You copied the Doc URL to your clipboard.

--cpu

This option enables the linker to determine the target ARM processor or architecture.

Note

This release of ARM Compiler is targeted at ARMv8 and ARMv7-A architectures and processors that implement these architectures.

Syntax

--cpu=list
--cpu=name

Where:

name

is the name of a processor or architecture. Table 1 lists the processor and architecture names that are supported.

Processor and architecture names are not case-sensitive in armlink.

Wildcard characters are not accepted.

Table 1. Supported ARM processors and architectures
Processor and architecture nameDescription
7ARMv7 with Thumb (Thumb-2 technology) only, and without hardware divide
7-AARMv7 application profile supporting virtual MMU-based memory systems, with ARM, Thumb (Thumb-2 technology) and ThumbEE, DSP support, and 32-bit Advanced SIMD support
7-A.securityEnables the use of the SMC instruction (formerly SMI) when assembling for the ARMv7-A architecture
8-A.32ARMv8, AArch32 state
8-A.32.cryptoARMv8, AArch32 state with cryptographic instructions
8-A.32.no_neonARMv8, AArch32 state without Advanced SIMD instructions
8-A.64ARMv8, AArch64 state
8-A.64.cryptoARMv8, AArch64 state with cryptographic instructions
8-A.64.no_neonARMv8, AArch64 state without Advanced SIMD instructions
Cortex-A5Cortex-A5 processor
Cortex-A5.vfpCortex-A5 processor with floating point instructions
Cortex-A5.neonCortex-A5 processor with Advanced SIMD instructions
Cortex-A7Cortex-A7 processor
Cortex-A7.no_neonCortex-A7 processor with floating point instructions
Cortex-A7.no_neon.no_vfpCortex-A7 processor without Advanced SIMD instructions and without floating point instructions
Cortex-A8Cortex-A8 processor
Cortex-A8.no_neonCortex-A8 processor without Advanced SIMD instructions
Cortex-A9Cortex-A9 processor
Cortex-A9.no_neonCortex-A9 processor with floating point instructions
Cortex-A9.no_neon.no_vfpCortex-A9 processor without Advanced SIMD instructions and without floating point instructions
Cortex-A12Cortex-A12 processor
Cortex-A12.no_neon.no_vfpCortex-A12 processor without Advanced SIMD instructions and without floating point instructions
Cortex-A15Cortex-A15 processor
Cortex-A15.no_neonCortex-A15 processor with floating point instructions
Cortex-A15.no_neon.no_vfpCortex-A15 processor without Advanced SIMD instructions and without floating point instructions
Cortex-A17Cortex-A17 processor
Cortex-A17.no_neon.no_vfpCortex-A17 processor without Advanced SIMD instructions and without floating point instructions

Usage

If you omit --cpu, the linker auto-detects the processor or architecture from the input object files.

Specify --cpu=list to list the supported processor and architecture names that you can use with --cpu=name.

The link phase fails if any of the component object files rely on features that are incompatible with the specified processor. The linker also uses this option to optimize the choice of system libraries and any veneers that have to be generated when building the final image.

See also

Reference