Conversation with #inferno at Mon Jan 7 18:09:28 2013 on powerman@irc.freenode.net (irc) (18:46:52) ravelont [~ravelont@187-127-85-204.user.veloxzone.com.br] entered the room. (18:53:03) ravelont is now known as raphaelsc (19:58:06) raphaelsc left the room (quit: Quit: Leaving). (20:13:18) vsrinivas [U2FsdGVkX1@batman.acm.jhu.edu] entered the room. (22:22:43) ravelont [~ravelont@187-127-85-204.user.veloxzone.com.br] entered the room. (22:27:44) ravelont left the room (quit: Quit: Leaving). (22:28:07) ravelont [~ravelont@187-127-85-204.user.veloxzone.com.br] entered the room. (23:11:10) Fish left the room (quit: Read error: Connection reset by peer). (23:13:58) Fish [~Fish@9fans.fr] entered the room. (23:18:53) Fish left the room (quit: Read error: Connection reset by peer). (23:20:20) Fish [~Fish@9fans.fr] entered the room. (23:22:11) Fish left the room (quit: Read error: Connection reset by peer). (23:22:59) Fish [~Fish@9fans.fr] entered the room. (23:25:38) Fish left the room (quit: Read error: Connection reset by peer). (23:26:47) Fish [~Fish@9fans.fr] entered the room. (23:28:24) Fish left the room (quit: Read error: Connection reset by peer). (23:29:34) Fish [~Fish@9fans.fr] entered the room. (23:51:21) Fish left the room (quit: Quit: WeeChat 0.3.9.2). (23:51:29) Fish [~Fish@9fans.fr] entered the room. (23:56:33) Fish left the room (quit: Read error: Connection reset by peer). (23:57:59) Fish [~Fish@9fans.fr] entered the room. (00:01:09) Fish left the room (quit: Read error: Connection reset by peer). (00:02:30) Fish [~Fish@9fans.fr] entered the room. (00:03:28) ravelont left the room (quit: Quit: Leaving). (00:36:49) ravelont [~ravelont@187-127-85-204.user.veloxzone.com.br] entered the room. (01:13:31) yshurik left the room (quit: Quit: Leaving.). (02:50:04) ravelont is now known as raphaelsc (03:29:03) doublec: leetspete: I get a segfault running wm/wm on emu-sdl on 64bit gentoo fyi (03:49:29) leetspete: doublec: brother could ye spare a coredump (04:03:45) doublec: leetspete: sure, I'm just about to try running it through the debugger (04:04:59) doublec: leetspete: http://pastebin.com/jcE99unN (04:10:25) doublec: leetspete: if I move SDL_Init from sdl_proc and put it in 'main' then things work (04:10:52) doublec: I assume some thread/stack thing is going wrong (04:12:37) doublec: leetspete: my hacky workaround: http://cd.pn/emu.patch (04:26:57) leetspete: doublec: Huh...pulseaudio. (04:27:12) leetspete: Wonder if it hates getting calls from non-main threads? (04:27:26) leetspete: I don't have a pulseaudio driver backing my SDL libs. (04:29:48) doublec: yeah that could be it (04:30:19) leetspete: doublec: Do you still get the segfault if you export SDL_AUDIODRIVER=dsp ? (04:32:17) doublec: No I don't get the segfault then (04:32:25) doublec: I did get an assertion though when exiting (04:32:29) doublec: but it's intermittent (04:33:08) doublec: leetspete: http://pastebin.com/DpnVxuDw is assertion on exit (04:34:13) leetspete: doublec: I should probably figure out what to do about that. atexit() doesn't work inside emu, ensuring SDL_Quit() gets called is somewhat tricky. (04:34:26) leetspete: Or it's really simple and I just don't know. (04:34:31) doublec: this says not to call SDL video functions on anything but the main thread: http://wiki.libsdl.org/moin.cgi/FAQDevelopment (04:38:23) leetspete: As long as it's always the same thread that called SDL_Init(), it is supposedly fine per the authors. (04:38:38) leetspete: On the other hand, I have no idea about how that works with PulseAudio. (04:39:51) leetspete: doublec: I'll need a bit of time to investigate, but extreme thanks. (04:40:55) doublec: np (04:41:49) doublec: SDL_AUDIODRIVER=alsa works too (04:41:55) doublec: Possibly a pulseaudio bug (04:53:00) doublec: leetspete: valgrind says possible stack overflow. Do inferno threads runs with a smaller stack? (04:53:54) leetspete: doublec: Ah, that makes sense. They do indeed, depending on the platform. (04:55:35) doublec: leetspete: that was it (04:55:53) doublec: in emu/port/kproc-pthreads.c it sets a stack size (04:56:06) doublec: leetspete: if I remove that then it runs fine so I assume it's setting it too small (04:56:11) leetspete: And PulseAudio happens to trigger this. (04:56:36) doublec: PulseAudio is using iconv_open for logging and it calls a function that uses a bit of stack unfortunately (04:57:31) leetspete: (╯°□°)╯︵ ┻━┻ (04:58:09) leetspete: I should probably actually just not enable sound until I write a sound driver for Inferno. (04:58:27) leetspete: And SDL_Net probably shouldn't get initialized, either. (04:58:36) leetspete: But the thread stack issue, that's sticky. (05:05:33) leetspete: Hm...the KPX11 flag to kproc() gives the thread extra stack... (05:07:15) leetspete: Heh, for the X thread's stack, Inferno allocates 256k. (05:07:47) doublec: No doubt they hit a similar issue :) (05:07:52) leetspete: I'd not seen this: CTRLC = 'C'-'@' (05:08:11) leetspete: That's clever and easy to remember. '@' happens to be 64 in ASCII. (05:08:38) doublec: heh (05:30:56) raphaelsc left the room (quit: Remote host closed the connection). (06:11:29) doublec left the room (quit: Quit: ircfs!). (06:14:29) doublec [~none@unaffiliated/doublec] entered the room. (06:27:45) doublec left the room (quit: Remote host closed the connection). (06:32:19) doublec [~none@unaffiliated/doublec] entered the room. (08:09:44) vsriniva1 [U2FsdGVkX1@batman.acm.jhu.edu] entered the room. (08:14:49) doublec left the room (quit: *.net *.split). (08:14:49) vsrinivas left the room (quit: *.net *.split). (08:21:34) rogpeppe left the room (quit: Ping timeout: 252 seconds). (10:02:33) rogpeppe [~rog@host-92-30-129-4.as13285.net] entered the room. (10:11:22) yshurik [~Adium@193.69.63.210] entered the room. (10:19:23) `jpi left the room (quit: Ping timeout: 255 seconds). (10:24:41) `jpi [~jpi@unaffiliated/pseud0cod3r] entered the room. (10:44:02) doublec [~none@unaffiliated/doublec] entered the room. (14:53:11) yshurik1 [~Adium@193.69.63.210] entered the room. (14:53:11) yshurik left the room (quit: Read error: Connection reset by peer). (16:19:30) yshurik1 left the room (quit: Quit: Leaving.). (17:03:10) yshurik [~Adium@223.65.202.84.customer.cdi.no] entered the room. (17:23:59) yshurik left the room (quit: Quit: Leaving.). (17:39:33) yshurik [~Adium@223.65.202.84.customer.cdi.no] entered the room. (17:41:57) yshurik left the room (quit: Client Quit). (18:27:03) The account has disconnected and you are no longer in this chat. You will automatically rejoin the chat when the account reconnects.