Udis86 - Disassembler Library for x86 and AMD64

Documentation · Project Page · Author · Changes

Udis86 is an easy-to-use minimalistic disassembler library (libudis86) for the x86 and AMD64 (x86-64) range of instruction set architectures. The primary intent of the design and development of udis86 is to aid software development projects that entail binary code analysis.

libudis86

1. Full support for the x86 and x86-64 (AMD64) range of instruction set architectures.
2. Full support for all AMD-V, INTEL-VMX, MMX, SSE, SSE2, SSE3, FPU(x87), and AMD 3Dnow! instructions.
3. Supports 16bit, 32bit, and 64bit disassembly modes.
4. Generates output in AT&T or INTEL assembler language syntaxes.
5. Supports flexbile input methods: File, Buffer, and Hooks.
6. Thread-safe and Reentrant.
7. Clean and very easy-to-use API.
8. Builds on *nix systems, Win32, DJGPP (new), Standalone, etc.

See Usage.

udcli

A front-end incarnation of this library, udcli is a small command-line tool for your quick disassembly needs.
See Usage.

Author

Udis86 is a creation of Vivek Mohan. You can reach me at vivek[at]sig9[dot]com. Please let me know if you are using udis86, have ideas for it, or would like to comment on it.

CHANGES

 * [Aug/05/07] Patch for DJGPP (Thanks to Robert Riebisch)
	       Removed "test" Target in tot Makefile (Thanks to Robert Riebisch)
	       New target:tests.
	       Misc. clean up top-of-tree Makefile.
	       Use config.h for package version
	       Review/cleanup FP Opcode map. 
 * [Dec/18/06] Fixed: Printf warnings.
	       Fixed: set_input_buffered to set_input_buffer in documentation.
	       Added support for INTEL VMX instructions.
	       API addition: ud_set_vendor(): UD_VENDOR_{INTEL, AMD}.
	       Fixed: Handling of invalid combination of bytes. No more input re-syncing.
	       Added Makefiles for win32 build. (Thanks to Xi Hang).
	       Added Makefile.standalone for "Standalone Udis86".
	       Fixed: u->inp_end initialization bug.
	       Fixed: swapgs only in 64bit mode.
 * [Oct/27/06] 0xF1 is now int1.
	       Fixed: cast prefix for Group1_op80/82/C0/D0 instructions.
	       Fixed: lmsw
	       Added support for Udis86 Standalone.
 * [Sep/18/06] Fixed: Missing support for o32 for in/out instructions.
 * [Jul/26/06] Fixed: Missing REX.W in Group1_op83 instructions.
 * [Jul/16/06] Fixed: rex.b on [rip+disp]
 * [Jul/15/06] Fixed: Wrongly printing "loopnz" as "loopn" (Thanks to Sanjay Patel)
 	       Fixed: Wrongly decoding "out Ib, AL" ((Thanks to Sanjay Patel)
 * [Jun/8/06]  Missing instructions added. (As per AMD64 Manual Vol3 Rev 3.11 
               Dec 2005)
               Fixed minor decoder issues. (Thanks to Sanjay Patel)
               Fixed issues with input stream buffer. (Thanks to Sanjay Patel)
 * [May/18/06] Fixed issues with handling of 64-bit values. 
               Fixed bug in syntax generation for immediate operands in branch
               instructions.
               Fixed decoding of (G, W) operands.
 * [May/12/06] Added support for data types in MS VC++.
 * [May/11/06] Fixed portability issues concerning inp_uintXX functions in 
               input.c and 64-bit casting in syn-intel/syn-att.c (Thanks to 
               Sanjay Patel)