arno11 | Wizzup: btw did you find time for serial ir-tx debug ? that's not urgent but i'd like to finish pierogi port properly :P | 15:56 |
---|---|---|
Wizzup | arno11: I can set it up later tonight, say 1900 or so (in 3.5 hours). I was just finishing up setting up the maemo part of my desk :D | 16:29 |
* Wizzup eyes the serial device | 16:29 | |
arno11 | ok cool :) | 16:31 |
Wizzup | arno11: 30-50 mins and I'll set it up fyi | 19:26 |
arno11 | ok | 19:31 |
Wizzup | hooking it up nopw. | 21:06 |
Wizzup | now | 21:06 |
arno11 | oki | 21:14 |
Wizzup | hard to get n900 on it properly ;) | 21:17 |
arno11 | :) | 21:19 |
Wizzup | ok | 21:19 |
Wizzup | Starting kernel ... | 21:19 |
Wizzup | let's see | 21:19 |
arno11 | cool | 21:19 |
Wizzup | [ 23.331420] rc rc0: Unspecified device as /devices/platform/n900-ir/rc/rc0 | 21:20 |
Wizzup | [ 23.332000] rc rc0: lirc_dev: driver ir_rx51 registered at minor = 0, no receiver, raw IR transmitter | 21:20 |
arno11 | i know this error | 21:20 |
Wizzup | Ah, I don't think it's an error | 21:21 |
Wizzup | I'm just sharing the progress. | 21:21 |
Wizzup | since the boot takes a while ;) | 21:21 |
arno11 | yes indeed just a message | 21:21 |
arno11 | yes | 21:21 |
Wizzup | so what command do I run again? | 21:21 |
arno11 | let me check | 21:21 |
sicelo | ir-ctl -S rc5:0x1e01 | 21:22 |
arno11 | yes | 21:22 |
arno11 | :) | 21:23 |
Wizzup | -bash: ir-ctl: command not found | 21:23 |
Wizzup | let me get it | 21:23 |
sicelo | provided by v4l-utils | 21:23 |
Wizzup | maemo decided to run apt worker first ;) | 21:24 |
arno11 | lol | 21:24 |
arno11 | the old fremantle tweak works to block it btw | 21:25 |
Wizzup | I like it :P | 21:25 |
Wizzup | it's been a long time wish for me to extend it to work for all packages too | 21:25 |
Wizzup | not just the ones with X-Maemo-etc | 21:25 |
arno11 | ok | 21:26 |
Wizzup | got a trace | 21:26 |
Wizzup | https://bpa.st/UAOFS | 21:27 |
Wizzup | so the first thing that happens after ir-ctl is: | 21:28 |
Wizzup | rc_core: Loaded IR protocol module ir-rc5-decoder, but protocol rc-5 still not available | 21:28 |
Wizzup | and after that there is an oops | 21:28 |
arno11 | rc5 is supposed to be available, weird | 21:29 |
arno11 | difficult to see something relevant | 21:30 |
freemangordon | arno11: tonegend | 21:32 |
arno11 | freemangordon: yes i found it :) | 21:32 |
freemangordon | https://github.com/maemo-leste/maemo-multimedia-tone-generator | 21:33 |
arno11 | yep | 21:33 |
freemangordon | I wonder why PA is so slow in resampling 8<->48 | 21:34 |
freemangordon | it should be optimized, in theory | 21:34 |
arno11 | i think that's because it is doing 8 to 48 to 8 | 21:35 |
arno11 | the modem accept only 8 and 16 | 21:35 |
freemangordon | ok, but fremantle is doing the same, no? | 21:35 |
arno11 | nope | 21:36 |
freemangordon | how's that? | 21:36 |
Wizzup | I think we have problems reading 8Kz from the mic | 21:36 |
freemangordon | it is the same modem | 21:36 |
Wizzup | probably in fremantle they just do that | 21:36 |
freemangordon | no | 21:36 |
arno11 | Wizzup: yes | 21:36 |
freemangordon | wait, why is 48<->8 code in fremantle then? | 21:36 |
arno11 | in fremantle th trick is to record at 48 | 21:36 |
arno11 | and then downsampling to 8 | 21:37 |
arno11 | in leste we record at 4 to avoid crash and then 4 to 48 | 21:37 |
freemangordon | crash? | 21:37 |
freemangordon | some kernel bug? | 21:38 |
arno11 | yes we can't record at 8 | 21:38 |
arno11 | none issue with PA | 21:38 |
Wizzup | freemangordon: do you have any idea about the stack trace above? | 21:38 |
Wizzup | we don't really know how to read it, as per usual :) | 21:38 |
freemangordon | ok, but can;t we do the same in leste? | 21:38 |
freemangordon | Wizzup: sec | 21:38 |
arno11 | we can do the same in leste but with a module like pulseaudio-nokia lol | 21:39 |
freemangordon | wait, we can;t tell PA to resmple? | 21:40 |
freemangordon | *resample | 21:40 |
arno11 | it does everything automatically | 21:40 |
arno11 | excepting if we use a specific module for that | 21:40 |
freemangordon | hmm | 21:41 |
arno11 | even if you ask PA to resample differently, daemon.conf will probably cancel it | 21:43 |
arno11 | we have the example with latency: impossible to set it manually | 21:44 |
freemangordon | Wizzup: may I have tho .ko file? | 21:45 |
freemangordon | *the | 21:45 |
freemangordon | ir_rx51.ko that is | 21:45 |
Wizzup | from my n900? | 21:46 |
freemangordon | mhm | 21:46 |
Wizzup | sure, let me get it | 21:46 |
freemangordon | so, "but protocol rc-5 still not available" is harmless? | 21:46 |
Wizzup | don't know | 21:47 |
arno11 | probably not | 21:47 |
arno11 | because even using other tools it crashes | 21:47 |
arno11 | like lirc or pierogi | 21:47 |
Wizzup | so is this a missing config? | 21:47 |
Wizzup | freemangordon: https://wizzup.org/is-rx51.ko | 21:48 |
arno11 | misunderstanding: rc5 not available is not a problem | 21:50 |
freemangordon | Wizzup: 404 | 21:50 |
freemangordon | wget https://wizzup.org/is-rx51.ko gives that | 21:50 |
Wizzup | probaly unix perms | 21:51 |
Wizzup | sec | 21:51 |
Wizzup | oh | 21:51 |
Wizzup | https://wizzup.org/ir-rx51.ko | 21:51 |
freemangordon | ok now | 21:51 |
Wizzup | arno11: do you need me to test anything else now? | 21:55 |
arno11 | i think eveything else is ok ;) tyvm | 21:56 |
freemangordon | Wizzup: arno11: is it possible that we have issue with sections: https://github.com/maemo-leste/droid4-linux/blob/master/drivers/media/rc/ir-rx51.c#L171 | 22:04 |
freemangordon | device private is a static struct | 22:04 |
freemangordon | that should be devm_kzalloc()-ed I think | 22:05 |
freemangordon | like here https://github.com/maemo-leste/droid4-linux/blob/master/drivers/media/rc/tango-ir.c#L178 | 22:06 |
Wizzup | seems likely given https://github.com/maemo-leste/droid4-linux/commit/a92def1becf33e91fc460c7ae575aa9210ba8f40 | 22:08 |
Wizzup | " | 22:08 |
Wizzup | Compile tested only. | 22:08 |
Wizzup | nightmare commits for niche projects :P | 22:08 |
freemangordon | well, if you have kernel build around, we can easily test | 22:09 |
freemangordon | just remove global struct declaration and allocate it with devm_kzalloc | 22:09 |
freemangordon | that is a pretty trivial | 22:10 |
freemangordon | fix | 22:10 |
arno11 | for you lol | 22:10 |
freemangordon | ok, I can provide you with a patch | 22:10 |
freemangordon | tomorrow :)O | 22:10 |
arno11 | arghhhh :) | 22:11 |
freemangordon | BTW, we are using the same tree for n900 and d4, right? | 22:11 |
freemangordon | arno11: so,, just came | 22:11 |
freemangordon | oops | 22:11 |
freemangordon | that was meant to be "sorry, just came home..." | 22:12 |
freemangordon | :) | 22:12 |
arno11 | ok no probs :) | 22:12 |
freemangordon | arno11: do you have machine you can build kernel on? | 22:25 |
arno11 | this evening no, i have a baby instead of the machine... | 22:27 |
freemangordon | heh | 22:28 |
freemangordon | ok :) | 22:28 |
freemangordon | will try to provide a patch tomorrow, maybe Wizzup will be able to test it | 22:28 |
arno11 | i can try tomorrow if you want | 22:29 |
freemangordon | ok | 22:29 |
arno11 | thx anyway | 22:30 |
sicelo | thanks for checking fmg :-) | 22:44 |
sicelo | i am able to build kernel tonight | 22:44 |
arno11 | nice !!! | 22:45 |
freemangordon | sicelo: I will provide patch tomorrow. build-tested though | 22:59 |
freemangordon | or, feel free to fix it by yourself | 22:59 |
Generated by irclog2html.py 2.17.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!