| 00:07.40 | *** join/#card10badge mmu_man (~revol@37.164.196.113) |
| 00:08.42 | *** join/#card10badge theAntimon (~matze@p54B09494.dip0.t-ipconnect.de) |
| 00:19.01 | *** join/#card10badge mmu_man_ (~revol@81.185.167.177) |
| 00:50.09 | frigginglorious | schneider: just writing python. whats a good way to open a serial connection from macOS? |
| 01:02.50 | M^[m] | frigginglorious: most people i've seen just use screen, it's pretty straightforward. you should be able to just plug in your running card10, make the connection and then ctrl+c into a prompt from within screen. |
| 01:03.05 | *** join/#card10badge theAntimon (~matze@p54B09494.dip0.t-ipconnect.de) |
| 01:07.40 | *** join/#card10badge ebi (~quassel@x4d0714f0.dyn.telefonica.de) |
| 01:08.29 | *** join/#card10badge mkie (~quassel@x4d0714f0.dyn.telefonica.de) |
| 03:19.29 | *** join/#card10badge PyroPeter (~pyropeter@april-fools/2013/tenth/pyropeter) |
| 03:30.49 | *** join/#card10badge user_51_ (~quassel@92.117.134.64) |
| 03:51.31 | *** join/#card10badge verin0x5 (~verin0x@x590e681c.dyn.telefonica.de) |
| 04:00.30 | *** join/#card10badge Stormwind_mobile (~Stormwind@dslb-084-062-109-126.084.062.pools.vodafone-ip.de) |
| 05:20.50 | *** join/#card10badge DocScrutinizer05 (~saturn@openmoko/engineers/joerg) |
| 08:15.17 | *** join/#card10badge mmu_man (~revol@9.164.185.81.rev.sfr.net) |
| 08:59.14 | *** join/#card10badge mntl2 (~mntl@clark.uberspace.de) |
| 08:59.21 | *** join/#card10badge vista (~vista@notx.ml) |
| 08:59.25 | *** join/#card10badge beneth (~beneth@lxcb.beneth.fr) |
| 08:59.40 | *** join/#card10badge verin0x9 (~verin0x@x590e681c.dyn.telefonica.de) |
| 09:00.35 | *** join/#card10badge mini0x00 (~minixor@ns2.kernelpanic.com.ar) |
| 09:00.52 | *** join/#card10badge Entroacc1ptor (~mad@mad.unserver.de) |
| 09:00.53 | *** join/#card10badge Rahix_ (~Rahix@rahix.de) |
| 09:01.20 | *** join/#card10badge lilafisch (~lilafisch@irc.xtort.eu) |
| 09:01.21 | *** join/#card10badge maxammann_ (~maxammann@v37187.1blu.de) |
| 09:05.13 | *** join/#card10badge atluxity (~atluxity@178.63.28.77) |
| 09:07.04 | *** join/#card10badge Twix (~None@twix.ws) |
| 09:38.36 | *** join/#card10badge mmu_man (~revol@37.167.179.15) |
| 09:48.18 | Stormwind_mobile | schneider: is it possible to write background programs in python, kind of a daemon or TSR, or do in need to dig into the driver level in C to be able to do that? |
| 09:48.52 | Stormwind_mobile | Also, where in the code is the part that defines the power off sequence? |
| 10:32.11 | *** join/#card10badge Rev_Illo (~revillo@unaffiliated/rev-illo/x-3122184) |
| 11:00.21 | rahix[m] | Stormwind_mobile: epicardium/modules/pmic.c near the end |
| 11:01.18 | rahix[m] | The current firmware architecture does not allow for background applications unfortunately |
| 11:06.45 | Stormwind_mobile | One (or both?) cores run a RTOS, right? Shouldn't that allow running user tasks? |
| 11:18.51 | *** join/#card10badge theAntimon (~matze@p54B09494.dip0.t-ipconnect.de) |
| 11:35.03 | *** join/#card10badge mmu_man_ (~revol@37.167.179.15) |
| 11:42.18 | *** join/#card10badge mmu_man (~revol@37.167.179.15) |
| 11:46.10 | *** join/#card10badge xdecimal (~Android@197.210.58.129) |
| 11:47.33 | *** join/#card10badge mmu_man_ (~revol@37.167.179.15) |
| 11:51.10 | *** join/#card10badge Rev_Illo (~revillo@unaffiliated/rev-illo/x-3122184) |
| 11:59.53 | *** join/#card10badge mmu_man (~revol@37.167.179.15) |
| 12:23.46 | *** join/#card10badge mmu_man (~revol@37.167.179.15) |
| 12:52.36 | DocScrutinizer05 | I'd assume RTOS supports preemptive multitasking |
| 12:53.11 | DocScrutinizer05 | some sort of, at least |
| 12:53.41 | DocScrutinizer05 | https://www.freertos.org/implementation/a00005.html |
| 12:54.33 | *** join/#card10badge theAntimon (~matze@p54B09494.dip0.t-ipconnect.de) |
| 12:56.40 | DocScrutinizer05 | if there's any viable mechanism for IPC is another question, so maybe your background task has a hard time making the epi|py-cardium tasks change their behavior in any way triggered or determined by your daemon |
| 12:59.50 | *** join/#card10badge mmu_man (~revol@252.172.185.81.rev.sfr.net) |
| 12:59.53 | DocScrutinizer05 | but you might save the complete context of the running (pycardium) task, take over display, LEDs etc to accomplish your thing while you lock/forbid scheduling, then after you did your thing you restore context and scheduling and go to sleep. Should work |
| 13:01.01 | DocScrutinizer05 | oops you'd need to save and restore context and stop/restart tasks on both cores |
| 13:05.51 | DocScrutinizer05 | Stormwind_mobile: the tricky part is you can't use the existing IPC between epicardium and pycardium, but on epi-core you probably simply could suspend the epicardium process so pycardium would just wait forever (or actually until epicardium task resumes) for a reply to any epicardium API call, while your process may take over control of hardware on epi-core |
| 13:08.13 | DocScrutinizer05 | after restoring epicardium context and handing over to the original epi-process everything should be back to normal unless queues got clogged |
| 13:22.16 | *** join/#card10badge theAntimon (~matze@p54B09494.dip0.t-ipconnect.de) |
| 13:23.37 | Rahix | DocScrutinizer05: I'd prefer if you leave answering such questions to people who are more familiar with the firmware internals. A lot of what you just said is incorrect for the architecture our firmware currently has |
| 13:24.35 | Rahix | Stormwind_mobile: Only Epicardium runs on FreeRTOS and yes, you can easily add new tasks to it. There are quite a few actually running already: https://firmware.card10.badge.events.ccc.de/overview.html#epicardium |
| 13:25.47 | Rahix | What I was referring to is the ability for user-code (that is, python apps or l0dables) to run concurrently, which is not possible as there is no RTOS on core 1 |
| 13:27.24 | Rahix | As for the API: You can easily use the API in Epicardium as well. There is no need for any additional synchronization here. |
| 13:27.52 | DocScrutinizer05 | Rahix: except for IPC my comments were generic and not related to the card10 firmware, but I agree an answer by somebody knowing the firmware internals would be of much higher value than my generic musing |
| 13:36.36 | *** join/#card10badge theAntimon (~matze@p54B09494.dip0.t-ipconnect.de) |
| 13:37.17 | DocScrutinizer05 | factoid for me to keep in mind: core1 (I stop calling it epicardium) already supports and uses preemptive scheduling, which is a great thing to have |
| 13:50.12 | DocScrutinizer05 | s/core1/core0/ |
| 13:51.21 | M^[m] | that's precisely why i still call it epic :) |
| 13:52.28 | Stormwind_mobile | I wish there was more documentation about firmware and hardware, especially simple "this plugs into these things" mindmap like stuff |
| 13:54.31 | Stormwind_mobile | I asked already a while ago: can anyone/approved contributors edit the wiki, tutorials and other docs? There are still tutorials with "TODO" parts and referring to states of things pre or during camp, pictures based on alpha hardware,... |
| 13:54.40 | *** join/#card10badge LostInWeb (~quassel@2a03:4000:15:1f2::9) |
| 14:09.06 | Rahix | You can edit the wiki yourself! Just fork the repo and send a MR |
| 14:43.38 | coon | is it possible to run java script node on the card10 badge? |
| 14:43.53 | coon | *node -> code |
| 14:44.26 | DocScrutinizer05 | TIL: no queues in API ("The API is strictly synchroneous."), according to docs. So what was that "my queues got filled" error report of a while ago? |
| 14:45.18 | Rahix | coon, we don't have any javascript interpreter in the firmware, so no |
| 14:45.58 | coon | can I use c code then or am I forced to use micropython? |
| 14:46.08 | Rahix | C code works fine (called l0dables) |
| 14:46.20 | Rahix | Take a look into the l0dables folder of the firmware for an example |
| 14:46.20 | DocScrutinizer05 | wait, how does "strictly sync" apply to callback functions and streams? |
| 14:46.27 | coon | so it works like on the r0ket / rad1o badge? |
| 14:46.34 | Rahix | similar, yes |
| 14:46.38 | coon | ok thanks :) |
| 14:47.07 | DocScrutinizer05 | checks if there actually are any callbacks |
| 14:47.08 | Rahix | coon: The C API is documented here: https://firmware.card10.badge.events.ccc.de/epicardium/api.html |
| 14:47.48 | Rahix | DocScrutinizer05: There are no callback functions and streams are read using a dedicated `read` api call (epic_stream_read to be exact) |
| 14:49.22 | Rahix | The error you got were most likely some sensors data buffers overrunning |
| 14:55.57 | DocScrutinizer05 | Rahix: many thanks! |
| 14:56.52 | DocScrutinizer05 | not me, some other user in here who told about card10 getting dog slow with sensors, so yes, prolly sensor buffer queues |
| 14:58.54 | Stormwind_mobile | Are l0adables capable of the same hardware access like the OS/drivers, or more like micropython, but in another language? |
| 14:59.38 | DocScrutinizer05 | l0adables run on core1/Pyc, no? |
| 15:00.16 | DocScrutinizer05 | so I'd assume the latter is correct |
| 15:00.55 | DocScrutinizer05 | aiui all hw access from epic/core0 |
| 15:01.52 | DocScrutinizer05 | interesting question what happens if a l0adable goes direct hw access from core1 |
| 15:02.29 | Stormwind_mobile | Btw: epicardium is probably not "epic" (though a fun abbreviation), but epi cardium, (wrapped) around the heart/core |
| 15:03.05 | Stormwind_mobile | DocScrutinizer05: BSOD |
| 15:03.11 | DocScrutinizer05 | hehe |
| 15:04.24 | DocScrutinizer05 | is eager to hear Rahix' qualified statement regarding l0adables accessing hw |
| 15:04.55 | DocScrutinizer05 | can't be a great idea anyway |
| 15:06.09 | DocScrutinizer05 | particularly when there is a chance for resource allocation collisions |
| 15:06.42 | Rahix | Just write a driver in Epicardium instead, it is definitely the saner solution |
| 15:06.52 | DocScrutinizer05 | :-) |
| 15:07.49 | *** join/#card10badge theAntimon (~matze@p54B09494.dip0.t-ipconnect.de) |
| 15:07.52 | DocScrutinizer05 | is there any hw resource arbitration / muxing between the two cores or is it a first come first lose aka last write sticks scheme? |
| 15:09.35 | DocScrutinizer05 | or let me reword that: is there any addr space translation also applicable to peripherals' registers that is specific to cores |
| 15:10.52 | DocScrutinizer05 | the doc's addr map didn't mention anything like that |
| 15:11.19 | DocScrutinizer05 | s/addr/memory/ |
| 15:12.28 | Stormwind_mobile | Rahix: are drivers also tasks? |
| 15:12.39 | DocScrutinizer05 | dedicated addr ranges for cores under "usage" but no overlaps or different addr levels for same physical memory |
| 15:13.14 | Rahix | Stormwind_mobile: for the most part, each driver also has a task but some don't |
| 15:13.26 | DocScrutinizer05 | Stormwind_mobile: I seen a BM160 driver https://git.card10.badge.events.ccc.de/card10/firmware/blob/master/epicardium/modules/bhi.c#L419 task, so some are |
| 15:14.01 | Rahix | Here is the overview ... file:///home/rahix/Documents/Development/Embedded/card10/firmware/Documentation/output/overview.html#epicardium |
| 15:14.25 | DocScrutinizer05 | sorry BHI160 |
| 15:28.51 | Stormwind_mobile | Rahix: so a background task that e.g.reads accelerometer and gyro values, processes those and and looks for a certain combination of thresholds, if found triggering a wakeup event, would need to go into epicardium as task |
| 15:29.16 | Stormwind_mobile | Has anyone used the gyroscope after all? |
| 15:29.34 | Rahix | yes |
| 15:29.37 | Rahix | and yes |
| 15:30.54 | Rahix | For gesture detection (I guess?) you might want to use the second accelerometer which apparently already has some stuff like this implemented in hardware |
| 15:31.00 | Rahix | That would be the BMA400 |
| 15:39.19 | *** join/#card10badge theAntimon (~matze@p54B09494.dip0.t-ipconnect.de) |
| 15:40.30 | Stormwind_mobile | Recognition of the typical "turn wrist and raise arm to look at the watch, then turn it on" would be the first thing that comes to my mind. |
| 15:41.49 | Stormwind_mobile | Other ideas would be something like recognition of prolonged vibration = bicycle or car riding on rough terrain. |
| 15:43.55 | Stormwind_mobile | Or try to implement a virtual touch input: imagine a grid of e.g. 2*4 or 3*5 squares on the screen. tapping on each of them ought to result in a discernible reading of accelerometer and gyroscope |
| 15:44.44 | Stormwind_mobile | I definitely read about someone proposing this, if not even implemented and written down in a paper. Have to look that up again. |
| 15:46.12 | Stormwind_mobile | Also tapping on the edges of the PCB should make for an alternative way of input. Which I think could prove very nice, considering the finicky hardware switches that will eventually break, if not already broken. |
| 15:46.29 | Stormwind_mobile | I think I heard a few complaints about broken buttons already. |
| 16:48.46 | *** join/#card10badge Stormwind_mobile (~Stormwind@dslb-084-062-109-126.084.062.pools.vodafone-ip.de) |
| 16:54.25 | *** join/#card10badge Stormwind_mobile (~Stormwind@dslb-084-062-109-126.084.062.pools.vodafone-ip.de) |
| 17:27.25 | *** join/#card10badge theAntimon (~matze@p54B09494.dip0.t-ipconnect.de) |
| 18:02.35 | *** join/#card10badge Stormwind_mobile (~Stormwind@dslb-084-062-109-126.084.062.pools.vodafone-ip.de) |
| 18:03.11 | *** join/#card10badge mmu_man (~revol@vaf26-2-82-244-111-82.fbx.proxad.net) |
| 18:13.25 | *** join/#card10badge theAntimon (~matze@p54B09494.dip0.t-ipconnect.de) |
| 18:36.18 | *** join/#card10badge infobot (ibot@c-174-52-60-165.hsd1.ut.comcast.net) |
| 18:36.18 | *** topic/#card10badge is Logix (Wiki): https://card10.badge.events.ccc.de/ git: https://git.card10.badge.events.ccc.de/card10 % current version: https://card10.badge.events.ccc.de/en/firmwareupdate/ % Display: https://www.aliexpress.com/item/4000018872259.html |
| 18:50.50 | *** join/#card10badge bionade24 (~oskar@server2.oscloud.info) |
| 18:57.39 | *** join/#card10badge bionade24 (~oskar@server2.oscloud.info) |
| 19:00.09 | *** join/#card10badge erlehmann (~erlehmann@46.114.32.163) |
| 19:04.24 | *** join/#card10badge bionade24 (~oskar@server2.oscloud.info) |
| 19:07.57 | *** join/#card10badge bionade24 (~oskar@server2.oscloud.info) |
| 19:15.09 | *** join/#card10badge erlehmann_ (~erlehmann@46.114.38.181) |
| 19:27.52 | *** join/#card10badge erlehmann_ (~erlehmann@46.114.36.188) |
| 19:30.53 | *** join/#card10badge erlehmann (~erlehmann@46.114.37.182) |
| 19:43.41 | *** join/#card10badge frigginglorious (~Thunderbi@71-82-147-112.dhcp.eucl.wi.charter.com) |
| 19:44.08 | *** join/#card10badge erlehmann (~erlehmann@46.114.37.182) |
| 20:29.39 | *** join/#card10badge erlehmann (~erlehmann@46.114.37.182) |
| 20:32.56 | *** join/#card10badge erlehmann_ (~erlehmann@46.114.36.71) |
| 20:40.46 | *** join/#card10badge kloenk (~quassel@mail.kloenk.de) |
| 21:10.40 | *** mode/#card10badge [+o Rahix] by ChanServ |
| 21:44.03 | *** join/#card10badge theAntimon (~matze@pD95D351B.dip0.t-ipconnect.de) |
| 21:49.00 | *** join/#card10badge artemis[m] (artemisasr@gateway/shell/matrix.org/x-xssjaqukxgkineze) |
| 21:49.22 | *** join/#card10badge erlehmann_ (~erlehmann@46.114.32.57) |
| 21:54.07 | *** join/#card10badge Benjojo (~sid4563@srv-69.nsa.me.uk) |
| 22:05.47 | *** join/#card10badge erlehmann_ (~erlehmann@46.114.38.188) |
| 22:29.09 | *** join/#card10badge derpeter (~derpeter@derpeter.net) |
| 22:30.33 | *** join/#card10badge erlehmann_ (~erlehmann@46.114.36.9) |
| 22:32.21 | *** join/#card10badge msvb-mob (~msvb-lab@monero/hardware/michael) |
| 23:11.57 | *** join/#card10badge erlehmann_ (~erlehmann@46.114.32.222) |
| 23:28.50 | *** join/#card10badge frigginglorious (~Thunderbi@71-82-147-112.dhcp.eucl.wi.charter.com) |
| 23:32.39 | *** join/#card10badge erlehmann (~erlehmann@46.114.32.222) |