--- Log for 10.06.114 Server: leguin.freenode.net Channel: #rockbox --- Nick: logbot Version: Dancer V4.16 Started: 28 days and 20 hours ago 00.07.04 Quit lebellium (Quit: ChatZilla 0.9.90.1 [Firefox 30.0/20140605174243]) 00.17.04 Quit preglow (*.net *.split) 00.17.30 *** Saving seen data "./dancer.seen" 00.28.10 Join preglow [0] (~thomj@2001:840:4243:3::100) 00.35.53 Quit bertrik (Remote host closed the connection) 00.44.42 Join Misanthropos [0] (~Misanthro@frnk-4d01eb3a.pool.mediaways.net) 00.51.16 Quit fs-bluebot (Ping timeout: 240 seconds) 00.52.46 Quit bluebrother^ (Ping timeout: 252 seconds) 00.54.08 Join fs-bluebot [0] (~fs-bluebo@g224239249.adsl.alicedsl.de) 00.54.45 Join bluebrother [0] (~dom@rockbox/developer/bluebrother) 00.54.50 Quit ender` (Quit: Smoking is one of the leading causes of statistics.) 01.26.07 Join ygrek [0] (~user@108.59.6.97) 01.31.37 Quit ZincAlloy (Quit: Leaving.) 01.46.56 Quit ygrek (Remote host closed the connection) 01.55.19 Join RMCHD [0] (~RMCHD@a89-154-230-205.cpe.netcabo.pt) 01.56.22 Part RMCHD ("Closing Window") 02.03.11 Quit Misanthropos (Quit: Ex-Chat) 02.17.33 *** Saving seen data "./dancer.seen" 02.21.19 Quit RiD (Quit: A good plan today is better than a perfect plan tomorrow.) 02.34.53 Quit evilnick (Ping timeout: 240 seconds) 02.35.42 Quit gevaerts (Ping timeout: 276 seconds) 02.55.31 Join evilnick [0] (~evilnick@rockbox/staff/evilnick) 02.57.46 Join gevaerts [0] (~fg@rockbox/developer/gevaerts) 03.00.00 Quit AlexP (Remote host closed the connection) 03.30.00 Quit Rower (Quit: Hmmm...) 04.17.35 *** Saving seen data "./dancer.seen" 04.19.30 # soap: you shoulda pinged me in here instead of trying in the forums :/ 04.25.55 Quit pixelma (Disconnected by services) 04.25.56 Join pixelma_ [0] (pixelma@rockbox/staff/pixelma) 04.25.58 Nick pixelma_ is now known as pixelma (pixelma@rockbox/staff/pixelma) 04.26.05 Join amiconn_ [0] (quassel@rockbox/developer/amiconn) 04.26.12 Quit amiconn (Disconnected by services) 04.28.48 Quit the-kyle (Remote host closed the connection) 05.06.58 Join the-kyle [0] (~kyle@kyle.tk) 05.07.13 Join steffengy [0] (~quassel@p57B48CF9.dip0.t-ipconnect.de) 05.07.48 Quit steffengy1 (Ping timeout: 276 seconds) 05.44.47 Quit TheSeven (Ping timeout: 260 seconds) 05.46.30 Join TheSeven [0] (~quassel@rockbox/developer/TheSeven) 06.06.37 Join pamaury [0] (~quassel@rockbox/developer/pamaury) 06.07.18 Join Strife89 [0] (~Strife89@adsl-98-80-200-103.mcn.bellsouth.net) 06.17.38 *** Saving seen data "./dancer.seen" 06.36.51 Quit pamaury (Ping timeout: 245 seconds) 07.12.37 Quit Strife89 (Ping timeout: 240 seconds) 07.16.31 Join pamaury [0] (~quassel@rockbox/developer/pamaury) 07.26.33 Join [Saint] [0] (~saint@rockbox/staff/saint) 07.35.15 # <[Saint]> Hmmmmm. 07.35.57 # <[Saint]> 8 core accounts is the apparent limit for quasselcore on a raspi if you want it to be tolerably responsive. 07.50.18 Join fragilematter [0] (~fragilema@unaffiliated/fragilematter) 08.15.48 Join ygrek [0] (~user@108.59.6.97) 08.17.41 *** Saving seen data "./dancer.seen" 08.26.01 Quit tertu (Ping timeout: 245 seconds) 08.29.23 Join ender` [0] (krneki@foo.eternallybored.org) 08.36.02 Quit [Saint] (Quit: Quit.) 08.53.54 Nick amiconn_ is now known as amiconn (quassel@rockbox/developer/amiconn) 08.55.57 Join tertu [0] (~quassel@143.44.65.14) 08.58.38 Join Zagor [242] (~bjst@rockbox/developer/Zagor) 09.00.30 Quit tertu (Ping timeout: 276 seconds) 09.00.34 Join tertu_ [0] (~quassel@143.44.65.14) 09.09.29 Join petur [0] (~petur@rockbox/developer/petur) 09.23.21 Join [Saint] [0] (~saint@rockbox/staff/saint) 09.33.23 Quit [Saint] (Quit: Quit.) 09.44.15 Join [Saint] [0] (~saint@rockbox/staff/saint) 09.48.03 Quit [Saint] (Client Quit) 09.50.42 Join [Saint] [0] (~saint@rockbox/staff/saint) 09.52.52 # <[Saint]> Woo. 09.53.34 # John, Woo? 09.54.10 # <[Saint]> Sorry - wrong channel. I'm far too excited about my minor accomplishment with an Arch image creation script. 09.54.32 # maybe other programs would be more performant on a raspi;) 09.55.44 # <[Saint]> This is an STE Snowball. 09.56.11 # <[Saint]> 1.2GH dual core, 1GB RAM, 8GB NAND, etc. 09.56.25 # <[Saint]> (IOW: takes a fat shit on a raspi) 09.56.37 # <[Saint]> ...but I digress, and this is off-topic. 09.57.55 # <[Saint]> I was playing around with a very similar setup on a raspi earlier, but, it just doesn't cut it compared to the Snowball. 09.58.02 # <[Saint]> anyhoo - topic. 10.01.44 # Topic's dead, baby, topic's dead. 10.03.19 Join kugel [0] (~kugel@avm-guido.avm.de) 10.03.19 Quit kugel (Changing host) 10.03.19 Join kugel [0] (~kugel@rockbox/developer/kugel) 10.08.52 Join einhirn [0] (~Miranda@bsod.rz.tu-clausthal.de) 10.17.45 *** Saving seen data "./dancer.seen" 10.18.48 Join wodz [0] (~wodz@iwl138.internetdsl.tpnet.pl) 10.23.17 # <[Saint]> Who's motorcycle is this? 10.24.06 # :) 10.36.57 Quit ygrek (Ping timeout: 265 seconds) 10.45.03 Quit jhMikeS (Ping timeout: 260 seconds) 10.51.14 Join b0hoon [0] (4e1e727f@gateway/web/freenode/ip.78.30.114.127) 10.57.30 Quit b0hoon (Ping timeout: 246 seconds) 10.58.41 Join [Saint_] [0] (~saint@rockbox/staff/saint) 10.58.44 Quit [Saint] (Read error: Connection reset by peer) 11.22.37 Quit [Saint_] (Read error: Connection reset by peer) 11.22.43 Join [Saint] [0] (~saint@rockbox/staff/saint) 11.24.34 Part [Saint] 11.25.17 Quit GodEater (Ping timeout: 240 seconds) 11.25.56 Join [Saint] [0] (~saint@rockbox/staff/saint) 11.28.37 # Zagor: are you going to be present in Warsaw? 11.29.37 # wodz: yes! I'm arriving with LO456 at 08:55 on saturday 11.29.50 Quit copper (Quit: ZNC - http://znc.in) 11.30.18 # cool 11.30.38 Join GodEater [0] (~whoknows@my83-216-94-31.mynow.co.uk) 11.30.38 Quit GodEater (Changing host) 11.30.38 Join GodEater [0] (~whoknows@rockbox/staff/GodEater) 11.35.27 Join copper [0] (~copper@unaffiliated/copper) 11.35.51 Quit GodEater (Ping timeout: 276 seconds) 11.57.30 Quit Cinos (Ping timeout: 240 seconds) 12.13.24 Join Cinos [0] (Cinos@cinos.biz) 12.17.47 *** Saving seen data "./dancer.seen" 13.06.28 # Build Server message: 3New build round started. Revision bc37665, 253 builds, 31 clients. 13.10.07 # Build Server message: 3Build round completed after 221 seconds. 13.51.05 Quit tertu_ (Read error: Connection reset by peer) 14.12.27 Quit [Saint] (Ping timeout: 252 seconds) 14.17.48 *** Saving seen data "./dancer.seen" 14.55.29 Join amayer [0] (~amayer@mail.weberadvertising.com) 15.02.15 Join ygrek [0] (~user@108.59.6.97) 15.02.23 Join [Saint] [0] (~saint@rockbox/staff/saint) 15.38.55 Quit wodz (Quit: Leaving) 15.44.20 Quit [Saint] (Read error: Connection reset by peer) 15.45.04 Join [Saint] [0] (~saint@rockbox/staff/saint) 15.56.19 Quit kugel (Quit: leaving) 16.08.07 Join RiD [0] (RiD@bl22-226-139.dsl.telepac.pt) 16.17.49 *** Saving seen data "./dancer.seen" 16.32.53 Quit igitoor (Ping timeout: 252 seconds) 16.34.02 Join igitoor [0] (igitur@2a00:d880:3:1::c1ca:a648) 16.37.52 # JdGordon, ping you? forums? 16.45.59 Quit igitoor (Changing host) 16.45.59 Join igitoor [0] (igitur@unaffiliated/contempt) 16.58.29 Quit Zagor (Quit: Clint excited) 17.12.32 Quit mortalis (Ping timeout: 276 seconds) 17.15.18 Quit einhirn (Quit: Miranda IM! Smaller, Faster, Easier. http://miranda-im.org) 17.22.33 Quit RiD (Quit: A good plan today is better than a perfect plan tomorrow.) 17.23.13 Join Rower [0] (husvagn@h176n2-aeg-a11.ias.bredband.telia.com) 17.46.03 Join AlexP [0] (~alex@rockbox/staff/AlexP) 17.46.55 Quit petur (Quit: *plop*) 17.48.09 # pamaury: seems like you can't disable EP0 on this USB core (it will never signal completion) 17.48.28 # that caused a nasty lockup with the new driver 17.48.40 # just something to keep in mind in case we run into it again 17.48.48 # TheSeven: I fail to see the link between the two (disable EP0 and signal completion) 17.49.00 # ah got it, it will never signal completion of disabling EP0 17.49.05 # yes 17.49.12 # that IRQ will never come 17.49.15 # but I knews that already, I think it's written somewhere on some manuals 17.49.34 # at least it says the disable bit of EP0 is not implemented or something similar 17.50.27 # yes, that was just something that I messed up while writing the glue to get the new driver compile inside rockbox 17.50.50 # nasty indeed, is it working now ? 17.51.15 # well, now the EP0 state machine panics for some reason 17.51.30 Join goa [0] (~light@ANancy-258-1-88-13.w90-13.abo.wanadoo.fr) 17.51.37 # but it seems like what the driver is doing is correct, just the state machine seems to be confused somehow 17.51.44 # likely a rather highlevel bug that's easy to nail down 17.52.33 # I can't really tell anything about whether this driver works better than the old one before we get to handle situations with concurrent TX traffic on multiple EPs 17.52.42 # but I'm hoping to get there within a few days now 17.52.50 Quit goa (Client Quit) 17.53.27 # steffengy is currently looking into sorting this out as a little excercise to get into kernel-level development ;) 17.57.07 # it's quite possible the multi-traffic thing or bad fifo setup is really the killer 17.58.34 # * pamaury needs to go 17.59.44 # pamaury: it's quite promising that I haven't seen anything that would point to multi-traffic problems on emcore or umsboot though, while the old umsboot (using rockbox's driver) definitely had trouble in that area 18.00.02 # indeed 18.00.26 Join jhMikeS [0] (~jethead71@rockbox/developer/jhMikeS) 18.00.32 # now thinking about it, we might have the same problem on rk27xx iirc, it's a completely different core though 18.01.16 # typical symptoms are lockups with HID activity during storage transfers, and failing to mount on windows (in many cases even without hid) 18.01.28 # and dmesg messages about bus resets on linux 18.01.41 # on rk27xx if you spam EP0 while UMS is running, you can get to kill it 18.01.48 # at least that's what I remember 18.01.58 # but there is no equivalent fifo setup 18.02.01 # try spamming it with hid traffic, that's the easiest way to test 18.07.51 Quit pamaury (Ping timeout: 252 seconds) 18.17.50 *** Saving seen data "./dancer.seen" 18.20.31 Join RiD [0] (RiD@2.83.226.139) 18.20.40 Quit Rower (Quit: Hmmm...) 18.38.07 Join bertrik [0] (~quassel@rockbox/developer/bertrik) 18.41.16 Quit ygrek (Ping timeout: 240 seconds) 18.41.17 Quit shamus (Ping timeout: 240 seconds) 19.23.23 Quit bertrik (Ping timeout: 252 seconds) 19.23.36 Join bertrik [0] (~quassel@ip117-49-211-87.adsl2.static.versatel.nl) 19.23.36 Quit bertrik (Changing host) 19.23.36 Join bertrik [0] (~quassel@rockbox/developer/bertrik) 19.26.39 Join Strife89 [0] (~Strife89@2602:306:bce1:8c20:dcde:1e23:608c:9b1e) 19.30.36 Join lebellium [0] (~chatzilla@89-93-178-161.hfc.dyn.abo.bbox.fr) 19.33.31 Join pamaury [0] (~quassel@rockbox/developer/pamaury) 19.45.12 Quit pamaury (Quit: No Ping reply in 180 seconds.) 19.46.31 Join pamaury [0] (~quassel@rockbox/developer/pamaury) 19.49.53 Quit pamaury (Client Quit) 19.55.36 Join pamaury [0] (~quassel@rockbox/developer/pamaury) 20.01.54 Quit pamaury (Quit: No Ping reply in 180 seconds.) 20.06.52 Join pamaury [0] (~quassel@rockbox/developer/pamaury) 20.14.18 Join rela [0] (~x@pdpc/supporter/active/rela) 20.17.53 *** Saving seen data "./dancer.seen" 20.49.33 Quit rela (Read error: Connection reset by peer) 21.03.50 Quit pamaury (Ping timeout: 265 seconds) 21.04.02 Join pamaury [0] (~quassel@rockbox/developer/pamaury) 21.25.20 Quit model (Read error: Connection reset by peer) 21.25.21 Quit model|gone (Read error: Connection reset by peer) 21.47.51 Quit pamaury (Ping timeout: 252 seconds) 21.50.45 Join ZincAlloy [0] (~Adium@pD9EEAC05.dip0.t-ipconnect.de) 21.53.22 Quit amayer (Remote host closed the connection) 21.53.23 Join shamus [0] (~shmaus@ip-206-192-193-180.marylandheights.ip.cablemo.net) 21.54.50 Join amayer [0] (~amayer@mail.weberadvertising.com) 21.59.29 Quit Strife89 (Ping timeout: 240 seconds) 22.03.06 Join rela [0] (~x@pdpc/supporter/active/rela) 22.17.55 *** Saving seen data "./dancer.seen" 22.33.17 Quit K1773R (*.net *.split) 22.33.19 Quit zu_ (*.net *.split) 22.33.29 Join zu [0] (~zu@ks387228.kimsufi.com) 22.34.49 Join K1773R [0] (~K1773R@unaffiliated/k1773r) 22.45.01 Join mc2739_ [0] (~mc2739@rockbox/developer/mc2739) 22.45.04 Quit mc2739 (Disconnected by services) 22.45.06 Nick mc2739_ is now known as mc2739 (~mc2739@rockbox/developer/mc2739) 22.51.03 # hm, this EP0 state machine in usb-s3c6400x is broken beyond repair ;) 22.51.38 Join wodz [0] (~wodz@89-75-149-112.dynamic.chello.pl) 22.52.26 # TheSeven: Do I understand correctly that you are working on new usb driver for n2g/classic? 22.52.47 # yes 22.52.55 Join Strife89 [0] (~Strife89@204.116.245.116) 22.53.20 # we're not quite sure what's broken with the old one, but it somehow doesn't work right 22.53.35 # that is something I know :-) 22.53.42 # and as the one we use in emcore (which is completely different) seems to work much better, steffengy and me are porting it to rockbox 22.54.21 Quit RiD (*.net *.split) 22.54.23 Quit Galois (*.net *.split) 22.54.26 # * wodz crosses fingers 22.54.46 # (it was originally ported to emcore from some other project on a completely different SoC) 22.59.29 # atj213x interrupt controller is strange. There seems to be no way to clear pending interrupt. I have stalled watchdog interrupt even if I mask this source and disable watchdog all together. If I don't service it the pending bit for watchdog is still present on the next irq. 23.00.02 # reading pending interrupt register doesn't clear it, writing to it also 23.01.31 # so that bit is completely stuck to 1? 23.02.14 # masking often doesn't affect the pending register, just controls which pending bits actually trigger a CPU-level IRQ 23.02.38 # and disabling a peripheral doesn't necessarily clear any pending IRQs from that 23.03.02 # yeah. That looks like this. But how am I supposed to clear pending interrupt then? 23.03.03 # you'll probably have to clear the IRQ at the source somehow, and possibly also clear it in the pending reg 23.03.34 # as long as it's masked you just shouldn't need to care about it being set in the pending reg though, it shouldn't actually cause a CPU IRQ 23.03.43 # Yes, clearing at peripherial level is the last thing I didn't try yet 23.03.57 # that's usually required to make it disappear from the pending reg 23.06.07 # it definitely doesn't cause 'irq storm' so it is not pushed to cpu level irq (or is pushed to the masked irq level - I didn't check) 23.08.35 Join RiD [0] (RiD@2.83.226.139) 23.08.35 Join Galois [0] (djao@efnet.math.uwaterloo.ca) 23.09.24 # I know that, in general, some interrupt bits can be edge-triggered and other level-triggered, if it's level-triggered you can't clear it unless you clear it at the source 23.20.36 Quit wodz (Quit: Leaving) 23.24.59 Quit JdGordon (Ping timeout: 276 seconds) 23.25.19 Quit rela (Read error: Connection reset by peer) 23.26.33 Join JdGordon [0] (~jonno@ppp118-209-108-2.lns20.mel4.internode.on.net) 23.26.33 Quit JdGordon (Changing host) 23.26.33 Join JdGordon [0] (~jonno@rockbox/developer/JdGordon) 23.26.55 # looks like we're surviving enumeration now \o/ 23.30.10 Quit amayer (Quit: Leaving) 23.46.26 Quit Strife89 (Ping timeout: 276 seconds) 23.50.22 # gevaerts: I've spotted some very suspiciously looking code in the USB core 23.51.15 Quit lebellium (Quit: ChatZilla 0.9.90.1 [Firefox 30.0/20140605174243]) 23.51.49 # it seems like usb_core_control_request_handler reallocates all non-EP0 endpoints every time it's run, before the device has an address 23.51.56 # can you explain the point of that? 23.52.32 # IIUC this should be done on SET_CONFIGURATION requests 23.52.54 # hm, I guess you need the EP numbers for the descriptors before that 23.53.55 # but I don't get why, if you're only doing this before SET_ADDRESS, you're redoing it on every request 23.54.10 # probably not broken, but it seems fishy somehow... 23.56.03 # Hmmm 23.56.11 # * gevaerts tries to remember the order of operations here... 23.57.00 # what happens on the bus is basically: RESET, GET_DEVICE_DESCRIPTOR, SET_ADDRESS, (move state to addressed), GET_CONFIGURATION_DESCRIPTOR, ..., SET_CONFIGURATION, (start using the endpoints) 23.57.07 # * gevaerts nods 23.57.16 # So it gets done twice 23.57.37 # so that if usb_state == DEFAULT will happen on GET_DEVICE_DESCRIPTOR and SET_ADDRESS AFAICS 23.57.45 # Yes, that block could have a separate flag, or usb_state could get an extra state 23.57.58 # or maybe this could just be moved into init 23.57.59 # Well 23.58.02 # Hmmmm 23.58.12 # I'm trying to remember what happens on bus reset 23.58.42 # I think this might have been implemented this way with dynamically changing configurations in mind 23.58.50 # No 23.59.31 # hm, then I really don't see a reason to not do this in init, as everything required by it should be constant (set of available endpoints, set of enabled drivers)