Merge remote-tracking branch 'qatar/master'

* qatar/master: (29 commits) amrwb: remove duplicate arguments from extrapolate_isf(). amrwb: error out early if mode is invalid. h264: change underread for 10bit QPEL to overread. matroska: check buffer size for RM-style byte reordering. vp8: disable mmx functions with sse/sse2 counterparts on x86-64. vp8: change int stride to ptrdiff_t stride. wma: fix invalid buffer size assumptions causing random overreads. Windows Media Audio Lossless decoder rv10/20: Fix slice overflow with checked bitstream reader. h263dec: Disallow width/height changing with frame threads. rv10/20: Fix a buffer overread caused by losing track of the remaining buffer size. rmdec: Honor .RMF tag size rather than assuming 18. g722: Fix the QMF scaling r3d: don't set codec timebase. electronicarts: set timebase for tgv video. electronicarts: parse the framerate for cmv video. ogg: don't set codec timebase electronicarts: don't set codec timebase avs: don't set codec timebase wavpack: Fix an integer overflow ... Conflicts: libavcodec/arm/vp8dsp_init_arm.c libavcodec/fraps.c libavcodec/h264.c libavcodec/mpeg4videodec.c libavcodec/mpegvideo.c libavcodec/msmpeg4.c libavcodec/pnmdec.c libavcodec/qpeg.c libavcodec/rawenc.c libavcodec/ulti.c libavcodec/vcr1.c libavcodec/version.h libavcodec/wmalosslessdec.c libavformat/electronicarts.c libswscale/ppc/yuv2rgb_altivec.c tests/ref/acodec/g722 tests/ref/fate/ea-cmv Merged-by: Michael Niedermayer <michaelni@gmx.at>
author: Michael Niedermayer <michaelni@gmx.at> 2012-03-03 00:09:47 +0100
committer: Michael Niedermayer <michaelni@gmx.at> 2012-03-03 00:23:10 +0100
commit: 268098d8b2a6e3dd84be788a2cd6fda10f7b3e71 (patch)
tree: afd9e17a980920b66edb38331e5a608308da711d /libavcodec
parent: 689f65126be8a55e8a1e706cb56b19bb975c20ce (diff)
parent: 9d87374ec0f382c8394ad511243db6980afa42af (diff)
download: ffmpeg-268098d8b2a6e3dd84be788a2cd6fda10f7b3e71.tar.gz
67 files changed, 1493 insertions, 1665 deletions
diff --git a/libavcodec/Makefile b/libavcodec/Makefile
index 880e30b07e..44dafb9626 100644
--- a/libavcodec/Makefile
+++ b/libavcodec/Makefile
@@ -299,12 +299,14 @@ OBJS-$(CONFIG_MPEG4_VAAPI_HWACCEL)     += vaapi_mpeg4.o
 OBJS-$(CONFIG_MSMPEG4V1_DECODER)       += msmpeg4.o msmpeg4data.o
 OBJS-$(CONFIG_MSMPEG4V2_DECODER)       += msmpeg4.o msmpeg4data.o h263dec.o \
                                           h263.o ituh263dec.o mpeg4videodec.o
-OBJS-$(CONFIG_MSMPEG4V2_ENCODER)       += msmpeg4.o msmpeg4data.o h263dec.o \
-                                          h263.o ituh263dec.o mpeg4videodec.o
+OBJS-$(CONFIG_MSMPEG4V2_ENCODER)       += msmpeg4.o msmpeg4enc.o msmpeg4data.o \
+                                          h263dec.o h263.o ituh263dec.o        \
+                                          mpeg4videodec.o
 OBJS-$(CONFIG_MSMPEG4V3_DECODER)       += msmpeg4.o msmpeg4data.o h263dec.o \
                                           h263.o ituh263dec.o mpeg4videodec.o
-OBJS-$(CONFIG_MSMPEG4V3_ENCODER)       += msmpeg4.o msmpeg4data.o h263dec.o \
-                                          h263.o ituh263dec.o mpeg4videodec.o
+OBJS-$(CONFIG_MSMPEG4V3_ENCODER)       += msmpeg4.o msmpeg4enc.o msmpeg4data.o \
+                                          h263dec.o h263.o ituh263dec.o        \
+                                          mpeg4videodec.o
 OBJS-$(CONFIG_MSRLE_DECODER)           += msrle.o msrledec.o
 OBJS-$(CONFIG_MSVIDEO1_DECODER)        += msvideo1.o
 OBJS-$(CONFIG_MSVIDEO1_ENCODER)        += msvideo1enc.o elbg.o
@@ -471,7 +473,7 @@ OBJS-$(CONFIG_WMV2_DECODER)            += wmv2dec.o wmv2.o        \
                                           msmpeg4.o msmpeg4data.o \
                                           intrax8.o intrax8dsp.o
 OBJS-$(CONFIG_WMV2_ENCODER)            += wmv2enc.o wmv2.o \
-                                          msmpeg4.o msmpeg4data.o \
+                                          msmpeg4.o msmpeg4enc.o msmpeg4data.o \
                                           mpeg4videodec.o ituh263dec.o h263dec.o
 OBJS-$(CONFIG_WNV1_DECODER)            += wnv1.o
 OBJS-$(CONFIG_WS_SND1_DECODER)         += ws-snd1.o
diff --git a/libavcodec/a64multienc.c b/libavcodec/a64multienc.c
index 0d88f72e34..a432069d26 100644
--- a/libavcodec/a64multienc.c
+++ b/libavcodec/a64multienc.c
@@ -246,7 +246,7 @@ static int a64multi_encode_frame(AVCodecContext *avctx, AVPacket *pkt,
                                  const AVFrame *pict, int *got_packet)
 {
     A64Context *c = avctx->priv_data;
-    AVFrame *const p = (AVFrame *) & c->picture;
+    AVFrame *const p = &c->picture;
 
     int frame;
     int x, y;
diff --git a/libavcodec/amrwbdec.c b/libavcodec/amrwbdec.c
index 524979d755..663fd0f2e9 100644
--- a/libavcodec/amrwbdec.c
+++ b/libavcodec/amrwbdec.c
@@ -898,10 +898,10 @@ static float auto_correlation(float *diff_isf, float mean, int lag)
  * Extrapolate a ISF vector to the 16kHz range (20th order LP)
  * used at mode 6k60 LP filter for the high frequency band.
  *
- * @param[out] out                 Buffer for extrapolated isf
- * @param[in]  isf                 Input isf vector
+ * @param[out] isf Buffer for extrapolated isf; contains LP_ORDER
+ *                 values on input
  */
-static void extrapolate_isf(float out[LP_ORDER_16k], float isf[LP_ORDER])
+static void extrapolate_isf(float isf[LP_ORDER_16k])
 {
     float diff_isf[LP_ORDER - 2], diff_mean;
     float *diff_hi = diff_isf - LP_ORDER + 1; // diff array for extrapolated indexes
@@ -909,8 +909,7 @@ static void extrapolate_isf(float out[LP_ORDER_16k], float isf[LP_ORDER])
     float est, scale;
     int i, i_max_corr;
 
-    memcpy(out, isf, (LP_ORDER - 1) * sizeof(float));
-    out[LP_ORDER_16k - 1] = isf[LP_ORDER - 1];
+    isf[LP_ORDER_16k - 1] = isf[LP_ORDER - 1];
 
     /* Calculate the difference vector */
     for (i = 0; i < LP_ORDER - 2; i++)
@@ -931,16 +930,16 @@ static void extrapolate_isf(float out[LP_ORDER_16k], float isf[LP_ORDER])
     i_max_corr++;
 
     for (i = LP_ORDER - 1; i < LP_ORDER_16k - 1; i++)
-        out[i] = isf[i - 1] + isf[i - 1 - i_max_corr]
+        isf[i] = isf[i - 1] + isf[i - 1 - i_max_corr]
                             - isf[i - 2 - i_max_corr];
 
     /* Calculate an estimate for ISF(18) and scale ISF based on the error */
-    est   = 7965 + (out[2] - out[3] - out[4]) / 6.0;
-    scale = 0.5 * (FFMIN(est, 7600) - out[LP_ORDER - 2]) /
-            (out[LP_ORDER_16k - 2] - out[LP_ORDER - 2]);
+    est   = 7965 + (isf[2] - isf[3] - isf[4]) / 6.0;
+    scale = 0.5 * (FFMIN(est, 7600) - isf[LP_ORDER - 2]) /
+            (isf[LP_ORDER_16k - 2] - isf[LP_ORDER - 2]);
 
     for (i = LP_ORDER - 1; i < LP_ORDER_16k - 1; i++)
-        diff_hi[i] = scale * (out[i] - out[i - 1]);
+        diff_hi[i] = scale * (isf[i] - isf[i - 1]);
 
     /* Stability insurance */
     for (i = LP_ORDER; i < LP_ORDER_16k - 1; i++)
@@ -952,11 +951,11 @@ static void extrapolate_isf(float out[LP_ORDER_16k], float isf[LP_ORDER])
         }
 
     for (i = LP_ORDER - 1; i < LP_ORDER_16k - 1; i++)
-        out[i] = out[i - 1] + diff_hi[i] * (1.0f / (1 << 15));
+        isf[i] = isf[i - 1] + diff_hi[i] * (1.0f / (1 << 15));
 
     /* Scale the ISF vector for 16000 Hz */
     for (i = 0; i < LP_ORDER_16k - 1; i++)
-        out[i] *= 0.8;
+        isf[i] *= 0.8;
 }
 
 /**
@@ -1003,7 +1002,7 @@ static void hb_synthesis(AMRWBContext *ctx, int subframe, float *samples,
         ff_weighted_vector_sumf(e_isf, isf_past, isf, isfp_inter[subframe],
                                 1.0 - isfp_inter[subframe], LP_ORDER);
 
-        extrapolate_isf(e_isf, e_isf);
+        extrapolate_isf(e_isf);
 
         e_isf[LP_ORDER_16k - 1] *= 2.0;
         ff_acelp_lsf2lspd(e_isp, e_isf, LP_ORDER_16k);
@@ -1095,23 +1094,27 @@ static int amrwb_decode_frame(AVCodecContext *avctx, void *data,
     buf_out = (float *)ctx->avframe.data[0];
 
     header_size      = decode_mime_header(ctx, buf);
+    if (ctx->fr_cur_mode > MODE_SID) {
+        av_log(avctx, AV_LOG_ERROR,
+               "Invalid mode %d\n", ctx->fr_cur_mode);
+        return AVERROR_INVALIDDATA;
+    }
     expected_fr_size = ((cf_sizes_wb[ctx->fr_cur_mode] + 7) >> 3) + 1;
 
     if (buf_size < expected_fr_size) {
         av_log(avctx, AV_LOG_ERROR,
             "Frame too small (%d bytes). Truncated file?\n", buf_size);
         *got_frame_ptr = 0;
-        return buf_size;
+        return AVERROR_INVALIDDATA;
     }
 
     if (!ctx->fr_quality || ctx->fr_cur_mode > MODE_SID)
         av_log(avctx, AV_LOG_ERROR, "Encountered a bad or corrupted frame\n");
 
-    if (ctx->fr_cur_mode == MODE_SID) /* Comfort noise frame */
+    if (ctx->fr_cur_mode == MODE_SID) { /* Comfort noise frame */
         av_log_missing_feature(avctx, "SID mode", 1);
-
-    if (ctx->fr_cur_mode >= MODE_SID)
         return -1;
+    }
 
     ff_amr_bit_reorder((uint16_t *) &ctx->frame, sizeof(AMRWBFrame),
         buf + header_size, amr_bit_orderings_by_mode[ctx->fr_cur_mode]);
diff --git a/libavcodec/arm/vp8dsp_init_arm.c b/libavcodec/arm/vp8dsp_init_arm.c
index 14021c954d..91f0dc95b4 100644
--- a/libavcodec/arm/vp8dsp_init_arm.c
+++ b/libavcodec/arm/vp8dsp_init_arm.c
@@ -23,87 +23,87 @@ void ff_vp8_luma_dc_wht_dc_armv6(DCTELEM block[4][4][16], DCTELEM dc[16]);
 
 #define idct_funcs(opt) \
 void ff_vp8_luma_dc_wht_ ## opt(DCTELEM block[4][4][16], DCTELEM dc[16]); \
-void ff_vp8_idct_add_ ## opt(uint8_t *dst, DCTELEM block[16], int stride); \
-void ff_vp8_idct_dc_add_ ## opt(uint8_t *dst, DCTELEM block[16], int stride); \
-void ff_vp8_idct_dc_add4y_ ## opt(uint8_t *dst, DCTELEM block[4][16], int stride); \
-void ff_vp8_idct_dc_add4uv_ ## opt(uint8_t *dst, DCTELEM block[4][16], int stride)
+void ff_vp8_idct_add_ ## opt(uint8_t *dst, DCTELEM block[16], ptrdiff_t stride); \
+void ff_vp8_idct_dc_add_ ## opt(uint8_t *dst, DCTELEM block[16], ptrdiff_t stride); \
+void ff_vp8_idct_dc_add4y_ ## opt(uint8_t *dst, DCTELEM block[4][16], ptrdiff_t stride); \
+void ff_vp8_idct_dc_add4uv_ ## opt(uint8_t *dst, DCTELEM block[4][16], ptrdiff_t stride)
 
 idct_funcs(neon);
 idct_funcs(armv6);
 
-void ff_vp8_v_loop_filter16_neon(uint8_t *dst, int stride,
+void ff_vp8_v_loop_filter16_neon(uint8_t *dst, ptrdiff_t stride,
                                  int flim_E, int flim_I, int hev_thresh);
-void ff_vp8_h_loop_filter16_neon(uint8_t *dst, int stride,
+void ff_vp8_h_loop_filter16_neon(uint8_t *dst, ptrdiff_t stride,
                                  int flim_E, int flim_I, int hev_thresh);
-void ff_vp8_v_loop_filter8uv_neon(uint8_t *dstU, uint8_t *dstV, int stride,
+void ff_vp8_v_loop_filter8uv_neon(uint8_t *dstU, uint8_t *dstV, ptrdiff_t stride,
                                   int flim_E, int flim_I, int hev_thresh);
-void ff_vp8_h_loop_filter8uv_neon(uint8_t *dstU, uint8_t *dstV, int stride,
+void ff_vp8_h_loop_filter8uv_neon(uint8_t *dstU, uint8_t *dstV, ptrdiff_t stride,
                                   int flim_E, int flim_I, int hev_thresh);
 
-void ff_vp8_v_loop_filter16_inner_neon(uint8_t *dst, int stride,
+void ff_vp8_v_loop_filter16_inner_neon(uint8_t *dst, ptrdiff_t stride,
                                        int flim_E, int flim_I, int hev_thresh);
-void ff_vp8_h_loop_filter16_inner_neon(uint8_t *dst, int stride,
+void ff_vp8_h_loop_filter16_inner_neon(uint8_t *dst, ptrdiff_t stride,
                                        int flim_E, int flim_I, int hev_thresh);
 void ff_vp8_v_loop_filter8uv_inner_neon(uint8_t *dstU, uint8_t *dstV,
-                                        int stride, int flim_E, int flim_I,
+                                        ptrdiff_t stride, int flim_E, int flim_I,
                                         int hev_thresh);
 void ff_vp8_h_loop_filter8uv_inner_neon(uint8_t *dstU, uint8_t *dstV,
-                                        int stride, int flim_E, int flim_I,
+                                        ptrdiff_t stride, int flim_E, int flim_I,
                                         int hev_thresh);
 
-void ff_vp8_v_loop_filter_inner_armv6(uint8_t *dst, int stride,
+void ff_vp8_v_loop_filter_inner_armv6(uint8_t *dst, ptrdiff_t stride,
                                       int flim_E, int flim_I,
                                       int hev_thresh, int count);
-void ff_vp8_h_loop_filter_inner_armv6(uint8_t *dst, int stride,
+void ff_vp8_h_loop_filter_inner_armv6(uint8_t *dst, ptrdiff_t stride,
                                       int flim_E, int flim_I,
                                       int hev_thresh, int count);
-void ff_vp8_v_loop_filter_armv6(uint8_t *dst, int stride,
+void ff_vp8_v_loop_filter_armv6(uint8_t *dst, ptrdiff_t stride,
                                 int flim_E, int flim_I,
                                 int hev_thresh, int count);
-void ff_vp8_h_loop_filter_armv6(uint8_t *dst, int stride,
+void ff_vp8_h_loop_filter_armv6(uint8_t *dst, ptrdiff_t stride,
                                 int flim_E, int flim_I,
                                 int hev_thresh, int count);
 
-static void ff_vp8_v_loop_filter16_armv6(uint8_t *dst, int stride,
+static void ff_vp8_v_loop_filter16_armv6(uint8_t *dst, ptrdiff_t stride,
                                          int flim_E, int flim_I, int hev_thresh)
 {
     ff_vp8_v_loop_filter_armv6(dst, stride, flim_E, flim_I, hev_thresh, 4);
 }
 
-static void ff_vp8_h_loop_filter16_armv6(uint8_t *dst, int stride,
+static void ff_vp8_h_loop_filter16_armv6(uint8_t *dst, ptrdiff_t stride,
                                          int flim_E, int flim_I, int hev_thresh)
 {
     ff_vp8_h_loop_filter_armv6(dst, stride, flim_E, flim_I, hev_thresh, 4);
 }
 
-static void ff_vp8_v_loop_filter8uv_armv6(uint8_t *dstU, uint8_t *dstV, int stride,
+static void ff_vp8_v_loop_filter8uv_armv6(uint8_t *dstU, uint8_t *dstV, ptrdiff_t stride,
                                           int flim_E, int flim_I, int hev_thresh)
 {
     ff_vp8_v_loop_filter_armv6(dstU, stride, flim_E, flim_I, hev_thresh, 2);
     ff_vp8_v_loop_filter_armv6(dstV, stride, flim_E, flim_I, hev_thresh, 2);
 }
 
-static void ff_vp8_h_loop_filter8uv_armv6(uint8_t *dstU, uint8_t *dstV, int stride,
+static void ff_vp8_h_loop_filter8uv_armv6(uint8_t *dstU, uint8_t *dstV, ptrdiff_t stride,
                                           int flim_E, int flim_I, int hev_thresh)
 {
     ff_vp8_h_loop_filter_armv6(dstU, stride, flim_E, flim_I, hev_thresh, 2);
     ff_vp8_h_loop_filter_armv6(dstV, stride, flim_E, flim_I, hev_thresh, 2);
 }
 
-static void ff_vp8_v_loop_filter16_inner_armv6(uint8_t *dst, int stride,
+static void ff_vp8_v_loop_filter16_inner_armv6(uint8_t *dst, ptrdiff_t stride,
                                                int flim_E, int flim_I, int hev_thresh)
 {
     ff_vp8_v_loop_filter_inner_armv6(dst, stride, flim_E, flim_I, hev_thresh, 4);
 }
 
-static void ff_vp8_h_loop_filter16_inner_armv6(uint8_t *dst, int stride,
+static void ff_vp8_h_loop_filter16_inner_armv6(uint8_t *dst, ptrdiff_t stride,
                                                int flim_E, int flim_I, int hev_thresh)
 {
     ff_vp8_h_loop_filter_inner_armv6(dst, stride, flim_E, flim_I, hev_thresh, 4);
 }
 
 static void ff_vp8_v_loop_filter8uv_inner_armv6(uint8_t *dstU, uint8_t *dstV,
-                                                int stride, int flim_E, int flim_I,
+                                                ptrdiff_t stride, int flim_E, int flim_I,
                                                 int hev_thresh)
 {
     ff_vp8_v_loop_filter_inner_armv6(dstU, stride, flim_E, flim_I, hev_thresh, 2);
@@ -111,7 +111,7 @@ static void ff_vp8_v_loop_filter8uv_inner_armv6(uint8_t *dstU, uint8_t *dstV,
 }
 
 static void ff_vp8_h_loop_filter8uv_inner_armv6(uint8_t *dstU, uint8_t *dstV,
-                                                int stride, int flim_E, int flim_I,
+                                                ptrdiff_t stride, int flim_E, int flim_I,
                                                 int hev_thresh)
 {
     ff_vp8_h_loop_filter_inner_armv6(dstU, stride, flim_E, flim_I, hev_thresh, 2);
@@ -119,16 +119,16 @@ static void ff_vp8_h_loop_filter8uv_inner_armv6(uint8_t *dstU, uint8_t *dstV,
 }
 
 #define simple_lf_funcs(opt) \
-void ff_vp8_v_loop_filter16_simple_ ## opt(uint8_t *dst, int stride, int flim); \
-void ff_vp8_h_loop_filter16_simple_ ## opt(uint8_t *dst, int stride, int flim)
+void ff_vp8_v_loop_filter16_simple_ ## opt(uint8_t *dst, ptrdiff_t stride, int flim); \
+void ff_vp8_h_loop_filter16_simple_ ## opt(uint8_t *dst, ptrdiff_t stride, int flim)
 
 simple_lf_funcs(neon);
 simple_lf_funcs(armv6);
 
-#define VP8_MC_OPT(n, opt)                                              \
-    void ff_put_vp8_##n##_##opt(uint8_t *dst, int dststride,            \
-                                uint8_t *src, int srcstride,            \
-                                int h, int x, int y)
+#define VP8_MC(n)                                                       \
+    void ff_put_vp8_##n##_neon(uint8_t *dst, ptrdiff_t dststride,       \
+                               uint8_t *src, ptrdiff_t srcstride,       \
+                               int h, int x, int y)
 
 #define VP8_MC(n) \
     VP8_MC_OPT(n, neon)
diff --git a/libavcodec/asv1.c b/libavcodec/asv1.c
index 4bcc8d72b1..c123125d46 100644
--- a/libavcodec/asv1.c
+++ b/libavcodec/asv1.c
@@ -454,7 +454,7 @@ static int decode_frame(AVCodecContext *avctx,
         }
     }
 
-    *picture= *(AVFrame*)&a->picture;
+    *picture   = a->picture;
     *data_size = sizeof(AVPicture);
 
     emms_c();
diff --git a/libavcodec/avs.c b/libavcodec/avs.c
index 05cb815fd8..4dcbb27fe0 100644
--- a/libavcodec/avs.c
+++ b/libavcodec/avs.c
@@ -51,7 +51,7 @@ avs_decode_frame(AVCodecContext * avctx,
     int buf_size = avpkt->size;
     AvsContext *const avs = avctx->priv_data;
     AVFrame *picture = data;
-    AVFrame *const p = (AVFrame *) & avs->picture;
+    AVFrame *const p =  &avs->picture;
     const uint8_t *table, *vect;
     uint8_t *out;
     int i, j, x, y, stride, vect_w = 3, vect_h = 3;
@@ -151,7 +151,7 @@ avs_decode_frame(AVCodecContext * avctx,
             align_get_bits(&change_map);
     }
 
-    *picture = *(AVFrame *) & avs->picture;
+    *picture   = avs->picture;
     *data_size = sizeof(AVPicture);
 
     return buf_size;
diff --git a/libavcodec/bmp.c b/libavcodec/bmp.c
index b7853d1093..b249f312d4 100644
--- a/libavcodec/bmp.c
+++ b/libavcodec/bmp.c
@@ -27,8 +27,8 @@
 static av_cold int bmp_decode_init(AVCodecContext *avctx){
     BMPContext *s = avctx->priv_data;
 
-    avcodec_get_frame_defaults((AVFrame*)&s->picture);
-    avctx->coded_frame = (AVFrame*)&s->picture;
+    avcodec_get_frame_defaults(&s->picture);
+    avctx->coded_frame = &s->picture;
 
     return 0;
 }
diff --git a/libavcodec/bmpenc.c b/libavcodec/bmpenc.c
index 4455af7d8c..1c40b13add 100644
--- a/libavcodec/bmpenc.c
+++ b/libavcodec/bmpenc.c
@@ -34,8 +34,8 @@ static const uint32_t rgb444_masks[]  = { 0x0F00, 0x00F0, 0x000F };
 static av_cold int bmp_encode_init(AVCodecContext *avctx){
     BMPContext *s = avctx->priv_data;
 
-    avcodec_get_frame_defaults((AVFrame*)&s->picture);
-    avctx->coded_frame = (AVFrame*)&s->picture;
+    avcodec_get_frame_defaults(&s->picture);
+    avctx->coded_frame = &s->picture;
 
     switch (avctx->pix_fmt) {
     case PIX_FMT_BGRA:
@@ -72,7 +72,7 @@ static int bmp_encode_frame(AVCodecContext *avctx, AVPacket *pkt,
                             const AVFrame *pict, int *got_packet)
 {
     BMPContext *s = avctx->priv_data;
-    AVFrame * const p= (AVFrame*)&s->picture;
+    AVFrame * const p = &s->picture;
     int n_bytes_image, n_bytes_per_row, n_bytes, i, n, hsize, ret;
     const uint32_t *pal = NULL;
     uint32_t palette256[256];
diff --git a/libavcodec/cavsdec.c b/libavcodec/cavsdec.c
index fb0ec841ce..b06bd53c00 100644
--- a/libavcodec/cavsdec.c
+++ b/libavcodec/cavsdec.c
@@ -501,9 +501,9 @@ static int decode_pic(AVSContext *h) {
     }
     /* release last B frame */
     if(h->picture.f.data[0])
-        s->avctx->release_buffer(s->avctx, (AVFrame *)&h->picture);
+        s->avctx->release_buffer(s->avctx, &h->picture.f);
 
-    s->avctx->get_buffer(s->avctx, (AVFrame *)&h->picture);
+    s->avctx->get_buffer(s->avctx, &h->picture.f);
     ff_cavs_init_pic(h);
     h->picture.poc = get_bits(&s->gb,8)*2;
 
@@ -592,7 +592,7 @@ static int decode_pic(AVSContext *h) {
     }
     if(h->pic_type != AV_PICTURE_TYPE_B) {
         if(h->DPB[1].f.data[0])
-            s->avctx->release_buffer(s->avctx, (AVFrame *)&h->DPB[1]);
+            s->avctx->release_buffer(s->avctx, &h->DPB[1].f);
         h->DPB[1] = h->DPB[0];
         h->DPB[0] = h->picture;
         memset(&h->picture,0,sizeof(Picture));
@@ -656,7 +656,7 @@ static int cavs_decode_frame(AVCodecContext * avctx,void *data, int *data_size,
     if (buf_size == 0) {
         if (!s->low_delay && h->DPB[0].f.data[0]) {
             *data_size = sizeof(AVPicture);
-            *picture = *(AVFrame *) &h->DPB[0];
+            *picture = h->DPB[0].f;
         }
         return 0;
     }
@@ -676,9 +676,9 @@ static int cavs_decode_frame(AVCodecContext * avctx,void *data, int *data_size,
         case PIC_I_START_CODE:
             if(!h->got_keyframe) {
                 if(h->DPB[0].f.data[0])
-                    avctx->release_buffer(avctx, (AVFrame *)&h->DPB[0]);
+                    avctx->release_buffer(avctx, &h->DPB[0].f);
                 if(h->DPB[1].f.data[0])
-                    avctx->release_buffer(avctx, (AVFrame *)&h->DPB[1]);
+                    avctx->release_buffer(avctx, &h->DPB[1].f);
                 h->got_keyframe = 1;
             }
         case PIC_PB_START_CODE:
@@ -692,12 +692,12 @@ static int cavs_decode_frame(AVCodecContext * avctx,void *data, int *data_size,
             *data_size = sizeof(AVPicture);
             if(h->pic_type != AV_PICTURE_TYPE_B) {
                 if(h->DPB[1].f.data[0]) {
-                    *picture = *(AVFrame *) &h->DPB[1];
+                    *picture = h->DPB[1].f;
                 } else {
                     *data_size = 0;
                 }
             } else
-                *picture = *(AVFrame *) &h->picture;
+                *picture = h->picture.f;
             break;
         case EXT_START_CODE:
             //mpeg_decode_extension(avctx,buf_ptr, input_size);
diff --git a/libavcodec/error_resilience.c b/libavcodec/error_resilience.c
index 2391e57f90..1c1420f93c 100644
--- a/libavcodec/error_resilience.c
+++ b/libavcodec/error_resilience.c
@@ -613,7 +613,7 @@ skip_mean_and_median:
                         if (s->avctx->codec_id == CODEC_ID_H264) {
                             // FIXME
                         } else {
-                            ff_thread_await_progress((AVFrame *) s->last_picture_ptr,
+                            ff_thread_await_progress(&s->last_picture_ptr->f,
                                                      mb_y, 0);
                         }
                         if (!s->last_picture.f.motion_val[0] ||
@@ -786,7 +786,7 @@ static int is_intra_more_likely(MpegEncContext *s)
                 if (s->avctx->codec_id == CODEC_ID_H264) {
                     // FIXME
                 } else {
-                    ff_thread_await_progress((AVFrame *) s->last_picture_ptr,
+                    ff_thread_await_progress(&s->last_picture_ptr->f,
                                              mb_y, 0);
                 }
                 is_intra_likely += s->dsp.sad[0](NULL, last_mb_ptr, mb_ptr                    , s->linesize, 16);
@@ -1170,7 +1170,7 @@ void ff_er_frame_end(MpegEncContext *s)
                     if (s->avctx->codec_id == CODEC_ID_H264) {
                         // FIXME
                     } else {
-                        ff_thread_await_progress((AVFrame *) s->next_picture_ptr, mb_y, 0);
+                        ff_thread_await_progress(&s->next_picture_ptr->f, mb_y, 0);
                     }
                     s->mv[0][0][0] = s->next_picture.f.motion_val[0][xy][0] *  time_pb            / time_pp;
                     s->mv[0][0][1] = s->next_picture.f.motion_val[0][xy][1] *  time_pb            / time_pp;
diff --git a/libavcodec/fraps.c b/libavcodec/fraps.c
index 45d95c9c24..e8c0b18b5d 100644
--- a/libavcodec/fraps.c
+++ b/libavcodec/fraps.c
@@ -62,7 +62,7 @@ static av_cold int decode_init(AVCodecContext *avctx)
     FrapsContext * const s = avctx->priv_data;
 
     avcodec_get_frame_defaults(&s->frame);
-    avctx->coded_frame = (AVFrame*)&s->frame;
+    avctx->coded_frame = &s->frame;
 
     s->avctx = avctx;
     s->tmpbuf = NULL;
@@ -132,7 +132,7 @@ static int decode_frame(AVCodecContext *avctx,
     int buf_size = avpkt->size;
     FrapsContext * const s = avctx->priv_data;
     AVFrame *frame = data;
-    AVFrame * const f = (AVFrame*)&s->frame;
+    AVFrame * const f = &s->frame;
     uint32_t header;
     unsigned int version,header_size;
     unsigned int x, y;
diff --git a/libavcodec/g722dec.c b/libavcodec/g722dec.c
index 50a224ba10..72bb0ef3c7 100644
--- a/libavcodec/g722dec.c
+++ b/libavcodec/g722dec.c
@@ -126,8 +126,8 @@ static int g722_decode_frame(AVCodecContext *avctx, void *data,
         c->prev_samples[c->prev_samples_pos++] = rlow - rhigh;
         ff_g722_apply_qmf(c->prev_samples + c->prev_samples_pos - 24,
                           &xout1, &xout2);
-        *out_buf++ = av_clip_int16(xout1 >> 12);
-        *out_buf++ = av_clip_int16(xout2 >> 12);
+        *out_buf++ = av_clip_int16(xout1 >> 11);
+        *out_buf++ = av_clip_int16(xout2 >> 11);
         if (c->prev_samples_pos >= PREV_SAMPLES_BUF_SIZE) {
             memmove(c->prev_samples, c->prev_samples + c->prev_samples_pos - 22,
                     22 * sizeof(c->prev_samples[0]));
diff --git a/libavcodec/g722enc.c b/libavcodec/g722enc.c
index a5ae0a5153..ba8ceeff86 100644
--- a/libavcodec/g722enc.c
+++ b/libavcodec/g722enc.c
@@ -136,8 +136,8 @@ static inline void filter_samples(G722Context *c, const int16_t *samples,
     c->prev_samples[c->prev_samples_pos++] = samples[0];
     c->prev_samples[c->prev_samples_pos++] = samples[1];
     ff_g722_apply_qmf(c->prev_samples + c->prev_samples_pos - 24, &xout1, &xout2);
-    *xlow  = xout1 + xout2 >> 13;
-    *xhigh = xout1 - xout2 >> 13;
+    *xlow  = xout1 + xout2 >> 14;
+    *xhigh = xout1 - xout2 >> 14;
     if (c->prev_samples_pos >= PREV_SAMPLES_BUF_SIZE) {
         memmove(c->prev_samples,
                 c->prev_samples + c->prev_samples_pos - 22,
diff --git a/libavcodec/gif.c b/libavcodec/gif.c
index 875c5b15dc..2ae63865d2 100644
--- a/libavcodec/gif.c
+++ b/libavcodec/gif.c
@@ -160,7 +160,7 @@ static int gif_encode_frame(AVCodecContext *avctx, AVPacket *pkt,
                             const AVFrame *pict, int *got_packet)
 {
     GIFContext *s = avctx->priv_data;
-    AVFrame *const p = (AVFrame *)&s->picture;
+    AVFrame *const p = &s->picture;
     uint8_t *outbuf_ptr, *end;
     int ret;
 
diff --git a/libavcodec/h261dec.c b/libavcodec/h261dec.c
index 68e3212a19..e97c76dab4 100644
--- a/libavcodec/h261dec.c
+++ b/libavcodec/h261dec.c
@@ -628,7 +628,8 @@ retry:
 
 assert(s->current_picture.f.pict_type == s->current_picture_ptr->f.pict_type);
 assert(s->current_picture.f.pict_type == s->pict_type);
-    *pict= *(AVFrame*)s->current_picture_ptr;
+
+    *pict = s->current_picture_ptr->f;
     ff_print_debug_info(s, pict);
 
     *data_size = sizeof(AVFrame);
diff --git a/libavcodec/h263dec.c b/libavcodec/h263dec.c
index ac2aeaf424..b5b6d8c863 100644
--- a/libavcodec/h263dec.c
+++ b/libavcodec/h263dec.c
@@ -357,7 +357,7 @@ uint64_t time= rdtsc();
     if (buf_size == 0) {
         /* special case for last picture */
         if (s->low_delay==0 && s->next_picture_ptr) {
-            *pict= *(AVFrame*)s->next_picture_ptr;
+            *pict = s->next_picture_ptr->f;
             s->next_picture_ptr= NULL;
 
             *data_size = sizeof(AVFrame);
@@ -727,9 +727,9 @@ intrax8_decoded:
     assert(s->current_picture.f.pict_type == s->current_picture_ptr->f.pict_type);
     assert(s->current_picture.f.pict_type == s->pict_type);
     if (s->pict_type == AV_PICTURE_TYPE_B || s->low_delay) {
-        *pict= *(AVFrame*)s->current_picture_ptr;
+        *pict = s->current_picture_ptr->f;
     } else if (s->last_picture_ptr != NULL) {
-        *pict= *(AVFrame*)s->last_picture_ptr;
+        *pict = s->last_picture_ptr->f;
     }
 
     if(s->last_picture_ptr || s->low_delay){
diff --git a/libavcodec/h264.c b/libavcodec/h264.c
index 20daa2e0a8..caf4b8ae5f 100644
--- a/libavcodec/h264.c
+++ b/libavcodec/h264.c
@@ -367,14 +367,14 @@ static void await_references(H264Context *h){
                 nrefs[list]--;
 
                 if(!FIELD_PICTURE && ref_field_picture){ // frame referencing two fields
-                    ff_thread_await_progress((AVFrame*)ref_pic, FFMIN((row >> 1) - !(row&1), pic_height-1), 1);
-                    ff_thread_await_progress((AVFrame*)ref_pic, FFMIN((row >> 1)           , pic_height-1), 0);
+                    ff_thread_await_progress(&ref_pic->f, FFMIN((row >> 1) - !(row & 1), pic_height - 1), 1);
+                    ff_thread_await_progress(&ref_pic->f, FFMIN((row >> 1),              pic_height - 1), 0);
                 }else if(FIELD_PICTURE && !ref_field_picture){ // field referencing one field of a frame
-                    ff_thread_await_progress((AVFrame*)ref_pic, FFMIN(row*2 + ref_field    , pic_height-1), 0);
+                    ff_thread_await_progress(&ref_pic->f, FFMIN(row * 2 + ref_field, pic_height - 1), 0);
                 }else if(FIELD_PICTURE){
-                    ff_thread_await_progress((AVFrame*)ref_pic, FFMIN(row, pic_height-1), ref_field);
+                    ff_thread_await_progress(&ref_pic->f, FFMIN(row, pic_height - 1), ref_field);
                 }else{
-                    ff_thread_await_progress((AVFrame*)ref_pic, FFMIN(row, pic_height-1), 0);
+                    ff_thread_await_progress(&ref_pic->f, FFMIN(row, pic_height - 1), 0);
                 }
             }
         }
@@ -2507,8 +2507,9 @@ static int field_end(H264Context *h, int in_setup){
     s->mb_y= 0;
 
     if (!in_setup && !s->dropable)
-        ff_thread_report_progress((AVFrame*)s->current_picture_ptr, (16*s->mb_height >> FIELD_PICTURE) - 1,
-                                 s->picture_structure==PICT_BOTTOM_FIELD);
+        ff_thread_report_progress(&s->current_picture_ptr->f,
+                                  (16 * s->mb_height >> FIELD_PICTURE) - 1,
+                                  s->picture_structure == PICT_BOTTOM_FIELD);
 
     if (CONFIG_H264_VDPAU_DECODER && s->avctx->codec->capabilities&CODEC_CAP_HWACCEL_VDPAU)
         ff_vdpau_h264_set_reference_frames(s);
@@ -2906,8 +2907,8 @@ static int decode_slice_header(H264Context *h, H264Context *h0){
             h->prev_frame_num++;
             h->prev_frame_num %= 1<<h->sps.log2_max_frame_num;
             s->current_picture_ptr->frame_num= h->prev_frame_num;
-            ff_thread_report_progress((AVFrame*)s->current_picture_ptr, INT_MAX, 0);
-            ff_thread_report_progress((AVFrame*)s->current_picture_ptr, INT_MAX, 1);
+            ff_thread_report_progress(&s->current_picture_ptr->f, INT_MAX, 0);
+            ff_thread_report_progress(&s->current_picture_ptr->f, INT_MAX, 1);
             ff_generate_sliding_window_mmcos(h);
             if (ff_h264_execute_ref_pic_marking(h, h->mmco, h->mmco_index) < 0 &&
                 (s->avctx->err_recognition & AV_EF_EXPLODE))
@@ -3577,8 +3578,8 @@ static void decode_finish_row(H264Context *h){
 
     if (s->dropable) return;
 
-    ff_thread_report_progress((AVFrame*)s->current_picture_ptr, top + height - 1,
-                             s->picture_structure==PICT_BOTTOM_FIELD);
+    ff_thread_report_progress(&s->current_picture_ptr->f, top + height - 1,
+                              s->picture_structure == PICT_BOTTOM_FIELD);
 }
 
 static int decode_slice(struct AVCodecContext *avctx, void *arg){
@@ -4067,7 +4068,7 @@ static int decode_frame(AVCodecContext *avctx,
 
         if(out){
             *data_size = sizeof(AVFrame);
-            *pict= *(AVFrame*)out;
+            *pict      = out->f;
         }
 
         return buf_index;
@@ -4121,7 +4122,7 @@ not_extra:
         *data_size = 0; /* Wait for second field. */
         if (h->next_output_pic && (h->next_output_pic->sync || h->sync>1)) {
                 *data_size = sizeof(AVFrame);
-                *pict = *(AVFrame*)h->next_output_pic;
+                *pict      = h->next_output_pic->f;
         }
     }
 
diff --git a/libavcodec/h264_direct.c b/libavcodec/h264_direct.c
index 079c665509..263832d829 100644
--- a/libavcodec/h264_direct.c
+++ b/libavcodec/h264_direct.c
@@ -154,7 +154,8 @@ static void await_reference_mb_row(H264Context * const h, Picture *ref, int mb_y
     //FIXME it can be safe to access mb stuff
     //even if pixels aren't deblocked yet
 
-    ff_thread_await_progress((AVFrame*)ref, FFMIN(16*mb_y >> ref_field_picture, ref_height-1),
+    ff_thread_await_progress(&ref->f,
+                             FFMIN(16 * mb_y >> ref_field_picture, ref_height - 1),
                              ref_field_picture && ref_field);
 }
 
diff --git a/libavcodec/indeo2.c b/libavcodec/indeo2.c
index eb58939338..ec5a86ad76 100644
--- a/libavcodec/indeo2.c
+++ b/libavcodec/indeo2.c
@@ -143,7 +143,7 @@ static int ir2_decode_frame(AVCodecContext *avctx,
     int buf_size = avpkt->size;
     Ir2Context * const s = avctx->priv_data;
     AVFrame *picture = data;
-    AVFrame * const p= (AVFrame*)&s->picture;
+    AVFrame * const p = &s->picture;
     int start;
 
     p->reference = 3;
@@ -188,7 +188,7 @@ static int ir2_decode_frame(AVCodecContext *avctx,
                          s->picture.data[1], s->picture.linesize[1], ir2_luma_table);
     }
 
-    *picture= *(AVFrame*)&s->picture;
+    *picture   = s->picture;
     *data_size = sizeof(AVPicture);
 
     return buf_size;
diff --git a/libavcodec/jpeglsenc.c b/libavcodec/jpeglsenc.c
index d5135adb30..62a2328a0a 100644
--- a/libavcodec/jpeglsenc.c
+++ b/libavcodec/jpeglsenc.c
@@ -232,7 +232,7 @@ static int encode_picture_ls(AVCodecContext *avctx, AVPacket *pkt,
                              const AVFrame *pict, int *got_packet)
 {
     JpeglsContext * const s = avctx->priv_data;
-    AVFrame * const p= (AVFrame*)&s->picture;
+    AVFrame * const p = &s->picture;
     const int near = avctx->prediction_method;
     PutBitContext pb, pb2;
     GetBitContext gb;
diff --git a/libavcodec/ljpegenc.c b/libavcodec/ljpegenc.c
index 00d34ecc9c..b5b443e7cc 100644
--- a/libavcodec/ljpegenc.c
+++ b/libavcodec/ljpegenc.c
@@ -45,7 +45,7 @@ static int encode_picture_lossless(AVCodecContext *avctx, AVPacket *pkt,
     MJpegContext * const m = s->mjpeg_ctx;
     const int width= s->width;
     const int height= s->height;
-    AVFrame * const p= (AVFrame*)&s->current_picture;
+    AVFrame * const p = &s->current_picture.f;
     const int predictor= avctx->prediction_method+1;
     const int mb_width  = (width  + s->mjpeg_hsample[0] - 1) / s->mjpeg_hsample[0];
     const int mb_height = (height + s->mjpeg_vsample[0] - 1) / s->mjpeg_vsample[0];
diff --git a/libavcodec/loco.c b/libavcodec/loco.c
index 75701e970b..eaf7e81cd4 100644
--- a/libavcodec/loco.c
+++ b/libavcodec/loco.c
@@ -166,7 +166,7 @@ static int decode_frame(AVCodecContext *avctx,
     const uint8_t *buf = avpkt->data;
     int buf_size = avpkt->size;
     LOCOContext * const l = avctx->priv_data;
-    AVFrame * const p= (AVFrame*)&l->pic;
+    AVFrame * const p = &l->pic;
     int decoded;
 
     if(p->data[0])
diff --git a/libavcodec/mdec.c b/libavcodec/mdec.c
index 9a417ee45b..03361c7c79 100644
--- a/libavcodec/mdec.c
+++ b/libavcodec/mdec.c
@@ -243,7 +243,7 @@ static av_cold int decode_init(AVCodecContext *avctx){
 
 static av_cold int decode_init_thread_copy(AVCodecContext *avctx){
     MDECContext * const a = avctx->priv_data;
-    AVFrame *p = (AVFrame*)&a->picture;
+    AVFrame *p = &a->picture;
 
     avctx->coded_frame= p;
     a->avctx= avctx;
diff --git a/libavcodec/mjpegdec.c b/libavcodec/mjpegdec.c
index 0a9d03061f..c5db0f1d41 100644
--- a/libavcodec/mjpegdec.c
+++ b/libavcodec/mjpegdec.c
@@ -1577,7 +1577,7 @@ int ff_mjpeg_decode_frame(AVCodecContext *avctx, void *data, int *data_size,
             goto the_end;
         } else if (unescaped_buf_size > (1U<<29)) {
             av_log(avctx, AV_LOG_ERROR, "MJPEG packet 0x%x too big (0x%x/0x%x), corrupt data?\n",
-                   start_code, unescaped_buf_ptr, buf_size);
+                   start_code, unescaped_buf_size, buf_size);
             return AVERROR_INVALIDDATA;
         } else {
             av_log(avctx, AV_LOG_DEBUG, "marker=%x avail_size_in_buf=%td\n",
diff --git a/libavcodec/mpeg12.c b/libavcodec/mpeg12.c
index a4e4661d18..5c5e09ec2d 100644
--- a/libavcodec/mpeg12.c
+++ b/libavcodec/mpeg12.c
@@ -1952,7 +1952,7 @@ static int slice_end(AVCodecContext *avctx, AVFrame *pict)
         ff_MPV_frame_end(s);
 
         if (s->pict_type == AV_PICTURE_TYPE_B || s->low_delay) {
-            *pict = *(AVFrame*)s->current_picture_ptr;
+            *pict = s->current_picture_ptr->f;
             ff_print_debug_info(s, pict);
         } else {
             if (avctx->active_thread_type & FF_THREAD_FRAME)
@@ -1960,7 +1960,7 @@ static int slice_end(AVCodecContext *avctx, AVFrame *pict)
             /* latency of 1 frame for I- and P-frames */
             /* XXX: use another variable than picture_number */
             if (s->last_picture_ptr != NULL) {
-                *pict = *(AVFrame*)s->last_picture_ptr;
+                *pict = s->last_picture_ptr->f;
                  ff_print_debug_info(s, pict);
             }
         }
@@ -2256,7 +2256,7 @@ static int mpeg_decode_frame(AVCodecContext *avctx,
     if (buf_size == 0 || (buf_size == 4 && AV_RB32(buf) == SEQ_END_CODE)) {
         /* special case for last picture */
         if (s2->low_delay == 0 && s2->next_picture_ptr) {
-            *picture = *(AVFrame*)s2->next_picture_ptr;
+            *picture = s2->next_picture_ptr->f;
             s2->next_picture_ptr = NULL;
 
             *data_size = sizeof(AVFrame);
diff --git a/libavcodec/mpeg4videodec.c b/libavcodec/mpeg4videodec.c
index a12ad99348..c87af8216e 100644
--- a/libavcodec/mpeg4videodec.c
+++ b/libavcodec/mpeg4videodec.c
@@ -1313,7 +1313,7 @@ static int mpeg4_decode_mb(MpegEncContext *s,
                 s->last_mv[i][1][1]= 0;
             }
 
-            ff_thread_await_progress((AVFrame*)s->next_picture_ptr, s->mb_y, 0);
+            ff_thread_await_progress(&s->next_picture_ptr->f, s->mb_y, 0);
         }
 
         /* if we skipped it in the future P Frame than skip it now too */
@@ -1500,7 +1500,7 @@ end:
 
             if(s->pict_type==AV_PICTURE_TYPE_B){
                 const int delta= s->mb_x + 1 == s->mb_width ? 2 : 1;
-                ff_thread_await_progress((AVFrame*)s->next_picture_ptr,
+                ff_thread_await_progress(&s->next_picture_ptr->f,
                                         (s->mb_x + delta >= s->mb_width) ? FFMIN(s->mb_y+1, s->mb_height-1) : s->mb_y, 0);
                 if (s->next_picture.f.mbskip_table[xy + delta])
                     return SLICE_OK;
diff --git a/libavcodec/mpegvideo.c b/libavcodec/mpegvideo.c
index efe4cae762..9577c9d38f 100644
--- a/libavcodec/mpegvideo.c
+++ b/libavcodec/mpegvideo.c
@@ -232,9 +232,9 @@ static void free_frame_buffer(MpegEncContext *s, Picture *pic)
      * dimensions; ignore user defined callbacks for these
      */
     if (s->codec_id != CODEC_ID_WMV3IMAGE && s->codec_id != CODEC_ID_VC1IMAGE)
-        ff_thread_release_buffer(s->avctx, (AVFrame *) pic);
+        ff_thread_release_buffer(s->avctx, &pic->f);
     else
-        avcodec_default_release_buffer(s->avctx, (AVFrame *) pic);
+        avcodec_default_release_buffer(s->avctx, &pic->f);
     av_freep(&pic->f.hwaccel_picture_private);
 }
 
@@ -257,9 +257,9 @@ static int alloc_frame_buffer(MpegEncContext *s, Picture *pic)
     }
 
     if (s->codec_id != CODEC_ID_WMV3IMAGE && s->codec_id != CODEC_ID_VC1IMAGE)
-        r = ff_thread_get_buffer(s->avctx, (AVFrame *) pic);
+        r = ff_thread_get_buffer(s->avctx, &pic->f);
     else
-        r = avcodec_default_get_buffer(s->avctx, (AVFrame *) pic);
+        r = avcodec_default_get_buffer(s->avctx, &pic->f);
 
     if (r < 0 || !pic->f.type || !pic->f.data[0]) {
         av_log(s->avctx, AV_LOG_ERROR, "get_buffer() failed (%d %d %p)\n",
@@ -729,7 +729,7 @@ av_cold int ff_MPV_common_init(MpegEncContext *s)
     s->codec_tag        = avpriv_toupper4(s->avctx->codec_tag);
     s->stream_codec_tag = avpriv_toupper4(s->avctx->stream_codec_tag);
 
-    s->avctx->coded_frame = (AVFrame*)&s->current_picture;
+    s->avctx->coded_frame = &s->current_picture.f;
 
     FF_ALLOCZ_OR_GOTO(s->avctx, s->mb_index2xy, (s->mb_num + 1) * sizeof(int), fail); // error ressilience code looks cleaner with this
     for (y = 0; y < s->mb_height; y++)
@@ -781,7 +781,7 @@ av_cold int ff_MPV_common_init(MpegEncContext *s)
     FF_ALLOCZ_OR_GOTO(s->avctx, s->picture,
                       s->picture_count * sizeof(Picture), fail);
     for (i = 0; i < s->picture_count; i++) {
-        avcodec_get_frame_defaults((AVFrame *) &s->picture[i]);
+        avcodec_get_frame_defaults(&s->picture[i].f);
     }
 
     FF_ALLOCZ_OR_GOTO(s->avctx, s->error_status_table, mb_array_size*sizeof(uint8_t), fail)
@@ -1247,10 +1247,8 @@ int ff_MPV_frame_start(MpegEncContext *s, AVCodecContext *avctx)
                     memset(s->last_picture_ptr->f.data[0] + s->last_picture_ptr->f.linesize[0]*i, 16, avctx->width);
             }
 
-            ff_thread_report_progress((AVFrame *) s->last_picture_ptr,
-                                      INT_MAX, 0);
-            ff_thread_report_progress((AVFrame *) s->last_picture_ptr,
-                                      INT_MAX, 1);
+            ff_thread_report_progress(&s->last_picture_ptr->f, INT_MAX, 0);
+            ff_thread_report_progress(&s->last_picture_ptr->f, INT_MAX, 1);
         }
         if ((s->next_picture_ptr == NULL ||
              s->next_picture_ptr->f.data[0] == NULL) &&
@@ -1263,10 +1261,8 @@ int ff_MPV_frame_start(MpegEncContext *s, AVCodecContext *avctx)
             s->next_picture_ptr->f.key_frame = 0;
             if (ff_alloc_picture(s, s->next_picture_ptr, 0) < 0)
                 return -1;
-            ff_thread_report_progress((AVFrame *) s->next_picture_ptr,
-                                      INT_MAX, 0);
-            ff_thread_report_progress((AVFrame *) s->next_picture_ptr,
-                                      INT_MAX, 1);
+            ff_thread_report_progress(&s->next_picture_ptr->f, INT_MAX, 0);
+            ff_thread_report_progress(&s->next_picture_ptr->f, INT_MAX, 1);
         }
     }
 
@@ -1391,10 +1387,10 @@ void ff_MPV_frame_end(MpegEncContext *s)
     memset(&s->next_picture,    0, sizeof(Picture));
     memset(&s->current_picture, 0, sizeof(Picture));
 #endif
-    s->avctx->coded_frame = (AVFrame *) s->current_picture_ptr;
+    s->avctx->coded_frame = &s->current_picture_ptr->f;
 
     if (s->codec_id != CODEC_ID_H264 && s->current_picture.f.reference) {
-        ff_thread_report_progress((AVFrame *) s->current_picture_ptr,
+        ff_thread_report_progress(&s->current_picture_ptr->f,
                                   s->mb_height - 1, 0);
     }
 }
@@ -2346,10 +2342,14 @@ void MPV_decode_mb_internal(MpegEncContext *s, DCTELEM block[12][64],
 
                 if(HAVE_THREADS && s->avctx->active_thread_type&FF_THREAD_FRAME) {
                     if (s->mv_dir & MV_DIR_FORWARD) {
-                        ff_thread_await_progress((AVFrame*)s->last_picture_ptr, ff_MPV_lowest_referenced_row(s, 0), 0);
+                        ff_thread_await_progress(&s->last_picture_ptr->f,
+                                                 ff_MPV_lowest_referenced_row(s, 0),
+                                                 0);
                     }
                     if (s->mv_dir & MV_DIR_BACKWARD) {
-                        ff_thread_await_progress((AVFrame*)s->next_picture_ptr, ff_MPV_lowest_referenced_row(s, 1), 0);
+                        ff_thread_await_progress(&s->next_picture_ptr->f,
+                                                 ff_MPV_lowest_referenced_row(s, 1),
+                                                 0);
                     }
                 }
 
@@ -2556,9 +2556,9 @@ void ff_draw_horiz_band(MpegEncContext *s, int y, int h){
         int i;
 
         if(s->pict_type==AV_PICTURE_TYPE_B || s->low_delay || (s->avctx->slice_flags&SLICE_FLAG_CODED_ORDER))
-            src= (AVFrame*)s->current_picture_ptr;
+            src = &s->current_picture_ptr->f;
         else if(s->last_picture_ptr)
-            src= (AVFrame*)s->last_picture_ptr;
+            src = &s->last_picture_ptr->f;
         else
             return;
 
@@ -2867,5 +2867,5 @@ void ff_set_qscale(MpegEncContext * s, int qscale)
 void ff_MPV_report_decode_progress(MpegEncContext *s)
 {
     if (s->pict_type != AV_PICTURE_TYPE_B && !s->partitioned_frame && !s->error_occurred)
-        ff_thread_report_progress((AVFrame*)s->current_picture_ptr, s->mb_y, 0);
+        ff_thread_report_progress(&s->current_picture_ptr->f, s->mb_y, 0);
 }
diff --git a/libavcodec/mpegvideo_enc.c b/libavcodec/mpegvideo_enc.c
index af57644626..9c7b889b97 100644
--- a/libavcodec/mpegvideo_enc.c
+++ b/libavcodec/mpegvideo_enc.c
@@ -981,7 +981,7 @@ static int load_input_picture(MpegEncContext *s, AVFrame *pic_arg)
         if (i < 0)
             return i;
 
-        pic = (AVFrame *) &s->picture[i];
+        pic = &s->picture[i].f;
         pic->reference = 3;
 
         for (i = 0; i < 4; i++) {
@@ -996,7 +996,7 @@ static int load_input_picture(MpegEncContext *s, AVFrame *pic_arg)
         if (i < 0)
             return i;
 
-        pic = (AVFrame *) &s->picture[i];
+        pic = &s->picture[i].f;
         pic->reference = 3;
 
         if (ff_alloc_picture(s, (Picture *) pic, 0) < 0) {
@@ -1252,7 +1252,7 @@ static int select_input_picture(MpegEncContext *s)
                                s->input_picture[0]->f.type == FF_BUFFER_TYPE_INTERNAL);
 
                         s->avctx->release_buffer(s->avctx,
-                                                 (AVFrame *) s->input_picture[0]);
+                                                 &s->input_picture[0]->f);
                     }
 
                     emms_c();
@@ -1385,13 +1385,13 @@ no_output_pic:
             /* mark us unused / free shared pic */
             if (s->reordered_input_picture[0]->f.type == FF_BUFFER_TYPE_INTERNAL)
                 s->avctx->release_buffer(s->avctx,
-                                         (AVFrame *) s->reordered_input_picture[0]);
+                                         &s->reordered_input_picture[0]->f);
             for (i = 0; i < 4; i++)
                 s->reordered_input_picture[0]->f.data[i] = NULL;
             s->reordered_input_picture[0]->f.type = 0;
 
-            copy_picture_attributes(s, (AVFrame *) pic,
-                                    (AVFrame *) s->reordered_input_picture[0]);
+            copy_picture_attributes(s, &pic->f,
+                                    &s->reordered_input_picture[0]->f);
 
             s->current_picture_ptr = pic;
         } else {
diff --git a/libavcodec/msmpeg4.c b/libavcodec/msmpeg4.c
index 232b82b219..b182da1f90 100644
--- a/libavcodec/msmpeg4.c
+++ b/libavcodec/msmpeg4.c
@@ -34,6 +34,7 @@
 #include "libavutil/x86_cpu.h"
 #include "h263.h"
 #include "mpeg4video.h"
+#include "msmpeg4data.h"
 #include "vc1data.h"
 
 /*
@@ -52,22 +53,8 @@
 #define V2_MV_VLC_BITS 9
 #define TEX_VLC_BITS 9
 
-#define II_BITRATE 128*1024
-#define MBAC_BITRATE 50*1024
-
 #define DEFAULT_INTER_INDEX 3
 
-static uint32_t v2_dc_lum_table[512][2];
-static uint32_t v2_dc_chroma_table[512][2];
-
-#include "msmpeg4data.h"
-
-#if CONFIG_ENCODERS //strangely gcc includes this even if it is not referenced
-static uint8_t rl_length[NB_RL_TABLES][MAX_LEVEL+1][MAX_RUN+1][2];
-#endif //CONFIG_ENCODERS
-
-static uint8_t static_rl_table_store[NB_RL_TABLES][2][2*MAX_RUN + MAX_LEVEL + 3];
-
 /* This table is practically identical to the one from h263
  * except that it is inverted. */
 static av_cold void init_h263_dc_for_msmpeg4(void)
@@ -102,8 +89,8 @@ static av_cold void init_h263_dc_for_msmpeg4(void)
                     uni_len++;
                 }
             }
-            v2_dc_lum_table[level+256][0]= uni_code;
-            v2_dc_lum_table[level+256][1]= uni_len;
+            ff_v2_dc_lum_table[level + 256][0] = uni_code;
+            ff_v2_dc_lum_table[level + 256][1] = uni_len;
 
             /* chrominance h263 */
             uni_code= ff_mpeg4_DCtab_chrom[size][0];
@@ -118,13 +105,13 @@ static av_cold void init_h263_dc_for_msmpeg4(void)
                     uni_len++;
                 }
             }
-            v2_dc_chroma_table[level+256][0]= uni_code;
-            v2_dc_chroma_table[level+256][1]= uni_len;
+            ff_v2_dc_chroma_table[level + 256][0] = uni_code;
+            ff_v2_dc_chroma_table[level + 256][1] = uni_len;
 
         }
 }
 
-static av_cold void common_init(MpegEncContext * s)
+av_cold void ff_msmpeg4_common_init(MpegEncContext *s)
 {
     static int initialized=0;
 
@@ -173,251 +160,6 @@ static av_cold void common_init(MpegEncContext * s)
     }
 }
 
-#if CONFIG_ENCODERS
-
-/* build the table which associate a (x,y) motion vector to a vlc */
-static void init_mv_table(MVTable *tab)
-{
-    int i, x, y;
-
-    tab->table_mv_index = av_malloc(sizeof(uint16_t) * 4096);
-    /* mark all entries as not used */
-    for(i=0;i<4096;i++)
-        tab->table_mv_index[i] = tab->n;
-
-    for(i=0;i<tab->n;i++) {
-        x = tab->table_mvx[i];
-        y = tab->table_mvy[i];
-        tab->table_mv_index[(x << 6) | y] = i;
-    }
-}
-
-void ff_msmpeg4_code012(PutBitContext *pb, int n)
-{
-    if (n == 0) {
-        put_bits(pb, 1, 0);
-    } else {
-        put_bits(pb, 1, 1);
-        put_bits(pb, 1, (n >= 2));
-    }
-}
-
-static int get_size_of_code(MpegEncContext * s, RLTable *rl, int last, int run, int level, int intra){
-    int size=0;
-    int code;
-    int run_diff= intra ? 0 : 1;
-
-    code = get_rl_index(rl, last, run, level);
-    size+= rl->table_vlc[code][1];
-    if (code == rl->n) {
-        int level1, run1;
-
-        level1 = level - rl->max_level[last][run];
-        if (level1 < 1)
-            goto esc2;
-        code = get_rl_index(rl, last, run, level1);
-        if (code == rl->n) {
-            esc2:
-            size++;
-            if (level > MAX_LEVEL)
-                goto esc3;
-            run1 = run - rl->max_run[last][level] - run_diff;
-            if (run1 < 0)
-                goto esc3;
-            code = get_rl_index(rl, last, run1, level);
-            if (code == rl->n) {
-            esc3:
-                /* third escape */
-                size+=1+1+6+8;
-            } else {
-                /* second escape */
-                size+= 1+1+ rl->table_vlc[code][1];
-            }
-        } else {
-            /* first escape */
-            size+= 1+1+ rl->table_vlc[code][1];
-        }
-    } else {
-        size++;
-    }
-    return size;
-}
-
-av_cold void ff_msmpeg4_encode_init(MpegEncContext *s)
-{
-    static int init_done=0;
-    int i;
-
-    common_init(s);
-    if(s->msmpeg4_version>=4){
-        s->min_qcoeff= -255;
-        s->max_qcoeff=  255;
-    }
-
-    if (!init_done) {
-        /* init various encoding tables */
-        init_done = 1;
-        init_mv_table(&ff_mv_tables[0]);
-        init_mv_table(&ff_mv_tables[1]);
-        for(i=0;i<NB_RL_TABLES;i++)
-            ff_init_rl(&ff_rl_table[i], static_rl_table_store[i]);
-
-        for(i=0; i<NB_RL_TABLES; i++){
-            int level;
-            for (level = 1; level <= MAX_LEVEL; level++) {
-                int run;
-                for(run=0; run<=MAX_RUN; run++){
-                    int last;
-                    for(last=0; last<2; last++){
-                        rl_length[i][level][run][last]= get_size_of_code(s, &ff_rl_table[  i], last, run, level, 0);
-                    }
-                }
-            }
-        }
-    }
-}
-
-static void find_best_tables(MpegEncContext * s)
-{
-    int i;
-    int best        = 0, best_size        = INT_MAX;
-    int chroma_best = 0, best_chroma_size = INT_MAX;
-
-    for(i=0; i<3; i++){
-        int level;
-        int chroma_size=0;
-        int size=0;
-
-        if(i>0){// ;)
-            size++;
-            chroma_size++;
-        }
-        for(level=0; level<=MAX_LEVEL; level++){
-            int run;
-            for(run=0; run<=MAX_RUN; run++){
-                int last;
-                const int last_size= size + chroma_size;
-                for(last=0; last<2; last++){
-                    int inter_count       = s->ac_stats[0][0][level][run][last] + s->ac_stats[0][1][level][run][last];
-                    int intra_luma_count  = s->ac_stats[1][0][level][run][last];
-                    int intra_chroma_count= s->ac_stats[1][1][level][run][last];
-
-                    if(s->pict_type==AV_PICTURE_TYPE_I){
-                        size       += intra_luma_count  *rl_length[i  ][level][run][last];
-                        chroma_size+= intra_chroma_count*rl_length[i+3][level][run][last];
-                    }else{
-                        size+=        intra_luma_count  *rl_length[i  ][level][run][last]
-                                     +intra_chroma_count*rl_length[i+3][level][run][last]
-                                     +inter_count       *rl_length[i+3][level][run][last];
-                    }
-                }
-                if(last_size == size+chroma_size) break;
-            }
-        }
-        if(size<best_size){
-            best_size= size;
-            best= i;
-        }
-        if(chroma_size<best_chroma_size){
-            best_chroma_size= chroma_size;
-            chroma_best= i;
-        }
-    }
-
-//    printf("type:%d, best:%d, qp:%d, var:%d, mcvar:%d, size:%d //\n",
-//           s->pict_type, best, s->qscale, s->mb_var_sum, s->mc_mb_var_sum, best_size);
-
-    if(s->pict_type==AV_PICTURE_TYPE_P) chroma_best= best;
-
-    memset(s->ac_stats, 0, sizeof(int)*(MAX_LEVEL+1)*(MAX_RUN+1)*2*2*2);
-
-    s->rl_table_index       =        best;
-    s->rl_chroma_table_index= chroma_best;
-
-    if(s->pict_type != s->last_non_b_pict_type){
-        s->rl_table_index= 2;
-        if(s->pict_type==AV_PICTURE_TYPE_I)
-            s->rl_chroma_table_index= 1;
-        else
-            s->rl_chroma_table_index= 2;
-    }
-
-}
-
-/* write MSMPEG4 compatible frame header */
-void ff_msmpeg4_encode_picture_header(MpegEncContext * s, int picture_number)
-{
-    find_best_tables(s);
-
-    avpriv_align_put_bits(&s->pb);
-    put_bits(&s->pb, 2, s->pict_type - 1);
-
-    put_bits(&s->pb, 5, s->qscale);
-    if(s->msmpeg4_version<=2){
-        s->rl_table_index = 2;
-        s->rl_chroma_table_index = 2;
-    }
-
-    s->dc_table_index = 1;
-    s->mv_table_index = 1; /* only if P frame */
-    s->use_skip_mb_code = 1; /* only if P frame */
-    s->per_mb_rl_table = 0;
-    if(s->msmpeg4_version==4)
-        s->inter_intra_pred= (s->width*s->height < 320*240 && s->bit_rate<=II_BITRATE && s->pict_type==AV_PICTURE_TYPE_P);
-//printf("%d %d %d %d %d\n", s->pict_type, s->bit_rate, s->inter_intra_pred, s->width, s->height);
-
-    if (s->pict_type == AV_PICTURE_TYPE_I) {
-        s->slice_height= s->mb_height/1;
-        put_bits(&s->pb, 5, 0x16 + s->mb_height/s->slice_height);
-
-        if(s->msmpeg4_version==4){
-            ff_msmpeg4_encode_ext_header(s);
-            if(s->bit_rate>MBAC_BITRATE)
-                put_bits(&s->pb, 1, s->per_mb_rl_table);
-        }
-
-        if(s->msmpeg4_version>2){
-            if(!s->per_mb_rl_table){
-                ff_msmpeg4_code012(&s->pb, s->rl_chroma_table_index);
-                ff_msmpeg4_code012(&s->pb, s->rl_table_index);
-            }
-
-            put_bits(&s->pb, 1, s->dc_table_index);
-        }
-    } else {
-        put_bits(&s->pb, 1, s->use_skip_mb_code);
-
-        if(s->msmpeg4_version==4 && s->bit_rate>MBAC_BITRATE)
-            put_bits(&s->pb, 1, s->per_mb_rl_table);
-
-        if(s->msmpeg4_version>2){
-            if(!s->per_mb_rl_table)
-                ff_msmpeg4_code012(&s->pb, s->rl_table_index);
-
-            put_bits(&s->pb, 1, s->dc_table_index);
-
-            put_bits(&s->pb, 1, s->mv_table_index);
-        }
-    }
-
-    s->esc3_level_length= 0;
-    s->esc3_run_length= 0;
-}
-
-void ff_msmpeg4_encode_ext_header(MpegEncContext * s)
-{
-        put_bits(&s->pb, 5, s->avctx->time_base.den / s->avctx->time_base.num); //yes 29.97 -> 29
-
-        put_bits(&s->pb, 11, FFMIN(s->bit_rate/1024, 2047));
-
-        if(s->msmpeg4_version>=3)
-            put_bits(&s->pb, 1, s->flipflop_rounding);
-        else
-            assert(s->flipflop_rounding==0);
-}
-
-#endif //CONFIG_ENCODERS
-
 /* predict coded block */
 int ff_msmpeg4_coded_block_pred(MpegEncContext * s, int n, uint8_t **coded_block_ptr)
 {
@@ -445,217 +187,6 @@ int ff_msmpeg4_coded_block_pred(MpegEncContext * s, int n, uint8_t **coded_block
     return pred;
 }
 
-#if CONFIG_ENCODERS
-
-void ff_msmpeg4_encode_motion(MpegEncContext * s,
-                                  int mx, int my)
-{
-    int code;
-    MVTable *mv;
-
-    /* modulo encoding */
-    /* WARNING : you cannot reach all the MVs even with the modulo
-       encoding. This is a somewhat strange compromise they took !!!  */
-    if (mx <= -64)
-        mx += 64;
-    else if (mx >= 64)
-        mx -= 64;
-    if (my <= -64)
-        my += 64;
-    else if (my >= 64)
-        my -= 64;
-
-    mx += 32;
-    my += 32;
-#if 0
-    if ((unsigned)mx >= 64 ||
-        (unsigned)my >= 64)
-        av_log(s->avctx, AV_LOG_ERROR, "error mx=%d my=%d\n", mx, my);
-#endif
-    mv = &ff_mv_tables[s->mv_table_index];
-
-    code = mv->table_mv_index[(mx << 6) | my];
-    put_bits(&s->pb,
-             mv->table_mv_bits[code],
-             mv->table_mv_code[code]);
-    if (code == mv->n) {
-        /* escape : code literally */
-        put_bits(&s->pb, 6, mx);
-        put_bits(&s->pb, 6, my);
-    }
-}
-
-void ff_msmpeg4_handle_slices(MpegEncContext *s){
-    if (s->mb_x == 0) {
-        if (s->slice_height && (s->mb_y % s->slice_height) == 0) {
-            if(s->msmpeg4_version < 4){
-                ff_mpeg4_clean_buffers(s);
-            }
-            s->first_slice_line = 1;
-        } else {
-            s->first_slice_line = 0;
-        }
-    }
-}
-
-static void msmpeg4v2_encode_motion(MpegEncContext * s, int val)
-{
-    int range, bit_size, sign, code, bits;
-
-    if (val == 0) {
-        /* zero vector */
-        code = 0;
-        put_bits(&s->pb, ff_mvtab[code][1], ff_mvtab[code][0]);
-    } else {
-        bit_size = s->f_code - 1;
-        range = 1 << bit_size;
-        if (val <= -64)
-            val += 64;
-        else if (val >= 64)
-            val -= 64;
-
-        if (val >= 0) {
-            sign = 0;
-        } else {
-            val = -val;
-            sign = 1;
-        }
-        val--;
-        code = (val >> bit_size) + 1;
-        bits = val & (range - 1);
-
-        put_bits(&s->pb, ff_mvtab[code][1] + 1, (ff_mvtab[code][0] << 1) | sign);
-        if (bit_size > 0) {
-            put_bits(&s->pb, bit_size, bits);
-        }
-    }
-}
-
-void ff_msmpeg4_encode_mb(MpegEncContext * s,
-                          DCTELEM block[6][64],
-                          int motion_x, int motion_y)
-{
-    int cbp, coded_cbp, i;
-    int pred_x, pred_y;
-    uint8_t *coded_block;
-
-    ff_msmpeg4_handle_slices(s);
-
-    if (!s->mb_intra) {
-        /* compute cbp */
-        cbp = 0;
-        for (i = 0; i < 6; i++) {
-            if (s->block_last_index[i] >= 0)
-                cbp |= 1 << (5 - i);
-        }
-        if (s->use_skip_mb_code && (cbp | motion_x | motion_y) == 0) {
-            /* skip macroblock */
-            put_bits(&s->pb, 1, 1);
-            s->last_bits++;
-            s->misc_bits++;
-            s->skip_count++;
-
-            return;
-        }
-        if (s->use_skip_mb_code)
-            put_bits(&s->pb, 1, 0);     /* mb coded */
-
-        if(s->msmpeg4_version<=2){
-            put_bits(&s->pb,
-                     ff_v2_mb_type[cbp&3][1],
-                     ff_v2_mb_type[cbp&3][0]);
-            if((cbp&3) != 3) coded_cbp= cbp ^ 0x3C;
-            else             coded_cbp= cbp;
-
-            put_bits(&s->pb,
-                     ff_h263_cbpy_tab[coded_cbp>>2][1],
-                     ff_h263_cbpy_tab[coded_cbp>>2][0]);
-
-            s->misc_bits += get_bits_diff(s);
-
-            ff_h263_pred_motion(s, 0, 0, &pred_x, &pred_y);
-            msmpeg4v2_encode_motion(s, motion_x - pred_x);
-            msmpeg4v2_encode_motion(s, motion_y - pred_y);
-        }else{
-            put_bits(&s->pb,
-                     ff_table_mb_non_intra[cbp + 64][1],
-                     ff_table_mb_non_intra[cbp + 64][0]);
-
-            s->misc_bits += get_bits_diff(s);
-
-            /* motion vector */
-            ff_h263_pred_motion(s, 0, 0, &pred_x, &pred_y);
-            ff_msmpeg4_encode_motion(s, motion_x - pred_x,
-                                  motion_y - pred_y);
-        }
-
-        s->mv_bits += get_bits_diff(s);
-
-        for (i = 0; i < 6; i++) {
-            ff_msmpeg4_encode_block(s, block[i], i);
-        }
-        s->p_tex_bits += get_bits_diff(s);
-    } else {
-        /* compute cbp */
-        cbp = 0;
-        coded_cbp = 0;
-        for (i = 0; i < 6; i++) {
-            int val, pred;
-            val = (s->block_last_index[i] >= 1);
-            cbp |= val << (5 - i);
-            if (i < 4) {
-                /* predict value for close blocks only for luma */
-                pred = ff_msmpeg4_coded_block_pred(s, i, &coded_block);
-                *coded_block = val;
-                val = val ^ pred;
-            }
-            coded_cbp |= val << (5 - i);
-        }
-
-        if(s->msmpeg4_version<=2){
-            if (s->pict_type == AV_PICTURE_TYPE_I) {
-                put_bits(&s->pb,
-                         ff_v2_intra_cbpc[cbp&3][1], ff_v2_intra_cbpc[cbp&3][0]);
-            } else {
-                if (s->use_skip_mb_code)
-                    put_bits(&s->pb, 1, 0);     /* mb coded */
-                put_bits(&s->pb,
-                         ff_v2_mb_type[(cbp&3) + 4][1],
-                         ff_v2_mb_type[(cbp&3) + 4][0]);
-            }
-            put_bits(&s->pb, 1, 0);             /* no AC prediction yet */
-            put_bits(&s->pb,
-                     ff_h263_cbpy_tab[cbp>>2][1],
-                     ff_h263_cbpy_tab[cbp>>2][0]);
-        }else{
-            if (s->pict_type == AV_PICTURE_TYPE_I) {
-                put_bits(&s->pb,
-                         ff_msmp4_mb_i_table[coded_cbp][1], ff_msmp4_mb_i_table[coded_cbp][0]);
-            } else {
-                if (s->use_skip_mb_code)
-                    put_bits(&s->pb, 1, 0);     /* mb coded */
-                put_bits(&s->pb,
-                         ff_table_mb_non_intra[cbp][1],
-                         ff_table_mb_non_intra[cbp][0]);
-            }
-            put_bits(&s->pb, 1, 0);             /* no AC prediction yet */
-            if(s->inter_intra_pred){
-                s->h263_aic_dir=0;
-                put_bits(&s->pb, ff_table_inter_intra[s->h263_aic_dir][1], ff_table_inter_intra[s->h263_aic_dir][0]);
-            }
-        }
-        s->misc_bits += get_bits_diff(s);
-
-        for (i = 0; i < 6; i++) {
-            ff_msmpeg4_encode_block(s, block[i], i);
-        }
-        s->i_tex_bits += get_bits_diff(s);
-        s->i_count++;
-    }
-}
-
-#endif //CONFIG_ENCODERS
-
 static inline int msmpeg4v1_pred_dc(MpegEncContext * s, int n,
                                     int32_t **dc_val_ptr)
 {
@@ -685,8 +216,8 @@ static int get_dc(uint8_t *src, int stride, int scale)
 }
 
 /* dir = 0: left, dir = 1: top prediction */
-static inline int msmpeg4_pred_dc(MpegEncContext * s, int n,
-                             int16_t **dc_val_ptr, int *dir_ptr)
+int ff_msmpeg4_pred_dc(MpegEncContext *s, int n,
+                       int16_t **dc_val_ptr, int *dir_ptr)
 {
     int a, b, c, wrap, pred, scale;
     int16_t *dc_val;
@@ -832,207 +363,6 @@ static inline int msmpeg4_pred_dc(MpegEncContext * s, int n,
     return pred;
 }
 
-#define DC_MAX 119
-
-static void msmpeg4_encode_dc(MpegEncContext * s, int level, int n, int *dir_ptr)
-{
-    int sign, code;
-    int pred, extquant;
-    int extrabits = 0;
-
-    int16_t *dc_val;
-    pred = msmpeg4_pred_dc(s, n, &dc_val, dir_ptr);
-
-    /* update predictor */
-    if (n < 4) {
-        *dc_val = level * s->y_dc_scale;
-    } else {
-        *dc_val = level * s->c_dc_scale;
-    }
-
-    /* do the prediction */
-    level -= pred;
-
-    if(s->msmpeg4_version<=2){
-        if (n < 4) {
-            put_bits(&s->pb,
-                     v2_dc_lum_table[level+256][1],
-                     v2_dc_lum_table[level+256][0]);
-        }else{
-            put_bits(&s->pb,
-                     v2_dc_chroma_table[level+256][1],
-                     v2_dc_chroma_table[level+256][0]);
-        }
-    }else{
-        sign = 0;
-        if (level < 0) {
-            level = -level;
-            sign = 1;
-        }
-        code = level;
-        if (code > DC_MAX)
-            code = DC_MAX;
-        else if( s->msmpeg4_version>=6 ) {
-            if( s->qscale == 1 ) {
-                extquant = (level + 3) & 0x3;
-                code  = ((level+3)>>2);
-            } else if( s->qscale == 2 ) {
-                extquant = (level + 1) & 0x1;
-                code  = ((level+1)>>1);
-            }
-        }
-
-        if (s->dc_table_index == 0) {
-            if (n < 4) {
-                put_bits(&s->pb, ff_table0_dc_lum[code][1], ff_table0_dc_lum[code][0]);
-            } else {
-                put_bits(&s->pb, ff_table0_dc_chroma[code][1], ff_table0_dc_chroma[code][0]);
-            }
-        } else {
-            if (n < 4) {
-                put_bits(&s->pb, ff_table1_dc_lum[code][1], ff_table1_dc_lum[code][0]);
-            } else {
-                put_bits(&s->pb, ff_table1_dc_chroma[code][1], ff_table1_dc_chroma[code][0]);
-            }
-        }
-
-        if(s->msmpeg4_version>=6 && s->qscale<=2)
-            extrabits = 3 - s->qscale;
-
-        if (code == DC_MAX)
-            put_bits(&s->pb, 8 + extrabits, level);
-        else if(extrabits > 0)//== VC1 && s->qscale<=2
-            put_bits(&s->pb, extrabits, extquant);
-
-        if (level != 0) {
-            put_bits(&s->pb, 1, sign);
-        }
-    }
-}
-
-/* Encoding of a block. Very similar to MPEG4 except for a different
-   escape coding (same as H263) and more vlc tables.
- */
-void ff_msmpeg4_encode_block(MpegEncContext * s, DCTELEM * block, int n)
-{
-    int level, run, last, i, j, last_index;
-    int last_non_zero, sign, slevel;
-    int code, run_diff, dc_pred_dir;
-    const RLTable *rl;
-    const uint8_t *scantable;
-
-    if (s->mb_intra) {
-        msmpeg4_encode_dc(s, block[0], n, &dc_pred_dir);
-        i = 1;
-        if (n < 4) {
-            rl = &ff_rl_table[s->rl_table_index];
-        } else {
-            rl = &ff_rl_table[3 + s->rl_chroma_table_index];
-        }
-        run_diff = s->msmpeg4_version>=4;
-        scantable= s->intra_scantable.permutated;
-    } else {
-        i = 0;
-        rl = &ff_rl_table[3 + s->rl_table_index];
-        if(s->msmpeg4_version<=2)
-            run_diff = 0;
-        else
-            run_diff = 1;
-        scantable= s->inter_scantable.permutated;
-    }
-
-    /* recalculate block_last_index for M$ wmv1 */
-    if(s->msmpeg4_version>=4 && s->msmpeg4_version<6 && s->block_last_index[n]>0){
-        for(last_index=63; last_index>=0; last_index--){
-            if(block[scantable[last_index]]) break;
-        }
-        s->block_last_index[n]= last_index;
-    }else
-        last_index = s->block_last_index[n];
-    /* AC coefs */
-    last_non_zero = i - 1;
-    for (; i <= last_index; i++) {
-        j = scantable[i];
-        level = block[j];
-        if (level) {
-            run = i - last_non_zero - 1;
-            last = (i == last_index);
-            sign = 0;
-            slevel = level;
-            if (level < 0) {
-                sign = 1;
-                level = -level;
-            }
-
-            if(level<=MAX_LEVEL && run<=MAX_RUN){
-                s->ac_stats[s->mb_intra][n>3][level][run][last]++;
-            }
-
-            s->ac_stats[s->mb_intra][n > 3][40][63][0]++; //esc3 like
-
-            code = get_rl_index(rl, last, run, level);
-            put_bits(&s->pb, rl->table_vlc[code][1], rl->table_vlc[code][0]);
-            if (code == rl->n) {
-                int level1, run1;
-
-                level1 = level - rl->max_level[last][run];
-                if (level1 < 1)
-                    goto esc2;
-                code = get_rl_index(rl, last, run, level1);
-                if (code == rl->n) {
-                esc2:
-                    put_bits(&s->pb, 1, 0);
-                    if (level > MAX_LEVEL)
-                        goto esc3;
-                    run1 = run - rl->max_run[last][level] - run_diff;
-                    if (run1 < 0)
-                        goto esc3;
-                    code = get_rl_index(rl, last, run1+1, level);
-                    if (s->msmpeg4_version == 4 && code == rl->n)
-                        goto esc3;
-                    code = get_rl_index(rl, last, run1, level);
-                    if (code == rl->n) {
-                    esc3:
-                        /* third escape */
-                        put_bits(&s->pb, 1, 0);
-                        put_bits(&s->pb, 1, last);
-                        if(s->msmpeg4_version>=4){
-                            if(s->esc3_level_length==0){
-                                s->esc3_level_length=8;
-                                s->esc3_run_length= 6;
-                                //ESCLVLSZ + ESCRUNSZ
-                                if(s->qscale<8)
-                                    put_bits(&s->pb, 6 + (s->msmpeg4_version>=6), 3);
-                                else
-                                    put_bits(&s->pb, 8, 3);
-                            }
-                            put_bits(&s->pb, s->esc3_run_length, run);
-                            put_bits(&s->pb, 1, sign);
-                            put_bits(&s->pb, s->esc3_level_length, level);
-                        }else{
-                            put_bits(&s->pb, 6, run);
-                            put_sbits(&s->pb, 8, slevel);
-                        }
-                    } else {
-                        /* second escape */
-                        put_bits(&s->pb, 1, 1);
-                        put_bits(&s->pb, rl->table_vlc[code][1], rl->table_vlc[code][0]);
-                        put_bits(&s->pb, 1, sign);
-                    }
-                } else {
-                    /* first escape */
-                    put_bits(&s->pb, 1, 1);
-                    put_bits(&s->pb, rl->table_vlc[code][1], rl->table_vlc[code][0]);
-                    put_bits(&s->pb, 1, sign);
-                }
-            } else {
-                put_bits(&s->pb, 1, sign);
-            }
-            last_non_zero = i;
-        }
-    }
-}
-
 /****************************************/
 /* decoding stuff */
 
@@ -1263,13 +593,13 @@ av_cold int ff_msmpeg4_decode_init(AVCodecContext *avctx)
     if (ff_h263_decode_init(avctx) < 0)
         return -1;
 
-    common_init(s);
+    ff_msmpeg4_common_init(s);
 
     if (!done) {
         done = 1;
 
         for(i=0;i<NB_RL_TABLES;i++) {
-            ff_init_rl(&ff_rl_table[i], static_rl_table_store[i]);
+            ff_init_rl(&ff_rl_table[i], ff_static_rl_table_store[i]);
         }
         INIT_VLC_RL(ff_rl_table[0], 642);
         INIT_VLC_RL(ff_rl_table[1], 1104);
@@ -1301,11 +631,11 @@ av_cold int ff_msmpeg4_decode_init(AVCodecContext *avctx)
                  &ff_table1_dc_chroma[0][0], 8, 4, 1216);
 
         INIT_VLC_STATIC(&v2_dc_lum_vlc, DC_VLC_BITS, 512,
-                 &v2_dc_lum_table[0][1], 8, 4,
-                 &v2_dc_lum_table[0][0], 8, 4, 1472);
+                 &ff_v2_dc_lum_table[0][1], 8, 4,
+                 &ff_v2_dc_lum_table[0][0], 8, 4, 1472);
         INIT_VLC_STATIC(&v2_dc_chroma_vlc, DC_VLC_BITS, 512,
-                 &v2_dc_chroma_table[0][1], 8, 4,
-                 &v2_dc_chroma_table[0][0], 8, 4, 1506);
+                 &ff_v2_dc_chroma_table[0][1], 8, 4,
+                 &ff_v2_dc_chroma_table[0][0], 8, 4, 1506);
 
         INIT_VLC_STATIC(&v2_intra_cbpc_vlc, V2_INTRA_CBPC_VLC_BITS, 4,
                  &ff_v2_intra_cbpc[0][1], 2, 1,
@@ -1588,7 +918,7 @@ static int msmpeg4_decode_dc(MpegEncContext * s, int n, int *dir_ptr)
         *dc_val= level;
     }else{
         int16_t *dc_val;
-        pred = msmpeg4_pred_dc(s, n, &dc_val, dir_ptr);
+        pred   = ff_msmpeg4_pred_dc(s, n, &dc_val, dir_ptr);
         level += pred;
 
         /* update predictor */
diff --git a/libavcodec/msmpeg4.h b/libavcodec/msmpeg4.h
index 463c72fa49..abc414cfdd 100644
--- a/libavcodec/msmpeg4.h
+++ b/libavcodec/msmpeg4.h
@@ -22,19 +22,29 @@
 #ifndef AVCODEC_MSMPEG4_H
 #define AVCODEC_MSMPEG4_H
 
+#include <stdint.h>
+
 #include "config.h"
 #include "avcodec.h"
 #include "dsputil.h"
 #include "mpegvideo.h"
+#include "msmpeg4data.h"
+#include "put_bits.h"
 
 #define INTER_INTRA_VLC_BITS 3
 #define MB_NON_INTRA_VLC_BITS 9
 #define MB_INTRA_VLC_BITS 9
 
+#define II_BITRATE 128*1024
+#define MBAC_BITRATE 50*1024
+
+#define DC_MAX 119
+
 extern VLC ff_mb_non_intra_vlc[4];
 extern VLC ff_inter_intra_vlc;
 
 void ff_msmpeg4_code012(PutBitContext *pb, int n);
+void ff_msmpeg4_common_init(MpegEncContext *s);
 void ff_msmpeg4_encode_block(MpegEncContext * s, DCTELEM * block, int n);
 void ff_msmpeg4_handle_slices(MpegEncContext *s);
 void ff_msmpeg4_encode_motion(MpegEncContext * s, int mx, int my);
@@ -43,6 +53,8 @@ int ff_msmpeg4_coded_block_pred(MpegEncContext * s, int n,
 int ff_msmpeg4_decode_motion(MpegEncContext * s, int *mx_ptr, int *my_ptr);
 int ff_msmpeg4_decode_block(MpegEncContext * s, DCTELEM * block,
                             int n, int coded, const uint8_t *scan_table);
+int ff_msmpeg4_pred_dc(MpegEncContext *s, int n,
+                       int16_t **dc_val_ptr, int *dir_ptr);
 int ff_wmv2_decode_mb(MpegEncContext *s, DCTELEM block[6][64]);
 
 #define CONFIG_MSMPEG4_DECODER (CONFIG_MSMPEG4V1_DECODER || \
diff --git a/libavcodec/msmpeg4data.c b/libavcodec/msmpeg4data.c
index 266ea2ec5d..50ba18c8cc 100644
--- a/libavcodec/msmpeg4data.c
+++ b/libavcodec/msmpeg4data.c
@@ -29,6 +29,10 @@
 
 #include "msmpeg4data.h"
 
+uint32_t ff_v2_dc_lum_table[512][2];
+uint32_t ff_v2_dc_chroma_table[512][2];
+uint8_t  ff_static_rl_table_store[NB_RL_TABLES][2][2 * MAX_RUN + MAX_LEVEL + 3];
+
 VLC ff_msmp4_mb_i_vlc;
 VLC ff_msmp4_dc_luma_vlc[2];
 VLC ff_msmp4_dc_chroma_vlc[2];
diff --git a/libavcodec/msmpeg4data.h b/libavcodec/msmpeg4data.h
index c32c09e241..24a10d9f2e 100644
--- a/libavcodec/msmpeg4data.h
+++ b/libavcodec/msmpeg4data.h
@@ -59,6 +59,10 @@ extern const uint8_t ff_wmv1_scantable[WMV1_SCANTABLE_COUNT][64];
 #define NB_RL_TABLES  6
 
 extern RLTable ff_rl_table[NB_RL_TABLES];
+extern uint8_t ff_static_rl_table_store[NB_RL_TABLES][2][2 * MAX_RUN + MAX_LEVEL + 3];
+
+extern uint32_t ff_v2_dc_lum_table[512][2];
+extern uint32_t ff_v2_dc_chroma_table[512][2];
 
 extern const uint8_t ff_wmv1_y_dc_scale_table[32];
 extern const uint8_t ff_wmv1_c_dc_scale_table[32];
diff --git a/libavcodec/msmpeg4enc.c b/libavcodec/msmpeg4enc.c
new file mode 100644
index 0000000000..527d2840e1
--- /dev/null
+++ b/libavcodec/msmpeg4enc.c
@@ -0,0 +1,692 @@
+/*
+ * MSMPEG4 encoder backend
+ * Copyright (c) 2001 Fabrice Bellard
+ * Copyright (c) 2002-2004 Michael Niedermayer <michaelni@gmx.at>
+ *
+ * msmpeg4v1 & v2 stuff by Michael Niedermayer <michaelni@gmx.at>
+ *
+ * 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
+ */
+
+/**
+ * @file
+ * MSMPEG4 encoder backend
+ */
+
+#include <stdint.h>
+#include <string.h>
+
+#include "libavutil/avutil.h"
+#include "libavutil/mem.h"
+#include "mpegvideo.h"
+#include "msmpeg4.h"
+#include "h263.h"
+#include "mpeg4video.h"
+#include "msmpeg4.h"
+#include "msmpeg4data.h"
+#include "put_bits.h"
+#include "rl.h"
+#include "vc1data.h"
+
+static uint8_t rl_length[NB_RL_TABLES][MAX_LEVEL+1][MAX_RUN+1][2];
+
+/* build the table which associate a (x,y) motion vector to a vlc */
+static void init_mv_table(MVTable *tab)
+{
+    int i, x, y;
+
+    tab->table_mv_index = av_malloc(sizeof(uint16_t) * 4096);
+    /* mark all entries as not used */
+    for(i=0;i<4096;i++)
+        tab->table_mv_index[i] = tab->n;
+
+    for(i=0;i<tab->n;i++) {
+        x = tab->table_mvx[i];
+        y = tab->table_mvy[i];
+        tab->table_mv_index[(x << 6) | y] = i;
+    }
+}
+
+void ff_msmpeg4_code012(PutBitContext *pb, int n)
+{
+    if (n == 0) {
+        put_bits(pb, 1, 0);
+    } else {
+        put_bits(pb, 1, 1);
+        put_bits(pb, 1, (n >= 2));
+    }
+}
+
+static int get_size_of_code(MpegEncContext * s, RLTable *rl, int last, int run, int level, int intra){
+    int size=0;
+    int code;
+    int run_diff= intra ? 0 : 1;
+
+    code = get_rl_index(rl, last, run, level);
+    size+= rl->table_vlc[code][1];
+    if (code == rl->n) {
+        int level1, run1;
+
+        level1 = level - rl->max_level[last][run];
+        if (level1 < 1)
+            goto esc2;
+        code = get_rl_index(rl, last, run, level1);
+        if (code == rl->n) {
+            esc2:
+            size++;
+            if (level > MAX_LEVEL)
+                goto esc3;
+            run1 = run - rl->max_run[last][level] - run_diff;
+            if (run1 < 0)
+                goto esc3;
+            code = get_rl_index(rl, last, run1, level);
+            if (code == rl->n) {
+            esc3:
+                /* third escape */
+                size+=1+1+6+8;
+            } else {
+                /* second escape */
+                size+= 1+1+ rl->table_vlc[code][1];
+            }
+        } else {
+            /* first escape */
+            size+= 1+1+ rl->table_vlc[code][1];
+        }
+    } else {
+        size++;
+    }
+    return size;
+}
+
+av_cold void ff_msmpeg4_encode_init(MpegEncContext *s)
+{
+    static int init_done=0;
+    int i;
+
+    ff_msmpeg4_common_init(s);
+    if(s->msmpeg4_version>=4){
+        s->min_qcoeff= -255;
+        s->max_qcoeff=  255;
+    }
+
+    if (!init_done) {
+        /* init various encoding tables */
+        init_done = 1;
+        init_mv_table(&ff_mv_tables[0]);
+        init_mv_table(&ff_mv_tables[1]);
+        for(i=0;i<NB_RL_TABLES;i++)
+            ff_init_rl(&ff_rl_table[i], ff_static_rl_table_store[i]);
+
+        for(i=0; i<NB_RL_TABLES; i++){
+            int level;
+            for (level = 1; level <= MAX_LEVEL; level++) {
+                int run;
+                for(run=0; run<=MAX_RUN; run++){
+                    int last;
+                    for(last=0; last<2; last++){
+                        rl_length[i][level][run][last]= get_size_of_code(s, &ff_rl_table[  i], last, run, level, 0);
+                    }
+                }
+            }
+        }
+    }
+}
+
+static void find_best_tables(MpegEncContext * s)
+{
+    int i;
+    int best        = 0, best_size        = INT_MAX;
+    int chroma_best = 0, best_chroma_size = INT_MAX;
+
+    for(i=0; i<3; i++){
+        int level;
+        int chroma_size=0;
+        int size=0;
+
+        if(i>0){// ;)
+            size++;
+            chroma_size++;
+        }
+        for(level=0; level<=MAX_LEVEL; level++){
+            int run;
+            for(run=0; run<=MAX_RUN; run++){
+                int last;
+                const int last_size= size + chroma_size;
+                for(last=0; last<2; last++){
+                    int inter_count       = s->ac_stats[0][0][level][run][last] + s->ac_stats[0][1][level][run][last];
+                    int intra_luma_count  = s->ac_stats[1][0][level][run][last];
+                    int intra_chroma_count= s->ac_stats[1][1][level][run][last];
+
+                    if(s->pict_type==AV_PICTURE_TYPE_I){
+                        size       += intra_luma_count  *rl_length[i  ][level][run][last];
+                        chroma_size+= intra_chroma_count*rl_length[i+3][level][run][last];
+                    }else{
+                        size+=        intra_luma_count  *rl_length[i  ][level][run][last]
+                                     +intra_chroma_count*rl_length[i+3][level][run][last]
+                                     +inter_count       *rl_length[i+3][level][run][last];
+                    }
+                }
+                if(last_size == size+chroma_size) break;
+            }
+        }
+        if(size<best_size){
+            best_size= size;
+            best= i;
+        }
+        if(chroma_size<best_chroma_size){
+            best_chroma_size= chroma_size;
+            chroma_best= i;
+        }
+    }
+
+//    printf("type:%d, best:%d, qp:%d, var:%d, mcvar:%d, size:%d //\n",
+//           s->pict_type, best, s->qscale, s->mb_var_sum, s->mc_mb_var_sum, best_size);
+
+    if(s->pict_type==AV_PICTURE_TYPE_P) chroma_best= best;
+
+    memset(s->ac_stats, 0, sizeof(int)*(MAX_LEVEL+1)*(MAX_RUN+1)*2*2*2);
+
+    s->rl_table_index       =        best;
+    s->rl_chroma_table_index= chroma_best;
+
+    if(s->pict_type != s->last_non_b_pict_type){
+        s->rl_table_index= 2;
+        if(s->pict_type==AV_PICTURE_TYPE_I)
+            s->rl_chroma_table_index= 1;
+        else
+            s->rl_chroma_table_index= 2;
+    }
+
+}
+
+/* write MSMPEG4 compatible frame header */
+void ff_msmpeg4_encode_picture_header(MpegEncContext * s, int picture_number)
+{
+    find_best_tables(s);
+
+    avpriv_align_put_bits(&s->pb);
+    put_bits(&s->pb, 2, s->pict_type - 1);
+
+    put_bits(&s->pb, 5, s->qscale);
+    if(s->msmpeg4_version<=2){
+        s->rl_table_index = 2;
+        s->rl_chroma_table_index = 2;
+    }
+
+    s->dc_table_index = 1;
+    s->mv_table_index = 1; /* only if P frame */
+    s->use_skip_mb_code = 1; /* only if P frame */
+    s->per_mb_rl_table = 0;
+    if(s->msmpeg4_version==4)
+        s->inter_intra_pred= (s->width*s->height < 320*240 && s->bit_rate<=II_BITRATE && s->pict_type==AV_PICTURE_TYPE_P);
+//printf("%d %d %d %d %d\n", s->pict_type, s->bit_rate, s->inter_intra_pred, s->width, s->height);
+
+    if (s->pict_type == AV_PICTURE_TYPE_I) {
+        s->slice_height= s->mb_height/1;
+        put_bits(&s->pb, 5, 0x16 + s->mb_height/s->slice_height);
+
+        if(s->msmpeg4_version==4){
+            ff_msmpeg4_encode_ext_header(s);
+            if(s->bit_rate>MBAC_BITRATE)
+                put_bits(&s->pb, 1, s->per_mb_rl_table);
+        }
+
+        if(s->msmpeg4_version>2){
+            if(!s->per_mb_rl_table){
+                ff_msmpeg4_code012(&s->pb, s->rl_chroma_table_index);
+                ff_msmpeg4_code012(&s->pb, s->rl_table_index);
+            }
+
+            put_bits(&s->pb, 1, s->dc_table_index);
+        }
+    } else {
+        put_bits(&s->pb, 1, s->use_skip_mb_code);
+
+        if(s->msmpeg4_version==4 && s->bit_rate>MBAC_BITRATE)
+            put_bits(&s->pb, 1, s->per_mb_rl_table);
+
+        if(s->msmpeg4_version>2){
+            if(!s->per_mb_rl_table)
+                ff_msmpeg4_code012(&s->pb, s->rl_table_index);
+
+            put_bits(&s->pb, 1, s->dc_table_index);
+
+            put_bits(&s->pb, 1, s->mv_table_index);
+        }
+    }
+
+    s->esc3_level_length= 0;
+    s->esc3_run_length= 0;
+}
+
+void ff_msmpeg4_encode_ext_header(MpegEncContext * s)
+{
+        put_bits(&s->pb, 5, s->avctx->time_base.den / s->avctx->time_base.num); //yes 29.97 -> 29
+
+        put_bits(&s->pb, 11, FFMIN(s->bit_rate/1024, 2047));
+
+        if(s->msmpeg4_version>=3)
+            put_bits(&s->pb, 1, s->flipflop_rounding);
+        else
+            assert(s->flipflop_rounding==0);
+}
+
+void ff_msmpeg4_encode_motion(MpegEncContext * s,
+                                  int mx, int my)
+{
+    int code;
+    MVTable *mv;
+
+    /* modulo encoding */
+    /* WARNING : you cannot reach all the MVs even with the modulo
+       encoding. This is a somewhat strange compromise they took !!!  */
+    if (mx <= -64)
+        mx += 64;
+    else if (mx >= 64)
+        mx -= 64;
+    if (my <= -64)
+        my += 64;
+    else if (my >= 64)
+        my -= 64;
+
+    mx += 32;
+    my += 32;
+#if 0
+    if ((unsigned)mx >= 64 ||
+        (unsigned)my >= 64)
+        av_log(s->avctx, AV_LOG_ERROR, "error mx=%d my=%d\n", mx, my);
+#endif
+    mv = &ff_mv_tables[s->mv_table_index];
+
+    code = mv->table_mv_index[(mx << 6) | my];
+    put_bits(&s->pb,
+             mv->table_mv_bits[code],
+             mv->table_mv_code[code]);
+    if (code == mv->n) {
+        /* escape : code literally */
+        put_bits(&s->pb, 6, mx);
+        put_bits(&s->pb, 6, my);
+    }
+}
+
+void ff_msmpeg4_handle_slices(MpegEncContext *s){
+    if (s->mb_x == 0) {
+        if (s->slice_height && (s->mb_y % s->slice_height) == 0) {
+            if(s->msmpeg4_version < 4){
+                ff_mpeg4_clean_buffers(s);
+            }
+            s->first_slice_line = 1;
+        } else {
+            s->first_slice_line = 0;
+        }
+    }
+}
+
+static void msmpeg4v2_encode_motion(MpegEncContext * s, int val)
+{
+    int range, bit_size, sign, code, bits;
+
+    if (val == 0) {
+        /* zero vector */
+        code = 0;
+        put_bits(&s->pb, ff_mvtab[code][1], ff_mvtab[code][0]);
+    } else {
+        bit_size = s->f_code - 1;
+        range = 1 << bit_size;
+        if (val <= -64)
+            val += 64;
+        else if (val >= 64)
+            val -= 64;
+
+        if (val >= 0) {
+            sign = 0;
+        } else {
+            val = -val;
+            sign = 1;
+        }
+        val--;
+        code = (val >> bit_size) + 1;
+        bits = val & (range - 1);
+
+        put_bits(&s->pb, ff_mvtab[code][1] + 1, (ff_mvtab[code][0] << 1) | sign);
+        if (bit_size > 0) {
+            put_bits(&s->pb, bit_size, bits);
+        }
+    }
+}
+
+void ff_msmpeg4_encode_mb(MpegEncContext * s,
+                          DCTELEM block[6][64],
+                          int motion_x, int motion_y)
+{
+    int cbp, coded_cbp, i;
+    int pred_x, pred_y;
+    uint8_t *coded_block;
+
+    ff_msmpeg4_handle_slices(s);
+
+    if (!s->mb_intra) {
+        /* compute cbp */
+        cbp = 0;
+        for (i = 0; i < 6; i++) {
+            if (s->block_last_index[i] >= 0)
+                cbp |= 1 << (5 - i);
+        }
+        if (s->use_skip_mb_code && (cbp | motion_x | motion_y) == 0) {
+            /* skip macroblock */
+            put_bits(&s->pb, 1, 1);
+            s->last_bits++;
+            s->misc_bits++;
+            s->skip_count++;
+
+            return;
+        }
+        if (s->use_skip_mb_code)
+            put_bits(&s->pb, 1, 0);     /* mb coded */
+
+        if(s->msmpeg4_version<=2){
+            put_bits(&s->pb,
+                     ff_v2_mb_type[cbp&3][1],
+                     ff_v2_mb_type[cbp&3][0]);
+            if((cbp&3) != 3) coded_cbp= cbp ^ 0x3C;
+            else             coded_cbp= cbp;
+
+            put_bits(&s->pb,
+                     ff_h263_cbpy_tab[coded_cbp>>2][1],
+                     ff_h263_cbpy_tab[coded_cbp>>2][0]);
+
+            s->misc_bits += get_bits_diff(s);
+
+            ff_h263_pred_motion(s, 0, 0, &pred_x, &pred_y);
+            msmpeg4v2_encode_motion(s, motion_x - pred_x);
+            msmpeg4v2_encode_motion(s, motion_y - pred_y);
+        }else{
+            put_bits(&s->pb,
+                     ff_table_mb_non_intra[cbp + 64][1],
+                     ff_table_mb_non_intra[cbp + 64][0]);
+
+            s->misc_bits += get_bits_diff(s);
+
+            /* motion vector */
+            ff_h263_pred_motion(s, 0, 0, &pred_x, &pred_y);
+            ff_msmpeg4_encode_motion(s, motion_x - pred_x,
+                                  motion_y - pred_y);
+        }
+
+        s->mv_bits += get_bits_diff(s);
+
+        for (i = 0; i < 6; i++) {
+            ff_msmpeg4_encode_block(s, block[i], i);
+        }
+        s->p_tex_bits += get_bits_diff(s);
+    } else {
+        /* compute cbp */
+        cbp = 0;
+        coded_cbp = 0;
+        for (i = 0; i < 6; i++) {
+            int val, pred;
+            val = (s->block_last_index[i] >= 1);
+            cbp |= val << (5 - i);
+            if (i < 4) {
+                /* predict value for close blocks only for luma */
+                pred = ff_msmpeg4_coded_block_pred(s, i, &coded_block);
+                *coded_block = val;
+                val = val ^ pred;
+            }
+            coded_cbp |= val << (5 - i);
+        }
+
+        if(s->msmpeg4_version<=2){
+            if (s->pict_type == AV_PICTURE_TYPE_I) {
+                put_bits(&s->pb,
+                         ff_v2_intra_cbpc[cbp&3][1], ff_v2_intra_cbpc[cbp&3][0]);
+            } else {
+                if (s->use_skip_mb_code)
+                    put_bits(&s->pb, 1, 0);     /* mb coded */
+                put_bits(&s->pb,
+                         ff_v2_mb_type[(cbp&3) + 4][1],
+                         ff_v2_mb_type[(cbp&3) + 4][0]);
+            }
+            put_bits(&s->pb, 1, 0);             /* no AC prediction yet */
+            put_bits(&s->pb,
+                     ff_h263_cbpy_tab[cbp>>2][1],
+                     ff_h263_cbpy_tab[cbp>>2][0]);
+        }else{
+            if (s->pict_type == AV_PICTURE_TYPE_I) {
+                put_bits(&s->pb,
+                         ff_msmp4_mb_i_table[coded_cbp][1], ff_msmp4_mb_i_table[coded_cbp][0]);
+            } else {
+                if (s->use_skip_mb_code)
+                    put_bits(&s->pb, 1, 0);     /* mb coded */
+                put_bits(&s->pb,
+                         ff_table_mb_non_intra[cbp][1],
+                         ff_table_mb_non_intra[cbp][0]);
+            }
+            put_bits(&s->pb, 1, 0);             /* no AC prediction yet */
+            if(s->inter_intra_pred){
+                s->h263_aic_dir=0;
+                put_bits(&s->pb, ff_table_inter_intra[s->h263_aic_dir][1], ff_table_inter_intra[s->h263_aic_dir][0]);
+            }
+        }
+        s->misc_bits += get_bits_diff(s);
+
+        for (i = 0; i < 6; i++) {
+            ff_msmpeg4_encode_block(s, block[i], i);
+        }
+        s->i_tex_bits += get_bits_diff(s);
+        s->i_count++;
+    }
+}
+
+static void msmpeg4_encode_dc(MpegEncContext * s, int level, int n, int *dir_ptr)
+{
+    int sign, code;
+    int pred, extquant;
+    int extrabits = 0;
+
+    int16_t *dc_val;
+    pred = ff_msmpeg4_pred_dc(s, n, &dc_val, dir_ptr);
+
+    /* update predictor */
+    if (n < 4) {
+        *dc_val = level * s->y_dc_scale;
+    } else {
+        *dc_val = level * s->c_dc_scale;
+    }
+
+    /* do the prediction */
+    level -= pred;
+
+    if(s->msmpeg4_version<=2){
+        if (n < 4) {
+            put_bits(&s->pb,
+                     ff_v2_dc_lum_table[level + 256][1],
+                     ff_v2_dc_lum_table[level + 256][0]);
+        }else{
+            put_bits(&s->pb,
+                     ff_v2_dc_chroma_table[level + 256][1],
+                     ff_v2_dc_chroma_table[level + 256][0]);
+        }
+    }else{
+        sign = 0;
+        if (level < 0) {
+            level = -level;
+            sign = 1;
+        }
+        code = level;
+        if (code > DC_MAX)
+            code = DC_MAX;
+        else if( s->msmpeg4_version>=6 ) {
+            if( s->qscale == 1 ) {
+                extquant = (level + 3) & 0x3;
+                code  = ((level+3)>>2);
+            } else if( s->qscale == 2 ) {
+                extquant = (level + 1) & 0x1;
+                code  = ((level+1)>>1);
+            }
+        }
+
+        if (s->dc_table_index == 0) {
+            if (n < 4) {
+                put_bits(&s->pb, ff_table0_dc_lum[code][1], ff_table0_dc_lum[code][0]);
+            } else {
+                put_bits(&s->pb, ff_table0_dc_chroma[code][1], ff_table0_dc_chroma[code][0]);
+            }
+        } else {
+            if (n < 4) {
+                put_bits(&s->pb, ff_table1_dc_lum[code][1], ff_table1_dc_lum[code][0]);
+            } else {
+                put_bits(&s->pb, ff_table1_dc_chroma[code][1], ff_table1_dc_chroma[code][0]);
+            }
+        }
+
+        if(s->msmpeg4_version>=6 && s->qscale<=2)
+            extrabits = 3 - s->qscale;
+
+        if (code == DC_MAX)
+            put_bits(&s->pb, 8 + extrabits, level);
+        else if(extrabits > 0)//== VC1 && s->qscale<=2
+            put_bits(&s->pb, extrabits, extquant);
+
+        if (level != 0) {
+            put_bits(&s->pb, 1, sign);
+        }
+    }
+}
+
+/* Encoding of a block. Very similar to MPEG4 except for a different
+   escape coding (same as H263) and more vlc tables.
+ */
+void ff_msmpeg4_encode_block(MpegEncContext * s, DCTELEM * block, int n)
+{
+    int level, run, last, i, j, last_index;
+    int last_non_zero, sign, slevel;
+    int code, run_diff, dc_pred_dir;
+    const RLTable *rl;
+    const uint8_t *scantable;
+
+    if (s->mb_intra) {
+        msmpeg4_encode_dc(s, block[0], n, &dc_pred_dir);
+        i = 1;
+        if (n < 4) {
+            rl = &ff_rl_table[s->rl_table_index];
+        } else {
+            rl = &ff_rl_table[3 + s->rl_chroma_table_index];
+        }
+        run_diff = s->msmpeg4_version>=4;
+        scantable= s->intra_scantable.permutated;
+    } else {
+        i = 0;
+        rl = &ff_rl_table[3 + s->rl_table_index];
+        if(s->msmpeg4_version<=2)
+            run_diff = 0;
+        else
+            run_diff = 1;
+        scantable= s->inter_scantable.permutated;
+    }
+
+    /* recalculate block_last_index for M$ wmv1 */
+    if(s->msmpeg4_version>=4 && s->msmpeg4_version<6 && s->block_last_index[n]>0){
+        for(last_index=63; last_index>=0; last_index--){
+            if(block[scantable[last_index]]) break;
+        }
+        s->block_last_index[n]= last_index;
+    }else
+        last_index = s->block_last_index[n];
+    /* AC coefs */
+    last_non_zero = i - 1;
+    for (; i <= last_index; i++) {
+        j = scantable[i];
+        level = block[j];
+        if (level) {
+            run = i - last_non_zero - 1;
+            last = (i == last_index);
+            sign = 0;
+            slevel = level;
+            if (level < 0) {
+                sign = 1;
+                level = -level;
+            }
+
+            if(level<=MAX_LEVEL && run<=MAX_RUN){
+                s->ac_stats[s->mb_intra][n>3][level][run][last]++;
+            }
+
+            s->ac_stats[s->mb_intra][n > 3][40][63][0]++; //esc3 like
+
+            code = get_rl_index(rl, last, run, level);
+            put_bits(&s->pb, rl->table_vlc[code][1], rl->table_vlc[code][0]);
+            if (code == rl->n) {
+                int level1, run1;
+
+                level1 = level - rl->max_level[last][run];
+                if (level1 < 1)
+                    goto esc2;
+                code = get_rl_index(rl, last, run, level1);
+                if (code == rl->n) {
+                esc2:
+                    put_bits(&s->pb, 1, 0);
+                    if (level > MAX_LEVEL)
+                        goto esc3;
+                    run1 = run - rl->max_run[last][level] - run_diff;
+                    if (run1 < 0)
+                        goto esc3;
+                    code = get_rl_index(rl, last, run1+1, level);
+                    if (s->msmpeg4_version == 4 && code == rl->n)
+                        goto esc3;
+                    code = get_rl_index(rl, last, run1, level);
+                    if (code == rl->n) {
+                    esc3:
+                        /* third escape */
+                        put_bits(&s->pb, 1, 0);
+                        put_bits(&s->pb, 1, last);
+                        if(s->msmpeg4_version>=4){
+                            if(s->esc3_level_length==0){
+                                s->esc3_level_length=8;
+                                s->esc3_run_length= 6;
+                                //ESCLVLSZ + ESCRUNSZ
+                                if(s->qscale<8)
+                                    put_bits(&s->pb, 6 + (s->msmpeg4_version>=6), 3);
+                                else
+                                    put_bits(&s->pb, 8, 3);
+                            }
+                            put_bits(&s->pb, s->esc3_run_length, run);
+                            put_bits(&s->pb, 1, sign);
+                            put_bits(&s->pb, s->esc3_level_length, level);
+                        }else{
+                            put_bits(&s->pb, 6, run);
+                            put_sbits(&s->pb, 8, slevel);
+                        }
+                    } else {
+                        /* second escape */
+                        put_bits(&s->pb, 1, 1);
+                        put_bits(&s->pb, rl->table_vlc[code][1], rl->table_vlc[code][0]);
+                        put_bits(&s->pb, 1, sign);
+                    }
+                } else {
+                    /* first escape */
+                    put_bits(&s->pb, 1, 1);
+                    put_bits(&s->pb, rl->table_vlc[code][1], rl->table_vlc[code][0]);
+                    put_bits(&s->pb, 1, sign);
+                }
+            } else {
+                put_bits(&s->pb, 1, sign);
+            }
+            last_non_zero = i;
+        }
+    }
+}
diff --git a/libavcodec/pamenc.c b/libavcodec/pamenc.c
index d1de7a0e33..7a3499e579 100644
--- a/libavcodec/pamenc.c
+++ b/libavcodec/pamenc.c
@@ -28,7 +28,7 @@ static int pam_encode_frame(AVCodecContext *avctx, AVPacket *pkt,
                             const AVFrame *pict, int *got_packet)
 {
     PNMContext *s     = avctx->priv_data;
-    AVFrame * const p = (AVFrame*)&s->picture;
+    AVFrame * const p = &s->picture;
     int i, h, w, n, linesize, depth, maxval, ret;
     const char *tuple_type;
     uint8_t *ptr;
diff --git a/libavcodec/pnm.c b/libavcodec/pnm.c
index 0e547deb5f..bfb4a25b7c 100644
--- a/libavcodec/pnm.c
+++ b/libavcodec/pnm.c
@@ -198,8 +198,8 @@ av_cold int ff_pnm_init(AVCodecContext *avctx)
 {
     PNMContext *s = avctx->priv_data;
 
-    avcodec_get_frame_defaults((AVFrame*)&s->picture);
-    avctx->coded_frame = (AVFrame*)&s->picture;
+    avcodec_get_frame_defaults(&s->picture);
+    avctx->coded_frame = &s->picture;
 
     return 0;
 }
diff --git a/libavcodec/pnmdec.c b/libavcodec/pnmdec.c
index 0be7ec9326..0c1fad868b 100644
--- a/libavcodec/pnmdec.c
+++ b/libavcodec/pnmdec.c
@@ -31,7 +31,7 @@ static int pnm_decode_frame(AVCodecContext *avctx, void *data,
     int buf_size         = avpkt->size;
     PNMContext * const s = avctx->priv_data;
     AVFrame *picture     = data;
-    AVFrame * const p    = (AVFrame*)&s->picture;
+    AVFrame * const p    = &s->picture;
     int i, j, n, linesize, h, upgrade = 0, is_mono = 0;
     unsigned char *ptr;
     int components, sample_len;
@@ -184,7 +184,7 @@ static int pnm_decode_frame(AVCodecContext *avctx, void *data,
         }
         break;
     }
-    *picture   = *(AVFrame*)&s->picture;
+    *picture   = s->picture;
     *data_size = sizeof(AVPicture);
 
     return s->bytestream - s->bytestream_start;
diff --git a/libavcodec/pnmenc.c b/libavcodec/pnmenc.c
index 1f96db51b9..b1f27b4a3e 100644
--- a/libavcodec/pnmenc.c
+++ b/libavcodec/pnmenc.c
@@ -28,7 +28,7 @@ static int pnm_encode_frame(AVCodecContext *avctx, AVPacket *pkt,
                             const AVFrame *pict, int *got_packet)
 {
     PNMContext *s     = avctx->priv_data;
-    AVFrame * const p = (AVFrame*)&s->picture;
+    AVFrame * const p = &s->picture;
     int i, h, h1, c, n, linesize, ret;
     uint8_t *ptr, *ptr1, *ptr2;
 
diff --git a/libavcodec/ppc/vp8dsp_altivec.c b/libavcodec/ppc/vp8dsp_altivec.c
index e9defbc67f..7f7d3bd35c 100644
--- a/libavcodec/ppc/vp8dsp_altivec.c
+++ b/libavcodec/ppc/vp8dsp_altivec.c
@@ -75,8 +75,8 @@ static const vec_s8 h_subpel_filters_outer[3] =
     dstv = vec_sra(dstv, c7)
 
 static av_always_inline
-void put_vp8_epel_h_altivec_core(uint8_t *dst, int dst_stride,
-                                 uint8_t *src, int src_stride,
+void put_vp8_epel_h_altivec_core(uint8_t *dst, ptrdiff_t dst_stride,
+                                 uint8_t *src, ptrdiff_t src_stride,
                                  int h, int mx, int w, int is6tap)
 {
     LOAD_H_SUBPEL_FILTER(mx-1);
@@ -161,8 +161,8 @@ static const vec_u8 v_subpel_filters[7] =
     dstv = vec_sra(dstv, c7)
 
 static av_always_inline
-void put_vp8_epel_v_altivec_core(uint8_t *dst, int dst_stride,
-                                 uint8_t *src, int src_stride,
+void put_vp8_epel_v_altivec_core(uint8_t *dst, ptrdiff_t dst_stride,
+                                 uint8_t *src, ptrdiff_t src_stride,
                                  int h, int my, int w, int is6tap)
 {
     LOAD_V_SUBPEL_FILTER(my-1);
@@ -226,19 +226,19 @@ void put_vp8_epel_v_altivec_core(uint8_t *dst, int dst_stride,
 
 #define EPEL_FUNCS(WIDTH, TAPS) \
 static av_noinline \
-void put_vp8_epel ## WIDTH ## _h ## TAPS ## _altivec(uint8_t *dst, int dst_stride, uint8_t *src, int src_stride, int h, int mx, int my) \
+void put_vp8_epel ## WIDTH ## _h ## TAPS ## _altivec(uint8_t *dst, ptrdiff_t dst_stride, uint8_t *src, ptrdiff_t src_stride, int h, int mx, int my) \
 { \
     put_vp8_epel_h_altivec_core(dst, dst_stride, src, src_stride, h, mx, WIDTH, TAPS == 6); \
 } \
 \
 static av_noinline \
-void put_vp8_epel ## WIDTH ## _v ## TAPS ## _altivec(uint8_t *dst, int dst_stride, uint8_t *src, int src_stride, int h, int mx, int my) \
+void put_vp8_epel ## WIDTH ## _v ## TAPS ## _altivec(uint8_t *dst, ptrdiff_t dst_stride, uint8_t *src, ptrdiff_t src_stride, int h, int mx, int my) \
 { \
     put_vp8_epel_v_altivec_core(dst, dst_stride, src, src_stride, h, my, WIDTH, TAPS == 6); \
 }
 
 #define EPEL_HV(WIDTH, HTAPS, VTAPS) \
-static void put_vp8_epel ## WIDTH ## _h ## HTAPS ## v ## VTAPS ## _altivec(uint8_t *dst, int stride, uint8_t *src, int s, int h, int mx, int my) \
+static void put_vp8_epel ## WIDTH ## _h ## HTAPS ## v ## VTAPS ## _altivec(uint8_t *dst, ptrdiff_t stride, uint8_t *src, ptrdiff_t s, int h, int mx, int my) \
 { \
     DECLARE_ALIGNED(16, uint8_t, tmp)[(2*WIDTH+5)*16]; \
     if (VTAPS == 6) { \
@@ -266,7 +266,7 @@ EPEL_HV(4,  4,6)
 EPEL_HV(4,  6,4)
 EPEL_HV(4,  4,4)
 
-static void put_vp8_pixels16_altivec(uint8_t *dst, int stride, uint8_t *src, int s, int h, int mx, int my)
+static void put_vp8_pixels16_altivec(uint8_t *dst, ptrdiff_t stride, uint8_t *src, ptrdiff_t s, int h, int mx, int my)
 {
     ff_put_pixels16_altivec(dst, src, stride, h);
 }
diff --git a/libavcodec/qdrw.c b/libavcodec/qdrw.c
index 54deec5c83..751e1a1909 100644
--- a/libavcodec/qdrw.c
+++ b/libavcodec/qdrw.c
@@ -40,7 +40,7 @@ static int decode_frame(AVCodecContext *avctx,
     const uint8_t *buf_end = avpkt->data + avpkt->size;
     int buf_size = avpkt->size;
     QdrawContext * const a = avctx->priv_data;
-    AVFrame * const p= (AVFrame*)&a->pic;
+    AVFrame * const p = &a->pic;
     uint8_t* outdata;
     int colors;
     int i;
diff --git a/libavcodec/qpeg.c b/libavcodec/qpeg.c
index bbb9f71aae..ca48b6bcf3 100644
--- a/libavcodec/qpeg.c
+++ b/libavcodec/qpeg.c
@@ -259,8 +259,8 @@ static int decode_frame(AVCodecContext *avctx,
     const uint8_t *buf = avpkt->data;
     int buf_size = avpkt->size;
     QpegContext * const a = avctx->priv_data;
-    AVFrame * p= (AVFrame*)&a->pic;
-    AVFrame * ref= (AVFrame*)&a->ref;
+    AVFrame *  p = &a->pic;
+    AVFrame * ref= &a->ref;
     uint8_t* outdata;
     int delta, ret = 0;
     const uint8_t *pal = av_packet_get_side_data(avpkt, AV_PKT_DATA_PALETTE, NULL);
@@ -311,8 +311,8 @@ static av_cold int decode_init(AVCodecContext *avctx){
 
 static av_cold int decode_end(AVCodecContext *avctx){
     QpegContext * const a = avctx->priv_data;
-    AVFrame * const p= (AVFrame*)&a->pic;
-    AVFrame * const ref= (AVFrame*)&a->ref;
+    AVFrame * const p = &a->pic;
+    AVFrame * const ref= &a->ref;
 
     if(p->data[0])
         avctx->release_buffer(avctx, p);
diff --git a/libavcodec/rawdec.c b/libavcodec/rawdec.c
index d912ca285d..71ab5df501 100644
--- a/libavcodec/rawdec.c
+++ b/libavcodec/rawdec.c
@@ -140,8 +140,8 @@ static int raw_decode(AVCodecContext *avctx,
     int linesize_align = 4;
     RawVideoContext *context = avctx->priv_data;
 
-    AVFrame * frame = (AVFrame *) data;
-    AVPicture * picture = (AVPicture *) data;
+    AVFrame   *frame   = data;
+    AVPicture *picture = data;
 
     frame->pict_type        = avctx->coded_frame->pict_type;
     frame->interlaced_frame = avctx->coded_frame->interlaced_frame;
diff --git a/libavcodec/rawenc.c b/libavcodec/rawenc.c
index 577a8fc98a..cb491e9094 100644
--- a/libavcodec/rawenc.c
+++ b/libavcodec/rawenc.c
@@ -32,7 +32,7 @@
 
 static av_cold int raw_init_encoder(AVCodecContext *avctx)
 {
-    avctx->coded_frame = (AVFrame *)avctx->priv_data;
+    avctx->coded_frame            = avctx->priv_data;
     avcodec_get_frame_defaults(avctx->coded_frame);
     avctx->coded_frame->pict_type = AV_PICTURE_TYPE_I;
     avctx->bits_per_coded_sample = av_get_bits_per_pixel(&av_pix_fmt_descriptors[avctx->pix_fmt]);
diff --git a/libavcodec/rv10.c b/libavcodec/rv10.c
index dd8715b300..bc86b69a2e 100644
--- a/libavcodec/rv10.c
+++ b/libavcodec/rv10.c
@@ -515,9 +515,10 @@ static int rv10_decode_packet(AVCodecContext *avctx,
                              const uint8_t *buf, int buf_size, int buf_size2)
 {
     MpegEncContext *s = avctx->priv_data;
-    int mb_count, mb_pos, left, start_mb_x;
+    int mb_count, mb_pos, left, start_mb_x, active_bits_size;
 
-    init_get_bits(&s->gb, buf, buf_size*8);
+    active_bits_size = buf_size * 8;
+    init_get_bits(&s->gb, buf, FFMAX(buf_size, buf_size2) * 8);
     if(s->codec_id ==CODEC_ID_RV10)
         mb_count = rv10_decode_picture_header(s);
     else
@@ -601,13 +602,26 @@ static int rv10_decode_packet(AVCodecContext *avctx,
         s->mv_type = MV_TYPE_16X16;
         ret=ff_h263_decode_mb(s, s->block);
 
-        if (ret != SLICE_ERROR && s->gb.size_in_bits < get_bits_count(&s->gb) && 8*buf_size2 >= get_bits_count(&s->gb)){
-            av_log(avctx, AV_LOG_DEBUG, "update size from %d to %d\n", s->gb.size_in_bits, 8*buf_size2);
-            s->gb.size_in_bits= 8*buf_size2;
+        // Repeat the slice end check from ff_h263_decode_mb with our active
+        // bitstream size
+        if (ret != SLICE_ERROR) {
+            int v = show_bits(&s->gb, 16);
+
+            if (get_bits_count(&s->gb) + 16 > active_bits_size)
+                v >>= get_bits_count(&s->gb) + 16 - active_bits_size;
+
+            if (!v)
+                ret = SLICE_END;
+        }
+        if (ret != SLICE_ERROR && active_bits_size < get_bits_count(&s->gb) &&
+            8 * buf_size2 >= get_bits_count(&s->gb)) {
+            active_bits_size = buf_size2 * 8;
+            av_log(avctx, AV_LOG_DEBUG, "update size from %d to %d\n",
+                   8 * buf_size, active_bits_size);
             ret= SLICE_OK;
         }
 
-        if (ret == SLICE_ERROR || s->gb.size_in_bits < get_bits_count(&s->gb)) {
+        if (ret == SLICE_ERROR || active_bits_size < get_bits_count(&s->gb)) {
             av_log(s->avctx, AV_LOG_ERROR, "ERROR at MB %d %d\n", s->mb_x, s->mb_y);
             return -1;
         }
@@ -629,7 +643,7 @@ static int rv10_decode_packet(AVCodecContext *avctx,
 
     ff_er_add_slice(s, start_mb_x, s->resync_mb_y, s->mb_x-1, s->mb_y, ER_MB_END);
 
-    return s->gb.size_in_bits;
+    return active_bits_size;
 }
 
 static int get_slice_offset(AVCodecContext *avctx, const uint8_t *buf, int n)
@@ -661,8 +675,12 @@ static int rv10_decode_frame(AVCodecContext *avctx,
 
     if(!avctx->slice_count){
         slice_count = (*buf++) + 1;
+        buf_size--;
         slices_hdr = buf + 4;
         buf += 8 * slice_count;
+        buf_size -= 8 * slice_count;
+        if (buf_size <= 0)
+            return AVERROR_INVALIDDATA;
     }else
         slice_count = avctx->slice_count;
 
@@ -696,9 +714,9 @@ static int rv10_decode_frame(AVCodecContext *avctx,
         ff_MPV_frame_end(s);
 
         if (s->pict_type == AV_PICTURE_TYPE_B || s->low_delay) {
-            *pict= *(AVFrame*)s->current_picture_ptr;
+            *pict = s->current_picture_ptr->f;
         } else if (s->last_picture_ptr != NULL) {
-            *pict= *(AVFrame*)s->last_picture_ptr;
+            *pict = s->last_picture_ptr->f;
         }
 
         if(s->last_picture_ptr || s->low_delay){
@@ -708,7 +726,7 @@ static int rv10_decode_frame(AVCodecContext *avctx,
         s->current_picture_ptr= NULL; //so we can detect if frame_end wasnt called (find some nicer solution...)
     }
 
-    return buf_size;
+    return avpkt->size;
 }
 
 AVCodec ff_rv10_decoder = {
diff --git a/libavcodec/rv34.c b/libavcodec/rv34.c
index 69273874c2..5550a74d0b 100644
--- a/libavcodec/rv34.c
+++ b/libavcodec/rv34.c
@@ -1660,7 +1660,7 @@ int ff_rv34_decode_frame(AVCodecContext *avctx,
     if (buf_size == 0) {
         /* special case for last picture */
         if (s->low_delay==0 && s->next_picture_ptr) {
-            *pict = *(AVFrame*)s->next_picture_ptr;
+            *pict = s->next_picture_ptr->f;
             s->next_picture_ptr = NULL;
 
             *data_size = sizeof(AVFrame);
@@ -1747,9 +1747,9 @@ int ff_rv34_decode_frame(AVCodecContext *avctx,
             ff_thread_report_progress(&s->current_picture_ptr->f, INT_MAX, 0);
 
         if (s->pict_type == AV_PICTURE_TYPE_B || s->low_delay) {
-            *pict = *(AVFrame*)s->current_picture_ptr;
+            *pict = s->current_picture_ptr->f;
         } else if (s->last_picture_ptr != NULL) {
-            *pict = *(AVFrame*)s->last_picture_ptr;
+            *pict = s->last_picture_ptr->f;
         }
 
         if(s->last_picture_ptr || s->low_delay){
diff --git a/libavcodec/svq1dec.c b/libavcodec/svq1dec.c
index cb89b3f619..381cd63937 100644
--- a/libavcodec/svq1dec.c
+++ b/libavcodec/svq1dec.c
@@ -735,7 +735,7 @@ static int svq1_decode_frame(AVCodecContext *avctx,
     }
   }
 
-  *pict = *(AVFrame*)&s->current_picture;
+  *pict = s->current_picture.f;
 
 
   ff_MPV_frame_end(s);
diff --git a/libavcodec/svq1enc.c b/libavcodec/svq1enc.c
index 84ff0d8814..a3936a8b37 100644
--- a/libavcodec/svq1enc.c
+++ b/libavcodec/svq1enc.c
@@ -473,7 +473,7 @@ static av_cold int svq1_encode_init(AVCodecContext *avctx)
     SVQ1Context * const s = avctx->priv_data;
 
     ff_dsputil_init(&s->dsp, avctx);
-    avctx->coded_frame= (AVFrame*)&s->picture;
+    avctx->coded_frame = &s->picture;
 
     s->frame_width = avctx->width;
     s->frame_height = avctx->height;
@@ -501,7 +501,7 @@ static int svq1_encode_frame(AVCodecContext *avctx, AVPacket *pkt,
                              const AVFrame *pict, int *got_packet)
 {
     SVQ1Context * const s = avctx->priv_data;
-    AVFrame * const p= (AVFrame*)&s->picture;
+    AVFrame * const p = &s->picture;
     AVFrame temp;
     int i, ret;
 
diff --git a/libavcodec/svq3.c b/libavcodec/svq3.c
index 3ebe95ef7f..8b112412cd 100644
--- a/libavcodec/svq3.c
+++ b/libavcodec/svq3.c
@@ -958,7 +958,7 @@ static int svq3_decode_frame(AVCodecContext *avctx,
     /* special case for last picture */
     if (buf_size == 0) {
         if (s->next_picture_ptr && !s->low_delay) {
-            *(AVFrame *) data = *(AVFrame *) &s->next_picture;
+            *(AVFrame *) data   = s->next_picture.f;
             s->next_picture_ptr = NULL;
             *data_size = sizeof(AVFrame);
         }
@@ -1101,9 +1101,9 @@ static int svq3_decode_frame(AVCodecContext *avctx,
     ff_MPV_frame_end(s);
 
     if (s->pict_type == AV_PICTURE_TYPE_B || s->low_delay) {
-        *(AVFrame *) data = *(AVFrame *) &s->current_picture;
+        *(AVFrame *) data = s->current_picture.f;
     } else {
-        *(AVFrame *) data = *(AVFrame *) &s->last_picture;
+        *(AVFrame *) data = s->last_picture.f;
     }
 
     /* Do not output the last pic after seeking. */
diff --git a/libavcodec/targa.c b/libavcodec/targa.c
index 573992ff8a..5ddd21bfef 100644
--- a/libavcodec/targa.c
+++ b/libavcodec/targa.c
@@ -106,7 +106,7 @@ static int decode_frame(AVCodecContext *avctx,
     const uint8_t *buf_end = avpkt->data + avpkt->size;
     TargaContext * const s = avctx->priv_data;
     AVFrame *picture = data;
-    AVFrame * const p= (AVFrame*)&s->picture;
+    AVFrame * const p = &s->picture;
     uint8_t *dst;
     int stride;
     int idlen, pal, compr, y, w, h, bpp, flags;
@@ -262,7 +262,7 @@ static int decode_frame(AVCodecContext *avctx,
         }
     }
 
-    *picture= *(AVFrame*)&s->picture;
+    *picture   = s->picture;
     *data_size = sizeof(AVPicture);
 
     return avpkt->size;
@@ -271,8 +271,8 @@ static int decode_frame(AVCodecContext *avctx,
 static av_cold int targa_init(AVCodecContext *avctx){
     TargaContext *s = avctx->priv_data;
 
-    avcodec_get_frame_defaults((AVFrame*)&s->picture);
-    avctx->coded_frame= (AVFrame*)&s->picture;
+    avcodec_get_frame_defaults(&s->picture);
+    avctx->coded_frame = &s->picture;
 
     return 0;
 }
diff --git a/libavcodec/tiff.c b/libavcodec/tiff.c
index 5d01e44ad8..5adf5a3b55 100644
--- a/libavcodec/tiff.c
+++ b/libavcodec/tiff.c
@@ -577,7 +577,7 @@ static int decode_frame(AVCodecContext *avctx,
     int buf_size = avpkt->size;
     TiffContext * const s = avctx->priv_data;
     AVFrame *picture = data;
-    AVFrame * const p= (AVFrame*)&s->picture;
+    AVFrame * const p = &s->picture;
     const uint8_t *orig_buf = buf, *end_buf = buf + buf_size;
     unsigned off;
     int id, le, ret;
@@ -691,7 +691,7 @@ static int decode_frame(AVCodecContext *avctx,
             dst += s->picture.linesize[0];
         }
     }
-    *picture= *(AVFrame*)&s->picture;
+    *picture   = s->picture;
     *data_size = sizeof(AVPicture);
 
     return buf_size;
@@ -703,8 +703,8 @@ static av_cold int tiff_init(AVCodecContext *avctx){
     s->width = 0;
     s->height = 0;
     s->avctx = avctx;
-    avcodec_get_frame_defaults((AVFrame*)&s->picture);
-    avctx->coded_frame= (AVFrame*)&s->picture;
+    avcodec_get_frame_defaults(&s->picture);
+    avctx->coded_frame = &s->picture;
     ff_lzw_decode_open(&s->lzw);
     ff_ccitt_unpack_init();
 
diff --git a/libavcodec/tiffenc.c b/libavcodec/tiffenc.c
index a0fecf7b0f..c061ac985f 100644
--- a/libavcodec/tiffenc.c
+++ b/libavcodec/tiffenc.c
@@ -206,7 +206,7 @@ static int encode_frame(AVCodecContext * avctx, AVPacket *pkt,
                         const AVFrame *pict, int *got_packet)
 {
     TiffEncoderContext *s = avctx->priv_data;
-    AVFrame *const p = (AVFrame *) & s->picture;
+    AVFrame *const p = &s->picture;
     int i;
     uint8_t *ptr;
     uint8_t *offset;
diff --git a/libavcodec/truemotion2.c b/libavcodec/truemotion2.c
index 985823dc25..ead7a39dbf 100644
--- a/libavcodec/truemotion2.c
+++ b/libavcodec/truemotion2.c
@@ -767,7 +767,7 @@ static int decode_frame(AVCodecContext *avctx,
     const uint8_t *buf = avpkt->data;
     int buf_size = avpkt->size;
     TM2Context * const l = avctx->priv_data;
-    AVFrame * const p= (AVFrame*)&l->pic;
+    AVFrame * const p = &l->pic;
     int i, skip, t;
 
     av_fast_padded_malloc(&l->buffer, &l->buffer_size, buf_size);
diff --git a/libavcodec/ulti.c b/libavcodec/ulti.c
index cd6f2d24ab..22b932a7e7 100644
--- a/libavcodec/ulti.c
+++ b/libavcodec/ulti.c
@@ -50,7 +50,7 @@ static av_cold int ulti_decode_init(AVCodecContext *avctx)
     s->height = avctx->height;
     s->blocks = (s->width / 8) * (s->height / 8);
     avctx->pix_fmt = PIX_FMT_YUV410P;
-    avcodec_get_frame_defaults(&s->frame);
+    avctx->coded_frame = &s->frame;
     avctx->coded_frame = (AVFrame*) &s->frame;
     s->ulti_codebook = ulti_codebook;
 
diff --git a/libavcodec/utils.c b/libavcodec/utils.c
index b612ca1ace..2b6e30e5bf 100644
--- a/libavcodec/utils.c
+++ b/libavcodec/utils.c
@@ -1226,6 +1226,12 @@ int attribute_align_arg avcodec_encode_video2(AVCodecContext *avctx,
                 avpkt->data = new_data;
         }
 
+        if (!user_packet && avpkt->data) {
+            uint8_t *new_data = av_realloc(avpkt->data, avpkt->size);
+            if (new_data)
+                avpkt->data = new_data;
+        }
+
         avctx->frame_number++;
     }
 
diff --git a/libavcodec/vc1dec.c b/libavcodec/vc1dec.c
index 6fb6b17ff8..15e5adb5c5 100644
--- a/libavcodec/vc1dec.c
+++ b/libavcodec/vc1dec.c
@@ -5456,7 +5456,7 @@ static int vc1_decode_frame(AVCodecContext *avctx, void *data,
     if (buf_size == 0 || (buf_size == 4 && AV_RB32(buf) == VC1_CODE_ENDOFSEQ)) {
         /* special case for last picture */
         if (s->low_delay == 0 && s->next_picture_ptr) {
-            *pict = *(AVFrame*)s->next_picture_ptr;
+            *pict = s->next_picture_ptr->f;
             s->next_picture_ptr = NULL;
 
             *data_size = sizeof(AVFrame);
@@ -5764,9 +5764,9 @@ image:
         *data_size = sizeof(AVFrame);
     } else {
         if (s->pict_type == AV_PICTURE_TYPE_B || s->low_delay) {
-            *pict = *(AVFrame*)s->current_picture_ptr;
+            *pict = s->current_picture_ptr->f;
         } else if (s->last_picture_ptr != NULL) {
-            *pict = *(AVFrame*)s->last_picture_ptr;
+            *pict = s->last_picture_ptr->f;
         }
         if (s->last_picture_ptr || s->low_delay) {
             *data_size = sizeof(AVFrame);
diff --git a/libavcodec/vcr1.c b/libavcodec/vcr1.c
index 1526215314..7b0e7a53b4 100644
--- a/libavcodec/vcr1.c
+++ b/libavcodec/vcr1.c
@@ -49,7 +49,7 @@ static int decode_frame(AVCodecContext *avctx,
     int buf_size = avpkt->size;
     VCR1Context * const a = avctx->priv_data;
     AVFrame *picture = data;
-    AVFrame * const p= (AVFrame*)&a->picture;
+    AVFrame * const p = &a->picture;
     const uint8_t *bytestream= buf;
     int i, x, y;
 
@@ -116,7 +116,7 @@ static int decode_frame(AVCodecContext *avctx,
         }
     }
 
-    *picture= *(AVFrame*)&a->picture;
+    *picture   = a->picture;
     *data_size = sizeof(AVPicture);
 
     return buf_size;
@@ -126,7 +126,7 @@ static int decode_frame(AVCodecContext *avctx,
 static int encode_frame(AVCodecContext *avctx, unsigned char *buf, int buf_size, void *data){
     VCR1Context * const a = avctx->priv_data;
     AVFrame *pict = data;
-    AVFrame * const p= (AVFrame*)&a->picture;
+    AVFrame * const p = &a->picture;
     int size;
 
     *p = *pict;
@@ -146,7 +146,7 @@ static int encode_frame(AVCodecContext *avctx, unsigned char *buf, int buf_size,
 static av_cold void common_init(AVCodecContext *avctx){
     VCR1Context * const a = avctx->priv_data;
 
-    avctx->coded_frame= (AVFrame*)&a->picture;
+    avctx->coded_frame = &a->picture;
     avcodec_get_frame_defaults(&a->picture);
     a->avctx= avctx;
 }
diff --git a/libavcodec/version.h b/libavcodec/version.h
index 1b72037572..1aa75eabbd 100644
--- a/libavcodec/version.h
+++ b/libavcodec/version.h
@@ -21,7 +21,7 @@
 #define AVCODEC_VERSION_H
 
 #define LIBAVCODEC_VERSION_MAJOR 54
-#define LIBAVCODEC_VERSION_MINOR  5
+#define LIBAVCODEC_VERSION_MINOR  6
 #define LIBAVCODEC_VERSION_MICRO 100
 
 #define LIBAVCODEC_VERSION_INT  AV_VERSION_INT(LIBAVCODEC_VERSION_MAJOR, \
diff --git a/libavcodec/vp8dsp.c b/libavcodec/vp8dsp.c
index ce90675d87..12f6988bff 100644
--- a/libavcodec/vp8dsp.c
+++ b/libavcodec/vp8dsp.c
@@ -77,7 +77,7 @@ static void vp8_luma_dc_wht_dc_c(DCTELEM block[4][4][16], DCTELEM dc[16])
 #define MUL_20091(a) ((((a)*20091) >> 16) + (a))
 #define MUL_35468(a)  (((a)*35468) >> 16)
 
-static void vp8_idct_add_c(uint8_t *dst, DCTELEM block[16], int stride)
+static void vp8_idct_add_c(uint8_t *dst, DCTELEM block[16], ptrdiff_t stride)
 {
     int i, t0, t1, t2, t3;
     uint8_t *cm = ff_cropTbl + MAX_NEG_CROP;
@@ -113,7 +113,7 @@ static void vp8_idct_add_c(uint8_t *dst, DCTELEM block[16], int stride)
     }
 }
 
-static void vp8_idct_dc_add_c(uint8_t *dst, DCTELEM block[16], int stride)
+static void vp8_idct_dc_add_c(uint8_t *dst, DCTELEM block[16], ptrdiff_t stride)
 {
     int i, dc = (block[0] + 4) >> 3;
     uint8_t *cm = ff_cropTbl + MAX_NEG_CROP + dc;
@@ -128,7 +128,7 @@ static void vp8_idct_dc_add_c(uint8_t *dst, DCTELEM block[16], int stride)
     }
 }
 
-static void vp8_idct_dc_add4uv_c(uint8_t *dst, DCTELEM block[4][16], int stride)
+static void vp8_idct_dc_add4uv_c(uint8_t *dst, DCTELEM block[4][16], ptrdiff_t stride)
 {
     vp8_idct_dc_add_c(dst+stride*0+0, block[0], stride);
     vp8_idct_dc_add_c(dst+stride*0+4, block[1], stride);
@@ -136,7 +136,7 @@ static void vp8_idct_dc_add4uv_c(uint8_t *dst, DCTELEM block[4][16], int stride)
     vp8_idct_dc_add_c(dst+stride*4+4, block[3], stride);
 }
 
-static void vp8_idct_dc_add4y_c(uint8_t *dst, DCTELEM block[4][16], int stride)
+static void vp8_idct_dc_add4y_c(uint8_t *dst, DCTELEM block[4][16], ptrdiff_t stride)
 {
     vp8_idct_dc_add_c(dst+ 0, block[0], stride);
     vp8_idct_dc_add_c(dst+ 4, block[1], stride);
@@ -157,7 +157,7 @@ static void vp8_idct_dc_add4y_c(uint8_t *dst, DCTELEM block[4][16], int stride)
 
 #define clip_int8(n) (cm[n+0x80]-0x80)
 
-static av_always_inline void filter_common(uint8_t *p, int stride, int is4tap)
+static av_always_inline void filter_common(uint8_t *p, ptrdiff_t stride, int is4tap)
 {
     LOAD_PIXELS
     int a, f1, f2;
@@ -188,7 +188,7 @@ static av_always_inline void filter_common(uint8_t *p, int stride, int is4tap)
     }
 }
 
-static av_always_inline int simple_limit(uint8_t *p, int stride, int flim)
+static av_always_inline int simple_limit(uint8_t *p, ptrdiff_t stride, int flim)
 {
     LOAD_PIXELS
     return 2*FFABS(p0-q0) + (FFABS(p1-q1) >> 1) <= flim;
@@ -198,7 +198,7 @@ static av_always_inline int simple_limit(uint8_t *p, int stride, int flim)
  * E - limit at the macroblock edge
  * I - limit for interior difference
  */
-static av_always_inline int normal_limit(uint8_t *p, int stride, int E, int I)
+static av_always_inline int normal_limit(uint8_t *p, ptrdiff_t stride, int E, int I)
 {
     LOAD_PIXELS
     return simple_limit(p, stride, E)
@@ -207,13 +207,13 @@ static av_always_inline int normal_limit(uint8_t *p, int stride, int E, int I)
 }
 
 // high edge variance
-static av_always_inline int hev(uint8_t *p, int stride, int thresh)
+static av_always_inline int hev(uint8_t *p, ptrdiff_t stride, int thresh)
 {
     LOAD_PIXELS
     return FFABS(p1-p0) > thresh || FFABS(q1-q0) > thresh;
 }
 
-static av_always_inline void filter_mbedge(uint8_t *p, int stride)
+static av_always_inline void filter_mbedge(uint8_t *p, ptrdiff_t stride)
 {
     int a0, a1, a2, w;
     uint8_t *cm = ff_cropTbl + MAX_NEG_CROP;
@@ -236,7 +236,7 @@ static av_always_inline void filter_mbedge(uint8_t *p, int stride)
 }
 
 #define LOOP_FILTER(dir, size, stridea, strideb, maybe_inline) \
-static maybe_inline void vp8_ ## dir ## _loop_filter ## size ## _c(uint8_t *dst, int stride,\
+static maybe_inline void vp8_ ## dir ## _loop_filter ## size ## _c(uint8_t *dst, ptrdiff_t stride,\
                                      int flim_E, int flim_I, int hev_thresh)\
 {\
     int i;\
@@ -250,7 +250,7 @@ static maybe_inline void vp8_ ## dir ## _loop_filter ## size ## _c(uint8_t *dst,
         }\
 }\
 \
-static maybe_inline void vp8_ ## dir ## _loop_filter ## size ## _inner_c(uint8_t *dst, int stride,\
+static maybe_inline void vp8_ ## dir ## _loop_filter ## size ## _inner_c(uint8_t *dst, ptrdiff_t stride,\
                                       int flim_E, int flim_I, int hev_thresh)\
 {\
     int i;\
@@ -270,13 +270,13 @@ LOOP_FILTER(h, 16, stride, 1,)
 
 #define UV_LOOP_FILTER(dir, stridea, strideb) \
 LOOP_FILTER(dir, 8, stridea, strideb, av_always_inline) \
-static void vp8_ ## dir ## _loop_filter8uv_c(uint8_t *dstU, uint8_t *dstV, int stride,\
+static void vp8_ ## dir ## _loop_filter8uv_c(uint8_t *dstU, uint8_t *dstV, ptrdiff_t stride,\
                                       int fE, int fI, int hev_thresh)\
 {\
   vp8_ ## dir ## _loop_filter8_c(dstU, stride, fE, fI, hev_thresh);\
   vp8_ ## dir ## _loop_filter8_c(dstV, stride, fE, fI, hev_thresh);\
 }\
-static void vp8_ ## dir ## _loop_filter8uv_inner_c(uint8_t *dstU, uint8_t *dstV, int stride,\
+static void vp8_ ## dir ## _loop_filter8uv_inner_c(uint8_t *dstU, uint8_t *dstV, ptrdiff_t stride,\
                                       int fE, int fI, int hev_thresh)\
 {\
   vp8_ ## dir ## _loop_filter8_inner_c(dstU, stride, fE, fI, hev_thresh);\
@@ -286,7 +286,7 @@ static void vp8_ ## dir ## _loop_filter8uv_inner_c(uint8_t *dstU, uint8_t *dstV,
 UV_LOOP_FILTER(v, 1, stride)
 UV_LOOP_FILTER(h, stride, 1)
 
-static void vp8_v_loop_filter_simple_c(uint8_t *dst, int stride, int flim)
+static void vp8_v_loop_filter_simple_c(uint8_t *dst, ptrdiff_t stride, int flim)
 {
     int i;
 
@@ -295,7 +295,7 @@ static void vp8_v_loop_filter_simple_c(uint8_t *dst, int stride, int flim)
             filter_common(dst+i, stride, 1);
 }
 
-static void vp8_h_loop_filter_simple_c(uint8_t *dst, int stride, int flim)
+static void vp8_h_loop_filter_simple_c(uint8_t *dst, ptrdiff_t stride, int flim)
 {
     int i;
 
@@ -315,7 +315,7 @@ static const uint8_t subpel_filters[7][6] = {
 };
 
 #define PUT_PIXELS(WIDTH) \
-static void put_vp8_pixels ## WIDTH ##_c(uint8_t *dst, int dststride, uint8_t *src, int srcstride, int h, int x, int y) { \
+static void put_vp8_pixels ## WIDTH ##_c(uint8_t *dst, ptrdiff_t dststride, uint8_t *src, ptrdiff_t srcstride, int h, int x, int y) { \
     int i; \
     for (i = 0; i < h; i++, dst+= dststride, src+= srcstride) { \
         memcpy(dst, src, WIDTH); \
@@ -335,7 +335,7 @@ PUT_PIXELS(4)
         F[3]*src[x+1*stride] - F[4]*src[x+2*stride] + 64) >> 7]
 
 #define VP8_EPEL_H(SIZE, TAPS) \
-static void put_vp8_epel ## SIZE ## _h ## TAPS ## _c(uint8_t *dst, int dststride, uint8_t *src, int srcstride, int h, int mx, int my) \
+static void put_vp8_epel ## SIZE ## _h ## TAPS ## _c(uint8_t *dst, ptrdiff_t dststride, uint8_t *src, ptrdiff_t srcstride, int h, int mx, int my) \
 { \
     const uint8_t *filter = subpel_filters[mx-1]; \
     uint8_t *cm = ff_cropTbl + MAX_NEG_CROP; \
@@ -349,7 +349,7 @@ static void put_vp8_epel ## SIZE ## _h ## TAPS ## _c(uint8_t *dst, int dststride
     } \
 }
 #define VP8_EPEL_V(SIZE, TAPS) \
-static void put_vp8_epel ## SIZE ## _v ## TAPS ## _c(uint8_t *dst, int dststride, uint8_t *src, int srcstride, int h, int mx, int my) \
+static void put_vp8_epel ## SIZE ## _v ## TAPS ## _c(uint8_t *dst, ptrdiff_t dststride, uint8_t *src, ptrdiff_t srcstride, int h, int mx, int my) \
 { \
     const uint8_t *filter = subpel_filters[my-1]; \
     uint8_t *cm = ff_cropTbl + MAX_NEG_CROP; \
@@ -363,7 +363,7 @@ static void put_vp8_epel ## SIZE ## _v ## TAPS ## _c(uint8_t *dst, int dststride
     } \
 }
 #define VP8_EPEL_HV(SIZE, HTAPS, VTAPS) \
-static void put_vp8_epel ## SIZE ## _h ## HTAPS ## v ## VTAPS ## _c(uint8_t *dst, int dststride, uint8_t *src, int srcstride, int h, int mx, int my) \
+static void put_vp8_epel ## SIZE ## _h ## HTAPS ## v ## VTAPS ## _c(uint8_t *dst, ptrdiff_t dststride, uint8_t *src, ptrdiff_t srcstride, int h, int mx, int my) \
 { \
     const uint8_t *filter = subpel_filters[mx-1]; \
     uint8_t *cm = ff_cropTbl + MAX_NEG_CROP; \
@@ -416,7 +416,7 @@ VP8_EPEL_HV(8,  6, 6)
 VP8_EPEL_HV(4,  6, 6)
 
 #define VP8_BILINEAR(SIZE) \
-static void put_vp8_bilinear ## SIZE ## _h_c(uint8_t *dst, int stride, uint8_t *src, int s2, int h, int mx, int my) \
+static void put_vp8_bilinear ## SIZE ## _h_c(uint8_t *dst, ptrdiff_t stride, uint8_t *src, ptrdiff_t s2, int h, int mx, int my) \
 { \
     int a = 8-mx, b = mx; \
     int x, y; \
@@ -428,7 +428,7 @@ static void put_vp8_bilinear ## SIZE ## _h_c(uint8_t *dst, int stride, uint8_t *
         src += stride; \
     } \
 } \
-static void put_vp8_bilinear ## SIZE ## _v_c(uint8_t *dst, int stride, uint8_t *src, int s2, int h, int mx, int my) \
+static void put_vp8_bilinear ## SIZE ## _v_c(uint8_t *dst, ptrdiff_t stride, uint8_t *src, ptrdiff_t s2, int h, int mx, int my) \
 { \
     int c = 8-my, d = my; \
     int x, y; \
@@ -441,7 +441,7 @@ static void put_vp8_bilinear ## SIZE ## _v_c(uint8_t *dst, int stride, uint8_t *
     } \
 } \
 \
-static void put_vp8_bilinear ## SIZE ## _hv_c(uint8_t *dst, int stride, uint8_t *src, int s2, int h, int mx, int my) \
+static void put_vp8_bilinear ## SIZE ## _hv_c(uint8_t *dst, ptrdiff_t stride, uint8_t *src, ptrdiff_t s2, int h, int mx, int my) \
 { \
     int a = 8-mx, b = mx; \
     int c = 8-my, d = my; \
diff --git a/libavcodec/vp8dsp.h b/libavcodec/vp8dsp.h
index 951a5566a7..d0c7d4aacc 100644
--- a/libavcodec/vp8dsp.h
+++ b/libavcodec/vp8dsp.h
@@ -29,40 +29,44 @@
 
 #include "dsputil.h"
 
-typedef void (*vp8_mc_func)(uint8_t *dst/*align 8*/, int dstStride,
-                            uint8_t *src/*align 1*/, int srcStride,
+typedef void (*vp8_mc_func)(uint8_t *dst/*align 8*/, ptrdiff_t dstStride,
+                            uint8_t *src/*align 1*/, ptrdiff_t srcStride,
                             int h, int x, int y);
 
 typedef struct VP8DSPContext {
     void (*vp8_luma_dc_wht)(DCTELEM block[4][4][16], DCTELEM dc[16]);
     void (*vp8_luma_dc_wht_dc)(DCTELEM block[4][4][16], DCTELEM dc[16]);
-    void (*vp8_idct_add)(uint8_t *dst, DCTELEM block[16], int stride);
-    void (*vp8_idct_dc_add)(uint8_t *dst, DCTELEM block[16], int stride);
-    void (*vp8_idct_dc_add4y)(uint8_t *dst, DCTELEM block[4][16], int stride);
-    void (*vp8_idct_dc_add4uv)(uint8_t *dst, DCTELEM block[4][16], int stride);
+    void (*vp8_idct_add)(uint8_t *dst, DCTELEM block[16], ptrdiff_t stride);
+    void (*vp8_idct_dc_add)(uint8_t *dst, DCTELEM block[16], ptrdiff_t stride);
+    void (*vp8_idct_dc_add4y)(uint8_t *dst, DCTELEM block[4][16],
+                              ptrdiff_t stride);
+    void (*vp8_idct_dc_add4uv)(uint8_t *dst, DCTELEM block[4][16],
+                               ptrdiff_t stride);
 
     // loop filter applied to edges between macroblocks
-    void (*vp8_v_loop_filter16y)(uint8_t *dst, int stride,
+    void (*vp8_v_loop_filter16y)(uint8_t *dst, ptrdiff_t stride,
                                  int flim_E, int flim_I, int hev_thresh);
-    void (*vp8_h_loop_filter16y)(uint8_t *dst, int stride,
+    void (*vp8_h_loop_filter16y)(uint8_t *dst, ptrdiff_t stride,
                                  int flim_E, int flim_I, int hev_thresh);
-    void (*vp8_v_loop_filter8uv)(uint8_t *dstU, uint8_t *dstV, int stride,
+    void (*vp8_v_loop_filter8uv)(uint8_t *dstU, uint8_t *dstV, ptrdiff_t stride,
                                  int flim_E, int flim_I, int hev_thresh);
-    void (*vp8_h_loop_filter8uv)(uint8_t *dstU, uint8_t *dstV, int stride,
+    void (*vp8_h_loop_filter8uv)(uint8_t *dstU, uint8_t *dstV, ptrdiff_t stride,
                                  int flim_E, int flim_I, int hev_thresh);
 
     // loop filter applied to inner macroblock edges
-    void (*vp8_v_loop_filter16y_inner)(uint8_t *dst, int stride,
+    void (*vp8_v_loop_filter16y_inner)(uint8_t *dst, ptrdiff_t stride,
                                        int flim_E, int flim_I, int hev_thresh);
-    void (*vp8_h_loop_filter16y_inner)(uint8_t *dst, int stride,
+    void (*vp8_h_loop_filter16y_inner)(uint8_t *dst, ptrdiff_t stride,
                                        int flim_E, int flim_I, int hev_thresh);
-    void (*vp8_v_loop_filter8uv_inner)(uint8_t *dstU, uint8_t *dstV, int stride,
+    void (*vp8_v_loop_filter8uv_inner)(uint8_t *dstU, uint8_t *dstV,
+                                       ptrdiff_t stride,
                                        int flim_E, int flim_I, int hev_thresh);
-    void (*vp8_h_loop_filter8uv_inner)(uint8_t *dstU, uint8_t *dstV, int stride,
+    void (*vp8_h_loop_filter8uv_inner)(uint8_t *dstU, uint8_t *dstV,
+                                       ptrdiff_t stride,
                                        int flim_E, int flim_I, int hev_thresh);
 
-    void (*vp8_v_loop_filter_simple)(uint8_t *dst, int stride, int flim);
-    void (*vp8_h_loop_filter_simple)(uint8_t *dst, int stride, int flim);
+    void (*vp8_v_loop_filter_simple)(uint8_t *dst, ptrdiff_t stride, int flim);
+    void (*vp8_h_loop_filter_simple)(uint8_t *dst, ptrdiff_t stride, int flim);
 
     /**
      * first dimension: width>>3, height is assumed equal to width
@@ -76,9 +80,12 @@ typedef struct VP8DSPContext {
     vp8_mc_func put_vp8_bilinear_pixels_tab[3][3][3];
 } VP8DSPContext;
 
-void ff_put_vp8_pixels16_c(uint8_t *dst, uint8_t *src, int stride, int h, int x, int y);
-void ff_put_vp8_pixels8_c(uint8_t *dst, uint8_t *src, int stride, int h, int x, int y);
-void ff_put_vp8_pixels4_c(uint8_t *dst, uint8_t *src, int stride, int h, int x, int y);
+void ff_put_vp8_pixels16_c(uint8_t *dst, uint8_t *src, ptrdiff_t stride,
+                           int h, int x, int y);
+void ff_put_vp8_pixels8_c(uint8_t *dst, uint8_t *src, ptrdiff_t stride,
+                          int h, int x, int y);
+void ff_put_vp8_pixels4_c(uint8_t *dst, uint8_t *src, ptrdiff_t stride,
+                          int h, int x, int y);
 
 void ff_vp8dsp_init(VP8DSPContext *c);
 void ff_vp8dsp_init_x86(VP8DSPContext *c);
diff --git a/libavcodec/wma.h b/libavcodec/wma.h
index 6c8e944b79..954153fac0 100644
--- a/libavcodec/wma.h
+++ b/libavcodec/wma.h
@@ -124,7 +124,7 @@ typedef struct WMACodecContext {
     /* output buffer for one frame and the last for IMDCT windowing */
     DECLARE_ALIGNED(32, float, frame_out)[MAX_CHANNELS][BLOCK_MAX_SIZE * 2];
     /* last frame info */
-    uint8_t last_superframe[MAX_CODED_SUPERFRAME_SIZE + 4]; /* padding added */
+    uint8_t last_superframe[MAX_CODED_SUPERFRAME_SIZE + FF_INPUT_BUFFER_PADDING_SIZE]; /* padding added */
     int last_bitoffset;
     int last_superframe_len;
     float noise_table[NOISE_TAB_SIZE];
diff --git a/libavcodec/wmadec.c b/libavcodec/wmadec.c
index 7dae379ba8..f26a1433a9 100644
--- a/libavcodec/wmadec.c
+++ b/libavcodec/wmadec.c
@@ -863,6 +863,12 @@ static int wma_decode_superframe(AVCodecContext *avctx, void *data,
 
     if (s->use_bit_reservoir) {
         bit_offset = get_bits(&s->gb, s->byte_offset_bits + 3);
+        if (bit_offset > get_bits_left(&s->gb)) {
+            av_log(avctx, AV_LOG_ERROR,
+                   "Invalid last frame bit offset %d > buf size %d (%d)\n",
+                   bit_offset, get_bits_left(&s->gb), buf_size);
+            goto fail;
+        }
 
         if (s->last_superframe_len > 0) {
             //        printf("skip=%d\n", s->last_bitoffset);
@@ -879,9 +885,10 @@ static int wma_decode_superframe(AVCodecContext *avctx, void *data,
             if (len > 0) {
                 *q++ = (get_bits)(&s->gb, len) << (8 - len);
             }
+            memset(q, 0, FF_INPUT_BUFFER_PADDING_SIZE);
 
             /* XXX: bit_offset bits into last frame */
-            init_get_bits(&s->gb, s->last_superframe, MAX_CODED_SUPERFRAME_SIZE*8);
+            init_get_bits(&s->gb, s->last_superframe, s->last_superframe_len * 8 + bit_offset);
             /* skip unused bits */
             if (s->last_bitoffset > 0)
                 skip_bits(&s->gb, s->last_bitoffset);
@@ -895,9 +902,9 @@ static int wma_decode_superframe(AVCodecContext *avctx, void *data,
 
         /* read each frame starting from bit_offset */
         pos = bit_offset + 4 + 4 + s->byte_offset_bits + 3;
-        if (pos >= MAX_CODED_SUPERFRAME_SIZE * 8)
+        if (pos >= MAX_CODED_SUPERFRAME_SIZE * 8 || pos > buf_size * 8)
             return AVERROR_INVALIDDATA;
-        init_get_bits(&s->gb, buf + (pos >> 3), (MAX_CODED_SUPERFRAME_SIZE - (pos >> 3))*8);
+        init_get_bits(&s->gb, buf + (pos >> 3), (buf_size - (pos >> 3))*8);
         len = pos & 7;
         if (len > 0)
             skip_bits(&s->gb, len);
diff --git a/libavcodec/wmalosslessdec.c b/libavcodec/wmalosslessdec.c
index b32731f96e..973edc438a 100644
--- a/libavcodec/wmalosslessdec.c
+++ b/libavcodec/wmalosslessdec.c
@@ -1,8 +1,9 @@
 /*
- * Wmall compatible decoder
+ * Windows Media Audio Lossless decoder
  * Copyright (c) 2007 Baptiste Coudurier, Benjamin Larsson, Ulion
  * Copyright (c) 2008 - 2011 Sascha Sommer, Benjamin Larsson
  * Copyright (c) 2011 Andreas Öman
+ * Copyright (c) 2011 - 2012 Mashiat Sarker Shakkhar
  *
  * This file is part of FFmpeg.
  *
@@ -21,210 +22,94 @@
  * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
  */
 
-/**
- * @file
- * @brief wmall decoder implementation
- * Wmall is an MDCT based codec comparable to wma standard or AAC.
- * The decoding therefore consists of the following steps:
- * - bitstream decoding
- * - reconstruction of per-channel data
- * - rescaling and inverse quantization
- * - IMDCT
- * - windowing and overlapp-add
- *
- * The compressed wmall bitstream is split into individual packets.
- * Every such packet contains one or more wma frames.
- * The compressed frames may have a variable length and frames may
- * cross packet boundaries.
- * Common to all wmall frames is the number of samples that are stored in
- * a frame.
- * The number of samples and a few other decode flags are stored
- * as extradata that has to be passed to the decoder.
- *
- * The wmall frames themselves are again split into a variable number of
- * subframes. Every subframe contains the data for 2^N time domain samples
- * where N varies between 7 and 12.
- *
- * Example wmall bitstream (in samples):
- *
- * ||   packet 0           || packet 1 || packet 2      packets
- * ---------------------------------------------------
- * || frame 0      || frame 1       || frame 2    ||    frames
- * ---------------------------------------------------
- * ||   |      |   ||   |   |   |   ||            ||    subframes of channel 0
- * ---------------------------------------------------
- * ||      |   |   ||   |   |   |   ||            ||    subframes of channel 1
- * ---------------------------------------------------
- *
- * The frame layouts for the individual channels of a wma frame does not need
- * to be the same.
- *
- * However, if the offsets and lengths of several subframes of a frame are the
- * same, the subframes of the channels can be grouped.
- * Every group may then use special coding techniques like M/S stereo coding
- * to improve the compression ratio. These channel transformations do not
- * need to be applied to a whole subframe. Instead, they can also work on
- * individual scale factor bands (see below).
- * The coefficients that carry the audio signal in the frequency domain
- * are transmitted as huffman-coded vectors with 4, 2 and 1 elements.
- * In addition to that, the encoder can switch to a runlevel coding scheme
- * by transmitting subframe_length / 128 zero coefficients.
- *
- * Before the audio signal can be converted to the time domain, the
- * coefficients have to be rescaled and inverse quantized.
- * A subframe is therefore split into several scale factor bands that get
- * scaled individually.
- * Scale factors are submitted for every frame but they might be shared
- * between the subframes of a channel. Scale factors are initially DPCM-coded.
- * Once scale factors are shared, the differences are transmitted as runlevel
- * codes.
- * Every subframe length and offset combination in the frame layout shares a
- * common quantization factor that can be adjusted for every channel by a
- * modifier.
- * After the inverse quantization, the coefficients get processed by an IMDCT.
- * The resulting values are then windowed with a sine window and the first half
- * of the values are added to the second half of the output from the previous
- * subframe in order to reconstruct the output samples.
- */
-
 #include "avcodec.h"
 #include "internal.h"
 #include "get_bits.h"
 #include "put_bits.h"
-#include "dsputil.h"
 #include "wma.h"
 
 /** current decoder limitations */
-#define WMALL_MAX_CHANNELS    8                             ///< max number of handled channels
-#define MAX_SUBFRAMES  32                                    ///< max number of subframes per channel
-#define MAX_BANDS      29                                    ///< max number of scale factor bands
-#define MAX_FRAMESIZE  32768                                 ///< maximum compressed frame size
-
-#define WMALL_BLOCK_MIN_BITS  6                                           ///< log2 of min block size
-#define WMALL_BLOCK_MAX_BITS 12                                           ///< log2 of max block size
-#define WMALL_BLOCK_MAX_SIZE (1 << WMALL_BLOCK_MAX_BITS)                 ///< maximum block size
+#define WMALL_MAX_CHANNELS      8                       ///< max number of handled channels
+#define MAX_SUBFRAMES          32                       ///< max number of subframes per channel
+#define MAX_BANDS              29                       ///< max number of scale factor bands
+#define MAX_FRAMESIZE       32768                       ///< maximum compressed frame size
+
+#define WMALL_BLOCK_MIN_BITS    6                       ///< log2 of min block size
+#define WMALL_BLOCK_MAX_BITS   12                       ///< log2 of max block size
+#define WMALL_BLOCK_MAX_SIZE (1 << WMALL_BLOCK_MAX_BITS)    ///< maximum block size
 #define WMALL_BLOCK_SIZES    (WMALL_BLOCK_MAX_BITS - WMALL_BLOCK_MIN_BITS + 1) ///< possible block sizes
 
 
-#define VLCBITS            9
-#define SCALEVLCBITS       8
-#define VEC4MAXDEPTH    ((HUFF_VEC4_MAXBITS+VLCBITS-1)/VLCBITS)
-#define VEC2MAXDEPTH    ((HUFF_VEC2_MAXBITS+VLCBITS-1)/VLCBITS)
-#define VEC1MAXDEPTH    ((HUFF_VEC1_MAXBITS+VLCBITS-1)/VLCBITS)
-#define SCALEMAXDEPTH   ((HUFF_SCALE_MAXBITS+SCALEVLCBITS-1)/SCALEVLCBITS)
-#define SCALERLMAXDEPTH ((HUFF_SCALE_RL_MAXBITS+VLCBITS-1)/VLCBITS)
-
-static float            sin64[33];        ///< sinus table for decorrelation
-
 /**
- * @brief frame specific decoder context for a single channel
+ * @brief frame-specific decoder context for a single channel
  */
 typedef struct {
-    int16_t  prev_block_len;                          ///< length of the previous block
-    uint8_t  transmit_coefs;
-    uint8_t  num_subframes;
-    uint16_t subframe_len[MAX_SUBFRAMES];             ///< subframe length in samples
-    uint16_t subframe_offset[MAX_SUBFRAMES];          ///< subframe positions in the current frame
-    uint8_t  cur_subframe;                            ///< current subframe number
-    uint16_t decoded_samples;                         ///< number of already processed samples
-    uint8_t  grouped;                                 ///< channel is part of a group
-    int      quant_step;                              ///< quantization step for the current subframe
-    int8_t   reuse_sf;                                ///< share scale factors between subframes
-    int8_t   scale_factor_step;                       ///< scaling step for the current subframe
-    int      max_scale_factor;                        ///< maximum scale factor for the current subframe
-    int      saved_scale_factors[2][MAX_BANDS];       ///< resampled and (previously) transmitted scale factor values
-    int8_t   scale_factor_idx;                        ///< index for the transmitted scale factor values (used for resampling)
-    int*     scale_factors;                           ///< pointer to the scale factor values used for decoding
-    uint8_t  table_idx;                               ///< index in sf_offsets for the scale factor reference block
-    float*   coeffs;                                  ///< pointer to the subframe decode buffer
-    uint16_t num_vec_coeffs;                          ///< number of vector coded coefficients
-    DECLARE_ALIGNED(16, float, out)[WMALL_BLOCK_MAX_SIZE + WMALL_BLOCK_MAX_SIZE / 2]; ///< output buffer
-    int      transient_counter;                       ///< number of transient samples from the beginning of transient zone
+    int16_t     prev_block_len;                         ///< length of the previous block
+    uint8_t     transmit_coefs;
+    uint8_t     num_subframes;
+    uint16_t    subframe_len[MAX_SUBFRAMES];            ///< subframe length in samples
+    uint16_t    subframe_offsets[MAX_SUBFRAMES];        ///< subframe positions in the current frame
+    uint8_t     cur_subframe;                           ///< current subframe number
+    uint16_t    decoded_samples;                        ///< number of already processed samples
+    int         quant_step;                             ///< quantization step for the current subframe
+    int         transient_counter;                      ///< number of transient samples from the beginning of the transient zone
 } WmallChannelCtx;
 
 /**
- * @brief channel group for channel transformations
- */
-typedef struct {
-    uint8_t num_channels;                                     ///< number of channels in the group
-    int8_t  transform;                                        ///< transform on / off
-    int8_t  transform_band[MAX_BANDS];                        ///< controls if the transform is enabled for a certain band
-    float   decorrelation_matrix[WMALL_MAX_CHANNELS*WMALL_MAX_CHANNELS];
-    float*  channel_data[WMALL_MAX_CHANNELS];                ///< transformation coefficients
-} WmallChannelGrp;
-
-/**
  * @brief main decoder context
  */
 typedef struct WmallDecodeCtx {
     /* generic decoder variables */
-    AVCodecContext*  avctx;                         ///< codec context for av_log
-    DSPContext       dsp;                           ///< accelerated DSP functions
-    AVFrame          frame;
-    uint8_t          frame_data[MAX_FRAMESIZE +
-                      FF_INPUT_BUFFER_PADDING_SIZE];///< compressed frame data
-    PutBitContext    pb;                            ///< context for filling the frame_data buffer
-    FFTContext       mdct_ctx[WMALL_BLOCK_SIZES];  ///< MDCT context per block size
-    DECLARE_ALIGNED(16, float, tmp)[WMALL_BLOCK_MAX_SIZE]; ///< IMDCT output buffer
-    float*           windows[WMALL_BLOCK_SIZES];   ///< windows for the different block sizes
+    AVCodecContext  *avctx;
+    AVFrame         frame;
+    uint8_t         frame_data[MAX_FRAMESIZE + FF_INPUT_BUFFER_PADDING_SIZE];  ///< compressed frame data
+    PutBitContext   pb;                             ///< context for filling the frame_data buffer
 
     /* frame size dependent frame information (set during initialization) */
-    uint32_t         decode_flags;                  ///< used compression features
-    uint8_t          len_prefix;                    ///< frame is prefixed with its length
-    uint8_t          dynamic_range_compression;     ///< frame contains DRC data
-    uint8_t          bits_per_sample;               ///< integer audio sample size for the unscaled IMDCT output (used to scale to [-1.0, 1.0])
-    uint16_t         samples_per_frame;             ///< number of samples to output
-    uint16_t         log2_frame_size;
-    int8_t           num_channels;                  ///< number of channels in the stream (same as AVCodecContext.num_channels)
-    int8_t           lfe_channel;                   ///< lfe channel index
-    uint8_t          max_num_subframes;
-    uint8_t          subframe_len_bits;             ///< number of bits used for the subframe length
-    uint8_t          max_subframe_len_bit;          ///< flag indicating that the subframe is of maximum size when the first subframe length bit is 1
-    uint16_t         min_samples_per_subframe;
-    int8_t           num_sfb[WMALL_BLOCK_SIZES];   ///< scale factor bands per block size
-    int16_t          sfb_offsets[WMALL_BLOCK_SIZES][MAX_BANDS];                    ///< scale factor band offsets (multiples of 4)
-    int8_t           sf_offsets[WMALL_BLOCK_SIZES][WMALL_BLOCK_SIZES][MAX_BANDS]; ///< scale factor resample matrix
-    int16_t          subwoofer_cutoffs[WMALL_BLOCK_SIZES]; ///< subwoofer cutoff values
+    uint32_t        decode_flags;                   ///< used compression features
+    int             len_prefix;                     ///< frame is prefixed with its length
+    int             dynamic_range_compression;      ///< frame contains DRC data
+    uint8_t         bits_per_sample;                ///< integer audio sample size for the unscaled IMDCT output (used to scale to [-1.0, 1.0])
+    uint16_t        samples_per_frame;              ///< number of samples to output
+    uint16_t        log2_frame_size;
+    int8_t          num_channels;                   ///< number of channels in the stream (same as AVCodecContext.num_channels)
+    int8_t          lfe_channel;                    ///< lfe channel index
+    uint8_t         max_num_subframes;
+    uint8_t         subframe_len_bits;              ///< number of bits used for the subframe length
+    uint8_t         max_subframe_len_bit;           ///< flag indicating that the subframe is of maximum size when the first subframe length bit is 1
+    uint16_t        min_samples_per_subframe;
 
     /* packet decode state */
-    GetBitContext    pgb;                           ///< bitstream reader context for the packet
-    int              next_packet_start;             ///< start offset of the next wma packet in the demuxer packet
-    uint8_t          packet_offset;                 ///< frame offset in the packet
-    uint8_t          packet_sequence_number;        ///< current packet number
-    int              num_saved_bits;                ///< saved number of bits
-    int              frame_offset;                  ///< frame offset in the bit reservoir
-    int              subframe_offset;               ///< subframe offset in the bit reservoir
-    uint8_t          packet_loss;                   ///< set in case of bitstream error
-    uint8_t          packet_done;                   ///< set when a packet is fully decoded
+    GetBitContext   pgb;                            ///< bitstream reader context for the packet
+    int             next_packet_start;              ///< start offset of the next WMA packet in the demuxer packet
+    uint8_t         packet_offset;                  ///< offset to the frame in the packet
+    uint8_t         packet_sequence_number;         ///< current packet number
+    int             num_saved_bits;                 ///< saved number of bits
+    int             frame_offset;                   ///< frame offset in the bit reservoir
+    int             subframe_offset;                ///< subframe offset in the bit reservoir
+    uint8_t         packet_loss;                    ///< set in case of bitstream error
+    uint8_t         packet_done;                    ///< set when a packet is fully decoded
 
     /* frame decode state */
-    uint32_t         frame_num;                     ///< current frame number (not used for decoding)
-    GetBitContext    gb;                            ///< bitstream reader context
-    int              buf_bit_size;                  ///< buffer size in bits
-    int16_t*         samples_16;                    ///< current samplebuffer pointer (16-bit)
-    int16_t*         samples_16_end;                ///< maximum samplebuffer pointer
+    uint32_t        frame_num;                      ///< current frame number (not used for decoding)
+    GetBitContext   gb;                             ///< bitstream reader context
+    int             buf_bit_size;                   ///< buffer size in bits
+    int16_t         *samples_16;                    ///< current samplebuffer pointer (16-bit)
+    int16_t         *samples_16_end;                ///< maximum samplebuffer pointer
     int             *samples_32;                    ///< current samplebuffer pointer (24-bit)
     int             *samples_32_end;                ///< maximum samplebuffer pointer
-    uint8_t          drc_gain;                      ///< gain for the DRC tool
-    int8_t           skip_frame;                    ///< skip output step
-    int8_t           parsed_all_subframes;          ///< all subframes decoded?
+    uint8_t         drc_gain;                       ///< gain for the DRC tool
+    int8_t          skip_frame;                     ///< skip output step
+    int8_t          parsed_all_subframes;           ///< all subframes decoded?
 
     /* subframe/block decode state */
-    int16_t          subframe_len;                  ///< current subframe length
-    int8_t           channels_for_cur_subframe;     ///< number of channels that contain the subframe
-    int8_t           channel_indexes_for_cur_subframe[WMALL_MAX_CHANNELS];
-    int8_t           num_bands;                     ///< number of scale factor bands
-    int8_t           transmit_num_vec_coeffs;       ///< number of vector coded coefficients is part of the bitstream
-    int16_t*         cur_sfb_offsets;               ///< sfb offsets for the current block
-    uint8_t          table_idx;                     ///< index for the num_sfb, sfb_offsets, sf_offsets and subwoofer_cutoffs tables
-    int8_t           esc_len;                       ///< length of escaped coefficients
-
-    uint8_t          num_chgroups;                  ///< number of channel groups
-    WmallChannelGrp chgroup[WMALL_MAX_CHANNELS];    ///< channel group information
+    int16_t         subframe_len;                   ///< current subframe length
+    int8_t          channels_for_cur_subframe;      ///< number of channels that contain the subframe
+    int8_t          channel_indexes_for_cur_subframe[WMALL_MAX_CHANNELS];
 
     WmallChannelCtx channel[WMALL_MAX_CHANNELS];    ///< per channel data
 
-    // WMA lossless
+    // WMA Lossless-specific
 
     uint8_t do_arith_coding;
     uint8_t do_ac_filter;
@@ -232,41 +117,38 @@ typedef struct WmallDecodeCtx {
     uint8_t do_mclms;
     uint8_t do_lpc;
 
-    int8_t acfilter_order;
-    int8_t acfilter_scaling;
+    int8_t  acfilter_order;
+    int8_t  acfilter_scaling;
     int64_t acfilter_coeffs[16];
-    int acfilter_prevvalues[2][16];
+    int     acfilter_prevvalues[2][16];
 
-    int8_t mclms_order;
-    int8_t mclms_scaling;
+    int8_t  mclms_order;
+    int8_t  mclms_scaling;
     int16_t mclms_coeffs[128];
     int16_t mclms_coeffs_cur[4];
-    int mclms_prevvalues[64];   // FIXME: should be 32-bit / 16-bit depending on bit-depth
+    int16_t mclms_prevvalues[64];
     int16_t mclms_updates[64];
-    int mclms_recent;
+    int     mclms_recent;
 
-    int movave_scaling;
-    int quant_stepsize;
+    int     movave_scaling;
+    int     quant_stepsize;
 
     struct {
-    int order;
-    int scaling;
-    int coefsend;
-    int bitsend;
-    int16_t coefs[256];
-    int lms_prevvalues[512];    // FIXME: see above
-    int16_t lms_updates[512];   // and here too
-    int recent;
-    } cdlms[2][9];              /* XXX: Here, 2 is the max. no. of channels allowed,
-                                        9 is the maximum no. of filters per channel.
-                                        Question is, why 2 if WMALL_MAX_CHANNELS == 8 */
-
+        int order;
+        int scaling;
+        int coefsend;
+        int bitsend;
+        int16_t coefs[256];
+        int16_t lms_prevvalues[512];
+        int16_t lms_updates[512];
+        int recent;
+    } cdlms[2][9];
 
     int cdlms_ttl[2];
 
     int bV3RTM;
 
-    int is_channel_coded[2];    // XXX: same question as above applies here too (and below)
+    int is_channel_coded[2];
     int update_speed[2];
 
     int transient[2];
@@ -277,139 +159,76 @@ typedef struct WmallDecodeCtx {
 
     int channel_residues[2][2048];
 
-
     int lpc_coefs[2][40];
     int lpc_order;
     int lpc_scaling;
     int lpc_intbits;
 
-    int channel_coeffs[2][2048]; // FIXME: should be 32-bit / 16-bit depending on bit-depth
-
+    int channel_coeffs[2][2048];
 } WmallDecodeCtx;
 
 
-#undef dprintf
-#define dprintf(pctx, ...) av_log(pctx, AV_LOG_DEBUG, __VA_ARGS__)
-
-
-static int num_logged_tiles = 0;
-static int num_logged_subframes = 0;
-static int num_lms_update_call = 0;
-
-/**
- *@brief helper function to print the most important members of the context
- *@param s context
- */
-static void av_cold dump_context(WmallDecodeCtx *s)
-{
-#define PRINT(a, b)     av_log(s->avctx, AV_LOG_DEBUG, " %s = %d\n", a, b);
-#define PRINT_HEX(a, b) av_log(s->avctx, AV_LOG_DEBUG, " %s = %x\n", a, b);
-
-    PRINT("ed sample bit depth", s->bits_per_sample);
-    PRINT_HEX("ed decode flags", s->decode_flags);
-    PRINT("samples per frame",   s->samples_per_frame);
-    PRINT("log2 frame size",     s->log2_frame_size);
-    PRINT("max num subframes",   s->max_num_subframes);
-    PRINT("len prefix",          s->len_prefix);
-    PRINT("num channels",        s->num_channels);
-}
-
-static void dump_int_buffer(uint8_t *buffer, int size, int length, int delimiter)
-{
-    int i;
-
-    for (i=0 ; i<length ; i++) {
-        if (!(i%delimiter))
-            av_log(0, 0, "\n[%d] ", i);
-        av_log(0, 0, "%d, ", *(int16_t *)(buffer + i * size));
-    }
-    av_log(0, 0, "\n");
-}
-
-/**
- *@brief Uninitialize the decoder and free all resources.
- *@param avctx codec context
- *@return 0 on success, < 0 otherwise
- */
-static av_cold int decode_end(AVCodecContext *avctx)
-{
-    WmallDecodeCtx *s = avctx->priv_data;
-    int i;
-
-    for (i = 0; i < WMALL_BLOCK_SIZES; i++)
-        ff_mdct_end(&s->mdct_ctx[i]);
-
-    return 0;
-}
-
-/**
- *@brief Initialize the decoder.
- *@param avctx codec context
- *@return 0 on success, -1 otherwise
- */
 static av_cold int decode_init(AVCodecContext *avctx)
 {
-    WmallDecodeCtx *s = avctx->priv_data;
+    WmallDecodeCtx *s  = avctx->priv_data;
     uint8_t *edata_ptr = avctx->extradata;
     unsigned int channel_mask;
-    int i;
-    int log2_max_num_subframes;
-    int num_possible_block_sizes;
+    int i, log2_max_num_subframes, num_possible_block_sizes;
 
     s->avctx = avctx;
-    dsputil_init(&s->dsp, avctx);
     init_put_bits(&s->pb, s->frame_data, MAX_FRAMESIZE);
 
     if (avctx->extradata_size >= 18) {
-        s->decode_flags    = AV_RL16(edata_ptr+14);
-        channel_mask       = AV_RL32(edata_ptr+2);
+        s->decode_flags    = AV_RL16(edata_ptr + 14);
+        channel_mask       = AV_RL32(edata_ptr +  2);
         s->bits_per_sample = AV_RL16(edata_ptr);
         if (s->bits_per_sample == 16)
             avctx->sample_fmt = AV_SAMPLE_FMT_S16;
-        else if (s->bits_per_sample == 24)
+        else if (s->bits_per_sample == 24) {
             avctx->sample_fmt = AV_SAMPLE_FMT_S32;
-        else {
+            av_log_missing_feature(avctx, "bit-depth higher than 16", 0);
+            return AVERROR_PATCHWELCOME;
+        } else {
             av_log(avctx, AV_LOG_ERROR, "Unknown bit-depth: %d\n",
                    s->bits_per_sample);
             return AVERROR_INVALIDDATA;
         }
-        /** dump the extradata */
+        /* dump the extradata */
         for (i = 0; i < avctx->extradata_size; i++)
-            dprintf(avctx, "[%x] ", avctx->extradata[i]);
-        dprintf(avctx, "\n");
+            av_dlog(avctx, AV_LOG_DEBUG, "[%x] ", avctx->extradata[i]);
+        av_dlog(avctx, AV_LOG_DEBUG, "\n");
 
     } else {
-        av_log_ask_for_sample(avctx, "Unknown extradata size\n");
+        av_log_ask_for_sample(avctx, "Unsupported extradata size\n");
         return AVERROR_INVALIDDATA;
     }
 
-    /** generic init */
+    /* generic init */
     s->log2_frame_size = av_log2(avctx->block_align) + 4;
 
-    /** frame info */
+    /* frame info */
     s->skip_frame  = 1; /* skip first frame */
     s->packet_loss = 1;
-    s->len_prefix  = (s->decode_flags & 0x40);
+    s->len_prefix  = s->decode_flags & 0x40;
 
-    /** get frame len */
+    /* get frame len */
     s->samples_per_frame = 1 << ff_wma_get_frame_len_bits(avctx->sample_rate,
                                                           3, s->decode_flags);
 
-    /** init previous block len */
+    /* init previous block len */
     for (i = 0; i < avctx->channels; i++)
         s->channel[i].prev_block_len = s->samples_per_frame;
 
-    /** subframe info */
-    log2_max_num_subframes  = ((s->decode_flags & 0x38) >> 3);
+    /* subframe info */
+    log2_max_num_subframes  = (s->decode_flags & 0x38) >> 3;
     s->max_num_subframes    = 1 << log2_max_num_subframes;
     s->max_subframe_len_bit = 0;
     s->subframe_len_bits    = av_log2(log2_max_num_subframes) + 1;
 
     num_possible_block_sizes     = log2_max_num_subframes + 1;
     s->min_samples_per_subframe  = s->samples_per_frame / s->max_num_subframes;
-    s->dynamic_range_compression = (s->decode_flags & 0x80);
-
-    s->bV3RTM = s->decode_flags & 0x100;
+    s->dynamic_range_compression = s->decode_flags & 0x80;
+    s->bV3RTM                    = s->decode_flags & 0x100;
 
     if (s->max_num_subframes > MAX_SUBFRAMES) {
         av_log(avctx, AV_LOG_ERROR, "invalid number of subframes %i\n",
@@ -419,19 +238,19 @@ static av_cold int decode_init(AVCodecContext *avctx)
 
     s->num_channels = avctx->channels;
 
-    /** extract lfe channel position */
+    /* extract lfe channel position */
     s->lfe_channel = -1;
 
     if (channel_mask & 8) {
         unsigned int mask;
-        for (mask = 1; mask < 16; mask <<= 1) {
+        for (mask = 1; mask < 16; mask <<= 1)
             if (channel_mask & mask)
                 ++s->lfe_channel;
-        }
     }
 
     if (s->num_channels < 0) {
-        av_log(avctx, AV_LOG_ERROR, "invalid number of channels %d\n", s->num_channels);
+        av_log(avctx, AV_LOG_ERROR, "invalid number of channels %d\n",
+               s->num_channels);
         return AVERROR_INVALIDDATA;
     } else if (s->num_channels > WMALL_MAX_CHANNELS) {
         av_log_ask_for_sample(avctx, "unsupported number of channels\n");
@@ -439,33 +258,30 @@ static av_cold int decode_init(AVCodecContext *avctx)
     }
 
     avcodec_get_frame_defaults(&s->frame);
-    avctx->coded_frame = &s->frame;
-
+    avctx->coded_frame    = &s->frame;
     avctx->channel_layout = channel_mask;
     return 0;
 }
 
 /**
- *@brief Decode the subframe length.
- *@param s context
- *@param offset sample offset in the frame
- *@return decoded subframe length on success, < 0 in case of an error
+ * @brief Decode the subframe length.
+ * @param s      context
+ * @param offset sample offset in the frame
+ * @return decoded subframe length on success, < 0 in case of an error
  */
 static int decode_subframe_length(WmallDecodeCtx *s, int offset)
 {
-    int frame_len_ratio;
-    int subframe_len, len;
+    int frame_len_ratio, subframe_len, len;
 
-    /** no need to read from the bitstream when only one length is possible */
+    /* no need to read from the bitstream when only one length is possible */
     if (offset == s->samples_per_frame - s->min_samples_per_subframe)
         return s->min_samples_per_subframe;
 
-    len = av_log2(s->max_num_subframes - 1) + 1;
+    len             = av_log2(s->max_num_subframes - 1) + 1;
     frame_len_ratio = get_bits(&s->gb, len);
+    subframe_len    = s->min_samples_per_subframe * (frame_len_ratio + 1);
 
-    subframe_len = s->min_samples_per_subframe * (frame_len_ratio + 1);
-
-    /** sanity check the length */
+    /* sanity check the length */
     if (subframe_len < s->min_samples_per_subframe ||
         subframe_len > s->samples_per_frame) {
         av_log(s->avctx, AV_LOG_ERROR, "broken frame: subframe_len %i\n",
@@ -476,7 +292,7 @@ static int decode_subframe_length(WmallDecodeCtx *s, int offset)
 }
 
 /**
- *@brief Decode how the data in the frame is split into subframes.
+ * @brief Decode how the data in the frame is split into subframes.
  *       Every WMA frame contains the encoded data for a fixed number of
  *       samples per channel. The data for every channel might be split
  *       into several subframes. This function will reconstruct the list of
@@ -492,38 +308,31 @@ static int decode_subframe_length(WmallDecodeCtx *s, int offset)
  *       The algorithm repeats these steps until the frame is properly divided
  *       between the individual channels.
  *
- *@param s context
- *@return 0 on success, < 0 in case of an error
+ * @param s context
+ * @return 0 on success, < 0 in case of an error
  */
 static int decode_tilehdr(WmallDecodeCtx *s)
 {
-    uint16_t num_samples[WMALL_MAX_CHANNELS];        /**< sum of samples for all currently known subframes of a channel */
-    uint8_t  contains_subframe[WMALL_MAX_CHANNELS];  /**< flag indicating if a channel contains the current subframe */
-    int channels_for_cur_subframe = s->num_channels;  /**< number of channels that contain the current subframe */
-    int fixed_channel_layout = 0;                     /**< flag indicating that all channels use the same subfra2me offsets and sizes */
-    int min_channel_len = 0;                          /**< smallest sum of samples (channels with this length will be processed first) */
-    int c;
-
-    /* Should never consume more than 3073 bits (256 iterations for the
-     * while loop when always the minimum amount of 128 samples is substracted
-     * from missing samples in the 8 channel case).
-     * 1 + BLOCK_MAX_SIZE * MAX_CHANNELS / BLOCK_MIN_SIZE * (MAX_CHANNELS  + 4)
-     */
-
-    /** reset tiling information */
+    uint16_t num_samples[WMALL_MAX_CHANNELS] = { 0 }; /* sum of samples for all currently known subframes of a channel */
+    uint8_t  contains_subframe[WMALL_MAX_CHANNELS];   /* flag indicating if a channel contains the current subframe */
+    int channels_for_cur_subframe = s->num_channels;  /* number of channels that contain the current subframe */
+    int fixed_channel_layout = 0;                     /* flag indicating that all channels use the same subfra2me offsets and sizes */
+    int min_channel_len = 0;                          /* smallest sum of samples (channels with this length will be processed first) */
+    int c, tile_aligned;
+
+    /* reset tiling information */
     for (c = 0; c < s->num_channels; c++)
         s->channel[c].num_subframes = 0;
 
-    memset(num_samples, 0, sizeof(num_samples));
-
-    if (s->max_num_subframes == 1 || get_bits1(&s->gb))
+    tile_aligned = get_bits1(&s->gb);
+    if (s->max_num_subframes == 1 || tile_aligned)
         fixed_channel_layout = 1;
 
-    /** loop until the frame data is split between the subframes */
+    /* loop until the frame data is split between the subframes */
     do {
         int subframe_len;
 
-        /** check which channels contain the subframe */
+        /* check which channels contain the subframe */
         for (c = 0; c < s->num_channels; c++) {
             if (num_samples[c] == min_channel_len) {
                 if (fixed_channel_layout || channels_for_cur_subframe == 1 ||
@@ -536,13 +345,13 @@ static int decode_tilehdr(WmallDecodeCtx *s)
                 contains_subframe[c] = 0;
         }
 
-        /** get subframe length, subframe_len == 0 is not allowed */
+        /* get subframe length, subframe_len == 0 is not allowed */
         if ((subframe_len = decode_subframe_length(s, min_channel_len)) <= 0)
             return AVERROR_INVALIDDATA;
-        /** add subframes to the individual channels and find new min_channel_len */
+        /* add subframes to the individual channels and find new min_channel_len */
         min_channel_len += subframe_len;
         for (c = 0; c < s->num_channels; c++) {
-            WmallChannelCtx* chan = &s->channel[c];
+            WmallChannelCtx *chan = &s->channel[c];
 
             if (contains_subframe[c]) {
                 if (chan->num_subframes >= MAX_SUBFRAMES) {
@@ -570,10 +379,9 @@ static int decode_tilehdr(WmallDecodeCtx *s)
     } while (min_channel_len < s->samples_per_frame);
 
     for (c = 0; c < s->num_channels; c++) {
-        int i;
-        int offset = 0;
+        int i, offset = 0;
         for (i = 0; i < s->channel[c].num_subframes; i++) {
-            s->channel[c].subframe_offset[i] = offset;
+            s->channel[c].subframe_offsets[i] = offset;
             offset += s->channel[c].subframe_len[i];
         }
     }
@@ -581,215 +389,175 @@ static int decode_tilehdr(WmallDecodeCtx *s)
     return 0;
 }
 
-
-static int my_log2(unsigned int i)
-{
-    unsigned int iLog2 = 0;
-    while ((i >> iLog2) > 1)
-        iLog2++;
-    return iLog2;
-}
-
-
-/**
- *
- */
 static void decode_ac_filter(WmallDecodeCtx *s)
 {
     int i;
-    s->acfilter_order = get_bits(&s->gb, 4) + 1;
+    s->acfilter_order   = get_bits(&s->gb, 4) + 1;
     s->acfilter_scaling = get_bits(&s->gb, 4);
 
-    for(i = 0; i < s->acfilter_order; i++) {
+    for (i = 0; i < s->acfilter_order; i++)
         s->acfilter_coeffs[i] = get_bits(&s->gb, s->acfilter_scaling) + 1;
-    }
 }
 
-
-/**
- *
- */
 static void decode_mclms(WmallDecodeCtx *s)
 {
-    s->mclms_order = (get_bits(&s->gb, 4) + 1) * 2;
+    s->mclms_order   = (get_bits(&s->gb, 4) + 1) * 2;
     s->mclms_scaling = get_bits(&s->gb, 4);
-    if(get_bits1(&s->gb)) {
-        // mclms_send_coef
-        int i;
-        int send_coef_bits;
+    if (get_bits1(&s->gb)) {
+        int i, send_coef_bits;
         int cbits = av_log2(s->mclms_scaling + 1);
         assert(cbits == my_log2(s->mclms_scaling + 1));
-        if(1 << cbits < s->mclms_scaling + 1)
+        if (1 << cbits < s->mclms_scaling + 1)
             cbits++;
 
         send_coef_bits = (cbits ? get_bits(&s->gb, cbits) : 0) + 2;
 
-        for(i = 0; i < s->mclms_order * s->num_channels * s->num_channels; i++) {
+        for (i = 0; i < s->mclms_order * s->num_channels * s->num_channels; i++)
             s->mclms_coeffs[i] = get_bits(&s->gb, send_coef_bits);
-        }
 
-        for(i = 0; i < s->num_channels; i++) {
+        for (i = 0; i < s->num_channels; i++) {
             int c;
-            for(c = 0; c < i; c++) {
+            for (c = 0; c < i; c++)
                 s->mclms_coeffs_cur[i * s->num_channels + c] = get_bits(&s->gb, send_coef_bits);
-            }
         }
     }
 }
 
-
-/**
- *
- */
 static void decode_cdlms(WmallDecodeCtx *s)
 {
     int c, i;
     int cdlms_send_coef = get_bits1(&s->gb);
 
-    for(c = 0; c < s->num_channels; c++) {
+    for (c = 0; c < s->num_channels; c++) {
         s->cdlms_ttl[c] = get_bits(&s->gb, 3) + 1;
-        for(i = 0; i < s->cdlms_ttl[c]; i++) {
+        for (i = 0; i < s->cdlms_ttl[c]; i++)
             s->cdlms[c][i].order = (get_bits(&s->gb, 7) + 1) * 8;
-        }
 
-        for(i = 0; i < s->cdlms_ttl[c]; i++) {
+        for (i = 0; i < s->cdlms_ttl[c]; i++)
             s->cdlms[c][i].scaling = get_bits(&s->gb, 4);
-        }
 
-        if(cdlms_send_coef) {
-            for(i = 0; i < s->cdlms_ttl[c]; i++) {
+        if (cdlms_send_coef) {
+            for (i = 0; i < s->cdlms_ttl[c]; i++) {
                 int cbits, shift_l, shift_r, j;
                 cbits = av_log2(s->cdlms[c][i].order);
-                if(1 << cbits < s->cdlms[c][i].order)
+                if ((1 << cbits) < s->cdlms[c][i].order)
                     cbits++;
                 s->cdlms[c][i].coefsend = get_bits(&s->gb, cbits) + 1;
 
                 cbits = av_log2(s->cdlms[c][i].scaling + 1);
-                if(1 << cbits < s->cdlms[c][i].scaling + 1)
+                if ((1 << cbits) < s->cdlms[c][i].scaling + 1)
                     cbits++;
 
                 s->cdlms[c][i].bitsend = get_bits(&s->gb, cbits) + 2;
                 shift_l = 32 - s->cdlms[c][i].bitsend;
-                shift_r = 32 - 2 - s->cdlms[c][i].scaling;
-                for(j = 0; j < s->cdlms[c][i].coefsend; j++) {
+                shift_r = 32 - s->cdlms[c][i].scaling - 2;
+                for (j = 0; j < s->cdlms[c][i].coefsend; j++)
                     s->cdlms[c][i].coefs[j] =
                         (get_bits(&s->gb, s->cdlms[c][i].bitsend) << shift_l) >> shift_r;
-                }
             }
         }
     }
 }
 
-/**
- *
- */
 static int decode_channel_residues(WmallDecodeCtx *s, int ch, int tile_size)
 {
     int i = 0;
     unsigned int ave_mean;
     s->transient[ch] = get_bits1(&s->gb);
-    if(s->transient[ch]) {
-            s->transient_pos[ch] = get_bits(&s->gb, av_log2(tile_size));
+    if (s->transient[ch]) {
+        s->transient_pos[ch] = get_bits(&s->gb, av_log2(tile_size));
         if (s->transient_pos[ch])
-                s->transient[ch] = 0;
-            s->channel[ch].transient_counter =
-                FFMAX(s->channel[ch].transient_counter, s->samples_per_frame / 2);
-        } else if (s->channel[ch].transient_counter)
-            s->transient[ch] = 1;
+            s->transient[ch] = 0;
+        s->channel[ch].transient_counter =
+            FFMAX(s->channel[ch].transient_counter, s->samples_per_frame / 2);
+    } else if (s->channel[ch].transient_counter)
+        s->transient[ch] = 1;
 
-    if(s->seekable_tile) {
+    if (s->seekable_tile) {
         ave_mean = get_bits(&s->gb, s->bits_per_sample);
         s->ave_sum[ch] = ave_mean << (s->movave_scaling + 1);
-//        s->ave_sum[ch] *= 2;
     }
 
-    if(s->seekable_tile) {
-        if(s->do_inter_ch_decorr)
+    if (s->seekable_tile) {
+        if (s->do_inter_ch_decorr)
             s->channel_residues[ch][0] = get_sbits(&s->gb, s->bits_per_sample + 1);
         else
             s->channel_residues[ch][0] = get_sbits(&s->gb, s->bits_per_sample);
         i++;
     }
-    //av_log(0, 0, "%8d: ", num_logged_tiles++);
-    for(; i < tile_size; i++) {
+    for (; i < tile_size; i++) {
         int quo = 0, rem, rem_bits, residue;
         while(get_bits1(&s->gb)) {
             quo++;
             if (get_bits_left(&s->gb) <= 0)
                 return -1;
         }
-        if(quo >= 32)
+        if (quo >= 32)
             quo += get_bits_long(&s->gb, get_bits(&s->gb, 5) + 1);
 
-    ave_mean = (s->ave_sum[ch] + (1 << s->movave_scaling)) >> (s->movave_scaling + 1);
-    if (ave_mean <= 1)
-        residue = quo;
-    else
-    {
-        rem_bits = av_ceil_log2(ave_mean);
-        rem = rem_bits ? get_bits(&s->gb, rem_bits) : 0;
-        residue = (quo << rem_bits) + rem;
-    }
+        ave_mean = (s->ave_sum[ch] + (1 << s->movave_scaling)) >> (s->movave_scaling + 1);
+        if (ave_mean <= 1)
+            residue = quo;
+        else {
+            rem_bits = av_ceil_log2(ave_mean);
+            rem      = rem_bits ? get_bits(&s->gb, rem_bits) : 0;
+            residue  = (quo << rem_bits) + rem;
+        }
 
-        s->ave_sum[ch] = residue + s->ave_sum[ch] - (s->ave_sum[ch] >> s->movave_scaling);
+        s->ave_sum[ch] = residue + s->ave_sum[ch] -
+                         (s->ave_sum[ch] >> s->movave_scaling);
 
-        if(residue & 1)
+        if (residue & 1)
             residue = -(residue >> 1) - 1;
         else
             residue = residue >> 1;
         s->channel_residues[ch][i] = residue;
     }
-    //dump_int_buffer(s->channel_residues[ch], 4, tile_size, 16);
 
     return 0;
 
 }
 
-
-/**
- *
- */
-static void
-decode_lpc(WmallDecodeCtx *s)
+static void decode_lpc(WmallDecodeCtx *s)
 {
     int ch, i, cbits;
-    s->lpc_order = get_bits(&s->gb, 5) + 1;
+    s->lpc_order   = get_bits(&s->gb, 5) + 1;
     s->lpc_scaling = get_bits(&s->gb, 4);
     s->lpc_intbits = get_bits(&s->gb, 3) + 1;
     cbits = s->lpc_scaling + s->lpc_intbits;
-    for(ch = 0; ch < s->num_channels; ch++) {
-        for(i = 0; i < s->lpc_order; i++) {
+    for (ch = 0; ch < s->num_channels; ch++)
+        for (i = 0; i < s->lpc_order; i++)
             s->lpc_coefs[ch][i] = get_sbits(&s->gb, cbits);
-        }
-    }
 }
 
-
 static void clear_codec_buffers(WmallDecodeCtx *s)
 {
     int ich, ilms;
 
-    memset(s->acfilter_coeffs    , 0, 16 * sizeof(int));
-    memset(s->acfilter_prevvalues, 0, 16 * 2 * sizeof(int)); // may be wrong
-    memset(s->lpc_coefs          , 0, 40 * 2 * sizeof(int));
+    memset(s->acfilter_coeffs,     0, sizeof(s->acfilter_coeffs));
+    memset(s->acfilter_prevvalues, 0, sizeof(s->acfilter_prevvalues));
+    memset(s->lpc_coefs,           0, sizeof(s->lpc_coefs));
 
-    memset(s->mclms_coeffs    , 0, 128 * sizeof(int16_t));
-    memset(s->mclms_coeffs_cur, 0,   4 * sizeof(int16_t));
-    memset(s->mclms_prevvalues, 0,  64 * sizeof(int));
-    memset(s->mclms_updates   , 0,  64 * sizeof(int16_t));
+    memset(s->mclms_coeffs,     0, sizeof(s->mclms_coeffs));
+    memset(s->mclms_coeffs_cur, 0, sizeof(s->mclms_coeffs_cur));
+    memset(s->mclms_prevvalues, 0, sizeof(s->mclms_prevvalues));
+    memset(s->mclms_updates,    0, sizeof(s->mclms_updates));
 
     for (ich = 0; ich < s->num_channels; ich++) {
         for (ilms = 0; ilms < s->cdlms_ttl[ich]; ilms++) {
-            memset(s->cdlms[ich][ilms].coefs         , 0, 256 * sizeof(int16_t));
-            memset(s->cdlms[ich][ilms].lms_prevvalues, 0, 512 * sizeof(int));
-            memset(s->cdlms[ich][ilms].lms_updates   , 0, 512 * sizeof(int16_t));
+            memset(s->cdlms[ich][ilms].coefs, 0,
+                   sizeof(s->cdlms[ich][ilms].coefs));
+            memset(s->cdlms[ich][ilms].lms_prevvalues, 0,
+                   sizeof(s->cdlms[ich][ilms].lms_prevvalues));
+            memset(s->cdlms[ich][ilms].lms_updates, 0,
+                   sizeof(s->cdlms[ich][ilms].lms_updates));
         }
         s->ave_sum[ich] = 0;
     }
 }
 
 /**
- *@brief Resets filter parameters and transient area at new seekable tile
+ * @brief Reset filter parameters and transient area at new seekable tile.
  */
 static void reset_codec(WmallDecodeCtx *s)
 {
@@ -799,23 +567,19 @@ static void reset_codec(WmallDecodeCtx *s)
         for (ilms = 0; ilms < s->cdlms_ttl[ich]; ilms++)
             s->cdlms[ich][ilms].recent = s->cdlms[ich][ilms].order;
         /* first sample of a seekable subframe is considered as the starting of
-           a transient area which is samples_per_frame samples long */
+            a transient area which is samples_per_frame samples long */
         s->channel[ich].transient_counter = s->samples_per_frame;
-        s->transient[ich] = 1;
+        s->transient[ich]     = 1;
         s->transient_pos[ich] = 0;
     }
 }
 
-
-
 static void mclms_update(WmallDecodeCtx *s, int icoef, int *pred)
 {
-    int i, j, ich;
-    int pred_error;
-    int order = s->mclms_order;
+    int i, j, ich, pred_error;
+    int order        = s->mclms_order;
     int num_channels = s->num_channels;
-    int range = 1 << (s->bits_per_sample - 1);
-    //int bps = s->bits_per_sample > 16 ? 4 : 2; // bytes per sample
+    int range        = 1 << (s->bits_per_sample - 1);
 
     for (ich = 0; ich < num_channels; ich++) {
         pred_error = s->channel_residues[ich][icoef] - pred[ich];
@@ -860,7 +624,7 @@ static void mclms_update(WmallDecodeCtx *s, int icoef, int *pred)
     if (s->mclms_recent == 0) {
         memcpy(&s->mclms_prevvalues[order * num_channels],
                s->mclms_prevvalues,
-               4 * order * num_channels);
+               2 * order * num_channels);
         memcpy(&s->mclms_updates[order * num_channels],
                s->mclms_updates,
                2 * order * num_channels);
@@ -871,7 +635,7 @@ static void mclms_update(WmallDecodeCtx *s, int icoef, int *pred)
 static void mclms_predict(WmallDecodeCtx *s, int icoef, int *pred)
 {
     int ich, i;
-    int order = s->mclms_order;
+    int order        = s->mclms_order;
     int num_channels = s->num_channels;
 
     for (ich = 0; ich < num_channels; ich++) {
@@ -892,7 +656,7 @@ static void mclms_predict(WmallDecodeCtx *s, int icoef, int *pred)
 
 static void revert_mclms(WmallDecodeCtx *s, int tile_size)
 {
-    int icoef, pred[WMALL_MAX_CHANNELS] = {0};
+    int icoef, pred[WMALL_MAX_CHANNELS] = { 0 };
     for (icoef = 0; icoef < tile_size; icoef++) {
         mclms_predict(s, icoef, pred);
         mclms_update(s, icoef, pred);
@@ -901,28 +665,22 @@ static void revert_mclms(WmallDecodeCtx *s, int tile_size)
 
 static int lms_predict(WmallDecodeCtx *s, int ich, int ilms)
 {
-    int pred = 0;
-    int icoef;
+    int pred = 0, icoef;
     int recent = s->cdlms[ich][ilms].recent;
 
     for (icoef = 0; icoef < s->cdlms[ich][ilms].order; icoef++)
         pred += s->cdlms[ich][ilms].coefs[icoef] *
-                    s->cdlms[ich][ilms].lms_prevvalues[icoef + recent];
+                s->cdlms[ich][ilms].lms_prevvalues[icoef + recent];
 
-    //pred += (1 << (s->cdlms[ich][ilms].scaling - 1));
-    /* XXX: Table 29 has:
-            iPred >= cdlms[iCh][ilms].scaling;
-       seems to me like a missing > */
-    //pred >>= s->cdlms[ich][ilms].scaling;
     return pred;
 }
 
-static void lms_update(WmallDecodeCtx *s, int ich, int ilms, int input, int residue)
+static void lms_update(WmallDecodeCtx *s, int ich, int ilms,
+                       int input, int residue)
 {
     int icoef;
     int recent = s->cdlms[ich][ilms].recent;
-    int range = 1 << s->bits_per_sample - 1;
-    //int bps = s->bits_per_sample > 16 ? 4 : 2; // bytes per sample
+    int range  = 1 << s->bits_per_sample - 1;
 
     if (residue < 0) {
         for (icoef = 0; icoef < s->cdlms[ich][ilms].order; icoef++)
@@ -931,18 +689,15 @@ static void lms_update(WmallDecodeCtx *s, int ich, int ilms, int input, int resi
     } else if (residue > 0) {
         for (icoef = 0; icoef < s->cdlms[ich][ilms].order; icoef++)
             s->cdlms[ich][ilms].coefs[icoef] +=
-                s->cdlms[ich][ilms].lms_updates[icoef + recent];    /* spec mistakenly
-                                                                    dropped the recent */
+                s->cdlms[ich][ilms].lms_updates[icoef + recent];
     }
 
     if (recent)
         recent--;
     else {
-        /* XXX: This memcpy()s will probably fail if a fixed 32-bit buffer is used.
-                follow kshishkov's suggestion of using a union. */
         memcpy(&s->cdlms[ich][ilms].lms_prevvalues[s->cdlms[ich][ilms].order],
                s->cdlms[ich][ilms].lms_prevvalues,
-               4 * s->cdlms[ich][ilms].order);
+               2 * s->cdlms[ich][ilms].order);
         memcpy(&s->cdlms[ich][ilms].lms_updates[s->cdlms[ich][ilms].order],
                s->cdlms[ich][ilms].lms_updates,
                2 * s->cdlms[ich][ilms].order);
@@ -957,14 +712,6 @@ static void lms_update(WmallDecodeCtx *s, int ich, int ilms, int input, int resi
     else
         s->cdlms[ich][ilms].lms_updates[recent] = s->update_speed[ich];
 
-    /* XXX: spec says:
-    cdlms[iCh][ilms].updates[iRecent + cdlms[iCh][ilms].order >> 4] >>= 2;
-    lms_updates[iCh][ilms][iRecent + cdlms[iCh][ilms].order >> 3] >>= 1;
-
-        Questions is - are cdlms[iCh][ilms].updates[] and lms_updates[][][] two
-        seperate buffers? Here I've assumed that the two are same which makes
-        more sense to me.
-    */
     s->cdlms[ich][ilms].lms_updates[recent + (s->cdlms[ich][ilms].order >> 4)] >>= 2;
     s->cdlms[ich][ilms].lms_updates[recent + (s->cdlms[ich][ilms].order >> 3)] >>= 1;
     s->cdlms[ich][ilms].recent = recent;
@@ -995,27 +742,23 @@ static void use_normal_update_speed(WmallDecodeCtx *s, int ich)
         recent = s->cdlms[ich][ilms].recent;
         if (s->update_speed[ich] == 8)
             continue;
-        if (s->bV3RTM) {
+        if (s->bV3RTM)
             for (icoef = 0; icoef < s->cdlms[ich][ilms].order; icoef++)
                 s->cdlms[ich][ilms].lms_updates[icoef + recent] /= 2;
-        } else {
+        else
             for (icoef = 0; icoef < s->cdlms[ich][ilms].order; icoef++)
                 s->cdlms[ich][ilms].lms_updates[icoef] /= 2;
-        }
     }
     s->update_speed[ich] = 8;
 }
 
-static void revert_cdlms(WmallDecodeCtx *s, int ch, int coef_begin, int coef_end)
+static void revert_cdlms(WmallDecodeCtx *s, int ch,
+                         int coef_begin, int coef_end)
 {
-    int icoef;
-    int pred;
-    int ilms, num_lms;
-    int residue, input;
+    int icoef, pred, ilms, num_lms, residue, input;
 
     num_lms = s->cdlms_ttl[ch];
     for (ilms = num_lms - 1; ilms >= 0; ilms--) {
-        //s->cdlms[ch][ilms].recent = s->cdlms[ch][ilms].order;
         for (icoef = coef_begin; icoef < coef_end; icoef++) {
             pred = 1 << (s->cdlms[ch][ilms].scaling - 1);
             residue = s->channel_residues[ch][icoef];
@@ -1029,10 +772,10 @@ static void revert_cdlms(WmallDecodeCtx *s, int ch, int coef_begin, int coef_end
 
 static void revert_inter_ch_decorr(WmallDecodeCtx *s, int tile_size)
 {
-    int icoef;
     if (s->num_channels != 2)
         return;
     else if (s->is_channel_coded[0] && s->is_channel_coded[1]) {
+        int icoef;
         for (icoef = 0; icoef < tile_size; icoef++) {
             s->channel_residues[0][icoef] -= s->channel_residues[1][icoef] >> 1;
             s->channel_residues[1][icoef] += s->channel_residues[0][icoef];
@@ -1042,12 +785,10 @@ static void revert_inter_ch_decorr(WmallDecodeCtx *s, int tile_size)
 
 static void revert_acfilter(WmallDecodeCtx *s, int tile_size)
 {
-    int ich, icoef;
-    int pred;
-    int i, j;
+    int ich, pred, i, j;
     int64_t *filter_coeffs = s->acfilter_coeffs;
-    int scaling = s->acfilter_scaling;
-    int order = s->acfilter_order;
+    int scaling            = s->acfilter_scaling;
+    int order              = s->acfilter_order;
 
     for (ich = 0; ich < s->num_channels; ich++) {
         int *prevvalues = s->acfilter_prevvalues[ich];
@@ -1074,28 +815,19 @@ static void revert_acfilter(WmallDecodeCtx *s, int tile_size)
     }
 }
 
-/**
- *@brief Decode a single subframe (block).
- *@param s codec context
- *@return 0 on success, < 0 when decoding failed
- */
 static int decode_subframe(WmallDecodeCtx *s)
 {
-    int offset = s->samples_per_frame;
-    int subframe_len = s->samples_per_frame;
-    int i, j;
-    int total_samples   = s->samples_per_frame * s->num_channels;
-    int rawpcm_tile;
-    int padding_zeroes;
+    int offset        = s->samples_per_frame;
+    int subframe_len  = s->samples_per_frame;
+    int total_samples = s->samples_per_frame * s->num_channels;
+    int i, j, rawpcm_tile, padding_zeroes;
 
     s->subframe_offset = get_bits_count(&s->gb);
 
-    /** reset channel context and find the next block offset and size
+    /* reset channel context and find the next block offset and size
         == the next block of the channel with the smallest number of
-        decoded samples
-    */
+        decoded samples */
     for (i = 0; i < s->num_channels; i++) {
-        s->channel[i].grouped = 0;
         if (offset > s->channel[i].decoded_samples) {
             offset = s->channel[i].decoded_samples;
             subframe_len =
@@ -1103,14 +835,14 @@ static int decode_subframe(WmallDecodeCtx *s)
         }
     }
 
-    /** get a list of all channels that contain the estimated block */
+    /* get a list of all channels that contain the estimated block */
     s->channels_for_cur_subframe = 0;
     for (i = 0; i < s->num_channels; i++) {
         const int cur_subframe = s->channel[i].cur_subframe;
-        /** substract already processed samples */
+        /* subtract already processed samples */
         total_samples -= s->channel[i].decoded_samples;
 
-        /** and count if there are multiple subframes that match our profile */
+        /* and count if there are multiple subframes that match our profile */
         if (offset == s->channel[i].decoded_samples &&
             subframe_len == s->channel[i].subframe_len[cur_subframe]) {
             total_samples -= s->channel[i].subframe_len[cur_subframe];
@@ -1121,95 +853,89 @@ static int decode_subframe(WmallDecodeCtx *s)
         }
     }
 
-    /** check if the frame will be complete after processing the
+    /* check if the frame will be complete after processing the
         estimated block */
     if (!total_samples)
         s->parsed_all_subframes = 1;
 
 
     s->seekable_tile = get_bits1(&s->gb);
-    if(s->seekable_tile) {
+    if (s->seekable_tile) {
         clear_codec_buffers(s);
 
         s->do_arith_coding    = get_bits1(&s->gb);
-        if(s->do_arith_coding) {
-            dprintf(s->avctx, "do_arith_coding == 1");
+        if (s->do_arith_coding) {
+            av_dlog(s->avctx, AV_LOG_DEBUG, "do_arith_coding == 1");
             abort();
         }
         s->do_ac_filter       = get_bits1(&s->gb);
         s->do_inter_ch_decorr = get_bits1(&s->gb);
         s->do_mclms           = get_bits1(&s->gb);
 
-        if(s->do_ac_filter)
+        if (s->do_ac_filter)
             decode_ac_filter(s);
 
-        if(s->do_mclms)
+        if (s->do_mclms)
             decode_mclms(s);
 
         decode_cdlms(s);
         s->movave_scaling = get_bits(&s->gb, 3);
         s->quant_stepsize = get_bits(&s->gb, 8) + 1;
 
-            reset_codec(s);
+        reset_codec(s);
     }
 
     rawpcm_tile = get_bits1(&s->gb);
 
-    for(i = 0; i < s->num_channels; i++) {
+    for (i = 0; i < s->num_channels; i++)
         s->is_channel_coded[i] = 1;
-    }
 
-    if(!rawpcm_tile) {
-
-        for(i = 0; i < s->num_channels; i++) {
+    if (!rawpcm_tile) {
+        for (i = 0; i < s->num_channels; i++)
             s->is_channel_coded[i] = get_bits1(&s->gb);
-        }
 
-        if(s->bV3RTM) {
+        if (s->bV3RTM) {
             // LPC
             s->do_lpc = get_bits1(&s->gb);
-            if(s->do_lpc) {
+            if (s->do_lpc) {
                 decode_lpc(s);
+                av_log_ask_for_sample(s->avctx, "Inverse LPC filter not "
+                                      "implemented. Expect wrong output.\n");
             }
-        } else {
+        } else
             s->do_lpc = 0;
-        }
     }
 
 
-    if(get_bits1(&s->gb)) {
+    if (get_bits1(&s->gb))
         padding_zeroes = get_bits(&s->gb, 5);
-    } else {
+    else
         padding_zeroes = 0;
-    }
-
-    if(rawpcm_tile) {
 
+    if (rawpcm_tile) {
         int bits = s->bits_per_sample - padding_zeroes;
-        dprintf(s->avctx, "RAWPCM %d bits per sample. total %d bits, remain=%d\n", bits,
+        av_dlog(s->avctx, AV_LOG_DEBUG, "RAWPCM %d bits per sample. "
+                "total %d bits, remain=%d\n", bits,
                 bits * s->num_channels * subframe_len, get_bits_count(&s->gb));
-        for(i = 0; i < s->num_channels; i++) {
-            for(j = 0; j < subframe_len; j++) {
+        for (i = 0; i < s->num_channels; i++)
+            for (j = 0; j < subframe_len; j++)
                 s->channel_coeffs[i][j] = get_sbits(&s->gb, bits);
-//                dprintf(s->avctx, "PCM[%d][%d] = 0x%04x\n", i, j, s->channel_coeffs[i][j]);
-            }
-        }
     } else {
-        for(i = 0; i < s->num_channels; i++)
-            if(s->is_channel_coded[i]) {
-            decode_channel_residues(s, i, subframe_len);
-            if (s->seekable_tile)
-                use_high_update_speed(s, i);
-            else
-                use_normal_update_speed(s, i);
-            revert_cdlms(s, i, 0, subframe_len);
-        }
+        for (i = 0; i < s->num_channels; i++)
+            if (s->is_channel_coded[i]) {
+                decode_channel_residues(s, i, subframe_len);
+                if (s->seekable_tile)
+                    use_high_update_speed(s, i);
+                else
+                    use_normal_update_speed(s, i);
+                revert_cdlms(s, i, 0, subframe_len);
+            }
     }
     if (s->do_mclms)
         revert_mclms(s, subframe_len);
     if (s->do_inter_ch_decorr)
         revert_inter_ch_decorr(s, subframe_len);
-    if(s->do_ac_filter)
+    if (s->do_ac_filter)
         revert_acfilter(s, subframe_len);
 
     /* Dequantize */
@@ -1218,7 +944,7 @@ static int decode_subframe(WmallDecodeCtx *s)
             for (j = 0; j < subframe_len; j++)
                 s->channel_residues[i][j] *= s->quant_stepsize;
 
-    // Write to proper output buffer depending on bit-depth
+    /* Write to proper output buffer depending on bit-depth */
     for (i = 0; i < subframe_len; i++)
         for (j = 0; j < s->num_channels; j++) {
             if (s->bits_per_sample == 16)
@@ -1227,8 +953,7 @@ static int decode_subframe(WmallDecodeCtx *s)
                 *s->samples_32++ = s->channel_residues[j][i];
         }
 
-    /** handled one subframe */
-
+    /* handled one subframe */
     for (i = 0; i < s->channels_for_cur_subframe; i++) {
         int c = s->channel_indexes_for_cur_subframe[i];
         if (s->channel[c].cur_subframe >= s->channel[c].num_subframes) {
@@ -1237,77 +962,72 @@ static int decode_subframe(WmallDecodeCtx *s)
         }
         ++s->channel[c].cur_subframe;
     }
-    num_logged_subframes++;
     return 0;
 }
 
 /**
- *@brief Decode one WMA frame.
- *@param s codec context
- *@return 0 if the trailer bit indicates that this is the last frame,
- *        1 if there are additional frames
+ * @brief Decode one WMA frame.
+ * @param s codec context
+ * @return 0 if the trailer bit indicates that this is the last frame,
+ *         1 if there are additional frames
  */
 static int decode_frame(WmallDecodeCtx *s)
 {
     GetBitContext* gb = &s->gb;
-    int more_frames = 0;
-    int len = 0;
-    int i, ret;
+    int more_frames = 0, len = 0, i, ret;
 
     s->frame.nb_samples = s->samples_per_frame;
     if ((ret = s->avctx->get_buffer(s->avctx, &s->frame)) < 0) {
-        /** return an error if no frame could be decoded at all */
+        /* return an error if no frame could be decoded at all */
         av_log(s->avctx, AV_LOG_ERROR,
                "not enough space for the output samples\n");
         s->packet_loss = 1;
-        return 0;
+        return ret;
     }
     s->samples_16 = (int16_t *)s->frame.data[0];
     s->samples_32 = (int32_t *)s->frame.data[0];
 
-    /** get frame length */
+    /* get frame length */
     if (s->len_prefix)
         len = get_bits(gb, s->log2_frame_size);
 
-    /** decode tile information */
+    /* decode tile information */
     if (decode_tilehdr(s)) {
         s->packet_loss = 1;
         return 0;
     }
 
-    /** read drc info */
-    if (s->dynamic_range_compression) {
+    /* read drc info */
+    if (s->dynamic_range_compression)
         s->drc_gain = get_bits(gb, 8);
-    }
 
-    /** no idea what these are for, might be the number of samples
-        that need to be skipped at the beginning or end of a stream */
+    /* no idea what these are for, might be the number of samples
+       that need to be skipped at the beginning or end of a stream */
     if (get_bits1(gb)) {
         int skip;
 
-        /** usually true for the first frame */
+        /* usually true for the first frame */
         if (get_bits1(gb)) {
             skip = get_bits(gb, av_log2(s->samples_per_frame * 2));
-            dprintf(s->avctx, "start skip: %i\n", skip);
+            av_dlog(s->avctx, AV_LOG_DEBUG, "start skip: %i\n", skip);
         }
 
-        /** sometimes true for the last frame */
+        /* sometimes true for the last frame */
         if (get_bits1(gb)) {
             skip = get_bits(gb, av_log2(s->samples_per_frame * 2));
-            dprintf(s->avctx, "end skip: %i\n", skip);
+            av_dlog(s->avctx, AV_LOG_DEBUG, "end skip: %i\n", skip);
         }
 
     }
 
-    /** reset subframe states */
+    /* reset subframe states */
     s->parsed_all_subframes = 0;
     for (i = 0; i < s->num_channels; i++) {
         s->channel[i].decoded_samples = 0;
         s->channel[i].cur_subframe    = 0;
-        s->channel[i].reuse_sf        = 0;
     }
 
-    /** decode all subframes */
+    /* decode all subframes */
     while (!s->parsed_all_subframes) {
         if (decode_subframe(s) < 0) {
             s->packet_loss = 1;
@@ -1315,15 +1035,14 @@ static int decode_frame(WmallDecodeCtx *s)
         }
     }
 
-    dprintf(s->avctx, "Frame done\n");
+    av_dlog(s->avctx, AV_LOG_DEBUG, "Frame done\n");
 
-    if (s->skip_frame) {
+    if (s->skip_frame)
         s->skip_frame = 0;
-    }
 
     if (s->len_prefix) {
         if (len != (get_bits_count(gb) - s->frame_offset) + 2) {
-            /** FIXME: not sure if this is always an error */
+            /* FIXME: not sure if this is always an error */
             av_log(s->avctx, AV_LOG_ERROR,
                    "frame[%i] would have to skip %i bits\n", s->frame_num,
                    len - (get_bits_count(gb) - s->frame_offset) - 1);
@@ -1331,27 +1050,21 @@ static int decode_frame(WmallDecodeCtx *s)
             return 0;
         }
 
-        /** skip the rest of the frame data */
+        /* skip the rest of the frame data */
         skip_bits_long(gb, len - (get_bits_count(gb) - s->frame_offset) - 1);
-    } else {
-/*
-        while (get_bits_count(gb) < s->num_saved_bits && get_bits1(gb) == 0) {
-            dprintf(s->avctx, "skip1\n");
-        }
-*/
     }
 
-    /** decode trailer bit */
+    /* decode trailer bit */
     more_frames = get_bits1(gb);
     ++s->frame_num;
     return more_frames;
 }
 
 /**
- *@brief Calculate remaining input buffer length.
- *@param s codec context
- *@param gb bitstream reader context
- *@return remaining size in bits
+ * @brief Calculate remaining input buffer length.
+ * @param s  codec context
+ * @param gb bitstream reader context
+ * @return remaining size in bits
  */
 static int remaining_bits(WmallDecodeCtx *s, GetBitContext *gb)
 {
@@ -1359,23 +1072,24 @@ static int remaining_bits(WmallDecodeCtx *s, GetBitContext *gb)
 }
 
 /**
- *@brief Fill the bit reservoir with a (partial) frame.
- *@param s codec context
- *@param gb bitstream reader context
- *@param len length of the partial frame
- *@param append decides wether to reset the buffer or not
+ * @brief Fill the bit reservoir with a (partial) frame.
+ * @param s      codec context
+ * @param gb     bitstream reader context
+ * @param len    length of the partial frame
+ * @param append decides whether to reset the buffer or not
  */
 static void save_bits(WmallDecodeCtx *s, GetBitContext* gb, int len,
                       int append)
 {
     int buflen;
+    PutBitContext tmp;
 
-    /** when the frame data does not need to be concatenated, the input buffer
-        is resetted and additional bits from the previous frame are copyed
+    /* when the frame data does not need to be concatenated, the input buffer
+        is reset and additional bits from the previous frame are copied
         and skipped later so that a fast byte copy is possible */
 
     if (!append) {
-        s->frame_offset = get_bits_count(gb) & 7;
+        s->frame_offset   = get_bits_count(gb) & 7;
         s->num_saved_bits = s->frame_offset;
         init_put_bits(&s->pb, s->frame_data, MAX_FRAMESIZE);
     }
@@ -1391,7 +1105,7 @@ static void save_bits(WmallDecodeCtx *s, GetBitContext* gb, int len,
     s->num_saved_bits += len;
     if (!append) {
         avpriv_copy_bits(&s->pb, gb->buffer + (get_bits_count(gb) >> 3),
-                     s->num_saved_bits);
+                         s->num_saved_bits);
     } else {
         int align = 8 - (get_bits_count(gb) & 7);
         align = FFMIN(align, len);
@@ -1401,57 +1115,52 @@ static void save_bits(WmallDecodeCtx *s, GetBitContext* gb, int len,
     }
     skip_bits_long(gb, len);
 
-    {
-        PutBitContext tmp = s->pb;
-        flush_put_bits(&tmp);
-    }
+    tmp = s->pb;
+    flush_put_bits(&tmp);
 
     init_get_bits(&s->gb, s->frame_data, s->num_saved_bits);
     skip_bits(&s->gb, s->frame_offset);
 }
 
 /**
- *@brief Decode a single WMA packet.
- *@param avctx codec context
- *@param data the output buffer
- *@param data_size number of bytes that were written to the output buffer
- *@param avpkt input packet
- *@return number of bytes that were read from the input buffer
+ * @brief Decode a single WMA packet.
+ * @param avctx     codec context
+ * @param data      the output buffer
+ * @param data_size number of bytes that were written to the output buffer
+ * @param avpkt     input packet
+ * @return number of bytes that were read from the input buffer
  */
-static int decode_packet(AVCodecContext *avctx,
-                         void *data, int *got_frame_ptr, AVPacket* avpkt)
+static int decode_packet(AVCodecContext *avctx, void *data, int *got_frame_ptr,
+                         AVPacket* avpkt)
 {
     WmallDecodeCtx *s = avctx->priv_data;
     GetBitContext* gb  = &s->pgb;
     const uint8_t* buf = avpkt->data;
     int buf_size       = avpkt->size;
-    int num_bits_prev_frame;
-    int packet_sequence_number;
-    int seekable_frame_in_packet;
-    int spliced_packet;
+    int num_bits_prev_frame, packet_sequence_number,
+        seekable_frame_in_packet, spliced_packet;
 
     if (s->packet_done || s->packet_loss) {
-        int seekable_frame_in_packet, spliced_packet;
         s->packet_done = 0;
 
-        /** sanity check for the buffer length */
+        /* sanity check for the buffer length */
         if (buf_size < avctx->block_align)
             return 0;
 
         s->next_packet_start = buf_size - avctx->block_align;
-        buf_size = avctx->block_align;
-        s->buf_bit_size = buf_size << 3;
+        buf_size             = avctx->block_align;
+        s->buf_bit_size      = buf_size << 3;
 
-        /** parse packet header */
+        /* parse packet header */
         init_get_bits(gb, buf, s->buf_bit_size);
-        packet_sequence_number = get_bits(gb, 4);
+        packet_sequence_number   = get_bits(gb, 4);
         seekable_frame_in_packet = get_bits1(gb);
-        spliced_packet = get_bits1(gb);
+        spliced_packet           = get_bits1(gb);
 
-        /** get number of bits that need to be added to the previous frame */
+        /* get number of bits that need to be added to the previous frame */
         num_bits_prev_frame = get_bits(gb, s->log2_frame_size);
 
-        /** check for packet loss */
+        /* check for packet loss */
         if (!s->packet_loss &&
             ((s->packet_sequence_number + 1) & 0xF) != packet_sequence_number) {
             s->packet_loss = 1;
@@ -1467,24 +1176,23 @@ static int decode_packet(AVCodecContext *avctx,
                 s->packet_done = 1;
             }
 
-            /** append the previous frame data to the remaining data from the
-                previous packet to create a full frame */
+            /* Append the previous frame data to the remaining data from the
+             * previous packet to create a full frame. */
             save_bits(s, gb, num_bits_prev_frame, 1);
 
-            /** decode the cross packet frame if it is valid */
+            /* decode the cross packet frame if it is valid */
             if (!s->packet_loss)
                 decode_frame(s);
         } else if (s->num_saved_bits - s->frame_offset) {
-            dprintf(avctx, "ignoring %x previously saved bits\n",
+            av_dlog(avctx, AV_LOG_DEBUG, "ignoring %x previously saved bits\n",
                     s->num_saved_bits - s->frame_offset);
         }
 
         if (s->packet_loss) {
-            /** reset number of saved bits so that the decoder
-                does not start to decode incomplete frames in the
-                s->len_prefix == 0 case */
+            /* Reset number of saved bits so that the decoder does not start
+             * to decode incomplete frames in the s->len_prefix == 0 case. */
             s->num_saved_bits = 0;
-            s->packet_loss = 0;
+            s->packet_loss    = 0;
         }
 
     } else {
@@ -1501,13 +1209,12 @@ static int decode_packet(AVCodecContext *avctx,
             s->packet_done = !decode_frame(s);
         } else if (!s->len_prefix
                    && s->num_saved_bits > get_bits_count(&s->gb)) {
-            /** when the frames do not have a length prefix, we don't know
-                the compressed length of the individual frames
-                however, we know what part of a new packet belongs to the
-                previous frame
-                therefore we save the incoming packet first, then we append
-                the "previous frame" data from the next packet so that
-                we get a buffer that only contains full frames */
+            /* when the frames do not have a length prefix, we don't know the
+             * compressed length of the individual frames however, we know what
+             * part of a new packet belongs to the previous frame therefore we
+             * save the incoming packet first, then we append the "previous
+             * frame" data from the next packet so that we get a buffer that
+             * only contains full frames */
             s->packet_done = !decode_frame(s);
         } else {
             s->packet_done = 1;
@@ -1516,47 +1223,26 @@ static int decode_packet(AVCodecContext *avctx,
 
     if (s->packet_done && !s->packet_loss &&
         remaining_bits(s, gb) > 0) {
-        /** save the rest of the data so that it can be decoded
-            with the next packet */
+        /* save the rest of the data so that it can be decoded
+         * with the next packet */
         save_bits(s, gb, remaining_bits(s, gb), 0);
     }
 
     *(AVFrame *)data = s->frame;
-    *got_frame_ptr = 1;
+    *got_frame_ptr   = 1;
     s->packet_offset = get_bits_count(gb) & 7;
 
     return (s->packet_loss) ? AVERROR_INVALIDDATA : get_bits_count(gb) >> 3;
 }
 
-/**
- *@brief Clear decoder buffers (for seeking).
- *@param avctx codec context
- */
-static void flush(AVCodecContext *avctx)
-{
-    WmallDecodeCtx *s = avctx->priv_data;
-    int i;
-    /** reset output buffer as a part of it is used during the windowing of a
-        new frame */
-    for (i = 0; i < s->num_channels; i++)
-        memset(s->channel[i].out, 0, s->samples_per_frame *
-               sizeof(*s->channel[i].out));
-    s->packet_loss = 1;
-}
-
 
-/**
- *@brief wmall decoder
- */
 AVCodec ff_wmalossless_decoder = {
     .name           = "wmalossless",
     .type           = AVMEDIA_TYPE_AUDIO,
     .id             = CODEC_ID_WMALOSSLESS,
     .priv_data_size = sizeof(WmallDecodeCtx),
     .init           = decode_init,
-    .close          = decode_end,
     .decode         = decode_packet,
-    .flush          = flush,
-    .capabilities = CODEC_CAP_SUBFRAMES | CODEC_CAP_EXPERIMENTAL | CODEC_CAP_DR1,
-    .long_name = NULL_IF_CONFIG_SMALL("Windows Media Audio 9 Lossless"),
+    .capabilities   = CODEC_CAP_SUBFRAMES | CODEC_CAP_DR1,
+    .long_name      = NULL_IF_CONFIG_SMALL("Windows Media Audio Lossless"),
 };
diff --git a/libavcodec/wnv1.c b/libavcodec/wnv1.c
index 6429a5b748..5ce5e9db51 100644
--- a/libavcodec/wnv1.c
+++ b/libavcodec/wnv1.c
@@ -64,7 +64,7 @@ static int decode_frame(AVCodecContext *avctx,
     const uint8_t *buf = avpkt->data;
     int buf_size = avpkt->size;
     WNV1Context * const l = avctx->priv_data;
-    AVFrame * const p= (AVFrame*)&l->pic;
+    AVFrame * const p = &l->pic;
     unsigned char *Y,*U,*V;
     int i, j;
     int prev_y = 0, prev_u = 0, prev_v = 0;
diff --git a/libavcodec/x86/h264_qpel_10bit.asm b/libavcodec/x86/h264_qpel_10bit.asm
index 51412e3977..bdacf9f472 100644
--- a/libavcodec/x86/h264_qpel_10bit.asm
+++ b/libavcodec/x86/h264_qpel_10bit.asm
@@ -619,7 +619,7 @@ MC MC33
 %define PAD 12
 %define COUNT 2
 %else
-%define PAD 0
+%define PAD 4
 %define COUNT 3
 %endif
 put_hv%2_10_%1:
diff --git a/libavcodec/x86/vp8dsp-init.c b/libavcodec/x86/vp8dsp-init.c
index a75fdf5bc5..e3b727d1b1 100644
--- a/libavcodec/x86/vp8dsp-init.c
+++ b/libavcodec/x86/vp8dsp-init.c
@@ -29,98 +29,98 @@
 /*
  * MC functions
  */
-extern void ff_put_vp8_epel4_h4_mmxext(uint8_t *dst, int dststride,
-                                       uint8_t *src, int srcstride,
+extern void ff_put_vp8_epel4_h4_mmxext(uint8_t *dst, ptrdiff_t dststride,
+                                       uint8_t *src, ptrdiff_t srcstride,
                                        int height, int mx, int my);
-extern void ff_put_vp8_epel4_h6_mmxext(uint8_t *dst, int dststride,
-                                       uint8_t *src, int srcstride,
+extern void ff_put_vp8_epel4_h6_mmxext(uint8_t *dst, ptrdiff_t dststride,
+                                       uint8_t *src, ptrdiff_t srcstride,
                                        int height, int mx, int my);
-extern void ff_put_vp8_epel4_v4_mmxext(uint8_t *dst, int dststride,
-                                       uint8_t *src, int srcstride,
+extern void ff_put_vp8_epel4_v4_mmxext(uint8_t *dst, ptrdiff_t dststride,
+                                       uint8_t *src, ptrdiff_t srcstride,
                                        int height, int mx, int my);
-extern void ff_put_vp8_epel4_v6_mmxext(uint8_t *dst, int dststride,
-                                       uint8_t *src, int srcstride,
+extern void ff_put_vp8_epel4_v6_mmxext(uint8_t *dst, ptrdiff_t dststride,
+                                       uint8_t *src, ptrdiff_t srcstride,
                                        int height, int mx, int my);
 
-extern void ff_put_vp8_epel8_h4_sse2  (uint8_t *dst, int dststride,
-                                       uint8_t *src, int srcstride,
+extern void ff_put_vp8_epel8_h4_sse2  (uint8_t *dst, ptrdiff_t dststride,
+                                       uint8_t *src, ptrdiff_t srcstride,
                                        int height, int mx, int my);
-extern void ff_put_vp8_epel8_h6_sse2  (uint8_t *dst, int dststride,
-                                       uint8_t *src, int srcstride,
+extern void ff_put_vp8_epel8_h6_sse2  (uint8_t *dst, ptrdiff_t dststride,
+                                       uint8_t *src, ptrdiff_t srcstride,
                                        int height, int mx, int my);
-extern void ff_put_vp8_epel8_v4_sse2  (uint8_t *dst, int dststride,
-                                       uint8_t *src, int srcstride,
+extern void ff_put_vp8_epel8_v4_sse2  (uint8_t *dst, ptrdiff_t dststride,
+                                       uint8_t *src, ptrdiff_t srcstride,
                                        int height, int mx, int my);
-extern void ff_put_vp8_epel8_v6_sse2  (uint8_t *dst, int dststride,
-                                       uint8_t *src, int srcstride,
+extern void ff_put_vp8_epel8_v6_sse2  (uint8_t *dst, ptrdiff_t dststride,
+                                       uint8_t *src, ptrdiff_t srcstride,
                                        int height, int mx, int my);
 
-extern void ff_put_vp8_epel4_h4_ssse3 (uint8_t *dst, int dststride,
-                                       uint8_t *src, int srcstride,
+extern void ff_put_vp8_epel4_h4_ssse3 (uint8_t *dst, ptrdiff_t dststride,
+                                       uint8_t *src, ptrdiff_t srcstride,
                                        int height, int mx, int my);
-extern void ff_put_vp8_epel4_h6_ssse3 (uint8_t *dst, int dststride,
-                                       uint8_t *src, int srcstride,
+extern void ff_put_vp8_epel4_h6_ssse3 (uint8_t *dst, ptrdiff_t dststride,
+                                       uint8_t *src, ptrdiff_t srcstride,
                                        int height, int mx, int my);
-extern void ff_put_vp8_epel4_v4_ssse3 (uint8_t *dst, int dststride,
-                                       uint8_t *src, int srcstride,
+extern void ff_put_vp8_epel4_v4_ssse3 (uint8_t *dst, ptrdiff_t dststride,
+                                       uint8_t *src, ptrdiff_t srcstride,
                                        int height, int mx, int my);
-extern void ff_put_vp8_epel4_v6_ssse3 (uint8_t *dst, int dststride,
-                                       uint8_t *src, int srcstride,
+extern void ff_put_vp8_epel4_v6_ssse3 (uint8_t *dst, ptrdiff_t dststride,
+                                       uint8_t *src, ptrdiff_t srcstride,
                                        int height, int mx, int my);
-extern void ff_put_vp8_epel8_h4_ssse3 (uint8_t *dst, int dststride,
-                                       uint8_t *src, int srcstride,
+extern void ff_put_vp8_epel8_h4_ssse3 (uint8_t *dst, ptrdiff_t dststride,
+                                       uint8_t *src, ptrdiff_t srcstride,
                                        int height, int mx, int my);
-extern void ff_put_vp8_epel8_h6_ssse3 (uint8_t *dst, int dststride,
-                                       uint8_t *src, int srcstride,
+extern void ff_put_vp8_epel8_h6_ssse3 (uint8_t *dst, ptrdiff_t dststride,
+                                       uint8_t *src, ptrdiff_t srcstride,
                                        int height, int mx, int my);
-extern void ff_put_vp8_epel8_v4_ssse3 (uint8_t *dst, int dststride,
-                                       uint8_t *src, int srcstride,
+extern void ff_put_vp8_epel8_v4_ssse3 (uint8_t *dst, ptrdiff_t dststride,
+                                       uint8_t *src, ptrdiff_t srcstride,
                                        int height, int mx, int my);
-extern void ff_put_vp8_epel8_v6_ssse3 (uint8_t *dst, int dststride,
-                                       uint8_t *src, int srcstride,
+extern void ff_put_vp8_epel8_v6_ssse3 (uint8_t *dst, ptrdiff_t dststride,
+                                       uint8_t *src, ptrdiff_t srcstride,
                                        int height, int mx, int my);
 
-extern void ff_put_vp8_bilinear4_h_mmxext(uint8_t *dst, int dststride,
-                                          uint8_t *src, int srcstride,
+extern void ff_put_vp8_bilinear4_h_mmxext(uint8_t *dst, ptrdiff_t dststride,
+                                          uint8_t *src, ptrdiff_t srcstride,
                                           int height, int mx, int my);
-extern void ff_put_vp8_bilinear8_h_sse2  (uint8_t *dst, int dststride,
-                                          uint8_t *src, int srcstride,
+extern void ff_put_vp8_bilinear8_h_sse2  (uint8_t *dst, ptrdiff_t dststride,
+                                          uint8_t *src, ptrdiff_t srcstride,
                                           int height, int mx, int my);
-extern void ff_put_vp8_bilinear4_h_ssse3 (uint8_t *dst, int dststride,
-                                          uint8_t *src, int srcstride,
+extern void ff_put_vp8_bilinear4_h_ssse3 (uint8_t *dst, ptrdiff_t dststride,
+                                          uint8_t *src, ptrdiff_t srcstride,
                                           int height, int mx, int my);
-extern void ff_put_vp8_bilinear8_h_ssse3 (uint8_t *dst, int dststride,
-                                          uint8_t *src, int srcstride,
+extern void ff_put_vp8_bilinear8_h_ssse3 (uint8_t *dst, ptrdiff_t dststride,
+                                          uint8_t *src, ptrdiff_t srcstride,
                                           int height, int mx, int my);
 
-extern void ff_put_vp8_bilinear4_v_mmxext(uint8_t *dst, int dststride,
-                                          uint8_t *src, int srcstride,
+extern void ff_put_vp8_bilinear4_v_mmxext(uint8_t *dst, ptrdiff_t dststride,
+                                          uint8_t *src, ptrdiff_t srcstride,
                                           int height, int mx, int my);
-extern void ff_put_vp8_bilinear8_v_sse2  (uint8_t *dst, int dststride,
-                                          uint8_t *src, int srcstride,
+extern void ff_put_vp8_bilinear8_v_sse2  (uint8_t *dst, ptrdiff_t dststride,
+                                          uint8_t *src, ptrdiff_t srcstride,
                                           int height, int mx, int my);
-extern void ff_put_vp8_bilinear4_v_ssse3 (uint8_t *dst, int dststride,
-                                          uint8_t *src, int srcstride,
+extern void ff_put_vp8_bilinear4_v_ssse3 (uint8_t *dst, ptrdiff_t dststride,
+                                          uint8_t *src, ptrdiff_t srcstride,
                                           int height, int mx, int my);
-extern void ff_put_vp8_bilinear8_v_ssse3 (uint8_t *dst, int dststride,
-                                          uint8_t *src, int srcstride,
+extern void ff_put_vp8_bilinear8_v_ssse3 (uint8_t *dst, ptrdiff_t dststride,
+                                          uint8_t *src, ptrdiff_t srcstride,
                                           int height, int mx, int my);
 
 
-extern void ff_put_vp8_pixels8_mmx (uint8_t *dst, int dststride,
-                                    uint8_t *src, int srcstride,
+extern void ff_put_vp8_pixels8_mmx (uint8_t *dst, ptrdiff_t dststride,
+                                    uint8_t *src, ptrdiff_t srcstride,
                                     int height, int mx, int my);
-extern void ff_put_vp8_pixels16_mmx(uint8_t *dst, int dststride,
-                                    uint8_t *src, int srcstride,
+extern void ff_put_vp8_pixels16_mmx(uint8_t *dst, ptrdiff_t dststride,
+                                    uint8_t *src, ptrdiff_t srcstride,
                                     int height, int mx, int my);
-extern void ff_put_vp8_pixels16_sse(uint8_t *dst, int dststride,
-                                    uint8_t *src, int srcstride,
+extern void ff_put_vp8_pixels16_sse(uint8_t *dst, ptrdiff_t dststride,
+                                    uint8_t *src, ptrdiff_t srcstride,
                                     int height, int mx, int my);
 
 #define TAP_W16(OPT, FILTERTYPE, TAPTYPE) \
 static void ff_put_vp8_ ## FILTERTYPE ## 16_ ## TAPTYPE ## _ ## OPT( \
-    uint8_t *dst,  int dststride, uint8_t *src, \
-    int srcstride, int height, int mx, int my) \
+    uint8_t *dst,  ptrdiff_t dststride, uint8_t *src, \
+    ptrdiff_t srcstride, int height, int mx, int my) \
 { \
     ff_put_vp8_ ## FILTERTYPE ## 8_ ## TAPTYPE ## _ ## OPT( \
         dst,     dststride, src,     srcstride, height, mx, my); \
@@ -129,8 +129,8 @@ static void ff_put_vp8_ ## FILTERTYPE ## 16_ ## TAPTYPE ## _ ## OPT( \
 }
 #define TAP_W8(OPT, FILTERTYPE, TAPTYPE) \
 static void ff_put_vp8_ ## FILTERTYPE ## 8_ ## TAPTYPE ## _ ## OPT( \
-    uint8_t *dst,  int dststride, uint8_t *src, \
-    int srcstride, int height, int mx, int my) \
+    uint8_t *dst,  ptrdiff_t dststride, uint8_t *src, \
+    ptrdiff_t srcstride, int height, int mx, int my) \
 { \
     ff_put_vp8_ ## FILTERTYPE ## 4_ ## TAPTYPE ## _ ## OPT( \
         dst,     dststride, src,     srcstride, height, mx, my); \
@@ -138,6 +138,7 @@ static void ff_put_vp8_ ## FILTERTYPE ## 8_ ## TAPTYPE ## _ ## OPT( \
         dst + 4, dststride, src + 4, srcstride, height, mx, my); \
 }
 
+#if ARCH_X86_32
 TAP_W8 (mmxext, epel, h4)
 TAP_W8 (mmxext, epel, h6)
 TAP_W16(mmxext, epel, h6)
@@ -148,6 +149,7 @@ TAP_W8 (mmxext, bilinear, h)
 TAP_W16(mmxext, bilinear, h)
 TAP_W8 (mmxext, bilinear, v)
 TAP_W16(mmxext, bilinear, v)
+#endif
 
 TAP_W16(sse2,   epel, h6)
 TAP_W16(sse2,   epel, v6)
@@ -161,8 +163,8 @@ TAP_W16(ssse3,  bilinear, v)
 
 #define HVTAP(OPT, ALIGN, TAPNUMX, TAPNUMY, SIZE, MAXHEIGHT) \
 static void ff_put_vp8_epel ## SIZE ## _h ## TAPNUMX ## v ## TAPNUMY ## _ ## OPT( \
-    uint8_t *dst, int dststride, uint8_t *src, \
-    int srcstride, int height, int mx, int my) \
+    uint8_t *dst, ptrdiff_t dststride, uint8_t *src, \
+    ptrdiff_t srcstride, int height, int mx, int my) \
 { \
     DECLARE_ALIGNED(ALIGN, uint8_t, tmp)[SIZE * (MAXHEIGHT + TAPNUMY - 1)]; \
     uint8_t *tmpptr = tmp + SIZE * (TAPNUMY / 2 - 1); \
@@ -173,15 +175,21 @@ static void ff_put_vp8_epel ## SIZE ## _h ## TAPNUMX ## v ## TAPNUMY ## _ ## OPT
         dst, dststride, tmpptr, SIZE,      height,               mx, my); \
 }
 
+#if ARCH_X86_32
 #define HVTAPMMX(x, y) \
 HVTAP(mmxext, 8, x, y,  4,  8) \
 HVTAP(mmxext, 8, x, y,  8, 16)
 
+HVTAP(mmxext, 8, 6, 6, 16, 16)
+#else
+#define HVTAPMMX(x, y) \
+HVTAP(mmxext, 8, x, y,  4,  8)
+#endif
+
 HVTAPMMX(4, 4)
 HVTAPMMX(4, 6)
 HVTAPMMX(6, 4)
 HVTAPMMX(6, 6)
-HVTAP(mmxext, 8, 6, 6, 16, 16)
 
 #define HVTAPSSE2(x, y, w) \
 HVTAP(sse2,  16, x, y, w, 16) \
@@ -200,8 +208,8 @@ HVTAP(ssse3, 16, 6, 6, 4, 8)
 
 #define HVBILIN(OPT, ALIGN, SIZE, MAXHEIGHT) \
 static void ff_put_vp8_bilinear ## SIZE ## _hv_ ## OPT( \
-    uint8_t *dst, int dststride, uint8_t *src, \
-    int srcstride, int height, int mx, int my) \
+    uint8_t *dst, ptrdiff_t dststride, uint8_t *src, \
+    ptrdiff_t srcstride, int height, int mx, int my) \
 { \
     DECLARE_ALIGNED(ALIGN, uint8_t, tmp)[SIZE * (MAXHEIGHT + 2)]; \
     ff_put_vp8_bilinear ## SIZE ## _h_ ## OPT( \
@@ -211,43 +219,68 @@ static void ff_put_vp8_bilinear ## SIZE ## _hv_ ## OPT( \
 }
 
 HVBILIN(mmxext, 8,  4,  8)
+#if ARCH_X86_32
 HVBILIN(mmxext, 8,  8, 16)
 HVBILIN(mmxext, 8, 16, 16)
+#endif
 HVBILIN(sse2,   8,  8, 16)
 HVBILIN(sse2,   8, 16, 16)
 HVBILIN(ssse3,  8,  4,  8)
 HVBILIN(ssse3,  8,  8, 16)
 HVBILIN(ssse3,  8, 16, 16)
 
-extern void ff_vp8_idct_dc_add_mmx(uint8_t *dst, DCTELEM block[16], int stride);
-extern void ff_vp8_idct_dc_add_sse4(uint8_t *dst, DCTELEM block[16], int stride);
-extern void ff_vp8_idct_dc_add4y_mmx(uint8_t *dst, DCTELEM block[4][16], int stride);
-extern void ff_vp8_idct_dc_add4y_sse2(uint8_t *dst, DCTELEM block[4][16], int stride);
-extern void ff_vp8_idct_dc_add4uv_mmx(uint8_t *dst, DCTELEM block[2][16], int stride);
+extern void ff_vp8_idct_dc_add_mmx(uint8_t *dst, DCTELEM block[16],
+                                   ptrdiff_t stride);
+extern void ff_vp8_idct_dc_add_sse4(uint8_t *dst, DCTELEM block[16],
+                                    ptrdiff_t stride);
+extern void ff_vp8_idct_dc_add4y_mmx(uint8_t *dst, DCTELEM block[4][16],
+                                      ptrdiff_t stride);
+extern void ff_vp8_idct_dc_add4y_sse2(uint8_t *dst, DCTELEM block[4][16],
+                                      ptrdiff_t stride);
+extern void ff_vp8_idct_dc_add4uv_mmx(uint8_t *dst, DCTELEM block[2][16],
+                                      ptrdiff_t stride);
 extern void ff_vp8_luma_dc_wht_mmx(DCTELEM block[4][4][16], DCTELEM dc[16]);
 extern void ff_vp8_luma_dc_wht_sse(DCTELEM block[4][4][16], DCTELEM dc[16]);
-extern void ff_vp8_idct_add_mmx(uint8_t *dst, DCTELEM block[16], int stride);
-extern void ff_vp8_idct_add_sse(uint8_t *dst, DCTELEM block[16], int stride);
+extern void ff_vp8_idct_add_mmx(uint8_t *dst, DCTELEM block[16],
+                                ptrdiff_t stride);
+extern void ff_vp8_idct_add_sse(uint8_t *dst, DCTELEM block[16],
+                                ptrdiff_t stride);
 
 #define DECLARE_LOOP_FILTER(NAME)\
-extern void ff_vp8_v_loop_filter_simple_ ## NAME(uint8_t *dst, int stride, int flim);\
-extern void ff_vp8_h_loop_filter_simple_ ## NAME(uint8_t *dst, int stride, int flim);\
-extern void ff_vp8_v_loop_filter16y_inner_ ## NAME (uint8_t *dst, int stride,\
+extern void ff_vp8_v_loop_filter_simple_ ## NAME(uint8_t *dst, \
+                                                 ptrdiff_t stride, \
+                                                 int flim);\
+extern void ff_vp8_h_loop_filter_simple_ ## NAME(uint8_t *dst, \
+                                                 ptrdiff_t stride, \
+                                                 int flim);\
+extern void ff_vp8_v_loop_filter16y_inner_ ## NAME (uint8_t *dst, \
+                                                     ptrdiff_t stride,\
+                                                    int e, int i, int hvt);\
+extern void ff_vp8_h_loop_filter16y_inner_ ## NAME (uint8_t *dst, \
+                                                    ptrdiff_t stride,\
+                                                    int e, int i, int hvt);\
+extern void ff_vp8_v_loop_filter8uv_inner_ ## NAME (uint8_t *dstU, \
+                                                    uint8_t *dstV,\
+                                                    ptrdiff_t s, \
+                                                    int e, int i, int hvt);\
+extern void ff_vp8_h_loop_filter8uv_inner_ ## NAME (uint8_t *dstU, \
+                                                    uint8_t *dstV,\
+                                                    ptrdiff_t s, \
                                                     int e, int i, int hvt);\
-extern void ff_vp8_h_loop_filter16y_inner_ ## NAME (uint8_t *dst, int stride,\
+extern void ff_vp8_v_loop_filter16y_mbedge_ ## NAME(uint8_t *dst, \
+                                                    ptrdiff_t stride,\
                                                     int e, int i, int hvt);\
-extern void ff_vp8_v_loop_filter8uv_inner_ ## NAME (uint8_t *dstU, uint8_t *dstV,\
-                                                    int s, int e, int i, int hvt);\
-extern void ff_vp8_h_loop_filter8uv_inner_ ## NAME (uint8_t *dstU, uint8_t *dstV,\
-                                                    int s, int e, int i, int hvt);\
-extern void ff_vp8_v_loop_filter16y_mbedge_ ## NAME(uint8_t *dst, int stride,\
+extern void ff_vp8_h_loop_filter16y_mbedge_ ## NAME(uint8_t *dst, \
+                                                    ptrdiff_t stride,\
                                                     int e, int i, int hvt);\
-extern void ff_vp8_h_loop_filter16y_mbedge_ ## NAME(uint8_t *dst, int stride,\
+extern void ff_vp8_v_loop_filter8uv_mbedge_ ## NAME(uint8_t *dstU, \
+                                                    uint8_t *dstV,\
+                                                    ptrdiff_t s, \
                                                     int e, int i, int hvt);\
-extern void ff_vp8_v_loop_filter8uv_mbedge_ ## NAME(uint8_t *dstU, uint8_t *dstV,\
-                                                    int s, int e, int i, int hvt);\
-extern void ff_vp8_h_loop_filter8uv_mbedge_ ## NAME(uint8_t *dstU, uint8_t *dstV,\
-                                                    int s, int e, int i, int hvt);
+extern void ff_vp8_h_loop_filter8uv_mbedge_ ## NAME(uint8_t *dstU, \
+                                                    uint8_t *dstV,\
+                                                    ptrdiff_t s, \
+                                                    int e, int i, int hvt);
 
 DECLARE_LOOP_FILTER(mmx)
 DECLARE_LOOP_FILTER(mmxext)
@@ -288,15 +321,18 @@ av_cold void ff_vp8dsp_init_x86(VP8DSPContext* c)
 
     if (mm_flags & AV_CPU_FLAG_MMX) {
         c->vp8_idct_dc_add    = ff_vp8_idct_dc_add_mmx;
-        c->vp8_idct_dc_add4y  = ff_vp8_idct_dc_add4y_mmx;
         c->vp8_idct_dc_add4uv = ff_vp8_idct_dc_add4uv_mmx;
+#if ARCH_X86_32
+        c->vp8_idct_dc_add4y  = ff_vp8_idct_dc_add4y_mmx;
         c->vp8_idct_add       = ff_vp8_idct_add_mmx;
         c->vp8_luma_dc_wht    = ff_vp8_luma_dc_wht_mmx;
         c->put_vp8_epel_pixels_tab[0][0][0]     =
         c->put_vp8_bilinear_pixels_tab[0][0][0] = ff_put_vp8_pixels16_mmx;
+#endif
         c->put_vp8_epel_pixels_tab[1][0][0]     =
         c->put_vp8_bilinear_pixels_tab[1][0][0] = ff_put_vp8_pixels8_mmx;
 
+#if ARCH_X86_32
         c->vp8_v_loop_filter_simple = ff_vp8_v_loop_filter_simple_mmx;
         c->vp8_h_loop_filter_simple = ff_vp8_h_loop_filter_simple_mmx;
 
@@ -309,17 +345,19 @@ av_cold void ff_vp8dsp_init_x86(VP8DSPContext* c)
         c->vp8_h_loop_filter16y       = ff_vp8_h_loop_filter16y_mbedge_mmx;
         c->vp8_v_loop_filter8uv       = ff_vp8_v_loop_filter8uv_mbedge_mmx;
         c->vp8_h_loop_filter8uv       = ff_vp8_h_loop_filter8uv_mbedge_mmx;
+#endif
     }
 
     /* note that 4-tap width=16 functions are missing because w=16
      * is only used for luma, and luma is always a copy or sixtap. */
     if (mm_flags & AV_CPU_FLAG_MMX2) {
+        VP8_MC_FUNC(2, 4, mmxext);
+        VP8_BILINEAR_MC_FUNC(2, 4, mmxext);
+#if ARCH_X86_32
         VP8_LUMA_MC_FUNC(0, 16, mmxext);
         VP8_MC_FUNC(1, 8, mmxext);
-        VP8_MC_FUNC(2, 4, mmxext);
         VP8_BILINEAR_MC_FUNC(0, 16, mmxext);
         VP8_BILINEAR_MC_FUNC(1, 8, mmxext);
-        VP8_BILINEAR_MC_FUNC(2, 4, mmxext);
 
         c->vp8_v_loop_filter_simple = ff_vp8_v_loop_filter_simple_mmxext;
         c->vp8_h_loop_filter_simple = ff_vp8_h_loop_filter_simple_mmxext;
@@ -333,6 +371,7 @@ av_cold void ff_vp8dsp_init_x86(VP8DSPContext* c)
         c->vp8_h_loop_filter16y       = ff_vp8_h_loop_filter16y_mbedge_mmxext;
         c->vp8_v_loop_filter8uv       = ff_vp8_v_loop_filter8uv_mbedge_mmxext;
         c->vp8_h_loop_filter8uv       = ff_vp8_h_loop_filter8uv_mbedge_mmxext;
+#endif
     }
 
     if (mm_flags & AV_CPU_FLAG_SSE) {
diff --git a/libavcodec/x86/vp8dsp.asm b/libavcodec/x86/vp8dsp.asm
index 833c88a1a0..140597031f 100644
--- a/libavcodec/x86/vp8dsp.asm
+++ b/libavcodec/x86/vp8dsp.asm
@@ -865,6 +865,7 @@ cglobal put_vp8_pixels8_mmx, 5,5
     jg .nextrow
     REP_RET
 
+%if ARCH_X86_32
 cglobal put_vp8_pixels16_mmx, 5,5
 .nextrow:
     movq  mm0, [r2+r3*0+0]
@@ -880,6 +881,7 @@ cglobal put_vp8_pixels16_mmx, 5,5
     sub   r4d, 2
     jg .nextrow
     REP_RET
+%endif
 
 cglobal put_vp8_pixels16_sse, 5,5,2
 .nextrow:
@@ -973,6 +975,7 @@ cglobal vp8_idct_dc_add_sse4, 3, 3, 6
 ; void vp8_idct_dc_add4y_<opt>(uint8_t *dst, DCTELEM block[4][16], int stride);
 ;-----------------------------------------------------------------------------
 
+%if ARCH_X86_32
 INIT_MMX
 cglobal vp8_idct_dc_add4y_mmx, 3, 3
     ; load data
@@ -1007,6 +1010,7 @@ cglobal vp8_idct_dc_add4y_mmx, 3, 3
     ADD_DC    m0, m6, 0, mova
     ADD_DC    m1, m7, 8, mova
     RET
+%endif
 
 INIT_XMM
 cglobal vp8_idct_dc_add4y_sse2, 3, 3, 6
@@ -1152,7 +1156,9 @@ cglobal vp8_idct_add_%1, 3, 3
     RET
 %endmacro
 
+%if ARCH_X86_32
 VP8_IDCT_ADD mmx
+%endif
 VP8_IDCT_ADD sse
 
 ;-----------------------------------------------------------------------------
@@ -1217,7 +1223,9 @@ cglobal vp8_luma_dc_wht_%1, 2,3
 %endmacro
 
 INIT_MMX
+%if ARCH_X86_32
 VP8_DC_WHT mmx
+%endif
 VP8_DC_WHT sse
 
 ;-----------------------------------------------------------------------------
@@ -1610,6 +1618,7 @@ cglobal vp8_%2_loop_filter_simple_%1, 3, %3, %4
 %endif
 %endmacro
 
+%if ARCH_X86_32
 INIT_MMX
 %define SPLATB_REG SPLATB_REG_MMX
 SIMPLE_LOOPFILTER mmx,    v, 4, 0
@@ -1617,6 +1626,8 @@ SIMPLE_LOOPFILTER mmx,    h, 5, 0
 %define SPLATB_REG SPLATB_REG_MMXEXT
 SIMPLE_LOOPFILTER mmxext, v, 4, 0
 SIMPLE_LOOPFILTER mmxext, h, 5, 0
+%endif
+
 INIT_XMM
 %define SPLATB_REG SPLATB_REG_SSE2
 %define WRITE_8W   WRITE_8W_SSE2
@@ -2118,6 +2129,7 @@ cglobal vp8_%2_loop_filter16y_inner_%1, 5, %3, %5
     RET
 %endmacro
 
+%if ARCH_X86_32
 INIT_MMX
 %define SPLATB_REG SPLATB_REG_MMX
 INNER_LOOPFILTER mmx,    v, 6, 16, 0
@@ -2130,6 +2142,7 @@ INNER_LOOPFILTER mmxext, v, 6, 16, 0
 INNER_LOOPFILTER mmxext, h, 6, 16, 0
 INNER_LOOPFILTER mmxext, v, 6,  8, 0
 INNER_LOOPFILTER mmxext, h, 6,  8, 0
+%endif
 
 INIT_XMM
 %define SPLATB_REG SPLATB_REG_SSE2
@@ -2814,6 +2827,7 @@ cglobal vp8_%2_loop_filter16y_mbedge_%1, 5, %3, %5
     RET
 %endmacro
 
+%if ARCH_X86_32
 INIT_MMX
 %define SPLATB_REG SPLATB_REG_MMX
 MBEDGE_LOOPFILTER mmx,    v, 6, 16, 0
@@ -2826,6 +2840,7 @@ MBEDGE_LOOPFILTER mmxext, v, 6, 16, 0
 MBEDGE_LOOPFILTER mmxext, h, 6, 16, 0
 MBEDGE_LOOPFILTER mmxext, v, 6,  8, 0
 MBEDGE_LOOPFILTER mmxext, h, 6,  8, 0
+%endif
 
 INIT_XMM
 %define SPLATB_REG SPLATB_REG_SSE2
diff --git a/libavcodec/xl.c b/libavcodec/xl.c
index 78f34afa4e..91a0df0683 100644
--- a/libavcodec/xl.c
+++ b/libavcodec/xl.c
@@ -45,7 +45,7 @@ static int decode_frame(AVCodecContext *avctx,
     const uint8_t *buf = avpkt->data;
     int buf_size = avpkt->size;
     VideoXLContext * const a = avctx->priv_data;
-    AVFrame * const p= (AVFrame*)&a->pic;
+    AVFrame * const p = &a->pic;
     uint8_t *Y, *U, *V;
     int i, j;
     int stride;
diff --git a/libavcodec/zmbvenc.c b/libavcodec/zmbvenc.c
index abb34a9bdb..a7fdc280a5 100644
--- a/libavcodec/zmbvenc.c
+++ b/libavcodec/zmbvenc.c
@@ -313,7 +313,7 @@ static av_cold int encode_init(AVCodecContext *avctx)
         return -1;
     }
 
-    avctx->coded_frame = (AVFrame*)&c->pic;
+    avctx->coded_frame = &c->pic;
 
     return 0;
 }
author	Michael Niedermayer <michaelni@gmx.at>	2012-03-03 00:09:47 +0100
committer	Michael Niedermayer <michaelni@gmx.at>	2012-03-03 00:23:10 +0100
commit	268098d8b2a6e3dd84be788a2cd6fda10f7b3e71 (patch)
tree	afd9e17a980920b66edb38331e5a608308da711d /libavcodec
parent	689f65126be8a55e8a1e706cb56b19bb975c20ce (diff)
parent	9d87374ec0f382c8394ad511243db6980afa42af (diff)
download	ffmpeg-268098d8b2a6e3dd84be788a2cd6fda10f7b3e71.tar.gz