Merge remote-tracking branch 'qatar/master'

[ffmpeg] / libavcodec / x86 / h264_qpel_mmx.c

diff --git a/libavcodec/x86/h264_qpel_mmx.c b/libavcodec/x86/h264_qpel_mmx.c
index d8ceca181912399c104c08d6c132a2c900806f2f..807d8548d6f44543fae2aea4330c3755ac12e523 100644 (file)
--- a/libavcodec/x86/h264_qpel_mmx.c
+++ b/libavcodec/x86/h264_qpel_mmx.c
@@ -1,5 +1,6 @@
 /*
  * Copyright (c) 2004-2005 Michael Niedermayer, Loren Merritt
+ * Copyright (c) 2011 Daniel Kang
  *
  * This file is part of FFmpeg.
  *
@@ -398,7 +399,7 @@ static av_noinline void OPNAME ## h264_qpel8or16_v_lowpass_ ## MMX(uint8_t *dst,
             "2:                         \n\t"\
             \
             : "+a"(src), "+c"(dst)\
-            : "S"((x86_reg)srcStride), "D"((x86_reg)dstStride), "g"(h)\
+            : "S"((x86_reg)srcStride), "D"((x86_reg)dstStride), "rm"(h)\
             : "memory"\
         );\
         src += 4-(h+5)*srcStride;\
@@ -446,7 +447,7 @@ static av_always_inline void OPNAME ## h264_qpel8or16_hv1_lowpass_ ## MMX(int16_
             QPEL_H264HV(%%mm3, %%mm4, %%mm5, %%mm0, %%mm1, %%mm2, 15*48)\
             "2:                     \n\t"\
             : "+a"(src)\
-            : "c"(tmp), "S"((x86_reg)srcStride), "g"(size)\
+            : "c"(tmp), "S"((x86_reg)srcStride), "rm"(size)\
             : "memory"\
             );\
         tmp += 4;\
@@ -823,7 +824,7 @@ static av_noinline void OPNAME ## h264_qpel8or16_v_lowpass_ ## MMX(uint8_t *dst,
         "2:                          \n\t"\
         \
         : "+a"(src), "+c"(dst)\
-        : "S"((x86_reg)srcStride), "D"((x86_reg)dstStride), "g"(h)\
+        : "S"((x86_reg)srcStride), "D"((x86_reg)dstStride), "rm"(h)\
         : XMM_CLOBBERS("%xmm0", "%xmm1", "%xmm2", "%xmm3", \
                        "%xmm4", "%xmm5", "%xmm6", "%xmm7",)\
           "memory"\
@@ -878,7 +879,7 @@ static av_always_inline void put_h264_qpel8or16_hv1_lowpass_sse2(int16_t *tmp, u
             QPEL_H264HV_XMM(%%xmm3, %%xmm4, %%xmm5, %%xmm0, %%xmm1, %%xmm2, 15*48)
             "2:                         \n\t"
             : "+a"(src)
-            : "c"(tmp), "S"((x86_reg)srcStride), "g"(size)
+            : "c"(tmp), "S"((x86_reg)srcStride), "rm"(size)
             : XMM_CLOBBERS("%xmm0", "%xmm1", "%xmm2", "%xmm3",
                            "%xmm4", "%xmm5", "%xmm6", "%xmm7",)
               "memory"
@@ -1199,3 +1200,100 @@ H264_MC_816(H264_MC_HV, sse2)
 H264_MC_816(H264_MC_H, ssse3)
 H264_MC_816(H264_MC_HV, ssse3)
 #endif
+
+
+
+//10bit
+#define LUMA_MC_OP(OP, NUM, DEPTH, TYPE, OPT) \
+void ff_ ## OP ## _h264_qpel ## NUM ## _ ## TYPE ## _ ## DEPTH ## _ ## OPT \
+    (uint8_t *dst, uint8_t *src, int stride);
+
+#define LUMA_MC_ALL(DEPTH, TYPE, OPT) \
+    LUMA_MC_OP(put,  4, DEPTH, TYPE, OPT) \
+    LUMA_MC_OP(avg,  4, DEPTH, TYPE, OPT) \
+    LUMA_MC_OP(put,  8, DEPTH, TYPE, OPT) \
+    LUMA_MC_OP(avg,  8, DEPTH, TYPE, OPT) \
+    LUMA_MC_OP(put, 16, DEPTH, TYPE, OPT) \
+    LUMA_MC_OP(avg, 16, DEPTH, TYPE, OPT)
+
+#define LUMA_MC_816(DEPTH, TYPE, OPT) \
+    LUMA_MC_OP(put,  8, DEPTH, TYPE, OPT) \
+    LUMA_MC_OP(avg,  8, DEPTH, TYPE, OPT) \
+    LUMA_MC_OP(put, 16, DEPTH, TYPE, OPT) \
+    LUMA_MC_OP(avg, 16, DEPTH, TYPE, OPT)
+
+LUMA_MC_ALL(10, mc00, mmxext)
+LUMA_MC_ALL(10, mc10, mmxext)
+LUMA_MC_ALL(10, mc20, mmxext)
+LUMA_MC_ALL(10, mc30, mmxext)
+LUMA_MC_ALL(10, mc01, mmxext)
+LUMA_MC_ALL(10, mc11, mmxext)
+LUMA_MC_ALL(10, mc21, mmxext)
+LUMA_MC_ALL(10, mc31, mmxext)
+LUMA_MC_ALL(10, mc02, mmxext)
+LUMA_MC_ALL(10, mc12, mmxext)
+LUMA_MC_ALL(10, mc22, mmxext)
+LUMA_MC_ALL(10, mc32, mmxext)
+LUMA_MC_ALL(10, mc03, mmxext)
+LUMA_MC_ALL(10, mc13, mmxext)
+LUMA_MC_ALL(10, mc23, mmxext)
+LUMA_MC_ALL(10, mc33, mmxext)
+
+LUMA_MC_816(10, mc00, sse2)
+LUMA_MC_816(10, mc10, sse2)
+LUMA_MC_816(10, mc10, sse2_cache64)
+LUMA_MC_816(10, mc10, ssse3_cache64)
+LUMA_MC_816(10, mc20, sse2)
+LUMA_MC_816(10, mc20, sse2_cache64)
+LUMA_MC_816(10, mc20, ssse3_cache64)
+LUMA_MC_816(10, mc30, sse2)
+LUMA_MC_816(10, mc30, sse2_cache64)
+LUMA_MC_816(10, mc30, ssse3_cache64)
+LUMA_MC_816(10, mc01, sse2)
+LUMA_MC_816(10, mc11, sse2)
+LUMA_MC_816(10, mc21, sse2)
+LUMA_MC_816(10, mc31, sse2)
+LUMA_MC_816(10, mc02, sse2)
+LUMA_MC_816(10, mc12, sse2)
+LUMA_MC_816(10, mc22, sse2)
+LUMA_MC_816(10, mc32, sse2)
+LUMA_MC_816(10, mc03, sse2)
+LUMA_MC_816(10, mc13, sse2)
+LUMA_MC_816(10, mc23, sse2)
+LUMA_MC_816(10, mc33, sse2)
+
+#define QPEL16_OPMC(OP, MC, MMX)\
+void ff_ ## OP ## _h264_qpel16_ ## MC ## _10_ ## MMX(uint8_t *dst, uint8_t *src, int stride){\
+    ff_ ## OP ## _h264_qpel8_ ## MC ## _10_ ## MMX(dst   , src   , stride);\
+    ff_ ## OP ## _h264_qpel8_ ## MC ## _10_ ## MMX(dst+16, src+16, stride);\
+    src += 8*stride;\
+    dst += 8*stride;\
+    ff_ ## OP ## _h264_qpel8_ ## MC ## _10_ ## MMX(dst   , src   , stride);\
+    ff_ ## OP ## _h264_qpel8_ ## MC ## _10_ ## MMX(dst+16, src+16, stride);\
+}
+
+#define QPEL16_OP(MC, MMX)\
+QPEL16_OPMC(put, MC, MMX)\
+QPEL16_OPMC(avg, MC, MMX)
+
+#define QPEL16(MMX)\
+QPEL16_OP(mc00, MMX)\
+QPEL16_OP(mc01, MMX)\
+QPEL16_OP(mc02, MMX)\
+QPEL16_OP(mc03, MMX)\
+QPEL16_OP(mc10, MMX)\
+QPEL16_OP(mc11, MMX)\
+QPEL16_OP(mc12, MMX)\
+QPEL16_OP(mc13, MMX)\
+QPEL16_OP(mc20, MMX)\
+QPEL16_OP(mc21, MMX)\
+QPEL16_OP(mc22, MMX)\
+QPEL16_OP(mc23, MMX)\
+QPEL16_OP(mc30, MMX)\
+QPEL16_OP(mc31, MMX)\
+QPEL16_OP(mc32, MMX)\
+QPEL16_OP(mc33, MMX)
+
+#if ARCH_X86_32 && HAVE_YASM // ARCH_X86_64 implies sse2+
+QPEL16(mmxext)
+#endif