Conversation with #inferno at Thu Dec  4 15:46:49 2008 on powerman-asdf@irc.freenode.net (irc)
(15:46:49) #inferno: The topic for #inferno is: Inferno and Limbo | Website: http://www.vitanuova.com/inferno/index.html | Documentation: http://www.vitanuova.com/inferno/docs.html | Wiki: http://canto.hopto.org/wiki/1/index.html | Tutorial: http://www.resc.rdg.ac.uk/twiki/bin/view/Resc/InfernoTutorial | Mailing list archives: http://dir.gmane.org/gmane.os.inferno.general
(15:55:25) sqweek: man, inferno isn't very nice to my netbsd box
(15:55:54) sqweek: ps: kvm_getproc2: Cannot allocate memory: Cannot allocate memory
(15:56:00) sqweek: i had a cron job fail a few hours ago with: 
(15:56:19) sqweek: (heh, pasting in wm/irc didn't work quite how i expected)
(15:57:28) mjl-: multi-line paste goes to data-file immediately, paste without newline goes to text-entry box
(15:57:35) mjl-: best i could come up with...
(15:57:56) mjl-: perhaps a multi-line paste should first flush the text in the input entry?
(15:58:09) mjl-: sqweek: but emu is using up all memory?
(15:58:18) sqweek: anyway, there's 143 emu procs running to power 4 instances of ircfs
(15:58:44) mjl-: hmm, a bit much indeed.
(15:58:44) mjl-: i wonder why that is
(15:59:12) sqweek: they all claim to have 7000k of RSS
(15:59:27) mjl-: i would expect:  4 for reading from network, then 4 (mostly idle) to network, and a few for inferno itself
(15:59:29) sqweek: but i have a feeling that is not accurate ;)
(15:59:40) mjl-: perhaps that's shared memory?
(16:00:02) mjl-: i think that emu on openbsd uses rfork that shares the memory
(16:00:06) sqweek: there must be something like that going on
(16:00:20) mjl-: but the 143 procs is worth investigating
(16:00:52) mjl-: it seems i have 511 emu's running on my linux box at this moment
(16:01:24) mjl-: sqweek: how many progs does ps(1) in the emu running the ircfs-en show?
(16:01:28) sqweek: it doesn't have much respect for the process table i guess ;)
(16:02:00) sqweek: 47
(16:03:04) mjl-: i have about the same for 4 ircfs'en, plus styxlisten
(16:03:15) sqweek: 16 Ircfs  5 Bufio  4 Styxlisten  9 Sh
(16:03:17) mjl-: perhaps the styxlisten spawns some procs too...
(16:04:35) sqweek: 10 Styx  2 Mntgen  2 Nametree  1 Cs  1 Ps  1 Export
(16:07:16) mjl-: well, i still don't see how that would cause 143 procs
(16:07:28) mjl-: more than one proc per prog doesn't make sense in general i think
(16:08:00) mjl-: though my reading of the inferno source stopped somewhere half-way half a year ago, and i haven't returned to it yet.
(16:08:51) mjl-: i thought procs were made on demand, and reused.  and that they are only created for host-os system calls (that block).
(16:09:16) mjl-: and progs (inferno programs) are just an internal emu/dis datastructure representing a thread.
(16:54:55) npe [n=npe@32.97.110.63] entered the room.
(18:14:33) KillerX left the room (quit: Remote closed the connection).
(18:43:02) gualteri left the room (quit: "'out'").
(22:00:07) ooooo [i=none@201.80.219.33] entered the room.
(23:52:00) newmanbe [n=btdn@138.74.131.25] entered the room.
(00:52:47) npe left the room (quit: Connection timed out).
(00:53:14) newmanbe left the room (quit: Read error: 104 (Connection reset by peer)).
(00:54:04) newmanbe [n=btdn@138.74.131.25] entered the room.