Re: FVWM: Re: Memory usage

From: Charles Hines <>
Date: Fri, 26 Apr 1996 15:13:42 -0400

>>>>> "Johnny" == Johnny Tolliver <jxt_at_ca04.CAD.ORNL.GOV> writes:

Johnny> On Fri, 19 Apr 1996 chuck_hines_at_VNET.IBM.COM wrote:
>> Not much actually. On my system, I just checked, restarted fvwm,
>> checked, went back to fvwm2, checked again, and here's the results:
>>
>> USER PID %CPU %MEM SZ RSS TTY STAT STIME TIME COMMAND
>> hines 43806 0.2 1.0 596 860 hft/1 S Apr 15 14:06 fvwm2 -version
>> ^^^ ^^^
>> hines 44000 3.7 1.0 424 656 hft/1 S 11:09:55 0:01 fvwm -version
>> ^^^ ^^^
>> hines 44011 12.0 1.0 520 784 hft/1 S 11:10:48 0:03 fvwm2 -version

Johnny> To be fair, you need to add in the cost of the
Johnny> "optional"FvwmPager and GoodStuff/FvwmButtons modules.

I don't agree. I don't think that it's necessary to show them as an
answer to what was asked for the mere fact that they are optional
(note, no quotes). I never use the FvwmButtons myself. I do use the
pager (the only module I really use, actually), but if you're just
comparing window managers memory usage, you don't need to look at the
Pager since it only gives a visual interface to things that can be
easily bound to keys.

Now, for 1.24r you could compile it so the only pager is the external
pager module as well. So, if 1.xx were compiled with the built in
pager then you could add the memory usage of the 2.x pager module into
the 2.x figures. Still, it's just simpler to show the size difference
of the main fvwm2 modules.

It still takes a lot less than mwm and dtwm do, even with several
modules running and adding their amounts in. The difference between
1.xx and 2.xx isn't that great (even with adding in the module sizes),
and I hope to be able to decrease the difference (or at least make
sure it doesn't increase any further).

Johnny> Anybody know why top and ps don't agree? Just what exactly are "SIZE",
Johnny> "RES", "SZ", and "RSS"?

>From AIX's InfoExplorer description for ps:

  SZ (-l, l) The size in 1K-byte units of the core image of the
                process.
  SIZE (v) Virtual size of the data section of the process (in
                1K-byte units).
  RSS (v) Real-memory (resident set) size of the process (in
                1K-byte units).

RES isn't described, but I would assume that it is really RSS since
the values match in your example.

Chuck


--
Visit the official FVWM web page at <URL:http://www.hpc.uh.edu/fvwm/>.
To unsubscribe from the list, send "unsubscribe fvwm" in the body of a
message to majordomo_at_hpc.uh.edu.
To report problems, send mail to fvwm-owner_at_hpc.uh.edu.
Received on Fri Apr 26 1996 - 14:11:40 BST

This archive was generated by hypermail 2.3.0 : Mon Aug 29 2016 - 19:37:58 BST