Merge remote-tracking branch 'qatar/master'

* qatar/master:
  cljr: K&R cosmetics
  cljr: return a more sensible value when encountering invalid headers
  cljr: drop unnecessary emms_c() calls without MMX code
  cljr: remove useless casts
  cljr: group encode/decode parts under single ifdefs
  cljr: remove stray semicolon
  cljr: add missing return statement in decode_end()
  doc: add pulseaudio to the input list
  avconv: remove unsubstantiated comment
  shorten: avoid abort() on unknown audio types
  cljr: add encoder
  build: merge lists of HTML documentation targets
  tests/examples: Mark some variables only used within their files as static.
  tests/tools/examples: Replace direct exit() calls by return.
  x86 cpuid: set vendor union members separately
  cljr: release picture at end of decoding
  rv40: NEON optimised rv40 qpel motion compensation

Conflicts:
	doc/examples/muxing.c
	libavcodec/cljr.c
	libavcodec/version.h

Merged-by: Michael Niedermayer <michaelni@gmx.at>

19 files changed, 849 insertions, 119 deletions

diff --git a/Changelog b/Changelog
index 6ab3e84b19..e863a086ad 100644
--- a/Changelog
+++ b/Changelog
@@ -130,6 +130,7 @@ easier to use. The changes are:
 - Microsoft Windows ICO demuxer
 - life source
 - PCM format support in OMA demuxer
+- CLJR encoder
 
 
 version 0.8:
diff --git a/avconv.c b/avconv.c
index d629e3f430..d0c9aabaf4 100644
--- a/avconv.c
+++ b/avconv.c
@@ -643,7 +643,7 @@ void exit_program(int ret)
         exit (255);
     }
 
-    exit(ret); /* not all OS-es handle main() return value */
+    exit(ret);
 }
 
 static void assert_avoptions(AVDictionary *m)
diff --git a/doc/Makefile b/doc/Makefile
index 01960b74b0..ca403d5bf7 100644
--- a/doc/Makefile
+++ b/doc/Makefile
@@ -1,8 +1,12 @@
 MANPAGES    = $(PROGS-yes:%=doc/%.1)
 PODPAGES    = $(PROGS-yes:%=doc/%.pod)
-HTMLPAGES   = $(PROGS-yes:%=doc/%.html)
+HTMLPAGES   = $(PROGS-yes:%=doc/%.html)                                 \
+              doc/developer.html                                        \
+              doc/faq.html                                              \
+              doc/general.html                                          \
+              doc/libavfilter.html                                      \
 
-DOCS = $(addprefix doc/, developer.html faq.html general.html libavfilter.html) $(HTMLPAGES) $(MANPAGES) $(PODPAGES)
+DOCS = $(HTMLPAGES) $(MANPAGES) $(PODPAGES)
 
 all-$(CONFIG_DOC): documentation
 
diff --git a/doc/examples/muxing.c b/doc/examples/muxing.c
index 6f88517be4..e72bfacc49 100644
--- a/doc/examples/muxing.c
+++ b/doc/examples/muxing.c
@@ -50,11 +50,11 @@ static int sws_flags = SWS_BICUBIC;
 /**************************************************************/
 /* audio output */
 
-float t, tincr, tincr2;
-int16_t *samples;
-uint8_t *audio_outbuf;
-int audio_outbuf_size;
-int audio_input_frame_size;
+static float t, tincr, tincr2;
+static int16_t *samples;
+static uint8_t *audio_outbuf;
+static int audio_outbuf_size;
+static int audio_input_frame_size;
 
 /*
  * add an audio output stream
@@ -190,9 +190,9 @@ static void close_audio(AVFormatContext *oc, AVStream *st)
 /**************************************************************/
 /* video output */
 
-AVFrame *picture, *tmp_picture;
-uint8_t *video_outbuf;
-int frame_count, video_outbuf_size;
+static AVFrame *picture, *tmp_picture;
+static uint8_t *video_outbuf;
+static int frame_count, video_outbuf_size;
 
 /* add a video output stream */
 static AVStream *add_video_stream(AVFormatContext *oc, enum CodecID codec_id)
@@ -454,7 +454,7 @@ int main(int argc, char **argv)
                "The output format is automatically guessed according to the file extension.\n"
                "Raw images can also be output by using '%%d' in the filename\n"
                "\n", argv[0]);
-        exit(1);
+        return 1;
     }
 
     filename = argv[1];
@@ -466,7 +466,7 @@ int main(int argc, char **argv)
         avformat_alloc_output_context2(&oc, NULL, "mpeg", filename);
     }
     if (!oc) {
-        exit(1);
+        return 1;
     }
     fmt = oc->oformat;
 
@@ -494,7 +494,7 @@ int main(int argc, char **argv)
     if (!(fmt->flags & AVFMT_NOFILE)) {
         if (avio_open(&oc->pb, filename, AVIO_FLAG_WRITE) < 0) {
             fprintf(stderr, "Could not open '%s'\n", filename);
-            exit(1);
+            return 1;
         }
     }
 
diff --git a/doc/general.texi b/doc/general.texi
index 5883e84109..660553d404 100644
--- a/doc/general.texi
+++ b/doc/general.texi
@@ -401,7 +401,7 @@ following image formats are supported:
     @tab Codec used in Delphine Software International games.
 @item Discworld II BMV Video @tab     @tab  X
 @item Cinepak                @tab     @tab  X
-@item Cirrus Logic AccuPak   @tab     @tab  X
+@item Cirrus Logic AccuPak   @tab  X  @tab  X
     @tab fourcc: CLJR
 @item Creative YUV (CYUV)    @tab     @tab  X
 @item DFA                    @tab     @tab  X
@@ -788,6 +788,7 @@ performance on systems without hardware floating point support).
 @item JACK              @tab X      @tab
 @item LIBDC1394         @tab X      @tab
 @item OSS               @tab X      @tab X
+@item Pulseaudio        @tab X      @tab
 @item Video4Linux       @tab X      @tab
 @item Video4Linux2      @tab X      @tab
 @item VfW capture       @tab X      @tab
diff --git a/libavcodec/allcodecs.c b/libavcodec/allcodecs.c
index 02df6f20ab..19afbd4895 100644
--- a/libavcodec/allcodecs.c
+++ b/libavcodec/allcodecs.c
@@ -89,7 +89,7 @@ void avcodec_register_all(void)
     REGISTER_DECODER (CAVS, cavs);
     REGISTER_DECODER (CDGRAPHICS, cdgraphics);
     REGISTER_DECODER (CINEPAK, cinepak);
-    REGISTER_DECODER (CLJR, cljr);
+    REGISTER_ENCDEC  (CLJR, cljr);
     REGISTER_DECODER (CSCD, cscd);
     REGISTER_DECODER (CYUV, cyuv);
     REGISTER_DECODER (DFA, dfa);
diff --git a/libavcodec/arm/rv40dsp_init_neon.c b/libavcodec/arm/rv40dsp_init_neon.c
index 3a863e1916..36d75e6fd8 100644
--- a/libavcodec/arm/rv40dsp_init_neon.c
+++ b/libavcodec/arm/rv40dsp_init_neon.c
@@ -23,6 +23,28 @@
 #include "libavcodec/avcodec.h"
 #include "libavcodec/rv34dsp.h"
 
+#define DECL_QPEL3(type, w, pos) \
+    void ff_##type##_rv40_qpel##w##_mc##pos##_neon(uint8_t *dst, uint8_t *src,\
+                                                   int stride)
+#define DECL_QPEL2(w, pos)                      \
+    DECL_QPEL3(put, w, pos);                    \
+    DECL_QPEL3(avg, w, pos)
+
+#define DECL_QPEL_XY(x, y)                      \
+    DECL_QPEL2(16, x ## y);                     \
+    DECL_QPEL2(8,  x ## y)
+
+#define DECL_QPEL_Y(y)                          \
+    DECL_QPEL_XY(0, y);                         \
+    DECL_QPEL_XY(1, y);                         \
+    DECL_QPEL_XY(2, y);                         \
+    DECL_QPEL_XY(3, y);                         \
+
+DECL_QPEL_Y(0);
+DECL_QPEL_Y(1);
+DECL_QPEL_Y(2);
+DECL_QPEL_Y(3);
+
 void ff_put_rv40_chroma_mc8_neon(uint8_t *, uint8_t *, int, int, int, int);
 void ff_put_rv40_chroma_mc4_neon(uint8_t *, uint8_t *, int, int, int, int);
 
@@ -34,6 +56,59 @@ void ff_rv40_weight_func_8_neon(uint8_t *, uint8_t *, uint8_t *, int, int, int);
 
 void ff_rv40dsp_init_neon(RV34DSPContext *c, DSPContext* dsp)
 {
+    c->put_pixels_tab[0][ 1] = ff_put_rv40_qpel16_mc10_neon;
+    c->put_pixels_tab[0][ 3] = ff_put_rv40_qpel16_mc30_neon;
+    c->put_pixels_tab[0][ 4] = ff_put_rv40_qpel16_mc01_neon;
+    c->put_pixels_tab[0][ 5] = ff_put_rv40_qpel16_mc11_neon;
+    c->put_pixels_tab[0][ 6] = ff_put_rv40_qpel16_mc21_neon;
+    c->put_pixels_tab[0][ 7] = ff_put_rv40_qpel16_mc31_neon;
+    c->put_pixels_tab[0][ 9] = ff_put_rv40_qpel16_mc12_neon;
+    c->put_pixels_tab[0][10] = ff_put_rv40_qpel16_mc22_neon;
+    c->put_pixels_tab[0][11] = ff_put_rv40_qpel16_mc32_neon;
+    c->put_pixels_tab[0][12] = ff_put_rv40_qpel16_mc03_neon;
+    c->put_pixels_tab[0][13] = ff_put_rv40_qpel16_mc13_neon;
+    c->put_pixels_tab[0][14] = ff_put_rv40_qpel16_mc23_neon;
+    c->put_pixels_tab[0][15] = ff_put_rv40_qpel16_mc33_neon;
+    c->avg_pixels_tab[0][ 1] = ff_avg_rv40_qpel16_mc10_neon;
+    c->avg_pixels_tab[0][ 3] = ff_avg_rv40_qpel16_mc30_neon;
+    c->avg_pixels_tab[0][ 4] = ff_avg_rv40_qpel16_mc01_neon;
+    c->avg_pixels_tab[0][ 5] = ff_avg_rv40_qpel16_mc11_neon;
+    c->avg_pixels_tab[0][ 6] = ff_avg_rv40_qpel16_mc21_neon;
+    c->avg_pixels_tab[0][ 7] = ff_avg_rv40_qpel16_mc31_neon;
+    c->avg_pixels_tab[0][ 9] = ff_avg_rv40_qpel16_mc12_neon;
+    c->avg_pixels_tab[0][10] = ff_avg_rv40_qpel16_mc22_neon;
+    c->avg_pixels_tab[0][11] = ff_avg_rv40_qpel16_mc32_neon;
+    c->avg_pixels_tab[0][12] = ff_avg_rv40_qpel16_mc03_neon;
+    c->avg_pixels_tab[0][13] = ff_avg_rv40_qpel16_mc13_neon;
+    c->avg_pixels_tab[0][14] = ff_avg_rv40_qpel16_mc23_neon;
+    c->avg_pixels_tab[0][15] = ff_avg_rv40_qpel16_mc33_neon;
+    c->put_pixels_tab[1][ 1] = ff_put_rv40_qpel8_mc10_neon;
+    c->put_pixels_tab[1][ 3] = ff_put_rv40_qpel8_mc30_neon;
+    c->put_pixels_tab[1][ 4] = ff_put_rv40_qpel8_mc01_neon;
+    c->put_pixels_tab[1][ 5] = ff_put_rv40_qpel8_mc11_neon;
+    c->put_pixels_tab[1][ 6] = ff_put_rv40_qpel8_mc21_neon;
+    c->put_pixels_tab[1][ 7] = ff_put_rv40_qpel8_mc31_neon;
+    c->put_pixels_tab[1][ 9] = ff_put_rv40_qpel8_mc12_neon;
+    c->put_pixels_tab[1][10] = ff_put_rv40_qpel8_mc22_neon;
+    c->put_pixels_tab[1][11] = ff_put_rv40_qpel8_mc32_neon;
+    c->put_pixels_tab[1][12] = ff_put_rv40_qpel8_mc03_neon;
+    c->put_pixels_tab[1][13] = ff_put_rv40_qpel8_mc13_neon;
+    c->put_pixels_tab[1][14] = ff_put_rv40_qpel8_mc23_neon;
+    c->put_pixels_tab[1][15] = ff_put_rv40_qpel8_mc33_neon;
+    c->avg_pixels_tab[1][ 1] = ff_avg_rv40_qpel8_mc10_neon;
+    c->avg_pixels_tab[1][ 3] = ff_avg_rv40_qpel8_mc30_neon;
+    c->avg_pixels_tab[1][ 4] = ff_avg_rv40_qpel8_mc01_neon;
+    c->avg_pixels_tab[1][ 5] = ff_avg_rv40_qpel8_mc11_neon;
+    c->avg_pixels_tab[1][ 6] = ff_avg_rv40_qpel8_mc21_neon;
+    c->avg_pixels_tab[1][ 7] = ff_avg_rv40_qpel8_mc31_neon;
+    c->avg_pixels_tab[1][ 9] = ff_avg_rv40_qpel8_mc12_neon;
+    c->avg_pixels_tab[1][10] = ff_avg_rv40_qpel8_mc22_neon;
+    c->avg_pixels_tab[1][11] = ff_avg_rv40_qpel8_mc32_neon;
+    c->avg_pixels_tab[1][12] = ff_avg_rv40_qpel8_mc03_neon;
+    c->avg_pixels_tab[1][13] = ff_avg_rv40_qpel8_mc13_neon;
+    c->avg_pixels_tab[1][14] = ff_avg_rv40_qpel8_mc23_neon;
+    c->avg_pixels_tab[1][15] = ff_avg_rv40_qpel8_mc33_neon;
+
     c->put_chroma_pixels_tab[0] = ff_put_rv40_chroma_mc8_neon;
     c->put_chroma_pixels_tab[1] = ff_put_rv40_chroma_mc4_neon;
     c->avg_chroma_pixels_tab[0] = ff_avg_rv40_chroma_mc8_neon;
diff --git a/libavcodec/arm/rv40dsp_neon.S b/libavcodec/arm/rv40dsp_neon.S
index cafd98add0..07ba8428c1 100644
--- a/libavcodec/arm/rv40dsp_neon.S
+++ b/libavcodec/arm/rv40dsp_neon.S
@@ -1,5 +1,6 @@
 /*
  * Copyright (c) 2011 Janne Grunau <janne-libav@jannau.net>
+ * Copyright (c) 2011 Mans Rullgard <mans@mansr.com>
  *
  * This file is part of Libav.
  *
@@ -19,6 +20,644 @@
  */
 
 #include "asm.S"
+#include "neon.S"
+
+.macro  qpel_lowpass    r0,  r1,  rc1, rc2, shift
+        vext.8          d25, \r0, \r1, #1       @ src[-1]
+        vext.8          d26, \r0, \r1, #4       @ src[ 2]
+        vext.8          d24, \r0, \r1, #5       @ src[ 3]
+        vaddl.u8        q9,  d25, d26
+        vaddl.u8        q8,  \r0, d24
+        vext.8          d27, \r0, \r1, #2       @ src[ 0]
+        vshl.s16        q12, q9,  #2
+        vsub.s16        q8,  q8,  q9
+        vext.8          d28, \r0, \r1, #3       @ src[ 1]
+        vsub.s16        q8,  q8,  q12
+        vmlal.u8        q8,  d27, \rc1
+        vmlal.u8        q8,  d28, \rc2
+        vqrshrun.s16    \r0, q8,  #\shift
+.endm
+
+.macro  qpel_lowpass_x2 r0,  r1,  r2,  r3,  rc1, rc2, shift
+        vext.8          d25, \r0, \r1, #1       @ src[-1]
+        vext.8          d26, \r0, \r1, #4       @ src[ 2]
+        vext.8          d24, \r0, \r1, #5       @ src[ 3]
+        vaddl.u8        q9,  d25, d26
+        vaddl.u8        q8,  \r0, d24
+        vext.8          d29, \r0, \r1, #2       @ src[ 0]
+        vext.8          d28, \r0, \r1, #3       @ src[ 1]
+        vshl.s16        q10, q9,  #2
+        vext.8          \r1, \r2, \r3, #1       @ src[-1]
+        vsub.s16        q8,  q8,  q9
+        vext.8          d22, \r2, \r3, #4       @ src[ 2]
+        vext.8          \r0, \r2, \r3, #5       @ src[ 3]
+        vaddl.u8        q13, \r1, d22
+        vaddl.u8        q12, \r2, \r0
+        vsub.s16        q8,  q8,  q10
+        vshl.s16        q9,  q13, #2
+        vsub.s16        q12, q12, q13
+        vmlal.u8        q8,  d29, \rc1
+        vmlal.u8        q8,  d28, \rc2
+        vsub.s16        q12, q12, q9
+        vext.8          d26, \r2, \r3, #2       @ src[ 0]
+        vext.8          d27, \r2, \r3, #3       @ src[ 1]
+        vmlal.u8        q12, d26, \rc1
+        vmlal.u8        q12, d27, \rc2
+        vqrshrun.s16    \r0, q8,  #\shift
+        vqrshrun.s16    \r2, q12, #\shift
+.endm
+
+.macro  rv40_qpel8_h    shift
+function put_rv40_qpel8_h_lp_packed_s\shift\()_neon
+1:
+        vld1.8          {q2},     [r1], r2
+        vld1.8          {q3},     [r1], r2
+        qpel_lowpass_x2 d4,  d5,  d6,  d7,  d0,  d1,  \shift
+        vst1.8          {d4},     [r12,:64]!
+        vst1.8          {d6},     [r12,:64]!
+        subs            r3,  r3,  #2
+        bgt             1b
+        vld1.8          {q2},     [r1]
+        qpel_lowpass    d4,  d5,  d0,  d1,  \shift
+        vst1.8          {d4},     [r12,:64]!
+        bx              lr
+endfunc
+.endm
+
+.macro  rv40_qpel8_v    shift, type
+function \type\()_rv40_qpel8_v_lp_packed_s\shift\()_neon
+        vld1.64         {d2},     [r1,:64]!
+        vld1.64         {d3},     [r1,:64]!
+        vld1.64         {d4},     [r1,:64]!
+        vld1.64         {d5},     [r1,:64]!
+        vld1.64         {d6},     [r1,:64]!
+        vld1.64         {d7},     [r1,:64]!
+        vld1.64         {d8},     [r1,:64]!
+        vld1.64         {d9},     [r1,:64]!
+        vld1.64         {d10},    [r1,:64]!
+        vld1.64         {d11},    [r1,:64]!
+        vld1.64         {d12},    [r1,:64]!
+        vld1.64         {d13},    [r1,:64]!
+        vld1.64         {d14},    [r1,:64]!
+        transpose_8x8   d2,  d3,  d4,  d5,  d6,  d7,  d8,  d9
+        transpose_8x8   d10, d11, d12, d13, d14, d15, d30, d31
+        qpel_lowpass_x2 d2,  d10, d3,  d11, d0,  d1,  \shift
+        qpel_lowpass_x2 d4,  d12, d5,  d13, d0,  d1,  \shift
+        qpel_lowpass_x2 d6,  d14, d7,  d15, d0,  d1,  \shift
+        qpel_lowpass_x2 d8,  d30, d9,  d31, d0,  d1,  \shift
+        transpose_8x8   d2,  d3,  d4,  d5,  d6,  d7,  d8,  d9
+  .ifc \type,avg
+        vld1.64         d12,      [r0,:64], r2
+        vld1.64         d13,      [r0,:64], r2
+        vld1.64         d14,      [r0,:64], r2
+        vld1.64         d15,      [r0,:64], r2
+        vld1.64         d16,      [r0,:64], r2
+        vld1.64         d17,      [r0,:64], r2
+        vld1.64         d18,      [r0,:64], r2
+        vld1.64         d19,      [r0,:64], r2
+        sub             r0,  r0,  r2,  lsl #3
+        vrhadd.u8       q1,  q1,  q6
+        vrhadd.u8       q2,  q2,  q7
+        vrhadd.u8       q3,  q3,  q8
+        vrhadd.u8       q4,  q4,  q9
+  .endif
+        vst1.64         d2,       [r0,:64], r2
+        vst1.64         d3,       [r0,:64], r2
+        vst1.64         d4,       [r0,:64], r2
+        vst1.64         d5,       [r0,:64], r2
+        vst1.64         d6,       [r0,:64], r2
+        vst1.64         d7,       [r0,:64], r2
+        vst1.64         d8,       [r0,:64], r2
+        vst1.64         d9,       [r0,:64], r2
+        bx              lr
+endfunc
+.endm
+
+        rv40_qpel8_h    5
+        rv40_qpel8_h    6
+
+.macro  rv40_qpel       type
+function \type\()_rv40_qpel8_h_lowpass_neon
+  .ifc \type,avg
+        mov             r12, r0
+  .endif
+1:
+        vld1.8          {q2},     [r1], r2
+        vld1.8          {q3},     [r1], r2
+        qpel_lowpass_x2 d4,  d5,  d6,  d7,  d0,  d1,  6
+  .ifc \type,avg
+        vld1.8          {d3},     [r12,:64], r2
+        vld1.8          {d16},    [r12,:64], r2
+        vrhadd.u8       d4,  d4,  d3
+        vrhadd.u8       d6,  d6,  d16
+  .endif
+        vst1.8          {d4},     [r0,:64], r2
+        vst1.8          {d6},     [r0,:64], r2
+        subs            r3,  r3,  #2
+        bgt             1b
+        bx              lr
+endfunc
+
+function \type\()_rv40_qpel8_v_lowpass_neon
+        vld1.64         {d2},     [r1], r2
+        vld1.64         {d3},     [r1], r2
+        vld1.64         {d4},     [r1], r2
+        vld1.64         {d5},     [r1], r2
+        vld1.64         {d6},     [r1], r2
+        vld1.64         {d7},     [r1], r2
+        vld1.64         {d8},     [r1], r2
+        vld1.64         {d9},     [r1], r2
+        vld1.64         {d10},    [r1], r2
+        vld1.64         {d11},    [r1], r2
+        vld1.64         {d12},    [r1], r2
+        vld1.64         {d13},    [r1], r2
+        vld1.64         {d14},    [r1]
+        transpose_8x8   d2,  d3,  d4,  d5,  d6,  d7,  d8,  d9
+        transpose_8x8   d10, d11, d12, d13, d14, d15, d30, d31
+        qpel_lowpass_x2 d2,  d10, d3,  d11, d0,  d1,  6
+        qpel_lowpass_x2 d4,  d12, d5,  d13, d0,  d1,  6
+        qpel_lowpass_x2 d6,  d14, d7,  d15, d0,  d1,  6
+        qpel_lowpass_x2 d8,  d30, d9,  d31, d0,  d1,  6
+        transpose_8x8   d2,  d3,  d4,  d5,  d6,  d7,  d8,  d9
+  .ifc \type,avg
+        vld1.64         d12,      [r0,:64], r2
+        vld1.64         d13,      [r0,:64], r2
+        vld1.64         d14,      [r0,:64], r2
+        vld1.64         d15,      [r0,:64], r2
+        vld1.64         d16,      [r0,:64], r2
+        vld1.64         d17,      [r0,:64], r2
+        vld1.64         d18,      [r0,:64], r2
+        vld1.64         d19,      [r0,:64], r2
+        sub             r0,  r0,  r2,  lsl #3
+        vrhadd.u8       q1,  q1,  q6
+        vrhadd.u8       q2,  q2,  q7
+        vrhadd.u8       q3,  q3,  q8
+        vrhadd.u8       q4,  q4,  q9
+  .endif
+        vst1.64         d2,       [r0,:64], r2
+        vst1.64         d3,       [r0,:64], r2
+        vst1.64         d4,       [r0,:64], r2
+        vst1.64         d5,       [r0,:64], r2
+        vst1.64         d6,       [r0,:64], r2
+        vst1.64         d7,       [r0,:64], r2
+        vst1.64         d8,       [r0,:64], r2
+        vst1.64         d9,       [r0,:64], r2
+        bx              lr
+endfunc
+
+        rv40_qpel8_v    5, \type
+        rv40_qpel8_v    6, \type
+
+function ff_\type\()_rv40_qpel8_mc10_neon, export=1
+        sub             r1,  r1,  #2
+        mov             r3,  #8
+        vmov.i8         d0,  #52
+        vmov.i8         d1,  #20
+        b               \type\()_rv40_qpel8_h_lowpass_neon
+endfunc
+
+function ff_\type\()_rv40_qpel8_mc30_neon, export=1
+        sub             r1,  r1,  #2
+        mov             r3,  #8
+        vmov.i8         d0,  #20
+        vmov.i8         d1,  #52
+        b               \type\()_rv40_qpel8_h_lowpass_neon
+endfunc
+
+function ff_\type\()_rv40_qpel8_mc01_neon, export=1
+        push            {r4, lr}
+        vpush           {d8-d15}
+        sub             r1,  r1,  r2,  lsl #1
+        vmov.i8         d0,  #52
+        vmov.i8         d1,  #20
+        bl              \type\()_rv40_qpel8_v_lowpass_neon
+        vpop            {d8-d15}
+        pop             {r4, pc}
+endfunc
+
+function ff_\type\()_rv40_qpel8_mc11_neon, export=1
+        push            {r4, lr}
+        vpush           {d8-d15}
+        sub             sp,  sp,  #14*8
+        add             r12, sp,  #7
+        bic             r12, r12, #7
+        sub             r1,  r1,  r2,  lsl #1
+        sub             r1,  r1,  #2
+        mov             r3,  #12
+        vmov.i8         d0,  #52
+        vmov.i8         d1,  #20
+        bl              put_rv40_qpel8_h_lp_packed_s6_neon
+        add             r1,  sp,  #7
+        bic             r1,  r1,  #7
+        bl              \type\()_rv40_qpel8_v_lp_packed_s6_neon
+        add             sp,  sp,  #14*8
+        vpop            {d8-d15}
+        pop             {r4, pc}
+endfunc
+
+function ff_\type\()_rv40_qpel8_mc21_neon, export=1
+        push            {r4, lr}
+        vpush           {d8-d15}
+        sub             sp,  sp,  #14*8
+        add             r12, sp,  #7
+        bic             r12, r12, #7
+        sub             r1,  r1,  r2,  lsl #1
+        sub             r1,  r1,  #2
+        mov             r3,  #12
+        vmov.i8         d0,  #20
+        vmov.i8         d1,  #20
+        bl              put_rv40_qpel8_h_lp_packed_s5_neon
+        add             r1,  sp,  #7
+        bic             r1,  r1,  #7
+        vmov.i8         d0,  #52
+        bl              \type\()_rv40_qpel8_v_lp_packed_s6_neon
+        add             sp,  sp,  #14*8
+        vpop            {d8-d15}
+        pop             {r4, pc}
+endfunc
+
+function ff_\type\()_rv40_qpel8_mc31_neon, export=1
+        push            {r4, lr}
+        vpush           {d8-d15}
+        sub             sp,  sp,  #14*8
+        add             r12, sp,  #7
+        bic             r12, r12, #7
+        sub             r1,  r1,  r2,  lsl #1
+        sub             r1,  r1,  #2
+        mov             r3,  #12
+        vmov.i8         d0,  #20
+        vmov.i8         d1,  #52
+        bl              put_rv40_qpel8_h_lp_packed_s6_neon
+        add             r1,  sp,  #7
+        bic             r1,  r1,  #7
+        vswp            d0,  d1
+        bl              \type\()_rv40_qpel8_v_lp_packed_s6_neon
+        add             sp,  sp,  #14*8
+        vpop            {d8-d15}
+        pop             {r4, pc}
+endfunc
+
+function ff_\type\()_rv40_qpel8_mc12_neon, export=1
+        push            {r4, lr}
+        vpush           {d8-d15}
+        sub             sp,  sp,  #14*8
+        add             r12, sp,  #7
+        bic             r12, r12, #7
+        sub             r1,  r1,  r2,  lsl #1
+        sub             r1,  r1,  #2
+        mov             r3,  #12
+        vmov.i8         d0,  #52
+        vmov.i8         d1,  #20
+        bl              put_rv40_qpel8_h_lp_packed_s6_neon
+        add             r1,  sp,  #7
+        bic             r1,  r1,  #7
+        vmov.i8         d0,  #20
+        bl              \type\()_rv40_qpel8_v_lp_packed_s5_neon
+        add             sp,  sp,  #14*8
+        vpop            {d8-d15}
+        pop             {r4, pc}
+endfunc
+
+function ff_\type\()_rv40_qpel8_mc22_neon, export=1
+        push            {r4, lr}
+        vpush           {d8-d15}
+        sub             sp,  sp,  #14*8
+        add             r12, sp,  #7
+        bic             r12, r12, #7
+        sub             r1,  r1,  r2,  lsl #1
+        sub             r1,  r1,  #2
+        mov             r3,  #12
+        vmov.i8         d0,  #20
+        vmov.i8         d1,  #20
+        bl              put_rv40_qpel8_h_lp_packed_s5_neon
+        add             r1,  sp,  #7
+        bic             r1,  r1,  #7
+        bl              \type\()_rv40_qpel8_v_lp_packed_s5_neon
+        add             sp,  sp,  #14*8
+        vpop            {d8-d15}
+        pop             {r4, pc}
+endfunc
+
+function ff_\type\()_rv40_qpel8_mc32_neon, export=1
+        push            {r4, lr}
+        vpush           {d8-d15}
+        sub             sp,  sp,  #14*8
+        add             r12, sp,  #7
+        bic             r12, r12, #7
+        sub             r1,  r1,  r2,  lsl #1
+        sub             r1,  r1,  #2
+        mov             r3,  #12
+        vmov.i8         d0,  #20
+        vmov.i8         d1,  #52
+        bl              put_rv40_qpel8_h_lp_packed_s6_neon
+        add             r1,  sp,  #7
+        bic             r1,  r1,  #7
+        vmov.i8         d1,  #20
+        bl              \type\()_rv40_qpel8_v_lp_packed_s5_neon
+        add             sp,  sp,  #14*8
+        vpop            {d8-d15}
+        pop             {r4, pc}
+endfunc
+
+function ff_\type\()_rv40_qpel8_mc03_neon, export=1
+        push            {r4, lr}
+        vpush           {d8-d15}
+        sub             r1,  r1,  r2,  lsl #1
+        vmov.i8         d0,  #20
+        vmov.i8         d1,  #52
+        bl              \type\()_rv40_qpel8_v_lowpass_neon
+        vpop            {d8-d15}
+        pop             {r4, pc}
+endfunc
+
+function ff_\type\()_rv40_qpel8_mc33_neon, export=1
+        mov             r3,  #8
+        b               ff_\type\()_pixels8_xy2_neon
+endfunc
+
+function ff_\type\()_rv40_qpel8_mc13_neon, export=1
+        push            {r4, lr}
+        vpush           {d8-d15}
+        sub             sp,  sp,  #14*8
+        add             r12, sp,  #7
+        bic             r12, r12, #7
+        sub             r1,  r1,  r2,  lsl #1
+        sub             r1,  r1,  #2
+        mov             r3,  #12
+        vmov.i8         d0,  #52
+        vmov.i8         d1,  #20
+        bl              put_rv40_qpel8_h_lp_packed_s6_neon
+        add             r1,  sp,  #7
+        bic             r1,  r1,  #7
+        vswp            d0,  d1
+        bl              \type\()_rv40_qpel8_v_lp_packed_s6_neon
+        add             sp,  sp,  #14*8
+        vpop            {d8-d15}
+        pop             {r4, pc}
+endfunc
+
+function ff_\type\()_rv40_qpel8_mc23_neon, export=1
+        push            {r4, lr}
+        vpush           {d8-d15}
+        sub             sp,  sp,  #14*8
+        add             r12, sp,  #7
+        bic             r12, r12, #7
+        sub             r1,  r1,  r2,  lsl #1
+        sub             r1,  r1,  #2
+        mov             r3,  #12
+        vmov.i8         d0,  #20
+        vmov.i8         d1,  #20
+        bl              put_rv40_qpel8_h_lp_packed_s5_neon
+        add             r1,  sp,  #7
+        bic             r1,  r1,  #7
+        vmov.i8         d1,  #52
+        bl              \type\()_rv40_qpel8_v_lp_packed_s6_neon
+        add             sp,  sp,  #14*8
+        vpop            {d8-d15}
+        pop             {r4, pc}
+endfunc
+
+function ff_\type\()_rv40_qpel16_mc10_neon, export=1
+        vmov.i8         d0,  #52
+        vmov.i8         d1,  #20
+.L\type\()_rv40_qpel16_h:
+        push            {r1, lr}
+        sub             r1,  r1,  #2
+        mov             r3,  #16
+        bl              \type\()_rv40_qpel8_h_lowpass_neon
+        pop             {r1, lr}
+        sub             r0,  r0,  r2,  lsl #4
+        add             r0,  r0,  #8
+        add             r1,  r1,  #6
+        mov             r3,  #16
+        b               \type\()_rv40_qpel8_h_lowpass_neon
+endfunc
+
+function ff_\type\()_rv40_qpel16_mc30_neon, export=1
+        vmov.i8         d0,  #20
+        vmov.i8         d1,  #52
+        b               .L\type\()_rv40_qpel16_h
+endfunc
+
+function ff_\type\()_rv40_qpel16_mc01_neon, export=1
+        vmov.i8         d0,  #52
+        vmov.i8         d1,  #20
+.L\type\()_rv40_qpel16_v:
+        sub             r1,  r1,  r2,  lsl #1
+        push            {r1, lr}
+        vpush           {d8-d15}
+        bl              \type\()_rv40_qpel8_v_lowpass_neon
+        sub             r1,  r1,  r2,  lsl #2
+        bl              \type\()_rv40_qpel8_v_lowpass_neon
+        ldr             r1,  [sp, #64]
+        sub             r0,  r0,  r2,  lsl #4
+        add             r0,  r0,  #8
+        add             r1,  r1,  #8
+        bl              \type\()_rv40_qpel8_v_lowpass_neon
+        sub             r1,  r1,  r2,  lsl #2
+        bl              \type\()_rv40_qpel8_v_lowpass_neon
+        vpop            {d8-d15}
+        pop             {r1, pc}
+endfunc
+
+function ff_\type\()_rv40_qpel16_mc11_neon, export=1
+        sub             r1,  r1,  r2,  lsl #1
+        sub             r1,  r1,  #2
+        push            {r1, lr}
+        vpush           {d8-d15}
+        sub             sp,  sp,  #44*8
+        add             r12, sp,  #7
+        bic             r12, r12, #7
+        mov             r3,  #20
+        vmov.i8         d0,  #52
+        vmov.i8         d1,  #20
+        bl              put_rv40_qpel8_h_lp_packed_s6_neon
+        ldr             r1,  [sp, #416]
+        add             r1,  r1,  #8
+        mov             r3,  #20
+        bl              put_rv40_qpel8_h_lp_packed_s6_neon
+.L\type\()_rv40_qpel16_v_s6:
+        add             r1,  sp,  #7
+        bic             r1,  r1,  #7
+        bl              \type\()_rv40_qpel8_v_lp_packed_s6_neon
+        sub             r1,  r1,  #40
+        bl              \type\()_rv40_qpel8_v_lp_packed_s6_neon
+        sub             r0,  r0,  r2,  lsl #4
+        add             r0,  r0,  #8
+        bl              \type\()_rv40_qpel8_v_lp_packed_s6_neon
+        sub             r1,  r1,  #40
+        bl              \type\()_rv40_qpel8_v_lp_packed_s6_neon
+        add             sp,  sp,  #44*8
+        vpop            {d8-d15}
+        pop             {r1, pc}
+endfunc
+
+function ff_\type\()_rv40_qpel16_mc21_neon, export=1
+        sub             r1,  r1,  r2,  lsl #1
+        sub             r1,  r1,  #2
+        push            {r1, lr}
+        vpush           {d8-d15}
+        sub             sp,  sp,  #44*8
+        add             r12, sp,  #7
+        bic             r12, r12, #7
+        mov             r3,  #20
+        vmov.i8         d0,  #20
+        vmov.i8         d1,  #20
+        bl              put_rv40_qpel8_h_lp_packed_s5_neon
+        ldr             r1,  [sp, #416]
+        add             r1,  r1,  #8
+        mov             r3,  #20
+        bl              put_rv40_qpel8_h_lp_packed_s5_neon
+        vmov.i8         d0,  #52
+        b               .L\type\()_rv40_qpel16_v_s6
+endfunc
+
+function ff_\type\()_rv40_qpel16_mc31_neon, export=1
+        sub             r1,  r1,  r2,  lsl #1
+        sub             r1,  r1,  #2
+        push            {r1, lr}
+        vpush           {d8-d15}
+        sub             sp,  sp,  #44*8
+        add             r12, sp,  #7
+        bic             r12, r12, #7
+        mov             r3,  #20
+        vmov.i8         d0,  #20
+        vmov.i8         d1,  #52
+        bl              put_rv40_qpel8_h_lp_packed_s6_neon
+        ldr             r1,  [sp, #416]
+        add             r1,  r1,  #8
+        mov             r3,  #20
+        bl              put_rv40_qpel8_h_lp_packed_s6_neon
+        vswp            d0,  d1
+        b               .L\type\()_rv40_qpel16_v_s6
+endfunc
+
+function ff_\type\()_rv40_qpel16_mc12_neon, export=1
+        sub             r1,  r1,  r2,  lsl #1
+        sub             r1,  r1,  #2
+        push            {r1, lr}
+        vpush           {d8-d15}
+        sub             sp,  sp,  #44*8
+        add             r12, sp,  #7
+        bic             r12, r12, #7
+        mov             r3,  #20
+        vmov.i8         d0,  #52
+        vmov.i8         d1,  #20
+        bl              put_rv40_qpel8_h_lp_packed_s6_neon
+        ldr             r1,  [sp, #416]
+        add             r1,  r1,  #8
+        mov             r3,  #20
+        bl              put_rv40_qpel8_h_lp_packed_s6_neon
+        vmov.i8         d0,  #20
+.L\type\()_rv40_qpel16_v_s5:
+        add             r1,  sp,  #7
+        bic             r1,  r1,  #7
+        bl              \type\()_rv40_qpel8_v_lp_packed_s5_neon
+        sub             r1,  r1,  #40
+        bl              \type\()_rv40_qpel8_v_lp_packed_s5_neon
+        sub             r0,  r0,  r2,  lsl #4
+        add             r0,  r0,  #8
+        bl              \type\()_rv40_qpel8_v_lp_packed_s5_neon
+        sub             r1,  r1,  #40
+        bl              \type\()_rv40_qpel8_v_lp_packed_s5_neon
+        add             sp,  sp,  #44*8
+        vpop            {d8-d15}
+        pop             {r1, pc}
+endfunc
+
+function ff_\type\()_rv40_qpel16_mc22_neon, export=1
+        sub             r1,  r1,  r2,  lsl #1
+        sub             r1,  r1,  #2
+        push            {r1, lr}
+        vpush           {d8-d15}
+        sub             sp,  sp,  #44*8
+        add             r12, sp,  #7
+        bic             r12, r12, #7
+        mov             r3,  #20
+        vmov.i8         d0,  #20
+        vmov.i8         d1,  #20
+        bl              put_rv40_qpel8_h_lp_packed_s5_neon
+        ldr             r1,  [sp, #416]
+        add             r1,  r1,  #8
+        mov             r3,  #20
+        bl              put_rv40_qpel8_h_lp_packed_s5_neon
+        b               .L\type\()_rv40_qpel16_v_s5
+endfunc
+
+function ff_\type\()_rv40_qpel16_mc32_neon, export=1
+        sub             r1,  r1,  r2,  lsl #1
+        sub             r1,  r1,  #2
+        push            {r1, lr}
+        vpush           {d8-d15}
+        sub             sp,  sp,  #44*8
+        add             r12, sp,  #7
+        bic             r12, r12, #7
+        mov             r3,  #20
+        vmov.i8         d0,  #20
+        vmov.i8         d1,  #52
+        bl              put_rv40_qpel8_h_lp_packed_s6_neon
+        ldr             r1,  [sp, #416]
+        add             r1,  r1,  #8
+        mov             r3,  #20
+        bl              put_rv40_qpel8_h_lp_packed_s6_neon
+        vmov.i8         d1,  #20
+        b               .L\type\()_rv40_qpel16_v_s5
+endfunc
+
+function ff_\type\()_rv40_qpel16_mc03_neon, export=1
+        vmov.i8         d0,  #20
+        vmov.i8         d1,  #52
+        b               .L\type\()_rv40_qpel16_v
+endfunc
+
+function ff_\type\()_rv40_qpel16_mc13_neon, export=1
+        sub             r1,  r1,  r2,  lsl #1
+        sub             r1,  r1,  #2
+        push            {r1, lr}
+        vpush           {d8-d15}
+        sub             sp,  sp,  #44*8
+        add             r12, sp,  #7
+        bic             r12, r12, #7
+        mov             r3,  #20
+        vmov.i8         d0,  #52
+        vmov.i8         d1,  #20
+        bl              put_rv40_qpel8_h_lp_packed_s6_neon
+        ldr             r1,  [sp, #416]
+        add             r1,  r1,  #8
+        mov             r3,  #20
+        bl              put_rv40_qpel8_h_lp_packed_s6_neon
+        vswp            d0,  d1
+        b               .L\type\()_rv40_qpel16_v_s6
+endfunc
+
+function ff_\type\()_rv40_qpel16_mc23_neon, export=1
+        sub             r1,  r1,  r2,  lsl #1
+        sub             r1,  r1,  #2
+        push            {r1, lr}
+        vpush           {d8-d15}
+        sub             sp,  sp,  #44*8
+        add             r12, sp,  #7
+        bic             r12, r12, #7
+        mov             r3,  #20
+        vmov.i8         d0,  #20
+        vmov.i8         d1,  #20
+        bl              put_rv40_qpel8_h_lp_packed_s5_neon
+        ldr             r1,  [sp, #416]
+        add             r1,  r1,  #8
+        mov             r3,  #20
+        bl              put_rv40_qpel8_h_lp_packed_s5_neon
+        vmov.i8         d1,  #52
+        b               .L\type\()_rv40_qpel16_v_s6
+endfunc
+
+function ff_\type\()_rv40_qpel16_mc33_neon, export=1
+        mov             r3,  #16
+        b               ff_\type\()_pixels16_xy2_neon
+endfunc
+.endm
+
+        rv40_qpel       put
+        rv40_qpel       avg
 
 .macro  rv40_weight
         vmovl.u8        q8,  d2
diff --git a/libavcodec/cljr.c b/libavcodec/cljr.c
index 53ab40cbe2..183652b3a1 100644
--- a/libavcodec/cljr.c
+++ b/libavcodec/cljr.c
@@ -25,119 +25,92 @@
  */
 
 #include "avcodec.h"
-#include "dsputil.h"
 #include "get_bits.h"
+#include "put_bits.h"
 
-/* Disable the encoder. */
-#undef CONFIG_CLJR_ENCODER
-#define CONFIG_CLJR_ENCODER 0
-
-typedef struct CLJRContext{
+typedef struct CLJRContext {
     AVCodecContext *avctx;
-    AVFrame picture;
+    AVFrame         picture;
 } CLJRContext;
 
+static av_cold int common_init(AVCodecContext *avctx)
+{
+    CLJRContext * const a = avctx->priv_data;
+
+    avcodec_get_frame_defaults(&a->picture);
+    avctx->coded_frame = &a->picture;
+    a->avctx = avctx;
+
+    return 0;
+}
+
+#if CONFIG_CLJR_DECODER
 static int decode_frame(AVCodecContext *avctx,
                         void *data, int *data_size,
                         AVPacket *avpkt)
 {
     const uint8_t *buf = avpkt->data;
-    int buf_size = avpkt->size;
+    int buf_size       = avpkt->size;
     CLJRContext * const a = avctx->priv_data;
     GetBitContext gb;
     AVFrame *picture = data;
-    AVFrame * const p= (AVFrame*)&a->picture;
+    AVFrame * const p = &a->picture;
     int x, y;
 
-    if(p->data[0])
+    if (p->data[0])
         avctx->release_buffer(avctx, p);
 
-    if(buf_size/avctx->height < avctx->width) {
-        av_log(avctx, AV_LOG_ERROR, "Resolution larger than buffer size. Invalid header?\n");
-        return -1;
+    if (buf_size / avctx->height < avctx->width) {
+        av_log(avctx, AV_LOG_ERROR,
+               "Resolution larger than buffer size. Invalid header?\n");
+        return AVERROR_INVALIDDATA;
     }
 
-    p->reference= 0;
-    if(avctx->get_buffer(avctx, p) < 0){
+    p->reference = 0;
+    if (avctx->get_buffer(avctx, p) < 0) {
         av_log(avctx, AV_LOG_ERROR, "get_buffer() failed\n");
         return -1;
     }
-    p->pict_type= AV_PICTURE_TYPE_I;
-    p->key_frame= 1;
+    p->pict_type = AV_PICTURE_TYPE_I;
+    p->key_frame = 1;
 
     init_get_bits(&gb, buf, buf_size * 8);
 
-    for(y=0; y<avctx->height; y++){
-        uint8_t *luma= &a->picture.data[0][ y*a->picture.linesize[0] ];
-        uint8_t *cb= &a->picture.data[1][ y*a->picture.linesize[1] ];
-        uint8_t *cr= &a->picture.data[2][ y*a->picture.linesize[2] ];
-        for(x=0; x<avctx->width; x+=4){
+    for (y = 0; y < avctx->height; y++) {
+        uint8_t *luma = &a->picture.data[0][y * a->picture.linesize[0]];
+        uint8_t *cb   = &a->picture.data[1][y * a->picture.linesize[1]];
+        uint8_t *cr   = &a->picture.data[2][y * a->picture.linesize[2]];
+        for (x = 0; x < avctx->width; x += 4) {
             luma[3] = get_bits(&gb, 5) << 3;
             luma[2] = get_bits(&gb, 5) << 3;
             luma[1] = get_bits(&gb, 5) << 3;
             luma[0] = get_bits(&gb, 5) << 3;
-            luma+= 4;
+            luma += 4;
             *(cb++) = get_bits(&gb, 6) << 2;
             *(cr++) = get_bits(&gb, 6) << 2;
         }
     }
 
-    *picture= *(AVFrame*)&a->picture;
+    *picture   = a->picture;
     *data_size = sizeof(AVPicture);
 
-    emms_c();
-
     return buf_size;
 }
 
-#if CONFIG_CLJR_ENCODER
-static int encode_frame(AVCodecContext *avctx, unsigned char *buf, int buf_size, void *data){
-    CLJRContext * const a = avctx->priv_data;
-    AVFrame *pict = data;
-    AVFrame * const p= (AVFrame*)&a->picture;
-    int size;
-
-    *p = *pict;
-    p->pict_type= AV_PICTURE_TYPE_I;
-    p->key_frame= 1;
-
-    emms_c();
-
-    avpriv_align_put_bits(&a->pb);
-    while(get_bit_count(&a->pb)&31)
-        put_bits(&a->pb, 8, 0);
-
-    size= get_bit_count(&a->pb)/32;
-
-    return size*4;
-}
-#endif
-
-static av_cold void common_init(AVCodecContext *avctx){
-    CLJRContext * const a = avctx->priv_data;
-
-    avcodec_get_frame_defaults(&a->picture);
-    avctx->coded_frame= (AVFrame*)&a->picture;
-    a->avctx= avctx;
-}
-
-static av_cold int decode_init(AVCodecContext *avctx){
-
-    common_init(avctx);
-
-    avctx->pix_fmt= PIX_FMT_YUV411P;
-
-    return 0;
+static av_cold int decode_init(AVCodecContext *avctx)
+{
+    avctx->pix_fmt = PIX_FMT_YUV411P;
+    return common_init(avctx);
 }
 
-#if CONFIG_CLJR_ENCODER
-static av_cold int encode_init(AVCodecContext *avctx){
-
-    common_init(avctx);
+static av_cold int decode_end(AVCodecContext *avctx)
+{
+    CLJRContext *a = avctx->priv_data;
 
+    if (a->picture.data[0])
+        avctx->release_buffer(avctx, &a->picture);
     return 0;
 }
-#endif
 
 AVCodec ff_cljr_decoder = {
     .name           = "cljr",
@@ -145,19 +118,55 @@ AVCodec ff_cljr_decoder = {
     .id             = CODEC_ID_CLJR,
     .priv_data_size = sizeof(CLJRContext),
     .init           = decode_init,
+    .close          = decode_end,
     .decode         = decode_frame,
     .capabilities   = CODEC_CAP_DR1,
-    .long_name = NULL_IF_CONFIG_SMALL("Cirrus Logic AccuPak"),
+    .long_name      = NULL_IF_CONFIG_SMALL("Cirrus Logic AccuPak"),
 };
+#endif
 
 #if CONFIG_CLJR_ENCODER
+static int encode_frame(AVCodecContext *avctx, unsigned char *buf,
+                        int buf_size, void *data)
+{
+    PutBitContext pb;
+    AVFrame *p = data;
+    int x, y;
+
+    p->pict_type = AV_PICTURE_TYPE_I;
+    p->key_frame = 1;
+
+    init_put_bits(&pb, buf, buf_size / 8);
+
+    for (y = 0; y < avctx->height; y++) {
+        uint8_t *luma = &p->data[0][y * p->linesize[0]];
+        uint8_t *cb   = &p->data[1][y * p->linesize[1]];
+        uint8_t *cr   = &p->data[2][y * p->linesize[2]];
+        for (x = 0; x < avctx->width; x += 4) {
+            put_bits(&pb, 5, luma[3] >> 3);
+            put_bits(&pb, 5, luma[2] >> 3);
+            put_bits(&pb, 5, luma[1] >> 3);
+            put_bits(&pb, 5, luma[0] >> 3);
+            luma += 4;
+            put_bits(&pb, 6, *(cb++) >> 2);
+            put_bits(&pb, 6, *(cr++) >> 2);
+        }
+    }
+
+    flush_put_bits(&pb);
+
+    return put_bits_count(&pb) / 8;
+}
+
 AVCodec ff_cljr_encoder = {
     .name           = "cljr",
     .type           = AVMEDIA_TYPE_VIDEO,
     .id             = CODEC_ID_CLJR,
     .priv_data_size = sizeof(CLJRContext),
-    .init           = encode_init,
+    .init           = common_init,
     .encode         = encode_frame,
-    .long_name = NULL_IF_CONFIG_SMALL("Cirrus Logic AccuPak"),
+    .pix_fmts       = (const enum PixelFormat[]) { PIX_FMT_YUV411P,
+                                                   PIX_FMT_NONE },
+    .long_name      = NULL_IF_CONFIG_SMALL("Cirrus Logic AccuPak"),
 };
 #endif
diff --git a/libavcodec/dct-test.c b/libavcodec/dct-test.c
index e3203efe89..dd6987b459 100644
--- a/libavcodec/dct-test.c
+++ b/libavcodec/dct-test.c
@@ -170,7 +170,7 @@ static const struct algo idct_tab[] = {
 
 #define AANSCALE_BITS 12
 
-uint8_t cropTbl[256 + 2 * MAX_NEG_CROP];
+static uint8_t cropTbl[256 + 2 * MAX_NEG_CROP];
 
 static int64_t gettime(void)
 {
diff --git a/libavcodec/fft-test.c b/libavcodec/fft-test.c
index be105fe834..1c88a53fdd 100644
--- a/libavcodec/fft-test.c
+++ b/libavcodec/fft-test.c
@@ -37,8 +37,6 @@
 #include <stdlib.h>
 #include <string.h>
 
-#undef exit
-
 /* reference fft */
 
 #define MUL16(a,b) ((a) * (b))
@@ -228,7 +226,6 @@ static void help(void)
            "-n b   set the transform size to 2^b\n"
            "-f x   set scale factor for output data of (I)MDCT to x\n"
            );
-    exit(1);
 }
 
 enum tf_transform {
@@ -267,7 +264,7 @@ int main(int argc, char **argv)
         switch(c) {
         case 'h':
             help();
-            break;
+            return 1;
         case 's':
             do_speed = 1;
             break;
diff --git a/libavcodec/motion-test.c b/libavcodec/motion-test.c
index 85ea1045b4..f187183c12 100644
--- a/libavcodec/motion-test.c
+++ b/libavcodec/motion-test.c
@@ -33,14 +33,13 @@
 #include "dsputil.h"
 #include "libavutil/lfg.h"
 
-#undef exit
 #undef printf
 
 #define WIDTH 64
 #define HEIGHT 64
 
-uint8_t img1[WIDTH * HEIGHT];
-uint8_t img2[WIDTH * HEIGHT];
+static uint8_t img1[WIDTH * HEIGHT];
+static uint8_t img2[WIDTH * HEIGHT];
 
 static void fill_random(uint8_t *tab, int size)
 {
@@ -61,7 +60,6 @@ static void help(void)
 {
     printf("motion-test [-h]\n"
            "test motion implementations\n");
-    exit(1);
 }
 
 static int64_t gettime(void)
@@ -138,7 +136,7 @@ int main(int argc, char **argv)
         switch(c) {
         case 'h':
             help();
-            break;
+            return 1;
         }
     }
 
diff --git a/libavcodec/shorten.c b/libavcodec/shorten.c
index f0a173cc7e..e0d3f6f986 100644
--- a/libavcodec/shorten.c
+++ b/libavcodec/shorten.c
@@ -176,7 +176,7 @@ static void fix_bitshift(ShortenContext *s, int32_t *buffer)
 }
 
 
-static void init_offset(ShortenContext *s)
+static int init_offset(ShortenContext *s)
 {
     int32_t mean = 0;
     int  chan, i;
@@ -190,12 +190,13 @@ static void init_offset(ShortenContext *s)
             break;
         default:
             av_log(s->avctx, AV_LOG_ERROR, "unknown audio type");
-            abort();
+            return AVERROR_INVALIDDATA;
     }
 
     for (chan = 0; chan < s->channels; chan++)
         for (i = 0; i < nblock; i++)
             s->offset[chan][i] = mean;
+    return 0;
 }
 
 static int decode_wave_header(AVCodecContext *avctx, const uint8_t *header,
@@ -367,7 +368,8 @@ static int read_header(ShortenContext *s)
     if ((ret = allocate_buffers(s)) < 0)
         return ret;
 
-    init_offset(s);
+    if ((ret = init_offset(s)) < 0)
+        return ret;
 
     if (s->version > 1)
         s->lpcqoffset = V2LPCQOFFSET;
diff --git a/libavcodec/version.h b/libavcodec/version.h
index 9042e26e2f..0e2c58e852 100644
--- a/libavcodec/version.h
+++ b/libavcodec/version.h
@@ -21,8 +21,8 @@
 #define AVCODEC_VERSION_H
 
 #define LIBAVCODEC_VERSION_MAJOR 53
-#define LIBAVCODEC_VERSION_MINOR 41
-#define LIBAVCODEC_VERSION_MICRO  2
+#define LIBAVCODEC_VERSION_MINOR 42
+#define LIBAVCODEC_VERSION_MICRO  0
 
 #define LIBAVCODEC_VERSION_INT  AV_VERSION_INT(LIBAVCODEC_VERSION_MAJOR, \
                                                LIBAVCODEC_VERSION_MINOR, \
diff --git a/libavformat/riff.c b/libavformat/riff.c
index 6c8d8abe40..8dad3c829b 100644
--- a/libavformat/riff.c
+++ b/libavformat/riff.c
@@ -266,7 +266,7 @@ const AVCodecTag ff_codec_bmp_tags[] = {
     { CODEC_ID_TARGA,        MKTAG('t', 'g', 'a', ' ') },
     { CODEC_ID_PNG,          MKTAG('M', 'P', 'N', 'G') },
     { CODEC_ID_PNG,          MKTAG('P', 'N', 'G', '1') },
-    { CODEC_ID_CLJR,         MKTAG('c', 'l', 'j', 'r') },
+    { CODEC_ID_CLJR,         MKTAG('C', 'L', 'J', 'R') },
     { CODEC_ID_DIRAC,        MKTAG('d', 'r', 'a', 'c') },
     { CODEC_ID_RPZA,         MKTAG('a', 'z', 'p', 'r') },
     { CODEC_ID_RPZA,         MKTAG('R', 'P', 'Z', 'A') },
diff --git a/libavformat/seek-test.c b/libavformat/seek-test.c
index 3a448cfe3b..ec88a0baf2 100644
--- a/libavformat/seek-test.c
+++ b/libavformat/seek-test.c
@@ -28,7 +28,6 @@
 #include "libavutil/mathematics.h"
 #include "libavformat/avformat.h"
 
-#undef exit
 #undef printf
 #undef fprintf
 
@@ -75,7 +74,7 @@ int main(int argc, char **argv)
     if (argc != 2) {
         printf("usage: %s input_file\n"
                "\n", argv[0]);
-        exit(1);
+        return 1;
     }
 
     filename = argv[1];
@@ -84,13 +83,13 @@ int main(int argc, char **argv)
     av_dict_free(&format_opts);
     if (ret < 0) {
         fprintf(stderr, "cannot open %s\n", filename);
-        exit(1);
+        return 1;
     }
 
     ret = avformat_find_stream_info(ic, NULL);
     if (ret < 0) {
         fprintf(stderr, "%s: could not find codec parameters\n", filename);
-        exit(1);
+        return 1;
     }
 
     for(i=0; ; i++){
diff --git a/libavutil/adler32.c b/libavutil/adler32.c
index 8ea5fc2958..f4f56ea9b6 100644
--- a/libavutil/adler32.c
+++ b/libavutil/adler32.c
@@ -61,7 +61,9 @@ unsigned long av_adler32_update(unsigned long adler, const uint8_t * buf,
 #include "log.h"
 #include "timer.h"
 #define LEN 7001
-volatile int checksum;
+
+static volatile int checksum;
+
 int main(int argc, char **argv)
 {
     int i;
diff --git a/libavutil/x86/cpu.c b/libavutil/x86/cpu.c
index bb3f1b325b..93df737c28 100644
--- a/libavutil/x86/cpu.c
+++ b/libavutil/x86/cpu.c
@@ -74,7 +74,10 @@ int ff_get_cpu_flags_x86(void)
         return 0; /* CPUID not supported */
 #endif
 
-    cpuid(0, max_std_level, vendor.i[0], vendor.i[2], vendor.i[1]);
+    cpuid(0, max_std_level, ebx, ecx, edx);
+    vendor.i[0] = ebx;
+    vendor.i[1] = edx;
+    vendor.i[2] = ecx;
 
     if(max_std_level >= 1){
         cpuid(1, eax, ebx, ecx, std_caps);
diff --git a/tools/cws2fws.c b/tools/cws2fws.c
index b8535feaa4..35fce35683 100644
--- a/tools/cws2fws.c
+++ b/tools/cws2fws.c
@@ -29,14 +29,14 @@ int main(int argc, char *argv[])
     if (argc < 3)
     {
         printf("Usage: %s <infile.swf> <outfile.swf>\n", argv[0]);
-        exit(1);
+        return 1;
     }
 
     fd_in = open(argv[1], O_RDONLY);
     if (fd_in < 0)
     {
         perror("Error opening input file");
-        exit(1);
+        return 1;
     }
 
     fd_out = open(argv[2], O_WRONLY|O_CREAT, 00644);
@@ -44,7 +44,7 @@ int main(int argc, char *argv[])
     {
         perror("Error opening output file");
         close(fd_in);
-        exit(1);
+        return 1;
     }
 
     if (read(fd_in, &buf_in, 8) != 8)
@@ -52,13 +52,13 @@ int main(int argc, char *argv[])
         printf("Header error\n");
         close(fd_in);
         close(fd_out);
-        exit(1);
+        return 1;
     }
 
     if (buf_in[0] != 'C' || buf_in[1] != 'W' || buf_in[2] != 'S')
     {
         printf("Not a compressed flash file\n");
-        exit(1);
+        return 1;
     }
 
     fstat(fd_in, &statbuf);
@@ -71,7 +71,7 @@ int main(int argc, char *argv[])
     buf_in[0] = 'F';
     if (write(fd_out, &buf_in, 8) < 8) {
         perror("Error writing output file");
-        exit(1);
+        return 1;
     }
 
     zstream.zalloc = NULL;
@@ -97,7 +97,7 @@ int main(int argc, char *argv[])
         {
             printf("Error while decompressing: %d\n", ret);
             inflateEnd(&zstream);
-            exit(1);
+            return 1;
         }
 
         dbgprintf("a_in: %d t_in: %lu a_out: %d t_out: %lu -- %lu out\n",
@@ -106,7 +106,7 @@ int main(int argc, char *argv[])
 
         if (write(fd_out, &buf_out, zstream.total_out - last_out) < zstream.total_out - last_out) {
             perror("Error writing output file");
-            exit(1);
+            return 1;
         }
 
         i += len;
@@ -128,7 +128,7 @@ int main(int argc, char *argv[])
         lseek(fd_out, 4, SEEK_SET);
         if (write(fd_out, &buf_in, 4) < 4) {
             perror("Error writing output file");
-            exit(1);
+            return 1;
         }
     }