Re: [PATCH 04/29] Allow rusage work on a big-endian 32bit-on-64bit target

[Mike Frysinger <vapier at gentoo dot org>]

On 27 Oct 2014 00:59, Andrew Pinski wrote:
> +#ifndef __RUSAGE_LONG
> +/* This definition works where __syscall_slong_t is the same as 'long int'
> +   and on little-endian when __syscall_slong_t is not 'long int' like x32. */

two spaces after the period

> +# define __RUSAGE_LONG(__field) 		\

there's a bad space after the ) and before the tabs

> +    __extension__ union				\
> +      {						\
> +	long int __field;			\
> +	__syscall_slong_t __##__field##_word;	\
> +      }

couldn't we also handle big endian here ?  we check __BYTE_ORDER in many common 
headers, so seems like we should be able to also provide a sensible default 
here.

otherwise, i like the clean up in this file.
-mike

Attachment: signature.asc
Description: Digital signature