Re: [gnugo-devel] trevor_1_12.2

[Trevor Morris]

Looks to me like the gg_cputime version of this is using the return
value of times, which is "Upon successful completion, times returns the
elapsed real    time, in clock     ticks, from an arbitrary point in the past
(e.g., system start-up time).     This point    does not change from one
invocation of times to  another."   i.e. the current unix implementation is
a just fancy way to get the elapse wall clock time!

My original implementation should work fine (the one now deleted):
-    struct tms t;
-    UNUSED(s);
-    times(&t);
-       return gtp_success(id, "%ld", t.tms_utime + t.tms_stime + t.tms_cutime +
t.tms_cstime);

I'll put together a patch tomorrow with this change, if no one else does.

-Trevor



At 06:34 PM 10/25/2001 +0200, you wrote:
>Trevor Morris wrote:
>
>>  - gtp_cputime now uses gg_cputime
>
>> This works on Mingw32 and VC builds.
>> However, it reports wall time rather than CPU time on Cygwin.
>
>Why is that? cygwin comes with both sys/times.h and times(),
>so gg_cputime invokes times()
>
>Teun
>
>_______________________________________________
>gnugo-devel mailing list
>address@hidden
>http://mail.gnu.org/mailman/listinfo/gnugo-devel
>