[PATCH] winsup/cygwin/times.cc (times): follow Linux and allow for a NULL buf argument

classic Classic list List threaded Threaded
2 messages Options
Reply | Threaded
Open this post in threaded view
|

[PATCH] winsup/cygwin/times.cc (times): follow Linux and allow for a NULL buf argument

ASSI

Adresses a problem reported on the Cygwin list.

---
 winsup/cygwin/times.cc | 17 +++++++++++------
 1 file changed, 11 insertions(+), 6 deletions(-)

diff --git a/winsup/cygwin/times.cc b/winsup/cygwin/times.cc
index 8908d44f1..909cae1f1 100644
--- a/winsup/cygwin/times.cc
+++ b/winsup/cygwin/times.cc
@@ -72,12 +72,17 @@ times (struct tms *buf)
       /* ticks is in in 100ns, convert to clock ticks. */
       tc = (clock_t) (ticks.QuadPart * CLOCKS_PER_SEC / NS100PERSEC);
 
-      buf->tms_stime = __to_clock_t (&kut.KernelTime, 0);
-      buf->tms_utime = __to_clock_t (&kut.UserTime, 0);
-      timeval_to_filetime (&myself->rusage_children.ru_stime, &kut.KernelTime);
-      buf->tms_cstime = __to_clock_t (&kut.KernelTime, 1);
-      timeval_to_filetime (&myself->rusage_children.ru_utime, &kut.UserTime);
-      buf->tms_cutime = __to_clock_t (&kut.UserTime, 1);
+      /* Linux allows a NULL buf and just returns tc in that case, so
+ mimic that */
+      if (buf)
+ {
+  buf->tms_stime = __to_clock_t (&kut.KernelTime, 0);
+  buf->tms_utime = __to_clock_t (&kut.UserTime, 0);
+  timeval_to_filetime (&myself->rusage_children.ru_stime, &kut.KernelTime);
+  buf->tms_cstime = __to_clock_t (&kut.KernelTime, 1);
+  timeval_to_filetime (&myself->rusage_children.ru_utime, &kut.UserTime);
+  buf->tms_cutime = __to_clock_t (&kut.UserTime, 1);
+ }
     }
   __except (EFAULT)
     {
--
2.23.0


Achim.
--
+<[Q+ Matrix-12 WAVE#46+305 Neuron microQkb Andromeda XTk Blofeld]>+

SD adaptation for Waldorf rackAttack V1.04R1:
http://Synth.Stromeko.net/Downloads.html#WaldorfSDada
Reply | Threaded
Open this post in threaded view
|

Re: [PATCH] winsup/cygwin/times.cc (times): follow Linux and allow for a NULL buf argument

Ken Brown-6
On 9/15/2019 12:28 PM, Achim Gratz wrote:

>
> Adresses a problem reported on the Cygwin list.
>
> ---
>   winsup/cygwin/times.cc | 17 +++++++++++------
>   1 file changed, 11 insertions(+), 6 deletions(-)
>
> diff --git a/winsup/cygwin/times.cc b/winsup/cygwin/times.cc
> index 8908d44f1..909cae1f1 100644
> --- a/winsup/cygwin/times.cc
> +++ b/winsup/cygwin/times.cc
> @@ -72,12 +72,17 @@ times (struct tms *buf)
>         /* ticks is in in 100ns, convert to clock ticks. */
>         tc = (clock_t) (ticks.QuadPart * CLOCKS_PER_SEC / NS100PERSEC);
>  
> -      buf->tms_stime = __to_clock_t (&kut.KernelTime, 0);
> -      buf->tms_utime = __to_clock_t (&kut.UserTime, 0);
> -      timeval_to_filetime (&myself->rusage_children.ru_stime, &kut.KernelTime);
> -      buf->tms_cstime = __to_clock_t (&kut.KernelTime, 1);
> -      timeval_to_filetime (&myself->rusage_children.ru_utime, &kut.UserTime);
> -      buf->tms_cutime = __to_clock_t (&kut.UserTime, 1);
> +      /* Linux allows a NULL buf and just returns tc in that case, so
> + mimic that */
> +      if (buf)
> + {
> +  buf->tms_stime = __to_clock_t (&kut.KernelTime, 0);
> +  buf->tms_utime = __to_clock_t (&kut.UserTime, 0);
> +  timeval_to_filetime (&myself->rusage_children.ru_stime, &kut.KernelTime);
> +  buf->tms_cstime = __to_clock_t (&kut.KernelTime, 1);
> +  timeval_to_filetime (&myself->rusage_children.ru_utime, &kut.UserTime);
> +  buf->tms_cutime = __to_clock_t (&kut.UserTime, 1);
> + }
>       }
>     __except (EFAULT)
>       {

Pushed, with a slight tweak to the commit message.  Thanks.

Ken