10.11 Extended notation extension for floating-point in A32/T32 code

armasm implements an extension to the architectural floating-point assembly syntax, called extended notation. This extension allows you to include datatype information or scalar indexes in register names.

Note:

Extended notation is not supported for A64 code.

If you use extended notation, you do not have to include the data type or scalar index information in every instruction.

Register names can be any of the following:

Untyped
The register name specifies the register, but not what datatype it contains, nor any index to a particular scalar within the register.
Untyped with scalar index
The register name specifies the register, but not what datatype it contains, It specifies an index to a particular scalar within the register.
Typed
The register name specifies the register, and what datatype it contains, but not any index to a particular scalar within the register.
Typed with scalar index
The register name specifies the register, what datatype it contains, and an index to a particular scalar within the register.

Use the SN and DN directives to define names for typed and scalar registers.

Non-ConfidentialPDF file icon PDF versionDUI0801J
Copyright © 2014–2017, 2019 Arm Limited or its affiliates. All rights reserved.