Deadlock on fragmented memory?

[Martin Husemann <martin%duskware.de@localhost>]

Hey folks,

I sometimes see my shark (arm v4, strongarm 110, 96 MB of ram, main
characteristics: very *slow* disk access) "hang" with many processes
blocked on "vmem". All of etc.daily is hit by that, lots of cron instances,
and no new exec(2) work (or so it seems).

Various suggestions have been raised off-list (like prepopulating the
exec pool), but nothing concrete ever came of it.

One option (which may be a too big hammer) is to cut down the number of
parallel exec calls, by adding

#define MAXEXEC 1

to sys/arch/shark/include/param.h, similar to what vax does.

I'm planning to at least test that, but would be open for other suggestions,
especially if other low-ram machines could benefit as well.

Martin