*** chbae has quit IRC | 00:09 | |
*** chbae has joined #automotive | 00:10 | |
*** chbae has quit IRC | 00:29 | |
*** JordonWu has joined #automotive | 00:31 | |
*** chbae has joined #automotive | 00:37 | |
*** mdurnev has joined #automotive | 02:25 | |
*** altendky has quit IRC | 02:31 | |
*** mkelly has joined #automotive | 03:00 | |
*** mkelly has quit IRC | 04:02 | |
*** AlisonChaiken has joined #automotive | 04:19 | |
*** _corrupt has quit IRC | 04:36 | |
*** _corrupt has joined #automotive | 04:58 | |
*** kooltux_ has joined #automotive | 06:04 | |
*** bartekt has joined #automotive | 06:11 | |
*** idonob has quit IRC | 06:43 | |
*** e8johan has joined #automotive | 06:48 | |
*** claneys has joined #automotive | 07:11 | |
*** rajm has joined #automotive | 07:13 | |
*** idonob has joined #automotive | 07:26 | |
*** bruce has joined #automotive | 07:42 | |
*** bruce is now known as ctbruce | 07:42 | |
*** toscalix has joined #automotive | 07:48 | |
*** dunderroffe_ has joined #automotive | 08:00 | |
*** dunderroffe has quit IRC | 08:03 | |
*** mfritzsch has joined #automotive | 08:07 | |
*** chbae has quit IRC | 08:12 | |
*** zeenix has joined #automotive | 08:37 | |
*** fagan has joined #automotive | 08:45 | |
jbocklage | Chron: indeed | 08:48 |
---|---|---|
*** jbocklage has quit IRC | 09:01 | |
*** jbocklage has joined #automotive | 09:02 | |
*** azanosch has joined #automotive | 09:12 | |
*** User3209 has joined #automotive | 09:15 | |
jbocklage | https://www.automotivelinux.org/software/agl-specification : spec link is broken | 09:18 |
*** dunderroffe_ has quit IRC | 09:19 | |
*** Chron has quit IRC | 09:51 | |
*** jbocklage123 has joined #automotive | 09:52 | |
*** jbocklage has quit IRC | 09:54 | |
*** fredw has joined #automotive | 10:06 | |
*** dunderroffe_ has joined #automotive | 10:15 | |
*** fredw has quit IRC | 10:28 | |
*** dunderroffe_ has quit IRC | 11:09 | |
*** fredw has joined #automotive | 11:12 | |
*** altendky has joined #automotive | 11:34 | |
*** dunderroffe has joined #automotive | 11:50 | |
*** wesam has joined #automotive | 12:14 | |
*** JordonWu has quit IRC | 12:35 | |
*** jbocklage123 has quit IRC | 13:00 | |
*** jbocklage has joined #automotive | 13:01 | |
*** e8johan has quit IRC | 13:09 | |
*** fredw has quit IRC | 13:25 | |
*** heeen has joined #automotive | 13:30 | |
*** fredw has joined #automotive | 13:57 | |
*** fredw has quit IRC | 14:01 | |
*** chbae has joined #automotive | 14:05 | |
*** fredw has joined #automotive | 14:08 | |
*** AlisonChaiken has quit IRC | 14:12 | |
*** nisha_m has joined #automotive | 14:16 | |
*** mdurnev has quit IRC | 14:26 | |
*** ctbruce has quit IRC | 14:31 | |
*** jobol has joined #automotive | 14:40 | |
*** chbae has quit IRC | 14:49 | |
*** jlrmagnus has joined #automotive | 14:53 | |
*** azanosch has quit IRC | 15:00 | |
*** araujo has joined #automotive | 15:08 | |
*** araujo has joined #automotive | 15:09 | |
*** mfritzsch has quit IRC | 15:19 | |
*** rajm has quit IRC | 15:48 | |
*** zeenix has quit IRC | 15:58 | |
*** claneys has quit IRC | 16:15 | |
*** toscalix has quit IRC | 16:23 | |
*** martinkelly has quit IRC | 16:33 | |
*** martinkelly has joined #automotive | 16:35 | |
*** AlisonChaiken has joined #automotive | 16:41 | |
mranostay | dl9pf_: somehow there is a dependency loop in the qemu image build now | 16:42 |
mranostay | dl9pf_: nm mdp mentioned something :) | 16:46 |
mdp | heh | 16:47 |
leon-anavi-cloud | hi | 16:49 |
*** martinkelly has quit IRC | 17:03 | |
*** martinkelly has joined #automotive | 17:22 | |
*** kooltux_ has quit IRC | 17:26 | |
*** platicus1 has joined #automotive | 17:34 | |
*** platicus1 has quit IRC | 17:41 | |
*** platicus1 has joined #automotive | 17:43 | |
*** platicus1 has quit IRC | 17:47 | |
*** fredw has quit IRC | 17:48 | |
*** jlrmagnus has quit IRC | 17:56 | |
*** platicus has quit IRC | 18:06 | |
*** jaragunde{M} has quit IRC | 18:06 | |
*** jbocklage has quit IRC | 18:07 | |
*** jaragunde{M} has joined #automotive | 18:08 | |
*** platicus1 has joined #automotive | 18:08 | |
*** jaragunde{M} has quit IRC | 18:10 | |
*** jaragunde{M} has joined #automotive | 18:11 | |
*** platicus has joined #automotive | 18:11 | |
*** jlrmagnus has joined #automotive | 18:15 | |
fury | what other hardware can i test the bluetooth audio streaming on, just to see if it's something specific to my hardware? | 18:28 |
mdp | qemu is the best hardware! ;) | 18:29 |
mdp | a lot of the builds are broken at runtime atm | 18:29 |
fury | sounds good - how do i hook up the cinglink dongle to qemu? | 18:30 |
fury | also i can't seem to actually "run" it on qemu, it never brings up the home screen :o | 18:30 |
fury | apologies if this is all already somewhere - docs.automotivelinux.org was kinda light in regards to getting it up and running on qemu | 18:31 |
mdp | this is how I'm running atm, https://www.irccloud.com/pastebin/mMY5tcXY/ | 18:31 |
fury | i see a terminal log and then briefly the splash screen, then it goes back to a terminal log and periodically jumps up in cpu usage | 18:31 |
mdp | note that I use some pending (need rebase and workaround for dependency loop with the netboot feature) patches to whack the root device and spit out a .qcow2. | 18:32 |
fury | cinolink, my bad. my eyes are failing me :P silly underlines | 18:32 |
mdp | a2dp was working well for me with that config | 18:33 |
mdp | SCO, not so much ;) | 18:33 |
fury | that's ok, nobody talks on the phone anymore, this is 2017 | 18:33 |
mdp | fury, NOOOO!!! all this work for nothing?!?? ;) | 18:33 |
mdp | hehe | 18:34 |
mdp | fury: fwiw, it's on my list to update the vbox/qemu docs but second after some telephony stuff I'm finishing up | 18:36 |
fury | oh man. reminds me of all the pain i went through trying to shoehorn 2 in/2 out a2dp streams into a tiny cortex M4 based product, phone call support was one of the trickier things. then the project was canceled. i spent a good year and a half, 2 years or so on that | 18:36 |
mdp | standard industry practice | 18:36 |
fury | naturally :D | 18:36 |
mdp | most people are permitted one project to go to market in a lifetime ;) | 18:37 |
mdp | smurray/mranostay, did rpi3 work on your last builds? | 18:39 |
mranostay | mdp: yesterday it did | 18:39 |
mranostay | not sure if a repo sync today will :) | 18:40 |
mdp | heh | 18:40 |
fury | so i don't have a .qcow2, i'm guessing that's why my agl qemu setup doesn't go very far | 18:40 |
mdp | mranostay: I hesitate to recommend that because of the pain of the limited gpu memory :( | 18:41 |
mranostay | audio was an issue as well UI responsiveness being limited to 255M | 18:41 |
mdp | fury, https://gerrit.automotivelinux.org/gerrit/#/c/9029/ and https://gerrit.automotivelinux.org/gerrit/#/c/9031/ work for now if you setup for agl-demo, leave netboot off. | 18:42 |
mdp | or you can cherrypick those basic usb passthrough options and use the runqemu method | 18:43 |
smurray | mdp: rpi3 didn't boot for me the last couple of days, not figured out why yet | 18:46 |
mranostay | smurray: works for me (tm) | 18:47 |
smurray | mranostay: it worked until I synced on Sunday | 18:48 |
fury | ok cool, merged those things into my thingy (hope i did it right O_o) and started up another bitbake. thanks :D | 18:53 |
fury | such gui pleb tho, i had to install beyond compare to feel comfortable with resolving the merge conflict on the first one | 18:56 |
fury | haven't figured out how to do that from command line yet | 18:56 |
mdp | yeah, sorry I didn't have them reworked..but was fighting sw on real h/w and had to set that aside. | 18:57 |
fury | no worries :D didn't hurt me to freshen up on my git skills | 18:57 |
fury | i'm a software engineer and it's been so long since i committed actual software to a repo at work that i had to google it | 18:58 |
fury | i need to do some git repos for the little one off fun projects i'm doing with arduinos and pis and whatnot | 18:59 |
*** jlrmagnus has quit IRC | 18:59 | |
fury | prolly a bad idea not to, what if a fireball hits my office?! | 18:59 |
fury | got this in console after booting up qemu https://www.irccloud.com/pastebin/3OHZq2SO/ | 19:10 |
*** jlrmagnus has joined #automotive | 19:11 | |
fury | n/m, forgot to sudo it | 19:12 |
fury | on second thought, it's still not firing up the home screen | 19:17 |
fury | https://pastebin.com/zQGpJnz4 | 19:17 |
fury | rebooted it a couple times but no dice | 19:17 |
fury | looks like the fun parts are: line 515 and 1349 | 19:20 |
mdp | looks like you need -cpu .... +sse4.1 ..., https://www.irccloud.com/pastebin/fIqUxksa/ | 19:20 |
fury | ooh | 19:21 |
*** jobol has quit IRC | 19:22 | |
fury | yay! | 19:23 |
* mdp bounces | 19:23 | |
fury | added +sse4.1,+sse4.2,+popcnt and the homescreen works. squishy tho :D | 19:23 |
fury | thanks :) | 19:23 |
mdp | ha, cool | 19:23 |
*** wesam has quit IRC | 19:23 | |
mdp | I'm able to resize it on the fly with no problems with the gfx emulation and driver I selected. | 19:24 |
mdp | fwiw | 19:24 |
mdp | though I run i3 so I leave it squished out of the way since I don't like floating windows | 19:24 |
fury | cool, resized the window, now it's just fuzzy instead of squishy | 19:25 |
fury | hmm. was able to scan for and pair to my iPhone, but then all devices disappeared from the bluetooth settings screen, i can't connect to it manually from the iPhone either. weird | 19:25 |
mdp | uh oh | 19:28 |
mdp | mranostay: over here | 19:28 |
fury | https://pastebin.com/cPUxHXst | 19:29 |
mdp | mranostay: since you have ios, did you see anything like the above? | 19:29 |
mdp | not necessarily ios related, mranostay, but since you mentioned apple's crazy timestamping issue..I consider it suspect. | 19:29 |
mranostay | that was avrcp related... | 19:30 |
mdp | I figured, didn't know if you anything beyond that. | 19:31 |
mdp | plus, my qemu build is weeks old so my experience is irrelevant now | 19:31 |
mdp | fury, I suspect the new cpu features became required because they recently tweaked build options, fwiw. | 19:31 |
mranostay | lot of setting app changes got in.. so chance something got broken | 19:32 |
fury | ah | 19:33 |
mdp | come to think of it..I saw it once on yesterday's build I'm running on porter do the thing where the 2-3 devices disappear, list blanks, then two of them reappear. | 19:33 |
fury | same problem with an android phone btw | 19:33 |
mdp | but nothing like what you describe exactly | 19:33 |
mdp | give it a few months to stabilize again ;) | 19:33 |
mranostay | the ones that reappear are likely ones you are paired too | 19:34 |
mdp | yeah, it was an existing pairing | 19:34 |
fury | the list completely disappears after i try to pair or connect | 19:34 |
fury | i saw it say "AV connection" to the android phone shortly before it disappeared, then it disconnected everything | 19:34 |
mranostay | wouldn't it be out of discovery or scan mode if it is connected? | 19:34 |
mranostay | not sure how the bluetooth spec is | 19:34 |
mdp | I do get the thing consistently where connect is unresponsive...I do need to get some time to fix that UI for that. can never tell if it got my mouse event | 19:35 |
mdp | due to the 2-3 second pause before it switches to disconnect state | 19:35 |
*** akilaw has joined #automotive | 19:37 | |
mdp | fury, surprised it doesn't say Handsfree & AV | 19:37 |
akilaw | hi | 19:37 |
akilaw | can anyone help me to install bitbake on my ubuntu pc? | 19:37 |
*** idonob has quit IRC | 19:38 | |
fury | few more logs of it - https://pastebin.com/aRWVUJgs | 19:44 |
mranostay | fury: this porter or rpi3? | 19:50 |
fury | qemu | 19:51 |
mranostay | ah duh | 19:52 |
mranostay | it is in the logs | 19:52 |
fury | i can connect and play "fine" on porter, but it's playing at the wrong rate (slower than the incoming audio) | 19:52 |
fury | so it plays at a lower pitch and eventually builds up a long delay | 19:53 |
mdp | line 1933 smacks of a permissions problem | 19:54 |
* mdp is very punny today | 19:54 | |
fury | listening to a 1khz sine on my phone, sounds like it's playing at about 923hz | 19:54 |
*** idonob has joined #automotive | 19:55 | |
mdp | which then results of bluez deregistering from dbus | 19:55 |
fury | so smack is "smacking" bluetooth :D | 19:56 |
mdp | s/of/in | 19:57 |
*** akilaw has quit IRC | 20:00 | |
*** caiortp has joined #automotive | 20:01 | |
fury | here's logs from my porter, currently up to about a minute and a half delay, it's fun to see the media player screen on AGL flash by "Advertisement" and me actually hear it a couple minutes later | 20:10 |
fury | https://pastebin.com/Q7ngXi5S | 20:10 |
*** praneeth has quit IRC | 20:14 | |
*** praneeth has joined #automotive | 20:15 | |
fury | prolly should add a jira but so far i'm the only person to be experiencing the bluetooth playback rate issue on porter | 20:17 |
fury | so dunno if it's something i'm doing wrong or what | 20:17 |
fury | quick question, to rule out my hub as part of the issue, i just yanked it out and plugged the dongle back in directly to the porter...how do i start the bluetooth stuff back up from the porter? i guess it didn't auto do it on reconnect of the dongle | 20:19 |
fury | (and now i have no touchscreen because it was hooked up to the hub) | 20:20 |
mranostay | better to put it up there and j-s will close it if it is something someone has resolved | 20:20 |
mranostay | mdp: seems pulseaudio isn't starting up now | 20:27 |
mranostay | another race condition? :/ | 20:27 |
fury | can't get bluetooth fired back up after dongle unplug/replug, bluetoothctl says this: | 20:28 |
fury | [bluetooth]# discoverable on | 20:28 |
fury | Failed to set discoverable on: org.bluez.Error.Failed | 20:28 |
fury | [bluetooth]# power on | 20:28 |
fury | Failed to set power on: org.bluez.Error.Blocked | 20:28 |
fury | systemctl restart bluetooth didn't do the trick either | 20:28 |
mranostay | yeah not sure we tested hotplugging that | 20:28 |
fury | but it says it does register a2dp endpoints | 20:28 |
fury | ah | 20:28 |
mranostay | since in an application it wouldn't be | 20:29 |
fury | hmm... does it turn bluetooth on on reboot? | 20:29 |
fury | i can boot it up with just the bluetooth dongle attached | 20:29 |
mranostay | fury: you have open the settings app | 20:29 |
fury | oh ok | 20:29 |
fury | guess i need to hunt down another hub | 20:29 |
mranostay | there is a bit of a workaround in it.. | 20:29 |
fury | hot damn, right next to me | 20:29 |
fury | that never works in real life | 20:30 |
fury | alright, gonna try this other hub to rule out the usb3 hub as causing a problem with the bluetooth dongle or something, thne i'll file a jira issue | 20:31 |
fury | ehhhh | 20:32 |
fury | the gechic display REALLY does not like that hub | 20:33 |
fury | holy shit | 20:33 |
fury | noises | 20:33 |
fury | hmm | 20:33 |
fury | that was not good | 20:33 |
mdp | mranostay: be sure that if using qemu that you have set your pulseaudio passthrough sink/source to match a real sink/source name on your host...not what's in my script ;) | 20:33 |
mdp | unless you have the same jabra headset | 20:34 |
mranostay | ah | 20:34 |
mranostay | i have them commented out | 20:34 |
fury | whew. i thought from the sounds it was making that i had just magic smoked my gechic display :X | 20:34 |
fury | guess that hub does not have enough juice to power it from the usb port alone | 20:34 |
mranostay | mdp: systemctl start --user pulseaudio works though | 20:36 |
fury | well, guess it's not specific to the hub, same deal with this one | 20:36 |
mdp | dunno without a log ;) | 20:36 |
mdp | afk | 20:37 |
fury | https://jira.automotivelinux.org/browse/SPEC-602 | 20:56 |
*** caiortp has quit IRC | 21:12 | |
fury | weird, i was playing bluetooth on the porter and then the a2dp endpoints just randomly unregistered | 21:20 |
fury | now it's not letting me play a2dp anymore. tried turning bluetooth off and on again... | 21:25 |
fury | https://www.irccloud.com/pastebin/1DNG3Tpr/ | 21:25 |
fury | have i ever mentioned how much i love bluetooth? | 21:25 |
fury | ok, it's not just Bluetooth. just plugged in a flash drive (and had to reboot the hub and the porter cause apparently I can't hot plug flash drives either O_o) | 21:49 |
fury | and it's also playing those files too slow | 21:49 |
fury | woo, think i found a bug in lightmediascanner | 21:59 |
fury | i've got a few big files in a directory on this flash drive and it vomits, won't scan any more files on the drive | 21:59 |
*** nisha_m has quit IRC | 22:20 | |
mranostay | fury: let me know if you find a bug in that. since i kinda own that part of AGL | 22:29 |
fury | sounds good | 22:42 |
fury | mranostay: here's the lightmediascanner having a problem with some large files on my flash drive https://www.irccloud.com/pastebin/6h7rARJT/ | 22:44 |
fury | i'm thinking it's a bug because why would it care about a .gz00 file, i'd think it'd only scan files it knew how to play | 22:44 |
mranostay | yeah it shouldn't be doing that at all | 22:45 |
mranostay | there is a patchset we have in agl to support scanning ogg headers correctly with metadata.. other than that it would what is in the lms project | 22:46 |
mranostay | fury: we only support ogg at this time | 22:46 |
fury | i played some wav files, don't have any ogg handy | 22:47 |
fury | figured from SPEC-312 that MP3 audio was intentionally not included due to patent | 22:48 |
mranostay | that would be my guess | 22:49 |
fury | upstream bug i guess - i can try to track it down and submit a patch if they have an active github project or something like that, if you'd like...either way, i'm eager to help :D | 22:50 |
mranostay | fury: yeah the lightmediascanner project is pretty dead | 22:50 |
mranostay | but really no other light weight media scanner it seems | 22:51 |
mranostay | fury: is there a home.tar.gz and does it try to scan that? | 22:51 |
fury | there is a .tar.xz in the folder but only 139 megs | 22:52 |
fury | the .tar.gz is split into 3 parts, 00 01 and 02 | 22:52 |
fury | 00 and 01 are 4,194,304,000 bytes long | 22:53 |
fury | gz00, gz01, gz02 | 22:53 |
fury | i had to back up the home folder of a linux laptop before i wiped it, just in case there was something there it turns out i needed, but the only thing i had handy was a flash drive that's fat32 formatted, so it couldn't write a larger file than 4 gigs | 22:53 |
mranostay | fury: seems it is failing in the stat() call from that log | 22:56 |
mranostay | fury: suspect it is a mix of a large file and 32-bit build.. | 22:58 |
fury | http://stackoverflow.com/questions/13893580/calling-stat-from-sys-stat-h-faills-with-value-too-large-for-defined-data-typ | 22:58 |
fury | sounds about right | 22:58 |
fury | it is larger than 2,147,483,647 bytes | 22:59 |
fury | cortex a15 is only 32 bit? | 22:59 |
fury | weird that stat would be using a signed data type for that tho | 22:59 |
fury | if it was using unsigned it'd be fine | 23:00 |
*** chbae has joined #automotive | 23:00 | |
mranostay | yeah it is | 23:00 |
fury | oh well. enough fun for today :D cya tomorrow | 23:02 |
mranostay | yeah it is a long type for st_size.. so it won't work | 23:02 |
fury | think the answer is to define that offset bits as 64 so that it compiles with a 64 bit type size | 23:03 |
*** jlrmagnus has quit IRC | 23:04 | |
fury | -D_FILE_OFFSET_BITS=64 | 23:06 |
mranostay | guess popping -D_FILE_OFFSET_BITS=64 on it will fix it | 23:06 |
mranostay | someone googled faster it seems | 23:06 |
*** chbae has quit IRC | 23:06 | |
mranostay | but of course could break somethings | 23:06 |
fury | guarantee it will break something :D | 23:08 |
fury | is it using the sys stat call or a stat call that is compiled into LMS? if it could be changed just for LMS it'd prolly break fewer (or no) things | 23:16 |
fury | I dunno if I'm using the right terminology, still a bit intermediate at Linux development | 23:17 |
fury | coming from microcontroller development with 1 meg of code space and 256 k of RAM, the sheer amount of code that goes into making AGL work is simply mind blowing :D | 23:20 |
*** jlrmagnus has joined #automotive | 23:21 | |
fury | I've submitted a couple patches and still feel like a noob :P | 23:22 |
fury | practice makes perfect tho. just gotta find a spot where I can really add some value | 23:23 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!