Re: [PATCH] cpufreq: make table sentinal macros unsigned to match use

[Simon Horman <horms@verge.net.au>]

On Fri, Jun 27, 2014 at 04:09:39PM -0500, Brian W Hart wrote:
> Commit 5eeaf1f18973 (cpufreq: Fix build error on some platforms that
> use cpufreq_for_each_*) moved function cpufreq_next_valid() to a public
> header.  Warnings are now generated when objects including that header
> are built with -Wsign-compare (as an out-of-tree module might be):
> 
> .../include/linux/cpufreq.h: In function ‘cpufreq_next_valid’:
> .../include/linux/cpufreq.h:519:27: warning: comparison between signed
> and unsigned integer expressions [-Wsign-compare]
>   while ((*pos)->frequency != CPUFREQ_TABLE_END)
>                            ^
> .../include/linux/cpufreq.h:520:25: warning: comparison between signed
> and unsigned integer expressions [-Wsign-compare]
>    if ((*pos)->frequency != CPUFREQ_ENTRY_INVALID)
>                          ^
> 
> Constants CPUFREQ_ENTRY_INVALID and CPUFREQ_TABLE_END are signed, but
> are used with unsigned member 'frequency' of cpufreq_frequency_table.
> Update the macro definitions to be explicitly unsigned to match their
> use.
> 
> This also corrects potentially wrong behavior of clk_rate_table_iter()
> if unsigned long is wider than usigned int.
> 
> Signed-off-by: Brian W Hart <hartb@linux.vnet.ibm.com>

Reviewed-by: Simon Horman <horms+renesas@verge.net.au>

> ---
> These macros are fairly broadly used in the kernel so I was bit leery
> of changing them, but after inspection I think it's fine.  I found 102
> uses of the macros, of which:
> 
> 99 are uses with cpufreq_frequency_table.frequency (95) or with local
>    variables of the same type as frequency (4).  These should be just
>    fine with this change--we're just making explicit a conversion that
>    was previously implicit.
> 
>  2 are uses with a local variable of different type (unsigned long) than
>    'frequency' (in drivers/sh/clk/core.c).  One of these uses is safe;
>    the other (in clk_rate_table_iter()) is broken if unsigned long
>    is wider than unsigned int.  As a side-effect, this patch corrects
>    the potential misbehavior there.
> 
>  1 is a use in macro cpufreq_for_each_entry() with what _should_ be the
>    frequency member of a cpufreq_frequency_table, provided the caller it
>    well-behaved.  There are 18 callers of this macro; all are well-behaved.
>    So these should also be safe.
> 
>  include/linux/cpufreq.h |    4 ++--
>  1 files changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/include/linux/cpufreq.h b/include/linux/cpufreq.h
> index ec4112d..8f8ae95 100644
> --- a/include/linux/cpufreq.h
> +++ b/include/linux/cpufreq.h
> @@ -482,8 +482,8 @@ extern struct cpufreq_governor cpufreq_gov_conservative;
>   *********************************************************************/
>  
>  /* Special Values of .frequency field */
> -#define CPUFREQ_ENTRY_INVALID	~0
> -#define CPUFREQ_TABLE_END	~1
> +#define CPUFREQ_ENTRY_INVALID	~0u
> +#define CPUFREQ_TABLE_END	~1u
>  /* Special Values of .flags field */
>  #define CPUFREQ_BOOST_FREQ	(1 << 0)
>  
> -- 
> 1.7.1
>