C3.5 Linker
reordering of execution regions containing T32 code
The linker reorders execution regions containing T32 code only if the size of the T32 code exceeds the branch range.
If the code size of an execution region exceeds the maximum branch range of a T32
instruction, then armlink reorders the input sections using a different
sorting algorithm. This sorting algorithm attempts to minimize the amount of veneers
generated.
The T32 branch instructions that can be veneered are always encoded as a pair of 16-bit
instructions. Processors that support Thumb-2 technology have a range of 16MB. Processors
that do not support Thumb-2 technology have a range of 4MB.
To disable section reordering, use the --no_largeregions command-line
option.
Arm’s Privacy Policy has been updated. By continuing to use our site, you consent to Arm’s Privacy Policy. Please review our Privacy Policy to learn more about our collection, use and transfers of your data.