daiduo/TurboPFor-Integer-Compression
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
Files
e01167f35debde8bcaea4c9841220c268d64aee4
TurboPFor-Integer-Compression/plugins.cc
x edf2a114c2 IcBench: Integer Compression codecs
2017-01-29 14:09:46 +01:00

1304 lines
69 KiB
C++
Raw Blame History

/**
Copyright (C) powturbo 2013-2017
GPL v2 License
This program is free software; you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation; either version 2 of the License, or
(at your option) any later version.
This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
GNU General Public License for more details.
You should have received a copy of the GNU General Public License along
with this program; if not, write to the Free Software Foundation, Inc.,
51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
- homepage : https://sites.google.com/site/powturbo/
- github : https://github.com/powturbo
- twitter : https://twitter.com/powturbo
- email : powturbo [_AT_] gmail [_DOT_] com
**/
// TurboPFor: Integer Compression - Benchmark App
#define _LARGEFILE64_SOURCE 1
#define _FILE_OFFSET_BITS 64
#include <stdlib.h>
#include <string.h>
#include <malloc.h>
#include <stdio.h>
#include <math.h>
#include <time.h>
#include <getopt.h>
#include <sys/stat.h>
#include <stdint.h>
#include "plugins.h"
#ifdef NCODEC0
#define CODEC0 0
#define CODEC0S 0
#else
#define CODEC0 1
#ifdef NSIMD
#define CODEC0V 0
#else
#define CODEC0V 1
#endif
#endif
#ifdef N
#define CODEC0 0
#else
#define CODEC0 1
#endif
#ifdef NCODEC1
#define CODEC1 0
#else
#define CODEC1 1
#endif
#ifdef NCODEC2
#define CODEC2 0
#else
#define CODEC2 1
#endif
#ifdef LZTURBO
#define C_LZTURBO CODEC2
#else
#define C_LZTURBO 0
#endif
#ifdef NTRANSFORM
#define TRANSFORM 0
#else
#define TRANSFORM 1
#endif
#ifdef BLOSC
#define C_C_BLOSC CODEC2
#else
#define C_C_BLOSC 0
#endif
enum {
#define C_MEMCPY 1
P_LMCPY, // must be 0
P_MCPY, // must be 1
P_COPY,
#define C_TURBOPFOR CODEC0
TB_PFDA, // actually not working w. mingw
TB_FOR,
TB_FORDA,
TB_BP,
TB_BPN,
TB_VBYTE,
TB_VSIMPLE,
TB_EF,
TB_VSIMPLE_S1,
TB_VSIMPLE_S4,
TB_VSIMPLE_S8,
#ifdef TURBOPFORH
#define C_TURBOPFORH CODEC0V
#else
#define C_TURBOPFORH 0
#endif
TB_BP128H,
#define C_TURBOPFORV CODEC0V
TB_FOR128V,
TB_PF128V,
TB_PF256V,
TB_PFN128V,
TB_PFN256V,
TB_BP128V,
TB_FOR256V,
TB_BP256H,
TB_BP256V,
TB_EF128V,
TB_EF256V,
#define C_BITSHUFFLE CODEC1
P_BITSHUFFLE,
P_BS_LZ,
P_BS_LZ4,
P_BS_ZLIB,
//P_BS_ZSTD,
P_BS_SHUFFLE,
#define C_FASTPFOR CODEC1
FP_VBYTE,
FP_FASTPFOR,
FP_SIMDFASTPFOR,
FP_SIMDOPTPFOR,
FP_OPTPFOR,
FP_SIMPLE8BRLE,
FP_SIMDPACK,
#define C_LIBFOR CODEC1
LF_FOR,
LF_FORX,
#define C_LITTLEPACK CODEC1
LI_PACK,
LI_BMIPACK,
LI_TURBOPACK,
LI_HORPACK,
LI_SCPACK,
#define C_LZ4 CODEC1
LZ4_,
LZ4_S1,
LZ4_S4,
LZ4_S8,
#define C_MASKEDVBYTE CODEC1
P_MASKEDVBYTE,
#define C_POLYCOM CODEC1
PC_OPTPFD, // compression too slow and limited to 28 bits. crashs on some lists
PC_VBYTE,
PC_RICE, // incl. only as demo, crash on some lists
PC_SIMPLE16, // limited to 28 bits.
#define C_QMX CODEC1
P_QMX,P_QMX2,P_QMX3,P_QMX4, // crash on gov2
#define C_SIMDCOMP CODEC1
SC_PACK,
SC_SIMDPACK128,
SC_SIMDPACK256,
SC_FOR,
SC_FORDA,
#define C_SIMPLE8B CODEC1 //crash on integers w. size 32 bits !
AM_SIMPLE8B,
#define C_STREAMVBYTE CODEC1
P_STREAMVBYTE,
#define C_VARINTG8IU CODEC1
P_VARINTG8IU,
#ifdef ZLIB
#define C_ZLIB CODEC2
#else
#define C_ZLIB 0
#endif
P_ZLIB,
#define C_TRANSFORM CODEC0
TB_TP8_32,
TB_TP8V_32,
TB_TP4V_32,
TB_ZIGZAG_32,
TB_DELTA_32,
#define C_LZTURBO 0
P_LZT,
P_VSHUF,
P_MAX
};
#ifndef __SSSE3__
#define C_VINTG8IU 0
#define C_MASKEDVBYTE 0
#endif
#if C_FASTPFOR
#include "ext/FastPFor/headers/variablebyte.h"
#include "ext/FastPFor/headers/simple16.h"
#include "ext/FastPFor/headers/simple8b_rle.h"
#include "ext/FastPFor/headers/fastpfor.h"
#include "ext/FastPFor/headers/simdfastpfor.h"
#include "ext/FastPFor/headers/optpfor.h"
#include "ext/FastPFor/headers/simdoptpfor.h"
//#include "ext/FastPFor/headers/compositecodec.h"
#endif
#if C_SIMPLE8B
#include "ext/simple8b.h" // optimized simple-8b by powturbo
#endif
#if C_C_BLOSC
#include "ext/c-blosc/blosc/shuffle.h"
#include "ext/c-blosc/blosc/blosc.h"
#endif
#if C_LIBFOR
#include "ext/libfor/for.h"
unsigned char *for_selectx( unsigned char *__restrict in, unsigned n, unsigned *__restrict out) { unsigned b = in[4], i; for(i = 0; i < n; i++) out[i] = for_select(in, i); return in + 5 + for_compressed_size_bits(n, b); }
#endif
#if C_LZ4
#include "ext/lz4/lib/lz4.h"
#include "ext/lz4/lib/lz4hc.h"
#endif
// #if C_SIMDCOMPLIB
//#include "ext/SIMDCompressionAndIntersection/include/bitpacking.h"
// #endif
#if C_POLYCOM
#include "ext/vas16c.h" // Simple 16
#include "ext/vas16d.h"
#include "ext/rc.h"
#include "ext/polycom/optpfd.h"
#include "ext/polycom/polyvbyte.h"
#endif
#if C_QMX
//#include "ext/qmx/compress_qmx.h"
//#include "ext/qmx.h"
#include "ext/bench_/bench/compress_qmx.h"
#include "ext/bench_/bench/compress_qmx_v2.h"
#include "ext/bench_/bench/compress_qmx_v3.h"
#include "ext/bench_/bench/compress_qmx_v4.h"
#endif
#if C_TURBOPFOR
#include "vint.h"
#include "vsimple.h"
#include "bitpack.h"
#include "eliasfano.h"
#include "bitutil.h"
#include "transpose.h"
#include "vp4.h"
#endif
#if C_ZLIB
#include <zlib.h>
#endif
#ifdef __cplusplus
extern "C" {
#endif
#if C_STREAMVBYTE
#include "ext/streamvbyte/include/streamvbyte.h"
#include "ext/streamvbyte/include/streamvbytedelta.h"
#undef VARINTDECODE_H_
#endif
#if C_MASKEDVBYTE
#include "ext/MaskedVByte/include/varintencode.h"
#undef VARINTDECODE_H_
#include "ext/MaskedVByte/include/varintdecode.h"
#endif
#if C_VARINTG8IU
#include "ext/varintg8iu.h" // SIMD Varint G8IU
#endif
#if C_LITTLEPACK
#include "ext/LittleIntPacker/include/bitpacking.h"
#include "ext/LittleIntPacker/include/util.h"
#endif
#include "ext/simdcomp/include/simdbitpacking.h"
#ifdef __cplusplus
}
#endif
#if C_SIMDCOMP
#undef SIMDBITPACKING_H_
#include "ext/vabyte.h" // Standard Variable Byte
#include "ext/simdcomp/include/simdcomp.h"
unsigned char *simdpackwithoutmaskd1n(uint32_t *in, uint32_t n, uint32_t *out, uint32_t start, uint32_t b) {//checkifdivisibleby(n, 128); const uint32_t * const initout(out); //while(needPaddingTo128Bits(out)) *out++ = 123456;
uint32_t *ip;
for(ip = in; ip != in+(n&~(128-1)); ip += 128,out += 4 * b)
simdpackwithoutmaskd1(start, ip, (__m128i *)out, b);
return (unsigned char *)out;
}
unsigned char *simdunpackd1n(uint32_t *in, uint32_t n, uint32_t *out, uint32_t start, uint32_t b) {
uint32_t k, *op;
for(op = out; op != out+(n&~(128-1)); op += 128,in += 4 * b)
simdunpackd1(start, (__m128i *)in, out, b);
return (unsigned char *)in;
}
unsigned char *simdfor_selectx( unsigned char *__restrict in, unsigned n, unsigned *__restrict out, unsigned start, unsigned b) {
unsigned i;
for(i=0; i < n;i++) out[i] = simdselectFOR(start, (const __m128i *)in, b, i); return in + simdpackFOR_compressedbytes(n, b);
}
#ifdef __AVX2__
unsigned char *avxpackwithoutmaskn(uint32_t *in, uint32_t n, uint32_t *out, uint32_t b) {
uint32_t *ip;
for(ip = in; ip != in+(n&~(256-1)); ip += 256,out += 8 * b)
avxpackwithoutmask(ip, (__m256i *)out, b);
return (unsigned char *)simdpack_shortlength(ip, n & (256-1), (__m128i *)out, b);
}
unsigned char *avxunpackn(uint32_t *in, uint32_t n, uint32_t *out, uint32_t b) {
uint32_t k, *op;
for(op = out; op != out+(n&~(256-1)); op += 256,in += 8 * b)
avxunpack((__m256i *)in, op, b);
return (unsigned char *)simdunpack_shortlength((__m128i *)in, n & (256-1), op, b);
}
#endif
#endif
#if C_LZTURBO
#include "../../lz/lz8.h"
int lz8c0( struct lzobj *lz);
int lz8c01(struct lzobj *lz);
int lz8d( struct lzobj *lz);
#include "../../lz/lzb.h"
int lzbc0( struct lzobj *lz);
int lzbc01(struct lzobj *lz);
int lzbc2( struct lzobj *lz);
int lzbd( struct lzobj *lz);
//#include "../../lz/lzh.h"
int lzhc2( struct lzobj *lz);
int lzhd( struct lzobj *lz);
int prunelen = 0x200, ans_seg = 12*1024;
#include "../../lz/anst.h"
#include "../../lz/ans.h"
#include "../../lz/mh.h"
#define MH_BLK (16*1024)
#endif
#if C_BITSHUFFLE
#define __STDC_VERSION__ 199901L
#include "ext/bitshuffle/src/bitshuffle.h"
#endif
#define BLK_SIZE (64*1024*4)
#define BLK_V128 (128*4)
#define BLK_V256 (256*4)
unsigned char sbuf[BLK_SIZE*2+64];
//------------------------------------------------- plugins: codes+function -------------------------------------------------
struct codecs codecs[] = {
{ C_C_BLOSC, "blosc", "https://github.com/Blosc/c-blosc" },
{ C_FASTPFOR, "FastPFor", "https://github.com/lemire/FastPFor" },
{ C_LZ4, "lz4", "https://github.com/Cyan4973/lz4" },
{ C_LITTLEPACK, "LittleIntPacker", "https://github.com/lemire/LittleIntPacker" },
{ C_LZTURBO, "LzTurbo", "https://sites.google.com/site/powturbo" },
{ C_MASKEDVBYTE, "MaskedVbyte", "http://maskedvbyte.org" },
{ C_POLYCOM, "Polycom", "https://github.com/encode84/bcm" },
{ C_QMX, "QMX", "https://bitbucket.org/andrewtrotman/bench" },
{ C_SIMDCOMP, "simdcomp", "https://github.com/lemire/simdcomp" },
{ C_SIMPLE8B, "Simple-8b optimized", "https://github.com/powturbo/TurboPFor" },
{ C_STREAMVBYTE, "Streamvbyte", "https://github.com/lemire/streamvbyte" },
{ C_TURBOPFOR, "TurboPFor", "https://github.com/powturbo/TurboPFor" },
{ C_VARINTG8IU, "VarintG8IU", "https://github.com/lemire/FastPFor" },
{ C_ZLIB, "zlib", "http://zlib.net\thttps://github.com/madler/zlib" },
{ C_MEMCPY, "memcpy/copy", "--------------------------------------" },
{ -1 }
};
struct plugs plugs[] = {
{ TB_PF128V, "TurboPFor", C_TURBOPFOR, BLK_V128,0,"","PFor (SSE2)" },
{ TB_PFN128V, "TurboPForN", C_TURBOPFOR, 0, 0,"","PFor (SSE2) large blocks" },
{ TB_PF256V, "TurboPFor256", C_TURBOPFOR, BLK_V256,0,"","PFor (AVX2)" },
{ TB_PFN256V, "TurboPFor256N", C_TURBOPFOR, 0, 0,"","PFor (AVX2) large blocks" },
{ TB_PFDA, "TurboPForDA", C_TURBOPFOR, BLK_V128,0,"","PFor direct access" },
{ TB_FOR, "TurboFor", C_TURBOPFOR, BLK_V128,0,"","FOR" },
{ TB_FORDA, "TurboForDA", C_TURBOPFOR, BLK_V128,0,"","FOR direct access" },
{ TB_FOR128V, "TurboForV", C_TURBOPFOR, BLK_V128,0,"","FOR (SIMD)" },
{ TB_FOR256V, "TurboFor256V", C_TURBOPFOR, BLK_V256,0,"","FOR (AVX2)" },
{ TB_BP, "TurboPack", C_TURBOPFOR, 0, 0,"","Bit packing (scalar)" },
{ TB_BPN, "TurboPackN", C_TURBOPFOR, 0, 0,"","Bit packing (scalar) large blocks" },
{ TB_BP128V, "TurboPackV", C_TURBOPFOR, BLK_V128,0,"","Bit packing (SSE2 Vertical)" },
{ TB_BP128H, "TurboPackH", C_TURBOPFOR, BLK_V128,0,"","Bit packing (SSE2 Horizontal)" },
{ TB_BP256V, "TurboPack256V", C_TURBOPFOR, BLK_V256,0,"","Bit packing (AVX2 Vertical)" },
{ TB_VBYTE, "TurboVByte", C_TURBOPFOR, 0, 0,"","Variable byte (scalar)" },
{ TB_VSIMPLE, "VSimple", C_TURBOPFOR, 0, 0,"","Variable simple (scalar)" },
{ TB_EF, "EliasFano", C_TURBOPFOR, 0, PG_DLT,"","Elias fano (scalar)" },
{ TB_EF128V, "EliasFanoV", C_TURBOPFOR, BLK_V128,0,"","Eliasfano (SSE2)" },
{ TB_EF256V, "EliasFano256V", C_TURBOPFOR, BLK_V256,0,"","Elias fano (AVX2" },
{ TB_VSIMPLE_S1, "vs_bitshuffle", C_LZ4, BLK_SIZE },
{ TB_VSIMPLE_S4, "vs_s4", C_LZ4, BLK_SIZE },
{ TB_VSIMPLE_S8, "vs_s8", C_LZ4, BLK_SIZE },
{ AM_SIMPLE8B, "Simple8b", C_SIMPLE8B, 1024, 0,"","64 bits Simple family (instable)" },
{ PC_SIMPLE16, "PC_Simple16", C_POLYCOM, BLK_V128,0,"","Simple 16. limited to 28 bits" },
{ PC_OPTPFD, "PC_OptPFD", C_POLYCOM, BLK_V128,0,"","OptPFD. limited to 28 bits" },
{ PC_VBYTE, "PC_Vbyte", C_POLYCOM, 0, 0,"","Variable byte" },
{ PC_RICE, "PC_Rice", C_POLYCOM, BLK_V128,0,"","Rice coding (instable)" },
{ P_VARINTG8IU, "VarintG8IU", C_VARINTG8IU, 0, 0,"","Variable byte SIMD" },
{ P_MASKEDVBYTE, "MaskedVbyte", C_MASKEDVBYTE, 0, 0,"","Variable byte SIMD" },
{ P_STREAMVBYTE, "StreamVbyte", C_STREAMVBYTE, 0, 0,"","Variable byte SIMD" },
{ FP_FASTPFOR, "FP_FastPFor", C_FASTPFOR, 0, 0,"","PFor scalar (inefficient for small blocks)" },
{ FP_SIMDFASTPFOR,"FP_SimdFastPFor", C_FASTPFOR, 0, 0,"","PFor SIMD (inefficient for small blocks)" },
{ FP_OPTPFOR, "FP_OptPFor", C_FASTPFOR, 0, 0,"","OptPFor scalar" },
{ FP_SIMDOPTPFOR, "FP_SIMDOptPFor", C_FASTPFOR, 0, 0,"","OptPFor SIMD" },
{ FP_VBYTE, "FP_VByte", C_FASTPFOR, 0, 0,"","Variable byte" },
{ FP_SIMPLE8BRLE, "FP_Simple8bRLE", C_FASTPFOR, 0, 0,"","Simple-8b + rle" },
{ SC_SIMDPACK128, "SC_SIMDPack128", C_SIMDCOMP, BLK_V128,0,"","Bit packing (SSE4.1)"},
{ SC_SIMDPACK256, "SC_SIMDPack256", C_SIMDCOMP, BLK_V256,0,"","Bit packing (SSE4.1)"},
{ SC_FOR, "SC_For", C_SIMDCOMP, BLK_V128,0,"","For (SSE4.1)"},
{ SC_FORDA, "SC_ForDA", C_SIMDCOMP, BLK_V128,0,"","For direct access (SSE4.1)"},
//{ CL_FASTPFORD1, "CL.SIMDPFORD1", C_SIMDCOMP, "", 0,BLK_V128},
{ P_BS_LZ, "blosc_lz", C_C_BLOSC, BLK_SIZE,0, "","TurboPFor's [delta]+transpose+lz4" },
{ P_BS_LZ4, "blosc_lz4", C_C_BLOSC, BLK_SIZE,0, "","TurboPFor's [delta]+transpose+lz4" },
{ P_BS_ZLIB, "blosc_zlib", C_C_BLOSC, BLK_SIZE,0, "","TurboPFor's [delta]+transpose+lz4" },
{ LF_FOR, "LibFor_For", C_SIMDCOMP, BLK_V128,0,"","For"},
{ LF_FORX, "LibFor_ForDA", C_SIMDCOMP, BLK_V128,0,"","For direct access"},
{ LI_PACK, "LI_Pack", C_LITTLEPACK, 0, 0,"","Bit packing (scalar)" },
{ LI_TURBOPACK, "LI_TurboPack", C_LITTLEPACK, 0, 0,"","Bit packing (scalar)" },
{ LI_SCPACK, "LI_SuperPack", C_LITTLEPACK, 0, 0,"","Bit packing (scalar)" },
{ LI_HORPACK, "LI_HorPack", C_LITTLEPACK, 0, 0,"","Bit packing (sse4.1 horizontal)" },
{ LI_BMIPACK, "LI_BMIPack256", C_LITTLEPACK, 0, 0,"","Bit packing (avx2)" },
{ P_QMX, "qmx", C_QMX, 0, 0,"","QMX SIMD (inefficient for small blocks)" },
{ P_QMX2, "qmx2", C_QMX, 0, 0,"","QMX SIMD (inefficient for small blocks)" },
{ P_QMX3, "qmx3", C_QMX, 0, 0,"","QMX SIMD (inefficient for small blocks)" },
{ P_QMX4, "qmx4", C_QMX, 0, 0,"","QMX SIMD (inefficient for small blocks)" },
//-------- lz77 + [delta] + transpose/shuffle ---------------
{ P_LZT, "LzTurbo", C_LZTURBO, BLK_SIZE, 0, "20,21,22,32" },
{ P_VSHUF, "VSimpleANS", C_LZTURBO, BLK_SIZE, 0, "20,21,22,32" },
{ LZ4_, "lz4", C_LZ4, BLK_SIZE, 0, "","lz4" },
{ LZ4_S1, "lz4_bitshufle", C_LZ4, BLK_SIZE, 0, "","Bitshuffle + lz4" },
{ LZ4_S4, "lz4_tp4", C_LZ4, BLK_SIZE, 0, "","TurboPFor's [delta]+byte transpose + lz4" },
{ LZ4_S8, "lz4_tp8", C_LZ4, BLK_SIZE, 0, "","TurboPFor's [delta]+transpose + lz4" },
{ P_ZLIB, "zlib", C_ZLIB, BLK_SIZE, 0, "","TurboPFor's [delta]+transpose + zlib" },
//----- Transform --------------
{ P_MCPY, "memcpy", C_MEMCPY },
{ P_COPY, "copy", C_MEMCPY },
{ TB_TP8_32, "TP8s_32", C_TURBOPFOR, BLK_SIZE,0,"","Byte Transpose (scalar)" },
{ TB_TP8V_32, "TP8_32", C_TURBOPFOR, BLK_SIZE,0,"","Byte transpose (sse)" },
{ TB_TP4V_32, "TP4_32", C_TURBOPFOR, BLK_SIZE,0,"","Nibble transpose (sse)" },
{ TB_ZIGZAG_32, "ZigZag_32", C_TURBOPFOR, BLK_SIZE,0,"","ZigZag encoding (sse2)" },
{ TB_DELTA_32, "Delta_32", C_TURBOPFOR, BLK_SIZE,0,"","Delta encoding (sse2)"},
{ P_BITSHUFFLE, "BitShuffle", C_TURBOPFOR, BLK_SIZE,0,"","Bit shuffle (SIMD)" },
{ P_BS_SHUFFLE, "Blosc_Shuffle", C_C_BLOSC, BLK_SIZE,0,"","Blosc's shuffle (SIMD)" },
#ifdef LZTURBO
// #include "../beplugr.h"
#endif
{ -1 }
};
struct plugg plugg[] = {
{ "TURBOPFOR","TURBOPFOR", "TurboPFor/TurboPFor256/TurboPackV256/TurboPackV/TurboVByte/TurboPack/TurboForDA/EliasFano/VSimple" },
{ "DEFAULT", "Default", "TurboPFor/TurboPFor256/TurboPackV256/TurboPackV/TurboVByte/TurboPack" },
{ "BENCH", "Benchmark", "TurboPFor/TurboPackV/TurboVByte/TurboPack/QMX/FP.SimdFastPfor/FP.SimdOptPFor/MaskedVbyte/StreamVbyte" },
{ "EFFICIENT","Efficient", "TurboPFor/vsimple/turbovbyte" },
{ "BITPACK", "Bit Packing", "TurboPackV/TurboPack256V/TurboPackH/TurboPack/SC.SimdPack128/SC.SimdPack256" },
{ "VBYTE", "Variable byte", "TurboVByte/FP.VByte/PC.Vbyte/VarintG8IU/MaskedVbyte/StreamVbyte" },
{ "SIMPLE", "Simple Family", "simple8b/simple16/vsimple/qmx" },
{ "LZ4", "lz4+bitshufle/transpose 4/8", "lz4_bitshuffle/lz4_tp4/lz4_tp8" },
{ "LI", "Little Integer", "LI.Pack/LI.TurboPack/LI.SuperPack/LI.HorPack/LI.BMIPack256" },
{ "" }
};
#define _TRANSFORM // Transform functions : transpose,zigzag
#define _TP_BITS 1 // transpose bits (4,8,16,32,64)
#if C_C_BLOSC
#define BITSHUFFLE(in,n,out) bitshuffle( 1, n, (unsigned char *)in, out, NULL)
#define BITUNSHUFFLE(in,n,out) bitunshuffle( 1, n, (unsigned char *)in, (unsigned char *)out,NULL)
#else
#define BITSHUFFLE(in,n,out) bshuf_bitshuffle(in, out, (n)/4, 4, 0); memcpy((char *)out+((n)&(~31)),(char *)in+((n)&(~31)),(n)&31)
#define BITUNSHUFFLE(in,n,out) bshuf_bitunshuffle(in, out, (n)/4, 4, 0);memcpy((char *)out+((n)&(~31)),(char *)in+((n)&(~31)),(n)&31)
#endif
#if _TP_BITS == 16
#define TRANSPOSE(in,n,out) transpose2(in,n,out)
#define UNTRANSPOSE(in,n,out) untranspose2(in,n,out)
#elif _TP_BITS == 64
#define TRANSPOSE(in,n,out) transpose8(in,n,out)
#define UNTRANSPOSE(in,n,out) untranspose8(in,n,out)
#elif _TP_BITS == 128
#define TRANSPOSE(in,n,out) transpose16(in,n,out)
#define UNTRANSPOSE(in,n,out) untranspose16(in,n,out)
#else
#define TRANSPOSE(in,n,out) transpose4(in,n,out)
#define UNTRANSPOSE(in,n,out) untranspose4(in,n,out)
#endif
#define PAD8(__x) (((__x)+7)/8)
unsigned char *u32enc(unsigned *__restrict in, int n, unsigned *__restrict out) { unsigned *ip;
for(ip = in; ip != in+(n&~3); ) {
*out++ = *ip++;
*out++ = *ip++;
*out++ = *ip++;
*out++ = *ip++;
}
while(ip < in+n) *out++ = *ip++;
return (unsigned char *)out;
}
unsigned char *u32dec(unsigned *__restrict in, int n, unsigned *__restrict out) { unsigned *op;
for(op = out; op != out+(n&~3); ) {
*op++ = *in++;
*op++ = *in++;
*op++ = *in++;
*op++ = *in++;
}
while(op < out+n) *op++ = *in++;
return (unsigned char *)in;
}
unsigned char *_bitunpackx32( unsigned char *__restrict in, unsigned n, unsigned *__restrict out , unsigned b) { unsigned i,k=0; for(i=0; i < n; i++,k+=b ) *out++ = _bitgetx32(in, k, b); return in + PAD8(n*b); }
// direct access functions : included for demonstration only. Use the bulk functions instead, if you are decompressing most of the values
unsigned char *bitf1unpackx32(unsigned char *__restrict in, unsigned n, unsigned *__restrict out, int start, unsigned b) { int i; for(i = 0; i < n; i++) out[i] = bitgetx32(in, i, b)+start+i+1; return in + PAD8(n*b); }
unsigned char *bitfunpackx32( unsigned char *__restrict in, unsigned n, unsigned *__restrict out, int start, unsigned b) { int i; for(i = 0; i < n; i++) out[i] = bitgetx32(in, i, b)+start; return in + PAD8(n*b); }
//---------------------------------------------- plugins --------------------------------------------------------
#include "conf.h"
int codini(size_t insize, int codec) {
switch(codec) {
#if C_C_BLOSC
case P_BS_LZ4: blosc_init(); blosc_set_nthreads(1); blosc_set_compressor(BLOSC_LZ4_COMPNAME);break;
case P_BS_ZLIB: blosc_init(); blosc_set_nthreads(1); blosc_set_compressor(BLOSC_ZLIB_COMPNAME); break;
case P_BS_LZ: blosc_init(); blosc_set_nthreads(1); blosc_set_compressor(BLOSC_BLOSCLZ_COMPNAME); break;
#endif
#if C_VARINTG8IU
case P_VARINTG8IU: VarIntG8IU();
#endif
#ifdef LZTURBO
// #include "../beplug0.h"
#endif
}
}
void codexit(int codec) {}
//--- Sorted integer array : Delta/Differential compression (inc=0 increasing, inc=1 strictly increasing sequence) ---------------
unsigned char *codcomps(unsigned char *_in, unsigned _n, unsigned char *out, int outsize, int codec, int lev, char *prm, int inc) {
unsigned *in = (unsigned *)_in, n = (_n+3) / 4, i, pa[BLK_SIZE+2048], x, b;
switch(codec) {
case TB_VSIMPLE: x = *in++; bitdelta32( in, --n, pa, x, inc);
vbxput32(out, x); return vsenc32( pa, n, out);
case TB_VBYTE: x = *in++; --n; vbxput32(out, x); return inc?vbd1enc32( in, n, out, x ):vbdenc32( in, n, out, x);
case TB_EF: x = *in++; --n; vbxput32(out, x); return inc?efano1enc32( in, n, out, x+1 ):efanoenc32( in, n, out, x);
case TB_PFDA: x = *in++; --n; vbxput32(out, x); DELTR(in,n,x,inc,pa);
return p4encx32(pa,n,out);
case TB_FOR :
case TB_FORDA: x = *in++; --n; vbxput32(out, x);
if(inc) { b = bitf132(in, n, x); *out++=b; return bitf1pack32(in, n, out, x, b); }
else { b = bitf32( in, n, x); *out++=b; return bitfpack32( in, n, out, x, b); }
case TB_BP128H:
case TB_BP: x = *in++; --n; vbxput32(out, x);
if(inc) { b = bitd132(in, n, x); *out++=b; return bitd1pack32(in, n, out, x, b); }
else { b = bitd32( in, n, x); *out++=b; return bitdpack32( in, n, out, x, b); }
case TB_BPN: return out+(inc?bitnd1pack32(in, n, out):bitndpack32( in, n, out));
#if C_TURBOPFORV
case TB_EF128V: x = *in++; --n; vbxput32(out, x);
if(inc) return n == 128?efano1enc128v32(in, n, out, x+1 ):efano1enc32( in, n, out, x+1);
else return n == 128?efanoenc128v32( in, n, out, x ):efanoenc32( in, n, out, x );
case TB_PF128V: x = *in++; --n; vbxput32(out, x);
if(inc) return n == 128?p4d1enc128v32( in, n, out, x ):p4d1enc32( in, n, out, x);
else return n == 128?p4denc128v32( in, n, out, x ):p4denc32( in, n, out, x);
case TB_PFN128V: return out+(inc?p4nd1enc128v32( in, n, out ):p4ndenc128v32(in, n, out));
case TB_FOR128V: x = *in++; --n; vbxput32(out, x);
if(inc) { b = bitf132(in, n, x); *out++=b; return n == 128?bitf1pack128v32(in, n, out, x, b):bitf1pack32( in, n, out, x, b); }
else { b = bitf32( in, n, x); *out++=b; return n == 128?bitfpack128v32( in, n, out, x, b):bitfpack32( in, n, out, x, b); }
case TB_BP128V: x = *in++; --n; vbxput32(out, x);
if(inc) { b = bitd132(in, n, x); *out++=b; return n == 128?bitd1pack128v32(in, n, out, x, b):bitd1pack32( in, n, out, x, b); }
else { b = bitd32( in, n, x); *out++=b; return n == 128?bitdpack128v32( in, n, out, x, b):bitdpack32( in, n, out, x, b); }
#ifdef __AVX2__
case TB_FOR256V: x = *in++; --n; vbxput32(out, x);
if(inc) { b = bitf132(in, n, x); *out++=b; return n == 256?bitf1pack256v32(in, n, out, x, b):bitf1pack32( in, n, out, x, b); }
else { b = bitf32( in, n, x); *out++=b; return n == 256?bitfpack256v32( in, n, out, x, b):bitfpack32( in, n, out, x, b); }
case TB_EF256V: x = *in++; --n; vbxput32(out, x);
if(inc) return n == 256?efano1enc256v32(in, n, out, x+1 ):efano1enc32( in, n, out, x+1);
else return n == 256?efanoenc256v32( in, n, out, x ):efanoenc32( in, n, out, x );
case TB_PF256V: x = *in++; bitdelta32( in, --n, pa, x, inc);
vbxput32(out, x); return n == 256?p4enc256v32( pa, n, out ):p4enc32( pa, n, out);
case TB_PFN256V: return out+(inc?p4nd1enc256v32( in, n, out ):p4ndenc256v32(in, n, out));
case TB_BP256V: x = *in++;bitdelta32( in, --n, pa, x, inc);
vbxput32(out, x);*out++=b; return n == 256?bitpack256v32( pa, n, out, b ):bitpack32( pa, n, out,b);
/*case TB_BP256V: x = *in++; --n;
if(inc) { b = bitd132(in, n, x); vbxput32(out, x); *out++=b; return n == 256?bitd1pack256v32(in, n, out, x, b):bitd1pack32( in, n, out, x, b); }
else { b = bitd32( in, n, x); vbxput32(out, x); *out++=b; return n == 256?bitdpack256v32( in, n, out, x, b):bitdpack32( in, n, out, x, b); }*/
#endif
#endif
//-------- External libs ----------
#if C_SIMPLE8B
case AM_SIMPLE8B: b = bitdelta32( in+1, --n, pa, in[0], inc); vbxput32(out, in[0]); if(b>28) die("simple-8b overflow.bits size>28\n");
return vs8benc( pa, n, out);
#endif
case P_COPY: return u32enc( in, n, (unsigned *)out);
case P_MCPY: memcpy(out, _in, _n); return out+_n;
#if C_FASTPFOR
case FP_VBYTE: bitdelta32( in, n, pa, -inc, inc); return vbyteenc( pa, n, (unsigned *)out);
#endif
#if C_MASKEDVBYTE
case P_MASKEDVBYTE: return out+vbyte_encode_delta(in, n, out, 0);
#endif
#if C_LIBFOR
case LF_FORX: //b = bitf32( in+1, --n, in[0]); *(unsigned *)out = in[0]; out+= 4; *out++=b; return out + for_compress_bits( in+1, out, n, in[0], b); //if(b < 0) b = maxbits(in), *out++ = b; return out + for_compress_bits(in, out, n, 0, b);
case LF_FOR: return out + for_compress_sorted(in, out, n);
#endif
#if C_POLYCOM
case PC_VBYTE: bitdelta32( in, n, pa, -inc, inc);return vbpolyenc(pa, n, out);
case PC_SIMPLE16: b = bitdelta32( in+1, --n, pa, in[0], inc); vbxput32(out, in[0]); if(b>28) die("simple16 overflow.bits size>28\n");
return vs16enc( pa, n, (unsigned *)out);
case PC_OPTPFD: b = bitdelta32( in+1, --n, pa, in[0], inc); vbxput32(out, in[0]); if(b>28) die("optp4 overflow.bits size>28\n");
return optpfdenc32(pa, n, out);
/*b = bitdelta32( in+1, --n, pa, in[0], inc); vbxput32(out, in[0]); if(b>28) die("optp4 overflow.bits size>28\n");
if(n < 128) return vbyteenc( pa, n, (unsigned *)out);
else { return out + OPT4(pa, n, (unsigned *)out); }*/
#endif
#if C_QMX
//case P_QMX: { bitdelta32( in+1, --n, pa, in[0], inc); vbxput32(out, in[0]); unsigned char *q = qmx_enc(pa, n, out+4); *(unsigned *)out = q - (out+4); return q; }
case P_QMX: { bitdelta32( in+1, --n, pa, in[0], inc); vbxput32(out, in[0]); ANT_compress_qmx qmx; unsigned r=qmx.compress(out+4, outsize, (uint32_t *)pa, (uint64_t)n); ctou32(out)=r; return out+4+r; }
case P_QMX2: { bitdelta32( in+1, --n, pa, in[0], inc); vbxput32(out, in[0]); ANT_compress_qmx_v2 qmx; unsigned r=qmx.compress(out+4, outsize, (uint32_t *)pa, (uint64_t)n); ctou32(out)=r; return out+4+r; }
case P_QMX3: { bitdelta32( in+1, --n, pa, in[0], inc); vbxput32(out, in[0]); ANT_compress_qmx_v3 qmx; unsigned r=qmx.compress(out+4, outsize, (uint32_t *)pa, (uint64_t)n); ctou32(out)=r; return out+4+r; }
case P_QMX4: { bitdelta32( in+1, --n, pa, in[0], inc); vbxput32(out, in[0]); ANT_compress_qmx_v4 qmx; unsigned r=qmx.compress(out+4, outsize, (uint32_t *)pa, (uint64_t)n); ctou32(out)=r; return out+4+r; }
#endif
#if C_SIMDCOMP
case SC_FOR:
case SC_FORDA: x = *in++; b = bitf32( in, --n, x); vbxput32(out, x); *out++=b; return (unsigned char *)simdpackFOR_length(x, in, n, (__m128i *)out, b);
case SC_SIMDPACK128:
if(n < 129) { bitdelta32( in, n, pa, -inc, inc); return vbyteenc((unsigned *)pa, n, (unsigned *)out); }
else { b = simdmaxbitsd1(in[0], in+1); vbxput32(out, in[0]); *out++=b; return simdpackwithoutmaskd1n(in+1, n-1, (unsigned *)out, in[0], b); }
#endif else { simdpackFOR(in[0], (in+1), (__m128i *)out, b); return out + simdpackFOR_compressedbytes(n, b); }
#if C_STREAMVBYTE
case P_STREAMVBYTE: return out + streamvbyte_delta_encode(in, n, out, inc);
#endif
#if C_VARINTG8IU
case P_VARINTG8IU: bitdelta32( in, n, pa, -inc, inc); return vintg8enc(pa, n, out);
#endif
// --------- transpose + lz77 ------------------------------------------------------------------------------------------------
#if C_TRANSFORM
case TB_ZIGZAG_32:b = bitzigzag32(in, n, (unsigned *)out, 0); return out + n*4;
case TB_TP8_32: bitdelta32(in, n, (unsigned *)sbuf, -inc, inc); _transpose4( (unsigned char *)sbuf, n*4, out); return out + n*4;
case TB_TP8V_32: bitdelta32(in, n, (unsigned *)sbuf, -inc, inc); TRANSPOSE( (unsigned char *)sbuf, n*4, out); return out + n*4;
case TB_TP4V_32: bitdelta32(in, n, (unsigned *)sbuf, -inc, inc); transposen4( (unsigned char *)sbuf, n*4, out); return out + n*4;
case TB_DELTA_32: bitdelta32(in, n, (unsigned *)out, -inc, inc); return out + n*4;
#endif
// --------- delta + transpose + lz77 ----------------------------------------------------------------------------------------
#if C_LZTURBO
case P_LZT10:{ bitdelta32(in, n, (unsigned *)out, -inc, inc); TRANSPOSE((unsigned char *)out, n*4, sbuf); struct lzobj lz; lz.srclen = n*4; lz.src = sbuf; lz.dst = out; lz.dstlen = n*4; lz.level = 0; lz.hbits = 16; return out + lz8c01(&lz); }
case P_LZT20:{ bitdelta32(in, n, (unsigned *)out, -inc, inc); TRANSPOSE((unsigned char *)out, n*4, sbuf); struct lzobj lz; lz.srclen = n*4; lz.src = sbuf; lz.dst = out; lz.dstlen = n*4; lz.level = 0; lz.hbits = 16; return out + lzbc01(&lz); }
case P_LZT22:{ bitdelta32(in, n, (unsigned *)out, -inc, inc); TRANSPOSE((unsigned char *)out, n*4, sbuf); struct lzobj lz; lz.srclen = n*4; lz.src = sbuf; lz.dst = out; lz.dstlen = n*4; lz.level = 2; lz.hbits = 26; return out + lzbc2(&lz); }
case P_VSHUF:{ bitdelta32(in, n, (unsigned *)out, -inc, inc); TRANSPOSE((unsigned char *)out, n*4, sbuf); unsigned char *p = mheenc(sbuf, n*4, MH_BLK, out+4, out+n*4+MH_BLK); ctou32(out) = p-(out+4); return p; }
//case P_VSHUF:{ bitdelta32( in+1, --n, pa, in[0], inc); vbxput32(out, in[0]); unsigned char *p=vsenc32(pa, n, sbuf); p = mheenc(sbuf, p-sbuf, MH_BLK, out+4, out+n*4+1024); ctou32(out) = p-(out+4); return p; }
#endif
#if C_C_BLOSC
case P_BS_LZ:
case P_BS_LZ4:
case P_BS_ZLIB: return out + blosc_compress(lev, BLOSC_SHUFFLE, 4/*typesize*/, n*4, in, out, n*4+BLOSC_MAX_OVERHEAD);
//case P_BS_ZLIB_1: return out + blosc_compress(lev, BLOSC_SHUFFLE/*BLOSC_DELTA*/, 4/*typesize*/, n*4, in, out, n*4+BLOSC_MAX_OVERHEAD);
#endif
#if C_LZ4
#if C_BITSHUFFLE
case LZ4_S1: { bitdelta32(in, n, (unsigned *)out, -inc, inc); BITSHUFFLE((unsigned char *)out, n*4, sbuf); return out + LZ4_compress((char *)sbuf, (char *)out, n*4); } // bshuf_bitshuffle(out, sbuf, n*4/32, 32, 0);
#endif
case LZ4_ : return out + LZ4_compress((char *)in, (char *)out, n*4);
case LZ4_S4: { bitdelta32(in, n, (unsigned *)out, -inc, inc); transposen4((unsigned char *)out, n*4, sbuf); return out + LZ4_compress((char *)sbuf, (char *)out, n*4); } // bshuf_bitshuffle(out, sbuf, n*4/32, 32, 0);
case LZ4_S8: { bitdelta32(in, n, (unsigned *)out, -inc, inc); TRANSPOSE((unsigned char *)out, n*4, sbuf); return out + LZ4_compress((char *)sbuf, (char *)out, n*4); } // bshuf_bitshuffle(out, sbuf, n*4/32, 32, 0);
#endif
#if C_ZLIB
case P_ZLIB1: case P_ZLIB2: case P_ZLIB3: case P_ZLIB4: case P_ZLIB5: case P_ZLIB6: case P_ZLIB7: case P_ZLIB8: case P_ZLIB9:
{ bitdelta32(in, n, (unsigned *)out, -inc, inc); TRANSPOSE((unsigned char *)out, n*4, sbuf); uLongf outlen = n*4; int rc = compress2(out+4, &outlen, sbuf, n*4, codec-P_ZLIB1+1); if(rc != Z_OK) die("zlib compress2 rc=%d\n", rc); *(unsigned *)out = outlen; return out + 4 + outlen; }
#endif
}
return out;
}
unsigned char *coddecomps(unsigned char *in, unsigned _n, unsigned char *_out, int outlen, int codec, int lev, char *prm, int inc) {
unsigned *out = (unsigned *)_out, n = (outlen+3) / 4,x,b;
switch(codec) {
case TB_VSIMPLE:vbxget32(in, x);*out++ = x; --n; in = vsdec32( in, n, out); bitundn32(out, n, x+inc, inc); return in;
case TB_EF: vbxget32(in, x);*out++ = x; --n; return inc?efano1dec32( in, n, out, x+1 ):efanodec32( in, n, out, x);
case TB_VBYTE: vbxget32(in, x);*out++ = x; --n; return inc?vbd1dec32( in, n, out, x ):vbddec32( in, n, out, x);
case TB_PFDA: vbxget32(in, x);*out++ = x; --n; return inc?p4f1decx32( in, n, out, x ):p4fdecx32( in, n, out, x);
case TB_FOR: vbxget32(in, x);*out++ = x; --n; b = *in++; return inc?bitf1unpack32( in, n, out, x, b):bitfunpack32( in, n, out, x, b);
case TB_FORDA: vbxget32(in, x);*out++ = x; --n; b = *in++; return inc?bitf1unpackx32( in, n, out, x, b):bitfunpackx32( in, n, out, x, b);
case TB_BP: vbxget32(in, x);*out++ = x; --n; b = *in++; return inc?bitd1unpack32( in, n, out, x, b):bitdunpack32( in, n, out, x, b);
case TB_BPN: return in+(inc?bitnd1unpack32(in, n, out ):bitndunpack32( in, n, out));
#if C_TURBOPFORV
case TB_FOR128V:vbxget32(in, x);*out++ = x; --n; b = *in++;
if(inc) { return n==128?bitf1unpack128v32( in, n, out, x, b):bitf1unpack32( in, n, out, x, b); }
else { return n==128?bitfunpack128v32( in, n, out, x, b):bitfunpack32(in, n, out, x, b); }
case TB_EF128V: vbxget32(in, x);*out++ = x; --n;
if(inc) { return n==128?efano1dec128v32(in, n, out, x+1 ):efano1dec32( in, n, out, x+1); }
else { return n==128?efanodec128v32( in, n, out, x ):efanodec32( in, n, out, x); }
case TB_PF128V: vbxget32(in, x);*out++ = x; --n; //__builtin_prefetch(in+256);
if(inc) { return n==128?p4d1dec128v32( in, n, out, x ):p4d1dec32( in, n, out, x); }
else { return n==128?p4ddec128v32( in, n, out, x ):p4ddec32( in, n, out, x); }
case TB_PFN128V: return in+(inc?p4nd1dec128v32(in, n, out ):p4nddec128v32( in, n, out));
case TB_BP128V: vbxget32(in, x);*out++ = x; --n; b = *in++;
if(inc) { return n==128?bitd1unpack128v32(in, n, out, x, b):bitd1unpack32( in, n, out, x, b); }
else { return n==128?bitdunpack128v32( in, n, out, x, b):bitdunpack32( in, n, out, x, b); }
#ifdef __AVX2__
case TB_EF256V: vbxget32(in, x);*out++ = x; --n;
if(inc) { return n==256?efano1dec256v32(in, n, out, x+1 ):efano1dec32( in, n, out, x+1); }
else { return n==256?efanodec256v32( in, n, out, x ):efanodec32( in, n, out, x); }
case TB_PF256V: vbxget32(in, x);*out++ = x; --n; //__builtin_prefetch(in+256);
if(inc) { return n==256?p4d1dec256v32( in, n, out, x ):p4d1dec32( in, n, out, x); }
else { return n==256?p4ddec256v32( in, n, out, x ):p4ddec32( in, n, out, x); }
case TB_PFN256V: return in+(inc?p4nd1dec256v32(in, n, out ):p4nddec256v32( in, n, out));
case TB_FOR256V:vbxget32(in, x);*out++ = x; --n; b = *in++;
if(inc) { return n==256?bitf1unpack256v32( in, n, out, x, b):bitf1unpack32( in, n, out, x, b); }
else { return n==256?bitfunpack256v32( in, n, out, x, b):bitfunpack32(in, n, out, x, b); }
case TB_BP256V: vbxget32(in, x);*out++ = x; --n; b = *in++;
if(inc) { return n==256?bitd1unpack256v32(in, n, out, x, b):bitd1unpack32( in, n, out, x, b); }
else { return n==256?bitdunpack256v32( in, n, out, x, b):bitdunpack32( in, n, out, x, b); }
/*case TB_BP256V:vbxget32(in, x);*out = x; --n; b = *in++;
if(n <= 256) return inc?bitd1unpack32( in, n-1, out+1, x, b):bitdunpack32( in, n-1, out+1, x, b);
else { in = bitunpack256v32( in, out+1, b);bitundn32(out, n, -inc, inc); } break;*/
#endif
#endif
#if C_TURBOPFORH
case TB_BP128H: vbxget32(in, x);*out++ = x; --n; b = *in++;
if(inc) { return n==128?bitd1unpack128h32(in, n, out, x, b):bitd1unpack32( in, n, out, x, b); }
else { return n==128?bitdunpack128h32( in, n, out, x, b):bitdunpack32( in, n, out, x, b); }
#endif
//-------- Externa libs
case P_COPY: return u32dec( (unsigned *)in, n, out);
case P_MCPY: memcpy(_out, in, _n); return _out+_n;
#if C_SIMPLE8B
case AM_SIMPLE8B: vbxget32(in, x);*out++ = x; --n; in = vs8bdec(in, n, out); bitundn32(out, n, x+inc, inc); break;
#endif
#if C_FASTPFOR
case FP_VBYTE: in = vbytedec( in, n, out); bitundn32(out, n, -inc, inc); break;
#endif
#if C_LIBFOR
case LF_FORX: return for_selectx(in, n, out); //{ out[0] = *(unsigned *)in, b = in[4]; return for_selectx(in+5, n-1, out+1, out[0], b); }
case LF_FOR: return in + for_uncompress(in, out, n);
#endif
#if C_MASKEDVBYTE
case P_MASKEDVBYTE: in += masked_vbyte_decode_delta(in, out, n, 0); break;
#endif
#if C_POLYCOM
case PC_VBYTE: in = vbpolydec( in, n, out); bitundn32(out, n, -inc, inc); break;
case PC_SIMPLE16: vbxget32(in, x); *out = x; in = vs16dec((unsigned *)in, n-1, out+1); bitundn32(out, n, -inc, inc); break;
case PC_OPTPFD: vbxget32(in, x); *out = x; in = optpfddec32(in,n-1,out+1); bitundn32(out+1, n-1, x, inc); break;
/*if(n < 129) in = vbytedec(in, n, out);
else { _VBGET32(in, x, *out = x); unsigned all_array[2048]; in = (unsigned char *)detailed_p4_decode(out+1, (unsigned *)in, all_array); }
bitundn32(out, n, -inc, inc); break;*/
#endif
//case P_QMX: { vbxget32(in, x); *out = x; unsigned l = *(unsigned *)in; in = qmx_dec(in+4, l, out+1, n-1); bitundn32(out+1, n-1, x, inc); break; }
#if C_QMX //case P_QMX: return qmx_dec(in+4, ctou32(in), out, n);
case P_QMX: { vbxget32(in, x); *out = x; unsigned l = *(unsigned *)in; ANT_compress_qmx qmx; qmx.decompress(out+1, n-1, in+4, ctou32(in)); bitundn32(out+1, n-1, x, inc); return in+4+ctou32(in);}
case P_QMX2: { vbxget32(in, x); *out = x; unsigned l = *(unsigned *)in; ANT_compress_qmx_v2 qmx; qmx.decompress(out+1, n-1, in+4, ctou32(in)); bitundn32(out+1, n-1, x, inc); return in+4+ctou32(in);}
case P_QMX3: { vbxget32(in, x); *out = x; unsigned l = *(unsigned *)in; ANT_compress_qmx_v3 qmx; qmx.decompress(out+1, n-1, in+4, ctou32(in)); bitundn32(out+1, n-1, x, inc); return in+4+ctou32(in);}
case P_QMX4: { vbxget32(in, x); *out = x; unsigned l = *(unsigned *)in; ANT_compress_qmx_v4 qmx; qmx.decompress(out+1, n-1, in+4, ctou32(in)); bitundn32(out+1, n-1, x, inc); return in+4+ctou32(in);}
#endif
#if C_SIMDCOMP
case SC_SIMDPACK128:
if(n < 129) { in = vbytedec(in, n, out); bitundn32(out, n, -inc, inc); }
else { vbxget32(in, x);*out = x; b = *in++; in = simdunpackd1n((uint32_t *)in, n-1, out+1, out[0], b); } break;
case SC_FOR: vbxget32(in, x);*out = x; b = *in++; if(n < 129) return (unsigned char *)simdunpackFOR_length(x, (const __m128i *)in, n-1, (unsigned *)(out+1), b); else { simdunpackFOR(x, (const __m128i *)in, (unsigned *)(out+1), b); return in + simdpackFOR_compressedbytes(n-1, b); }
case SC_FORDA: { vbxget32(in, x);*out++ = x; --n; b = *in++; unsigned i; for(i=0; i < n;i++) out[i] = simdselectFOR(x, (const __m128i *)in, b, i); return in + simdpackFOR_compressedbytes(n, b); }
#endif
#if C_STREAMVBYTE
case P_STREAMVBYTE: return in + streamvbyte_delta_decode(in, out, n, inc);
#endif
#if C_VARINTG8IU
case P_VARINTG8IU: in = vintg8dec( in, n, out); bitundn32(out, n, -inc, inc); break;
#endif
//---------- transpose + lz77 ----------------------
#if C_TRANSFORM
case TB_ZIGZAG_32:memcpy(out, in, n*4); bitunzigzag32(out, n, 0); return in + n*4;
case TB_TP8_32: _untranspose4((unsigned char *)in, n*4, (unsigned char *)out); bitundn32(out, n, -inc, inc); return in + n*4;
case TB_TP8V_32: UNTRANSPOSE( (unsigned char *)in, n*4, (unsigned char *)out); bitundn32(out, n, -inc, inc); return in + n*4;
case TB_TP4V_32: untransposen4((unsigned char *)in, n*4, (unsigned char *)out); bitundn32(out, n, -inc, inc); return in + n*4;
case TB_DELTA_32: memcpy(out, in, n*4); bitundn32(out, n, -inc, inc); return in + n*4;
#endif
//---------- delta + transpose + lz77 ----------------------
#if C_LZ4
#if C_BITSHUFFLE
case LZ4_S1: in += LZ4_decompress_fast((char *)in, (char *)sbuf, n*4); BITUNSHUFFLE(sbuf, n*4, (unsigned char *)out); bitundn32(out, n, -inc, inc); break;
#endif
case LZ4_ : in += LZ4_decompress_fast((char *)in, (char *)out, n*4); break;
case LZ4_S4: in += LZ4_decompress_fast((char *)in, (char *)sbuf, n*4); untransposen4(sbuf, n*4, (unsigned char *)out); bitundn32(out, n, -inc, inc); break;
case LZ4_S8: in += LZ4_decompress_fast((char *)in, (char *)sbuf, n*4); UNTRANSPOSE(sbuf, n*4, (unsigned char *)out); bitundn32(out, n, -inc, inc); break;
#endif
#if C_ZLIB
case P_ZLIB1: case P_ZLIB2: case P_ZLIB3: case P_ZLIB4: case P_ZLIB5: case P_ZLIB6: case P_ZLIB7: case P_ZLIB8: case P_ZLIB9:
{ uLongf outsize = n*4; int l = *(unsigned *)in, rc = uncompress(sbuf, &outsize, in+4, l); in += 4 + l; UNTRANSPOSE(sbuf, n*4, (unsigned char *)out); bitundn32(out, n, -inc, inc); } break;
#endif
#if C_C_BLOSC
case P_BS_LZ:
case P_BS_LZ4:
case P_BS_ZLIB: { blosc_decompress(in, out, n*4); size_t nbytes, cbytes,blocksize; blosc_cbuffer_sizes(in, &nbytes, &cbytes, &blocksize); return in+cbytes; }
#endif
#if C_LZTURBO
case P_LZT10: { struct lzobj lz; lz.dstlen = n*4; lz.src = in; lz.dst = sbuf; lz.level = 0; in += lz8d(&lz); UNTRANSPOSE(sbuf, n*4, (unsigned char *)out); bitundn32(out, n, -inc, inc); } break;
case P_LZT20:
case P_LZT22: { struct lzobj lz; lz.dstlen = n*4; lz.src = in; lz.dst = sbuf; lz.level = 0; in += lzbd(&lz); UNTRANSPOSE(sbuf, n*4, (unsigned char *)out); bitundn32(out, n, -inc, inc); } break;
//case P_VSHUF: vbxget32(in, x);*out = x; in += ansdecompress( in, (n-1)*4, sbuf); vsdec32( sbuf, n-1, out+1); bitundn32(out, n, -inc, inc); break;
case P_VSHUF: { unsigned inlen = ctou32(in); in+=4; in = mhdec(in, inlen, MH_BLK, sbuf, n*4); UNTRANSPOSE(sbuf, n*4, (unsigned char *)out); bitundn32(out, n, -inc, inc); } break;
//case P_VSHUF: { vbxget32(in, x);*out = x; unsigned inlen = ctou32(in); in+=4; in = mhdec(in, inlen, MH_BLK, sbuf, n*4); vsdec32(sbuf, n-1, out+1); bitundn32(out+1, n-1, x, inc); } break;
#endif
}
return in;
}
unsigned char *codcompz(unsigned char *_in, unsigned _n, unsigned char *out, int outsize, int codec, int lev, char *prm, int inc) {
unsigned *in = (unsigned *)_in, n = (_n+3) / 4, i, pa[BLK_SIZE+2048], x, b;
switch(codec) {
case TB_VBYTE: x = *in++; --n; vbxput32(out, x); return vbzenc32( in, n, out, x);
case TB_BP: x = *in++; --n; vbxput32(out, x);b = bitz32(in, n, x); *out++=b;return bitzpack32(in, n, out, x, b);
case TB_BPN: x = *in++; --n; vbxput32(out, x); return out+bitnzpack32(in, n, out);
#if C_TURBOPFORV
case TB_BP128V: x = *in++; --n; vbxput32(out, x);b = bitz32(in, n, x); *out++=b;return n == 128?bitzpack128v32(in, n, out, x, b):bitzpack32(in, n, out, x, b);
#ifdef __AVX2__
//case TB_BP256V: b = bitz32(in, n, x); *out++=b; return n == 256?bitzpack256v32(in, n, out, x, b ):bitzpack32(in, n, out,x, b);
#endif
#endif
}
return out;
}
unsigned char *coddecompz(unsigned char *in, unsigned _n, unsigned char *_out, int outlen, int codec, int lev, char *prm, int inc) {
unsigned *out = (unsigned *)_out, n = (outlen+3) / 4,x,b;
switch(codec) {
case TB_VBYTE: vbxget32(in, x); *out++ = x; --n; return vbzdec32(in, n, out, x);
case TB_BP: vbxget32(in, x); *out++ = x; --n; b = *in++; return bitzunpack32(in, n, out, x, b);
case TB_BPN: vbxget32(in, x); *out++ = x; --n; return in+bitnzunpack32(in, n, out);
#if C_TURBOPFORV
case TB_BP128V:vbxget32(in, x); *out++ = x; --n; b = *in++; return n == 128?bitzunpack128v32(in, n, out, x, b):bitzunpack32(in, n, out, x, b);
#ifdef __AVX2__
//case TB_BP256V: b = *in++; return n == 256?bitzunpack256v32(in, n, out, x, b):bitzunpack32(in, n, out, x, b);
#endif
#endif
}
return in;
}
unsigned char *codcomp(unsigned char *_in, unsigned _n, unsigned char *out, int outsize, int codec, int lev, char *prm, int b) {
unsigned *in = (unsigned *)_in, n = (_n+3) / 4, x;
switch(codec) {
case TB_EF: return out;
case TB_VBYTE: return vbenc32( in, n, out);
case TB_VSIMPLE: return vsenc32( in, n, out);
case TB_PFDA: return p4encx32( in, n, out);
case TB_FOR :
case TB_FORDA: b = bitfm32( in, n, &x); vbxput32(out, x); *out++=b; return bitfpack32( in, n, out, x, b);
case TB_BP128H:
case TB_BP: if(b < 0) { BITSIZE32(in, n, b); *out++ = b; }
return bitpack32(in, n, out, b);
case TB_BPN: return out+bitnpack32(in, n, out);
#if C_TURBOPFORV
case TB_EF128V: return out;
case TB_PF128V: return n == 128?p4enc128v32(in, n, out):p4enc32(in, n, out);
case TB_PFN128V: return out+p4nenc128v32(in, n, out);
case TB_FOR128V: b = bitfm32( in, n, &x); vbxput32(out, x); *out++=b; return bitfpack128v32( in, n, out, x, b);
case TB_BP128V: if(b < 0) { BITSIZE32(in, n, b); *out++ = b; }
return n == 128?bitpack128v32(in, n, out, b):bitpack32(in, n, out, b);
#ifdef __AVX2__
case TB_PF256V: return n == 256?p4enc256v32(in, n, out):p4enc32(in, n, out);
case TB_PFN256V: return out+p4nenc256v32(in, n, out);
case TB_FOR256V: b = bitfm32( in, n, &x); vbxput32(out, x); *out++=b; return bitfpack256v32( in, n, out, x, b);
case TB_BP256V: if(b < 0) { BITSIZE32(in, n, b); *out++ = b; }
return n == 256?bitpack256v32(in, n, out, b):bitpack32(in, n, out, b);
#endif
#endif
//----- External libs ----------
case P_COPY: return u32enc( in, n, (unsigned *)out);
case P_MCPY: memcpy(out, _in, _n); return out+_n;
#if C_SIMPLE8B
case AM_SIMPLE8B:return vs8benc( in, n, out);
#endif
#if C_FASTPFOR
case FP_FASTPFOR: {
size_t nvalue = outsize/4;
FastPForLib::FastPFor<4> ic; ic.encodeArray((const int32_t *)in, n & (~127), (uint32_t *)(out+4), nvalue);
if(n & 127) {
size_t nvalue2 = outsize/4 - nvalue;
FastPForLib::VariableByte vc; vc.encodeArray((const int32_t *)(in + (n & (~127))), n & 127, (uint32_t *)(out + 4 + nvalue*4), nvalue2);
nvalue += nvalue2;
}
ctou32(out) = nvalue;
return out+4+nvalue*4;
}
case FP_SIMDFASTPFOR: {
size_t nvalue = outsize/4;
FastPForLib::SIMDFastPFor<4> ic; ic.encodeArray((const int32_t *)in, n & (~127), (uint32_t *)(out+4), nvalue);
if(n & 127) {
size_t nvalue2 = outsize/4 - nvalue;
FastPForLib::VariableByte vc; vc.encodeArray((const int32_t *)(in + (n & (~127))), n & 127, (uint32_t *)(out + 4 + nvalue*4), nvalue2);
nvalue += nvalue2;
}
ctou32(out) = nvalue;
return out+4+nvalue*4;
}
case FP_SIMDOPTPFOR: {
size_t nvalue = outsize/4;
FastPForLib::SIMDOPTPFor<4> ic; ic.encodeArray((const int32_t *)in, n & (~127), (uint32_t *)(out+4), nvalue);
if(n & 127) {
size_t nvalue2 = outsize/4 - nvalue;
FastPForLib::VariableByte vc; vc.encodeArray((const int32_t *)(in + (n & (~127))), n & 127, (uint32_t *)(out + 4 + nvalue*4), nvalue2);
nvalue += nvalue2;
}
ctou32(out) = nvalue;
return out+4+nvalue*4;
}
case FP_VBYTE: { size_t nvalue=outsize/4; FastPForLib::VariableByte ic; ic.encodeArray((const int32_t *)in, (const size_t)n, (uint32_t *)(out+4), nvalue); ctou32(out)=nvalue; return out+4+nvalue*4; }
case FP_SIMPLE8BRLE:{ size_t nvalue=outsize/4; FastPForLib::Simple8b_RLE<true> ic; ic.encodeArray((const int32_t *)in, (const size_t)n, (uint32_t *)(out+4), nvalue); ctou32(out)=nvalue; return out+4+nvalue*4; }
#endif
#if C_LIBFOR
case LF_FORX: //if(b < 0) b = maxbits(in), *out++ = b; return out + for_compress_bits(in, out, n, 0, b);
case LF_FOR: return out + for_compress_unsorted(in, out, n);
#endif
#if C_LITTLEPACK
case LI_PACK: if(b < 0) { b = maxbits_length(in,n); *out++ = b; } pack32( in, n, b, out); return out + byte_count(n,b);
case LI_HORPACK:
case LI_TURBOPACK: if(b < 0) { b = maxbits_length(in,n); *out++ = b; } turbopack32(in, n, b, out); return out + byte_count(n,b);
case LI_SCPACK: if(b < 0) { b = maxbits_length(in,n); *out++ = b; } scpack32( in, n, b, out); return out + byte_count(n,b);
#ifdef __AVX2__
case LI_BMIPACK: if(b < 0) { b = maxbits_length(in,n); *out++ = b; } bmipack32( in, n, b, out); return out + byte_count(n,b);
#endif
#endif
#if C_MASKEDVBYTE
case P_MASKEDVBYTE: return out + vbyte_encode(in, n, out);
#endif
#if C_POLYCOM
case PC_VBYTE: return vbpolyenc(in, n, out);
case PC_SIMPLE16: return vs16enc( in, n, (unsigned *)out);
case PC_RICE: return rcenc32( in, n, (unsigned *)out);
case PC_OPTPFD: return optpfdenc32(in, n, out); //if(n < 128) return vbyteenc(in, n, (unsigned *)out); else { unsigned tmp[2048]; for(i = 0; i < n; i++) tmp[i] = in[i]; return out + OPT4(tmp, n, (unsigned *)out); }
#endif
#if C_QMX
case P_QMX: { ANT_compress_qmx qmx; unsigned r=qmx.compress(out+4, outsize, (uint32_t *)in, (uint64_t)n); ctou32(out)=r; return out+4+r; } // { unsigned char *q = qmx_enc(in, n, out+4); ctou32(out) = q - (out+4); return q;
case P_QMX2: { ANT_compress_qmx_v2 qmx; unsigned r=qmx.compress(out+4, outsize, (uint32_t *)in, (uint64_t)n); ctou32(out)=r; return out+4+r; }
case P_QMX3: { ANT_compress_qmx_v3 qmx; unsigned r=qmx.compress(out+4, outsize, (uint32_t *)in, (uint64_t)n); ctou32(out)=r; return out+4+r; }
case P_QMX4: { ANT_compress_qmx_v4 qmx; unsigned r=qmx.compress(out+4, outsize, (uint32_t *)in, (uint64_t)n); ctou32(out)=r; return out+4+r; }
#endif
#if C_SIMDCOMP
//case SC_PACK: if(b < 0) { b = maxbits_length(in,n); *out++ = b; } return fastpackwithoutmask32(in, n, (uint32_t *)out, b);
case SC_FOR:
case SC_FORDA:
case SC_SIMDPACK128:if(b < 0) b = maxbits(in), *out++ = b; return (unsigned char *)simdpack_length(in, n, (__m128i *)out, b);
#ifdef __AVX2__
case SC_SIMDPACK256:if(b < 0) b = avxmaxbits(in), *out++ = b; return (unsigned char *)avxpackwithoutmaskn(in, n, (unsigned *)out, b);
#endif
#endif
#if C_STREAMVBYTE
case P_STREAMVBYTE: return out + streamvbyte_encode(in, n, out);
#endif
#if C_VARINTG8IU
case P_VARINTG8IU: return vintg8enc(in, n, out);
#endif
// --------- transform ----------------------------------------
#if C_TRANSFORM
case TB_ZIGZAG_32: bitzigzag32(in, n, (unsigned *)out, 0); return out + n*4;
case TB_TP8_32: _transpose4( (unsigned char *)in, n*4, out); return out + n*4;
case TB_TP8V_32: TRANSPOSE( (unsigned char *)in, n*4, out); return out + n*4;
case TB_TP4V_32: transposen4( (unsigned char *)in, n*4, out); return out + n*4;
#endif
#if C_C_BLOSC
case P_BS_SHUFFLE: shuffle( 4, n*4, (unsigned char *)in, out); return out + n*4;
#endif
#if C_BITSHUFFLE
case P_BITSHUFFLE: BITSHUFFLE( (unsigned char *)in, n*4, out);/* bshuf_bitshuffle(in, out, n, 4, 0);*/ return out + n*4;
#endif
// --------- transpose + lz77 ----------------------------------------
#if C_LZTURBO
case P_LZT10: { n *= 4; TRANSPOSE( (unsigned char *)in, n, sbuf); struct lzobj lz; lz.srclen = n; lz.src = sbuf; lz.dst = out; lz.dstlen = n; lz.level = 0; lz.hbits = 16; return out + lz8c01(&lz); }
case P_LZT20: { n *= 4; TRANSPOSE( (unsigned char *)in, n, sbuf); struct lzobj lz; lz.srclen = n; lz.src = sbuf; lz.dst = out; lz.dstlen = n; lz.level = 0; lz.hbits = 16; return out + lzbc01(&lz); }
case P_LZT22: { n *= 4; TRANSPOSE( (unsigned char *)in, n, sbuf); struct lzobj lz; lz.srclen = n; lz.src = sbuf; lz.dst = out; lz.dstlen = n; lz.level = 2; lz.hbits = 26; return out + lzbc2(&lz); }
case P_LZT32: { n *= 4; TRANSPOSE( (unsigned char *)in, n, sbuf); struct lzobj lz; lz.srclen = n; lz.src = sbuf; lz.dst = out; lz.dstlen = n; lz.level = 2; lz.hbits = 26; return out + lzhc2(&lz); }
//case P_VSHUF: { unsigned char *p = vsenc32(in, n, sbuf); p = mheenc(sbuf, p-sbuf, MH_BLK, out+2, out+n*4+256); ctou16(out) = p-(out+2); printf("L=%d ", p-(out+2)); return p; } //out + anscompress( sbuf, xb, out, out+n*4);
//case P_VSHUF: { unsigned char *p = mheenc(in, n*4, MH_BLK, out+4, out+n*4+256); ctou32(out) = p-(out+4); return p; } //out + anscompress( sbuf, xb, out, out+n*4);
case P_VSHUF: { unsigned char *p;/* = vsenc32(in, n, sbuf)*/ n *= 4; TRANSPOSE( (unsigned char *)in, n, sbuf); p = mheenc(sbuf, n/*p-sbuf*/, MH_BLK, out+4, out+n+256); ctou32(out) = p-(out+4); return p; } //out + anscompress( sbuf, xb, out, out+n*4);
#endif
#if C_BITSHUFFLE
case TB_VSIMPLE_S1: BITSHUFFLE( (unsigned char *)in, n*4, sbuf); return vsenc32((unsigned *)sbuf, n, out);
#endif
case TB_VSIMPLE_S4: transposen4( (unsigned char *)in, n*4, sbuf); return vsenc32((unsigned *)sbuf, n, out);
case TB_VSIMPLE_S8: TRANSPOSE( (unsigned char *)in, n*4, sbuf); return vsenc32((unsigned *)sbuf, n, out);
#if C_LZ4
#if C_BITSHUFFLE
case LZ4_S1: BITSHUFFLE( (unsigned char *)in, n*4, sbuf); return out + LZ4_compress((char *)sbuf, (char *)out, n*4);
#endif
case LZ4_ : return out + LZ4_compress((char *)in, (char *)out, n*4);
case LZ4_S4: transposen4( (unsigned char *)in, n*4, sbuf); return out + LZ4_compress((char *)sbuf, (char *)out, n*4);
case LZ4_S8: TRANSPOSE( (unsigned char *)in, n*4, sbuf); return out + LZ4_compress((char *)sbuf, (char *)out, n*4);
#endif
#if C_C_BLOSC
case P_BS_LZ:
case P_BS_LZ4:
case P_BS_ZLIB: return out + blosc_compress(lev, BLOSC_SHUFFLE, 4/*typesize*/, n*4, in, out, n*4+BLOSC_MAX_OVERHEAD);
//return out + blosc_compress(1/*clevel*/, BLOSC_BITSHUFFLE, 4/*typesize*/, n*4, in, out, n*4+BLOSC_MAX_OVERHEAD);
#endif
#if C_ZLIB
case P_ZLIB1: case P_ZLIB2: case P_ZLIB3: case P_ZLIB4: case P_ZLIB5: case P_ZLIB6: case P_ZLIB7: case P_ZLIB8: case P_ZLIB9:
{ n *= 4; TRANSPOSE( (unsigned char *)in, n, sbuf); uLongf outlen = n; int rc = compress2(out+4, &outlen, sbuf, n, codec-P_ZLIB1+1); if(rc != Z_OK) die("zlib compress2 rc=%d\n", rc); *(unsigned *)out = outlen; return out + 4 + outlen; }
#endif
default : die("library '%d' not included\n", codec);
}
return out;
}
unsigned char *coddecomp(unsigned char *in, unsigned _n, unsigned char *_out, int outlen, int codec, int lev, char *prm, int b) {
unsigned *out = (unsigned *)_out, n = (outlen+3) / 4,x;
switch(codec) {
case TB_VBYTE: return vbdec32( in, n, out); // case P_VU: return vudec32( in, n, out);
case TB_VSIMPLE: return vsdec32( in, n, out);
case TB_EF: return in;
case TB_PFDA : return p4decx32( in, n, out);
case TB_FOR: vbxget32(in, x); b = *in++; return bitfunpack32( in, n, out, x, b);
case TB_FORDA: vbxget32(in, x); b = *in++; return bitfunpackx32(in, n, out, x, b);
case TB_BP: if(b < 0) b = *in++; return bitunpack32( in, n, out, b);
case TB_BPN: return in+bitnunpack32( in, n, out);
#if C_TURBOPFORV
case TB_PF128V : __builtin_prefetch(in+256);return n == 128?p4dec128v32(in, n, out):p4dec32(in, n, out);
case TB_PFN128V : return in+p4ndec128v32(in, n, out);
case TB_EF128V: return in;
case TB_FOR128V: vbxget32(in, x); b = *in++;return bitfunpack128v32( in, n, out, x, b);
case TB_BP128V: if(b < 0) b = *in++; return n != 128?bitunpack32(in, n, out, b):bitunpack128v32(in, n, out, b);
#ifdef __AVX2__
case TB_FOR256V: vbxget32(in, x); b = *in++;return bitfunpack256v32( in, n, out, x, b);
case TB_PF256V : __builtin_prefetch(in+256); return n == 256?p4dec256v32(in, n, out):p4dec32(in, n, out);
case TB_PFN256V : return in+p4ndec256v32(in, n, out);
case TB_BP256V: if(b < 0) b = *in++; return n != 256?bitunpack32(in, n, out, b):bitunpack256v32(in, n, out, b);
#endif
#endif
#if C_TURBOPFORH
case TB_BP128H: if(b < 0) b = *in++; return n == 128?bitunpack128h32(in, n, out, b):bitunpack32(in, n, out, b);
#endif
//---- external libs -------------
case P_COPY: return u32dec( (unsigned *)in, n, out);
case P_MCPY: memcpy(_out, in, _n); return in+_n;
#if C_SIMPLE8B
case AM_SIMPLE8B: return vs8bdec( in, n, out);
#endif
#if C_FASTPFOR
case FP_FASTPFOR: {
size_t nvalue = n;
FastPForLib::FastPFor<4> ic; const uint32_t *ip = ic.decodeArray((const int32_t *)(in+4), ctou32(in), out, nvalue);
if(n & 127) {
nvalue = n - nvalue;
FastPForLib::VariableByte vc;
return (unsigned char *)vc.decodeArray(ip, (const uint32_t *)in+1+ctou32(in) - ip, out + (n&(~127)), nvalue); //return vbdec32((unsigned char *)ip, n & 127, out + mynvalue1);
}
return (unsigned char *)ip;
}
case FP_SIMDFASTPFOR: {
size_t nvalue = n;
FastPForLib::SIMDFastPFor<4> ic; const uint32_t *ip = ic.decodeArray((const int32_t *)(in+4), ctou32(in), out, nvalue);
if(n & 127) {
nvalue = n - nvalue;
FastPForLib::VariableByte vc;
return (unsigned char *)vc.decodeArray(ip, (const uint32_t *)in+1+ctou32(in) - ip, out + (n&(~127)), nvalue); //return vbdec32((unsigned char *)ip, n & 127, out + mynvalue1);
}
return (unsigned char *)ip;
}
case FP_SIMDOPTPFOR: {
size_t nvalue = n;
FastPForLib::SIMDOPTPFor<4> ic; const uint32_t *ip = ic.decodeArray((const int32_t *)(in+4), ctou32(in), out, nvalue);
if(n & 127) {
nvalue = n - nvalue;
FastPForLib::VariableByte vc;
return (unsigned char *)vc.decodeArray(ip, (const uint32_t *)in+1+ctou32(in) - ip, out + (n&(~127)), nvalue); //return vbdec32((unsigned char *)ip, n & 127, out + mynvalue1);
}
return (unsigned char *)ip;
}
case FP_VBYTE: //return vbytedec( in, n, out);
{ size_t nvalue=n; FastPForLib::VariableByte ic; return (unsigned char *)ic.decodeArray((const int32_t *)(in+4), ctou32(in), (uint32_t *)out, nvalue); }
case FP_SIMPLE8BRLE:
{ size_t nvalue=n; FastPForLib::Simple8b_RLE<true> ic; ic.decodeArray((const int32_t *)(in+4), ctou32(in), (uint32_t *)out, nvalue); return in+4+ctou32(in)*4; }
#endif
#if C_LIBFOR
case LF_FORX: return for_selectx(in, n, out); //if(b < 0) b = *in++; return for_selectx(in, n, out, 0, b);//
case LF_FOR: return in + for_uncompress(in, out, n); //if(b < 0) b = *in++; return in + for_uncompress_bits(in, out, n, 0, b); //return in + for_uncompress(in, out, n);
#endif
#if C_LITTLEPACK
case LI_PACK: if(b < 0) b = *in++; unpack32( in, n, b, out); return in + byte_count(n,b);
case LI_TURBOPACK: if(b < 0) b = *in++; turbounpack32( in, n, b, out); return in + byte_count(n,b);
case LI_SCPACK: if(b < 0) b = *in++; scunpack32( in, n, b, out); return in + byte_count(n,b);
case LI_HORPACK: if(b < 0) b = *in++; horizontalunpack32(in, n, b, out); return in + byte_count(n,b);
#ifdef __AVX2__
case LI_BMIPACK: if(b < 0) b = *in++; bmiunpack32( in, n, b, out); return in + byte_count(n,b);
#endif
#endif
#if C_MASKEDVBYTE
case P_MASKEDVBYTE: return in + masked_vbyte_decode(in, out, n);
#endif
#if C_POLYCOM
case PC_VBYTE: return vbpolydec(in, n, out);
case PC_SIMPLE16: return vs16dec( (unsigned *)in, n, out);
case PC_RICE: return rcdec32( (unsigned *)in, n, out);
case PC_OPTPFD : return optpfddec32( in, n, out); //if(n < 128) return vbytedec(in, n, out); else { unsigned all_array[2048]; return (unsigned char *)detailed_p4_decode(out, (unsigned *)in, all_array); }
#endif
#if C_SIMDCOMP
//case SC_PACK: if(b < 0) b = *in++; return fastunpack32((uint32_t *)in, n, out, b);
case SC_FOR:
case SC_SIMDPACK128: if(b < 0) b = *in++; return (unsigned char *)simdunpack_length( (__m128i *)in, n, out, b);
case SC_FORDA: if(b < 0) b = *in++; return simdfor_selectx(in, n, out, 0, b);
#ifdef __AVX2__
case SC_SIMDPACK256: if(b < 0) b = *in++; return (unsigned char *)avxunpackn( (unsigned *)in, n, out, b);
#endif
#endif
#if C_STREAMVBYTE
case P_STREAMVBYTE: return in + streamvbyte_decode(in, out, n);
#endif
#if C_QMX //case P_QMX: return qmx_dec(in+4, ctou32(in), out, n);
case P_QMX: { ANT_compress_qmx qmx; qmx.decompress(out, n, in+4, ctou32(in)); return in+4+ctou32(in);} // { unsigned char *q = qmx_enc(in, n, out+4); ctou32(out) = q - (out+4); return q;
case P_QMX2: { ANT_compress_qmx_v2 qmx; qmx.decompress(out, n, in+4, ctou32(in)); return in+4+ctou32(in);}
case P_QMX3: { ANT_compress_qmx_v3 qmx; qmx.decompress(out, n, in+4, ctou32(in)); return in+4+ctou32(in);}
case P_QMX4: { ANT_compress_qmx_v4 qmx; qmx.decompress(out, n, in+4, ctou32(in)); return in+4+ctou32(in);}
#endif
#if C_VARINTG8IU
case P_VARINTG8IU: return vintg8dec(in, n, out);
#endif
//---------- transform ----------------------
#if C_TRANSFORM
case TB_ZIGZAG_32: memcpy(out, in, n*4); bitunzigzag32(out, n, 0); return in + n*4;
case TB_TP8_32: _untranspose4( (unsigned char *)in, n*4, (unsigned char *)out); return in + n*4;
case TB_TP8V_32: UNTRANSPOSE( (unsigned char *)in, n*4, (unsigned char *)out); return in + n*4;
case TB_TP4V_32: untransposen4( (unsigned char *)in, n*4, (unsigned char *)out); return in + n*4;
#endif
#if C_C_BLOSC
case P_BS_SHUFFLE: unshuffle( 4, n*4, (unsigned char *)in, (unsigned char *)out); return in + n*4;
#endif
#if C_BITSHUFFLE
case P_BITSHUFFLE: BITUNSHUFFLE(in, n*4, (unsigned char *)out); /*bshuf_bitunshuffle(in, out, n, 4, 0);*/ return in + n*4;
#endif
//---------- transpose + lz77 ----------------------
#if C_LZTURBO
case P_LZT10: { struct lzobj lz; lz.dstlen = n*4; lz.src = in; lz.dst = sbuf; lz.level = 0; in += lz8d(&lz); UNTRANSPOSE(sbuf, n*4, (unsigned char *)out); } break;
case P_LZT20:
case P_LZT22: { struct lzobj lz; lz.dstlen = n*4; lz.src = in; lz.dst = sbuf; lz.level = 0; in += lzbd(&lz); UNTRANSPOSE(sbuf, n*4, (unsigned char *)out); } break;
case P_LZT32: { struct lzobj lz; lz.dstlen = n*4; lz.src = in; lz.dst = sbuf; lz.level = 0; in += lzhd(&lz); UNTRANSPOSE(sbuf, n*4, (unsigned char *)out); } break;
//case P_VSHUF: { unsigned inlen = ctou16(in); printf("l=%d ", inlen); in+=2; in = mhdec(in, inlen, MH_BLK, sbuf, n*4); vsdec32(sbuf, n, out); } break;// in = ansdecompress( in, n*4, sbuf);
//case P_VSHUF: { unsigned inlen = ctou32(in); in+=4; in = mhdec(in, inlen, MH_BLK, out, n*4); } break;// in = ansdecompress( in, n*4, sbuf);
case P_VSHUF: { unsigned inlen = ctou32(in); in+=4; in = mhdec(in, inlen, MH_BLK, sbuf, n*4); UNTRANSPOSE(sbuf, n*4, (unsigned char *)out); /*vsdec32(sbuf, n, out);*/ } break;// in = ansdecompress( in, n*4, sbuf);
#endif
#if C_BITSHUFFLE
case TB_VSIMPLE_S1: in = vsdec32(in, n, (unsigned *)sbuf); BITUNSHUFFLE(sbuf, n*4, (unsigned char *)out); break;/*bshuf_bitunshuffle(sbuf, out, n*4/32, 32, 0); */
#endif
case TB_VSIMPLE_S4: in = vsdec32(in, n, (unsigned *)sbuf); untransposen4(sbuf, n*4, (unsigned char *)out); break;
case TB_VSIMPLE_S8: in = vsdec32(in, n, (unsigned *)sbuf); UNTRANSPOSE(sbuf, n*4, (unsigned char *)out); break;/*bshuf_bitunshuffle(sbuf, out, n*4/32, 32, 0); */
#if C_LZ4
#if C_BITSHUFFLE
case LZ4_S1: in += LZ4_decompress_fast((char *)in, (char *)sbuf, n*4); BITUNSHUFFLE(sbuf, n*4, (unsigned char *)out); break;/*bshuf_bitunshuffle(sbuf, out, n*4/32, 32, 0); */
#endif
case LZ4_ : in += LZ4_decompress_fast((char *)in, (char *)out, n*4); break;
case LZ4_S4: in += LZ4_decompress_fast((char *)in, (char *)sbuf, n*4); untransposen4(sbuf, n*4, (unsigned char *)out); break;
case LZ4_S8: in += LZ4_decompress_fast((char *)in, (char *)sbuf, n*4); UNTRANSPOSE(sbuf, n*4, (unsigned char *)out); break;/*bshuf_bitunshuffle(sbuf, out, n*4/32, 32, 0); */
#endif
#if C_C_BLOSC
case P_BS_LZ:
case P_BS_LZ4:
case P_BS_ZLIB: { blosc_decompress(in, out, n*4); size_t nbytes, cbytes,blocksize; blosc_cbuffer_sizes(in, &nbytes, &cbytes, &blocksize); return in+cbytes; }
#endif
#if C_ZLIB
case P_ZLIB1: case P_ZLIB2: case P_ZLIB3: case P_ZLIB4: case P_ZLIB5: case P_ZLIB6: case P_ZLIB7: case P_ZLIB8: case P_ZLIB9:
{ uLongf outsize = n*4; int l = *(unsigned *)in, rc = uncompress(sbuf, &outsize, in+4, l); in += 4 + l; UNTRANSPOSE(sbuf, n*4, (unsigned char *)out); } break;
#endif
default : die("library '%d' not included\n", codec);
}
return in;
}
char *codver(int codec, char *v, char *s) {
switch(codec) {
#if C_C_BLOSC
return BLOSC_VERSION_STRING;
#endif
#if C_LZ4
case LZ4_: sprintf(s,"%d.%d.%d", LZ4_VERSION_MAJOR, LZ4_VERSION_MINOR, LZ4_VERSION_RELEASE); return s;
#endif
#if C_ZSTD
case P_ZSTD: sprintf(s,"%d.%d.%d", ZSTD_VERSION_MAJOR, ZSTD_VERSION_MINOR, ZSTD_VERSION_RELEASE); return s;
#endif
default: if(v) strcpy(s,v);
}
return s;
}
Reference in New Issue View Git Blame Copy Permalink