Conversation with #inferno at Thu Nov 9 04:54:17 2017 on powerman@chat.freenode.net (irc) (04:54:17) #inferno: Topic for #inferno set by anth at 19:45:51 on 04/07/15 (04:59:19) darvon [~darvon@107.191.100.185] entered the room. (05:14:56) voxadam left the room (quit: Ping timeout: 250 seconds). (05:28:59) voxadam [~adam@fedora/voxadam] entered the room. (06:35:20) qrstuv left the room (quit: Remote host closed the connection). (08:42:28) qrstuv [~heh@50.44.163.72] entered the room. (10:22:12) rogpeppe1 [~rog@79-75-244-167.dynamic.dsl.as9105.com] entered the room. (20:00:52) voxadam left the room (quit: Quit: WeeChat 1.9.1). (01:30:56) leg [~jorge@190.106.51.215] entered the room. (03:06:14) voxadam [~adam@fedora/voxadam] entered the room. (03:06:28) voxadam left the room. (04:53:51) leg left the room (quit: Quit: Leaving.). (20:10:27) rogpeppe1 left the room (quit: Ping timeout: 246 seconds). (00:02:37) Fish- left the room (quit: Quit: page fault). (00:04:53) Fish [~Fish@cartwheel.9grid.fr] entered the room. (01:49:35) leg [~jorge@190.106.51.215] entered the room. (06:31:00) leetspete left the room (quit: Ping timeout: 246 seconds). (06:34:01) leetspete [~pete@cpe-98-148-93-186.socal.res.rr.com] entered the room. (06:36:32) leg left the room (quit: Quit: Leaving.). (12:42:07) sirjofri [~sirjofri@natfh58.fh-hof.de] entered the room. (13:58:41) sirjofri left the room (quit: Quit: leaving). (00:41:54) leg [~jorge@190.106.51.215] entered the room. (08:22:37) leg left the room (quit: Quit: Leaving.). (11:09:49) visof [~visof@unaffiliated/visof] entered the room. (16:47:37) rogpeppe1 [~rog@79-75-244-167.dynamic.dsl.as9105.com] entered the room. (18:15:16) visof left the room (quit: Ping timeout: 268 seconds). (19:01:30) leg [~jorge@190.106.51.215] entered the room. (19:07:01) rogpeppe1 left the room (quit: Ping timeout: 240 seconds). (20:21:30) visof [~visof@unaffiliated/visof] entered the room. (21:16:06) rogpeppe1 [~rog@79-75-244-167.dynamic.dsl.as9105.com] entered the room. (21:28:54) visof left the room (quit: Ping timeout: 268 seconds). (21:32:59) visof [~visof@41.69.6.15] entered the room. (21:32:59) visof left the room (quit: Changing host). (21:32:59) visof [~visof@unaffiliated/visof] entered the room. (21:37:57) visof left the room (quit: Ping timeout: 240 seconds). (22:56:46) leg left the room (quit: Quit: Leaving.). (23:05:31) leg [~jorge@190.106.51.215] entered the room. (23:22:19) leg left the room (quit: Quit: Leaving.). (01:27:27) rogpeppe1 left the room (quit: Ping timeout: 248 seconds). (07:10:03) fgudin left the room (quit: Ping timeout: 250 seconds). (07:13:50) fgudin [fgudin@odin.sdf-eu.org] entered the room. (10:27:53) rogpeppe1 [~rog@79-75-244-167.dynamic.dsl.as9105.com] entered the room. (20:37:22) leetspete1 left the room (quit: Ping timeout: 258 seconds). (20:38:15) leetspete left the room (quit: Ping timeout: 268 seconds). (20:48:57) leetspete [~pete@cpe-98-148-93-186.socal.res.rr.com] entered the room. (22:04:02) qrstuv left the room (quit: Ping timeout: 260 seconds). (22:08:01) leetspete left the room (quit: Ping timeout: 248 seconds). (22:19:17) leetspete [~pete@cpe-98-148-93-186.socal.res.rr.com] entered the room. (23:04:23) leetspete1 [~none@cpe-98-148-93-186.socal.res.rr.com] entered the room. (23:15:41) qrstuv [~heh@50.44.163.72] entered the room. (23:51:45) Id- left the room (quit: Ping timeout: 255 seconds). (23:51:45) horrified left the room (quit: Ping timeout: 255 seconds). (23:51:45) Hazelesque left the room (quit: Ping timeout: 255 seconds). (23:51:48) Hazelesque_ [~hazel@lopsa/member/hazelesque] entered the room. (23:52:12) horrified [89r@farlander.net] entered the room. (23:52:50) Id- [~Id-@dpwalters.powered.by.lunarbnc.net] entered the room. (18:10:39) anth_x [~a@pabetcvpn01b.synchronoss.com] entered the room. (19:00:24) that1guy [~glenda@129.138.6.16] entered the room. (20:00:01) that1guy left the room (quit: Ping timeout: 248 seconds). (20:43:21) anth_x left the room (quit: Quit: Leaving.). (22:02:22) that1guy [~glenda@129.138.6.16] entered the room. (22:05:34) that1guy: I was curious if anyone here has Inferno running on top of Plan9 amd64. 9front more specifically. (22:05:55) that1guy: The same question was asked on the mailing list almost a year ago but was never settled. (22:10:48) qrstuv: no (22:12:42) qrstuv: inferno more or less does not have code for running on the amd64 (22:13:30) qrstuv: linux and others can run 386 inferno binaries on amd64 processors (22:13:44) qrstuv: but 9front can't run 386 binaries on their amd64 kernel (22:15:35) qrstuv: adapting emu and libinterp for 64 bit pointers is doable, but more work than anyone feels like doing (22:19:19) that1guy: Ok thanks. I'm very new to inferno, I am running it on some Linux machines for the purpose of exporting the network to try and use in Plan9. Haven't had success yet (22:20:41) that1guy: but I was also curious about running a local browser with css, etc in Plan9 (22:21:50) that1guy: nvm that now though. (22:21:53) leetspete1: I think most of the 9front guys are doing that in an emulator. (22:23:48) leetspete1: that1guy: I think the mailing list nearly died when Charles moved it to Google. Not that it was super active. (22:24:04) leetspete1: But I've tried replying to stuff, email gets silently discarded by Google. (22:25:29) that1guy: Yeah I believe its called vmx. I have nothing against it. I just got excited about how light Inferno seemed to be. (22:26:24) qrstuv: vmx is pretty new (22:26:55) qrstuv: most people have two computers, or vnc to a second computer, or just don't use web browsers (22:27:53) qrstuv: inferno will not help you browse the web (22:28:22) leetspete1: Indeed. Needs the newer cyphers for HTTPS. (22:28:51) leetspete1: Plus the JS engine is a bit old, but I can't find a reasonable suite of headless tests. (22:30:18) rogpeppe [~rog@79-75-244-167.dynamic.dsl.as9105.com] entered the room. (22:30:25) rogpeppe1 left the room (quit: Ping timeout: 250 seconds). (22:30:33) leetspete1: I don't use charon much; I do all of my dev work in Inferno (judicious use of os(10) to actually run code), though. (22:31:18) leetspete1: Inferno: an entire OS with a GUI that takes fewer resources than vim. (22:34:38) that1guy: leetspete1: That was the point that I am trying to reach as well. I guess using vnc doesn't really break that notion at least for now. But I wanted more immersion into Plan9/Inferno (22:35:20) leetspete1: Yeah, the way I got it was they gave me a Mac at work and I just couldn't deal so I went all Inferno except for the browser. (22:37:35) leetspete1: Mycroftiv has been documenting his ANTS thing like crazy lately, if you'd prefer to live inside drawterm than inside Inferno: https://doc.9gridchan.org/blog/ (22:39:40) leetspete1: (It's not so well-organized, but somewhere in there is a step-by-step guide to setting up a 9front/ANTS system.) (22:41:06) leetspete1: The Inferno codebase could use some love, honestly. (22:42:03) qrstuv: some programs thrive on neglect and abuse (22:43:37) qrstuv: i was working on it, but i took a four month break and completely forgot how computers work (22:43:41) qrstuv: so progress has stalled (22:45:30) qrstuv: the classic inferno story (22:45:57) leetspete1: A story that has played out often. (22:47:14) leetspete1: I keep getting tied up in paying the bills. I had to do a 9P thing in Go recently, it's kind of appalling how obscure doing non-web stuff is in Go. (22:47:44) leetspete1: Like, just a reasonable 9P library took a while to find, and then it turned out not to be reasonable. (22:48:01) qrstuv: i've only used lionkov and ron's (22:48:06) qrstuv: it wasn't too good (22:49:04) leetspete1: I tried out Docker's. Theirs is terrible, it seems like they have only used it to do occasional Linux filesystems. (22:49:27) qrstuv: what simple fool did you hoodwink into paying you to do 9p? (22:50:38) leetspete1: Haha, it was ancillary monitoring work, under the radar. (22:51:28) leetspete1: A small thing to monitor some specific bits of Redis for a project for the company that we rent our office space from. (22:51:57) because[m] [becausewaf@gateway/shell/matrix.org/x-jyxwvbcwahcwfqvd] entered the room. (22:52:14) leetspete1: I wanted to do it in Tk instead of a web thing. (22:53:18) leetspete1: That's also on the list of stuff Inferno stuff that could use some hacking. Unpatched Tk on Inferno is nearly impossible to see on a laptop, so I hacked that a little. (22:54:10) qrstuv: i'd like to throw all that crap out (22:54:16) qrstuv: there's way too much c code (22:55:37) leetspete1: That's reasonable. (22:55:39) qrstuv: but most of it can't be done with dis because the interpreter is too slow and the compilers are too dumb (22:56:10) qrstuv: which is what i was working on before real life interrupted me (22:56:40) leetspete1: I think you'd mentioned attempting a new compiler. That was you? It's been a long time. (22:57:02) qrstuv: it has indeed been a long time (22:57:16) qrstuv: whenever i really get going on it, something else comes up (22:58:03) leetspete1: Are your notes or any code up somewhere public? (22:58:18) qrstuv: no (22:58:25) qrstuv: it's not useful at the moment (22:59:25) qrstuv: so far i've written a toy control flow analyser and optimizer and halfway rewrote the 386 compiler as a limbo program (22:59:46) leetspete1: Ah, cool. (23:00:11) qrstuv: i wouldn't say that (23:00:59) qrstuv: dis instructions aren't well suited for optimizing, so i have to come up with a microcode and then compile that (23:01:34) leetspete1: Yeah, it seems like there are some bits that could have been reworked. (23:01:41) qrstuv: i'm only moving the existing compiler to dis to make testing easier (23:02:38) qrstuv: memory allocations, frees, increfs, and decrefs are all implicit in instructions that do actual work (23:03:43) qrstuv: so eliminating reference counts and allocations from loops isn't possible when optimizing only dis instructions (23:04:15) leetspete1: Ouch. (23:04:25) leetspete1: Is there a good reference besides xec.c? (23:04:27) qrstuv: slow slicing and array accesses are the main thing slowing down graphics processing (23:04:34) qrstuv: no (23:05:12) leetspete1: I suspected as much. (23:05:26) leetspete1: (The "no", not the big slowdown.) (23:06:56) qrstuv: so the idea is for the optimizer to read a dis program, convert it to a micro-dis program, optimize it, and then the compiler turn that into 386 program (23:11:46) leetspete1: Interesting! (23:12:35) rogpeppe left the room (quit: Ping timeout: 240 seconds). (23:13:49) qrstuv: as it is now, the optimizer turns dis programs into a mess of graphviz, and the compiler turns a useless subset of dis programs into 386 programs (23:33:09) qrstuv: http://www.9paste.net/qrstuv/tree/comp/ (23:33:11) qrstuv: http://www.9paste.net/qrstuv/tree/comp/?tar (23:36:31) that1guy left the room (quit: Ping timeout: 248 seconds). (00:15:31) leetspete1: qrstuv: Awesome! (00:17:20) leetspete1: Thanks. Looking over it. (00:37:16) rogpeppe [~rog@79-75-244-167.dynamic.dsl.as9105.com] entered the room. (01:12:41) qrstuv left the room (quit: Ping timeout: 240 seconds). (01:52:47) rogpeppe left the room (quit: Ping timeout: 250 seconds). (02:15:57) qrstuv [~heh@50.44.163.72] entered the room. (10:09:36) rogpeppe [~rog@79-75-244-167.dynamic.dsl.as9105.com] entered the room. (17:02:18) horrified left the room (quit: Quit: quit). (17:57:03) rogpeppe left the room (quit: Remote host closed the connection). (18:04:12) rogpeppe [~rog@79-75-244-167.dynamic.dsl.as9105.com] entered the room. (18:17:07) horrified [89r@farlander.net] entered the room. (23:27:04) rogpeppe left the room (quit: Ping timeout: 268 seconds). (23:27:08) rogpeppe1 [~rog@79-75-244-167.dynamic.dsl.as9105.com] entered the room. (00:43:27) The account has disconnected and you are no longer in this chat. You will automatically rejoin the chat when the account reconnects.