Save this as sc.d:
--------------------------- sc.d ---------------------------
#!/usr/sbin/dtrace -qs
syscall:::entry
/pid == $target/
{
@count[probefunc] = count();
self->st[probefunc] = timestamp;
}
syscall:::return
/self->st[probefunc]/
{
this->t = (timestamp - self->st[probefunc]) / 1000;
@times[probefunc] = sum(this->t);
self->st[probefunc] = 0; this->t = 0;
}
END
{
printf("%-16s %-8s %-8s\n","SYSCALL","COUNT","TIME(us)");
printa("%-16s %-@8d %-@8d\n",@count,@times);
}
--------------------------------------------------------------------------------------
Run as:
#./sc.d -p [PID_OF_TARGET_PROCESS]
You'll get something like this when you ^c to terminate:
^C
SYSCALL COUNT TIME(us)
chdir 3 21
faccessat 4 13
forksys 6 1855
waitsys 6 14956
fstatat64 7 67
gtime 18 26
setpgrp 29 61
read 51 13289380
write 60 1037
ioctl 97 701
sigaction 216 358
lwp_sigmask 232 443
Sorry about the formatting - I can't seem to get a fixed-width font to work
in
$%$^&#$%$@#$ gmail.
Anyway, the Time is in microseconds, and represents the total amount
of wall clock time during the collection period for each system call.
A lot more is of course possible.
Jim
On Wed, Mar 5, 2014 at 12:53 PM, Mahmood Naderan <[email protected]>wrote:
> Ok I found 'truss' command to be useful for my problem.
> Just one question, hope that somebody knows the answer. I know that this
> list is not directly related to DTrace list.
>
> I used 'truss -c -p 1010' and the output is
>
> syscall seconds calls errors
> write .005 77
> close .000 1
> brk .000 8
> lwp_park .407 6631
> lwp_unpark .246 5832
> sysconfig .000 1
> yield .012 212
> llseek .000 1
> lwp_mutex_timedlock .000 1
> open64 .000 2
> -------- ------ ----
> sys totals: .672 12766 0
> usr time: 310.172
> elapsed: 33.990
>
>
> I see that elapsed time is less that (sys+user).
> Does anybody know the reason?
>
> Regards,
> Mahmood
>
>
> On Sunday, March 2, 2014 12:08 PM, Mahmood Naderan <[email protected]>
> wrote:
> Hi,
> How can I find the execution time of each syscall using DTrace (or any
> other developers related tool)?
> For example, I can see
>
> fstat 100
> sendto 400
> brk 50
>
> I want to know which one has the largest overhead in the run time.
> 100 * (execution of one fstat)
> or
> 400 * (execution of one sendto)
> or
> 50 * (execution of one brk)
>
> Any feedback is appreciated.
>
> Regards,
> Mahmood
> *dtrace-discuss* |
> Archives<https://www.listbox.com/member/archive/184261/=now>
> <https://www.listbox.com/member/archive/rss/184261/24744495-1acb86e1> |
> Modify <https://www.listbox.com/member/?&;> Your Subscription
> <http://www.listbox.com/>
>
>
> *dtrace-discuss* |
> Archives<https://www.listbox.com/member/archive/184261/=now>
> <https://www.listbox.com/member/archive/rss/184261/21493819-f797b643> |
> Modify<https://www.listbox.com/member/?&;>Your Subscription
> <http://www.listbox.com>
>
-------------------------------------------
dtrace-discuss
Archives: https://www.listbox.com/member/archive/184261/=now
RSS Feed: https://www.listbox.com/member/archive/rss/184261/25769126-e243886f
Modify Your Subscription:
https://www.listbox.com/member/?member_id=25769126&id_secret=25769126-8d47a7b2
Powered by Listbox: http://www.listbox.com
