[coldfire-gnu-discuss] floating point instructions enabled by default for gcc-4.3-43 on mcf5472 processor?

[Wong, Ken]

Thanks for advice Maxim, our builds are currently omitting any type of
-mcpu flag so that has me wondering even more which default flags I'm
compiling with. :)

I'll setup the mcpu flag correctly today and see if the generating
objects differ significantly from my usual build.

If anyone else has any advice about how to figure out what the default
flags are for the toolchain, I'm all ears.

Thanks,

Ken.

-----Original Message-----
From: Maxim Kuvyrkov [mailto:maxim at codesourcery.com] 
Sent: Friday, February 13, 2009 2:29 AM
To: Wong, Ken
Cc: coldfire-gnu-discuss at codesourcery.com
Subject: Re: [coldfire-gnu-discuss] floating point instructions enabled
by default for gcc-4.3-43 on mcf5472 processor?

Wong, Ken wrote:
> I was wondering if anyone could tell me an easy way of determining 
> whether or not code generated using the gcc-4.3-43 toolchain is 
> generating hardware floating point instructions...
> 
>  
> 
> I'm pretty sure it is, and I think I can go through the objdump of my 
> code but being able to draw up a list of defaults that the compiler is

> using would be great.

It depends on -mcpu= setting: if selected CPU has FPU unit (which, I 
believe, at the moment are only devices from 547x/548x family) then FP 
instructions can be generated.

There is also option -msoft-float that unconditionally turns off use of 
FP instructions.

--
Maxim