x86/hevc: add 12bits support for MC

cherry picked from commit 3fcb7a4595a6f40100a22110a5805e3b7510c0fd

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

1 files changed, 58 insertions, 5 deletions

diff --git a/libavcodec/x86/hevc_mc.asm b/libavcodec/x86/hevc_mc.asm
index 5cf37d0..81b7d9e 100644
--- a/libavcodec/x86/hevc_mc.asm
+++ b/libavcodec/x86/hevc_mc.asm
@@ -21,11 +21,14 @@
 %include "libavutil/x86/x86util.asm"
 
 SECTION_RODATA
-pw_8:                   times 8 dw 512
-pw_10:                  times 8 dw 2048
-pw_bi_8:                times 8 dw 256
-pw_bi_10:               times 8 dw 1024
-max_pixels_10:          times 8  dw 1023
+pw_8:                   times 8 dw (1 << 9)
+pw_10:                  times 8 dw (1 << 11)
+pw_12:                  times 8 dw (1 << 13)
+pw_bi_8:                times 8 dw (1 << 8)
+pw_bi_10:               times 8 dw (1 << 10)
+pw_bi_12:               times 8 dw (1 << 12)
+max_pixels_10:          times 8  dw ((1 << 10)-1)
+max_pixels_12:          times 8  dw ((1 << 12)-1)
 zero:                   times 4  dd 0
 one_per_32:             times 4  dd 1
 
@@ -51,6 +54,7 @@ hevc_epel_filters_%4_%1 times %2 d%3 -2, 58
 
 EPEL_TABLE  8, 8, b, sse4
 EPEL_TABLE 10, 4, w, sse4
+EPEL_TABLE 12, 4, w, sse4
 
 %macro QPEL_TABLE 4
 hevc_qpel_filters_%4_%1 times %2 d%3  -1,  4
@@ -69,6 +73,7 @@ hevc_qpel_filters_%4_%1 times %2 d%3  -1,  4
 
 QPEL_TABLE  8, 8, b, sse4
 QPEL_TABLE 10, 4, w, sse4
+QPEL_TABLE 12, 4, w, sse4
 
 %define hevc_qpel_filters_sse4_14 hevc_qpel_filters_sse4_10
 
@@ -295,6 +300,29 @@ QPEL_TABLE 10, 4, w, sse4
 %endif
 %endmacro
 
+%macro PEL_12STORE2 3
+    movd           [%1], %2
+%endmacro
+%macro PEL_12STORE4 3
+    movq           [%1], %2
+%endmacro
+%macro PEL_12STORE6 3
+    movq           [%1], %2
+    psrldq            %2, 8
+    movd         [%1+8], %2
+%endmacro
+%macro PEL_12STORE8 3
+    movdqa         [%1], %2
+%endmacro
+%macro PEL_12STORE12 3
+    movdqa         [%1], %2
+    movq        [%1+16], %3
+%endmacro
+%macro PEL_12STORE16 3
+    PEL_12STORE8      %1, %2, %3
+    movdqa       [%1+16], %3
+%endmacro
+
 %macro PEL_10STORE2 3
     movd           [%1], %2
 %endmacro
@@ -1229,6 +1257,11 @@ WEIGHTING_FUNCS 4, 10
 WEIGHTING_FUNCS 6, 10
 WEIGHTING_FUNCS 8, 10
 
+WEIGHTING_FUNCS 2, 12
+WEIGHTING_FUNCS 4, 12
+WEIGHTING_FUNCS 6, 12
+WEIGHTING_FUNCS 8, 12
+
 HEVC_PUT_HEVC_PEL_PIXELS  2, 8
 HEVC_PUT_HEVC_PEL_PIXELS  4, 8
 HEVC_PUT_HEVC_PEL_PIXELS  6, 8
@@ -1241,6 +1274,10 @@ HEVC_PUT_HEVC_PEL_PIXELS 4, 10
 HEVC_PUT_HEVC_PEL_PIXELS 6, 10
 HEVC_PUT_HEVC_PEL_PIXELS 8, 10
 
+HEVC_PUT_HEVC_PEL_PIXELS 2, 12
+HEVC_PUT_HEVC_PEL_PIXELS 4, 12
+HEVC_PUT_HEVC_PEL_PIXELS 6, 12
+HEVC_PUT_HEVC_PEL_PIXELS 8, 12
 
 HEVC_PUT_HEVC_EPEL 2,  8
 HEVC_PUT_HEVC_EPEL 4,  8
@@ -1255,6 +1292,10 @@ HEVC_PUT_HEVC_EPEL 4, 10
 HEVC_PUT_HEVC_EPEL 6, 10
 HEVC_PUT_HEVC_EPEL 8, 10
 
+HEVC_PUT_HEVC_EPEL 2, 12
+HEVC_PUT_HEVC_EPEL 4, 12
+HEVC_PUT_HEVC_EPEL 6, 12
+HEVC_PUT_HEVC_EPEL 8, 12
 
 HEVC_PUT_HEVC_EPEL_HV 2,  8
 HEVC_PUT_HEVC_EPEL_HV 4,  8
@@ -1266,6 +1307,10 @@ HEVC_PUT_HEVC_EPEL_HV 4, 10
 HEVC_PUT_HEVC_EPEL_HV 6, 10
 HEVC_PUT_HEVC_EPEL_HV 8, 10
 
+HEVC_PUT_HEVC_EPEL_HV 2, 12
+HEVC_PUT_HEVC_EPEL_HV 4, 12
+HEVC_PUT_HEVC_EPEL_HV 6, 12
+HEVC_PUT_HEVC_EPEL_HV 8, 12
 
 HEVC_PUT_HEVC_QPEL 4,  8
 HEVC_PUT_HEVC_QPEL 8,  8
@@ -1275,6 +1320,9 @@ HEVC_PUT_HEVC_QPEL 16, 8
 HEVC_PUT_HEVC_QPEL 4, 10
 HEVC_PUT_HEVC_QPEL 8, 10
 
+HEVC_PUT_HEVC_QPEL 4, 12
+HEVC_PUT_HEVC_QPEL 8, 12
+
 HEVC_PUT_HEVC_QPEL_HV 2, 8
 HEVC_PUT_HEVC_QPEL_HV 4, 8
 HEVC_PUT_HEVC_QPEL_HV 6, 8
@@ -1285,4 +1333,9 @@ HEVC_PUT_HEVC_QPEL_HV 4, 10
 HEVC_PUT_HEVC_QPEL_HV 6, 10
 HEVC_PUT_HEVC_QPEL_HV 8, 10
 
+HEVC_PUT_HEVC_QPEL_HV 2, 12
+HEVC_PUT_HEVC_QPEL_HV 4, 12
+HEVC_PUT_HEVC_QPEL_HV 6, 12
+HEVC_PUT_HEVC_QPEL_HV 8, 12
+
 %endif ; ARCH_X86_64