Re: Strange variable declarations in Kernel

[Adrian Bunk <bunk@xxxxxxxxx>]

On Fri, Sep 02, 2005 at 05:25:06PM +0200, Paul Duplys wrote:
> Hi all, 
> 
> in the file "efi.h" located in /fs/partitions in the kernel tree I found
> this structure definition:
> 
> 87.  typedef struct _gpt_entry_attributes {
> 88.    u64 required_to_function:1;
> 89.    u64 reserved:47;
> 90.    u64 type_guid_specific:16;
> 91.  } __attribute__ ((packed)) gpt_entry_attributes;
> 
> 
> Unfortunatelly, my compiler (perhaps?) can't handle this:
>  
>  In file included from check.c:36:
>  efi.h:91: Illegal bitfield width
>  make[4]: *** [check.o] Error 1
> 
> I've never seen a variable definition like this before (and google doesn't
> work...):
> 
> required_to_function:1;
> 
> What does the collon after the variable name mean? What does the number
> after the collon mean? 
> 
> In the same file there are few other typedefs of the same form and they use
> the datatype u64 (which is defined as unsigned long long in asm/types.h) and
> the attribute packed too. But with them there are no compile problems. So I
> guess, the problem is the collon and the number after the collon.

As your compiler is telling you they are bitfields.

What compiler are you using that has a problem with this code?

> Regards and thanks in advance,
> Paul

cu
Adrian

-- 

       "Is there not promise of rain?" Ling Tan asked suddenly out
        of the darkness. There had been need of rain for many days.
       "Only a promise," Lao Er said.
                                       Pearl S. Buck - Dragon Seed


--
Kernelnewbies: Help each other learn about the Linux kernel.
Archive:       http://mail.nl.linux.org/kernelnewbies/
FAQ:           http://kernelnewbies.org/faq/