Subject: Re: further vm adventures
To: None <jiho@postal.c-zone.net>
From: Chuck Cranor <chuck@maria.wustl.edu>
List: tech-kern
Date: 04/27/1998 10:28:23
>I am unable to use ddb in this situation, because although I can switch VTs to
>enter ddb, once leaving ddb there is no way to get back to the X desktop.
you've just got to find the correct key. ctrl-alt-f7 works for me (using
pcvt).
>Suppose I come up on the X desktop running a file manager, open an xterm, and
>run 'systat vmstat' (Chuck's heart sinks). After running an image-intesive
>client a few times with the same data, things settle out into the following
>consistent pattern:
> exited execed exited execed exited
> ------ ------ ------ ------ ------
> wired: 2332K 2356K 2332K 2356K 2332K
> active: 7972K 9604K 8292K 9924K 8616K
> inactive: 1376K 988K 1376K 988K 1376K
> free: 11456K 10188K 11136K 9868K 10812K
>The X server gains 8K in its RSS across each interation. No other programs
>undergo any change. The net result is that about 320K (give or take a 4K page)
>is being lost from the free list to the active list on each iteration. This
>amount will vary depending on the data fed the client, but it will become
>consistent for any given set of data. I call this a leak because the client
>has exited, and the memory at issue cannot be found associated with any other
>program.
that's interesting, but there is not enough info to know what is going
on yet. you say the 320K returns to the free list when the X server
exits? in that case, it would be useful to look at the output of
ddb's "show map/f" on the X server's map from "exited" to "execed" to
"exited" and see what mappings changed. i'd prob. comment out the
call to _vm_object_print() in _vm_map_print() to reduce the noise in
ddb's output.
chuck