To strip, or not to strip...

[Daryl Morse]

Several weeks ago, there was a brief discussion on stripping
executables. dxterm was given as an example of an unstripped
executable. Because of the memory greediness of the dx* executables, I
thought I would take try stripping dxterm to see if there were any
noticable differences in performance/memory utilization.

  This is what I found:

1. Size of executable *is* significantly smaller. (BTW, I'm running
   MIPS-based machine.)

	unstripped dxterm - 4423920 bytes
	stripped   dxterm - 2842624 bytes

2. Size of the virtual address space as reported by  ps -u (SZ)

	unstripped dxterm - 3244
	stripped   dxterm - 3244

TSIZE was also unchanged. Am I missing something here? Not being a
Unix guru, I am not sure if there are other VM stats which might
better indicate altered memory utilization. (ie. TRS, RSS, etc.) Are
the savings that result from stripping executables only in disk space?

I would appreciate knowing whether or not strip actually reduces
memory utilization of executables. Please reply by email. If there
appears to be interest in this issue, I will summarize and repost.

Thanks, very much.
--
Daryl Morse                     | Voice : (604) 293-5476
MPR Teltech Ltd. 		| Fax   : (604) 293-5787
8999 Nelson Way, Burnaby, BC    | E-Mail: morse at quark.mpr.ca
Canada, V5A 4B5                 |         quark.mpr.ca!morse at uunet.uu.net