,   19266, 
-   8867,  -21407,    4520,  -12873, 
-  21407,   -8867,   19266,  -22725, 
-
-  22725,   22725,   31521,   26722, 
-  22725,   22725,   17855,    6270, 
-  29692,   12299,   26722,   -6270, 
- -12299,  -29692,  -31521,  -17855, 
-  22725,  -22725,   17855,  -31521, 
- -22725,   22725,    6270,   26722, 
-  12299,  -29692,    6270,  -17855, 
-  29692,  -12299,   26722,  -31521, 
-
-  21407,   21407,   29692,   25172, 
-  21407,   21407,   16819,    5906, 
-  27969,   11585,   25172,   -5906, 
- -11585,  -27969,  -29692,  -16819, 
-  21407,  -21407,   16819,  -29692, 
- -21407,   21407,    5906,   25172, 
-  11585,  -27969,    5906,  -16819, 
-  27969,  -11585,   25172,  -29692, 
-
-  19266,   19266,   26722,   22654, 
-  19266,   19266,   15137,    5315, 
-  25172,   10426,   22654,   -5315, 
- -10426,  -25172,  -26722,  -15137, 
-  19266,  -19266,   15137,  -26722, 
- -19266,   19266,    5315,   22654, 
-  10426,  -25172,    5315,  -15137, 
-  25172,  -10426,   22654,  -26722, 
-
-  16384,   16384,   22725,   19266, 
-  16384,   16384,   12873,    4520, 
-  21407,    8867,   19266,   -4520, 
-  -8867,  -21407,  -22725,  -12873, 
-  16384,  -16384,   12873,  -22725, 
- -16384,   16384,    4520,   19266, 
-   8867,  -21407,    4520,  -12873, 
-  21407,   -8867,   19266,  -22725, 
-
-  19266,   19266,   26722,   22654, 
-  19266,   19266,   15137,    5315, 
-  25172,   10426,   22654,   -5315, 
- -10426,  -25172,  -26722,  -15137, 
-  19266,  -19266,   15137,  -26722, 
- -19266,   19266,    5315,   22654, 
-  10426,  -25172,    5315,  -15137, 
-  25172,  -10426,   22654,  -26722, 
-
-  21407,   21407,   29692,   25172, 
-  21407,   21407,   16819,    5906, 
-  27969,   11585,   25172,   -5906, 
- -11585,  -27969,  -29692,  -16819, 
-  21407,  -21407,   16819,  -29692, 
- -21407,   21407,    5906,   25172, 
-  11585,  -27969,    5906,  -16819, 
-  27969,  -11585,   25172,  -29692, 
-
-  22725,   22725,   31521,   26722, 
-  22725,   22725,   17855,    6270, 
-  29692,   12299,   26722,   -6270, 
- -12299,  -29692,  -31521,  -17855, 
-  22725,  -22725,   17855,  -31521, 
- -22725,   22725,    6270,   26722, 
-  12299,  -29692,    6270,  -17855, 
-  29692,  -12299,   26722,  -31521, 
+  16384,   16384,   22725,   19266,
+  16384,   16384,   12873,    4520,
+  21407,    8867,   19266,   -4520,
+  -8867,  -21407,  -22725,  -12873,
+  16384,  -16384,   12873,  -22725,
+ -16384,   16384,    4520,   19266,
+   8867,  -21407,    4520,  -12873,
+  21407,   -8867,   19266,  -22725,
+
+  22725,   22725,   31521,   26722,
+  22725,   22725,   17855,    6270,
+  29692,   12299,   26722,   -6270,
+ -12299,  -29692,  -31521,  -17855,
+  22725,  -22725,   17855,  -31521,
+ -22725,   22725,    6270,   26722,
+  12299,  -29692,    6270,  -17855,
+  29692,  -12299,   26722,  -31521,
+
+  21407,   21407,   29692,   25172,
+  21407,   21407,   16819,    5906,
+  27969,   11585,   25172,   -5906,
+ -11585,  -27969,  -29692,  -16819,
+  21407,  -21407,   16819,  -29692,
+ -21407,   21407,    5906,   25172,
+  11585,  -27969,    5906,  -16819,
+  27969,  -11585,   25172,  -29692,
+
+  19266,   19266,   26722,   22654,
+  19266,   19266,   15137,    5315,
+  25172,   10426,   22654,   -5315,
+ -10426,  -25172,  -26722,  -15137,
+  19266,  -19266,   15137,  -26722,
+ -19266,   19266,    5315,   22654,
+  10426,  -25172,    5315,  -15137,
+  25172,  -10426,   22654,  -26722,
+
+  16384,   16384,   22725,   19266,
+  16384,   16384,   12873,    4520,
+  21407,    8867,   19266,   -4520,
+  -8867,  -21407,  -22725,  -12873,
+  16384,  -16384,   12873,  -22725,
+ -16384,   16384,    4520,   19266,
+   8867,  -21407,    4520,  -12873,
+  21407,   -8867,   19266,  -22725,
+
+  19266,   19266,   26722,   22654,
+  19266,   19266,   15137,    5315,
+  25172,   10426,   22654,   -5315,
+ -10426,  -25172,  -26722,  -15137,
+  19266,  -19266,   15137,  -26722,
+ -19266,   19266,    5315,   22654,
+  10426,  -25172,    5315,  -15137,
+  25172,  -10426,   22654,  -26722,
+
+  21407,   21407,   29692,   25172,
+  21407,   21407,   16819,    5906,
+  27969,   11585,   25172,   -5906,
+ -11585,  -27969,  -29692,  -16819,
+  21407,  -21407,   16819,  -29692,
+ -21407,   21407,    5906,   25172,
+  11585,  -27969,    5906,  -16819,
+  27969,  -11585,   25172,  -29692,
+
+  22725,   22725,   31521,   26722,
+  22725,   22725,   17855,    6270,
+  29692,   12299,   26722,   -6270,
+ -12299,  -29692,  -31521,  -17855,
+  22725,  -22725,   17855,  -31521,
+ -22725,   22725,    6270,   26722,
+  12299,  -29692,    6270,  -17855,
+  29692,  -12299,   26722,  -31521,
 };
 
-struct 
+struct
 {
  const int16_t tab_frw_01234567_sse2[256] ATTR_ALIGN(16);
 } tab_frw_01234567_sse2 ATTR_ALIGN(16) =
 {{
-//static const int16_t tab_frw_01234567_sse2[] ATTR_ALIGN(16) = {  // forward_dct coeff table  
+//static const int16_t tab_frw_01234567_sse2[] ATTR_ALIGN(16) = {  // forward_dct coeff table
 #define TABLE_SSE2 C4,  C4,  C1,  C3, -C6, -C2, -C1, -C5, \
                    C4,  C4,  C5,  C7,  C2,  C6,  C3, -C7, \
                   -C4,  C4,  C7,  C3,  C6, -C2,  C7, -C5, \
-                   C4, -C4,  C5, -C1,  C2, -C6,  C3, -C1, 
-// c1..c7 * cos(pi/4) * 2^15 
+                   C4, -C4,  C5, -C1,  C2, -C6,  C3, -C1,
+// c1..c7 * cos(pi/4) * 2^15
 #define C1 22725
 #define C2 21407
 #define C3 19266
@@ -355,17 +355,17 @@ static always_inline void fdct_row_sse2(const int16_t *in, int16_t *out)
 	"movq      \\i(%0), %%xmm2     \n\t"
 	"movq      \\i+8(%0), %%xmm0   \n\t"
 	"movdqa    \\t+32(%1), %%xmm3  \n\t"
-	"movdqa    \\t+48(%1), %%xmm7  \n\t"	
+	"movdqa    \\t+48(%1), %%xmm7  \n\t"
 	"movdqa    \\t(%1), %%xmm4     \n\t"
-	"movdqa    \\t+16(%1), %%xmm5  \n\t"	
+	"movdqa    \\t+16(%1), %%xmm5  \n\t"
 	".endm                         \n\t"
         ".macro FDCT_ROW_SSE2_H2 i t   \n\t"
 	"movq      \\i(%0), %%xmm2     \n\t"
 	"movq      \\i+8(%0), %%xmm0   \n\t"
 	"movdqa    \\t+32(%1), %%xmm3  \n\t"
-	"movdqa    \\t+48(%1), %%xmm7  \n\t"	
+	"movdqa    \\t+48(%1), %%xmm7  \n\t"
 	".endm                         \n\t"
-	".macro FDCT_ROW_SSE2 i        \n\t"	
+	".macro FDCT_ROW_SSE2 i        \n\t"
 	"movq      %%xmm2, %%xmm1      \n\t"
 	"pshuflw   $27, %%xmm0, %%xmm0 \n\t"
 	"paddsw    %%xmm0, %%xmm1      \n\t"
@@ -376,7 +376,7 @@ static always_inline void fdct_row_sse2(const int16_t *in, int16_t *out)
 	"pmaddwd   %%xmm1, %%xmm7      \n\t"
 	"pmaddwd   %%xmm5, %%xmm2      \n\t"
 	"pmaddwd   %%xmm4, %%xmm1      \n\t"
-	"paddd     %%xmm7, %%xmm3      \n\t"	
+	"paddd     %%xmm7, %%xmm3      \n\t"
 	"paddd     %%xmm2, %%xmm1      \n\t"
 	"paddd     %%xmm6, %%xmm3      \n\t"
 	"paddd     %%xmm6, %%xmm1      \n\t"
@@ -384,8 +384,8 @@ static always_inline void fdct_row_sse2(const int16_t *in, int16_t *out)
 	"psrad     %3, %%xmm1          \n\t"
 	"packssdw  %%xmm3, %%xmm1      \n\t"
 	"movdqa    %%xmm1, \\i(%4)     \n\t"
-	".endm                         \n\t"	
-	"movdqa    (%2), %%xmm6        \n\t"		
+	".endm                         \n\t"
+	"movdqa    (%2), %%xmm6        \n\t"
 	"FDCT_ROW_SSE2_H1 0 0 \n\t"
 	"FDCT_ROW_SSE2 0 \n\t"
 	"FDCT_ROW_SSE2_H2 64 0 \n\t"
@@ -411,7 +411,7 @@ static always_inline void fdct_row_sse2(const int16_t *in, int16_t *out)
 }
 
 static always_inline void fdct_row_mmx2(const int16_t *in, int16_t *out, const int16_t *table)
-{ 
+{
     pshufw_m2r(*(in + 4), mm5, 0x1B);
     movq_m2r(*(in + 0), mm0);
     movq_r2r(mm0, mm1);
@@ -454,7 +454,7 @@ static always_inline void fdct_row_mmx2(const int16_t *in, int16_t *out, const i
 }
 
 static always_inline void fdct_row_mmx(const int16_t *in, int16_t *out, const int16_t *table)
-{ 
+{
 //FIXME reorder (i dont have a old mmx only cpu here to benchmark ...)
     movd_m2r(*(in + 6), mm1);
     punpcklwd_m2r(*(in + 4), mm1);
@@ -547,7 +547,7 @@ void ff_fdct_mmx2(int16_t *block)
     }
 }
 
-void ff_fdct_sse2(int16_t *block) 
+void ff_fdct_sse2(int16_t *block)
 {
     int64_t align_tmp[16] ATTR_ALIGN(8);
     int16_t * const block_tmp= (int16_t*)align_tmp;
diff --git a/libavcodec/i386/fft_sse.c b/libavcodec/i386/fft_sse.c
index d07c943e91..f8be644a3b 100644
--- a/libavcodec/i386/fft_sse.c
+++ b/libavcodec/i386/fft_sse.c
@@ -23,13 +23,13 @@
 
 #include <xmmintrin.h>
 
-static const float p1p1p1m1[4] __attribute__((aligned(16))) = 
+static const float p1p1p1m1[4] __attribute__((aligned(16))) =
     { 1.0, 1.0, 1.0, -1.0 };
 
-static const float p1p1m1p1[4] __attribute__((aligned(16))) = 
+static const float p1p1m1p1[4] __attribute__((aligned(16))) =
     { 1.0, 1.0, -1.0, 1.0 };
 
-static const float p1p1m1m1[4] __attribute__((aligned(16))) = 
+static const float p1p1m1m1[4] __attribute__((aligned(16))) =
     { 1.0, 1.0, -1.0, -1.0 };
 
 #if 0
@@ -107,27 +107,27 @@ void ff_fft_calc_sse(FFTContext *s, FFTComplex *z)
 
                 a = *(__m128 *)p;
                 b = *(__m128 *)q;
-                
+
                 /* complex mul */
                 c = *(__m128 *)cptr;
                 /*  cre*re cim*re */
-                t1 = _mm_mul_ps(c, 
-                                _mm_shuffle_ps(b, b, _MM_SHUFFLE(2, 2, 0, 0))); 
+                t1 = _mm_mul_ps(c,
+                                _mm_shuffle_ps(b, b, _MM_SHUFFLE(2, 2, 0, 0)));
                 c = *(__m128 *)(cptr + 2);
                 /*  -cim*im cre*im */
                 t2 = _mm_mul_ps(c,
-                                _mm_shuffle_ps(b, b, _MM_SHUFFLE(3, 3, 1, 1))); 
+                                _mm_shuffle_ps(b, b, _MM_SHUFFLE(3, 3, 1, 1)));
                 b = _mm_add_ps(t1, t2);
-                
+
                 /* butterfly */
                 *(__m128 *)p = _mm_add_ps(a, b);
                 *(__m128 *)q = _mm_sub_ps(a, b);
-                
+
                 p += 2;
                 q += 2;
                 cptr += 4;
             } while (--k);
-        
+
             p += nloops;
             q += nloops;
         } while (--j);
diff --git a/libavcodec/i386/h264dsp_mmx.c b/libavcodec/i386/h264dsp_mmx.c
index b5e9baa3a6..45a3c02f35 100644
--- a/libavcodec/i386/h264dsp_mmx.c
+++ b/libavcodec/i386/h264dsp_mmx.c
@@ -384,7 +384,7 @@ static void h264_h_loop_filter_chroma_intra_mmx2(uint8_t *pix, int stride, int a
         "psraw $5, %%mm6		\n\t"\
         "packuswb %%mm6, %%mm6		\n\t"\
         OP(%%mm6, (%1), A, d)\
-        "add %3, %1			\n\t"     
+        "add %3, %1			\n\t"
 
 #define QPEL_H264HV(A,B,C,D,E,F,OF)\
         "movd (%0), "#F"		\n\t"\
@@ -399,7 +399,7 @@ static void h264_h_loop_filter_chroma_intra_mmx2(uint8_t *pix, int stride, int a
         "paddw "#F", "#A"		\n\t"\
         "paddw "#A", %%mm6		\n\t"\
         "movq %%mm6, "#OF"(%1)		\n\t"
-        
+
 #define QPEL_H264(OPNAME, OP, MMX)\
 static void OPNAME ## h264_qpel4_h_lowpass_ ## MMX(uint8_t *dst, uint8_t *src, int dstStride, int srcStride){\
     int h=4;\
diff --git a/libavcodec/i386/idct_mmx_xvid.c b/libavcodec/i386/idct_mmx_xvid.c
index 943c50f92b..219260ed88 100644
--- a/libavcodec/i386/idct_mmx_xvid.c
+++ b/libavcodec/i386/idct_mmx_xvid.c
@@ -72,13 +72,13 @@
 //-----------------------------------------------------------------------------
 
 
-static const int16_t tg_1_16[4*4] attribute_used __attribute__ ((aligned(8))) = { 
+static const int16_t tg_1_16[4*4] attribute_used __attribute__ ((aligned(8))) = {
   13036,13036,13036,13036,        // tg * (2<<16) + 0.5
   27146,27146,27146,27146,        // tg * (2<<16) + 0.5
   -21746,-21746,-21746,-21746,    // tg * (2<<16) + 0.5
   23170,23170,23170,23170};       // cos * (2<<15) + 0.5
 
-static const int32_t rounder_0[2*8] attribute_used __attribute__ ((aligned(8))) = { 
+static const int32_t rounder_0[2*8] attribute_used __attribute__ ((aligned(8))) = {
   65536,65536,
   3597,3597,
   2260,2260,
@@ -148,7 +148,7 @@ static const int32_t rounder_0[2*8] attribute_used __attribute__ ((aligned(8)))
 //-----------------------------------------------------------------------------
 
 // Table for rows 0,4 - constants are multiplied by cos_4_16
-static const int16_t tab_i_04_mmx[32*4] attribute_used __attribute__ ((aligned(8))) = { 
+static const int16_t tab_i_04_mmx[32*4] attribute_used __attribute__ ((aligned(8))) = {
   16384,16384,16384,-16384,       // movq-> w06 w04 w02 w00
   21407,8867,8867,-21407,         // w07 w05 w03 w01
   16384,-16384,16384,16384,       // w14 w12 w10 w08
@@ -190,7 +190,7 @@ static const int16_t tab_i_04_mmx[32*4] attribute_used __attribute__ ((aligned(8
 //-----------------------------------------------------------------------------
 
 // %3 for rows 0,4 - constants are multiplied by cos_4_16
-static const int16_t tab_i_04_xmm[32*4] attribute_used __attribute__ ((aligned(8))) = { 
+static const int16_t tab_i_04_xmm[32*4] attribute_used __attribute__ ((aligned(8))) = {
   16384,21407,16384,8867,      // movq-> w05 w04 w01 w00
   16384,8867,-16384,-21407,    // w07 w06 w03 w02
   16384,-8867,16384,-21407,    // w13 w12 w09 w08
@@ -501,7 +501,7 @@ asm volatile(
     DCT_8_INV_ROW_MMX(5*16(%0), 5*16(%0), 64*3(%2), 8*5(%1))
     DCT_8_INV_ROW_MMX(6*16(%0), 6*16(%0), 64*2(%2), 8*6(%1))
     DCT_8_INV_ROW_MMX(7*16(%0), 7*16(%0), 64*1(%2), 8*7(%1))
-    
+
             //# Process the columns (4 at a time)
     DCT_8_INV_COL(0(%0), 0(%0))
     DCT_8_INV_COL(8(%0), 8(%0))
@@ -524,7 +524,7 @@ asm volatile(
     DCT_8_INV_ROW_XMM(5*16(%0), 5*16(%0), 64*3(%2), 8*5(%1))
     DCT_8_INV_ROW_XMM(6*16(%0), 6*16(%0), 64*2(%2), 8*6(%1))
     DCT_8_INV_ROW_XMM(7*16(%0), 7*16(%0), 64*1(%2), 8*7(%1))
-    
+
             //# Process the columns (4 at a time)
     DCT_8_INV_COL(0(%0), 0(%0))
     DCT_8_INV_COL(8(%0), 8(%0))
diff --git a/libavcodec/i386/motion_est_mmx.c b/libavcodec/i386/motion_est_mmx.c
index 1b90f8e40f..69e10f628b 100644
--- a/libavcodec/i386/motion_est_mmx.c
+++ b/libavcodec/i386/motion_est_mmx.c
@@ -393,7 +393,7 @@ void dsputil_init_pix_mmx(DSPContext* c, AVCodecContext *avctx)
 
 	c->sad[0]= sad16_mmx2;
 	c->sad[1]= sad8_mmx2;
-        
+
         if(!(avctx->flags & CODEC_FLAG_BITEXACT)){
             c->pix_abs[0][1] = sad16_x2_mmx2;
             c->pix_abs[0][2] = sad16_y2_mmx2;
diff --git a/libavcodec/i386/mpegvideo_mmx.c b/libavcodec/i386/mpegvideo_mmx.c
index 70c81f6754..af799b6b00 100644
--- a/libavcodec/i386/mpegvideo_mmx.c
+++ b/libavcodec/i386/mpegvideo_mmx.c
@@ -40,7 +40,7 @@ static void dct_unquantize_h263_intra_mmx(MpegEncContext *s,
     qmul = qscale << 1;
 
     assert(s->block_last_index[n]>=0 || s->h263_aic);
-        
+
     if (!s->h263_aic) {
         if (n < 4)
             level = block[0] * s->y_dc_scale;
@@ -116,7 +116,7 @@ static void dct_unquantize_h263_inter_mmx(MpegEncContext *s,
     qadd = (qscale - 1) | 1;
 
     assert(s->block_last_index[n]>=0 || s->h263_aic);
-        
+
     nCoeffs= s->inter_scantable.raster_end[ s->block_last_index[n] ];
 //printf("%d %d  ", qmul, qadd);
 asm volatile(
@@ -209,7 +209,7 @@ static void dct_unquantize_mpeg1_intra_mmx(MpegEncContext *s,
 
     nCoeffs= s->intra_scantable.raster_end[ s->block_last_index[n] ]+1;
 
-    if (n < 4) 
+    if (n < 4)
         block0 = block[0] * s->y_dc_scale;
     else
         block0 = block[0] * s->c_dc_scale;
@@ -263,7 +263,7 @@ asm volatile(
 		"js 1b				\n\t"
 		::"r" (block+nCoeffs), "r"(quant_matrix+nCoeffs), "g" (qscale), "g" (-2*nCoeffs)
 		: "%"REG_a, "memory"
-	);    
+	);
     block[0]= block0;
 }
 
@@ -339,13 +339,13 @@ static void dct_unquantize_mpeg2_intra_mmx(MpegEncContext *s,
     long nCoeffs;
     const uint16_t *quant_matrix;
     int block0;
-    
+
     assert(s->block_last_index[n]>=0);
 
     if(s->alternate_scan) nCoeffs= 63; //FIXME
     else nCoeffs= s->intra_scantable.raster_end[ s->block_last_index[n] ];
 
-    if (n < 4) 
+    if (n < 4)
         block0 = block[0] * s->y_dc_scale;
     else
         block0 = block[0] * s->c_dc_scale;
@@ -394,7 +394,7 @@ asm volatile(
 		"jng 1b				\n\t"
 		::"r" (block+nCoeffs), "r"(quant_matrix+nCoeffs), "g" (qscale), "g" (-2*nCoeffs)
 		: "%"REG_a, "memory"
-	);    
+	);
     block[0]= block0;
         //Note, we dont do mismatch control for intra as errors cannot accumulate
 }
@@ -404,7 +404,7 @@ static void dct_unquantize_mpeg2_inter_mmx(MpegEncContext *s,
 {
     long nCoeffs;
     const uint16_t *quant_matrix;
-    
+
     assert(s->block_last_index[n]>=0);
 
     if(s->alternate_scan) nCoeffs= 63; //FIXME
@@ -470,13 +470,13 @@ asm volatile(
                 "psrlq $15, %%mm7		\n\t"
                 "pxor %%mm7, %%mm0		\n\t"
                 "movd %%mm0, 124(%0, %3)	\n\t"
-                
+
 		::"r" (block+nCoeffs), "r"(quant_matrix+nCoeffs), "g" (qscale), "r" (-2*nCoeffs)
 		: "%"REG_a, "memory"
 	);
 }
 
-/* draw the edges of width 'w' of an image of size width, height 
+/* draw the edges of width 'w' of an image of size width, height
    this mmx version can only handle w==8 || w==16 */
 static void draw_edges_mmx(uint8_t *buf, int wrap, int width, int height, int w)
 {
@@ -491,7 +491,7 @@ static void draw_edges_mmx(uint8_t *buf, int wrap, int width, int height, int w)
 	asm volatile(
 		"1:				\n\t"
 		"movd (%0), %%mm0		\n\t"
-		"punpcklbw %%mm0, %%mm0		\n\t" 
+		"punpcklbw %%mm0, %%mm0		\n\t"
 		"punpcklwd %%mm0, %%mm0		\n\t"
 		"punpckldq %%mm0, %%mm0		\n\t"
 		"movq %%mm0, -8(%0)		\n\t"
@@ -512,7 +512,7 @@ static void draw_edges_mmx(uint8_t *buf, int wrap, int width, int height, int w)
 	asm volatile(
 		"1:				\n\t"
 		"movd (%0), %%mm0		\n\t"
-		"punpcklbw %%mm0, %%mm0		\n\t" 
+		"punpcklbw %%mm0, %%mm0		\n\t"
 		"punpcklwd %%mm0, %%mm0		\n\t"
 		"punpckldq %%mm0, %%mm0		\n\t"
 		"movq %%mm0, -8(%0)		\n\t"
@@ -525,12 +525,12 @@ static void draw_edges_mmx(uint8_t *buf, int wrap, int width, int height, int w)
 		"movq %%mm1, 8(%0, %2)		\n\t"
 		"add %1, %0			\n\t"
 		"cmp %3, %0			\n\t"
-		" jb 1b				\n\t"		
+		" jb 1b				\n\t"
 		: "+r" (ptr)
 		: "r" ((long)wrap), "r" ((long)width), "r" (ptr + wrap*height)
 	);
     }
-    
+
     for(i=0;i<w;i+=4) {
         /* top and bottom (and hopefully also the corners) */
 	ptr= buf - (i + 1) * wrap - w;
@@ -694,7 +694,7 @@ void MPV_common_init_mmx(MpegEncContext *s)
 {
     if (mm_flags & MM_MMX) {
         const int dct_algo = s->avctx->dct_algo;
-        
+
         s->dct_unquantize_h263_intra = dct_unquantize_h263_intra_mmx;
         s->dct_unquantize_h263_inter = dct_unquantize_h263_inter_mmx;
         s->dct_unquantize_mpeg1_intra = dct_unquantize_mpeg1_intra_mmx;
@@ -703,7 +703,7 @@ void MPV_common_init_mmx(MpegEncContext *s)
         s->dct_unquantize_mpeg2_inter = dct_unquantize_mpeg2_inter_mmx;
 
         draw_edges = draw_edges_mmx;
-        
+
         if (mm_flags & MM_SSE2) {
 	    s->denoise_dct= denoise_dct_sse2;
 	} else {
diff --git a/libavcodec/i386/mpegvideo_mmx_template.c b/libavcodec/i386/mpegvideo_mmx_template.c
index 93f156ee55..28afdeef0c 100644
--- a/libavcodec/i386/mpegvideo_mmx_template.c
+++ b/libavcodec/i386/mpegvideo_mmx_template.c
@@ -52,7 +52,7 @@ static int RENAME(dct_quantize)(MpegEncContext *s,
     int level=0, q; //=0 is cuz gcc says uninitalized ...
     const uint16_t *qmat, *bias;
     __align8 int16_t temp_block[64];
-    
+
     assert((7&(int)(&temp_block[0])) == 0); //did gcc align it correctly?
 
     //s->fdct (block);
@@ -88,7 +88,7 @@ static int RENAME(dct_quantize)(MpegEncContext *s,
         } else
             /* For AIC we skip quant/dequant of INTRADC */
             level = (block[0] + 4)>>3;
-            
+
         block[0]=0; //avoid fake overflow
 //        temp_block[0] = (block[0] + (q >> 1)) / q;
         last_non_zero_p1 = 1;
@@ -101,7 +101,7 @@ static int RENAME(dct_quantize)(MpegEncContext *s,
     }
 
     if((s->out_format == FMT_H263 || s->out_format == FMT_H261) && s->mpeg_quant==0){
-    
+
         asm volatile(
             "movd %%"REG_a", %%mm3		\n\t" // last_non_zero_p1
             SPREADW(%%mm3)
@@ -116,16 +116,16 @@ static int RENAME(dct_quantize)(MpegEncContext *s,
             "pxor %%mm1, %%mm1			\n\t" // 0
             "movq (%1, %%"REG_a"), %%mm0	\n\t" // block[i]
             "pcmpgtw %%mm0, %%mm1		\n\t" // block[i] <= 0 ? 0xFF : 0x00
-            "pxor %%mm1, %%mm0			\n\t" 
+            "pxor %%mm1, %%mm0			\n\t"
             "psubw %%mm1, %%mm0			\n\t" // ABS(block[i])
             "psubusw %%mm6, %%mm0		\n\t" // ABS(block[i]) + bias[0]
             "pmulhw %%mm5, %%mm0		\n\t" // (ABS(block[i])*qmat[0] - bias[0]*qmat[0])>>16
-            "por %%mm0, %%mm4			\n\t" 
-            "pxor %%mm1, %%mm0			\n\t" 
+            "por %%mm0, %%mm4			\n\t"
+            "pxor %%mm1, %%mm0			\n\t"
             "psubw %%mm1, %%mm0			\n\t" // out=((ABS(block[i])*qmat[0] - bias[0]*qmat[0])>>16)*sign(block[i])
             "movq %%mm0, (%5, %%"REG_a")	\n\t"
             "pcmpeqw %%mm7, %%mm0		\n\t" // out==0 ? 0xFF : 0x00
-            "movq (%4, %%"REG_a"), %%mm1	\n\t" 
+            "movq (%4, %%"REG_a"), %%mm1	\n\t"
             "movq %%mm7, (%1, %%"REG_a")	\n\t" // 0
             "pandn %%mm1, %%mm0			\n\t"
 	    PMAXW(%%mm0, %%mm3)
@@ -142,7 +142,7 @@ static int RENAME(dct_quantize)(MpegEncContext *s,
         asm volatile(
             "movd %1, %%mm1			\n\t" // max_qcoeff
 	    SPREADW(%%mm1)
-            "psubusw %%mm1, %%mm4		\n\t" 
+            "psubusw %%mm1, %%mm4		\n\t"
             "packuswb %%mm4, %%mm4		\n\t"
             "movd %%mm4, %0			\n\t" // *overflow
         : "=g" (*overflow)
@@ -160,18 +160,18 @@ static int RENAME(dct_quantize)(MpegEncContext *s,
             "pxor %%mm1, %%mm1			\n\t" // 0
             "movq (%1, %%"REG_a"), %%mm0	\n\t" // block[i]
             "pcmpgtw %%mm0, %%mm1		\n\t" // block[i] <= 0 ? 0xFF : 0x00
-            "pxor %%mm1, %%mm0			\n\t" 
+            "pxor %%mm1, %%mm0			\n\t"
             "psubw %%mm1, %%mm0			\n\t" // ABS(block[i])
             "movq (%3, %%"REG_a"), %%mm6	\n\t" // bias[0]
             "paddusw %%mm6, %%mm0		\n\t" // ABS(block[i]) + bias[0]
             "movq (%2, %%"REG_a"), %%mm5		\n\t" // qmat[i]
             "pmulhw %%mm5, %%mm0		\n\t" // (ABS(block[i])*qmat[0] + bias[0]*qmat[0])>>16
-            "por %%mm0, %%mm4			\n\t" 
-            "pxor %%mm1, %%mm0			\n\t" 
+            "por %%mm0, %%mm4			\n\t"
+            "pxor %%mm1, %%mm0			\n\t"
             "psubw %%mm1, %%mm0			\n\t" // out=((ABS(block[i])*qmat[0] - bias[0]*qmat[0])>>16)*sign(block[i])
             "movq %%mm0, (%5, %%"REG_a")	\n\t"
             "pcmpeqw %%mm7, %%mm0		\n\t" // out==0 ? 0xFF : 0x00
-            "movq (%4, %%"REG_a"), %%mm1		\n\t" 
+            "movq (%4, %%"REG_a"), %%mm1		\n\t"
             "movq %%mm7, (%1, %%"REG_a")		\n\t" // 0
             "pandn %%mm1, %%mm0			\n\t"
 	    PMAXW(%%mm0, %%mm3)
@@ -188,7 +188,7 @@ static int RENAME(dct_quantize)(MpegEncContext *s,
         asm volatile(
             "movd %1, %%mm1			\n\t" // max_qcoeff
 	    SPREADW(%%mm1)
-            "psubusw %%mm1, %%mm4		\n\t" 
+            "psubusw %%mm1, %%mm4		\n\t"
             "packuswb %%mm4, %%mm4		\n\t"
             "movd %%mm4, %0			\n\t" // *overflow
         : "=g" (*overflow)
@@ -201,135 +201,135 @@ static int RENAME(dct_quantize)(MpegEncContext *s,
 
     if(s->dsp.idct_permutation_type == FF_SIMPLE_IDCT_PERM){
         if(last_non_zero_p1 <= 1) goto end;
-        block[0x08] = temp_block[0x01]; block[0x10] = temp_block[0x08]; 
-        block[0x20] = temp_block[0x10]; 
+        block[0x08] = temp_block[0x01]; block[0x10] = temp_block[0x08];
+        block[0x20] = temp_block[0x10];
         if(last_non_zero_p1 <= 4) goto end;
-        block[0x18] = temp_block[0x09]; block[0x04] = temp_block[0x02]; 
-        block[0x09] = temp_block[0x03]; 
+        block[0x18] = temp_block[0x09]; block[0x04] = temp_block[0x02];
+        block[0x09] = temp_block[0x03];
         if(last_non_zero_p1 <= 7) goto end;
-        block[0x14] = temp_block[0x0A]; block[0x28] = temp_block[0x11]; 
-        block[0x12] = temp_block[0x18]; block[0x02] = temp_block[0x20]; 
+        block[0x14] = temp_block[0x0A]; block[0x28] = temp_block[0x11];
+        block[0x12] = temp_block[0x18]; block[0x02] = temp_block[0x20];
         if(last_non_zero_p1 <= 11) goto end;
-        block[0x1A] = temp_block[0x19]; block[0x24] = temp_block[0x12]; 
-        block[0x19] = temp_block[0x0B]; block[0x01] = temp_block[0x04]; 
-        block[0x0C] = temp_block[0x05]; 
+        block[0x1A] = temp_block[0x19]; block[0x24] = temp_block[0x12];
+        block[0x19] = temp_block[0x0B]; block[0x01] = temp_block[0x04];
+        block[0x0C] = temp_block[0x05];
         if(last_non_zero_p1 <= 16) goto end;
-        block[0x11] = temp_block[0x0C]; block[0x29] = temp_block[0x13]; 
-        block[0x16] = temp_block[0x1A]; block[0x0A] = temp_block[0x21]; 
-        block[0x30] = temp_block[0x28]; block[0x22] = temp_block[0x30]; 
-        block[0x38] = temp_block[0x29]; block[0x06] = temp_block[0x22]; 
+        block[0x11] = temp_block[0x0C]; block[0x29] = temp_block[0x13];
+        block[0x16] = temp_block[0x1A]; block[0x0A] = temp_block[0x21];
+        block[0x30] = temp_block[0x28]; block[0x22] = temp_block[0x30];
+        block[0x38] = temp_block[0x29]; block[0x06] = temp_block[0x22];
         if(last_non_zero_p1 <= 24) goto end;
-        block[0x1B] = temp_block[0x1B]; block[0x21] = temp_block[0x14]; 
-        block[0x1C] = temp_block[0x0D]; block[0x05] = temp_block[0x06]; 
-        block[0x0D] = temp_block[0x07]; block[0x15] = temp_block[0x0E]; 
-        block[0x2C] = temp_block[0x15]; block[0x13] = temp_block[0x1C]; 
+        block[0x1B] = temp_block[0x1B]; block[0x21] = temp_block[0x14];
+        block[0x1C] = temp_block[0x0D]; block[0x05] = temp_block[0x06];
+        block[0x0D] = temp_block[0x07]; block[0x15] = temp_block[0x0E];
+        block[0x2C] = temp_block[0x15]; block[0x13] = temp_block[0x1C];
         if(last_non_zero_p1 <= 32) goto end;
-        block[0x0B] = temp_block[0x23]; block[0x34] = temp_block[0x2A]; 
-        block[0x2A] = temp_block[0x31]; block[0x32] = temp_block[0x38]; 
-        block[0x3A] = temp_block[0x39]; block[0x26] = temp_block[0x32]; 
-        block[0x39] = temp_block[0x2B]; block[0x03] = temp_block[0x24]; 
+        block[0x0B] = temp_block[0x23]; block[0x34] = temp_block[0x2A];
+        block[0x2A] = temp_block[0x31]; block[0x32] = temp_block[0x38];
+        block[0x3A] = temp_block[0x39]; block[0x26] = temp_block[0x32];
+        block[0x39] = temp_block[0x2B]; block[0x03] = temp_block[0x24];
         if(last_non_zero_p1 <= 40) goto end;
-        block[0x1E] = temp_block[0x1D]; block[0x25] = temp_block[0x16]; 
-        block[0x1D] = temp_block[0x0F]; block[0x2D] = temp_block[0x17]; 
-        block[0x17] = temp_block[0x1E]; block[0x0E] = temp_block[0x25]; 
-        block[0x31] = temp_block[0x2C]; block[0x2B] = temp_block[0x33]; 
+        block[0x1E] = temp_block[0x1D]; block[0x25] = temp_block[0x16];
+        block[0x1D] = temp_block[0x0F]; block[0x2D] = temp_block[0x17];
+        block[0x17] = temp_block[0x1E]; block[0x0E] = temp_block[0x25];
+        block[0x31] = temp_block[0x2C]; block[0x2B] = temp_block[0x33];
         if(last_non_zero_p1 <= 48) goto end;
-        block[0x36] = temp_block[0x3A]; block[0x3B] = temp_block[0x3B]; 
-        block[0x23] = temp_block[0x34]; block[0x3C] = temp_block[0x2D]; 
-        block[0x07] = temp_block[0x26]; block[0x1F] = temp_block[0x1F]; 
-        block[0x0F] = temp_block[0x27]; block[0x35] = temp_block[0x2E]; 
+        block[0x36] = temp_block[0x3A]; block[0x3B] = temp_block[0x3B];
+        block[0x23] = temp_block[0x34]; block[0x3C] = temp_block[0x2D];
+        block[0x07] = temp_block[0x26]; block[0x1F] = temp_block[0x1F];
+        block[0x0F] = temp_block[0x27]; block[0x35] = temp_block[0x2E];
         if(last_non_zero_p1 <= 56) goto end;
-        block[0x2E] = temp_block[0x35]; block[0x33] = temp_block[0x3C]; 
-        block[0x3E] = temp_block[0x3D]; block[0x27] = temp_block[0x36]; 
-        block[0x3D] = temp_block[0x2F]; block[0x2F] = temp_block[0x37]; 
+        block[0x2E] = temp_block[0x35]; block[0x33] = temp_block[0x3C];
+        block[0x3E] = temp_block[0x3D]; block[0x27] = temp_block[0x36];
+        block[0x3D] = temp_block[0x2F]; block[0x2F] = temp_block[0x37];
         block[0x37] = temp_block[0x3E]; block[0x3F] = temp_block[0x3F];
     }else if(s->dsp.idct_permutation_type == FF_LIBMPEG2_IDCT_PERM){
         if(last_non_zero_p1 <= 1) goto end;
-        block[0x04] = temp_block[0x01]; 
-        block[0x08] = temp_block[0x08]; block[0x10] = temp_block[0x10]; 
+        block[0x04] = temp_block[0x01];
+        block[0x08] = temp_block[0x08]; block[0x10] = temp_block[0x10];
         if(last_non_zero_p1 <= 4) goto end;
-        block[0x0C] = temp_block[0x09]; block[0x01] = temp_block[0x02]; 
-        block[0x05] = temp_block[0x03]; 
+        block[0x0C] = temp_block[0x09]; block[0x01] = temp_block[0x02];
+        block[0x05] = temp_block[0x03];
         if(last_non_zero_p1 <= 7) goto end;
-        block[0x09] = temp_block[0x0A]; block[0x14] = temp_block[0x11]; 
-        block[0x18] = temp_block[0x18]; block[0x20] = temp_block[0x20]; 
+        block[0x09] = temp_block[0x0A]; block[0x14] = temp_block[0x11];
+        block[0x18] = temp_block[0x18]; block[0x20] = temp_block[0x20];
         if(last_non_zero_p1 <= 11) goto end;
-        block[0x1C] = temp_block[0x19]; 
-        block[0x11] = temp_block[0x12]; block[0x0D] = temp_block[0x0B]; 
-        block[0x02] = temp_block[0x04]; block[0x06] = temp_block[0x05]; 
+        block[0x1C] = temp_block[0x19];
+        block[0x11] = temp_block[0x12]; block[0x0D] = temp_block[0x0B];
+        block[0x02] = temp_block[0x04]; block[0x06] = temp_block[0x05];
         if(last_non_zero_p1 <= 16) goto end;
-        block[0x0A] = temp_block[0x0C]; block[0x15] = temp_block[0x13]; 
-        block[0x19] = temp_block[0x1A]; block[0x24] = temp_block[0x21]; 
-        block[0x28] = temp_block[0x28]; block[0x30] = temp_block[0x30]; 
-        block[0x2C] = temp_block[0x29]; block[0x21] = temp_block[0x22]; 
+        block[0x0A] = temp_block[0x0C]; block[0x15] = temp_block[0x13];
+        block[0x19] = temp_block[0x1A]; block[0x24] = temp_block[0x21];
+        block[0x28] = temp_block[0x28]; block[0x30] = temp_block[0x30];
+        block[0x2C] = temp_block[0x29]; block[0x21] = temp_block[0x22];
         if(last_non_zero_p1 <= 24) goto end;
-        block[0x1D] = temp_block[0x1B]; block[0x12] = temp_block[0x14]; 
-        block[0x0E] = temp_block[0x0D]; block[0x03] = temp_block[0x06]; 
-        block[0x07] = temp_block[0x07]; block[0x0B] = temp_block[0x0E]; 
-        block[0x16] = temp_block[0x15]; block[0x1A] = temp_block[0x1C]; 
+        block[0x1D] = temp_block[0x1B]; block[0x12] = temp_block[0x14];
+        block[0x0E] = temp_block[0x0D]; block[0x03] = temp_block[0x06];
+        block[0x07] = temp_block[0x07]; block[0x0B] = temp_block[0x0E];
+        block[0x16] = temp_block[0x15]; block[0x1A] = temp_block[0x1C];
         if(last_non_zero_p1 <= 32) goto end;
-        block[0x25] = temp_block[0x23]; block[0x29] = temp_block[0x2A]; 
-        block[0x34] = temp_block[0x31]; block[0x38] = temp_block[0x38]; 
-        block[0x3C] = temp_block[0x39]; block[0x31] = temp_block[0x32]; 
-        block[0x2D] = temp_block[0x2B]; block[0x22] = temp_block[0x24]; 
+        block[0x25] = temp_block[0x23]; block[0x29] = temp_block[0x2A];
+        block[0x34] = temp_block[0x31]; block[0x38] = temp_block[0x38];
+        block[0x3C] = temp_block[0x39]; block[0x31] = temp_block[0x32];
+        block[0x2D] = temp_block[0x2B]; block[0x22] = temp_block[0x24];
         if(last_non_zero_p1 <= 40) goto end;
-        block[0x1E] = temp_block[0x1D]; block[0x13] = temp_block[0x16]; 
-        block[0x0F] = temp_block[0x0F]; block[0x17] = temp_block[0x17]; 
-        block[0x1B] = temp_block[0x1E]; block[0x26] = temp_block[0x25]; 
-        block[0x2A] = temp_block[0x2C]; block[0x35] = temp_block[0x33]; 
+        block[0x1E] = temp_block[0x1D]; block[0x13] = temp_block[0x16];
+        block[0x0F] = temp_block[0x0F]; block[0x17] = temp_block[0x17];
+        block[0x1B] = temp_block[0x1E]; block[0x26] = temp_block[0x25];
+        block[0x2A] = temp_block[0x2C]; block[0x35] = temp_block[0x33];
         if(last_non_zero_p1 <= 48) goto end;
-        block[0x39] = temp_block[0x3A]; block[0x3D] = temp_block[0x3B]; 
-        block[0x32] = temp_block[0x34]; block[0x2E] = temp_block[0x2D]; 
-            block[0x23] = temp_block[0x26]; block[0x1F] = temp_block[0x1F]; 
-        block[0x27] = temp_block[0x27]; block[0x2B] = temp_block[0x2E]; 
+        block[0x39] = temp_block[0x3A]; block[0x3D] = temp_block[0x3B];
+        block[0x32] = temp_block[0x34]; block[0x2E] = temp_block[0x2D];
+            block[0x23] = temp_block[0x26]; block[0x1F] = temp_block[0x1F];
+        block[0x27] = temp_block[0x27]; block[0x2B] = temp_block[0x2E];
         if(last_non_zero_p1 <= 56) goto end;
-        block[0x36] = temp_block[0x35]; block[0x3A] = temp_block[0x3C]; 
-        block[0x3E] = temp_block[0x3D]; block[0x33] = temp_block[0x36]; 
-        block[0x2F] = temp_block[0x2F]; block[0x37] = temp_block[0x37]; 
+        block[0x36] = temp_block[0x35]; block[0x3A] = temp_block[0x3C];
+        block[0x3E] = temp_block[0x3D]; block[0x33] = temp_block[0x36];
+        block[0x2F] = temp_block[0x2F]; block[0x37] = temp_block[0x37];
         block[0x3B] = temp_block[0x3E]; block[0x3F] = temp_block[0x3F];
     }else{
         if(last_non_zero_p1 <= 1) goto end;
-        block[0x01] = temp_block[0x01]; 
-        block[0x08] = temp_block[0x08]; block[0x10] = temp_block[0x10]; 
+        block[0x01] = temp_block[0x01];
+        block[0x08] = temp_block[0x08]; block[0x10] = temp_block[0x10];
         if(last_non_zero_p1 <= 4) goto end;
-        block[0x09] = temp_block[0x09]; block[0x02] = temp_block[0x02]; 
-        block[0x03] = temp_block[0x03]; 
+        block[0x09] = temp_block[0x09]; block[0x02] = temp_block[0x02];
+        block[0x03] = temp_block[0x03];
         if(last_non_zero_p1 <= 7) goto end;
-        block[0x0A] = temp_block[0x0A]; block[0x11] = temp_block[0x11]; 
-        block[0x18] = temp_block[0x18]; block[0x20] = temp_block[0x20]; 
+        block[0x0A] = temp_block[0x0A]; block[0x11] = temp_block[0x11];
+        block[0x18] = temp_block[0x18]; block[0x20] = temp_block[0x20];
         if(last_non_zero_p1 <= 11) goto end;
-        block[0x19] = temp_block[0x19]; 
-        block[0x12] = temp_block[0x12]; block[0x0B] = temp_block[0x0B]; 
-        block[0x04] = temp_block[0x04]; block[0x05] = temp_block[0x05]; 
+        block[0x19] = temp_block[0x19];
+        block[0x12] = temp_block[0x12]; block[0x0B] = temp_block[0x0B];
+        block[0x04] = temp_block[0x04]; block[0x05] = temp_block[0x05];
         if(last_non_zero_p1 <= 16) goto end;
-        block[0x0C] = temp_block[0x0C]; block[0x13] = temp_block[0x13]; 
-        block[0x1A] = temp_block[0x1A]; block[0x21] = temp_block[0x21]; 
-        block[0x28] = temp_block[0x28]; block[0x30] = temp_block[0x30]; 
-        block[0x29] = temp_block[0x29]; block[0x22] = temp_block[0x22]; 
+        block[0x0C] = temp_block[0x0C]; block[0x13] = temp_block[0x13];
+        block[0x1A] = temp_block[0x1A]; block[0x21] = temp_block[0x21];
+        block[0x28] = temp_block[0x28]; block[0x30] = temp_block[0x30];
+        block[0x29] = temp_block[0x29]; block[0x22] = temp_block[0x22];
         if(last_non_zero_p1 <= 24) goto end;
-        block[0x1B] = temp_block[0x1B]; block[0x14] = temp_block[0x14]; 
-        block[0x0D] = temp_block[0x0D]; block[0x06] = temp_block[0x06]; 
-        block[0x07] = temp_block[0x07]; block[0x0E] = temp_block[0x0E]; 
-        block[0x15] = temp_block[0x15]; block[0x1C] = temp_block[0x1C]; 
+        block[0x1B] = temp_block[0x1B]; block[0x14] = temp_block[0x14];
+        block[0x0D] = temp_block[0x0D]; block[0x06] = temp_block[0x06];
+        block[0x07] = temp_block[0x07]; block[0x0E] = temp_block[0x0E];
+        block[0x15] = temp_block[0x15]; block[0x1C] = temp_block[0x1C];
         if(last_non_zero_p1 <= 32) goto end;
-        block[0x23] = temp_block[0x23]; block[0x2A] = temp_block[0x2A]; 
-        block[0x31] = temp_block[0x31]; block[0x38] = temp_block[0x38]; 
-        block[0x39] = temp_block[0x39]; block[0x32] = temp_block[0x32]; 
-        block[0x2B] = temp_block[0x2B]; block[0x24] = temp_block[0x24]; 
+        block[0x23] = temp_block[0x23]; block[0x2A] = temp_block[0x2A];
+        block[0x31] = temp_block[0x31]; block[0x38] = temp_block[0x38];
+        block[0x39] = temp_block[0x39]; block[0x32] = temp_block[0x32];
+        block[0x2B] = temp_block[0x2B]; block[0x24] = temp_block[0x24];
         if(last_non_zero_p1 <= 40) goto end;
-        block[0x1D] = temp_block[0x1D]; block[0x16] = temp_block[0x16]; 
-        block[0x0F] = temp_block[0x0F]; block[0x17] = temp_block[0x17]; 
-        block[0x1E] = temp_block[0x1E]; block[0x25] = temp_block[0x25]; 
-        block[0x2C] = temp_block[0x2C]; block[0x33] = temp_block[0x33]; 
+        block[0x1D] = temp_block[0x1D]; block[0x16] = temp_block[0x16];
+        block[0x0F] = temp_block[0x0F]; block[0x17] = temp_block[0x17];
+        block[0x1E] = temp_block[0x1E]; block[0x25] = temp_block[0x25];
+        block[0x2C] = temp_block[0x2C]; block[0x33] = temp_block[0x33];
         if(last_non_zero_p1 <= 48) goto end;
-        block[0x3A] = temp_block[0x3A]; block[0x3B] = temp_block[0x3B]; 
-        block[0x34] = temp_block[0x34]; block[0x2D] = temp_block[0x2D]; 
-        block[0x26] = temp_block[0x26]; block[0x1F] = temp_block[0x1F]; 
-        block[0x27] = temp_block[0x27]; block[0x2E] = temp_block[0x2E]; 
+        block[0x3A] = temp_block[0x3A]; block[0x3B] = temp_block[0x3B];
+        block[0x34] = temp_block[0x34]; block[0x2D] = temp_block[0x2D];
+        block[0x26] = temp_block[0x26]; block[0x1F] = temp_block[0x1F];
+        block[0x27] = temp_block[0x27]; block[0x2E] = temp_block[0x2E];
         if(last_non_zero_p1 <= 56) goto end;
-        block[0x35] = temp_block[0x35]; block[0x3C] = temp_block[0x3C]; 
-        block[0x3D] = temp_block[0x3D]; block[0x36] = temp_block[0x36]; 
-        block[0x2F] = temp_block[0x2F]; block[0x37] = temp_block[0x37]; 
+        block[0x35] = temp_block[0x35]; block[0x3C] = temp_block[0x3C];
+        block[0x3D] = temp_block[0x3D]; block[0x36] = temp_block[0x36];
+        block[0x2F] = temp_block[0x2F]; block[0x37] = temp_block[0x37];
         block[0x3E] = temp_block[0x3E]; block[0x3F] = temp_block[0x3F];
     }
     end:
diff --git a/libavcodec/i386/simple_idct_mmx.c b/libavcodec/i386/simple_idct_mmx.c
index 92a366f217..7b5084c7eb 100644
--- a/libavcodec/i386/simple_idct_mmx.c
+++ b/libavcodec/i386/simple_idct_mmx.c
@@ -60,19 +60,19 @@ static const int16_t __attribute__((aligned(8))) coeffs[]= {
 
  C4,  C4,  C4,  C4,
  C4, -C4,  C4, -C4,
- 
+
  C2,  C6,  C2,  C6,
  C6, -C2,  C6, -C2,
- 
+
  C1,  C3,  C1,  C3,
  C5,  C7,  C5,  C7,
- 
+
  C3, -C7,  C3, -C7,
 -C1, -C5, -C1, -C5,
- 
+
  C5, -C1,  C5, -C1,
  C7,  C3,  C7,  C3,
- 
+
  C7, -C5,  C7, -C5,
  C3, -C1,  C3, -C1
 };
@@ -357,7 +357,7 @@ static inline void idct(int16_t *block)
 	"movd %%mm4, 64+" #dst "		\n\t"\
 	"movd %%mm5, 80+" #dst "		\n\t"\
 
-	
+
 #define DC_COND_ROW_IDCT(src0, src4, src1, src5, dst, rounder, shift) \
 	"movq " #src0 ", %%mm0			\n\t" /* R4	R0	r4	r0 */\
 	"movq " #src4 ", %%mm1			\n\t" /* R6	R2	r6	r2 */\
@@ -857,7 +857,7 @@ Z_COND_IDCT(  96(%0),104(%0),112(%0),120(%0), 96(%1),paddd (%2), 11, 5f)
 	"packssdw %%mm5, %%mm5			\n\t" /* A2-B2	a2-b2 */\
 	"movd %%mm6, 48+" #dst "		\n\t"\
 	"movd %%mm1, 64+" #dst "		\n\t"\
-	"movd %%mm5, 80+" #dst "		\n\t"	
+	"movd %%mm5, 80+" #dst "		\n\t"
 
 //IDCT(  src0,   src4,   src1,    src5,    dst, rounder, shift)
 IDCT(    (%1), 64(%1), 32(%1),  96(%1),  0(%0),/nop, 20)
@@ -924,7 +924,7 @@ Z_COND_IDCT(  96(%0),104(%0),112(%0),120(%0), 96(%1),paddd (%2), 11, 7f)
 	"packssdw %%mm5, %%mm5			\n\t" /* A2-B2	a2-b2 */\
 	"movd %%mm6, 48+" #dst "		\n\t"\
 	"movd %%mm1, 64+" #dst "		\n\t"\
-	"movd %%mm5, 80+" #dst "		\n\t"	
+	"movd %%mm5, 80+" #dst "		\n\t"
 
 
 //IDCT(  src0,   src4,   src1,    src5,    dst, rounder, shift)
@@ -1137,8 +1137,8 @@ IDCT(  24(%1), 88(%1), 56(%1), 120(%1), 12(%0),/nop, 20)
 	"packssdw %%mm1, %%mm6			\n\t" /* A3+B3	a3+b3 */\
 	"movq %%mm6, 48+" #dst "		\n\t"\
 	"movq %%mm6, 64+" #dst "		\n\t"\
-	"movq %%mm5, 80+" #dst "		\n\t"	
-	
+	"movq %%mm5, 80+" #dst "		\n\t"
+
 
 //IDCT(  src0,   src4,   src1,    src5,    dst, rounder, shift)
 IDCT(    0(%1), 64(%1), 32(%1),  96(%1),  0(%0),/nop, 20)
@@ -1214,7 +1214,7 @@ IDCT(  16(%1), 80(%1), 48(%1), 112(%1),  8(%0),/nop, 20)
 	"packssdw %%mm5, %%mm5			\n\t" /* A2-B2	a2-b2 */\
 	"movd %%mm4, 64+" #dst "		\n\t"\
 	"movd %%mm5, 80+" #dst "		\n\t"
-	
+
 
 //IDCT(  src0,   src4,   src1,    src5,    dst, rounder, shift)
 IDCT(    (%1), 64(%1), 32(%1),  96(%1),  0(%0),/nop, 20)
@@ -1256,7 +1256,7 @@ IDCT(  24(%1), 88(%1), 56(%1), 120(%1), 12(%0),/nop, 20)
 	"movq %%mm0, 32+" #dst "		\n\t"\
 	"movq %%mm4, 48+" #dst "		\n\t"\
 	"movq %%mm4, 64+" #dst "		\n\t"\
-	"movq %%mm0, 80+" #dst "		\n\t"	
+	"movq %%mm0, 80+" #dst "		\n\t"
 
 //IDCT(  src0,   src4,   src1,    src5,    dst, rounder, shift)
 IDCT(   0(%1), 64(%1), 32(%1),  96(%1),  0(%0),/nop, 20)
@@ -1277,7 +1277,7 @@ Input
  12 32 16 36 52 72 56 76
  05 45 07 47 25 65 27 67
  15 35 17 37 55 75 57 77
-  
+
 Temp
  00 04 10 14 20 24 30 34
  40 44 50 54 60 64 70 74
diff --git a/libavcodec/i386/vp3dsp_mmx.c b/libavcodec/i386/vp3dsp_mmx.c
index 3d220c1d45..4aa1a5f403 100644
--- a/libavcodec/i386/vp3dsp_mmx.c
+++ b/libavcodec/i386/vp3dsp_mmx.c
@@ -208,7 +208,7 @@ static const uint16_t idct_cosine_table[7] = {
     I(1) = d1 c1 b1 a1
     I(2) = d2 c2 b2 a2
     I(3) = d3 c3 b3 a3
-    
+
     J(4) = h0 g0 f0 e0
     J(5) = h1 g1 f1 e1
     J(6) = h2 g2 f2 e2
diff --git a/libavcodec/i386/vp3dsp_sse2.c b/libavcodec/i386/vp3dsp_sse2.c
index ed17891bfa..fcc511b651 100644
--- a/libavcodec/i386/vp3dsp_sse2.c
+++ b/libavcodec/i386/vp3dsp_sse2.c
@@ -36,21 +36,21 @@ static const unsigned short __align16 SSE2_dequant_const[] =
 };
 
 static const unsigned int __align16 eight_data[] =
-{ 
-    0x00080008, 
+{
+    0x00080008,
+    0x00080008,
     0x00080008,
-    0x00080008, 
-    0x00080008 
-}; 
+    0x00080008
+};
 
 static const unsigned short __align16 SSE2_idct_data[7 * 8] =
 {
-    64277,64277,64277,64277,64277,64277,64277,64277, 
-    60547,60547,60547,60547,60547,60547,60547,60547, 
-    54491,54491,54491,54491,54491,54491,54491,54491, 
-    46341,46341,46341,46341,46341,46341,46341,46341, 
-    36410,36410,36410,36410,36410,36410,36410,36410, 
-    25080,25080,25080,25080,25080,25080,25080,25080, 
+    64277,64277,64277,64277,64277,64277,64277,64277,
+    60547,60547,60547,60547,60547,60547,60547,60547,
+    54491,54491,54491,54491,54491,54491,54491,54491,
+    46341,46341,46341,46341,46341,46341,46341,46341,
+    36410,36410,36410,36410,36410,36410,36410,36410,
+    25080,25080,25080,25080,25080,25080,25080,25080,
     12785,12785,12785,12785,12785,12785,12785,12785
 };
 
@@ -820,6 +820,6 @@ void ff_vp3_idct_sse2(int16_t *input_data)
     SSE2_Row_IDCT();
 
     SSE2_Transpose();
-        
+
     SSE2_Column_IDCT();
 }