Pc Video Games

click here and donwload

Dalvik is that the method virtual machine (VM) in Google's robot software package, which, specifically, executes applications written for robot. This makes Dalvik AN integral a part of the robot software system stack, that is often used on mobile devices like mobile phones and pill computers, likewise as a lot of recently on devices like sensible TVs and wearables.

Programs square measure unremarkably written in Java and compiled to bytecode for the Java virtual machine, that is then translated to Dalvik bytecode and keep in .dex (Dalvik EXecutable) and .odex (Optimized Dalvik EXecutable) files; connected terms odex and de-odex square measure related to various bytecode conversions. The compact Dalvik possible format is intended for systems that square measure forced in terms of memory and processor speed.

Dalvik is ASCII text file software system. it had been originally written by Dan Bornstein, World Health Organization named it when the fishing village of Dalvík in Eyjafjörður, Iceland.[1][2]

An alternative runtime surroundings referred to as robot Runtime (ART) was enclosed in robot four.4 "KitKat" as a technology preview.[3][4] ART replaces Dalvik entirely in robot five.0 "Lollipop".[5]

Contents  [hide] 

1 design

2 Performance

3 Licensing and patents

4 See also

5 References

6 External links

Architecture[edit]

A comparison of Dalvik and ART architectures

Unlike Java VMs, that square measure stack machines, the Dalvik VM uses a register-based design that needs fewer, generally a lot of advanced virtual machine directions. Dalvik programs square measure written in Java exploitation the robot application programming interface (API), compiled to Java bytecode, and reborn to Dalvik directions as necessary.

A tool referred to as dx is employed to convert Java .class files into the .dex format. Multiple categories square measure enclosed in a very single .dex file. Duplicate strings and alternative constants employed in multiple category files square measure enclosed just once within the .dex output to conserve area. Java bytecode is additionally reborn into another instruction set utilized by the Dalvik VM. AN uncompressed .dex file is often a number of % smaller in size than a compressed Java archive (JAR) derived from a similar .class files.[6]

The Dalvik executables could also be changed once more once put in onto a mobile device. so as to achieve additional optimizations, computer memory unit order could also be swapped in sure knowledge, straightforward knowledge structures and performance libraries could also be joined inline, and empty category objects could also be short-circuited, for instance.

As of robot a pair of.2, Dalvik encompasses a just-in-time compiler.[7]

Being optimized for low memory necessities, Dalvik has some specific characteristics that differentiate it from alternative customary VMs:[8]

The VM was slimmed right down to use less area.

The constant pool has been changed to use solely 32-bit indices to modify the interpreter.

Standard Java bytecode executes 8-bit stack directions. native variables should be derived to or from the quantity stack by separate directions. Dalvik instead uses its own 16-bit instruction set that works directly on native variables. The native variable is usually picked by a 4-bit 'virtual register' field. This lowers Dalvik's instruction count and raises its interpreter speed.

According to Google, the planning of Dalvik permits a tool to run multiple instances of the VM with efficiency.[9]

Performance[edit]

The relative deserves of stack machines versus register-based approaches square measure an issue of in progress discussion.[10]

Generally, stack-based machines should use directions to load knowledge on the stack and manipulate that knowledge, and, thus, need a lot of directions than register machines to implement a similar high level code, however the directions in a very register machine should cypher the supply and destination registers and, therefore, tend to be larger. This distinction is primarily of importance to VM interpreters that opcode dispatch tends to be expensive  beside alternative factors equally relevant to just-in-time compilation.

However, tests performed on ARM devices by Oracle (owner of the Java technology) in 2010 with customary non-graphical Java benchmarks on Java SE embedded appeared to show it to be quicker, by a pair of to three times, than robot a pair of.2 (the initial unleash to incorporate a just-in-time compiler) that they compared to (both supported Java SE 6).[11][dated info]

In 2012, tutorial benchmarks confirmed the issue of three between HotSpot and Dalvik on a similar robot board, additionally noting that Dalvik code wasn't smaller than Hotspot.[12]

Furthermore, benchmarks performed on robot device still show (as of March 2014) up to an element one hundred between native applications and a Dalvik application on a similar robot device.[13][original research?