[x265-commits] [x265] doc: make a note about how to ensure the version number i...

Steve Borho steve at borho.org
Fri Oct 31 19:53:39 CET 2014


details:   http://hg.videolan.org/x265/rev/f807d346663e
branches:  
changeset: 8762:f807d346663e
user:      Steve Borho <steve at borho.org>
date:      Thu Oct 30 15:42:27 2014 -0500
description:
doc: make a note about how to ensure the version number is accurate
Subject: [x265] lowres: save intra mode in lowres

details:   http://hg.videolan.org/x265/rev/7047fec7140a
branches:  
changeset: 8763:7047fec7140a
user:      Nicolas Morey-Chaisemartin <nmorey at kalray.eu>
date:      Thu Oct 30 11:27:51 2014 +0100
description:
lowres: save intra mode in lowres
Subject: [x265] asm: replace constant table tab_c_128 by pb_128

details:   http://hg.videolan.org/x265/rev/7bb7f03d3e6b
branches:  
changeset: 8764:7bb7f03d3e6b
user:      Min Chen <chenm003 at 163.com>
date:      Thu Oct 30 15:46:09 2014 -0700
description:
asm: replace constant table tab_c_128 by pb_128
Subject: [x265] asm: replace constant table tab_c_512 by pw_512

details:   http://hg.videolan.org/x265/rev/b39313659e71
branches:  
changeset: 8765:b39313659e71
user:      Min Chen <chenm003 at 163.com>
date:      Thu Oct 30 15:46:23 2014 -0700
description:
asm: replace constant table tab_c_512 by pw_512
Subject: [x265] asm: AVX2 version luma_vpp[4x4], improve 391c -> 302c

details:   http://hg.videolan.org/x265/rev/f0f073deb207
branches:  
changeset: 8766:f0f073deb207
user:      Min Chen <chenm003 at 163.com>
date:      Thu Oct 30 15:46:40 2014 -0700
description:
asm: AVX2 version luma_vpp[4x4], improve 391c -> 302c
Subject: [x265] slicetype: fix an msvc warning

details:   http://hg.videolan.org/x265/rev/c369e6a12710
branches:  
changeset: 8767:c369e6a12710
user:      Steve Borho <steve at borho.org>
date:      Thu Oct 30 17:48:11 2014 -0500
description:
slicetype: fix an msvc warning
Subject: [x265] encoder: emit an Active Parameter Sets SEI in stream headers if interlaced

details:   http://hg.videolan.org/x265/rev/5e604833c5aa
branches:  stable
changeset: 8768:5e604833c5aa
user:      Steve Borho <steve at borho.org>
date:      Fri Oct 31 13:24:01 2014 -0500
description:
encoder: emit an Active Parameter Sets SEI in stream headers if interlaced

The APS is technically required if we're going to be emitting picture timing SEI
which we do for HRD and for interlaced inputs. The lack of APS for interlaced
content was causing a warning from the HM decoder and on Windows the decoder
would later crash.
Subject: [x265] Added tag 1.4 for changeset 5e604833c5aa

details:   http://hg.videolan.org/x265/rev/d2db9c1ab44b
branches:  stable
changeset: 8769:d2db9c1ab44b
user:      Steve Borho <steve at borho.org>
date:      Fri Oct 31 13:52:08 2014 -0500
description:
Added tag 1.4 for changeset 5e604833c5aa
Subject: [x265] Merge with stable

details:   http://hg.videolan.org/x265/rev/ae8a661acdc4
branches:  
changeset: 8770:ae8a661acdc4
user:      Steve Borho <steve at borho.org>
date:      Fri Oct 31 13:53:15 2014 -0500
description:
Merge with stable

diffstat:

 .hgtags                              |    1 +
 doc/reST/api.rst                     |    6 +
 source/common/lowres.cpp             |    2 +
 source/common/lowres.h               |    1 +
 source/common/x86/asm-primitives.cpp |    1 +
 source/common/x86/ipfilter8.asm      |  147 +++++++++++++++++++++++++++-------
 source/encoder/encoder.cpp           |    2 +-
 source/encoder/slicetype.cpp         |    1 +
 8 files changed, 127 insertions(+), 34 deletions(-)

diffs (truncated from 482 to 300 lines):

diff -r de28d1b07e6f -r ae8a661acdc4 .hgtags
--- a/.hgtags	Thu Oct 30 13:09:57 2014 -0500
+++ b/.hgtags	Fri Oct 31 13:53:15 2014 -0500
@@ -15,3 +15,4 @@ cea97c4d79456842e00ade6be6fd5ec34610e5f8
 ae9609aeebdc3271114168ece003679e9b1dca1b 1.1
 d6257335c5370ee54317a0426a12c1f0724b18b9 1.2
 c1e4fc0162c14fdb84f5c3bd404fb28cfe10a17f 1.3
+5e604833c5aa605d0b6efbe5234492b5e7d8ac61 1.4
diff -r de28d1b07e6f -r ae8a661acdc4 doc/reST/api.rst
--- a/doc/reST/api.rst	Thu Oct 30 13:09:57 2014 -0500
+++ b/doc/reST/api.rst	Fri Oct 31 13:53:15 2014 -0500
@@ -32,6 +32,12 @@ library was compiled (it will contain a 
 x265 which was compiled, and **x265_build_info_str** is a pointer to a
 string identifying the compiler and build options.
 
+.. Note::
+
+	**x265_version_str** is only updated when **cmake** runs. If you are
+	making binaries for others to use, it is recommended to run
+	**cmake** prior to **make** in your build scripts.
+
 x265 will accept input pixels of any depth between 8 and 16 bits
 regardless of the depth of its internal pixels (8 or 10).  It will shift
 and mask input pixels as required to reach the internal depth. If
diff -r de28d1b07e6f -r ae8a661acdc4 source/common/lowres.cpp
--- a/source/common/lowres.cpp	Thu Oct 30 13:09:57 2014 -0500
+++ b/source/common/lowres.cpp	Fri Oct 31 13:53:15 2014 -0500
@@ -69,6 +69,7 @@ bool Lowres::create(PicYuv *origPic, int
     lowresPlane[3] = buffer[3] + padoffset;
 
     CHECKED_MALLOC(intraCost, int32_t, cuCount);
+    CHECKED_MALLOC(intraMode, uint8_t, cuCount);
 
     for (int i = 0; i < bframes + 2; i++)
     {
@@ -99,6 +100,7 @@ void Lowres::destroy()
         X265_FREE(buffer[i]);
 
     X265_FREE(intraCost);
+    X265_FREE(intraMode);
 
     for (int i = 0; i < bframes + 2; i++)
     {
diff -r de28d1b07e6f -r ae8a661acdc4 source/common/lowres.h
--- a/source/common/lowres.h	Thu Oct 30 13:09:57 2014 -0500
+++ b/source/common/lowres.h	Fri Oct 31 13:53:15 2014 -0500
@@ -116,6 +116,7 @@ struct Lowres : public ReferencePlanes
     int32_t*  rowSatds[X265_BFRAME_MAX + 2][X265_BFRAME_MAX + 2];
     int       intraMbs[X265_BFRAME_MAX + 2];
     int32_t*  intraCost;
+    uint8_t*  intraMode;
     int64_t   satdCost;
     uint16_t* lowresCostForRc;
     uint16_t(*lowresCosts[X265_BFRAME_MAX + 2][X265_BFRAME_MAX + 2]);
diff -r de28d1b07e6f -r ae8a661acdc4 source/common/x86/asm-primitives.cpp
--- a/source/common/x86/asm-primitives.cpp	Thu Oct 30 13:09:57 2014 -0500
+++ b/source/common/x86/asm-primitives.cpp	Fri Oct 31 13:53:15 2014 -0500
@@ -1799,6 +1799,7 @@ void Setup_Assembly_Primitives(EncoderPr
         p.transpose[BLOCK_64x64] = x265_transpose64_avx2;
 #endif
         p.luma_hpp[LUMA_4x4] = x265_interp_8tap_horiz_pp_4x4_avx2;
+        p.luma_vpp[LUMA_4x4] = x265_interp_8tap_vert_pp_4x4_avx2;
     }
 #endif // if HIGH_BIT_DEPTH
 }
diff -r de28d1b07e6f -r ae8a661acdc4 source/common/x86/ipfilter8.asm
--- a/source/common/x86/ipfilter8.asm	Thu Oct 30 13:09:57 2014 -0500
+++ b/source/common/x86/ipfilter8.asm	Fri Oct 31 13:53:15 2014 -0500
@@ -42,7 +42,6 @@ tab_Vm:    db 0, 1, 0, 1, 0, 1, 0, 1, 0,
 
 tab_Cm:    db 0, 2, 1, 3, 0, 2, 1, 3, 0, 2, 1, 3, 0, 2, 1, 3
 
-tab_c_512:      times 8 dw 512
 tab_c_526336:   times 4 dd 8192*64+2048
 
 tab_ChromaCoeff: db  0, 64,  0,  0
@@ -123,13 +122,13 @@ tab_LumaCoeffVer: times 8 db 0, 0
                   times 8 db 58, -10
                   times 8 db 4, -1
 
-tab_c_128:      times 16 db 0x80
 tab_c_64_n64:   times 8 db 64, -64
 
 
 SECTION .text
 
 cextern idct4_shuf1
+cextern pb_128
 cextern pw_1
 cextern pw_512
 cextern pw_2000
@@ -171,7 +170,7 @@ movd        coef2,       [tab_ChromaCoef
 %endif
 
 pshufd      coef2,       coef2,      0
-mova        t2,          [tab_c_512]
+mova        t2,          [pw_512]
 mova        Tm0,         [tab_Tm]
 
 %rep 2
@@ -203,7 +202,7 @@ movd        coef2,       [tab_ChromaCoef
 %endif
 
 pshufd      coef2,       coef2,      0
-mova        t2,          [tab_c_512]
+mova        t2,          [pw_512]
 mova        Tm0,         [tab_Tm]
 
 %rep 4
@@ -235,7 +234,7 @@ movd        coef2,       [tab_ChromaCoef
 %endif
 
 pshufd      coef2,       coef2,      0
-mova        t2,          [tab_c_512]
+mova        t2,          [pw_512]
 mova        Tm0,         [tab_Tm]
 
 mov         r5d,        16/2
@@ -285,7 +284,7 @@ movd        coef2,       [tab_ChromaCoef
 %endif
 
 pshufd      coef2,       coef2,      0
-mova        t2,          [tab_c_512]
+mova        t2,          [pw_512]
 mova        Tm0,         [tab_Tm]
 
 FILTER_H4_w4_2   t0, t1, t2
@@ -313,7 +312,7 @@ movd        coef2,       [tab_ChromaCoef
 %endif
 
 pshufd      coef2,       coef2,      0
-mova        t2,          [tab_c_512]
+mova        t2,          [pw_512]
 mova        Tm0,         [tab_Tm]
 
 %rep 2
@@ -345,7 +344,7 @@ movd        coef2,       [tab_ChromaCoef
 %endif
 
 pshufd      coef2,       coef2,      0
-mova        t2,          [tab_c_512]
+mova        t2,          [pw_512]
 mova        Tm0,         [tab_Tm]
 
 %rep 4
@@ -377,7 +376,7 @@ movd        coef2,       [tab_ChromaCoef
 %endif
 
 pshufd      coef2,       coef2,      0
-mova        t2,          [tab_c_512]
+mova        t2,          [pw_512]
 mova        Tm0,         [tab_Tm]
 
 %rep 8
@@ -409,7 +408,7 @@ movd        coef2,       [tab_ChromaCoef
 %endif
 
 pshufd      coef2,       coef2,      0
-mova        t2,          [tab_c_512]
+mova        t2,          [pw_512]
 mova        Tm0,         [tab_Tm]
 
 mov         r5d,        32/2
@@ -606,7 +605,7 @@ movd        coef2,       [tab_ChromaCoef
 mov           r5d,       %2
 
 pshufd      coef2,       coef2,      0
-mova        t2,          [tab_c_512]
+mova        t2,          [pw_512]
 mova        Tm0,         [tab_Tm]
 mova        Tm1,         [tab_Tm + 16]
 
@@ -662,7 +661,7 @@ movd        coef2,       [tab_ChromaCoef
 mov         r5d,          %2
 
 pshufd      coef2,       coef2,      0
-mova        t2,          [tab_c_512]
+mova        t2,          [pw_512]
 mova        Tm0,         [tab_Tm]
 mova        Tm1,         [tab_Tm + 16]
 
@@ -749,7 +748,7 @@ cglobal interp_8tap_horiz_%3_%1x%2, 4,7,
     punpcklqdq  m3, m3
 
 %ifidn %3, pp 
-    mova      m2, [tab_c_512]
+    mova      m2, [pw_512]
 %else
     mova      m2, [pw_2000]
 %endif
@@ -1040,7 +1039,7 @@ cglobal interp_8tap_hv_pp_8x8, 4, 7, 8, 
     mov         r4,         rsp
 
 .loopH:
-    FILTER_H8_W8 m0, m1, m2, m3, coef, [tab_c_512], [r0 - 3]
+    FILTER_H8_W8 m0, m1, m2, m3, coef, [pw_512], [r0 - 3]
     psubw       m1,         [pw_2000]
     mova        [r4],       m1
 
@@ -1108,7 +1107,7 @@ movd        m0,        [tab_ChromaCoeff 
 lea         r4,        [r1 * 3]
 lea         r5,        [r0 + 4 * r1]
 pshufb      m0,        [tab_Cm]
-mova        m1,        [tab_c_512]
+mova        m1,        [pw_512]
 
 movd        m2,        [r0]
 movd        m3,        [r0 + r1]
@@ -1181,7 +1180,7 @@ movd        m0,        [tab_ChromaCoeff 
 
 pshufb      m0,        [tab_Cm]
 
-mova        m1,        [tab_c_512]
+mova        m1,        [pw_512]
 
 mov         r4d,       %2
 lea         r5,        [3 * r1]
@@ -1289,7 +1288,7 @@ pmaddubsw   m3,        m0
 
 phaddw      m2,        m3
 
-pmulhrsw    m2,        [tab_c_512]
+pmulhrsw    m2,        [pw_512]
 packuswb    m2,        m2
 movd        [r2],      m2
 pextrd      [r2 + r3], m2,  1
@@ -1313,7 +1312,7 @@ movd        m0,        [tab_ChromaCoeff 
 %endif
 
 pshufb      m0,        [tab_Cm]
-mova        m1,        [tab_c_512]
+mova        m1,        [pw_512]
 lea         r5,        [r0 + 4 * r1]
 lea         r4,        [r1 * 3]
 
@@ -1388,7 +1387,7 @@ movd        m0,        [tab_ChromaCoeff 
 
 pshufb      m0,        [tab_Cm]
 
-mova        m1,        [tab_c_512]
+mova        m1,        [pw_512]
 
 mov         r4d,       %2
 
@@ -1590,7 +1589,7 @@ pmaddubsw   m4,        m5
 
 paddw       m0,        m4
 
-mova        m4,        [tab_c_512]
+mova        m4,        [pw_512]
 
 pmulhrsw    m0,        m4
 packuswb    m0,        m0
@@ -2495,7 +2494,7 @@ movd        m5,        [tab_ChromaCoeff 
 
 pshufb      m6,        m5,       [tab_Vm]
 pshufb      m5,        [tab_Vm + 16]
-mova        m4,        [tab_c_512]
+mova        m4,        [pw_512]
 lea         r5,        [r1 * 3]
 
 mov         r4d,       %2
@@ -2593,7 +2592,7 @@ movd        m5,        [tab_ChromaCoeff 
 
 pshufb      m6,        m5,       [tab_Vm]
 pshufb      m5,        [tab_Vm + 16]
-mova        m4,        [tab_c_512]
+mova        m4,        [pw_512]
 
 mov         r4d,       %2
 lea         r5,        [3 * r1]
@@ -2716,7 +2715,7 @@ punpckhbw   m6,        m5,        m7
 pmaddubsw   m6,        m0
 paddw       m2,        m6
 
-mova        m6,        [tab_c_512]
+mova        m6,        [pw_512]
 
 pmulhrsw    m4,        m6
 pmulhrsw    m2,        m6
@@ -2806,7 +2805,7 @@ punpcklbw   m7,        m5,        m6
 pmaddubsw   m7,        m0
 paddw       m4,        m7
 
-mova        m7,        [tab_c_512]
+mova        m7,        [pw_512]
 
 pmulhrsw    m4,        m7
 pmulhrsw    m2,        m7
@@ -2899,7 +2898,7 @@ punpckhbw   m6,        m5,        m7
 pmaddubsw   m6,        m0
 paddw       m2,        m6
 
-mova        m6,        [tab_c_512]
+mova        m6,        [pw_512]
 
 pmulhrsw    m4,        m6
 pmulhrsw    m2,        m6
@@ -2998,7 +2997,7 @@ movd        m0,        [tab_ChromaCoeff 
 pshufb      m1,        m0,       [tab_Vm]


More information about the x265-commits mailing list