SPARC Assembly/SPARC Architecture

From Wikibooks, open books for an open world
Jump to navigation Jump to search

SPARC History[edit | edit source]

A Sun 4 SPARCstation.

The first version of the architecture, SPARCv7 (for Scalable Processor Architecture, version 7), was originally developed by Sun Microsystems in 1986 ; the first implementation became available in 1987. In 1989, SPARC International, an independent, non-profit organization, was created to promote the SPARC and provide conformance testing. Implementations of the original 32-bit SPARC architecture were initially designed and used for Sun's Sun-4 workstation and server systems, replacing their earlier Sun-3 systems based on the Motorola 68000 family of processors. Later, SPARC processors were used in SMP servers produced by Sun Microsystems, Solbourne and Fujitsu, among others.

A new version, SPARCv8, was released in 1990. It added integer multiply and divide instructions, which weren't previously available, and quad-precision floating point numbers, encoded on 128 bits.

The specification of the first 64-bit version of the SPARC, SPARCv9, was published in 1994, and Sun and Fujitsu both started shipping SPARCv9 processors at the end of 1995.

SPARC International (http://sparc.org) was formed in 1989 to open up the SPARC architecture to make a larger ecosystem for the design. The architecture is freely licensable for a nominal fee[citation needed]. SPARC processors have been designed and produced by several manufacturers, including Texas Instruments, Atmel, Cypress Semiconductor, and Fujitsu. As a result of SPARC International, the SPARC architecture is fully open and non-proprietary.

SPARC Architecture[edit | edit source]

The SPARC is of the Reduced Instruction Set Computing (RISC) architecture. The theory is that having the bare minimum of instructions needed to complete a job, the resulting architecture is faster, as most instructions take only one clock cycle to decode, leading to rapid execution. This is in contrast to CISC machines, which have specialized variable length instructions, and can take multiple clock cycles to decode and execute.

The SPARC architecture also prefetches instructions, having the next instruction fetched while the current one is executed. This has implication for branch instructions, as the next instruction might not be executed if the branch is taken, and this must be dealt with accordingly.

Why SPARC?[edit | edit source]

Being a RISC architecture, SPARC assembly is very clean and simple, while still being powerful. This makes it easy to learn, and a good choice before learning more complicated assembly languages if needed. Furthermore, it provides a closer look of what is going on "under the hood" of higher-level programming languages, and at the very least should foster an appreciation of the large amount of work being done by compilers when compiling such languages. If you don't have a SPARC computer, you can use a SPARC emulator.