10.8 Conditional execution of A32/T32 floating-point instructions

You can execute floating-point instructions conditionally, in the same way as most A32 and T32 instructions.

You cannot use any of the following floating-point instructions in an IT block:

  • VRINT{A, N, P, M}.
  • VSEL.
  • VCVT{A, N, P, M}.
  • VMAXNM.
  • VMINNM.

In addition, specifying any other floating-point instruction in an IT block is deprecated.

Most A32 floating-point instructions can be conditionally executed, by appending a condition code suffix to the instruction.

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