View source for Bytecode

{{Short description|Form of instruction set designed to be run by a software interpreter}}
{{redirect2|Portable code|P-code|other uses|software portability|and|P-Code (disambiguation)}}
{{More citations needed|date=January 2009}}
{{Use dmy dates|date=May 2019|cs1-dates=y}}
{{Program execution}}

'''Bytecode''' (also called '''portable code''' or '''p-code''') is a form of [[instruction set]] designed for efficient execution by a software [[interpreter (computing)|interpreter]]. Unlike [[Human-readable code|human-readable]]<ref name="Dynamic_Machine_Code"/> [[source code]], bytecodes are compact numeric codes, constants, and references (normally numeric addresses) that encode the result of [[compiler]] parsing and performing [[Semantic analysis (compilers)|semantic analysis]] of things like type, scope, and nesting depths of program objects.

The name ''bytecode'' stems from instruction sets that have one-[[byte]] [[opcode]]s followed by optional parameters. [[Intermediate representation]]s such as bytecode may be output by [[programming language]] implementations to ease [[interpreter (computing)|interpretation]], or it may be used to reduce hardware and [[operating system]] dependence by allowing the same code to run [[cross-platform]], on different devices. Bytecode may often be either directly executed on a [[virtual machine]] (a [[p-code machine]], i.e., interpreter), or it may be further compiled into [[machine code]] for better performance.

Since bytecode instructions are processed by software, they may be arbitrarily complex, but are nonetheless often akin to traditional hardware instructions: virtual [[stack machine]]s are the most common, but virtual [[register machine]]s have been built also.<ref name="Jucs_Lua"/><ref name="Dalvik"/> Different parts may often be stored in separate files, similar to [[object file|object modules]], but dynamically loaded during execution.

==Execution==
A bytecode program may be executed by parsing and ''directly'' executing the instructions, one at a time. This kind of ''bytecode interpreter'' is very portable. Some systems, called dynamic translators, or ''[[just-in-time compilation|just-in-time]]'' (JIT) compilers, translate bytecode into [[machine code]] as necessary at [[Run time (program lifecycle phase)|runtime]]. This makes the virtual machine hardware-specific but does not lose the portability of the bytecode. For example, [[Java (programming language)|Java]] and [[Smalltalk]] code is typically stored in bytecode format, which is typically then JIT compiled to translate the bytecode to machine code before execution. This introduces a delay before a program is run, when the bytecode is compiled to native machine code, but improves execution speed considerably compared to interpreting source code directly, normally by around an order of magnitude (10x).<ref name="Byte_Machine"/>

Because of its performance advantage, today many language implementations execute a program in two phases, first compiling the source code into bytecode, and then passing the bytecode to the virtual machine. There are bytecode based virtual machines of this sort for [[Java (programming language)|Java]], [[Raku (programming language)|Raku]], [[Python (programming language)|Python]], [[PHP]],{{efn|PHP has [[just-in-time compilation]] in PHP 8,<ref>{{Cite web|last=O’Phinney|first=Matthew Weier|title=Exploring the New PHP JIT Compiler|url=https://www.zend.com/blog/exploring-new-php-jit-compiler|access-date=2021-02-19|website=Zend by Perforce|language=en}}</ref><ref>{{Cite web|title=PHP 8: The JIT - stitcher.io|url=https://stitcher.io/blog/php-jit|access-date=2021-02-19|website=stitcher.io|language=en}}</ref> and before while not on in the default version, had options like [[HHVM]]. For older versions of PHP: Although [[PHP]] opcodes are generated each time the program is launched, and are always interpreted and not [[just-in-time compilation|just-in-time compiled]].}} [[Tcl]], [[AWK|mawk]] and [[Forth (programming language)|Forth]] (however, Forth is seldom compiled via bytecodes in this way, and its virtual machine is more generic instead). The implementation of [[Perl]] and [[Ruby (programming language)|Ruby]] 1.8 instead work by walking an [[abstract syntax tree]] representation derived from the source code.

More recently, the authors of [[V8 (JavaScript engine)|V8]]<ref name="Dynamic_Machine_Code"/> and [[Dart (programming language)|Dart]]<ref name="Loitsch_Bytecode"/> have challenged the notion that intermediate bytecode is needed for fast and efficient VM implementation. Both of these language implementations currently do direct JIT compiling from source code to machine code with no bytecode intermediary.<ref name="Javascript"/>

==Examples==
*[[ActionScript]] executes in the ActionScript Virtual Machine (AVM), which is part of Flash Player and [[Adobe AIR|AIR]]. ActionScript code is typically transformed into bytecode format by a [[compiler]]. Examples of compilers include one built into Adobe Flash Professional and one built into Adobe Flash Builder and available in the [[Apache Flex|Adobe Flex SDK]].
*[[Adobe Flash]] objects
*[[BANCStar programming language|BANCStar]], originally bytecode for an interface-building tool but used also as a language
*[[Berkeley Packet Filter]]
*[[EBPF]]
*Berkeley Pascal<ref>{{cite web|last=G.|first=Adam Y.|title=Berkeley Pascal|website=[[GitHub]] |date=2022-07-11|url=https://github.com/adamyg/berkeley_pascal|access-date=2022-01-08}}</ref>
*Byte Code Engineering Library
*C to [[Java virtual machine]] compilers
*[[CLISP]] implementation of [[Common Lisp]] used to compile only to bytecode for many years; however, now it also supports compiling to native code with the help of [[GNU lightning]]
*[[CMUCL]] and Scieneer Common Lisp implementations of [[Common Lisp]] can compile either to native code or to bytecode, which is far more compact
*[[Common Intermediate Language]] executed by [[Common Language Runtime]], used by [[.NET]] languages such as [[C Sharp (programming language)|C#]]
*[[Dalvik (software)|Dalvik]] bytecode, designed for the [[Android (operating system)|Android]] platform, is executed by the [[Dalvik (software)|Dalvik virtual machine]]
*Dis bytecode, designed for the [[Inferno (operating system)]], is executed by the [[Dis virtual machine]]
*[[EiffelStudio]] for the [[Eiffel (programming language)|Eiffel]] programming language
*EM, the [[Amsterdam Compiler Kit]] virtual machine used as an intermediate compiling language and as a modern bytecode language
*[[Emacs]] is a text editor with most of its functions implemented by [[Emacs Lisp]], its built-in dialect of [[Lisp (programming language)|Lisp]]. These features are compiled into bytecode. This architecture allows users to customize the editor with a high level language, which after compiling into bytecode yields reasonable performance.
*[[Embeddable Common Lisp]] implementation of [[Common Lisp]] can compile to bytecode or C code
*[[Common Lisp]] provides a <code>disassemble</code> function<ref>{{cite web|url=http://www.lispworks.com/documentation/HyperSpec/Body/f_disass.htm|title=CLHS: Function DISASSEMBLE|website=www.lispworks.com}}</ref> which prints to the standard output the underlying code of a specified function. The result is implementation-dependent and may or may not resolve to bytecode. Its inspection can be utilized for debugging and optimization purposes.<ref>{{cite web|url=https://lispcookbook.github.io/cl-cookbook/performance.html|title=The Common Lisp Cookbook – Performance Tuning and Tips|author=Collective|date=2023-12-13|website=lispcookbook.github.io}}</ref> [[Steel Bank Common Lisp]], for instance, produces:
:<syntaxhighlight lang="lisp">
(disassemble '(lambda (x) (print x)))
; disassembly for (LAMBDA (X))
; 2436F6DF:       850500000F22     TEST EAX, [#x220F0000]     ; no-arg-parsing entry point
;       E5:       8BD6             MOV EDX, ESI
;       E7:       8B05A8F63624     MOV EAX, [#x2436F6A8]      ; #<FDEFINITION object for PRINT>
;       ED:       B904000000       MOV ECX, 4
;       F2:       FF7504           PUSH DWORD PTR [EBP+4]
;       F5:       FF6005           JMP DWORD PTR [EAX+5]
;       F8:       CC0A             BREAK 10                   ; error trap
;       FA:       02               BYTE #X02
;       FB:       18               BYTE #X18                  ; INVALID-ARG-COUNT-ERROR
;       FC:       4F               BYTE #X4F                  ; ECX
</syntaxhighlight>
*Ericsson implementation of [[Erlang (programming language)|Erlang]] uses BEAM bytecodes
*[[Ethereum]]'s Virtual Machine (EVM) is the runtime environment, using its own bytecode, for transaction execution in Ethereum (smart contracts).
*[[Icon (programming language)|Icon]]<ref name="Arizona_Icom" /> and [[Unicon (programming language)|Unicon]]<ref name="Icon_Unicon"/> programming languages
*[[Infocom]] used the [[Z-machine]] to make its software applications more portable
*[[Java bytecode]], which is executed by the [[Java virtual machine]]
**[[ObjectWeb ASM|ASM]]
**[[BCEL]]
**Javassist
*[[Keiko bytecode]] used by the [[Oberon-2]] programming language to make it and the [[Oberon operating system]] more portable.
*[[KEYB (command)|KEYB]], the [[MS-DOS]]/[[PC&nbsp;DOS]] keyboard driver with its resource file [[KEYBOARD.SYS]] containing layout information and short [[p-code machine|p-code]] sequences executed by an interpreter inside the resident driver.<ref name="Paul_2001_KEYBOARD"/><ref name="Mendelson_2001_KEYBOARD"/>
*[[LLVM IR]]
*LSL, a scripting language used in virtual worlds compiles into bytecode running on a virtual machine. Second Life has the original Mono version, Inworldz developed the Phlox version.
*[[Lua (programming language)|Lua]] language uses a register-based bytecode virtual machine
*m-code of the [[MATLAB]] language<ref name="Patent_6973644"/>
*[[Malbolge]] is an [[esoteric programming language|esoteric]] [[machine language]] for a ternary virtual machine.
*[[P-code machine#Microsoft P-code|Microsoft P-code]] used in [[Visual C++]] and [[Visual Basic (classic)|Visual Basic]]
*[[Multiplan]]<ref name="Multiplan"/>
*[[O-code]] of the [[BCPL]] programming language
*[[OCaml]] language optionally compiles to a compact bytecode form
*[[p-code machine|p-code]] of [[UCSD Pascal]] implementation of the [[Pascal (programming language)|Pascal]] language
*[[Parrot virtual machine]]
*[[Pick operating system|Pick BASIC]] also referred to as Data BASIC or [[MultiValue#MultiValue DataBASIC|MultiValue BASIC]]
*The [[R (programming language)|R environment for statistical computing]] offers a bytecode compiler through the compiler package, now standard with R version 2.13.0. It is possible to compile this version of R so that the base and recommended packages exploit this.<ref name="cran_r"/>
*[[Pyramid 2000]] adventure game
*[[Python (programming language)|Python]] scripts are being compiled on execution to Python's bytecode language, and the compiled files (.pyc) are cached inside the script's folder
:Compiled code can be analysed and investigated using a built-in tool for debugging the low-level bytecode. The tool can be initialized from the shell, for example:
:<syntaxhighlight lang="pycon">
>>> import dis # "dis" - Disassembler of Python byte code into mnemonics.
>>> dis.dis('print("Hello, World!")')
  1           0 LOAD_NAME                0 (print)
              2 LOAD_CONST               0 ('Hello, World!')
              4 CALL_FUNCTION            1
              6 RETURN_VALUE
</syntaxhighlight>
*[[Scheme 48]] implementation of Scheme using bytecode interpreter
*Bytecodes of many implementations of the [[Smalltalk]] language
*The [[Parallax Propeller#Built in Spin bytecode interpreter|Spin interpreter]] built into the [[Parallax, Inc. (company)|Parallax]] Propeller [[microcontroller]]
*The [[SQLite]] database engine translates SQL statements into a bespoke byte-code format.<ref name="SQLite"/>
*Apple [[SWEET16]]
*[[Tcl]]
*[[IBM i#TIMI|TIMI]] is used by compilers on the [[IBM i]] platform.
*[[Tiny BASIC#Implementation in a virtual machine|Tiny BASIC]]
*[[Visual Basic for Applications]] compiles to bytecode.
*[[Visual FoxPro]] compiles to bytecode
*[[WebAssembly]]
*[[YARV]] and [[Rubinius]] for [[Ruby (programming language)|Ruby]]
*[[ZCODE]]
*[[Zend Engine]] opcodes for [[PHP]]

==See also==
{{wiktionary|bytecode}}
* {{Annotated link |Computing platform}}
* {{Annotated link |Intermediate representation}}
* {{Annotated link |Runtime system}}

==Notes==
{{notelist}}

==References==
<references>
<ref name="Jucs_Lua">{{cite web |url=http://www.jucs.org/jucs_11_7/the_implementation_of_lua/jucs_11_7_1159_1176_defigueiredo.html |title=The Implementation of Lua 5.0}} (NB. This involves a register-based virtual machine.)</ref>
<ref name="Dalvik">{{Cite web |url=http://source.android.com/tech/dalvik/dalvik-bytecode.html |title=Dalvik VM |url-status=dead |archive-url=https://web.archive.org/web/20130518021154/http://source.android.com/tech/dalvik/dalvik-bytecode.html |archive-date=2013-05-18 |access-date=2012-10-29}} (NB. This VM is register based.)</ref>
<ref name="Byte_Machine">{{cite web |title=Byte Code Vs Machine Code |website=www.allaboutcomputing.net |url=http://www.allaboutcomputing.net/2014/07/byte-code-vs-machine-code.html |access-date=2017-10-23}}</ref>
<ref name="Dynamic_Machine_Code">{{cite web |title=Dynamic Machine Code Generation |publisher=Google Inc. |url=https://chromium.googlesource.com/external/github.com/v8/v8.wiki/+/eb80fb157da30e8c838e758f178de674e47648ed/Design-Elements.md |access-date=2024-12-01  |archive-date=2017-03-05  |archive-url=https://web.archive.org/web/20170305155607/https://github.com/v8/v8/wiki/Design%20Elements |url-status=live }}</ref>
<ref name="Loitsch_Bytecode">{{Cite web |url=http://www.dartlang.org/articles/why-not-bytecode/ |title=Why Not a Bytecode VM? |last=Loitsch |first=Florian |publisher=[[Google]] |url-status=dead |archive-url=https://web.archive.org/web/20130512215811/http://www.dartlang.org/articles/why-not-bytecode/ |archive-date=2013-05-12}}</ref>
<ref name="Javascript">{{Cite web|url=https://2ality.com/2012/01/bytecode-myth.html|title=JavaScript myth: JavaScript needs a standard bytecode|website=2ality.com}}</ref>
<ref name="Arizona_Icom">{{Cite web |url=http://www.cs.arizona.edu/icon/ftp/doc/ib1up.pdf |title=The Implementation of the Icon Programming Language |url-status=dead |archive-url=https://web.archive.org/web/20160305123148/http://www.cs.arizona.edu/icon/ftp/doc/ib1up.pdf |archive-date=5 March 2016 |access-date=9 September 2011}}</ref>
<ref name="Icon_Unicon">{{Cite web|url=https://unicon.sourceforge.net/book/ib.pdf |archive-url=https://ghostarchive.org/archive/20221009/http://unicon.sourceforge.net/book/ib.pdf |archive-date=2022-10-09 |url-status=live|title=The Implementation of Icon and Unicon a Compendium}}</ref>
<ref name="Paul_2001_KEYBOARD">{{cite newsgroup |title=KEYBOARD.SYS internal structure |newsgroup=comp.os.msdos.programmer |author-first=Matthias R. |author-last=Paul |date=2001-12-30 |url=https://groups.google.com/d/msg/comp.os.msdos.programmer/l_IuSHsBDWQ/887rJF9IYmMJ |access-date=2016-09-17 |url-status=live |archive-url=https://archive.today/20170909082257/https://groups.google.com/forum/%23!msg/comp.os.msdos.programmer/l_IuSHsBDWQ/887rJF9IYmMJ |archive-date=2017-09-09 |quote=[…] In fact, the format is basically the same in [[MS-DOS]] 3.3&nbsp;- 8.0, [[PC&nbsp;DOS]] 3.3&nbsp;- 2000, including Russian, Lithuanian, Chinese and Japanese issues, as well as in Windows NT, 2000, and XP […]. There are minor differences and incompatibilities, but the general format has not changed over the years. […] Some of the data entries contain normal tables […] However, most entries contain ''executable code'' interpreted by some kind of [[p-code machine|p-code interpreter]] at *[[run time (program lifecycle phase)|runtime]]*, including conditional branches and the like. This is why the [[KEYB (DOS command)|KEYB]] driver has such a huge memory footprint compared to table-driven keyboard drivers which can be done in 3 - 4 Kb getting the same level of function except for the interpreter. […]}}</ref>
<ref name="Mendelson_2001_KEYBOARD">{{Cite web |url=http://www.columbia.edu/~em36/wpdos/eurodos.html |title=How to Display the Euro in MS-DOS and Windows DOS |last=Mendelson |first=Edward |author-link=Edward Mendelson |date=2001-07-20 |at=Display the euro symbol in full-screen MS-DOS (including Windows 95 or Windows 98 full-screen DOS) |url-status=live |archive-url=https://web.archive.org/web/20160917201248/http://www.columbia.edu/~em36/wpdos/eurodos.html |archive-date=2016-09-17 |access-date=2016-09-17 |quote=[…] Matthias [R.] Paul […] warns that the [[IBM PC DOS]] version of the keyboard driver uses some internal procedures that are not recognized by the [[Microsoft]] driver, so, if possible, you should use the [[IBM]] versions of both [[KEYB.COM]] and [[KEYBOARD.SYS]] instead of mixing Microsoft and IBM versions […]}} (NB. What is meant by "procedures" here are some additional bytecodes in the IBM KEYBOARD.SYS file not supported by the Microsoft version of the KEYB driver.)</ref>
<ref name="Patent_6973644">{{cite web |title=United States Patent 6,973,644 |url=http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO1&Sect2=HITOFF&d=PALL&p=1&u=%2Fnetahtml%2FPTO%2Fsrchnum.htm&r=1&f=G&l=50&s1=6973644.PN.&OS=PN/6973644&RS=PN/6973644 |access-date=2009-05-21  |archive-date=2017-03-05  |archive-url=https://web.archive.org/web/20170305001731/http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO1&Sect2=HITOFF&d=PALL&p=1&u=%2Fnetahtml%2FPTO%2Fsrchnum.htm&r=1&f=G&l=50&s1=6973644.PN.&OS=PN/6973644&RS=PN/6973644 |url-status=dead }}</ref>
<ref name="cran_r">{{Cite web|url=https://cran.r-project.org/doc/manuals/R-admin.html#Byte_002dcompiler|title=R Installation and Administration|website=cran.r-project.org}}</ref>
<ref name="SQLite">{{cite web |title=The SQLite Bytecode Engine |url=https://www.sqlite.org/opcode.html |access-date=29 August 2016 |archive-url=https://web.archive.org/web/20170414044139/http://sqlite.org/opcode.html |archive-date=14 April 2017 |url-status=dead }}</ref>
<ref name="Multiplan">{{cite book |title=Microsoft C Pcode Specifications |page=13 |quote=[[Multiplan]] wasn't compiled to [[machine code]], but to a kind of byte-code which was run by an [[interpreter (computing)|interpreter]], in order to make Multiplan portable across the widely varying hardware of the time. This byte-code distinguished between the machine-specific [[floating point format]] to calculate on, and an external (standard) format, which was [[binary-coded decimal|binary coded decimal]] (BCD). The PACK and UNPACK instructions converted between the two.}}</ref>
</references>

[[Category:Bytecodes| ]]
[[Category:Virtualization]]