I am targeting an STM32F767ZI which is an M7 arm core. The hardware FPU flags are incorrect. They currently are “-mcpu=cortex-m7 -mthumb -mfpu=fpv4-sp-d16 -mfloat-abi=hard” but the FPU flag should be “-mfpu=fpv5-d16” to properly enable single and double precision hardware floating point.
Thanks for reporting this. We have just released an updated version of the STM32 BSP that defines the FPU for all the devices correctly. You can install it via Tools->VisualGDB->Manage VisualGDB Packages.
If you still need help overriding the settings, please let us know.