Instructions do not work with custom floatt

Altera_Forum · ‎12-15-2009

Hi Everybody!!!

I created a custom instruction and it is not working for floating point. It worked only with integer value. I've used the macro that was defined in the Altera Manual:

# define ALT_CI_TESTE_INST_N 0x00000000

# define ALT_CI_TESTE_INST (A) __builtin_custom_fnf(ALT_CI_TESTE_INST_N, (A))

By the manual should be working.

Excerpt from my test code

# include <stdio.h>

# include "system.HA"

int main ()

(

float x = ALT_CI_TESTE_INST (0.0);

printf ( "Value:% f", x);

return 0;

)

The following appears on the console:

-----------------

Value:

-----------------

If anyone can help me, please respond. I am Brazilian and I do not know English very well, then no doubt, explain again.

Thank you!!!!

Altera_Forum · ‎01-05-2010

Try this instead:

float x = ALT_CI_TESTE_INST(0.0f);

The 'f' tells the compiler to use a single precision value instead of double precision (default). There is also a flag you can pass to the compiler to tell it to treat all floating point constants as type 'float' instead of 'double' but I forget what to use (I don't recommend it either since you might accidentally break some code that needed to use double precision constants).