ARM reveals details of 64-bit ARMv8 architecture

On October 27, ARM disclosed technical details of a 64-bit ARMv8 architecture. ARMv8 expands on the 32-bit ARMv7 architecture, used as the base of some market leading cores such as Cortex-A9 and Cortex-A15, by adding a 64-bit instruction set. The ARM 32-bit architecture is used in a wide range of electronic devices, from tiny sensors to large scale equipment, making it unique. Building on this existing architecture, ARMv8 will widen the range of applications for ARM, making it suitable for many consumer and enterprise applications, with 64-bit data processing and virtual addressing adding increased functionality.

There are two main executable states in ARMv8, known as AArch64 and AArch32. AArch32 support current ARMv7 instructions, while AArch64 adds the new A64 instruction set, for 64-bit processing. ARMv8 will either maintain or extend all the key features of ARMv7, such as Trustzone, virtualization and NEON advanced SIMD. This ensures the platform will be backward compatible with applications written for ARMv7, as well as meaning new applications can be written that work on both platforms.

Mike Muller, CTO at ARM, said "With our increasingly connected world, the market for 32-bit processing continues to expand and evolve creating new opportunities for 32-bit ARMv7 based processors in embedded, real-time and open application platforms. We believe the ARMv8 architecture is ideally suited to enable the ARM partnership to continue to grow in 32-bit application spaces and bring diverse, innovative and energy-efficient solutions to 64-bit processing markets."

ARM have already released the ARM compiler and Fast Modules to key partners, so they can ensure a robust environment supporting the 64-bit instruction set. The ARM partnership are working together to ensure the availability of, and support for, a range of open-source operating systems, applications and third party tools. Much of this development is a natural extension to the broad based support for ARMv7 based devices currently available.

"ARM is an important partner for Microsoft," said KD Hallman, general manager, Microsoft Corp. "The evolution of ARM to support a 64-bit architecture is a significant development for ARM and for the ARM ecosystem. We look forward to witnessing this technology's potential to enhance future ARM-based solutions."

"The combination of NVIDIA’s leadership in energy-efficient, high-performance processing and the new ARMv8 architecture will enable game-shifting breakthroughs in devices across the full range of computing – from smartphones through to supercomputers," said Dan Vivoli, senior vice president, NVIDIA.

"The current growth trajectory of data centers, driven by the viral explosion of social media and cloud computing, will continue to accelerate. The ability to handle this data increase with energy-efficient solutions is vital," said Vinay Ravuri, vice president and general manager of AppliedMicro’s Processor Business Unit. "The ARM 64-bit architecture provides the right balance of performance, efficiency and cost to scale to meet these growing demands and we are very excited to be a leading partner in implementing solutions based on the ARMv8 architecture."

ARM compatible devices can be designed to maximize the benefits of both AArch32 and AArch64, with the help of the ARMv8 architecture design. This means that high-end servers and computers will gain advantages from energy-efficient 64-bit computing, while maintaining backward compatibility and migration opportunities for existing software.

ARM will disclose details of processors based on ARMv8 during 2012, and expect prototype systems for both consumer and enterprise markets in 2014.

Primarily motivated by a move toward 64-bit applications, development of the architecture began in 2007. This will provide the ability to access a much larger virtual memory space, as well as enabling the expansion of ARM's current market presence, where a future need is foreseen. ARM are already seeing a lot of interest from partners.