ARM has announced an ultra-efficient Cortex-A32 processor, targeting next-generation embedded products. The Cortex-A32 processor brings the benefits of the ARMv8-A architecture into power-constrained, 32-bit embedded applications. The Cortex-A5 and Cortex-A7 processors, which are based on the ARMv7-A architecture, are among the most widely adopted application processors, powering embedded applications.

The Cortex-A32 is based on an 8-stage in-order pipeline that has been optimized to implement the 32-bit instruction set of the ARMv8-A architecture profile in the smallest possible die area while significantly reducing dynamic power consumption compared to the currently available the Cortex-A7 processor. According to ARM, the Cortex-A32 delivers 25% more efficiency (performance per mW) than the Cortex-A7 in the same process node. In its smallest configuration, the Cortex-A32 occupies less than 0.25 mm2 of silicon area while consuming less than 4mW of total power at 100 MHz in a 28nm process node.

The ARMv8-A architecture supports both 32-bit and 64-bit compute capabilities in the AArch32 and AArch64 execution states respectively. All previously released ARMv8-A processors have supported both execution states. The Cortex-A32 implements only the AArch32 execution state. This state maintains compatibility with existing ARMv7-A software while adding more than 100 new instructions to the A32/T32 instruction set. These additional instructions mean that embedded applications using Cortex-A32 will benefit from improvements in floating point performance, faster software encryption and enhanced media performance.

Reducing the idle power consumption is a key requirement for many embedded applications. The Cortex-A32 processor incorporates new power management features compared to Cortex-A7 and Cortex-A5 processors.

The Cortex-A32 can be configured in multiple ways from single- to quad-core. It includes ARM TrustZone technology to provide a secure foundation in the SoC hardware. TrustZone is a security technology, providing banking-class trust capability in devices such as smartphones. While it includes cryptographic instructions for efficient authentication and protection, the Cortex-A32 can also be coupled with TrustZone CryptoCell-700 series products to enable enhanced cryptographic hardware acceleration and advanced root of trust.

The first devices based the chip could come out late this year or early next. It will run Android Wear as well as other operating systems. Another target could be single-board computers in the vein of the Raspberry Pi 2.