Location:
Search - sse3
Search list
Description: 我所有的dct变换程序,包括mmx,sse,sse2等,对于大家应该有用。-all my A1501 transform procedures, including mmx, SSE, SSE3, should be useful for everyone.
Platform: |
Size: 38168 |
Author: 潘敏 |
Hits:
Description: Please excuse my english... It's so bad :)
Hacker Disassembler Engine, or HDE, is small disassembler engine, which intend to
code analyse. HDE get length of command, prefixes, ModR/M and SIB bytes, opcode,
immediate, displacement, relative address, etc. For example, you can use HDE when
writing unpackers or decryptors executable files, because more others disassemblers
too big, get only disasm listing and aren't intended for code analys, but more simple
length disassemblers usually get too little info. HDE get enough info to analyse, but
it has very small size. HDE package include DLL, objects, headers files and and source.
+ support MMX, SSE, SSE2, SSE3, 3DNow! instructions
+ high-speed & small size (coded in assembler ;)
+ compatibility with most coding language
To disassemble should call hde_disasm function:
DWORD hde_disasm(
void *pCode // pointer to code
HDE_STRUCT *pHDE_STRUCT // pointer to structure HDE_STRUCT
);
Function return length of command and fill structure HDE_STRUCT:
struct HDE_STRUCT
{
BYTE len; // length of command
BYTE p_rep; // rep/repnz/.. prefix: 0xF2 or 0xF3
BYTE p_lock; // lock prefix 0xF0
BYTE p_seg; // segment prefix: 0x2E, 0x36, 0x3E, 0x26, 0x64, 0x65
BYTE p_66; // prefix 0x66
BYTE p_67; // prefix 0x67
BYTE opcode; // opcode
BYTE opcode2; // second opcode, if first opcode equal 0x0F
BYTE modrm; // ModR/M byte
BYTE modrm_mod; // - mod byte of ModR/M
BYTE modrm_reg; // - reg byte of ModR/M
BYTE modrm_rm; // - r/m byte of ModR/M
BYTE sib; // SIB byte
BYTE sib_scale; // - scale (ss) byte of SIB
BYTE sib_index; // - index byte of SIB
BYTE sib_base; // - base byte of SIB
BYTE imm8; // immediate imm8
WORD imm16; // immediate imm16
DWORD imm32; // immediate imm32
BYTE disp8; // displacement disp8
WORD disp16; // displacement disp16, if prefix 0x67 exist
DWORD disp32; // displacement disp32
BYTE rel8; // relative address rel8
WORD rel16; // relative address rel16, if prefix 0x66 exist
DWORD rel32; // relative address rel32
};
Opcode and len fields always exist, others are optional and depend of command. If
field's value equal zero, then it isn't existing.
Note: HDE work only with 32-bit commands of x86 processors !
Platform: |
Size: 23447 |
Author: sys0007 |
Hits:
Description: 我所有的dct变换程序,包括mmx,sse,sse2等,对于大家应该有用。-all my A1501 transform procedures, including mmx, SSE, SSE3, should be useful for everyone.
Platform: |
Size: 37888 |
Author: 潘敏 |
Hits:
Description: The fastest Nokia H.264 decoder and fixed bug. Using method is same as xvid mpeg4,support ARGB MMX output, support VC6.0, VC2008 and Wince. You also can install Intel C++ compiler 10.1 or 11.0 with VC2008,
set command line option /QxP or SSE3 in "Enable Enhanced Instruction Set" make it much more faster.
I am Jesse Stone from Taiwan. msn:fatalfeel@hotmail.com
reference from http://media.library.ku.edu.tr/theses/msctheses/A/akkus_istemi,The fastest Nokia H.264 decoder and fixed bug. Using method is same as xvid mpeg4,support ARGB MMX output, support VC6.0, VC2008 and Wince. You also can install Intel C++ compiler 10.1 or 11.0 with VC2008,
set command line option /QxP or SSE3 in "Enable Enhanced Instruction Set" make it much more faster.
I am Jesse Stone from Taiwan. msn:fatalfeel@hotmail.com
reference from http://media.library.ku.edu.tr/theses/msctheses/A/akkus_istemi
Platform: |
Size: 607232 |
Author: 小誠 |
Hits:
Description: The fastest Nokia H.264 decoder and fixed bug. Using method is same as xvid mpeg4,support ARGB MMX output, support VC6.0, VC2008 and Wince. You also can install Intel C++ compiler 10.1 or 11.0 with VC2008, set command line option /QxP or SSE3 in "Enable Enhanced Instruction Set" make it much more faster. I am Jesse Stone from Taiwan. msn:fatalfeel@hotmail.com reference from http://media.library.ku.edu.tr/theses/msctheses/A/akkus_istemi ,The fastest Nokia H.264 decoder and fixed bug. Using method is same as xvid mpeg4, support ARGB MMX output, support VC6.0, VC2008 and Wince. You also can install Intel C++ Compiler 10.1 or 11.0 with VC2008, set command line option/QxP or SSE3 in
Platform: |
Size: 615424 |
Author: 小誠 |
Hits:
Description: The fastest Nokia H.264 decoder and fixed bug. Using method is same as xvid mpeg4,support ARGB MMX output, support VC6.0, VC2008 and Wince. You also can install Intel C++ compiler 10.1 or 11.0 with VC2008, set command line option /QxP or SSE3 in "Enable Enhanced Instruction Set" make it much more faster. I am Jesse Stone from Taiwan. msn:fatalfeel@hotmail.com reference from http://media.library.ku.edu.tr/theses/msctheses/A/akkus_istemi,The fastest Nokia H.264 decoder and fixed bug. Using method is same as xvid mpeg4, support ARGB MMX output, support VC6.0, VC2008 and Wince. You also can install Intel C++ Compiler 10.1 or 11.0 with VC2008, set command line option/QxP or SSE3 in Enable Enhanced Instruction Set make it much more faster. I am Jesse Stone from Taiwan. msn: fatalfeel@hotmail.com reference from http://media.library.ku.edu.tr/ theses/msctheses/A/akkus_istemi
Platform: |
Size: 614400 |
Author: 小誠 |
Hits:
Description: hde32 is a small disassembler engine, intended for analysis of x86-32 code. It gets length of command, prefixes, ModR/M, SIB, opcode, etc. For example, you can use hde32 when writing unpackers of executables, viruses, because most other disassemblers too big, get only assembler listing and are not intended for analysis of code, but most simple length disassemblers get too little info. hde32 gets enough info for analysis, but it has very small size.
* support FPU, MMX, SSE, SSE2, SSE3, 3DNow! instructions
* high-speed and small size (628 bytes)
* position independent code
* operating system independent code
* doesn t use commands of processor higher than i386
-hde32 is a small disassembler engine, intended for analysis of x86-32 code. It gets length of command, prefixes, ModR/M, SIB, opcode, etc. For example, you can use hde32 when writing unpackers of executables, viruses, because most other disassemblers too big, get only assembler listing and are not intended for analysis of code, but most simple length disassemblers get too little info. hde32 gets enough info for analysis, but it has very small size.
* support FPU, MMX, SSE, SSE2, SSE3, 3DNow! instructions
* high-speed and small size (628 bytes)
* position independent code
* operating system independent code
* doesn t use commands of processor higher than i386
Platform: |
Size: 4096 |
Author: Merc |
Hits:
Description: ffh264 in VC or Mac. Use C++ class rebuild it, and add MMX,SSE3 asm code.
Win32 need preinstall MinGw32. My name:Jesse Stone Taiwan. email:fatalfeel@hotmail.com-ffh264 in VC or Mac. Use C++ class rebuild it, and add MMX,SSE3 asm code.
need preinstall MinGw32. My name:Jesse Stone Taiwan. email:fatalfeel@hotmail.com
Platform: |
Size: 456704 |
Author: 小誠 |
Hits:
Description: ffh264 decoder in vc. add MMX and SSE3
need install mingw.
by Jesse Stone, Taiwan
email:fatalfeel@hotmail.com
Platform: |
Size: 420864 |
Author: 小誠 |
Hits:
Description: H264 decode in VC 9 include sse3 codec
1. install MinGW
2. set executable files path for VC EX. C:\Program Files\MinGW\bin
3. build path name only in ascii code, can not unicode EX. c:\hi_ff264 or d:\hi_ff264 is good
4. Choice register in config.h
#define HAVE_MMX 1
#define HAVE_MMX2 1
#define HAVE_SSE 1
#define HAVE_SSSE3 1
5. this code extract from ffmpeg-r26400-swscale-r32676 can decode h.264 profile format
6. my email: fatalfeel@hotmail.com by Jesse stone on Taiwan
Platform: |
Size: 420864 |
Author: 小誠 |
Hits:
Description: x86指令模擬器,可在C++開發環境下模擬x86/x64指令的運行-AsmJit is a complete x86/x64 JIT Assembler for C++ language. It supports 32-bit and 64-bit mode, FPU, MMX, 3dNow, SSE, SSE2, SSE3 and SSE4 through type-safe API which mimics an Intel assembler syntax and eliminates nearly all common mistakes which can be done by developers at compile time or run-time.
AsmJit has a high-level code generation classes which can be used as a portable way to create JIT code. It abstracts differences caused by 32-bit/64-bit mode, function calling conventions, and platform specific ABI.
AsmJit has been successfully tested by various C++ compilers (including MSVC, GCC, Clang, and BorlandC++) under all major operating systems (including Windows, Linux and Mac).
Platform: |
Size: 953344 |
Author: 去去去 |
Hits:
Description: vp8 sad 16x16x3 sse3 for Embedded Linux.
Platform: |
Size: 2048 |
Author: pengjuson |
Hits:
Description: Glue code for the SHA1 Secure Hash Algorithm assembler implementation using Supplemental SSE3 instructions.
Platform: |
Size: 2048 |
Author: riufengtao |
Hits:
Description: Glue code for the SHA1 Secure Hash Algorithm assembler implementation using Supplemental SSE3 instructions.
Platform: |
Size: 2048 |
Author: brchwb |
Hits:
Description: SSE(SSE2,SSE3,SSE4)指令集,及AVX指令集,算法加速必备文档-SSE (SSE2, SSE3, SSE4) instruction set, and AVX instruction set, algorithm acceleration necessary documents
Platform: |
Size: 4315136 |
Author: 于由 |
Hits:
Description: SSE的指令集,包括SSE/SSE2/SSE3和SSE4。每条指令的用法和说明-SSE instruction set, including SSE/SSE2/SSE3 and SSE4. Each instruction and usage instructions
Platform: |
Size: 13312 |
Author: 史巍 |
Hits: