Conversation with #inferno at Fri Oct 31 18:26:06 2008 on powerman-asdf@irc.freenode.net (irc)
(18:26:06) #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
(18:33:01) newmanbe left the room (quit: Read error: 104 (Connection reset by peer)).
(18:33:32) newmanbe [n=btdn@138.74.131.25] entered the room.
(18:56:29) btdn [n=btdn@138.74.131.25] entered the room.
(19:15:19) newmanbe left the room (quit: Read error: 110 (Connection timed out)).
(19:36:59) btdn left the room (quit: Read error: 110 (Connection timed out)).
(21:21:30) btdn [n=btdn@138.74.131.25] entered the room.
(03:26:03) underspecified left the room (quit: ).
(04:35:49) underspecified [n=eric@isa7-dhcp-116-122.naist.jp] entered the room.
(08:29:10) underspecified left the room (quit: ).
(08:47:17) underspecified [n=eric@isa7-dhcp-116-122.naist.jp] entered the room.
(09:06:34) anothy left the room (quit: Read error: 110 (Connection timed out)).
(09:22:37) btdn left the room (quit: Read error: 110 (Connection timed out)).
(10:15:29) KillerX [n=anant@gentoo/developer/KillerX] entered the room.
(10:16:27) KillerX left the room (quit: Read error: 54 (Connection reset by peer)).
(11:10:54) underspecified left the room (quit: ).
(12:36:18) underspecified [n=eric@softbank220043052011.bbtec.net] entered the room.
(13:56:21) btdn [n=btdn@138.74.131.25] entered the room.
(14:18:23) btdn left the room (quit: Read error: 110 (Connection timed out)).
(14:44:39) KillerX [n=anant@gentoo/developer/KillerX] entered the room.
(15:28:53) anothy [n=a@c-98-221-104-176.hsd1.nj.comcast.net] entered the room.
(16:15:44) ***powerman wonder is there any somethingfs service in Inferno which able to support concurrent operations...
(16:17:32) powerman: there styxflush.m, but it's not used by even single app
(16:18:06) powerman: the styxservers.m use 3 threads to serve non-concurrent blocking I/O...
(16:18:53) powerman: the ftpfs.b Tflush handler leave no chances it support concurrent requests too...
(16:23:30) powerman: Maybe I miss something and support for concurrent requests is something redundant and/or needless and considered harmful because it lead to overcomplicated implementation?
(16:24:48) powerman: In this case, can anybody enlighten me and give me any idea how to design servicefs I need to develop right now without support for concurrent requests?
(17:02:30) anothy: look at the dns server. it supports multiple outstanding (concurrent) requests.
(17:03:18) anothy: (i'm pretty sure, because i wrote the first draft of it)
(17:03:41) powerman: anothy: dns doesn't do styx :) it do file2chan
(17:04:36) anothy: it's all styx. that's how all file access is done.
(17:05:40) powerman: I know. But question is how to write service which able to serve a little more than just r/w for one file...
(17:09:53) anothy: well, there's lots of things which support concurrent access. just scanning through the index of section 4 for ideas, there's 9660srv, dossrv, and factotum that certainly do. others likely do, depending on what exactly you mean.
(17:13:31) powerman: anothy: no one of them support concurrent access. 
(17:13:55) powerman: 9660srv:
(17:13:55) powerman: rflush(t: ref Tmsg.Flush): ref Rmsg
(17:13:55) powerman: {
(17:13:55) powerman:          return ref Rmsg.Flush(t.tag);
(17:13:55) powerman: }
(17:14:08) powerman: dossrv: same thing
(17:14:48) powerman: factotum: I think it use keyfs, which use styxservers.m, which doesn't support concurrent access
(17:16:41) powerman: actually there a little reason for, say, 9660srv or dossrv or even keyfs to support concurrent access - they all should be able to serve incoming requests without delays (and, especially, delays for unpredictable time like when using network in ftpfs)
(17:20:45) powerman: I may, in theory, also avoid needs in concurrent styx by using file2chan, but, sadly, I need directory listing (and so files in it) created and updated on-the-fly, based on information from network.
(17:24:13) powerman: I've network service, which serve two types of requests: 1) get list of currently available applications and 2) run application by it name. I wish to provide access to that service from Inferno. The idea was: mount my appfs in some directory, on directory read request do "get list" and return available application names and file names in that directory; on file open+read request do "run application" and return data from that application as file content.
(17:25:05) powerman: And it should be possible to open+read many files in that directory in parallel, to have many application running at same time.
(17:26:49) powerman: Without requirement for dynamic directory listings, with stable list of available applications, it should be possible to have this done with file2chan (there about 800 apps, and I hope so many file2chan's isn't a problem for Inferno). But list of available apps is dynamic, so...
(17:27:09) anothy: maybe i don't understand what you mean by concurrent access.
(17:28:34) powerman: I mean if I run: cat /mnt/appfs/app1 & cat /mnt/appfs/app2 & then both files should be opened in parallel (and so my network service should receive 2 requests and start both apps in parallel).
(17:29:31) powerman: Current styxservers.m implementation require Rmsg send after receiving Tmsg. It will not receive next Tmsg before Rmsg to prev Tmsg will be sent. 
(17:31:49) powerman: That, it turn, mean, after my appfs will receive Tmsg.Read from 'cat /mnt/appfs/app1' it will block waiting from reply from my network service which actually run that app and will not be able to receive Tmsg.Read from 'cat /mnt/appfs/app2' until app1 will be finished and network service will send it reply.
(17:32:45) powerman: But the goal is to have both app1 and app2 running in parallel by that network service if I start two such 'cat' processes in Inferno.
(17:33:59) anothy: i don't have time to check this right now (trying to get something done before noon), but i think you're wrong.
(17:34:12) anothy: rebuild dossrv (for example) with a 10 second sleep before it replies to things.
(17:34:17) powerman: I will be happy if I wrong. :)
(17:34:48) anothy: i think you'll find "cat /n/dosdisk/a && cat /n/dosdisk/b" works fine still (assuming both files exist)
(17:35:07) anothy: if i'm right, you're noting that it's single-threaded *per fd*, which you get a new one of per open.
(17:35:37) powerman: I will try it, but I think rudimentary Tflush handlers are make things clear enough.
(17:35:40) anothy: that's how the file2chan stuff tracks it, anyway, and i think it's the same for more "full-featured" styx implementations.
(17:36:44) anothy: anyway, back to work for me. catch me tomorrow or monday and i'm happy to help if you've still not got it sorted.
(17:42:19) acmeuser [n=acmeuser@host81-159-96-222.range81-159.btcentralplus.com] entered the room.
(17:43:13) acmeuser left the room.
(17:54:07) powerman: anothy: I did this test.
(17:54:21) powerman: Everything works as I expected - non-concurrent.
(17:54:51) powerman: Look:
(17:54:51) powerman: ; ./dossrv -f disk -m /n/local
(17:54:51) powerman: ; ls -l /n/local
(17:54:51) powerman: --rw-rw-rw- M 6 dos srv 2 Nov 01 18:48 /n/local/file1
(17:54:51) powerman: --rw-rw-rw- M 6 dos srv 2 Nov 01 18:48 /n/local/file2
(17:54:51) powerman: ; grep . /n/local/*
(17:54:51) powerman: /n/local/file1: 1
(17:54:51) powerman: /n/local/file2: 2
(17:54:51) powerman: ; time cat /n/local/file1
(17:54:51) powerman: 1
(17:54:51) powerman: 0.009l 10r 10.01t
(17:55:44) powerman: (it took 10 seconds because I added sys->sleep(5000) before return ref Rmsg.Read(t.tag, data[0:r]); line, so this delay happens twice - when returning file content and when returning eof, I think.
(17:56:07) powerman: And now I run two cat's in parallel:
(17:56:07) powerman: ; time cat /n/local/file1 & time cat /n/local/file2 &
(17:56:07) powerman: ; 1
(17:56:07) powerman: 0l 10r 10t
(17:56:07) powerman: 2
(17:56:07) powerman: 0.002l 19.98r 19.98t
(18:58:23) anothy left the room (quit: "Leaving.").
(20:04:01) ***powerman send this question to mail list, but don't really expect any answers, as usually :(
(21:32:34) anothy [n=a@c-98-221-26-115.hsd1.nj.comcast.net] entered the room.
(22:58:01) anothy left the room (quit: "Leaving.").
(23:08:07) C-Keen [i=ckeen@pestilenz.org] entered the room.
(03:19:41) KillerX left the room (quit: ).
(05:54:51) uriel__ [n=uriel@84-216-52-47.sprayadsl.telenor.se] entered the room.
(05:56:36) uriel left the room (quit: Read error: 60 (Operation timed out)).
(08:15:27) uriel_ is now known as uriel
(09:53:35) paigeadele left the room (quit: Client Quit).
(12:25:18) vt3 [n=vt3@m016020.ppp.asahi-net.or.jp] entered the room.
(12:26:30) vt3 left the room (quit: "Leaving").
(16:23:10) KillerX [n=anant@gentoo/developer/KillerX] entered the room.
(21:41:18) powerman: one more nasty bug with #p :( http://code.google.com/p/inferno-os/issues/detail?id=147
(21:45:15) mycroftiv left the room.
(21:56:32) mycroftiv [n=infernus@h69-128-47-243.mdsnwi.dedicated.static.tds.net] entered the room.
(00:49:42) KillerX left the room (quit: Success).
(02:39:35) paigeadele [i=erratic@75.146.55.252] entered the room.
(03:40:51) anothy [n=a@c-98-221-104-176.hsd1.nj.comcast.net] entered the room.
(05:11:08) ooooo [i=none@201.80.219.33] entered the room.
(06:32:51) underspecified left the room (quit: ).
(08:17:05) underspecified [n=eric@isa7-dhcp-116-122.naist.jp] entered the room.
(08:56:41) powerman: is anybody know how qproduce (emu/port/qio.c) designed to be used?
(08:57:32) powerman: looks like it return -1 on error, but no one caller of qproduce try to check for this error and handle it in some way...
(08:57:51) powerman: so, messages which should be added into queue are just lost
(08:59:31) powerman: ... thus result in unreliable behaviour. Like that bug with /prog/pid/wait. :(
(09:00:16) powerman: that very sad...
(12:57:00) underspecified left the room (quit: ).
(13:49:32) underspecified [n=eric@softbank220043052011.bbtec.net] entered the room.
(17:10:05) anothy left the room (quit: wolfe.freenode.net irc.freenode.net).
(17:11:22) anothy [n=a@c-98-221-104-176.hsd1.nj.comcast.net] entered the room.
(17:23:59) npe [n=npe@32.97.110.64] entered the room.
(17:33:14) npe_ [n=npe@32.97.110.64] entered the room.
(17:33:14) npe_ left the room (quit: Read error: 104 (Connection reset by peer)).
(17:34:29) anothy left the room (quit: wolfe.freenode.net irc.freenode.net).
(17:34:53) anothy [n=a@c-98-221-104-176.hsd1.nj.comcast.net] entered the room.
(17:38:05) uriel_ [n=uriel@84-216-52-47.sprayadsl.telenor.se] entered the room.
(17:41:33) npe left the room (quit: Connection timed out).
(17:42:26) npe [n=npe@32.97.110.64] entered the room.
(17:42:26) npe left the room (quit: Connection reset by peer).
(17:50:54) uriel__ left the room (quit: Read error: 110 (Connection timed out)).
(17:51:38) npe [n=npe@32.97.110.64] entered the room.
(17:52:15) npe_ [n=npe@32.97.110.64] entered the room.
(17:52:15) npe_ left the room (quit: Success).
(18:01:27) npe_ [n=npe@32.97.110.64] entered the room.
(18:03:02) npe_ left the room (quit: Read error: 60 (Operation timed out)).
(18:09:23) npe left the room (quit: Connection timed out).
(18:10:38) npe [n=npe@32.97.110.64] entered the room.
(18:19:49) npe_ [n=npe@32.97.110.64] entered the room.
(18:19:49) npe_ left the room (quit: Read error: 104 (Connection reset by peer)).
(18:26:57) npe left the room (quit: Read error: 110 (Connection timed out)).
(18:29:00) npe [n=npe@32.97.110.64] entered the room.
(18:30:33) npe_ [n=npe@32.97.110.64] entered the room.
(18:39:46) npe__ [n=npe@32.97.110.64] entered the room.
(18:45:23) npe left the room (quit: Connection timed out).
(18:47:57) npe_ left the room (quit: Connection timed out).
(18:49:03) npe [n=npe@32.97.110.64] entered the room.
(18:49:03) npe left the room (quit: Connection reset by peer).
(18:57:23) npe__ left the room (quit: Connection timed out).
(18:58:14) npe [n=npe@32.97.110.64] entered the room.
(18:58:14) npe left the room (quit: Connection reset by peer).
(19:07:26) npe [n=npe@32.97.110.64] entered the room.
(19:07:26) npe left the room (quit: Connection reset by peer).
(19:16:37) npe [n=npe@32.97.110.64] entered the room.
(19:33:51) npe left the room (quit: Connection timed out).
(20:26:43) mjl-: evening
(21:44:16) anothy_lite [n=user@ool-44c4fe31.dyn.optonline.net] entered the room.
(21:48:01) npe [n=npe@32.97.110.64] entered the room.
(21:48:37) npe: test
(21:51:41) olegfink: passed.
(21:56:33) npe left the room (quit: Connection reset by peer).
(21:57:10) anothy_lit1 [n=user@ool-44c4fe31.dyn.optonline.net] entered the room.
(21:57:13) npe [n=npe@32.97.110.64] entered the room.
(21:57:13) npe left the room (quit: Read error: 104 (Connection reset by peer)).
(21:58:47) npe [n=npe@32.97.110.64] entered the room.
(21:58:55) KillerX [n=anant@gentoo/developer/KillerX] entered the room.
(21:59:06) KillerX left the room (quit: Remote closed the connection).
(22:03:13) npe left the room (quit: Read error: 60 (Operation timed out)).
(22:03:53) npe [n=npe@32.97.110.64] entered the room.
(22:06:19) anothy_lite left the room (quit: Read error: 110 (Connection timed out)).
(22:08:38) anothy_lit1 left the room (quit: Read error: 104 (Connection reset by peer)).
(22:13:07) npe_ [n=npe@32.97.110.64] entered the room.
(22:13:07) npe_ left the room (quit: Connection reset by peer).
(22:21:15) npe left the room (quit: Connection timed out).
(22:22:19) npe [n=npe@32.97.110.64] entered the room.
(22:25:43) npe left the room (quit: Read error: 60 (Operation timed out)).
(22:26:23) npe [n=npe@32.97.110.64] entered the room.
(22:26:23) npe left the room (quit: Connection reset by peer).
(22:35:35) npe [n=npe@32.97.110.64] entered the room.
(22:35:35) npe left the room (quit: Read error: 104 (Connection reset by peer)).
(22:43:42) ooooo left the room (quit: Remote closed the connection).
(04:11:31) anothy left the room (quit: wolfe.freenode.net irc.freenode.net).
(04:11:35) anothy [n=a@c-98-221-104-176.hsd1.nj.comcast.net] entered the room.
(05:40:45) C-Keen left the room (quit: Read error: 110 (Connection timed out)).
(07:19:23) npe [n=npe@66.112.249.148] entered the room.
(08:23:47) npe left the room (quit: ).
(09:38:14) C-Keen [i=ckeen@pestilenz.org] entered the room.
(12:58:16) C-Keen left the room (quit: Read error: 60 (Operation timed out)).
(13:54:52) mjl-: howdy
(14:45:34) C-Keen [i=ckeen@pestilenz.org] entered the room.
(15:39:26) KillerX [n=anant@gentoo/developer/KillerX] entered the room.
(17:07:17) npe [n=npe@32.97.110.64] entered the room.
(17:16:30) npe_ [n=npe@32.97.110.64] entered the room.
(17:22:00) npe_ left the room (quit: Read error: 60 (Operation timed out)).
(17:22:48) npe_ [n=npe@32.97.110.64] entered the room.
(17:22:48) npe_ left the room (quit: Read error: 104 (Connection reset by peer)).
(17:25:14) npe left the room (quit: Connection timed out).
(17:31:59) npe [n=npe@32.97.110.64] entered the room.
(17:41:10) npe_ [n=npe@32.97.110.64] entered the room.
(17:41:10) npe_ left the room (quit: Connection reset by peer).
(17:48:18) npe left the room (quit: Read error: 110 (Connection timed out)).
(17:50:22) npe [n=npe@32.97.110.64] entered the room.
(17:50:22) npe left the room (quit: Connection reset by peer).
(17:59:33) npe [n=npe@32.97.110.64] entered the room.
(18:08:46) npe_ [n=npe@32.97.110.64] entered the room.
(18:12:23) npe_ left the room (quit: Read error: 60 (Operation timed out)).
(18:13:03) npe_ [n=npe@32.97.110.64] entered the room.
(18:13:03) npe_ left the room (quit: Read error: 104 (Connection reset by peer)).
(18:16:49) npe left the room (quit: Connection timed out).
(18:22:14) npe [n=npe@32.97.110.64] entered the room.
(18:31:25) npe_ [n=npe@32.97.110.64] entered the room.
(18:36:40) npe_ left the room (quit: Read error: 60 (Operation timed out)).
(18:37:25) npe_ [n=npe@32.97.110.64] entered the room.
(18:38:44) npe left the room (quit: Connection timed out).
(18:45:56) npe_ left the room (quit: Connection reset by peer).
(18:46:36) npe [n=npe@32.97.110.64] entered the room.
(18:55:47) npe_ [n=npe@32.97.110.64] entered the room.
(18:55:48) npe_ left the room (quit: Connection reset by peer).
(19:02:49) npe left the room (quit: Connection timed out).
(19:04:58) npe [n=npe@32.97.110.64] entered the room.
(19:08:53) npe left the room (quit: Read error: 60 (Operation timed out)).
(19:09:33) npe [n=npe@32.97.110.64] entered the room.
(19:18:47) npe_ [n=npe@32.97.110.64] entered the room.
(19:27:10) npe left the room (quit: Connection timed out).
(19:28:02) npe [n=npe@32.97.110.64] entered the room.
(19:32:50) npe left the room (quit: Read error: 60 (Operation timed out)).
(19:36:18) npe_ left the room (quit: Connection timed out).
(19:37:12) npe [n=npe@32.97.110.64] entered the room.
(19:41:23) npe left the room (quit: Operation timed out).
(19:42:04) npe [n=npe@32.97.110.64] entered the room.
(19:51:14) npe_ [n=npe@32.97.110.64] entered the room.
(19:58:20) npe left the room (quit: Connection timed out).
(22:53:59) uriel__ [n=uriel@84-216-52-47.sprayadsl.telenor.se] entered the room.
(22:56:18) uriel_ left the room (quit: Read error: 60 (Operation timed out)).
(00:12:22) npe_ left the room (quit: ).
(03:08:27) KillerX left the room (quit: ).
(04:13:30) underspecified left the room (quit: ).
(04:17:13) ooooo [i=none@201.80.219.33] entered the room.
(04:21:02) eno__ [n=eno@adsl-70-137-145-135.dsl.snfc21.sbcglobal.net] entered the room.
(04:32:29) eno left the room (quit: Read error: 110 (Connection timed out)).
(04:36:53) npe [i=npe@66.112.249.18] entered the room.
(06:53:53) npe left the room (quit: ).
(09:52:29) npe [n=npe@66.112.249.148] entered the room.
(10:03:48) npe left the room (quit: ).
(15:39:26) anothy left the room (quit: wolfe.freenode.net irc.freenode.net).
(15:39:44) anothy [n=a@c-98-221-104-176.hsd1.nj.comcast.net] entered the room.
(16:41:55) underspecified [n=eric@softbank220043052011.bbtec.net] entered the room.
(16:48:19) npe [n=npe@32.97.110.64] entered the room.
(16:49:53) KillerX [n=anant@gentoo/developer/KillerX] entered the room.
(16:57:31) npe_ [n=npe@32.97.110.64] entered the room.
(16:59:01) sqweek: home sweet home
(17:00:35) soul9: wb
(17:04:36) npe left the room (quit: Connection timed out).
(17:06:43) npe [n=npe@32.97.110.64] entered the room.
(17:13:54) npe_ left the room (quit: Connection timed out).
(17:15:57) npe_ [n=npe@32.97.110.64] entered the room.