Cannot link mpllib when using hardware floating point

I'm using the 6.1 mpl library in combination with a LPC-4078 (Cortex M4) and I would like to get the benefits from the hardware floating point unit.

However, when I link the library with HWFP option on I get the following error:

RDB4078.axf uses VFP register arguments, libmpllib.a(compass_vec_cal.o) does not..(and so on).

When I read the readme.txt included with the library it says it is compiled with HWFP option:

M4 Compiler Options -

-Os -ffunction-sections -fdata-sections -Wall -Wa,-adhlns="$@.lst" -c -fmessage-length=0 -mcpu=cortex-m4 -mthumb -mfloat-abi=hard -mfpu=vfpv4 -g3 -gdwarf-2

Anyone knows a solution? I would like to boost my program.

Thanks in advance