Help with profiling some code?

Joshua 'Schwa' Gargus schwa at cc.gatech.edu
Thu May 15 21:07:26 UTC 2003


In the same order as before (faster execution first):

**Memory**
	old			+306,344 bytes
	young		+274,872 bytes
	used		+581,216 bytes
	free		-581,216 bytes

**GCs**
	full			0 totalling 0ms (0.0% uptime)
	incr		541 totalling 298ms (11.0% uptime), avg 1.0ms
	tenures		2 (avg 270 GCs/tenure)
	root table	0 overflows


and:

**Memory**
	old			+144,920 bytes
	young		+498,844 bytes
	used		+643,764 bytes
	free		-643,764 bytes

**GCs**
	full			0 totalling 0ms (0.0% uptime)
	incr		526 totalling 522ms (8.0% uptime), avg 1.0ms
	tenures		1 (avg 526 GCs/tenure)
	root table	0 overflows


Both images are based on Croquet (update 364).  There is not much
difference between the two, except that the code in one is quite a bit
newer.  However, everything below transformTransform: is identical.

Thanks,
Joshua




On Thu, May 15, 2003 at 04:56:31PM -0400, Bob Arning wrote:
> On Thu, 15 May 2003 16:47:56 -0400 "Joshua 'Schwa' Gargus" <schwa at cc.gatech.edu> wrote:
> >I tried garbage-collecting, and restarting each image.  Neither
> >made any difference.  Does anyone have any hints about how to 
> >approach this?
> 
> A couple of thoughts:
> 
> 1. What do the memory statistics at the bottom of the MessageTally output say?
> 
> 2. What is different abut the images?
> 
> Cheers,
> Bob



More information about the Squeak-dev mailing list