@scrypted/libav 1.0.179 → 1.0.181

package/openh264/codec/common/inc/wels_common_defs.h

@@ -0,0 +1,373 @@
+/*!
+ * \copy
+ *     Copyright (c)  2013, Cisco Systems
+ *     All rights reserved.
+ *
+ *     Redistribution and use in source and binary forms, with or without
+ *     modification, are permitted provided that the following conditions
+ *     are met:
+ *
+ *        * Redistributions of source code must retain the above copyright
+ *          notice, this list of conditions and the following disclaimer.
+ *
+ *        * Redistributions in binary form must reproduce the above copyright
+ *          notice, this list of conditions and the following disclaimer in
+ *          the documentation and/or other materials provided with the
+ *          distribution.
+ *
+ *     THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ *     "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ *     LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
+ *     FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
+ *     COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
+ *     INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING,
+ *     BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
+ *     LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
+ *     CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+ *     LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN
+ *     ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+ *     POSSIBILITY OF SUCH DAMAGE.
+ *
+ */
+
+//wels_common_defs.h
+#ifndef WELS_COMMON_DEFS_H__
+#define WELS_COMMON_DEFS_H__
+
+#include "typedefs.h"
+#include "macros.h"
+#include "codec_app_def.h"
+
+
+namespace WelsCommon {
+/*common use table*/
+
+#define  CTX_NA 0
+#define  WELS_CONTEXT_COUNT 460
+#define LEVEL_NUMBER 17
+typedef struct TagLevelLimits {
+  ELevelIdc uiLevelIdc;  // level idc
+  uint32_t uiMaxMBPS; // Max macroblock processing rate(MB/s)
+  uint32_t uiMaxFS;   // Max frame sizea(MBs)
+  uint32_t uiMaxDPBMbs;// Max decoded picture buffer size(MBs)
+  uint32_t uiMaxBR; // Max video bit rate
+  uint32_t uiMaxCPB; // Max CPB size
+  int16_t iMinVmv; // Vertical MV component range upper bound
+  int16_t iMaxVmv; // Vertical MV component range lower bound
+  uint16_t uiMinCR;  // Min compression ration
+  int16_t iMaxMvsPer2Mb; // Max number of motion vectors per two consecutive MBs
+} SLevelLimits;
+
+#define CpbBrNalFactor 1200  //baseline,main,and extended profiles.
+extern const SLevelLimits g_ksLevelLimits[LEVEL_NUMBER];
+extern const uint32_t g_kuiLevelMaps[LEVEL_NUMBER];
+extern const uint8_t g_kuiMbCountScan4Idx[24];
+extern const uint8_t g_kuiCache30ScanIdx[16];
+extern const uint8_t g_kuiCache48CountScan4Idx[24];
+
+extern const uint8_t g_kuiMatrixV[6][8][8];
+
+extern const uint8_t g_kuiDequantScaling4x4Default[2][16];
+extern const uint8_t g_kuiDequantScaling8x8Default[2][64];
+extern const ALIGNED_DECLARE (uint16_t, g_kuiDequantCoeff[52][8], 16);
+extern const ALIGNED_DECLARE (uint16_t, g_kuiDequantCoeff8x8[52][64], 16);
+extern const uint8_t g_kuiChromaQpTable[52];
+
+extern const uint8_t g_kuiCabacRangeLps[64][4];
+extern const int8_t g_kiCabacGlobalContextIdx[WELS_CONTEXT_COUNT][4][2];
+extern const uint8_t g_kuiStateTransTable[64][2];
+extern const uint32_t g_kuiGolombUELength[256];
+/*
+ *  NAL Unit Type (5 Bits)
+ */
+enum EWelsNalUnitType {
+  NAL_UNIT_UNSPEC_0             = 0,
+  NAL_UNIT_CODED_SLICE          = 1,
+  NAL_UNIT_CODED_SLICE_DPA      = 2,
+  NAL_UNIT_CODED_SLICE_DPB      = 3,
+  NAL_UNIT_CODED_SLICE_DPC      = 4,
+  NAL_UNIT_CODED_SLICE_IDR      = 5,
+  NAL_UNIT_SEI                  = 6,
+  NAL_UNIT_SPS                  = 7,
+  NAL_UNIT_PPS                  = 8,
+  NAL_UNIT_AU_DELIMITER         = 9,
+  NAL_UNIT_END_OF_SEQ           = 10,
+  NAL_UNIT_END_OF_STR           = 11,
+  NAL_UNIT_FILLER_DATA          = 12,
+  NAL_UNIT_SPS_EXT              = 13,
+  NAL_UNIT_PREFIX               = 14,
+  NAL_UNIT_SUBSET_SPS           = 15,
+  NAL_UNIT_DEPTH_PARAM          = 16, // NAL_UNIT_RESV_16
+  NAL_UNIT_RESV_17              = 17,
+  NAL_UNIT_RESV_18              = 18,
+  NAL_UNIT_AUX_CODED_SLICE      = 19,
+  NAL_UNIT_CODED_SLICE_EXT      = 20,
+  NAL_UNIT_MVC_SLICE_EXT        = 21, // NAL_UNIT_RESV_21
+  NAL_UNIT_RESV_22              = 22,
+  NAL_UNIT_RESV_23              = 23,
+  NAL_UNIT_UNSPEC_24            = 24,
+  NAL_UNIT_UNSPEC_25            = 25,
+  NAL_UNIT_UNSPEC_26            = 26,
+  NAL_UNIT_UNSPEC_27            = 27,
+  NAL_UNIT_UNSPEC_28            = 28,
+  NAL_UNIT_UNSPEC_29            = 29,
+  NAL_UNIT_UNSPEC_30            = 30,
+  NAL_UNIT_UNSPEC_31            = 31
+};
+
+/*
+ *  NAL Reference IDC (2 Bits)
+ */
+
+enum EWelsNalRefIdc {
+  NRI_PRI_LOWEST        = 0,
+  NRI_PRI_LOW           = 1,
+  NRI_PRI_HIGH          = 2,
+  NRI_PRI_HIGHEST       = 3
+};
+
+/*
+ * VCL TYPE
+ */
+
+enum EVclType {
+  NON_VCL   = 0,
+  VCL       = 1,
+  NOT_APP   = 2
+};
+
+/*
+ *  vcl type map for given NAL unit type and corresponding H264 type (0: AVC; 1: SVC).
+ */
+extern const EVclType g_keTypeMap[32][2];
+
+#define IS_VCL_NAL(t, ext_idx)                  (g_keTypeMap[t][ext_idx] == VCL)
+#define IS_PARAM_SETS_NALS(t)                   ( (t) == NAL_UNIT_SPS || (t) == NAL_UNIT_PPS || (t) == NAL_UNIT_SUBSET_SPS )
+#define IS_SPS_NAL(t)                           ( (t) == NAL_UNIT_SPS )
+#define IS_SUBSET_SPS_NAL(t)                    ( (t) == NAL_UNIT_SUBSET_SPS )
+#define IS_PPS_NAL(t)                           ( (t) == NAL_UNIT_PPS )
+#define IS_SEI_NAL(t)                           ( (t) == NAL_UNIT_SEI )
+#define IS_AU_DELIMITER_NAL(t)                  ( (t) == NAL_UNIT_AU_DELIMITER )
+#define IS_PREFIX_NAL(t)                        ( (t) == NAL_UNIT_PREFIX )
+#define IS_SUBSET_SPS_USED(t)                   ( (t) == NAL_UNIT_SUBSET_SPS || (t) == NAL_UNIT_CODED_SLICE_EXT )
+#define IS_VCL_NAL_AVC_BASE(t)                  ( (t) == NAL_UNIT_CODED_SLICE || (t) == NAL_UNIT_CODED_SLICE_IDR )
+#define IS_NEW_INTRODUCED_SVC_NAL(t)            ( (t) == NAL_UNIT_PREFIX || (t) == NAL_UNIT_CODED_SLICE_EXT )
+
+
+/* Base SSlice Types
+ * Invalid in case of eSliceType exceeds 9,
+ * Need trim when eSliceType > 4 as fixed SliceType(eSliceType-4),
+ * meaning mapped version after eSliceType minus 4.
+ */
+
+enum EWelsSliceType {
+  P_SLICE       = 0,
+  B_SLICE       = 1,
+  I_SLICE       = 2,
+  SP_SLICE      = 3,
+  SI_SLICE      = 4,
+  UNKNOWN_SLICE = 5
+};
+
+/* SSlice Types in scalable extension */
+enum ESliceTypeExt {
+  EP_SLICE = 0, // EP_SLICE: 0, 5
+  EB_SLICE = 1, // EB_SLICE: 1, 6
+  EI_SLICE = 2  // EI_SLICE: 2, 7
+};
+
+/* List Index */
+enum EListIndex {
+  LIST_0    = 0,
+  LIST_1    = 1,
+  LIST_A    = 2
+};
+
+
+
+/* Motion Vector components */
+enum EMvComp {
+  MV_X  = 0,
+  MV_Y  = 1,
+  MV_A  = 2
+};
+
+/* Chroma Components */
+
+enum EChromaComp {
+  CHROMA_CB     = 0,
+  CHROMA_CR     = 1,
+  CHROMA_A      = 2
+};
+
+
+
+/*
+ *  Memory Management Control Operation (MMCO) code
+ */
+enum EMmcoCode {
+  MMCO_END          = 0,
+  MMCO_SHORT2UNUSED = 1,
+  MMCO_LONG2UNUSED  = 2,
+  MMCO_SHORT2LONG   = 3,
+  MMCO_SET_MAX_LONG = 4,
+  MMCO_RESET        = 5,
+  MMCO_LONG         = 6
+};
+
+enum EVuiVideoFormat {
+  VUI_COMPONENT   = 0,
+  VUI_PAL         = 1,
+  VUI_NTSC        = 2,
+  VUI_SECAM       = 3,
+  VUI_MAC         = 4,
+  VUI_UNSPECIFIED = 5,
+  VUI_RESERVED1   = 6,
+  VUI_RESERVED2   = 7
+};
+
+/*
+ *  Bit-stream auxiliary reading / writing
+ */
+typedef struct TagBitStringAux {
+  uint8_t* pStartBuf;   // buffer to start position
+  uint8_t* pEndBuf;     // buffer + length
+  int32_t  iBits;       // count bits of overall bitstreaming input
+
+  intX_t   iIndex;      //only for cavlc usage
+  uint8_t* pCurBuf;     // current reading position
+  uint32_t uiCurBits;
+  int32_t  iLeftBits;   // count number of available bits left ([1, 8]),
+  // need pointer to next byte start position in case 0 bit left then 8 instead
+} SBitStringAux, *PBitStringAux;
+
+/* NAL Unix Header in AVC, refer to Page 56 in JVT X201wcm */
+typedef struct TagNalUnitHeader {
+  uint8_t             uiForbiddenZeroBit;
+  uint8_t             uiNalRefIdc;
+  EWelsNalUnitType    eNalUnitType;
+  uint8_t             uiReservedOneByte;                // only padding usage
+} SNalUnitHeader, *PNalUnitHeader;
+
+/* NAL Unit Header in scalable extension syntax, refer to Page 390 in JVT X201wcm */
+typedef struct TagNalUnitHeaderExt {
+  SNalUnitHeader      sNalUnitHeader;
+
+  // uint8_t   reserved_one_bit;
+  bool      bIdrFlag;
+  uint8_t   uiPriorityId;
+  int8_t    iNoInterLayerPredFlag;      // change as int8_t to support 3 values probably in encoder
+  uint8_t   uiDependencyId;
+
+  uint8_t   uiQualityId;
+  uint8_t   uiTemporalId;
+  bool      bUseRefBasePicFlag;
+  bool      bDiscardableFlag;
+
+  bool      bOutputFlag;
+  uint8_t   uiReservedThree2Bits;
+  // Derived variable(s)
+  uint8_t   uiLayerDqId;
+  bool      bNalExtFlag;
+} SNalUnitHeaderExt, *PNalUnitHeaderExt;
+
+/* AVC MB types*/
+#define MB_TYPE_INTRA4x4    0x00000001
+#define MB_TYPE_INTRA16x16  0x00000002
+#define MB_TYPE_INTRA8x8    0x00000004
+#define MB_TYPE_16x16       0x00000008
+#define MB_TYPE_16x8        0x00000010
+#define MB_TYPE_8x16        0x00000020
+#define MB_TYPE_8x8         0x00000040
+#define MB_TYPE_8x8_REF0    0x00000080
+#define MB_TYPE_SKIP        0x00000100
+#define MB_TYPE_INTRA_PCM   0x00000200
+#define MB_TYPE_INTRA_BL    0x00000400
+#define MB_TYPE_DIRECT      0x00000800
+#define MB_TYPE_P0L0        0x00001000
+#define MB_TYPE_P1L0        0x00002000
+#define MB_TYPE_P0L1        0x00004000
+#define MB_TYPE_P1L1        0x00008000
+#define MB_TYPE_L0        (MB_TYPE_P0L0 | MB_TYPE_P1L0)
+#define MB_TYPE_L1        (MB_TYPE_P0L1 | MB_TYPE_P1L1)
+
+#define SUB_MB_TYPE_8x8     0x00000001
+#define SUB_MB_TYPE_8x4     0x00000002
+#define SUB_MB_TYPE_4x8     0x00000004
+#define SUB_MB_TYPE_4x4     0x00000008
+
+#define MB_TYPE_INTRA     (MB_TYPE_INTRA4x4 | MB_TYPE_INTRA16x16 | MB_TYPE_INTRA8x8 | MB_TYPE_INTRA_PCM)
+#define MB_TYPE_INTER     (MB_TYPE_16x16 | MB_TYPE_16x8 | MB_TYPE_8x16 | MB_TYPE_8x8 | MB_TYPE_8x8_REF0 | MB_TYPE_SKIP | MB_TYPE_DIRECT)
+#define IS_INTRA4x4(type) ( MB_TYPE_INTRA4x4 == (type) )
+#define IS_INTRA8x8(type) ( MB_TYPE_INTRA8x8 == (type) )
+#define IS_INTRANxN(type) ( MB_TYPE_INTRA4x4 == (type) || MB_TYPE_INTRA8x8 == (type) )
+#define IS_INTRA16x16(type) ( MB_TYPE_INTRA16x16 == (type) )
+#define IS_INTRA(type) ( (type)&MB_TYPE_INTRA )
+#define IS_INTER(type) ( (type)&MB_TYPE_INTER )
+#define IS_INTER_16x16(type) ( (type)&MB_TYPE_16x16 )
+#define IS_INTER_16x8(type) ( (type)&MB_TYPE_16x8 )
+#define IS_INTER_8x16(type) ( (type)&MB_TYPE_8x16 )
+#define IS_TYPE_L0(type) ( (type)&MB_TYPE_L0 )
+#define IS_TYPE_L1(type) ( (type)&MB_TYPE_L1 )
+#define IS_DIR(a, part, list) ((a) & (MB_TYPE_P0L0<<((part)+2*(list))))
+
+
+#define IS_SKIP(type) ( ((type)&MB_TYPE_SKIP) != 0 )
+#define IS_DIRECT(type) ( ((type)&MB_TYPE_DIRECT) != 0 )
+#define IS_SVC_INTER(type) IS_INTER(type)
+#define IS_I_BL(type) ( (type) == MB_TYPE_INTRA_BL )
+#define IS_SVC_INTRA(type) ( IS_I_BL(type) || IS_INTRA(type) )
+#define IS_Inter_8x8(type) ( ((type)&MB_TYPE_8x8) != 0)
+#define IS_SUB_8x8(sub_type) (((sub_type)&SUB_MB_TYPE_8x8) != 0)
+#define IS_SUB_8x4(sub_type) (((sub_type)&SUB_MB_TYPE_8x4) != 0)
+#define IS_SUB_4x8(sub_type) (((sub_type)&SUB_MB_TYPE_4x8) != 0)
+#define IS_SUB_4x4(sub_type) (((sub_type)&SUB_MB_TYPE_4x4) != 0)
+
+#define REF_NOT_AVAIL   -2
+#define REF_NOT_IN_LIST -1  //intra
+
+/////////intra16x16  Luma
+#define I16_PRED_INVALID   -1
+#define I16_PRED_V       0
+#define I16_PRED_H       1
+#define I16_PRED_DC      2
+#define I16_PRED_P       3
+
+#define I16_PRED_DC_L    4
+#define I16_PRED_DC_T    5
+#define I16_PRED_DC_128  6
+#define I16_PRED_DC_A  7
+//////////intra4x4   Luma
+// Here, I8x8 also use these definitions
+#define I4_PRED_INVALID    0
+#define I4_PRED_V        0
+#define I4_PRED_H        1
+#define I4_PRED_DC       2
+#define I4_PRED_DDL      3 //diagonal_down_left
+#define I4_PRED_DDR      4 //diagonal_down_right
+#define I4_PRED_VR       5 //vertical_right
+#define I4_PRED_HD       6 //horizon_down
+#define I4_PRED_VL       7 //vertical_left
+#define I4_PRED_HU       8 //horizon_up
+
+#define I4_PRED_DC_L     9
+#define I4_PRED_DC_T     10
+#define I4_PRED_DC_128   11
+
+#define I4_PRED_DDL_TOP  12 //right-top replacing by padding rightmost pixel of top
+#define I4_PRED_VL_TOP   13 //right-top replacing by padding rightmost pixel of top
+#define I4_PRED_A   14
+
+//////////intra Chroma
+#define C_PRED_INVALID   -1
+#define C_PRED_DC        0
+#define C_PRED_H         1
+#define C_PRED_V         2
+#define C_PRED_P         3
+
+#define C_PRED_DC_L      4
+#define C_PRED_DC_T      5
+#define C_PRED_DC_128    6
+#define C_PRED_A    7
+}
+#endif//WELS_COMMON_DEFS_H__

package/openh264/codec/common/inc/wels_const_common.h

@@ -0,0 +1,63 @@
+/*!
+ * \copy
+ *     Copyright (c)  2013, Cisco Systems
+ *     All rights reserved.
+ *
+ *     Redistribution and use in source and binary forms, with or without
+ *     modification, are permitted provided that the following conditions
+ *     are met:
+ *
+ *        * Redistributions of source code must retain the above copyright
+ *          notice, this list of conditions and the following disclaimer.
+ *
+ *        * Redistributions in binary form must reproduce the above copyright
+ *          notice, this list of conditions and the following disclaimer in
+ *          the documentation and/or other materials provided with the
+ *          distribution.
+ *
+ *     THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ *     "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ *     LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
+ *     FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
+ *     COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
+ *     INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING,
+ *     BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
+ *     LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
+ *     CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+ *     LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN
+ *     ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+ *     POSSIBILITY OF SUCH DAMAGE.
+ *
+ */
+
+#ifndef WELS_CONST_COMMON_H__
+#define WELS_CONST_COMMON_H__
+
+// Miscellaneous sizing infos
+#ifndef MAX_FNAME_LEN
+#define MAX_FNAME_LEN           256     // maximal length of file name in char size
+#endif//MAX_FNAME_LEN
+
+#ifndef WELS_LOG_BUF_SIZE
+#define WELS_LOG_BUF_SIZE       4096
+#endif//WELS_LOG_BUF_SIZE
+
+#ifndef MAX_TRACE_LOG_SIZE
+#define MAX_TRACE_LOG_SIZE      (50 * (1<<20))  // max trace log size: 50 MB, overwrite occur if log file size exceeds this size
+#endif//MAX_TRACE_LOG_SIZE
+
+/* MB width in pixels for specified colorspace I420 usually used in codec */
+#define MB_WIDTH_LUMA           16
+#define MB_WIDTH_CHROMA         (MB_WIDTH_LUMA>>1)
+/* MB height in pixels for specified colorspace I420 usually used in codec */
+#define MB_HEIGHT_LUMA          16
+#define MB_HEIGHT_CHROMA        (MB_HEIGHT_LUMA>>1)
+#define MB_COEFF_LIST_SIZE      (256+((MB_WIDTH_CHROMA*MB_HEIGHT_CHROMA)<<1))
+#define MB_PARTITION_SIZE       4       // Macroblock partition size in 8x8 sub-blocks
+#define MB_BLOCK4x4_NUM         16
+#define MB_BLOCK8x8_NUM         4
+#define MAX_SPS_COUNT           32      // Count number of SPS
+#define BASE_QUALITY_ID         0
+
+
+#endif//WELS_CONST_COMMON_H__

package/openh264/codec/common/loongarch/copy_mb_lsx.c

@@ -0,0 +1,194 @@
+/*!
+ * \copy
+ *     Copyright (c)  2009-2018, Cisco Systems
+ *     All rights reserved.
+ *
+ *     Redistribution and use in source and binary forms, with or without
+ *     modification, are permitted provided that the following conditions
+ *     are met:
+ *
+ *        * Redistributions of source code must retain the above copyright
+ *          notice, this list of conditions and the following disclaimer.
+ *
+ *        * Redistributions in binary form must reproduce the above copyright
+ *          notice, this list of conditions and the following disclaimer in
+ *          the documentation and/or other materials provided with the
+ *          distribution.
+ *
+ *     THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ *     "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ *     LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
+ *     FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
+ *     COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
+ *     INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING,
+ *     BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
+ *     LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
+ *     CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+ *     LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN
+ *     ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+ *     POSSIBILITY OF SUCH DAMAGE.
+ *
+ *
+ * \file    copy_mb_lsx.c
+ *
+ * \brief   Loongson optimization
+ *
+ * \date    12/18/2021 Created
+ *
+ *************************************************************************************
+ */
+
+#include <stdint.h>
+#include "loongson_intrinsics.h"
+
+void WelsCopy8x8_lsx (uint8_t* pDst, int32_t iStrideD,
+                      uint8_t* pSrc, int32_t iStrideS) {
+
+  int32_t iStride0 = 0;
+  int32_t iStride1 = iStrideS << 1;
+  int32_t iStride2 = iStride1 << 1;
+
+  __m128i src0, src1, src2, src3, src4 ,src5, src6, src7;
+
+  DUP4_ARG2(__lsx_vldx,
+            pSrc, iStride0,
+            pSrc, iStrideS,
+            pSrc, iStride1,
+            pSrc, iStride1 + iStrideS,
+            src0, src1, src2, src3);
+  pSrc += iStride2;
+  DUP4_ARG2(__lsx_vldx,
+            pSrc, iStride0,
+            pSrc, iStrideS,
+            pSrc, iStride1,
+            pSrc, iStride1 + iStrideS,
+            src4, src5, src6, src7);
+
+  iStride1 = iStrideD << 1;
+
+  __lsx_vstelm_d(src0, pDst, 0, 0);
+  __lsx_vstelm_d(src1, pDst + iStrideD, 0, 0);
+  pDst += iStride1;
+  __lsx_vstelm_d(src2, pDst, 0, 0);
+  __lsx_vstelm_d(src3, pDst + iStrideD, 0, 0);
+  pDst += iStride1;
+  __lsx_vstelm_d(src4, pDst, 0, 0);
+  __lsx_vstelm_d(src5, pDst + iStrideD, 0, 0);
+  pDst += iStride1;
+  __lsx_vstelm_d(src6, pDst, 0, 0);
+  __lsx_vstelm_d(src7, pDst + iStrideD, 0, 0);
+}
+
+void WelsCopy16x16_lsx (uint8_t* pDst, int32_t iStrideD,
+                        uint8_t* pSrc, int32_t iStrideS) {
+  int32_t iStride0 = 0;
+  int32_t iStride1 = iStrideS;
+  int32_t iStride2 = iStrideS << 1;
+  int32_t iStride3 = iStride2 + iStrideS;
+  int32_t iStride4 = iStrideS << 2;
+
+  __m128i src0, src1, src2, src3, src4, src5, src6, src7;
+  __m128i src8, src9, src10, src11, src12, src13, src14, src15;
+
+  DUP4_ARG2(__lsx_vldx,
+            pSrc, iStride0, pSrc, iStride1,
+            pSrc, iStride2, pSrc, iStride3,
+            src0, src1, src2, src3);
+  pSrc += iStride4;
+  DUP4_ARG2(__lsx_vldx,
+            pSrc, iStride0, pSrc, iStride1,
+            pSrc, iStride2, pSrc, iStride3,
+            src4, src5, src6, src7);
+  pSrc += iStride4;
+  DUP4_ARG2(__lsx_vldx,
+            pSrc, iStride0, pSrc, iStride1,
+            pSrc, iStride2, pSrc, iStride3,
+            src8, src9, src10, src11);
+  pSrc += iStride4;
+  DUP4_ARG2(__lsx_vldx,
+            pSrc, iStride0, pSrc, iStride1,
+            pSrc, iStride2, pSrc, iStride3,
+            src12, src13, src14, src15);
+
+  iStride1 = iStrideD;
+  iStride2 = iStrideD << 1;
+  iStride3 = iStride2 + iStrideD;
+  iStride4 = iStrideD << 2;
+
+  __lsx_vstx(src0, pDst, iStride0);
+  __lsx_vstx(src1, pDst, iStride1);
+  __lsx_vstx(src2, pDst, iStride2);
+  __lsx_vstx(src3, pDst, iStride3);
+  pDst += iStride4;
+  __lsx_vstx(src4, pDst, iStride0);
+  __lsx_vstx(src5, pDst, iStride1);
+  __lsx_vstx(src6, pDst, iStride2);
+  __lsx_vstx(src7, pDst, iStride3);
+  pDst += iStride4;
+  __lsx_vstx(src8, pDst, iStride0);
+  __lsx_vstx(src9, pDst, iStride1);
+  __lsx_vstx(src10, pDst, iStride2);
+  __lsx_vstx(src11, pDst, iStride3);
+  pDst += iStride4;
+  __lsx_vstx(src12, pDst, iStride0);
+  __lsx_vstx(src13, pDst, iStride1);
+  __lsx_vstx(src14, pDst, iStride2);
+  __lsx_vstx(src15, pDst, iStride3);
+}
+
+void WelsCopy16x16NotAligned_lsx (uint8_t* pDst, int32_t iStrideD,
+                                  uint8_t* pSrc, int32_t iStrideS) {
+  int32_t iStride0 = 0;
+  int32_t iStride1 = iStrideS;
+  int32_t iStride2 = iStrideS << 1;
+  int32_t iStride3 = iStride2 + iStrideS;
+  int32_t iStride4 = iStrideS << 2;
+
+  v16u8_b src0, src1, src2, src3, src4, src5, src6, src7;
+  v16u8_b src8, src9, src10, src11, src12, src13, src14, src15;
+
+  DUP4_ARG2((v16u8_b)__lsx_vldx,
+            pSrc, iStride0, pSrc, iStride1,
+            pSrc, iStride2, pSrc, iStride3,
+            src0, src1, src2, src3);
+  pSrc += iStride4;
+  DUP4_ARG2((v16u8_b)__lsx_vldx,
+            pSrc, iStride0, pSrc, iStride1,
+            pSrc, iStride2, pSrc, iStride3,
+            src4, src5, src6, src7);
+  pSrc += iStride4;
+  DUP4_ARG2((v16u8_b)__lsx_vldx,
+            pSrc, iStride0, pSrc, iStride1,
+            pSrc, iStride2, pSrc, iStride3,
+            src8, src9, src10, src11);
+  pSrc += iStride4;
+  DUP4_ARG2((v16u8_b)__lsx_vldx,
+            pSrc, iStride0, pSrc, iStride1,
+            pSrc, iStride2, pSrc, iStride3,
+            src12, src13, src14, src15);
+
+  iStride1 = iStrideD;
+  iStride2 = iStrideD << 1;
+  iStride3 = iStride2 + iStrideD;
+  iStride4 = iStrideD << 2;
+
+  __lsx_vstx((__m128i)src0, pDst, iStride0);
+  __lsx_vstx((__m128i)src1, pDst, iStride1);
+  __lsx_vstx((__m128i)src2, pDst, iStride2);
+  __lsx_vstx((__m128i)src3, pDst, iStride3);
+  pDst += iStride4;
+  __lsx_vstx((__m128i)src4, pDst, iStride0);
+  __lsx_vstx((__m128i)src5, pDst, iStride1);
+  __lsx_vstx((__m128i)src6, pDst, iStride2);
+  __lsx_vstx((__m128i)src7, pDst, iStride3);
+  pDst += iStride4;
+  __lsx_vstx((__m128i)src8, pDst, iStride0);
+  __lsx_vstx((__m128i)src9, pDst, iStride1);
+  __lsx_vstx((__m128i)src10, pDst, iStride2);
+  __lsx_vstx((__m128i)src11, pDst, iStride3);
+  pDst += iStride4;
+  __lsx_vstx((__m128i)src12, pDst, iStride0);
+  __lsx_vstx((__m128i)src13, pDst, iStride1);
+  __lsx_vstx((__m128i)src14, pDst, iStride2);
+  __lsx_vstx((__m128i)src15, pDst, iStride3);
+}