[PATCH 1/1] ktime: Use macro NSEC_PER_USEC instead of a magic number

[PATCH 1/1] ktime: Use macro NSEC_PER_USEC instead of a magic number

[y]

From: Liu Ying <Ying.Liu@freescale.com>

We've got the macro NSEC_PER_USEC defined in header file
include/linux/time.h. To make the code decent, this patch
replaces the magic number 1000 to convert bewteen a time
value in microseconds and one in nanoseconds with the
macro NSEC_PER_USEC.

Signed-off-by: Liu Ying <Ying.Liu@freescale.com>
Cc: David S. Miller <davem@davemloft.net>
Cc: Daniel Borkmann <dborkman@redhat.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: John Stultz <john.stultz@linaro.org>
---
 include/linux/ktime.h |    7 ++++---
 1 files changed, 4 insertions(+), 3 deletions(-)

diff --git a/include/linux/ktime.h b/include/linux/ktime.h
index bbca128..5de963d 100644
--- a/include/linux/ktime.h
+++ b/include/linux/ktime.h
@@ -229,7 +229,8 @@ static inline ktime_t timespec_to_ktime(const struct timespec ts)
 static inline ktime_t timeval_to_ktime(const struct timeval tv)
 {
 	return (ktime_t) { .tv = { .sec = (s32)tv.tv_sec,
-				   .nsec = (s32)tv.tv_usec * 1000 } };
+				   .nsec = (s32)tv.tv_usec *
+					   (s32)NSEC_PER_USEC} };
 }
 
 /**
@@ -320,12 +321,12 @@ static inline s64 ktime_us_delta(const ktime_t later, const ktime_t earlier)
 
 static inline ktime_t ktime_add_us(const ktime_t kt, const u64 usec)
 {
-	return ktime_add_ns(kt, usec * 1000);
+	return ktime_add_ns(kt, usec * NSEC_PER_USEC);
 }
 
 static inline ktime_t ktime_sub_us(const ktime_t kt, const u64 usec)
 {
-	return ktime_sub_ns(kt, usec * 1000);
+	return ktime_sub_ns(kt, usec * NSEC_PER_USEC);
 }
 
 extern ktime_t ktime_add_safe(const ktime_t lhs, const ktime_t rhs);
-- 
1.7.1

Re: [PATCH 1/1] ktime: Use macro NSEC_PER_USEC instead of a magic number

[Daniel Borkmann]

On 05/06/2013 02:42 PM, y@shlinux1.ap.freescale.net wrote:
> From: Liu Ying <Ying.Liu@freescale.com>

Nitpick:

If you only have 1 patch, then you don't need "1/1" in the subject.
The term magic number here might be a little exaggerated, not so
magic actually. :-)

But, in general I agree that we could do that also since it is used
elsewhere in the same header file. Thus, this might be more consistent.

> We've got the macro NSEC_PER_USEC defined in header file
> include/linux/time.h. To make the code decent, this patch
> replaces the magic number 1000 to convert bewteen a time
> value in microseconds and one in nanoseconds with the
> macro NSEC_PER_USEC.
>
> Signed-off-by: Liu Ying <Ying.Liu@freescale.com>
> Cc: David S. Miller <davem@davemloft.net>
> Cc: Daniel Borkmann <dborkman@redhat.com>
> Cc: Thomas Gleixner <tglx@linutronix.de>
> Cc: John Stultz <john.stultz@linaro.org>
> ---
>   include/linux/ktime.h |    7 ++++---
>   1 files changed, 4 insertions(+), 3 deletions(-)
>
> diff --git a/include/linux/ktime.h b/include/linux/ktime.h
> index bbca128..5de963d 100644
> --- a/include/linux/ktime.h
> +++ b/include/linux/ktime.h
> @@ -229,7 +229,8 @@ static inline ktime_t timespec_to_ktime(const struct timespec ts)
>   static inline ktime_t timeval_to_ktime(const struct timeval tv)
>   {
>   	return (ktime_t) { .tv = { .sec = (s32)tv.tv_sec,
> -				   .nsec = (s32)tv.tv_usec * 1000 } };
> +				   .nsec = (s32)tv.tv_usec *
> +					   (s32)NSEC_PER_USEC} };

Nitpick: Likely, this could be reduced to one cast only.

Otherwise looks good.

>   }
>
>   /**
> @@ -320,12 +321,12 @@ static inline s64 ktime_us_delta(const ktime_t later, const ktime_t earlier)
>
>   static inline ktime_t ktime_add_us(const ktime_t kt, const u64 usec)
>   {
> -	return ktime_add_ns(kt, usec * 1000);
> +	return ktime_add_ns(kt, usec * NSEC_PER_USEC);
>   }
>
>   static inline ktime_t ktime_sub_us(const ktime_t kt, const u64 usec)
>   {
> -	return ktime_sub_ns(kt, usec * 1000);
> +	return ktime_sub_ns(kt, usec * NSEC_PER_USEC);
>   }
>
>   extern ktime_t ktime_add_safe(const ktime_t lhs, const ktime_t rhs);
>

Re: [PATCH 1/1] ktime: Use macro NSEC_PER_USEC instead of a magic number

[Daniel Borkmann]

On 05/07/2013 11:08 AM, Liu Ying wrote:
> 2013/5/7 Liu Ying <liu.y.victor@gmail.com>
>> 2013/5/7 Daniel Borkmann <dborkman@redhat.com>
>>> On 05/06/2013 02:42 PM, y@shlinux1.ap.freescale.net wrote:
>>>> From: Liu Ying <Ying.Liu@freescale.com>

>>> If you only have 1 patch, then you don't need "1/1" in the subject.
>>> The term magic number here might be a little exaggerated, not so
>>> magic actually. :-)
>>>
>> Thanks. I will update this. How about turning "magic number" to "immediate
>> number"?

Maybe like ``ktime: Use macro NSEC_PER_USEC where appropriate''.