Merge remote-tracking branch 'qatar/master' into master

* qatar/master: (27 commits)
  ac3enc: fix LOCAL_ALIGNED usage in count_mantissa_bits()
  ac3dsp: do not use the ff_* prefix when referencing ff_ac3_bap_bits.
  ac3dsp: fix loop condition in ac3_update_bap_counts_c()
  ARM: unbreak build
  ac3enc: modify mantissa bit counting to keep bap counts for all values of bap instead of just 0 to 4.
  ac3enc: split mantissa bit counting into a separate function.
  ac3enc: store per-block/channel bap pointers by reference block in a 2D array rather than in the AC3Block struct.
  get_bits: add av_unused tag to cache variable
  sws: replace all long with int.
  ARM: aacdec: fix constraints on inline asm
  ARM: remove unnecessary volatile from inline asm
  ARM: add "cc" clobbers to inline asm where needed
  ARM: improve FASTDIV asm
  ac3enc: use LOCAL_ALIGNED macro
  APIchanges: fill in git hash for av_get_pix_fmt_name (0420bd7).
  lavu: add av_get_pix_fmt_name() convenience function
  cmdutils: remove OPT_FUNC2
  swscale: fix crash in bilinear scaling.
  vpxenc: add VP8E_SET_STATIC_THRESHOLD mapping
  webm: support stereo videos in matroska/webm muxer
  ...

Conflicts:
	Changelog
	cmdutils.c
	cmdutils.h
	doc/APIchanges
	doc/muxers.texi
	ffmpeg.c
	ffplay.c
	libavcodec/ac3enc.c
	libavcodec/ac3enc_float.c
	libavcodec/avcodec.h
	libavcodec/get_bits.h
	libavcodec/libvpxenc.c
	libavcodec/version.h
	libavdevice/libdc1394.c
	libavformat/matroskaenc.c
	libavutil/avutil.h
	libswscale/rgb2rgb.c
	libswscale/swscale.c
	libswscale/swscale_template.c
	libswscale/x86/swscale_template.c

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

1 files changed, 78 insertions, 75 deletions

diff --git a/libavcodec/arm/aac.h b/libavcodec/arm/aac.h
index 390cdbfcf9..3b14c094c6 100644
--- a/libavcodec/arm/aac.h
+++ b/libavcodec/arm/aac.h
@@ -30,17 +30,17 @@ static inline float *VMUL2(float *dst, const float *v, unsigned idx,
                            const float *scale)
 {
     unsigned v0, v1;
-    __asm__ volatile ("ubfx     %0,  %4,  #0, #4      \n\t"
-                      "ubfx     %1,  %4,  #4, #4      \n\t"
-                      "ldr      %0,  [%3, %0, lsl #2] \n\t"
-                      "ldr      %1,  [%3, %1, lsl #2] \n\t"
-                      "vld1.32  {d1[]},   [%5,:32]    \n\t"
-                      "vmov     d0,  %0,  %1          \n\t"
-                      "vmul.f32 d0,  d0,  d1          \n\t"
-                      "vst1.32  {d0},     [%2,:64]!   \n\t"
-                      : "=&r"(v0), "=&r"(v1), "+r"(dst)
-                      : "r"(v), "r"(idx), "r"(scale)
-                      : "d0", "d1");
+    __asm__ ("ubfx     %0,  %6,  #0, #4      \n\t"
+             "ubfx     %1,  %6,  #4, #4      \n\t"
+             "ldr      %0,  [%5, %0, lsl #2] \n\t"
+             "ldr      %1,  [%5, %1, lsl #2] \n\t"
+             "vld1.32  {d1[]},   [%7,:32]    \n\t"
+             "vmov     d0,  %0,  %1          \n\t"
+             "vmul.f32 d0,  d0,  d1          \n\t"
+             "vst1.32  {d0},     [%2,:64]!   \n\t"
+             : "=&r"(v0), "=&r"(v1), "+r"(dst), "=m"(dst[0]), "=m"(dst[1])
+             : "r"(v), "r"(idx), "r"(scale)
+             : "d0", "d1");
     return dst;
 }
 
@@ -49,22 +49,23 @@ static inline float *VMUL4(float *dst, const float *v, unsigned idx,
                            const float *scale)
 {
     unsigned v0, v1, v2, v3;
-    __asm__ volatile ("ubfx     %0,  %6,  #0, #2      \n\t"
-                      "ubfx     %1,  %6,  #2, #2      \n\t"
-                      "ldr      %0,  [%5, %0, lsl #2] \n\t"
-                      "ubfx     %2,  %6,  #4, #2      \n\t"
-                      "ldr      %1,  [%5, %1, lsl #2] \n\t"
-                      "ubfx     %3,  %6,  #6, #2      \n\t"
-                      "ldr      %2,  [%5, %2, lsl #2] \n\t"
-                      "vmov     d0,  %0,  %1          \n\t"
-                      "ldr      %3,  [%5, %3, lsl #2] \n\t"
-                      "vld1.32  {d2[],d3[]},[%7,:32]  \n\t"
-                      "vmov     d1,  %2,  %3          \n\t"
-                      "vmul.f32 q0,  q0,  q1          \n\t"
-                      "vst1.32  {q0},     [%4,:128]!  \n\t"
-                      : "=&r"(v0), "=&r"(v1), "=&r"(v2), "=&r"(v3), "+r"(dst)
-                      : "r"(v), "r"(idx), "r"(scale)
-                      : "d0", "d1", "d2", "d3");
+    __asm__ ("ubfx     %0,  %10, #0, #2      \n\t"
+             "ubfx     %1,  %10, #2, #2      \n\t"
+             "ldr      %0,  [%9, %0, lsl #2] \n\t"
+             "ubfx     %2,  %10, #4, #2      \n\t"
+             "ldr      %1,  [%9, %1, lsl #2] \n\t"
+             "ubfx     %3,  %10, #6, #2      \n\t"
+             "ldr      %2,  [%9, %2, lsl #2] \n\t"
+             "vmov     d0,  %0,  %1          \n\t"
+             "ldr      %3,  [%9, %3, lsl #2] \n\t"
+             "vld1.32  {d2[],d3[]},[%11,:32] \n\t"
+             "vmov     d1,  %2,  %3          \n\t"
+             "vmul.f32 q0,  q0,  q1          \n\t"
+             "vst1.32  {q0},     [%4,:128]!  \n\t"
+             : "=&r"(v0), "=&r"(v1), "=&r"(v2), "=&r"(v3), "+r"(dst),
+               "=m"(dst[0]), "=m"(dst[1]), "=m"(dst[2]), "=m"(dst[3])
+             : "r"(v), "r"(idx), "r"(scale)
+             : "d0", "d1", "d2", "d3");
     return dst;
 }
 
@@ -73,22 +74,23 @@ static inline float *VMUL2S(float *dst, const float *v, unsigned idx,
                             unsigned sign, const float *scale)
 {
     unsigned v0, v1, v2, v3;
-    __asm__ volatile ("ubfx     %0,  %6,  #0, #4      \n\t"
-                      "ubfx     %1,  %6,  #4, #4      \n\t"
-                      "ldr      %0,  [%5, %0, lsl #2] \n\t"
-                      "lsl      %2,  %8,  #30         \n\t"
-                      "ldr      %1,  [%5, %1, lsl #2] \n\t"
-                      "lsl      %3,  %8,  #31         \n\t"
-                      "vmov     d0,  %0,  %1          \n\t"
-                      "bic      %2,  %2,  #1<<30      \n\t"
-                      "vld1.32  {d1[]},   [%7,:32]    \n\t"
-                      "vmov     d2,  %2,  %3          \n\t"
-                      "veor     d0,  d0,  d2          \n\t"
-                      "vmul.f32 d0,  d0,  d1          \n\t"
-                      "vst1.32  {d0},     [%4,:64]!   \n\t"
-                      : "=&r"(v0), "=&r"(v1), "=&r"(v2), "=&r"(v3), "+r"(dst)
-                      : "r"(v), "r"(idx), "r"(scale), "r"(sign)
-                      : "d0", "d1", "d2");
+    __asm__ ("ubfx     %0,  %8,  #0, #4      \n\t"
+             "ubfx     %1,  %8,  #4, #4      \n\t"
+             "ldr      %0,  [%7, %0, lsl #2] \n\t"
+             "lsl      %2,  %10, #30         \n\t"
+             "ldr      %1,  [%7, %1, lsl #2] \n\t"
+             "lsl      %3,  %10, #31         \n\t"
+             "vmov     d0,  %0,  %1          \n\t"
+             "bic      %2,  %2,  #1<<30      \n\t"
+             "vld1.32  {d1[]},   [%9,:32]    \n\t"
+             "vmov     d2,  %2,  %3          \n\t"
+             "veor     d0,  d0,  d2          \n\t"
+             "vmul.f32 d0,  d0,  d1          \n\t"
+             "vst1.32  {d0},     [%4,:64]!   \n\t"
+             : "=&r"(v0), "=&r"(v1), "=&r"(v2), "=&r"(v3), "+r"(dst),
+               "=m"(dst[0]), "=m"(dst[1])
+             : "r"(v), "r"(idx), "r"(scale), "r"(sign)
+             : "d0", "d1", "d2");
     return dst;
 }
 
@@ -97,38 +99,39 @@ static inline float *VMUL4S(float *dst, const float *v, unsigned idx,
                             unsigned sign, const float *scale)
 {
     unsigned v0, v1, v2, v3, nz;
-    __asm__ volatile ("vld1.32  {d2[],d3[]},[%9,:32]  \n\t"
-                      "ubfx     %0,  %8,  #0, #2      \n\t"
-                      "ubfx     %1,  %8,  #2, #2      \n\t"
-                      "ldr      %0,  [%7, %0, lsl #2] \n\t"
-                      "ubfx     %2,  %8,  #4, #2      \n\t"
-                      "ldr      %1,  [%7, %1, lsl #2] \n\t"
-                      "ubfx     %3,  %8,  #6, #2      \n\t"
-                      "ldr      %2,  [%7, %2, lsl #2] \n\t"
-                      "vmov     d0,  %0,  %1          \n\t"
-                      "ldr      %3,  [%7, %3, lsl #2] \n\t"
-                      "lsr      %6,  %8,  #12         \n\t"
-                      "rbit     %6,  %6               \n\t"
-                      "vmov     d1,  %2,  %3          \n\t"
-                      "lsls     %6,  %6,  #1          \n\t"
-                      "and      %0,  %5,  #1<<31      \n\t"
-                      "lslcs    %5,  %5,  #1          \n\t"
-                      "lsls     %6,  %6,  #1          \n\t"
-                      "and      %1,  %5,  #1<<31      \n\t"
-                      "lslcs    %5,  %5,  #1          \n\t"
-                      "lsls     %6,  %6,  #1          \n\t"
-                      "and      %2,  %5,  #1<<31      \n\t"
-                      "lslcs    %5,  %5,  #1          \n\t"
-                      "vmov     d4,  %0,  %1          \n\t"
-                      "and      %3,  %5,  #1<<31      \n\t"
-                      "vmov     d5,  %2,  %3          \n\t"
-                      "veor     q0,  q0,  q2          \n\t"
-                      "vmul.f32 q0,  q0,  q1          \n\t"
-                      "vst1.32  {q0},     [%4,:128]!  \n\t"
-                      : "=&r"(v0), "=&r"(v1), "=&r"(v2), "=&r"(v3), "+r"(dst),
-                        "+r"(sign), "=r"(nz)
-                      : "r"(v), "r"(idx), "r"(scale)
-                      : "d0", "d1", "d2", "d3", "d4", "d5");
+    __asm__ ("vld1.32  {d2[],d3[]},[%13,:32] \n\t"
+             "ubfx     %0,  %12, #0, #2      \n\t"
+             "ubfx     %1,  %12, #2, #2      \n\t"
+             "ldr      %0,  [%11,%0, lsl #2] \n\t"
+             "ubfx     %2,  %12, #4, #2      \n\t"
+             "ldr      %1,  [%11,%1, lsl #2] \n\t"
+             "ubfx     %3,  %12, #6, #2      \n\t"
+             "ldr      %2,  [%11,%2, lsl #2] \n\t"
+             "vmov     d0,  %0,  %1          \n\t"
+             "ldr      %3,  [%11,%3, lsl #2] \n\t"
+             "lsr      %6,  %12, #12         \n\t"
+             "rbit     %6,  %6               \n\t"
+             "vmov     d1,  %2,  %3          \n\t"
+             "lsls     %6,  %6,  #1          \n\t"
+             "and      %0,  %5,  #1<<31      \n\t"
+             "lslcs    %5,  %5,  #1          \n\t"
+             "lsls     %6,  %6,  #1          \n\t"
+             "and      %1,  %5,  #1<<31      \n\t"
+             "lslcs    %5,  %5,  #1          \n\t"
+             "lsls     %6,  %6,  #1          \n\t"
+             "and      %2,  %5,  #1<<31      \n\t"
+             "lslcs    %5,  %5,  #1          \n\t"
+             "vmov     d4,  %0,  %1          \n\t"
+             "and      %3,  %5,  #1<<31      \n\t"
+             "vmov     d5,  %2,  %3          \n\t"
+             "veor     q0,  q0,  q2          \n\t"
+             "vmul.f32 q0,  q0,  q1          \n\t"
+             "vst1.32  {q0},     [%4,:128]!  \n\t"
+             : "=&r"(v0), "=&r"(v1), "=&r"(v2), "=&r"(v3), "+r"(dst),
+               "+r"(sign), "=r"(nz),
+               "=m"(dst[0]), "=m"(dst[1]), "=m"(dst[2]), "=m"(dst[3])
+             : "r"(v), "r"(idx), "r"(scale)
+             : "cc", "d0", "d1", "d2", "d3", "d4", "d5");
     return dst;
 }