What is the advantage and disadvantage of assembly language?

Answer 1

Assembly language is a symbolic representation of a processor's native code. Using machine code allows the programmer to control precisely what the processor does. It offers a great deal of power to use all of the features of the processor. The resulting program is normally very fast and very compact. In small programs it is also very predictable. Timings, for example, can be calculated very precisely and program flow is easily controlled. It is often used for small, real time applications. However, the programmer needs to have a good understanding of the hardware being used. As programs become larger, assembly language get very cumbersome. Maintenance of assembly language is notoriously difficult, especially if another programmer is brought in to carry out modifications after the code has been written. Assembly langauge also has no support of an operating system, nor does it have any complex instructions. Storing and retrieving data is a simple task with high level languages; assembly needs the whole process to be programmed step by step. Mathmatical processes also have to be performed with binary addition and subtraction when using assembly which can get very complex. Finally, every processor has its own assembly language. Use a new processor and you need to learn a new language each time. Assembly is a great language to use for certain applications, rotten for others and never for the faint hearted.

Answer 2

The advantages of coding in assembly is far less today than before. High level programming languages are constantly being improved and given features that make assembler less relevant.

Assembler is used primarily in three circumstances.

1) to perform low level operations which can not be done in a high level language. This for example includes the code responsible for the first few instructions needed to start a computer up and load code written in a high level language. In addition, it might be used for making calls between operating modes of the computer's processor where strict adherence to a structure is required.

2) it may be used in areas where there is extreme hardware constraints. For example, a PIC microcontroller might only have a few kilobytes (or even less) of memory to store a program. Therefore, manually choosing each specific processor operation can make the difference between whether code will fit in the CPU or not.

3) finally, and less relevant than ever. Hand coding assembler may in some cases allow faster code to be written than if it were coded in a high level language. In modern times, it is used most heavily to implement "instruction level parallelism" such as making use of SSE, AVX, ARM NEON, etc... instructions. But with the advent of languages like OpenCL which included special types for this specific task, it will become less relevant over time. Popular projects such as the x264 video encoder make heavy use of assembler to squeeze every little extra bit of performance out of the CPU that it can. In many cases, x264s bottleneck now lays in the speed which the CPU can access memory than in the CPU's ability to perform calculations.

On the final point, it's also worth mentioning that with "out of order" execution, the instructions fed to the processor becomes less important since it is not always possible to predict how the code itself will be executed as it once was. This is because different processor cores will perform a step similar to recompiling code differently than each other before executing it.

Therefore, in most cases, only the first two reasons make sense in a modern context. Some might say that any new projects coded with assembler for the sake of performance would likely take longer to write than to wait for a more optimal implementation of a compiler which provides instruction level parallelization to arrive.

Answer 3

The main advantage: it can directly communicate with hardware. The main disadvantage: it's very hard to write in terms of machine instructions.