• Ary Manzana (44/44) Aug 11 2007 Take a look at this trace.log file:
• Lutger (10/17) Aug 11 2007 My guess is that is because the timings are very short. In general I
• Bill Baxter (6/26) Aug 12 2007 I was thinking it was just some sort of rounding.

Ary Manzana <ary esperanto.org.ar> writes:

Take a look at this trace.log file:

------------------
	    1	foo
	    6	foo2
bar	7	199	49
	    4	someMethod
------------------
	    1	main
foo	1	50	9
	    1	bar
------------------
	    5	foo3
	    1	main
foo2	6	212	54
	    6	bar
------------------
	    5	someMethod
foo3	5	209	45
	    5	foo2
------------------
	    4	bar
	    1	main
someMethod	5	254	45
	    5	foo3
------------------
main	0	230	28
	    1	foo
	    1	foo2
	    1	someMethod

======== Timer Is 3579545 Ticks/Sec, Times are in Microsecs ========

   Num          Tree        Func        Per
   Calls        Time        Time        Call

       6          59          15           2     foo2
       7          55          13           1     bar
       5          70          12           2     someMethod
       5          58          12           2     foo3
       1          64           7           7     main
       1          13           2           2     foo
---

I can't understand why main has a tree time of 64 and someMethod, which 
is invoked by main, has 70. An invoked method can't spend more time in 
total than the function that invoked it. Or yes?

Well, maybe looking at the tree ticks... main has 230, but then again, 
someMethod has 254.

Lutger <lutger.blijdestijn gmail.com> writes:

Ary Manzana wrote:
...
 
 I can't understand why main has a tree time of 64 and someMethod, which 
 is invoked by main, has 70. An invoked method can't spend more time in 
 total than the function that invoked it. Or yes?
 
 Well, maybe looking at the tree ticks... main has 230, but then again, 
 someMethod has 254.

My guess is that is because the timings are very short. In general I 
have noticed some irregular behavior, it is a good idea to profile 
multiple times.

There are some limitations and reasons for irregular timings written 
under the 'notes' section on digitalmars: 
http://www.digitalmars.com/ctg/trace.html

Finally, on some machines, the windows high performance api can have 
bugs (I have such a machine, motherboard issue).

Bill Baxter <dnewsgroup billbaxter.com> writes:

Lutger wrote:
 Ary Manzana wrote:
 ...
 I can't understand why main has a tree time of 64 and someMethod, 
 which is invoked by main, has 70. An invoked method can't spend more 
 time in total than the function that invoked it. Or yes?

 Well, maybe looking at the tree ticks... main has 230, but then again, 
 someMethod has 254.

 
 My guess is that is because the timings are very short. In general I 
 have noticed some irregular behavior, it is a good idea to profile 
 multiple times.
 
 There are some limitations and reasons for irregular timings written 
 under the 'notes' section on digitalmars: 
 http://www.digitalmars.com/ctg/trace.html
 
 Finally, on some machines, the windows high performance api can have 
 bugs (I have such a machine, motherboard issue).

I was thinking it was just some sort of rounding.
I see things like Per Call = 0, with Func Time = <not zero> very often. 
   So I was just assuming the numbers were rounded to the nearest unit, 
whatever the unit is.

--bb