diff --git a/Makefile b/Makefile
index cb23b917b5..f4057c538c 100644
--- a/Makefile
+++ b/Makefile
@@ -27,8 +27,6 @@ ALLPROGS    = $(BASENAMES:%=%$(EXESUF))
 ALLPROGS_G  = $(BASENAMES:%=%_g$(EXESUF))
 ALLMANPAGES = $(BASENAMES:%=%.1)
 
-ALLFFLIBS = avcodec avdevice avfilter avformat avutil postproc swscale
-
 FFLIBS-$(CONFIG_AVDEVICE) += avdevice
 FFLIBS-$(CONFIG_AVFILTER) += avfilter
 FFLIBS-$(CONFIG_AVFORMAT) += avformat
diff --git a/common.mak b/common.mak
index 8daf5945d7..b1f62d90f7 100644
--- a/common.mak
+++ b/common.mak
@@ -20,6 +20,8 @@ $(foreach VAR,$(SILENT),$(eval override $(VAR) = @$($(VAR))))
 $(eval INSTALL = @$(call ECHO,INSTALL,$$(^:$(SRC_DIR)/%=%)); $(INSTALL))
 endif
 
+ALLFFLIBS = avcodec avdevice avfilter avformat avutil postproc swscale
+
 # NASM requires -I path terminated with /
 IFLAGS     := -I. -I$(SRC_PATH)/
 CPPFLAGS   := $(IFLAGS) $(CPPFLAGS)
@@ -27,7 +29,7 @@ CFLAGS     += $(ECFLAGS)
 CCFLAGS     = $(CFLAGS)
 YASMFLAGS  += $(IFLAGS) -Pconfig.asm
 HOSTCFLAGS += $(IFLAGS)
-LDFLAGS    += $(ALLFFLIBS:%=-Llib%)
+LDFLAGS    := $(ALLFFLIBS:%=-Llib%) $(LDFLAGS)
 
 define COMPILE
        $($(1)DEP)
diff --git a/configure b/configure
index 80a160e22b..5fa1d96eb7 100755
--- a/configure
+++ b/configure
@@ -2049,6 +2049,7 @@ elif $cc -V 2>&1 | grep -q Sun; then
     cc_ident=$($cc -V 2>&1 | head -n1 | cut -d' ' -f 2-)
     DEPEND_CMD='$(DEPCC) $(DEPFLAGS) $< | sed -e "1s,^.*: ,$@: ," -e "\$$!s,\$$, \\\," -e "1!s,^.*: , ," > $(@:.o=.d)'
     DEPFLAGS='$(CPPFLAGS) $(CFLAGS) -xM1'
+    add_ldflags -xc99
     speed_cflags='-O5'
     size_cflags='-O5 -xspace'
     filter_cflags=suncc_flags
diff --git a/libavcodec/acelp_pitch_delay.c b/libavcodec/acelp_pitch_delay.c
index 66d65a6903..c2155a5c0c 100644
--- a/libavcodec/acelp_pitch_delay.c
+++ b/libavcodec/acelp_pitch_delay.c
@@ -105,20 +105,9 @@ int16_t ff_acelp_decode_gain_code(
     for(i=0; i<ma_pred_order; i++)
         mr_energy += quant_energy[i] * ma_prediction_coeff[i];
 
-#ifdef G729_BITEXACT
-    mr_energy += (((-6165LL * ff_log2(dsp->scalarproduct_int16(fc_v, fc_v, subframe_size, 0))) >> 3) & ~0x3ff);
-
-    mr_energy = (5439 * (mr_energy >> 15)) >> 8;           // (0.15) = (0.15) * (7.23)
-
-    return bidir_sal(
-               ((ff_exp2(mr_energy & 0x7fff) + 16) >> 5) * (gain_corr_factor >> 1),
-               (mr_energy >> 15) - 25
-           );
-#else
     mr_energy = gain_corr_factor * exp(M_LN10 / (20 << 23) * mr_energy) /
                 sqrt(dsp->scalarproduct_int16(fc_v, fc_v, subframe_size, 0));
     return mr_energy >> 12;
-#endif
 }
 
 float ff_amr_set_fixed_gain(float fixed_gain_factor, float fixed_mean_energy,
diff --git a/libavcodec/acelp_vectors.c b/libavcodec/acelp_vectors.c
index a44ab8cfe6..3d6ef83964 100644
--- a/libavcodec/acelp_vectors.c
+++ b/libavcodec/acelp_vectors.c
@@ -48,26 +48,6 @@ const uint8_t ff_fc_2pulses_9bits_track1_gray[16] =
   28, 26,
 };
 
-const uint8_t ff_fc_2pulses_9bits_track2_gray[32] =
-{
-  0,  2,
-  5,  4,
-  12, 10,
-  7,  9,
-  25, 24,
-  20, 22,
-  14, 15,
-  19, 17,
-  36, 31,
-  21, 26,
-  1,  6,
-  16, 11,
-  27, 29,
-  32, 30,
-  39, 37,
-  34, 35,
-};
-
 const uint8_t ff_fc_4pulses_8bits_tracks_13[16] =
 {
   0, 5, 10, 15, 20, 25, 30, 35, 40, 45, 50, 55, 60, 65, 70, 75,
diff --git a/libavcodec/acelp_vectors.h b/libavcodec/acelp_vectors.h
index f3bc781446..557c07b35e 100644
--- a/libavcodec/acelp_vectors.h
+++ b/libavcodec/acelp_vectors.h
@@ -81,37 +81,6 @@ extern const uint8_t ff_fc_4pulses_8bits_track_4[32];
 extern const uint8_t ff_fc_2pulses_9bits_track1[16];
 extern const uint8_t ff_fc_2pulses_9bits_track1_gray[16];
 
-/**
- * Track|Pulse|        Positions
- * -----------------------------------------
- *  2   | 1   | 0, 7, 14, 20, 27, 34,  1, 21
- *      |     | 2, 9, 15, 22, 29, 35,  6, 26
- *      |     | 4,10, 17, 24, 30, 37, 11, 31
- *      |     | 5,12, 19, 25, 32, 39, 16, 36
- * -----------------------------------------
- *
- * @remark Track in the table should be read top-to-bottom, left-to-right.
- *
- * @note (EE.1) This table (from the reference code) does not comply with
- *              the specification.
- *              The specification contains the following table:
- *
- * Track|Pulse|        Positions
- * -----------------------------------------
- *  2   | 1   | 0, 5, 10, 15, 20, 25, 30, 35
- *      |     | 1, 6, 11, 16, 21, 26, 31, 36
- *      |     | 2, 7, 12, 17, 22, 27, 32, 37
- *      |     | 4, 9, 14, 19, 24, 29, 34, 39
- *
- * -----------------------------------------
- *
- * @note (EE.2) Reference G.729D code also uses gray decoding for each
- *              pulse index before looking up the value in the table.
- *
- * Used in G.729 @@6.4k (with gray coding)
- */
-extern const uint8_t ff_fc_2pulses_9bits_track2_gray[32];
-
 /**
  * b60 hamming windowed sinc function coefficients
  */
diff --git a/libavcodec/alpha/dsputil_alpha.c b/libavcodec/alpha/dsputil_alpha.c
index 6ce3f4bf15..d0d2a621c2 100644
--- a/libavcodec/alpha/dsputil_alpha.c
+++ b/libavcodec/alpha/dsputil_alpha.c
@@ -335,7 +335,7 @@ void dsputil_init_alpha(DSPContext* c, AVCodecContext *avctx)
     put_pixels_clamped_axp_p = c->put_pixels_clamped;
     add_pixels_clamped_axp_p = c->add_pixels_clamped;
 
-    if (!avctx->lowres &&
+    if (!avctx->lowres && avctx->bits_per_raw_sample <= 8 &&
         (avctx->idct_algo == FF_IDCT_AUTO ||
          avctx->idct_algo == FF_IDCT_SIMPLEALPHA)) {
         c->idct_put = ff_simple_idct_put_axp;
diff --git a/libavcodec/amrnbdec.c b/libavcodec/amrnbdec.c
index b8d826e139..8a9e77709a 100644
--- a/libavcodec/amrnbdec.c
+++ b/libavcodec/amrnbdec.c
@@ -83,7 +83,7 @@
 /** Maximum sharpening factor
  *
  * The specification says 0.8, which should be 13107, but the reference C code
- * uses 13017 instead. (Amusingly the same applies to SHARP_MAX in g729dec.c.)
+ * uses 13017 instead. (Amusingly the same applies to SHARP_MAX in bitexact G.729.)
  */
 #define SHARP_MAX 0.79449462890625
 
diff --git a/libavcodec/arm/dsputil_init_arm.c b/libavcodec/arm/dsputil_init_arm.c
index 0351412761..5728d709e7 100644
--- a/libavcodec/arm/dsputil_init_arm.c
+++ b/libavcodec/arm/dsputil_init_arm.c
@@ -80,7 +80,7 @@ void dsputil_init_arm(DSPContext* c, AVCodecContext *avctx)
     ff_put_pixels_clamped = c->put_pixels_clamped;
     ff_add_pixels_clamped = c->add_pixels_clamped;
 
-    if (!avctx->lowres) {
+    if (!avctx->lowres && avctx->bits_per_raw_sample <= 8) {
         if(avctx->idct_algo == FF_IDCT_AUTO ||
            avctx->idct_algo == FF_IDCT_ARM){
             c->idct_put              = j_rev_dct_arm_put;
diff --git a/libavcodec/arm/dsputil_init_armv5te.c b/libavcodec/arm/dsputil_init_armv5te.c
index 750e5147b4..2390aabb62 100644
--- a/libavcodec/arm/dsputil_init_armv5te.c
+++ b/libavcodec/arm/dsputil_init_armv5te.c
@@ -29,8 +29,9 @@ void ff_prefetch_arm(void *mem, int stride, int h);
 
 void av_cold ff_dsputil_init_armv5te(DSPContext* c, AVCodecContext *avctx)
 {
-    if (!avctx->lowres && (avctx->idct_algo == FF_IDCT_AUTO ||
-                           avctx->idct_algo == FF_IDCT_SIMPLEARMV5TE)) {
+    if (!avctx->lowres && avctx->bits_per_raw_sample <= 8 &&
+        (avctx->idct_algo == FF_IDCT_AUTO ||
+         avctx->idct_algo == FF_IDCT_SIMPLEARMV5TE)) {
         c->idct_put              = ff_simple_idct_put_armv5te;
         c->idct_add              = ff_simple_idct_add_armv5te;
         c->idct                  = ff_simple_idct_armv5te;
diff --git a/libavcodec/arm/dsputil_init_armv6.c b/libavcodec/arm/dsputil_init_armv6.c
index 9acea4a1d6..1fc636b12d 100644
--- a/libavcodec/arm/dsputil_init_armv6.c
+++ b/libavcodec/arm/dsputil_init_armv6.c
@@ -74,8 +74,9 @@ void av_cold ff_dsputil_init_armv6(DSPContext* c, AVCodecContext *avctx)
 {
     const int high_bit_depth = avctx->codec_id == CODEC_ID_H264 && avctx->bits_per_raw_sample > 8;
 
-    if (!avctx->lowres && (avctx->idct_algo == FF_IDCT_AUTO ||
-                           avctx->idct_algo == FF_IDCT_SIMPLEARMV6)) {
+    if (!avctx->lowres && avctx->bits_per_raw_sample <= 8 &&
+        (avctx->idct_algo == FF_IDCT_AUTO ||
+         avctx->idct_algo == FF_IDCT_SIMPLEARMV6)) {
         c->idct_put              = ff_simple_idct_put_armv6;
         c->idct_add              = ff_simple_idct_add_armv6;
         c->idct                  = ff_simple_idct_armv6;
diff --git a/libavcodec/arm/dsputil_init_neon.c b/libavcodec/arm/dsputil_init_neon.c
index c818b33fef..c8b6b6ebae 100644
--- a/libavcodec/arm/dsputil_init_neon.c
+++ b/libavcodec/arm/dsputil_init_neon.c
@@ -177,7 +177,7 @@ void ff_dsputil_init_neon(DSPContext *c, AVCodecContext *avctx)
 {
     const int high_bit_depth = avctx->codec_id == CODEC_ID_H264 && avctx->bits_per_raw_sample > 8;
 
-    if (!avctx->lowres) {
+    if (!avctx->lowres && avctx->bits_per_raw_sample <= 8) {
         if (avctx->idct_algo == FF_IDCT_AUTO ||
             avctx->idct_algo == FF_IDCT_SIMPLENEON) {
             c->idct_put              = ff_simple_idct_put_neon;
diff --git a/libavcodec/bfin/dsputil_bfin.c b/libavcodec/bfin/dsputil_bfin.c
index 5b94472326..eade153440 100644
--- a/libavcodec/bfin/dsputil_bfin.c
+++ b/libavcodec/bfin/dsputil_bfin.c
@@ -256,16 +256,18 @@ void dsputil_init_bfin( DSPContext* c, AVCodecContext *avctx )
     if (avctx->dct_algo == FF_DCT_AUTO)
         c->fdct               = ff_bfin_fdct;
 
-    if (avctx->idct_algo==FF_IDCT_VP3) {
-        c->idct_permutation_type = FF_NO_IDCT_PERM;
-        c->idct               = ff_bfin_vp3_idct;
-        c->idct_add           = ff_bfin_vp3_idct_add;
-        c->idct_put           = ff_bfin_vp3_idct_put;
-    } else if (avctx->idct_algo == FF_IDCT_AUTO) {
-        c->idct_permutation_type = FF_NO_IDCT_PERM;
-        c->idct               = ff_bfin_idct;
-        c->idct_add           = bfin_idct_add;
-        c->idct_put           = bfin_idct_put;
+    if (avctx->bits_per_raw_sample <= 8) {
+        if (avctx->idct_algo == FF_IDCT_VP3) {
+            c->idct_permutation_type = FF_NO_IDCT_PERM;
+            c->idct                  = ff_bfin_vp3_idct;
+            c->idct_add              = ff_bfin_vp3_idct_add;
+            c->idct_put              = ff_bfin_vp3_idct_put;
+        } else if (avctx->idct_algo == FF_IDCT_AUTO) {
+            c->idct_permutation_type = FF_NO_IDCT_PERM;
+            c->idct                  = ff_bfin_idct;
+            c->idct_add              = bfin_idct_add;
+            c->idct_put              = bfin_idct_put;
+        }
     }
 }
 
diff --git a/libavcodec/celp_math.c b/libavcodec/celp_math.c
index 09111da819..de33109b44 100644
--- a/libavcodec/celp_math.c
+++ b/libavcodec/celp_math.c
@@ -27,82 +27,6 @@
 #include "avcodec.h"
 #include "celp_math.h"
 
-#ifdef G729_BITEXACT
-/**
- * Cosine table: base_cos[i] = (1<<15) * cos(i*PI/64)
- */
-static const int16_t base_cos[64] =
-{
-  32767,  32729,  32610,  32413,  32138,  31786,  31357,  30853,
-  30274,  29622,  28899,  28106,  27246,  26320,  25330,  24279,
-  23170,  22006,  20788,  19520,  18205,  16846,  15447,  14010,
-  12540,  11039,   9512,   7962,   6393,   4808,   3212,   1608,
-      0,  -1608,  -3212,  -4808,  -6393,  -7962,  -9512, -11039,
- -12540, -14010, -15447, -16846, -18205, -19520, -20788, -22006,
- -23170, -24279, -25330, -26320, -27246, -28106, -28899, -29622,
- -30274, -30853, -31357, -31786, -32138, -32413, -32610, -32729
-};
-
-/**
- * Slope used to compute cos(x)
- *
- * cos(ind*64+offset) = base_cos[ind]+offset*slope_cos[ind]
- * values multiplied by 1<<19
- */
-static const int16_t slope_cos[64] =
-{
-   -632,  -1893,  -3150,  -4399,  -5638,  -6863,  -8072,  -9261,
- -10428, -11570, -12684, -13767, -14817, -15832, -16808, -17744,
- -18637, -19486, -20287, -21039, -21741, -22390, -22986, -23526,
- -24009, -24435, -24801, -25108, -25354, -25540, -25664, -25726,
- -25726, -25664, -25540, -25354, -25108, -24801, -24435, -24009,
- -23526, -22986, -22390, -21741, -21039, -20287, -19486, -18637,
- -17744, -16808, -15832, -14817, -13767, -12684, -11570, -10428,
-  -9261,  -8072,  -6863,  -5638,  -4399,  -3150,  -1893,   -632
-};
-
-/**
- * Table used to compute exp2(x)
- *
- * tab_exp2[i] = (1<<14) * exp2(i/32) = 2^(i/32) i=0..32
- */
-static const uint16_t tab_exp2[33] =
-{
-  16384, 16743, 17109, 17484, 17867, 18258, 18658, 19066, 19484, 19911,
-  20347, 20792, 21247, 21713, 22188, 22674, 23170, 23678, 24196, 24726,
-  25268, 25821, 26386, 26964, 27554, 28158, 28774, 29405, 30048, 30706,
-  31379, 32066, 32767
-};
-
-int16_t ff_cos(uint16_t arg)
-{
-    uint8_t offset= arg;
-    uint8_t ind = arg >> 8;
-
-    assert(arg < 0x4000);
-
-    return FFMAX(base_cos[ind] + ((slope_cos[ind] * offset) >> 12), -0x8000);
-}
-
-int ff_exp2(uint16_t power)
-{
-    uint16_t frac_x0;
-    uint16_t frac_dx;
-    int result;
-
-    assert(power <= 0x7fff);
-
-    frac_x0 = power >> 10;
-    frac_dx = (power & 0x03ff) << 5;
-
-    result = tab_exp2[frac_x0] << 15;
-    result += frac_dx * (tab_exp2[frac_x0+1] - tab_exp2[frac_x0]);
-
-    return result >> 10;
-}
-
-#else // G729_BITEXACT
-
 /**
  * Cosine table: base_cos[i] = (1<<15) * cos(i*PI/64)
  */
@@ -154,8 +78,6 @@ int ff_exp2(uint16_t power)
     return result + ((result*(power&31)*89)>>22);
 }
 
-#endif // else G729_BITEXACT
-
 /**
  * Table used to compute log2(x)
  *
@@ -163,17 +85,10 @@ int ff_exp2(uint16_t power)
  */
 static const uint16_t tab_log2[33] =
 {
-#ifdef G729_BITEXACT
-      0,   1455,   2866,   4236,   5568,   6863,   8124,   9352,
-  10549,  11716,  12855,  13967,  15054,  16117,  17156,  18172,
-  19167,  20142,  21097,  22033,  22951,  23852,  24735,  25603,
-  26455,  27291,  28113,  28922,  29716,  30497,  31266,  32023,  32767,
-#else
       4,   1459,   2870,   4240,   5572,   6867,   8127,   9355,
   10552,  11719,  12858,  13971,  15057,  16120,  17158,  18175,
   19170,  20145,  21100,  22036,  22954,  23854,  24738,  25605,
   26457,  27294,  28116,  28924,  29719,  30500,  31269,  32025,  32769,
-#endif
 };
 
 int ff_log2(uint32_t value)
diff --git a/libavcodec/dct-test.c b/libavcodec/dct-test.c
index d9ec0a5194..de6582ca08 100644
--- a/libavcodec/dct-test.c
+++ b/libavcodec/dct-test.c
@@ -111,7 +111,7 @@ static const struct algo idct_tab[] = {
     { "FAANI",          ff_faanidct,           NO_PERM  },
     { "REF-DBL",        ff_ref_idct,           NO_PERM  },
     { "INT",            j_rev_dct,             MMX_PERM },
-    { "SIMPLE-C",       ff_simple_idct,        NO_PERM  },
+    { "SIMPLE-C",       ff_simple_idct_8,      NO_PERM  },
 
 #if HAVE_MMX
 #if CONFIG_GPL
diff --git a/libavcodec/dnxhdenc.c b/libavcodec/dnxhdenc.c
index 8b63bbb19b..57ae8e205b 100644
--- a/libavcodec/dnxhdenc.c
+++ b/libavcodec/dnxhdenc.c
@@ -41,7 +41,7 @@ static const AVClass class = { "dnxhd", av_default_item_name, options, LIBAVUTIL
 
 #define LAMBDA_FRAC_BITS 10
 
-static av_always_inline void dnxhd_get_pixels_8x4(DCTELEM *restrict block, const uint8_t *pixels, int line_size)
+static void dnxhd_get_pixels_8x4(DCTELEM *restrict block, const uint8_t *pixels, int line_size)
 {
     int i;
     for (i = 0; i < 4; i++) {
@@ -52,10 +52,10 @@ static av_always_inline void dnxhd_get_pixels_8x4(DCTELEM *restrict block, const
         pixels += line_size;
         block += 8;
     }
-    memcpy(block   , block- 8, sizeof(*block)*8);
-    memcpy(block+ 8, block-16, sizeof(*block)*8);
-    memcpy(block+16, block-24, sizeof(*block)*8);
-    memcpy(block+24, block-32, sizeof(*block)*8);
+    memcpy(block,      block -  8, sizeof(*block) * 8);
+    memcpy(block +  8, block - 16, sizeof(*block) * 8);
+    memcpy(block + 16, block - 24, sizeof(*block) * 8);
+    memcpy(block + 24, block - 32, sizeof(*block) * 8);
 }
 
 static int dnxhd_init_vlc(DNXHDEncContext *ctx)
@@ -64,9 +64,9 @@ static int dnxhd_init_vlc(DNXHDEncContext *ctx)
     int max_level = 1<<(ctx->cid_table->bit_depth+2);
 
     FF_ALLOCZ_OR_GOTO(ctx->m.avctx, ctx->vlc_codes, max_level*4*sizeof(*ctx->vlc_codes), fail);
-    FF_ALLOCZ_OR_GOTO(ctx->m.avctx, ctx->vlc_bits , max_level*4*sizeof(*ctx->vlc_bits ), fail);
-    FF_ALLOCZ_OR_GOTO(ctx->m.avctx, ctx->run_codes, 63*2                               , fail);
-    FF_ALLOCZ_OR_GOTO(ctx->m.avctx, ctx->run_bits , 63                                 , fail);
+    FF_ALLOCZ_OR_GOTO(ctx->m.avctx, ctx->vlc_bits,  max_level*4*sizeof(*ctx->vlc_bits) , fail);
+    FF_ALLOCZ_OR_GOTO(ctx->m.avctx, ctx->run_codes, 63*2,                                fail);
+    FF_ALLOCZ_OR_GOTO(ctx->m.avctx, ctx->run_bits,  63,                                  fail);
 
     ctx->vlc_codes += max_level*2;
     ctx->vlc_bits  += max_level*2;
@@ -119,8 +119,8 @@ static int dnxhd_init_qmat(DNXHDEncContext *ctx, int lbias, int cbias)
     uint16_t weight_matrix[64] = {1,}; // convert_matrix needs uint16_t*
     int qscale, i;
 
-    FF_ALLOCZ_OR_GOTO(ctx->m.avctx, ctx->qmatrix_l,   (ctx->m.avctx->qmax+1) * 64 *     sizeof(int)     , fail);
-    FF_ALLOCZ_OR_GOTO(ctx->m.avctx, ctx->qmatrix_c,   (ctx->m.avctx->qmax+1) * 64 *     sizeof(int)     , fail);
+    FF_ALLOCZ_OR_GOTO(ctx->m.avctx, ctx->qmatrix_l,   (ctx->m.avctx->qmax+1) * 64 *     sizeof(int),      fail);
+    FF_ALLOCZ_OR_GOTO(ctx->m.avctx, ctx->qmatrix_c,   (ctx->m.avctx->qmax+1) * 64 *     sizeof(int),      fail);
     FF_ALLOCZ_OR_GOTO(ctx->m.avctx, ctx->qmatrix_l16, (ctx->m.avctx->qmax+1) * 64 * 2 * sizeof(uint16_t), fail);
     FF_ALLOCZ_OR_GOTO(ctx->m.avctx, ctx->qmatrix_c16, (ctx->m.avctx->qmax+1) * 64 * 2 * sizeof(uint16_t), fail);
 
@@ -218,7 +218,7 @@ static int dnxhd_encode_init(AVCodecContext *avctx)
     FF_ALLOCZ_OR_GOTO(ctx->m.avctx, ctx->slice_size, ctx->m.mb_height*sizeof(uint32_t), fail);
     FF_ALLOCZ_OR_GOTO(ctx->m.avctx, ctx->slice_offs, ctx->m.mb_height*sizeof(uint32_t), fail);
     FF_ALLOCZ_OR_GOTO(ctx->m.avctx, ctx->mb_bits,    ctx->m.mb_num   *sizeof(uint16_t), fail);
-    FF_ALLOCZ_OR_GOTO(ctx->m.avctx, ctx->mb_qscale,  ctx->m.mb_num   *sizeof(uint8_t) , fail);
+    FF_ALLOCZ_OR_GOTO(ctx->m.avctx, ctx->mb_qscale,  ctx->m.mb_num   *sizeof(uint8_t),  fail);
 
     ctx->frame.key_frame = 1;
     ctx->frame.pict_type = AV_PICTURE_TYPE_I;
@@ -341,7 +341,7 @@ static av_always_inline int dnxhd_ssd_block(DCTELEM *qblock, DCTELEM *block)
     int score = 0;
     int i;
     for (i = 0; i < 64; i++)
-        score += (block[i]-qblock[i])*(block[i]-qblock[i]);
+        score += (block[i] - qblock[i]) * (block[i] - qblock[i]);
     return score;
 }
 
@@ -369,26 +369,28 @@ static av_always_inline void dnxhd_get_blocks(DNXHDEncContext *ctx, int mb_x, in
     const uint8_t *ptr_v = ctx->thread[0]->src[2] + ((mb_y << 4) * ctx->m.uvlinesize) + (mb_x << 3);
     DSPContext *dsp = &ctx->m.dsp;
 
-    dsp->get_pixels(ctx->blocks[0], ptr_y    , ctx->m.linesize);
+    dsp->get_pixels(ctx->blocks[0], ptr_y,     ctx->m.linesize);
     dsp->get_pixels(ctx->blocks[1], ptr_y + 8, ctx->m.linesize);
-    dsp->get_pixels(ctx->blocks[2], ptr_u    , ctx->m.uvlinesize);
-    dsp->get_pixels(ctx->blocks[3], ptr_v    , ctx->m.uvlinesize);
+    dsp->get_pixels(ctx->blocks[2], ptr_u,     ctx->m.uvlinesize);
+    dsp->get_pixels(ctx->blocks[3], ptr_v,     ctx->m.uvlinesize);
 
     if (mb_y+1 == ctx->m.mb_height && ctx->m.avctx->height == 1080) {
         if (ctx->interlaced) {
-            ctx->get_pixels_8x4_sym(ctx->blocks[4], ptr_y + ctx->dct_y_offset    , ctx->m.linesize);
+            ctx->get_pixels_8x4_sym(ctx->blocks[4], ptr_y + ctx->dct_y_offset,     ctx->m.linesize);
             ctx->get_pixels_8x4_sym(ctx->blocks[5], ptr_y + ctx->dct_y_offset + 8, ctx->m.linesize);
-            ctx->get_pixels_8x4_sym(ctx->blocks[6], ptr_u + ctx->dct_uv_offset   , ctx->m.uvlinesize);
-            ctx->get_pixels_8x4_sym(ctx->blocks[7], ptr_v + ctx->dct_uv_offset   , ctx->m.uvlinesize);
+            ctx->get_pixels_8x4_sym(ctx->blocks[6], ptr_u + ctx->dct_uv_offset,    ctx->m.uvlinesize);
+            ctx->get_pixels_8x4_sym(ctx->blocks[7], ptr_v + ctx->dct_uv_offset,    ctx->m.uvlinesize);
         } else {
-            dsp->clear_block(ctx->blocks[4]); dsp->clear_block(ctx->blocks[5]);
-            dsp->clear_block(ctx->blocks[6]); dsp->clear_block(ctx->blocks[7]);
+            dsp->clear_block(ctx->blocks[4]);
+            dsp->clear_block(ctx->blocks[5]);
+            dsp->clear_block(ctx->blocks[6]);
+            dsp->clear_block(ctx->blocks[7]);
         }
     } else {
-        dsp->get_pixels(ctx->blocks[4], ptr_y + ctx->dct_y_offset    , ctx->m.linesize);
+        dsp->get_pixels(ctx->blocks[4], ptr_y + ctx->dct_y_offset,     ctx->m.linesize);
         dsp->get_pixels(ctx->blocks[5], ptr_y + ctx->dct_y_offset + 8, ctx->m.linesize);
-        dsp->get_pixels(ctx->blocks[6], ptr_u + ctx->dct_uv_offset   , ctx->m.uvlinesize);
-        dsp->get_pixels(ctx->blocks[7], ptr_v + ctx->dct_uv_offset   , ctx->m.uvlinesize);
+        dsp->get_pixels(ctx->blocks[6], ptr_u + ctx->dct_uv_offset,    ctx->m.uvlinesize);
+        dsp->get_pixels(ctx->blocks[7], ptr_v + ctx->dct_uv_offset,    ctx->m.uvlinesize);
     }
 }
 
@@ -496,14 +498,14 @@ static void dnxhd_setup_threads_slices(DNXHDEncContext *ctx)
     for (mb_y = 0; mb_y < ctx->m.mb_height; mb_y++) {
         int thread_size;
         ctx->slice_offs[mb_y] = offset;
-            ctx->slice_size[mb_y] = 0;
-            for (mb_x = 0; mb_x < ctx->m.mb_width; mb_x++) {
-                unsigned mb = mb_y * ctx->m.mb_width + mb_x;
-                ctx->slice_size[mb_y] += ctx->mb_bits[mb];
-            }
-            ctx->slice_size[mb_y] = (ctx->slice_size[mb_y]+31)&~31;
-            ctx->slice_size[mb_y] >>= 3;
-            thread_size = ctx->slice_size[mb_y];
+        ctx->slice_size[mb_y] = 0;
+        for (mb_x = 0; mb_x < ctx->m.mb_width; mb_x++) {
+            unsigned mb = mb_y * ctx->m.mb_width + mb_x;
+            ctx->slice_size[mb_y] += ctx->mb_bits[mb];
+        }
+        ctx->slice_size[mb_y] = (ctx->slice_size[mb_y]+31)&~31;
+        ctx->slice_size[mb_y] >>= 3;
+        thread_size = ctx->slice_size[mb_y];
         offset += thread_size;
     }
 }
diff --git a/libavcodec/dsputil.c b/libavcodec/dsputil.c
index 26e23cd84b..5c5f9db47c 100644
--- a/libavcodec/dsputil.c
+++ b/libavcodec/dsputil.c
@@ -184,7 +184,7 @@ static int pix_norm1_c(uint8_t * pix, int line_size)
             s += sq[pix[6]];
             s += sq[pix[7]];
 #else
-#if LONG_MAX > 2147483647
+#if HAVE_FAST_64BIT
             register uint64_t x=*(uint64_t*)pix;
             s += sq[x&0xff];
             s += sq[(x>>8)&0xff];
@@ -2225,7 +2225,7 @@ static int quant_psnr8x8_c(/*MpegEncContext*/ void *c, uint8_t *src1, uint8_t *s
 
     s->block_last_index[0/*FIXME*/]= s->fast_dct_quantize(s, temp, 0/*FIXME*/, s->qscale, &i);
     s->dct_unquantize_inter(s, temp, 0, s->qscale);
-    ff_simple_idct(temp); //FIXME
+    ff_simple_idct_8(temp); //FIXME
 
     for(i=0; i<64; i++)
         sum+= (temp[i]-bak[i])*(temp[i]-bak[i]);
@@ -2866,6 +2866,12 @@ av_cold void dsputil_init(DSPContext* c, AVCodecContext *avctx)
         c->idct    = j_rev_dct1;
         c->idct_permutation_type= FF_NO_IDCT_PERM;
     }else{
+        if (avctx->bits_per_raw_sample == 10) {
+            c->idct_put              = ff_simple_idct_put_10;
+            c->idct_add              = ff_simple_idct_add_10;
+            c->idct                  = ff_simple_idct_10;
+            c->idct_permutation_type = FF_NO_IDCT_PERM;
+        } else {
         if(avctx->idct_algo==FF_IDCT_INT){
             c->idct_put= ff_jref_idct_put;
             c->idct_add= ff_jref_idct_add;
@@ -2896,11 +2902,12 @@ av_cold void dsputil_init(DSPContext* c, AVCodecContext *avctx)
             c->idct_put = ff_bink_idct_put_c;
             c->idct_permutation_type = FF_NO_IDCT_PERM;
         }else{ //accurate/default
-            c->idct_put= ff_simple_idct_put;
-            c->idct_add= ff_simple_idct_add;
-            c->idct    = ff_simple_idct;
+            c->idct_put = ff_simple_idct_put_8;
+            c->idct_add = ff_simple_idct_add_8;
+            c->idct     = ff_simple_idct_8;
             c->idct_permutation_type= FF_NO_IDCT_PERM;
         }
+        }
     }
 
     c->get_pixels = get_pixels_c;
@@ -3169,21 +3176,18 @@ av_cold void dsputil_init(DSPContext* c, AVCodecContext *avctx)
     dspfunc2(avg_h264_qpel, 1,  8, depth);\
     dspfunc2(avg_h264_qpel, 2,  4, depth);
 
-    if (avctx->codec_id != CODEC_ID_H264 || avctx->bits_per_raw_sample == 8) {
-        BIT_DEPTH_FUNCS(8)
-    } else {
-        switch (avctx->bits_per_raw_sample) {
-            case 9:
-                BIT_DEPTH_FUNCS(9)
-                break;
-            case 10:
-                BIT_DEPTH_FUNCS(10)
-                break;
-            default:
-                av_log(avctx, AV_LOG_DEBUG, "Unsupported bit depth: %d\n", avctx->bits_per_raw_sample);
-                BIT_DEPTH_FUNCS(8)
-                break;
-        }
+    switch (avctx->bits_per_raw_sample) {
+    case 9:
+        BIT_DEPTH_FUNCS(9);
+        break;
+    case 10:
+        BIT_DEPTH_FUNCS(10);
+        break;
+    default:
+        av_log(avctx, AV_LOG_DEBUG, "Unsupported bit depth: %d\n", avctx->bits_per_raw_sample);
+    case 8:
+        BIT_DEPTH_FUNCS(8);
+        break;
     }
 
 
@@ -3243,4 +3247,3 @@ av_cold void dsputil_init(DSPContext* c, AVCodecContext *avctx)
         av_log(avctx, AV_LOG_ERROR, "Internal error, IDCT permutation not set\n");
     }
 }
-
diff --git a/libavcodec/dsputil_template.c b/libavcodec/dsputil_template.c
index 5f7aa4fdc6..17f05e6280 100644
--- a/libavcodec/dsputil_template.c
+++ b/libavcodec/dsputil_template.c
@@ -230,150 +230,6 @@ static void FUNCC(add_pixels4)(uint8_t *restrict p_pixels, DCTELEM *p_block, int
     }
 }
 
-#if 0
-
-#define PIXOP2(OPNAME, OP) \
-static void OPNAME ## _pixels(uint8_t *block, const uint8_t *pixels, int line_size, int h)\
-{\
-    int i;\
-    for(i=0; i<h; i++){\
-        OP(*((uint64_t*)block), AV_RN64(pixels));\
-        pixels+=line_size;\
-        block +=line_size;\
-    }\
-}\
-\
-static void OPNAME ## _no_rnd_pixels_x2_c(uint8_t *block, const uint8_t *pixels, int line_size, int h)\
-{\
-    int i;\
-    for(i=0; i<h; i++){\
-        const uint64_t a= AV_RN64(pixels  );\
-        const uint64_t b= AV_RN64(pixels+1);\
-        OP(*((uint64_t*)block), (a&b) + (((a^b)&0xFEFEFEFEFEFEFEFEULL)>>1));\
-        pixels+=line_size;\
-        block +=line_size;\
-    }\
-}\
-\
-static void OPNAME ## _pixels_x2_c(uint8_t *block, const uint8_t *pixels, int line_size, int h)\
-{\
-    int i;\
-    for(i=0; i<h; i++){\
-        const uint64_t a= AV_RN64(pixels  );\
-        const uint64_t b= AV_RN64(pixels+1);\
-        OP(*((uint64_t*)block), (a|b) - (((a^b)&0xFEFEFEFEFEFEFEFEULL)>>1));\
-        pixels+=line_size;\
-        block +=line_size;\
-    }\
-}\
-\
-static void OPNAME ## _no_rnd_pixels_y2_c(uint8_t *block, const uint8_t *pixels, int line_size, int h)\
-{\
-    int i;\
-    for(i=0; i<h; i++){\
-        const uint64_t a= AV_RN64(pixels          );\
-        const uint64_t b= AV_RN64(pixels+line_size);\
-        OP(*((uint64_t*)block), (a&b) + (((a^b)&0xFEFEFEFEFEFEFEFEULL)>>1));\
-        pixels+=line_size;\
-        block +=line_size;\
-    }\
-}\
-\
-static void OPNAME ## _pixels_y2_c(uint8_t *block, const uint8_t *pixels, int line_size, int h)\
-{\
-    int i;\
-    for(i=0; i<h; i++){\
-        const uint64_t a= AV_RN64(pixels          );\
-        const uint64_t b= AV_RN64(pixels+line_size);\
-        OP(*((uint64_t*)block), (a|b) - (((a^b)&0xFEFEFEFEFEFEFEFEULL)>>1));\
-        pixels+=line_size;\
-        block +=line_size;\
-    }\
-}\
-\
-static void OPNAME ## _pixels_xy2_c(uint8_t *block, const uint8_t *pixels, int line_size, int h)\
-{\
-        int i;\
-        const uint64_t a= AV_RN64(pixels  );\
-        const uint64_t b= AV_RN64(pixels+1);\
-        uint64_t l0=  (a&0x0303030303030303ULL)\
-                    + (b&0x0303030303030303ULL)\
-                    + 0x0202020202020202ULL;\
-        uint64_t h0= ((a&0xFCFCFCFCFCFCFCFCULL)>>2)\
-                   + ((b&0xFCFCFCFCFCFCFCFCULL)>>2);\
-        uint64_t l1,h1;\
-\
-        pixels+=line_size;\
-        for(i=0; i<h; i+=2){\
-            uint64_t a= AV_RN64(pixels  );\
-            uint64_t b= AV_RN64(pixels+1);\
-            l1=  (a&0x0303030303030303ULL)\
-               + (b&0x0303030303030303ULL);\
-            h1= ((a&0xFCFCFCFCFCFCFCFCULL)>>2)\
-              + ((b&0xFCFCFCFCFCFCFCFCULL)>>2);\
-            OP(*((uint64_t*)block), h0+h1+(((l0+l1)>>2)&0x0F0F0F0F0F0F0F0FULL));\
-            pixels+=line_size;\
-            block +=line_size;\
-            a= AV_RN64(pixels  );\
-            b= AV_RN64(pixels+1);\
-            l0=  (a&0x0303030303030303ULL)\
-               + (b&0x0303030303030303ULL)\
-               + 0x0202020202020202ULL;\
-            h0= ((a&0xFCFCFCFCFCFCFCFCULL)>>2)\
-              + ((b&0xFCFCFCFCFCFCFCFCULL)>>2);\
-            OP(*((uint64_t*)block), h0+h1+(((l0+l1)>>2)&0x0F0F0F0F0F0F0F0FULL));\
-            pixels+=line_size;\
-            block +=line_size;\
-        }\
-}\
-\
-static void OPNAME ## _no_rnd_pixels_xy2_c(uint8_t *block, const uint8_t *pixels, int line_size, int h)\
-{\
-        int i;\
-        const uint64_t a= AV_RN64(pixels  );\
-        const uint64_t b= AV_RN64(pixels+1);\
-        uint64_t l0=  (a&0x0303030303030303ULL)\
-                    + (b&0x0303030303030303ULL)\
-                    + 0x0101010101010101ULL;\
-        uint64_t h0= ((a&0xFCFCFCFCFCFCFCFCULL)>>2)\
-                   + ((b&0xFCFCFCFCFCFCFCFCULL)>>2);\
-        uint64_t l1,h1;\
-\
-        pixels+=line_size;\
-        for(i=0; i<h; i+=2){\
-            uint64_t a= AV_RN64(pixels  );\
-            uint64_t b= AV_RN64(pixels+1);\
-            l1=  (a&0x0303030303030303ULL)\
-               + (b&0x0303030303030303ULL);\
-            h1= ((a&0xFCFCFCFCFCFCFCFCULL)>>2)\
-              + ((b&0xFCFCFCFCFCFCFCFCULL)>>2);\
-            OP(*((uint64_t*)block), h0+h1+(((l0+l1)>>2)&0x0F0F0F0F0F0F0F0FULL));\
-            pixels+=line_size;\
-            block +=line_size;\
-            a= AV_RN64(pixels  );\
-            b= AV_RN64(pixels+1);\
-            l0=  (a&0x0303030303030303ULL)\
-               + (b&0x0303030303030303ULL)\
-               + 0x0101010101010101ULL;\
-            h0= ((a&0xFCFCFCFCFCFCFCFCULL)>>2)\
-              + ((b&0xFCFCFCFCFCFCFCFCULL)>>2);\
-            OP(*((uint64_t*)block), h0+h1+(((l0+l1)>>2)&0x0F0F0F0F0F0F0F0FULL));\
-            pixels+=line_size;\
-            block +=line_size;\
-        }\
-}\
-\
-CALL_2X_PIXELS(OPNAME ## _pixels16_c    , OPNAME ## _pixels_c    , 8*sizeof(pixel))\
-CALL_2X_PIXELS(OPNAME ## _pixels16_x2_c , OPNAME ## _pixels_x2_c , 8*sizeof(pixel))\
-CALL_2X_PIXELS(OPNAME ## _pixels16_y2_c , OPNAME ## _pixels_y2_c , 8*sizeof(pixel))\
-CALL_2X_PIXELS(OPNAME ## _pixels16_xy2_c, OPNAME ## _pixels_xy2_c, 8*sizeof(pixel))\
-CALL_2X_PIXELS(OPNAME ## _no_rnd_pixels16_x2_c , OPNAME ## _no_rnd_pixels_x2_c , 8*sizeof(pixel))\
-CALL_2X_PIXELS(OPNAME ## _no_rnd_pixels16_y2_c , OPNAME ## _no_rnd_pixels_y2_c , 8*sizeof(pixel))\
-CALL_2X_PIXELS(OPNAME ## _no_rnd_pixels16_xy2_c, OPNAME ## _no_rnd_pixels_xy2_c, 8*sizeof(pixel))
-
-#define op_avg(a, b) a = ( ((a)|(b)) - ((((a)^(b))&0xFEFEFEFEFEFEFEFEULL)>>1) )
-#else // 64 bit variant
-
 #define PIXOP2(OPNAME, OP) \
 static void FUNCC(OPNAME ## _pixels2)(uint8_t *block, const uint8_t *pixels, int line_size, int h){\
     int i;\
@@ -749,7 +605,6 @@ CALL_2X_PIXELS(FUNCC(OPNAME ## _no_rnd_pixels16_y2) , FUNCC(OPNAME ## _no_rnd_pi
 CALL_2X_PIXELS(FUNCC(OPNAME ## _no_rnd_pixels16_xy2), FUNCC(OPNAME ## _no_rnd_pixels8_xy2), 8*sizeof(pixel))\
 
 #define op_avg(a, b) a = rnd_avg_pixel4(a, b)
-#endif
 #define op_put(a, b) a = b
 
 PIXOP2(avg, op_avg)
diff --git a/libavcodec/g729data.h b/libavcodec/g729data.h
deleted file mode 100644
index 1d64553f20..0000000000
--- a/libavcodec/g729data.h
+++ /dev/null
@@ -1,278 +0,0 @@
-/*
- * data for G.729 decoder
- * Copyright (c) 2007 Vladimir Voroshilov
- *
- * This file is part of FFmpeg.
- *
- * FFmpeg 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,
- * 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
- * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
- */
-
-#ifndef AVCODEC_G729DATA_H
-#define AVCODEC_G729DATA_H
-
-#include <stdint.h>
-
-#define MA_NP                4  ///< Moving Average (MA) prediction order
-
-#define VQ_1ST_BITS          7  ///< first stage vector of quantizer (size in bits)
-#define VQ_2ND_BITS          5  ///< second stage vector of quantizer (size in bits)
-
-#define GC_1ST_IDX_BITS_8K   3  ///< gain codebook (first stage) index, 8k mode (size in bits)
-#define GC_2ND_IDX_BITS_8K   4  ///< gain codebook (second stage) index, 8k mode (size in bits)
-
-#define GC_1ST_IDX_BITS_6K4  3  ///< gain codebook (first stage) index, 6.4k mode (size in bits)
-#define GC_2ND_IDX_BITS_6K4  3  ///< gain codebook (second stage) index, 6.4k mode (size in bits)
-
-/**
- * first stage LSP codebook
- * (10-dimensional, with 128 entries (3.24 of G.729)
- */
-static const int16_t cb_lsp_1st[1<<VQ_1ST_BITS][10] = { /* (2.13) */
-  { 1486,  2168,  3751,  9074, 12134, 13944, 17983, 19173, 21190, 21820},
-  { 1730,  2640,  3450,  4870,  6126,  7876, 15644, 17817, 20294, 21902},
-  { 1568,  2256,  3088,  4874, 11063, 13393, 18307, 19293, 21109, 21741},
-  { 1733,  2512,  3357,  4708,  6977, 10296, 17024, 17956, 19145, 20350},
-  { 1744,  2436,  3308,  8731, 10432, 12007, 15614, 16639, 21359, 21913},
-  { 1786,  2369,  3372,  4521,  6795, 12963, 17674, 18988, 20855, 21640},
-  { 1631,  2433,  3361,  6328, 10709, 12013, 13277, 13904, 19441, 21088},
-  { 1489,  2364,  3291,  6250,  9227, 10403, 13843, 15278, 17721, 21451},
-  { 1869,  2533,  3475,  4365,  9152, 14513, 15908, 17022, 20611, 21411},
-  { 2070,  3025,  4333,  5854,  7805,  9231, 10597, 16047, 20109, 21834},
-  { 1910,  2673,  3419,  4261, 11168, 15111, 16577, 17591, 19310, 20265},
-  { 1141,  1815,  2624,  4623,  6495,  9588, 13968, 16428, 19351, 21286},
-  { 2192,  3171,  4707,  5808, 10904, 12500, 14162, 15664, 21124, 21789},
-  { 1286,  1907,  2548,  3453,  9574, 11964, 15978, 17344, 19691, 22495},
-  { 1921,  2720,  4604,  6684, 11503, 12992, 14350, 15262, 16997, 20791},
-  { 2052,  2759,  3897,  5246,  6638, 10267, 15834, 16814, 18149, 21675},
-  { 1798,  2497,  5617, 11449, 13189, 14711, 17050, 18195, 20307, 21182},
-  { 1009,  1647,  2889,  5709,  9541, 12354, 15231, 18494, 20966, 22033},
-  { 3016,  3794,  5406,  7469, 12488, 13984, 15328, 16334, 19952, 20791},
-  { 2203,  3040,  3796,  5442, 11987, 13512, 14931, 16370, 17856, 18803},
-  { 2912,  4292,  7988,  9572, 11562, 13244, 14556, 16529, 20004, 21073},
-  { 2861,  3607,  5923,  7034,  9234, 12054, 13729, 18056, 20262, 20974},
-  { 3069,  4311,  5967,  7367, 11482, 12699, 14309, 16233, 18333, 19172},
-  { 2434,  3661,  4866,  5798, 10383, 11722, 13049, 15668, 18862, 19831},
-  { 2020,  2605,  3860,  9241, 13275, 14644, 16010, 17099, 19268, 20251},
-  { 1877,  2809,  3590,  4707, 11056, 12441, 15622, 17168, 18761, 19907},
-  { 2107,  2873,  3673,  5799, 13579, 14687, 15938, 17077, 18890, 19831},
-  { 1612,  2284,  2944,  3572,  8219, 13959, 15924, 17239, 18592, 20117},
-  { 2420,  3156,  6542, 10215, 12061, 13534, 15305, 16452, 18717, 19880},
-  { 1667,  2612,  3534,  5237, 10513, 11696, 12940, 16798, 18058, 19378},
-  { 2388,  3017,  4839,  9333, 11413, 12730, 15024, 16248, 17449, 18677},
-  { 1875,  2786,  4231,  6320,  8694, 10149, 11785, 17013, 18608, 19960},
-  {  679,  1411,  4654,  8006, 11446, 13249, 15763, 18127, 20361, 21567},
-  { 1838,  2596,  3578,  4608,  5650, 11274, 14355, 15886, 20579, 21754},
-  { 1303,  1955,  2395,  3322, 12023, 13764, 15883, 18077, 20180, 21232},
-  { 1438,  2102,  2663,  3462,  8328, 10362, 13763, 17248, 19732, 22344},
-  {  860,  1904,  6098,  7775,  9815, 12007, 14821, 16709, 19787, 21132},
-  { 1673,  2723,  3704,  6125,  7668,  9447, 13683, 14443, 20538, 21731},
-  { 1246,  1849,  2902,  4508,  7221, 12710, 14835, 16314, 19335, 22720},
-  { 1525,  2260,  3862,  5659,  7342, 11748, 13370, 14442, 18044, 21334},
-  { 1196,  1846,  3104,  7063, 10972, 12905, 14814, 17037, 19922, 22636},
-  { 2147,  3106,  4475,  6511,  8227,  9765, 10984, 12161, 18971, 21300},
-  { 1585,  2405,  2994,  4036, 11481, 13177, 14519, 15431, 19967, 21275},
-  { 1778,  2688,  3614,  4680,  9465, 11064, 12473, 16320, 19742, 20800},
-  { 1862,  2586,  3492,  6719, 11708, 13012, 14364, 16128, 19610, 20425},
-  { 1395,  2156,  2669,  3386, 10607, 12125, 13614, 16705, 18976, 21367},
-  { 1444,  2117,  3286,  6233,  9423, 12981, 14998, 15853, 17188, 21857},
-  { 2004,  2895,  3783,  4897,  6168,  7297, 12609, 16445, 19297, 21465},
-  { 1495,  2863,  6360,  8100, 11399, 14271, 15902, 17711, 20479, 22061},
-  { 2484,  3114,  5718,  7097,  8400, 12616, 14073, 14847, 20535, 21396},
-  { 2424,  3277,  5296,  6284, 11290, 12903, 16022, 17508, 19333, 20283},
-  { 2565,  3778,  5360,  6989,  8782, 10428, 14390, 15742, 17770, 21734},
-  { 2727,  3384,  6613,  9254, 10542, 12236, 14651, 15687, 20074, 21102},
-  { 1916,  2953,  6274,  8088,  9710, 10925, 12392, 16434, 20010, 21183},
-  { 3384,  4366,  5349,  7667, 11180, 12605, 13921, 15324, 19901, 20754},
-  { 3075,  4283,  5951,  7619,  9604, 11010, 12384, 14006, 20658, 21497},
-  { 1751,  2455,  5147,  9966, 11621, 13176, 14739, 16470, 20788, 21756},
-  { 1442,  2188,  3330,  6813,  8929, 12135, 14476, 15306, 19635, 20544},
-  { 2294,  2895,  4070,  8035, 12233, 13416, 14762, 17367, 18952, 19688},
-  { 1937,  2659,  4602,  6697,  9071, 12863, 14197, 15230, 16047, 18877},
-  { 2071,  2663,  4216,  9445, 10887, 12292, 13949, 14909, 19236, 20341},
-  { 1740,  2491,  3488,  8138,  9656, 11153, 13206, 14688, 20896, 21907},
-  { 2199,  2881,  4675,  8527, 10051, 11408, 14435, 15463, 17190, 20597},
-  { 1943,  2988,  4177,  6039,  7478,  8536, 14181, 15551, 17622, 21579},
-  { 1825,  3175,  7062,  9818, 12824, 15450, 18330, 19856, 21830, 22412},
-  { 2464,  3046,  4822,  5977,  7696, 15398, 16730, 17646, 20588, 21320},
-  { 2550,  3393,  5305,  6920, 10235, 14083, 18143, 19195, 20681, 21336},
-  { 3003,  3799,  5321,  6437,  7919, 11643, 15810, 16846, 18119, 18980},
-  { 3455,  4157,  6838,  8199,  9877, 12314, 15905, 16826, 19949, 20892},
-  { 3052,  3769,  4891,  5810,  6977, 10126, 14788, 15990, 19773, 20904},
-  { 3671,  4356,  5827,  6997,  8460, 12084, 14154, 14939, 19247, 20423},
-  { 2716,  3684,  5246,  6686,  8463, 10001, 12394, 14131, 16150, 19776},
-  { 1945,  2638,  4130,  7995, 14338, 15576, 17057, 18206, 20225, 20997},
-  { 2304,  2928,  4122,  4824,  5640, 13139, 15825, 16938, 20108, 21054},
-  { 1800,  2516,  3350,  5219, 13406, 15948, 17618, 18540, 20531, 21252},
-  { 1436,  2224,  2753,  4546,  9657, 11245, 15177, 16317, 17489, 19135},
-  { 2319,  2899,  4980,  6936,  8404, 13489, 15554, 16281, 20270, 20911},
-  { 2187,  2919,  4610,  5875,  7390, 12556, 14033, 16794, 20998, 21769},
-  { 2235,  2923,  5121,  6259,  8099, 13589, 15340, 16340, 17927, 20159},
-  { 1765,  2638,  3751,  5730,  7883, 10108, 13633, 15419, 16808, 18574},
-  { 3460,  5741,  9596, 11742, 14413, 16080, 18173, 19090, 20845, 21601},
-  { 3735,  4426,  6199,  7363,  9250, 14489, 16035, 17026, 19873, 20876},
-  { 3521,  4778,  6887,  8680, 12717, 14322, 15950, 18050, 20166, 21145},
-  { 2141,  2968,  6865,  8051, 10010, 13159, 14813, 15861, 17528, 18655},
-  { 4148,  6128,  9028, 10871, 12686, 14005, 15976, 17208, 19587, 20595},
-  { 4403,  5367,  6634,  8371, 10163, 11599, 14963, 16331, 17982, 18768},
-  { 4091,  5386,  6852,  8770, 11563, 13290, 15728, 16930, 19056, 20102},
-  { 2746,  3625,  5299,  7504, 10262, 11432, 13172, 15490, 16875, 17514},
-  { 2248,  3556,  8539, 10590, 12665, 14696, 16515, 17824, 20268, 21247},
-  { 1279,  1960,  3920,  7793, 10153, 14753, 16646, 18139, 20679, 21466},
-  { 2440,  3475,  6737,  8654, 12190, 14588, 17119, 17925, 19110, 19979},
-  { 1879,  2514,  4497,  7572, 10017, 14948, 16141, 16897, 18397, 19376},
-  { 2804,  3688,  7490, 10086, 11218, 12711, 16307, 17470, 20077, 21126},
-  { 2023,  2682,  3873,  8268, 10255, 11645, 15187, 17102, 18965, 19788},
-  { 2823,  3605,  5815,  8595, 10085, 11469, 16568, 17462, 18754, 19876},
-  { 2851,  3681,  5280,  7648,  9173, 10338, 14961, 16148, 17559, 18474},
-  { 1348,  2645,  5826,  8785, 10620, 12831, 16255, 18319, 21133, 22586},
-  { 2141,  3036,  4293,  6082,  7593, 10629, 17158, 18033, 21466, 22084},
-  { 1608,  2375,  3384,  6878,  9970, 11227, 16928, 17650, 20185, 21120},
-  { 2774,  3616,  5014,  6557,  7788,  8959, 17068, 18302, 19537, 20542},
-  { 1934,  4813,  6204,  7212,  8979, 11665, 15989, 17811, 20426, 21703},
-  { 2288,  3507,  5037,  6841,  8278,  9638, 15066, 16481, 21653, 22214},
-  { 2951,  3771,  4878,  7578,  9016, 10298, 14490, 15242, 20223, 20990},
-  { 3256,  4791,  6601,  7521,  8644,  9707, 13398, 16078, 19102, 20249},
-  { 1827,  2614,  3486,  6039, 12149, 13823, 16191, 17282, 21423, 22041},
-  { 1000,  1704,  3002,  6335,  8471, 10500, 14878, 16979, 20026, 22427},
-  { 1646,  2286,  3109,  7245, 11493, 12791, 16824, 17667, 18981, 20222},
-  { 1708,  2501,  3315,  6737,  8729,  9924, 16089, 17097, 18374, 19917},
-  { 2623,  3510,  4478,  5645,  9862, 11115, 15219, 18067, 19583, 20382},
-  { 2518,  3434,  4728,  6388,  8082,  9285, 13162, 18383, 19819, 20552},
-  { 1726,  2383,  4090,  6303,  7805, 12845, 14612, 17608, 19269, 20181},
-  { 2860,  3735,  4838,  6044,  7254,  8402, 14031, 16381, 18037, 19410},
-  { 4247,  5993,  7952,  9792, 12342, 14653, 17527, 18774, 20831, 21699},
-  { 3502,  4051,  5680,  6805,  8146, 11945, 16649, 17444, 20390, 21564},
-  { 3151,  4893,  5899,  7198, 11418, 13073, 15124, 17673, 20520, 21861},
-  { 3960,  4848,  5926,  7259,  8811, 10529, 15661, 16560, 18196, 20183},
-  { 4499,  6604,  8036,  9251, 10804, 12627, 15880, 17512, 20020, 21046},
-  { 4251,  5541,  6654,  8318,  9900, 11686, 15100, 17093, 20572, 21687},
-  { 3769,  5327,  7865,  9360, 10684, 11818, 13660, 15366, 18733, 19882},
-  { 3083,  3969,  6248,  8121,  9798, 10994, 12393, 13686, 17888, 19105},
-  { 2731,  4670,  7063,  9201, 11346, 13735, 16875, 18797, 20787, 22360},
-  { 1187,  2227,  4737,  7214,  9622, 12633, 15404, 17968, 20262, 23533},
-  { 1911,  2477,  3915, 10098, 11616, 12955, 16223, 17138, 19270, 20729},
-  { 1764,  2519,  3887,  6944,  9150, 12590, 16258, 16984, 17924, 18435},
-  { 1400,  3674,  7131,  8718, 10688, 12508, 15708, 17711, 19720, 21068},
-  { 2322,  3073,  4287,  8108,  9407, 10628, 15862, 16693, 19714, 21474},
-  { 2630,  3339,  4758,  8360, 10274, 11333, 12880, 17374, 19221, 19936},
-  { 1721,  2577,  5553,  7195,  8651, 10686, 15069, 16953, 18703, 19929}
-};
-
-/**
- * second stage LSP codebook, high and low parts
-   (both 5-dimensional, with 32 entries (3.2.4 of G.729)
- */
-static const int16_t cb_lsp_2nd[1<<VQ_2ND_BITS][10] = { /* (2.13) */
-  { -435,  -815,  -742,  1033,  -518,   582, -1201,   829,    86,   385},
-  { -833,  -891,   463,    -8, -1251,  1450,    72,  -231,   864,   661},
-  {-1021,   231,  -306,   321,  -220,  -163,  -526,  -754, -1633,   267},
-  {   57,  -198,  -339,   -33, -1468,   573,   796,  -169,  -631,   816},
-  {  171,  -350,   294,  1660,   453,   519,   291,   159,  -640, -1296},
-  { -701,  -842,   -58,   950,   892,  1549,   715,   527,  -714,  -193},
-  {  584,    31,  -289,   356,  -333,  -457,   612,  -283, -1381,  -741},
-  { -109,  -808,   231,    77,   -87,  -344,  1341,  1087,  -654,  -569},
-  { -859,  1236,   550,   854,   714,  -543, -1752,  -195,   -98,  -276},
-  { -877,  -954, -1248,  -299,   212,  -235,  -728,   949,  1517,   895},
-  {  -77,   344,  -620,   763,   413,   502,  -362,  -960,  -483,  1386},
-  { -314,  -307,  -256, -1260,  -429,   450,  -466,  -108,  1010,  2223},
-  {  711,   693,   521,   650,  1305,   -28,  -378,   744, -1005,   240},
-  { -112,  -271,  -500,   946,  1733,   271,   -15,   909,  -259,  1688},
-  {  575,   -10,  -468,  -199,  1101, -1011,   581,   -53,  -747,   878},
-  {  145,  -285, -1280,  -398,    36,  -498, -1377,    18,  -444,  1483},
-  {-1133,  -835,  1350,  1284,   -95,  1015,  -222,   443,   372,  -354},
-  {-1459, -1237,   416,  -213,   466,   669,   659,  1640,   932,   534},
-  {  -15,    66,   468,  1019,  -748,  1385,  -182,  -907,  -721,  -262},
-  { -338,   148,  1445,    75,  -760,   569,  1247,   337,   416,  -121},
-  {  389,   239,  1568,   981,   113,   369, -1003,  -507,  -587,  -904},
-  { -312,   -98,   949,    31,  1104,    72,  -141,  1465,    63,  -785},
-  { 1127,   584,   835,   277, -1159,   208,   301,  -882,   117,  -404},
-  {  539,  -114,   856,  -493,   223,  -912,   623,   -76,   276,  -440},
-  { 2197,  2337,  1268,   670,   304,  -267,  -525,   140,   882,  -139},
-  {-1596,   550,   801,  -456,   -56,  -697,   865,  1060,   413,   446},
-  { 1154,   593,   -77,  1237,   -31,   581, -1037,  -895,   669,   297},
-  {  397,   558,   203,  -797,  -919,     3,   692,  -292,  1050,   782},
-  {  334,  1475,   632,   -80,    48, -1061,  -484,   362,  -597,  -852},
-  { -545,  -330,  -429,  -680,  1133, -1182,  -744,  1340,   262,    63},
-  { 1320,   827,  -398,  -576,   341,  -774,  -483, -1247,   -70,    98},
-  { -163,   674,   -11,  -886,   531, -1125,  -265,  -242,   724,   934}
-};
-
-/**
- * gain codebook (first stage), 8k mode (3.9.2 of G.729)
- */
-static const int16_t cb_gain_1st_8k[1<<GC_1ST_IDX_BITS_8K][2] = { /*(0.14) (2.13) */
-  { 3242 ,  9949 },
-  { 1551 ,  2425 },
-  { 2678 , 27162 },
-  { 1921 ,  9291 },
-  { 1831 ,  5022 },
-  {    1 ,  1516 },
-  {  356 , 14756 },
-  {   57 ,  5404 },
-};
-
-/**
- * gain codebook (second stage), 8k mode (3.9.2 of G.729)
- */
-static const int16_t cb_gain_2nd_8k[1<<GC_2ND_IDX_BITS_8K][2] = { /*(1.14) (1.13) */
-  {  5142 ,   592 },
-  { 17299 ,  1861 },
-  {  6160 ,  2395 },
-  { 16112 ,  3392 },
-  {   826 ,  2005 },
-  { 18973 ,  5935 },
-  {  1994 ,     0 },
-  { 15434 ,   237 },
-  { 10573 ,  2966 },
-  { 15132 ,  4914 },
-  { 11569 ,  1196 },
-  { 14194 ,  1630 },
-  {  8091 ,  4861 },
-  { 15161 , 14276 },
-  {  9120 ,   525 },
-  { 13260 ,  3256 },
-};
-
-/**
- * 4th order Moving Average (MA) Predictor codebook (3.2.4 of G.729)
- */
-static const int16_t cb_ma_predictor[2][MA_NP][10] = { /* (0.15) */
-  {
-    { 8421,  9109,  9175,  8965,  9034,  9057,  8765,  8775,  9106,  8673},
-    { 7018,  7189,  7638,  7307,  7444,  7379,  7038,  6956,  6930,  6868},
-    { 5472,  4990,  5134,  5177,  5246,  5141,  5206,  5095,  4830,  5147},
-    { 4056,  3031,  2614,  3024,  2916,  2713,  3309,  3237,  2857,  3473}
-  },
-  {
-    { 7733,  7880,  8188,  8175,  8247,  8490,  8637,  8601,  8359,  7569},
-    { 4210,  3031,  2552,  3473,  3876,  3853,  4184,  4154,  3909,  3968},
-    { 3214,  1930,  1313,  2143,  2493,  2385,  2755,  2706,  2542,  2919},
-    { 3024,  1592,   940,  1631,  1723,  1579,  2034,  2084,  1913,  2601}
-  }
-};
-
-static const int16_t cb_ma_predictor_sum[2][10] = { /* (0.15) */
-  { 7798,  8447,  8205,  8293,  8126,  8477,  8447,  8703,  9043,  8604},
-  {14585, 18333, 19772, 17344, 16426, 16459, 15155, 15220, 16043, 15708}
-};
-
-/**
- * initial LSP coefficients belongs to virtual frame preceding  the
- * first frame of the stream
- */
-static const int16_t lsp_init[10]= { /* (0.15) */
-   30000, 26000, 21000, 15000, 8000, 0, -8000,-15000,-21000,-26000
-};
-#endif /* AVCODEC_G729DATA_H */
diff --git a/libavcodec/g729dec.c b/libavcodec/g729dec.c
deleted file mode 100644
index c4a883f392..0000000000
--- a/libavcodec/g729dec.c
+++ /dev/null
@@ -1,331 +0,0 @@
-/*
- * G.729 decoder
- * Copyright (c) 2008 Vladimir Voroshilov
- *
- * This file is part of FFmpeg.
- *
- * FFmpeg 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,
- * 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
- * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
- */
-#include <stdlib.h>
-#include <inttypes.h>
-#include <limits.h>
-#include <stdio.h>
-#include <string.h>
-#include <math.h>
-#include <assert.h>
-
-#include "avcodec.h"
-#include "libavutil/avutil.h"
-#include "get_bits.h"
-
-#include "lsp.h"
-#include "celp_math.h"
-#include "acelp_filters.h"
-#include "acelp_pitch_delay.h"
-#include "acelp_vectors.h"
-#include "g729data.h"
-
-/**
- * minimum quantized LSF value (3.2.4)
- * 0.005 in Q13
- */
-#define LSFQ_MIN                   40
-
-/**
- * maximum quantized LSF value (3.2.4)
- * 3.135 in Q13
- */
-#define LSFQ_MAX                   25681
-
-/**
- * minimum LSF distance (3.2.4)
- * 0.0391 in Q13
- */
-#define LSFQ_DIFF_MIN              321
-
-/**
- * minimum gain pitch value (3.8, Equation 47)
- * 0.2 in (1.14)
- */
-#define SHARP_MIN                  3277
-
-/**
- * maximum gain pitch value (3.8, Equation 47)
- * (EE) This does not comply with the specification.
- * Specification says about 0.8, which should be
- * 13107 in (1.14), but reference C code uses
- * 13017 (equals to 0.7945) instead of it.
- */
-#define SHARP_MAX                  13017
-
-/**
- * subframe size
- */
-#define SUBFRAME_SIZE              40
-
-
-typedef struct {
-    uint8_t ac_index_bits[2];   ///< adaptive codebook index for second subframe (size in bits)
-    uint8_t parity_bit;         ///< parity bit for pitch delay
-    uint8_t gc_1st_index_bits;  ///< gain codebook (first stage) index (size in bits)
-    uint8_t gc_2nd_index_bits;  ///< gain codebook (second stage) index (size in bits)
-    uint8_t fc_signs_bits;      ///< number of pulses in fixed-codebook vector
-    uint8_t fc_indexes_bits;    ///< size (in bits) of fixed-codebook index entry
-} G729FormatDescription;
-
-typedef struct {
-    int pitch_delay_int_prev;   ///< integer part of previous subframe's pitch delay (4.1.3)
-
-    /// (2.13) LSP quantizer outputs
-    int16_t  past_quantizer_output_buf[MA_NP + 1][10];
-    int16_t* past_quantizer_outputs[MA_NP + 1];
-
-    int16_t lsfq[10];           ///< (2.13) quantized LSF coefficients from previous frame
-    int16_t lsp_buf[2][10];     ///< (0.15) LSP coefficients (previous and current frames) (3.2.5)
-    int16_t *lsp[2];            ///< pointers to lsp_buf
-}  G729Context;
-
-static const G729FormatDescription format_g729_8k = {
-    .ac_index_bits     = {8,5},
-    .parity_bit        = 1,
-    .gc_1st_index_bits = GC_1ST_IDX_BITS_8K,
-    .gc_2nd_index_bits = GC_2ND_IDX_BITS_8K,
-    .fc_signs_bits     = 4,
-    .fc_indexes_bits   = 13,
-};
-
-static const G729FormatDescription format_g729d_6k4 = {
-    .ac_index_bits     = {8,4},
-    .parity_bit        = 0,
-    .gc_1st_index_bits = GC_1ST_IDX_BITS_6K4,
-    .gc_2nd_index_bits = GC_2ND_IDX_BITS_6K4,
-    .fc_signs_bits     = 2,
-    .fc_indexes_bits   = 9,
-};
-
-/**
- * @brief pseudo random number generator
- */
-static inline uint16_t g729_prng(uint16_t value)
-{
-    return 31821 * value + 13849;
-}
-
-/**
- * Get parity bit of bit 2..7
- */
-static inline int get_parity(uint8_t value)
-{
-   return (0x6996966996696996ULL >> (value >> 2)) & 1;
-}
-
-static void lsf_decode(int16_t* lsfq, int16_t* past_quantizer_outputs[MA_NP + 1],
-                       int16_t ma_predictor,
-                       int16_t vq_1st, int16_t vq_2nd_low, int16_t vq_2nd_high)
-{
-    int i,j;
-    static const uint8_t min_distance[2]={10, 5}; //(2.13)
-    int16_t* quantizer_output = past_quantizer_outputs[MA_NP];
-
-    for (i = 0; i < 5; i++) {
-        quantizer_output[i]     = cb_lsp_1st[vq_1st][i    ] + cb_lsp_2nd[vq_2nd_low ][i    ];
-        quantizer_output[i + 5] = cb_lsp_1st[vq_1st][i + 5] + cb_lsp_2nd[vq_2nd_high][i + 5];
-    }
-
-    for (j = 0; j < 2; j++) {
-        for (i = 1; i < 10; i++) {
-            int diff = (quantizer_output[i - 1] - quantizer_output[i] + min_distance[j]) >> 1;
-            if (diff > 0) {
-                quantizer_output[i - 1] -= diff;
-                quantizer_output[i    ] += diff;
-            }
-        }
-    }
-
-    for (i = 0; i < 10; i++) {
-        int sum = quantizer_output[i] * cb_ma_predictor_sum[ma_predictor][i];
-        for (j = 0; j < MA_NP; j++)
-            sum += past_quantizer_outputs[j][i] * cb_ma_predictor[ma_predictor][j][i];
-
-        lsfq[i] = sum >> 15;
-    }
-
-    /* Rotate past_quantizer_outputs. */
-    memmove(past_quantizer_outputs + 1, past_quantizer_outputs, MA_NP * sizeof(int16_t*));
-    past_quantizer_outputs[0] = quantizer_output;
-
-    ff_acelp_reorder_lsf(lsfq, LSFQ_DIFF_MIN, LSFQ_MIN, LSFQ_MAX, 10);
-}
-
-static av_cold int decoder_init(AVCodecContext * avctx)
-{
-    G729Context* ctx = avctx->priv_data;
-    int i,k;
-
-    if (avctx->channels != 1) {
-        av_log(avctx, AV_LOG_ERROR, "Only mono sound is supported (requested channels: %d).\n", avctx->channels);
-        return AVERROR(EINVAL);
-    }
-
-    /* Both 8kbit/s and 6.4kbit/s modes uses two subframes per frame. */
-    avctx->frame_size = SUBFRAME_SIZE << 1;
-
-    for (k = 0; k < MA_NP + 1; k++) {
-        ctx->past_quantizer_outputs[k] = ctx->past_quantizer_output_buf[k];
-        for (i = 1; i < 11; i++)
-            ctx->past_quantizer_outputs[k][i - 1] = (18717 * i) >> 3;
-    }
-
-    ctx->lsp[0] = ctx->lsp_buf[0];
-    ctx->lsp[1] = ctx->lsp_buf[1];
-    memcpy(ctx->lsp[0], lsp_init, 10 * sizeof(int16_t));
-
-    return 0;
-}
-
-static int decode_frame(AVCodecContext *avctx, void *data, int *data_size,
-                        AVPacket *avpkt)
-{
-    const uint8_t *buf = avpkt->data;
-    int buf_size       = avpkt->size;
-    int16_t *out_frame = data;
-    GetBitContext gb;
-    G729FormatDescription format;
-    int frame_erasure = 0;    ///< frame erasure detected during decoding
-    int bad_pitch = 0;        ///< parity check failed
-    int i;
-    G729Context *ctx = avctx->priv_data;
-    int16_t lp[2][11];           // (3.12)
-    uint8_t ma_predictor;     ///< switched MA predictor of LSP quantizer
-    uint8_t quantizer_1st;    ///< first stage vector of quantizer
-    uint8_t quantizer_2nd_lo; ///< second stage lower vector of quantizer (size in bits)
-    uint8_t quantizer_2nd_hi; ///< second stage higher vector of quantizer (size in bits)
-
-    int pitch_delay_int;         // pitch delay, integer part
-    int pitch_delay_3x;          // pitch delay, multiplied by 3
-
-    if (*data_size < SUBFRAME_SIZE << 2) {
-        av_log(avctx, AV_LOG_ERROR, "Error processing packet: output buffer too small\n");
-        return AVERROR(EIO);
-    }
-
-    if (buf_size == 10) {
-        format = format_g729_8k;
-        av_log(avctx, AV_LOG_DEBUG, "Packet type: %s\n", "G.729 @ 8kbit/s");
-    } else if (buf_size == 8) {
-        format = format_g729d_6k4;
-        av_log(avctx, AV_LOG_DEBUG, "Packet type: %s\n", "G.729D @ 6.4kbit/s");
-    } else {
-        av_log(avctx, AV_LOG_ERROR, "Packet size %d is unknown.\n", buf_size);
-        return AVERROR_INVALIDDATA;
-    }
-
-    for (i=0; i < buf_size; i++)
-        frame_erasure |= buf[i];
-    frame_erasure = !frame_erasure;
-
-    init_get_bits(&gb, buf, buf_size);
-
-    ma_predictor     = get_bits(&gb, 1);
-    quantizer_1st    = get_bits(&gb, VQ_1ST_BITS);
-    quantizer_2nd_lo = get_bits(&gb, VQ_2ND_BITS);
-    quantizer_2nd_hi = get_bits(&gb, VQ_2ND_BITS);
-
-    lsf_decode(ctx->lsfq, ctx->past_quantizer_outputs,
-               ma_predictor,
-               quantizer_1st, quantizer_2nd_lo, quantizer_2nd_hi);
-
-    ff_acelp_lsf2lsp(ctx->lsp[1], ctx->lsfq, 10);
-
-    ff_acelp_lp_decode(&lp[0][0], &lp[1][0], ctx->lsp[1], ctx->lsp[0], 10);
-
-    FFSWAP(int16_t*, ctx->lsp[1], ctx->lsp[0]);
-
-    for (i = 0; i < 2; i++) {
-        uint8_t ac_index;      ///< adaptive codebook index
-        uint8_t pulses_signs;  ///< fixed-codebook vector pulse signs
-        int fc_indexes;        ///< fixed-codebook indexes
-        uint8_t gc_1st_index;  ///< gain codebook (first stage) index
-        uint8_t gc_2nd_index;  ///< gain codebook (second stage) index
-
-        ac_index      = get_bits(&gb, format.ac_index_bits[i]);
-        if(!i && format.parity_bit)
-            bad_pitch = get_parity(ac_index) == get_bits1(&gb);
-        fc_indexes    = get_bits(&gb, format.fc_indexes_bits);
-        pulses_signs  = get_bits(&gb, format.fc_signs_bits);
-        gc_1st_index  = get_bits(&gb, format.gc_1st_index_bits);
-        gc_2nd_index  = get_bits(&gb, format.gc_2nd_index_bits);
-
-        if(!i) {
-            if (bad_pitch)
-                pitch_delay_3x   = 3 * ctx->pitch_delay_int_prev;
-            else
-                pitch_delay_3x = ff_acelp_decode_8bit_to_1st_delay3(ac_index);
-        } else {
-            int pitch_delay_min = av_clip(ctx->pitch_delay_int_prev - 5,
-                                          PITCH_DELAY_MIN, PITCH_DELAY_MAX - 9);
-
-            if(packet_type == FORMAT_G729D_6K4)
-                pitch_delay_3x = ff_acelp_decode_4bit_to_2nd_delay3(ac_index, pitch_delay_min);
-            else
-                pitch_delay_3x = ff_acelp_decode_5_6_bit_to_2nd_delay3(ac_index, pitch_delay_min);
-        }
-
-        /* Round pitch delay to nearest (used everywhere except ff_acelp_interpolate). */
-        pitch_delay_int  = (pitch_delay_3x + 1) / 3;
-
-        ff_acelp_weighted_vector_sum(fc + pitch_delay_int,
-                                     fc + pitch_delay_int,
-                                     fc, 1 << 14,
-                                     av_clip(ctx->gain_pitch, SHARP_MIN, SHARP_MAX),
-                                     0, 14,
-                                     SUBFRAME_SIZE - pitch_delay_int);
-
-        if (frame_erasure) {
-            ctx->gain_pitch = (29491 * ctx->gain_pitch) >> 15; // 0.90 (0.15)
-            ctx->gain_code  = ( 2007 * ctx->gain_code ) >> 11; // 0.98 (0.11)
-
-            gain_corr_factor = 0;
-        } else {
-            ctx->gain_pitch  = cb_gain_1st_8k[gc_1st_index][0] +
-                               cb_gain_2nd_8k[gc_2nd_index][0];
-            gain_corr_factor = cb_gain_1st_8k[gc_1st_index][1] +
-                               cb_gain_2nd_8k[gc_2nd_index][1];
-
-        ff_acelp_weighted_vector_sum(ctx->exc + i * SUBFRAME_SIZE,
-                                     ctx->exc + i * SUBFRAME_SIZE, fc,
-                                     (!voicing && frame_erasure) ? 0 : ctx->gain_pitch,
-                                     ( voicing && frame_erasure) ? 0 : ctx->gain_code,
-                                     1 << 13, 14, SUBFRAME_SIZE);
-
-            ctx->pitch_delay_int_prev = pitch_delay_int;
-    }
-
-    *data_size = SUBFRAME_SIZE << 2;
-    return buf_size;
-}
-
-AVCodec ff_g729_decoder =
-{
-    "g729",
-    AVMEDIA_TYPE_AUDIO,
-    CODEC_ID_G729,
-    sizeof(G729Context),
-    decoder_init,
-    NULL,
-    NULL,
-    decode_frame,
-    .long_name = NULL_IF_CONFIG_SMALL("G.729"),
-};
diff --git a/libavcodec/lsp.c b/libavcodec/lsp.c
index 374539a029..42a32d7971 100644
--- a/libavcodec/lsp.c
+++ b/libavcodec/lsp.c
@@ -150,11 +150,7 @@ void ff_acelp_lp_decode(int16_t* lp_1st, int16_t* lp_2nd, const int16_t* lsp_2nd
 
     /* LSP values for first subframe (3.2.5 of G.729, Equation 24)*/
     for(i=0; i<lp_order; i++)
-#ifdef G729_BITEXACT
-        lsp_1st[i] = (lsp_2nd[i] >> 1) + (lsp_prev[i] >> 1);
-#else
         lsp_1st[i] = (lsp_2nd[i] + lsp_prev[i]) >> 1;
-#endif
 
     ff_acelp_lsp2lpc(lp_1st, lsp_1st, lp_order >> 1);
 
diff --git a/libavcodec/mpegvideo.c b/libavcodec/mpegvideo.c
index f44824117d..6ebaef8341 100644
--- a/libavcodec/mpegvideo.c
+++ b/libavcodec/mpegvideo.c
@@ -318,7 +318,7 @@ int ff_alloc_picture(MpegEncContext *s, Picture *pic, int shared){
     s->prev_pict_types[0]= s->dropable ? AV_PICTURE_TYPE_B : s->pict_type;
     if (pic->f.age < PREV_PICT_TYPES_BUFFER_SIZE && s->prev_pict_types[pic->f.age] == AV_PICTURE_TYPE_B)
         pic->f.age = INT_MAX; // Skipped MBs in B-frames are quite rare in MPEG-1/2 and it is a bit tricky to skip them anyway.
-    pic->owner2 = NULL;
+    pic->owner2 = s;
 
     return 0;
 fail: //for the FF_ALLOCZ_OR_GOTO macro
@@ -1036,14 +1036,16 @@ int MPV_frame_start(MpegEncContext *s, AVCodecContext *avctx)
     /* mark&release old frames */
     if (s->pict_type != AV_PICTURE_TYPE_B && s->last_picture_ptr && s->last_picture_ptr != s->next_picture_ptr && s->last_picture_ptr->f.data[0]) {
       if(s->out_format != FMT_H264 || s->codec_id == CODEC_ID_SVQ3){
-          free_frame_buffer(s, s->last_picture_ptr);
+          if (s->last_picture_ptr->owner2 == s)
+              free_frame_buffer(s, s->last_picture_ptr);
 
         /* release forgotten pictures */
         /* if(mpeg124/h263) */
         if(!s->encoding){
             for(i=0; i<s->picture_count; i++){
-                if (s->picture[i].f.data[0] && &s->picture[i] != s->next_picture_ptr && s->picture[i].f.reference) {
-                    av_log(avctx, AV_LOG_ERROR, "releasing zombie picture\n");
+                if (s->picture[i].owner2 == s && s->picture[i].f.data[0] && &s->picture[i] != s->next_picture_ptr && s->picture[i].f.reference) {
+                    if (!(avctx->active_thread_type & FF_THREAD_FRAME))
+                        av_log(avctx, AV_LOG_ERROR, "releasing zombie picture\n");
                     free_frame_buffer(s, &s->picture[i]);
                 }
             }
diff --git a/libavcodec/ppc/dsputil_ppc.c b/libavcodec/ppc/dsputil_ppc.c
index c683725f41..8229fb5bf7 100644
--- a/libavcodec/ppc/dsputil_ppc.c
+++ b/libavcodec/ppc/dsputil_ppc.c
@@ -178,7 +178,7 @@ void dsputil_init_ppc(DSPContext* c, AVCodecContext *avctx)
         }
 #endif //CONFIG_ENCODERS
 
-        if (avctx->lowres==0) {
+        if (avctx->lowres == 0 && avctx->bits_per_raw_sample <= 8) {
             if ((avctx->idct_algo == FF_IDCT_AUTO) ||
                 (avctx->idct_algo == FF_IDCT_ALTIVEC)) {
                 c->idct_put = idct_put_altivec;
diff --git a/libavcodec/ps2/dsputil_mmi.c b/libavcodec/ps2/dsputil_mmi.c
index 349583f1ba..707d1c9f5e 100644
--- a/libavcodec/ps2/dsputil_mmi.c
+++ b/libavcodec/ps2/dsputil_mmi.c
@@ -156,7 +156,8 @@ void dsputil_init_mmi(DSPContext* c, AVCodecContext *avctx)
 
     c->get_pixels = get_pixels_mmi;
 
-    if(idct_algo==FF_IDCT_AUTO || idct_algo==FF_IDCT_PS2){
+    if (avctx->bits_per_raw_sample <= 8 &&
+        (idct_algo==FF_IDCT_AUTO || idct_algo==FF_IDCT_PS2)) {
         c->idct_put= ff_mmi_idct_put;
         c->idct_add= ff_mmi_idct_add;
         c->idct    = ff_mmi_idct;
diff --git a/libavcodec/sh4/dsputil_sh4.c b/libavcodec/sh4/dsputil_sh4.c
index d254e1db6b..88502677d2 100644
--- a/libavcodec/sh4/dsputil_sh4.c
+++ b/libavcodec/sh4/dsputil_sh4.c
@@ -97,7 +97,8 @@ void dsputil_init_sh4(DSPContext* c, AVCodecContext *avctx)
 
         if (!high_bit_depth)
         c->clear_blocks = clear_blocks_sh4;
-        if(idct_algo==FF_IDCT_AUTO || idct_algo==FF_IDCT_SH4){
+        if (avctx->bits_per_raw_sample <= 8 &&
+            (idct_algo==FF_IDCT_AUTO || idct_algo==FF_IDCT_SH4)) {
                 c->idct_put = idct_put;
                 c->idct_add = idct_add;
                c->idct     = idct_sh4;
diff --git a/libavcodec/simple_idct.c b/libavcodec/simple_idct.c
index 06f3624ba3..ffe5a0b070 100644
--- a/libavcodec/simple_idct.c
+++ b/libavcodec/simple_idct.c
@@ -25,339 +25,19 @@
  * simpleidct in C.
  */
 
-/*
-  based upon some outcommented c code from mpeg2dec (idct_mmx.c
-  written by Aaron Holtzman <aholtzma@ess.engr.uvic.ca>)
- */
-
 #include "libavutil/intreadwrite.h"
 #include "avcodec.h"
 #include "dsputil.h"
 #include "mathops.h"
 #include "simple_idct.h"
 
-#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
-#define W3  19266  //cos(i*M_PI/16)*sqrt(2)*(1<<14) + 0.5
-#define W4  16383  //cos(i*M_PI/16)*sqrt(2)*(1<<14) + 0.5
-#define W5  12873  //cos(i*M_PI/16)*sqrt(2)*(1<<14) + 0.5
-#define W6  8867   //cos(i*M_PI/16)*sqrt(2)*(1<<14) + 0.5
-#define W7  4520   //cos(i*M_PI/16)*sqrt(2)*(1<<14) + 0.5
-#define ROW_SHIFT 11
-#define COL_SHIFT 20 // 6
+#define BIT_DEPTH 8
+#include "simple_idct_template.c"
+#undef BIT_DEPTH
 
-static inline void idctRowCondDC (DCTELEM * row)
-{
-        int a0, a1, a2, a3, b0, b1, b2, b3;
-
-#if HAVE_FAST_64BIT
-#define ROW0_MASK (0xffffLL << 48 * HAVE_BIGENDIAN)
-        if (((((uint64_t *)row)[0] & ~ROW0_MASK) | ((uint64_t *)row)[1]) == 0) {
-            uint64_t temp = (row[0] << 3) & 0xffff;
-            temp += temp << 16;
-            temp += temp << 32;
-            ((uint64_t *)row)[0] = temp;
-            ((uint64_t *)row)[1] = temp;
-            return;
-        }
-#else
-        if (!(((uint32_t*)row)[1] |
-              ((uint32_t*)row)[2] |
-              ((uint32_t*)row)[3] |
-              row[1])) {
-            uint32_t temp = (row[0] << 3) & 0xffff;
-            temp += temp << 16;
-            ((uint32_t*)row)[0]=((uint32_t*)row)[1] =
-                ((uint32_t*)row)[2]=((uint32_t*)row)[3] = temp;
-            return;
-        }
-#endif
-
-        a0 = (W4 * row[0]) + (1 << (ROW_SHIFT - 1));
-        a1 = a0;
-        a2 = a0;
-        a3 = a0;
-
-        /* no need to optimize : gcc does it */
-        a0 += W2 * row[2];
-        a1 += W6 * row[2];
-        a2 -= W6 * row[2];
-        a3 -= W2 * row[2];
-
-        b0 = MUL16(W1, row[1]);
-        MAC16(b0, W3, row[3]);
-        b1 = MUL16(W3, row[1]);
-        MAC16(b1, -W7, row[3]);
-        b2 = MUL16(W5, row[1]);
-        MAC16(b2, -W1, row[3]);
-        b3 = MUL16(W7, row[1]);
-        MAC16(b3, -W5, row[3]);
-
-        if (AV_RN64A(row + 4)) {
-            a0 += W4*row[4] + W6*row[6];
-            a1 += - W4*row[4] - W2*row[6];
-            a2 += - W4*row[4] + W2*row[6];
-            a3 += W4*row[4] - W6*row[6];
-
-            MAC16(b0, W5, row[5]);
-            MAC16(b0, W7, row[7]);
-
-            MAC16(b1, -W1, row[5]);
-            MAC16(b1, -W5, row[7]);
-
-            MAC16(b2, W7, row[5]);
-            MAC16(b2, W3, row[7]);
-
-            MAC16(b3, W3, row[5]);
-            MAC16(b3, -W1, row[7]);
-        }
-
-        row[0] = (a0 + b0) >> ROW_SHIFT;
-        row[7] = (a0 - b0) >> ROW_SHIFT;
-        row[1] = (a1 + b1) >> ROW_SHIFT;
-        row[6] = (a1 - b1) >> ROW_SHIFT;
-        row[2] = (a2 + b2) >> ROW_SHIFT;
-        row[5] = (a2 - b2) >> ROW_SHIFT;
-        row[3] = (a3 + b3) >> ROW_SHIFT;
-        row[4] = (a3 - b3) >> ROW_SHIFT;
-}
-
-static inline void idctSparseColPut (uint8_t *dest, int line_size,
-                                     DCTELEM * col)
-{
-        int a0, a1, a2, a3, b0, b1, b2, b3;
-        uint8_t *cm = ff_cropTbl + MAX_NEG_CROP;
-
-        /* XXX: I did that only to give same values as previous code */
-        a0 = W4 * (col[8*0] + ((1<<(COL_SHIFT-1))/W4));
-        a1 = a0;
-        a2 = a0;
-        a3 = a0;
-
-        a0 +=  + W2*col[8*2];
-        a1 +=  + W6*col[8*2];
-        a2 +=  - W6*col[8*2];
-        a3 +=  - W2*col[8*2];
-
-        b0 = MUL16(W1, col[8*1]);
-        b1 = MUL16(W3, col[8*1]);
-        b2 = MUL16(W5, col[8*1]);
-        b3 = MUL16(W7, col[8*1]);
-
-        MAC16(b0, + W3, col[8*3]);
-        MAC16(b1, - W7, col[8*3]);
-        MAC16(b2, - W1, col[8*3]);
-        MAC16(b3, - W5, col[8*3]);
-
-        if(col[8*4]){
-            a0 += + W4*col[8*4];
-            a1 += - W4*col[8*4];
-            a2 += - W4*col[8*4];
-            a3 += + W4*col[8*4];
-        }
-
-        if (col[8*5]) {
-            MAC16(b0, + W5, col[8*5]);
-            MAC16(b1, - W1, col[8*5]);
-            MAC16(b2, + W7, col[8*5]);
-            MAC16(b3, + W3, col[8*5]);
-        }
-
-        if(col[8*6]){
-            a0 += + W6*col[8*6];
-            a1 += - W2*col[8*6];
-            a2 += + W2*col[8*6];
-            a3 += - W6*col[8*6];
-        }
-
-        if (col[8*7]) {
-            MAC16(b0, + W7, col[8*7]);
-            MAC16(b1, - W5, col[8*7]);
-            MAC16(b2, + W3, col[8*7]);
-            MAC16(b3, - W1, col[8*7]);
-        }
-
-        dest[0] = cm[(a0 + b0) >> COL_SHIFT];
-        dest += line_size;
-        dest[0] = cm[(a1 + b1) >> COL_SHIFT];
-        dest += line_size;
-        dest[0] = cm[(a2 + b2) >> COL_SHIFT];
-        dest += line_size;
-        dest[0] = cm[(a3 + b3) >> COL_SHIFT];
-        dest += line_size;
-        dest[0] = cm[(a3 - b3) >> COL_SHIFT];
-        dest += line_size;
-        dest[0] = cm[(a2 - b2) >> COL_SHIFT];
-        dest += line_size;
-        dest[0] = cm[(a1 - b1) >> COL_SHIFT];
-        dest += line_size;
-        dest[0] = cm[(a0 - b0) >> COL_SHIFT];
-}
-
-static inline void idctSparseColAdd (uint8_t *dest, int line_size,
-                                     DCTELEM * col)
-{
-        int a0, a1, a2, a3, b0, b1, b2, b3;
-        uint8_t *cm = ff_cropTbl + MAX_NEG_CROP;
-
-        /* XXX: I did that only to give same values as previous code */
-        a0 = W4 * (col[8*0] + ((1<<(COL_SHIFT-1))/W4));
-        a1 = a0;
-        a2 = a0;
-        a3 = a0;
-
-        a0 +=  + W2*col[8*2];
-        a1 +=  + W6*col[8*2];
-        a2 +=  - W6*col[8*2];
-        a3 +=  - W2*col[8*2];
-
-        b0 = MUL16(W1, col[8*1]);
-        b1 = MUL16(W3, col[8*1]);
-        b2 = MUL16(W5, col[8*1]);
-        b3 = MUL16(W7, col[8*1]);
-
-        MAC16(b0, + W3, col[8*3]);
-        MAC16(b1, - W7, col[8*3]);
-        MAC16(b2, - W1, col[8*3]);
-        MAC16(b3, - W5, col[8*3]);
-
-        if(col[8*4]){
-            a0 += + W4*col[8*4];
-            a1 += - W4*col[8*4];
-            a2 += - W4*col[8*4];
-            a3 += + W4*col[8*4];
-        }
-
-        if (col[8*5]) {
-            MAC16(b0, + W5, col[8*5]);
-            MAC16(b1, - W1, col[8*5]);
-            MAC16(b2, + W7, col[8*5]);
-            MAC16(b3, + W3, col[8*5]);
-        }
-
-        if(col[8*6]){
-            a0 += + W6*col[8*6];
-            a1 += - W2*col[8*6];
-            a2 += + W2*col[8*6];
-            a3 += - W6*col[8*6];
-        }
-
-        if (col[8*7]) {
-            MAC16(b0, + W7, col[8*7]);
-            MAC16(b1, - W5, col[8*7]);
-            MAC16(b2, + W3, col[8*7]);
-            MAC16(b3, - W1, col[8*7]);
-        }
-
-        dest[0] = cm[dest[0] + ((a0 + b0) >> COL_SHIFT)];
-        dest += line_size;
-        dest[0] = cm[dest[0] + ((a1 + b1) >> COL_SHIFT)];
-        dest += line_size;
-        dest[0] = cm[dest[0] + ((a2 + b2) >> COL_SHIFT)];
-        dest += line_size;
-        dest[0] = cm[dest[0] + ((a3 + b3) >> COL_SHIFT)];
-        dest += line_size;
-        dest[0] = cm[dest[0] + ((a3 - b3) >> COL_SHIFT)];
-        dest += line_size;
-        dest[0] = cm[dest[0] + ((a2 - b2) >> COL_SHIFT)];
-        dest += line_size;
-        dest[0] = cm[dest[0] + ((a1 - b1) >> COL_SHIFT)];
-        dest += line_size;
-        dest[0] = cm[dest[0] + ((a0 - b0) >> COL_SHIFT)];
-}
-
-static inline void idctSparseCol (DCTELEM * col)
-{
-        int a0, a1, a2, a3, b0, b1, b2, b3;
-
-        /* XXX: I did that only to give same values as previous code */
-        a0 = W4 * (col[8*0] + ((1<<(COL_SHIFT-1))/W4));
-        a1 = a0;
-        a2 = a0;
-        a3 = a0;
-
-        a0 +=  + W2*col[8*2];
-        a1 +=  + W6*col[8*2];
-        a2 +=  - W6*col[8*2];
-        a3 +=  - W2*col[8*2];
-
-        b0 = MUL16(W1, col[8*1]);
-        b1 = MUL16(W3, col[8*1]);
-        b2 = MUL16(W5, col[8*1]);
-        b3 = MUL16(W7, col[8*1]);
-
-        MAC16(b0, + W3, col[8*3]);
-        MAC16(b1, - W7, col[8*3]);
-        MAC16(b2, - W1, col[8*3]);
-        MAC16(b3, - W5, col[8*3]);
-
-        if(col[8*4]){
-            a0 += + W4*col[8*4];
-            a1 += - W4*col[8*4];
-            a2 += - W4*col[8*4];
-            a3 += + W4*col[8*4];
-        }
-
-        if (col[8*5]) {
-            MAC16(b0, + W5, col[8*5]);
-            MAC16(b1, - W1, col[8*5]);
-            MAC16(b2, + W7, col[8*5]);
-            MAC16(b3, + W3, col[8*5]);
-        }
-
-        if(col[8*6]){
-            a0 += + W6*col[8*6];
-            a1 += - W2*col[8*6];
-            a2 += + W2*col[8*6];
-            a3 += - W6*col[8*6];
-        }
-
-        if (col[8*7]) {
-            MAC16(b0, + W7, col[8*7]);
-            MAC16(b1, - W5, col[8*7]);
-            MAC16(b2, + W3, col[8*7]);
-            MAC16(b3, - W1, col[8*7]);
-        }
-
-        col[0 ] = ((a0 + b0) >> COL_SHIFT);
-        col[8 ] = ((a1 + b1) >> COL_SHIFT);
-        col[16] = ((a2 + b2) >> COL_SHIFT);
-        col[24] = ((a3 + b3) >> COL_SHIFT);
-        col[32] = ((a3 - b3) >> COL_SHIFT);
-        col[40] = ((a2 - b2) >> COL_SHIFT);
-        col[48] = ((a1 - b1) >> COL_SHIFT);
-        col[56] = ((a0 - b0) >> COL_SHIFT);
-}
-
-void ff_simple_idct_put(uint8_t *dest, int line_size, DCTELEM *block)
-{
-    int i;
-    for(i=0; i<8; i++)
-        idctRowCondDC(block + i*8);
-
-    for(i=0; i<8; i++)
-        idctSparseColPut(dest + i, line_size, block + i);
-}
-
-void ff_simple_idct_add(uint8_t *dest, int line_size, DCTELEM *block)
-{
-    int i;
-    for(i=0; i<8; i++)
-        idctRowCondDC(block + i*8);
-
-    for(i=0; i<8; i++)
-        idctSparseColAdd(dest + i, line_size, block + i);
-}
-
-void ff_simple_idct(DCTELEM *block)
-{
-    int i;
-    for(i=0; i<8; i++)
-        idctRowCondDC(block + i*8);
-
-    for(i=0; i<8; i++)
-        idctSparseCol(block + i);
-}
+#define BIT_DEPTH 10
+#include "simple_idct_template.c"
+#undef BIT_DEPTH
 
 /* 2x4x8 idct */
 
@@ -428,7 +108,7 @@ void ff_simple_idct248_put(uint8_t *dest, int line_size, DCTELEM *block)
 
     /* IDCT8 on each line */
     for(i=0; i<8; i++) {
-        idctRowCondDC(block + i*8);
+        idctRowCondDC_8(block + i*8);
     }
 
     /* IDCT4 and store */
@@ -503,7 +183,7 @@ void ff_simple_idct84_add(uint8_t *dest, int line_size, DCTELEM *block)
 
     /* IDCT8 on each line */
     for(i=0; i<4; i++) {
-        idctRowCondDC(block + i*8);
+        idctRowCondDC_8(block + i*8);
     }
 
     /* IDCT4 and store */
@@ -523,7 +203,7 @@ void ff_simple_idct48_add(uint8_t *dest, int line_size, DCTELEM *block)
 
     /* IDCT8 and store */
     for(i=0; i<4; i++){
-        idctSparseColAdd(dest + i, line_size, block + i);
+        idctSparseColAdd_8(dest + i, line_size, block + i);
     }
 }
 
diff --git a/libavcodec/simple_idct.h b/libavcodec/simple_idct.h
index 24f6a6d5db..10ac7da5e6 100644
--- a/libavcodec/simple_idct.h
+++ b/libavcodec/simple_idct.h
@@ -31,12 +31,17 @@
 #include <stdint.h>
 #include "dsputil.h"
 
-void ff_simple_idct_put(uint8_t *dest, int line_size, DCTELEM *block);
-void ff_simple_idct_add(uint8_t *dest, int line_size, DCTELEM *block);
+void ff_simple_idct_put_8(uint8_t *dest, int line_size, DCTELEM *block);
+void ff_simple_idct_add_8(uint8_t *dest, int line_size, DCTELEM *block);
+void ff_simple_idct_8(DCTELEM *block);
+
+void ff_simple_idct_put_10(uint8_t *dest, int line_size, DCTELEM *block);
+void ff_simple_idct_add_10(uint8_t *dest, int line_size, DCTELEM *block);
+void ff_simple_idct_10(DCTELEM *block);
+
 void ff_simple_idct_mmx(int16_t *block);
 void ff_simple_idct_add_mmx(uint8_t *dest, int line_size, int16_t *block);
 void ff_simple_idct_put_mmx(uint8_t *dest, int line_size, int16_t *block);
-void ff_simple_idct(DCTELEM *block);
 
 void ff_simple_idct248_put(uint8_t *dest, int line_size, DCTELEM *block);
 
diff --git a/libavcodec/simple_idct_template.c b/libavcodec/simple_idct_template.c
new file mode 100644
index 0000000000..6d3f6f764d
--- /dev/null
+++ b/libavcodec/simple_idct_template.c
@@ -0,0 +1,316 @@
+/*
+ * Simple IDCT
+ *
+ * Copyright (c) 2001 Michael Niedermayer <michaelni@gmx.at>
+ *
+ * This file is part of Libav.
+ *
+ * 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.
+ *
+ * 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 Libav; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
+ */
+
+/**
+ * @file
+ * simpleidct in C.
+ */
+
+/*
+  based upon some outcommented c code from mpeg2dec (idct_mmx.c
+  written by Aaron Holtzman <aholtzma@ess.engr.uvic.ca>)
+ */
+
+#include "bit_depth_template.c"
+
+#undef W1
+#undef W2
+#undef W3
+#undef W4
+#undef W5
+#undef W6
+#undef W7
+#undef ROW_SHIFT
+#undef COL_SHIFT
+#undef DC_SHIFT
+#undef MUL
+#undef MAC
+
+#if BIT_DEPTH == 8
+
+#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
+#define W3  19266  //cos(i*M_PI/16)*sqrt(2)*(1<<14) + 0.5
+#define W4  16383  //cos(i*M_PI/16)*sqrt(2)*(1<<14) + 0.5
+#define W5  12873  //cos(i*M_PI/16)*sqrt(2)*(1<<14) + 0.5
+#define W6  8867   //cos(i*M_PI/16)*sqrt(2)*(1<<14) + 0.5
+#define W7  4520   //cos(i*M_PI/16)*sqrt(2)*(1<<14) + 0.5
+
+#define ROW_SHIFT 11
+#define COL_SHIFT 20
+#define DC_SHIFT 3
+
+#define MUL(a, b)    MUL16(a, b)
+#define MAC(a, b, c) MAC16(a, b, c)
+
+#elif BIT_DEPTH == 10
+
+#define W1 90901
+#define W2 85627
+#define W3 77062
+#define W4 65535
+#define W5 51491
+#define W6 35468
+#define W7 18081
+
+#define ROW_SHIFT 15
+#define COL_SHIFT 20
+#define DC_SHIFT 1
+
+#define MUL(a, b)    ((a) * (b))
+#define MAC(a, b, c) ((a) += (b) * (c))
+
+#else
+
+#error "Unsupported bitdepth"
+
+#endif
+
+static inline void FUNC(idctRowCondDC)(DCTELEM *row)
+{
+    int a0, a1, a2, a3, b0, b1, b2, b3;
+
+#if HAVE_FAST_64BIT
+#define ROW0_MASK (0xffffLL << 48 * HAVE_BIGENDIAN)
+    if (((((uint64_t *)row)[0] & ~ROW0_MASK) | ((uint64_t *)row)[1]) == 0) {
+        uint64_t temp = (row[0] << DC_SHIFT) & 0xffff;
+        temp += temp << 16;
+        temp += temp << 32;
+        ((uint64_t *)row)[0] = temp;
+        ((uint64_t *)row)[1] = temp;
+        return;
+    }
+#else
+    if (!(((uint32_t*)row)[1] |
+          ((uint32_t*)row)[2] |
+          ((uint32_t*)row)[3] |
+          row[1])) {
+        uint32_t temp = (row[0] << DC_SHIFT) & 0xffff;
+        temp += temp << 16;
+        ((uint32_t*)row)[0]=((uint32_t*)row)[1] =
+            ((uint32_t*)row)[2]=((uint32_t*)row)[3] = temp;
+        return;
+    }
+#endif
+
+    a0 = (W4 * row[0]) + (1 << (ROW_SHIFT - 1));
+    a1 = a0;
+    a2 = a0;
+    a3 = a0;
+
+    a0 += W2 * row[2];
+    a1 += W6 * row[2];
+    a2 -= W6 * row[2];
+    a3 -= W2 * row[2];
+
+    b0 = MUL(W1, row[1]);
+    MAC(b0, W3, row[3]);
+    b1 = MUL(W3, row[1]);
+    MAC(b1, -W7, row[3]);
+    b2 = MUL(W5, row[1]);
+    MAC(b2, -W1, row[3]);
+    b3 = MUL(W7, row[1]);
+    MAC(b3, -W5, row[3]);
+
+    if (AV_RN64A(row + 4)) {
+        a0 +=   W4*row[4] + W6*row[6];
+        a1 += - W4*row[4] - W2*row[6];
+        a2 += - W4*row[4] + W2*row[6];
+        a3 +=   W4*row[4] - W6*row[6];
+
+        MAC(b0,  W5, row[5]);
+        MAC(b0,  W7, row[7]);
+
+        MAC(b1, -W1, row[5]);
+        MAC(b1, -W5, row[7]);
+
+        MAC(b2,  W7, row[5]);
+        MAC(b2,  W3, row[7]);
+
+        MAC(b3,  W3, row[5]);
+        MAC(b3, -W1, row[7]);
+    }
+
+    row[0] = (a0 + b0) >> ROW_SHIFT;
+    row[7] = (a0 - b0) >> ROW_SHIFT;
+    row[1] = (a1 + b1) >> ROW_SHIFT;
+    row[6] = (a1 - b1) >> ROW_SHIFT;
+    row[2] = (a2 + b2) >> ROW_SHIFT;
+    row[5] = (a2 - b2) >> ROW_SHIFT;
+    row[3] = (a3 + b3) >> ROW_SHIFT;
+    row[4] = (a3 - b3) >> ROW_SHIFT;
+}
+
+#define IDCT_COLS do {                                  \
+        a0 = W4 * (col[8*0] + ((1<<(COL_SHIFT-1))/W4)); \
+        a1 = a0;                                        \
+        a2 = a0;                                        \
+        a3 = a0;                                        \
+                                                        \
+        a0 +=  W2*col[8*2];                             \
+        a1 +=  W6*col[8*2];                             \
+        a2 += -W6*col[8*2];                             \
+        a3 += -W2*col[8*2];                             \
+                                                        \
+        b0 = MUL(W1, col[8*1]);                         \
+        b1 = MUL(W3, col[8*1]);                         \
+        b2 = MUL(W5, col[8*1]);                         \
+        b3 = MUL(W7, col[8*1]);                         \
+                                                        \
+        MAC(b0,  W3, col[8*3]);                         \
+        MAC(b1, -W7, col[8*3]);                         \
+        MAC(b2, -W1, col[8*3]);                         \
+        MAC(b3, -W5, col[8*3]);                         \
+                                                        \
+        if (col[8*4]) {                                 \
+            a0 +=  W4*col[8*4];                         \
+            a1 += -W4*col[8*4];                         \
+            a2 += -W4*col[8*4];                         \
+            a3 +=  W4*col[8*4];                         \
+        }                                               \
+                                                        \
+        if (col[8*5]) {                                 \
+            MAC(b0,  W5, col[8*5]);                     \
+            MAC(b1, -W1, col[8*5]);                     \
+            MAC(b2,  W7, col[8*5]);                     \
+            MAC(b3,  W3, col[8*5]);                     \
+        }                                               \
+                                                        \
+        if (col[8*6]) {                                 \
+            a0 +=  W6*col[8*6];                         \
+            a1 += -W2*col[8*6];                         \
+            a2 +=  W2*col[8*6];                         \
+            a3 += -W6*col[8*6];                         \
+        }                                               \
+                                                        \
+        if (col[8*7]) {                                 \
+            MAC(b0,  W7, col[8*7]);                     \
+            MAC(b1, -W5, col[8*7]);                     \
+            MAC(b2,  W3, col[8*7]);                     \
+            MAC(b3, -W1, col[8*7]);                     \
+        }                                               \
+    } while (0)
+
+static inline void FUNC(idctSparseColPut)(pixel *dest, int line_size,
+                                          DCTELEM *col)
+{
+    int a0, a1, a2, a3, b0, b1, b2, b3;
+    INIT_CLIP;
+
+    IDCT_COLS;
+
+    dest[0] = CLIP((a0 + b0) >> COL_SHIFT);
+    dest += line_size;
+    dest[0] = CLIP((a1 + b1) >> COL_SHIFT);
+    dest += line_size;
+    dest[0] = CLIP((a2 + b2) >> COL_SHIFT);
+    dest += line_size;
+    dest[0] = CLIP((a3 + b3) >> COL_SHIFT);
+    dest += line_size;
+    dest[0] = CLIP((a3 - b3) >> COL_SHIFT);
+    dest += line_size;
+    dest[0] = CLIP((a2 - b2) >> COL_SHIFT);
+    dest += line_size;
+    dest[0] = CLIP((a1 - b1) >> COL_SHIFT);
+    dest += line_size;
+    dest[0] = CLIP((a0 - b0) >> COL_SHIFT);
+}
+
+static inline void FUNC(idctSparseColAdd)(pixel *dest, int line_size,
+                                          DCTELEM *col)
+{
+    int a0, a1, a2, a3, b0, b1, b2, b3;
+    INIT_CLIP;
+
+    IDCT_COLS;
+
+    dest[0] = CLIP(dest[0] + ((a0 + b0) >> COL_SHIFT));
+    dest += line_size;
+    dest[0] = CLIP(dest[0] + ((a1 + b1) >> COL_SHIFT));
+    dest += line_size;
+    dest[0] = CLIP(dest[0] + ((a2 + b2) >> COL_SHIFT));
+    dest += line_size;
+    dest[0] = CLIP(dest[0] + ((a3 + b3) >> COL_SHIFT));
+    dest += line_size;
+    dest[0] = CLIP(dest[0] + ((a3 - b3) >> COL_SHIFT));
+    dest += line_size;
+    dest[0] = CLIP(dest[0] + ((a2 - b2) >> COL_SHIFT));
+    dest += line_size;
+    dest[0] = CLIP(dest[0] + ((a1 - b1) >> COL_SHIFT));
+    dest += line_size;
+    dest[0] = CLIP(dest[0] + ((a0 - b0) >> COL_SHIFT));
+}
+
+static inline void FUNC(idctSparseCol)(DCTELEM *col)
+{
+    int a0, a1, a2, a3, b0, b1, b2, b3;
+
+    IDCT_COLS;
+
+    col[0 ] = ((a0 + b0) >> COL_SHIFT);
+    col[8 ] = ((a1 + b1) >> COL_SHIFT);
+    col[16] = ((a2 + b2) >> COL_SHIFT);
+    col[24] = ((a3 + b3) >> COL_SHIFT);
+    col[32] = ((a3 - b3) >> COL_SHIFT);
+    col[40] = ((a2 - b2) >> COL_SHIFT);
+    col[48] = ((a1 - b1) >> COL_SHIFT);
+    col[56] = ((a0 - b0) >> COL_SHIFT);
+}
+
+void FUNC(ff_simple_idct_put)(uint8_t *dest_, int line_size, DCTELEM *block)
+{
+    pixel *dest = (pixel *)dest_;
+    int i;
+
+    line_size /= sizeof(pixel);
+
+    for (i = 0; i < 8; i++)
+        FUNC(idctRowCondDC)(block + i*8);
+
+    for (i = 0; i < 8; i++)
+        FUNC(idctSparseColPut)(dest + i, line_size, block + i);
+}
+
+void FUNC(ff_simple_idct_add)(uint8_t *dest_, int line_size, DCTELEM *block)
+{
+    pixel *dest = (pixel *)dest_;
+    int i;
+
+    line_size /= sizeof(pixel);
+
+    for (i = 0; i < 8; i++)
+        FUNC(idctRowCondDC)(block + i*8);
+
+    for (i = 0; i < 8; i++)
+        FUNC(idctSparseColAdd)(dest + i, line_size, block + i);
+}
+
+void FUNC(ff_simple_idct)(DCTELEM *block)
+{
+    int i;
+
+    for (i = 0; i < 8; i++)
+        FUNC(idctRowCondDC)(block + i*8);
+
+    for (i = 0; i < 8; i++)
+        FUNC(idctSparseCol)(block + i);
+}
diff --git a/libavcodec/sparc/dsputil_vis.c b/libavcodec/sparc/dsputil_vis.c
index e4236602f6..28061822ba 100644
--- a/libavcodec/sparc/dsputil_vis.c
+++ b/libavcodec/sparc/dsputil_vis.c
@@ -3956,7 +3956,8 @@ void dsputil_init_vis(DSPContext* c, AVCodecContext *avctx)
   const int high_bit_depth = avctx->codec_id == CODEC_ID_H264 && avctx->bits_per_raw_sample > 8;
 
   if (accel & ACCEL_SPARC_VIS) {
-      if(avctx->idct_algo==FF_IDCT_SIMPLEVIS){
+      if (avctx->bits_per_raw_sample <= 8 &&
+          avctx->idct_algo == FF_IDCT_SIMPLEVIS) {
           c->idct_put = ff_simple_idct_put_vis;
           c->idct_add = ff_simple_idct_add_vis;
           c->idct     = ff_simple_idct_vis;
diff --git a/libavcodec/vc1.c b/libavcodec/vc1.c
index d6b48b46d0..9504238b86 100644
--- a/libavcodec/vc1.c
+++ b/libavcodec/vc1.c
@@ -337,11 +337,11 @@ int vc1_decode_sequence_header(AVCodecContext *avctx, VC1Context *v, GetBitConte
     v->res_fasttx = get_bits1(gb);
     if (!v->res_fasttx)
     {
-        v->vc1dsp.vc1_inv_trans_8x8 = ff_simple_idct;
+        v->vc1dsp.vc1_inv_trans_8x8 = ff_simple_idct_8;
         v->vc1dsp.vc1_inv_trans_8x4 = ff_simple_idct84_add;
         v->vc1dsp.vc1_inv_trans_4x8 = ff_simple_idct48_add;
         v->vc1dsp.vc1_inv_trans_4x4 = ff_simple_idct44_add;
-        v->vc1dsp.vc1_inv_trans_8x8_dc = ff_simple_idct_add;
+        v->vc1dsp.vc1_inv_trans_8x8_dc = ff_simple_idct_add_8;
         v->vc1dsp.vc1_inv_trans_8x4_dc = ff_simple_idct84_add;
         v->vc1dsp.vc1_inv_trans_4x8_dc = ff_simple_idct48_add;
         v->vc1dsp.vc1_inv_trans_4x4_dc = ff_simple_idct44_add;
diff --git a/libavcodec/x86/dsputil_mmx.c b/libavcodec/x86/dsputil_mmx.c
index 12cd3c35cf..53ffacaf3e 100644
--- a/libavcodec/x86/dsputil_mmx.c
+++ b/libavcodec/x86/dsputil_mmx.c
@@ -2369,7 +2369,7 @@ void dsputil_init_mmx(DSPContext* c, AVCodecContext *avctx)
     if (mm_flags & AV_CPU_FLAG_MMX) {
         const int idct_algo= avctx->idct_algo;
 
-        if(avctx->lowres==0){
+        if (avctx->lowres == 0 && avctx->bits_per_raw_sample <= 8) {
             if(idct_algo==FF_IDCT_AUTO || idct_algo==FF_IDCT_SIMPLEMMX){
                 c->idct_put= ff_simple_idct_put_mmx;
                 c->idct_add= ff_simple_idct_add_mmx;
diff --git a/libswscale/utils.c b/libswscale/utils.c
index 480c5a2a6d..aa887310a1 100644
--- a/libswscale/utils.c
+++ b/libswscale/utils.c
@@ -182,11 +182,13 @@ int sws_isSupportedInput(enum PixelFormat pix_fmt)
         || (x)==PIX_FMT_YUV420P9LE    \
         || (x)==PIX_FMT_YUV420P10LE   \
         || (x)==PIX_FMT_YUV420P16LE   \
+        || (x)==PIX_FMT_YUV422P10LE   \
         || (x)==PIX_FMT_YUV422P16LE   \
         || (x)==PIX_FMT_YUV444P16LE   \
         || (x)==PIX_FMT_YUV422P10BE   \
         || (x)==PIX_FMT_YUV420P9BE    \
         || (x)==PIX_FMT_YUV420P10BE   \
+        || (x)==PIX_FMT_YUV422P10BE   \
         || (x)==PIX_FMT_YUV420P16BE   \
         || (x)==PIX_FMT_YUV422P16BE   \
         || (x)==PIX_FMT_YUV444P16BE   \