Re: [PATCH 1/8] Use __kernel_long_t in struct timex

[Catalin Marinas <catalin.marinas@arm.com>]

On Fri, Dec 27, 2013 at 05:25:05PM +0000, H.J. Lu wrote:
> X32 adjtimex system call is the same as x86-64 adjtimex system call,
> which uses 64-bit integer for long in struct timex. But x32 long is
> 32 bit.  This patch replaces long in struct timex with __kernel_long_t
> if __BITS_PER_LONG == 64.
> 
> Signed-off-by: H.J. Lu <hjl.tools@gmail.com>
> ---
>  include/uapi/linux/timex.h | 46 ++++++++++++++++++++++++++++++++++++++--------
>  1 file changed, 38 insertions(+), 8 deletions(-)
> 
> diff --git a/include/uapi/linux/timex.h b/include/uapi/linux/timex.h
> index a7ea81f..98314e9 100644
> --- a/include/uapi/linux/timex.h
> +++ b/include/uapi/linux/timex.h
> @@ -63,28 +63,58 @@
>   */
>  struct timex {
>  	unsigned int modes;	/* mode selector */
> +#if __BITS_PER_LONG == 64
> +	__kernel_long_t offset;	/* time offset (usec) */
> +	__kernel_long_t freq;	/* frequency offset (scaled ppm) */
> +	__kernel_long_t maxerror;/* maximum error (usec) */
> +	__kernel_long_t esterror;/* estimated error (usec) */
> +#else
>  	long offset;		/* time offset (usec) */
>  	long freq;		/* frequency offset (scaled ppm) */
>  	long maxerror;		/* maximum error (usec) */
>  	long esterror;		/* estimated error (usec) */
> +#endif

These changes should work on arm64 ILP32 as well.

>  	int status;		/* clock command/status */
> +#if __BITS_PER_LONG == 64
> +	__kernel_long_t constant;/* pll time constant */
> +	__kernel_long_t precision;/* clock precision (usec) (read only) */
> +	__kernel_long_t tolerance;/* clock frequency tolerance (ppm)
> +				   * (read only)
> +				   */
> +#else
>  	long constant;		/* pll time constant */
>  	long precision;		/* clock precision (usec) (read only) */
>  	long tolerance;		/* clock frequency tolerance (ppm)
>  				 * (read only)
>  				 */
> +#endif
>  	struct timeval time;	/* (read only, except for ADJ_SETOFFSET) */

struct timeval is already defined in terms of __kernel_long_t, so no
issues here either.

BTW, could we not avoid the #if and always use __kernel_long_t? This
wouldn't break the user ABI.

-- 
Catalin