Date: Mon, 17 Jan 2000 15:41:34 +0100 (MET) From: Ariel Burbaickij <Ariel.Burbaickij@mni.fh-giessen.de> To: freebsd-questions@freebsd.org Subject: understanding timing Message-ID: <Pine.GSO.4.10.10001171521100.807-100000@sun9>
next in thread | raw e-mail | index | archive | help
We got homework to implement shell.I am ready so far .So this mail is not
begging for doing my work by anyone else :)).We got some solution pattern
from out teacher suggesting following:
void do_with_time(int argc,char * argv [])//function for doing it.
{
clock_t start ,end ;
struct tm tnt ;
pid_t pid1 ,pid2 //YTou already see here it is about using two forks.
if(argc > 1)//we expect the coomand proper as well
{
pid1 = fork()
switch(pid1)
{
case -1 :
/* fear ,fire ,foes */
case 0 :
{
start = times(&tnt);
pid2 = fork()
switch(pid2)
{
case -1 :
/* fear fire ,foes */
case 0 :
{
execvp(argv[1],&argv[1]);
printf("You should never see me!");
}
default:
waitpid(pid,NULL,NULL) //quick and dirty
end = times(&tnt);
/* kind
printf(Real time is difference between end and start);
printf(System time is &time.tms_cstime);//evaluation order is
//surely against us
}//end of second switch
default :
waitpid(pid,NULL,NULL);
}//end of first switch
}//end of if
else
printf(You need at least two to dance tango);
What i do not understand is why you need two fork's() already provided
with waitpid()and so having mechanism for waiting for well-defined
child,so eliminating the danger of cumulating the childrens' time
in tms structrue.I would understand with wait() only.But here I guess
I need a little help from you.
kind regards,
Ariel
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-questions" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.GSO.4.10.10001171521100.807-100000>