Previous day | Jump to hour: 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 | Next day

Seconds: Show Hide | Joins: Show Hide | View raw
Font: Serif Sans-Serif Monospace | Size: Small Medium Large

Click in the nick column to highlight everything a person has said.
The Logo icon identifies that the person is a core developer (has commit access).

#rockbox log for 2021-11-09

00:58:21***Saving seen data "./dancer.seen"
01:20:16 Join ZincAlloy [0] (~Adium@2a02:8108:943f:d824:5125:465d:71e8:d3bc)
01:24:41 Quit ZincAlloy (Ping timeout: 245 seconds)
02:00:47 Join kugel [0] (
02:58:25***Saving seen data "./dancer.seen"
03:17:14 Join tchan1 [0] (
03:18:32 Quit tchan (Ping timeout: 240 seconds)
04:58:28***Saving seen data "./dancer.seen"
06:23:54speachy__builtin: I can generate a new set of ssh keys for gerrit but it's goign to require manual intervention for everyone.
06:33:10 Join advcomp2019_ [0] (~advcomp20@user/advcomp2019)
06:37:07 Quit advcomp2019 (Ping timeout: 256 seconds)
06:58:31***Saving seen data "./dancer.seen"
07:16:01speachybecause the ssh protocol doesn't know anything about rotating keys
08:05:29 Quit tchan1 (Read error: Connection reset by peer)
08:07:22 Join tchan1 [0] (
08:12:06 Quit tchan1 (Client Quit)
08:12:23 Join tchan [0] (
08:25:26 Quit tchan (Ping timeout: 246 seconds)
08:35:40 Quit ufdm (Read error: Connection reset by peer)
08:35:58 Join ufdm [0] (
08:58:32***Saving seen data "./dancer.seen"
09:14:50 Join massiveH [0] (
09:50:55rb-bluebotBuild Server message: New build round started. Revision 7af98ce6bb, 303 builds, 11 clients.
09:52:31 Join amachronic [0] (~amachroni@user/amachronic)
10:02:52rb-bluebotBuild Server message: Build round completed after 717 seconds.
10:02:54rb-bluebotBuild Server message: Revision 7af98ce6bb result: All green
10:07:00amachronicspeachy: that last commit may improve the USB situation
10:07:42amachronicI screwed up set address requests when I converted them to the new API and they could've been getting corrupted.
10:30:34 Join Genju [0] (~Genju@user/genju)
10:30:53rb-bluebotBuild Server message: New build round started. Revision b14df9077f, 303 builds, 11 clients.
10:31:09Genju <−− "The RockboxUtility fully supports these players (note that the current 1.4.1 release is from 2019 and will not work)."
10:31:20Genjuis it just me or is this conflicting information?
10:32:03amachronicthe support is there I believe, but only in dev builds which aren't released.
10:32:26Genjuhm, is there any way to get a git build of rbutil without compiling it myself?
10:32:45Genju(build for windows, that is)
10:32:46amachronictry here,53588.0.html
10:33:03Genjuoh, sweet, thank you!
10:33:07amachronicI'm not sure if there's anything newer.
10:33:17amachronicyou're welcome! :)
10:35:13Genjuthat one has the aigo eros q and k players in the device list, at least
10:35:44Genjuguess i'll have to measure my device to figure out which one to use
10:35:54Genjuor put it on a scale :]
10:36:18Genju(ordered a surfans f20 unit today)
10:37:25 Quit massiveH (Quit: Leaving)
10:37:34Genjuoh, never mind, surfans f20 is in the device list, too!
10:42:41rb-bluebotBuild Server message: Build round completed after 709 seconds.
10:42:42rb-bluebotBuild Server message: Revision b14df9077f result: All green
10:53:17Genjuwell, look at that
10:53:30Genjuactually managed to compile a git version of rbutil :D
10:54:18Genjux86_64 even!
10:55:18Genjuit depends on a bunch of external dll files, couldn't be bothered to make a fully static build, but it hasn't exploded yet
10:55:35 Join tchan [0] (
10:58:35***Saving seen data "./dancer.seen"
11:07:06_bilgusdoing better than I Genju I just never use rbutil :p
11:07:29Genjuthe manual installation process for the F20 seems a bit tedious
11:07:54Genjulike installing a special driver
11:08:23Genjuah, wait, that's just for the native port
11:08:33Genjuinstalling hosted rockbox seems easy enough without rbutil
11:17:35 Quit amachronic (Read error: Connection reset by peer)
11:37:32 Quit tchan (Ping timeout: 240 seconds)
11:38:52 Join tchan [0] (
12:46:08 Join ZincAlloy [0] (
12:58:38***Saving seen data "./dancer.seen"
13:02:50 Join lebellium [0] (
13:52:30 Quit Genju (Quit: Genju)
14:53:19 Join asabas [0] (~asabas@
14:55:37 Join kugel_ [0] (
14:58:43***Saving seen data "./dancer.seen"
15:00:32 Join salty-horse [0] (~salty-hor@user/salty-horse)
15:00:46 Quit kugel (*.net *.split)
15:00:46 Quit Xeha (*.net *.split)
15:00:46 Quit asaba (*.net *.split)
15:00:46 Quit Galois (*.net *.split)
15:00:46 Quit dys (*.net *.split)
15:00:53salty-horsespeachy, my headphones button issue is unrelated to your patch. I'll investigate it.
15:01:09 Join Xeha [0] (
15:02:14salty-horsewho is the webmaster for the rockbox website? It seems to be blocking my firefox user agent: returning http 403 when I use it, and 200 when I change it.
15:05:53speachysalty-horse: that would be me. send me the IP address you're coming from and I can look it up.
16:07:59speachyok, turns out firefox nightly is experimenting with "Firefox/100" in the user-agent and that was tripping over one of the rules in the banlist.
16:08:16speachy(a lot of bots claim to be firefox/1something)
16:28:13 Quit salty-horse (Quit: Leaving)
16:51:57braewoodsfinally, firefox has beaten chrome at something.
16:52:02braewoodsversion numbers. :P
16:58:47***Saving seen data "./dancer.seen"
17:10:04tertuokay so
17:10:08tertunow that i have my eros k
17:10:15tertuwhere can i grab the native builds?
17:15:29 Join dconrad_web [0] (
17:16:31 Quit braewoods (Quit: WeeChat 2.8)
17:19:32dconrad_webtertu: I think generally we're not keeping the native eros q/k builds around in order to reduce confusion, but the automatic builds after each commit are kept at for the time being (linked to from
17:19:44dconrad_webjust be aware, things might break from commit-to-commit
17:20:41dconrad_webthe bootloader isn't saved in the same way, but if you're ready to try it out I can get it to you when I get home from work
17:21:49dconrad_webas far as I'm aware I'm the only one who has tried it yet, so... it is definitely "your mileage may vary"
17:25:06 Join braewoods [0] (~braewoods@user/braewoods)
17:25:22dconrad_webthough I will say it has been remarkably stable for me
17:28:10 Quit lebellium (Quit: Leaving)
17:36:41munkiswell reverting dpc fixed the brightness changes, but not the panics.
17:48:14tertui'd definitely like to give it a shot
17:50:15dconrad_webexcellent, I should be able to get the bootloader to you in an hour or so
17:52:07dconrad_webI assume you've seen the writeup on and,53858.0.html/ (first post, mostly for the jztool downloads)?
18:02:01 Join amachronic [0] (~amachroni@user/amachronic)
18:07:27munkisamachronic: have you seen fs#13316?
18:07:28rb-bluebot Copying files is extremely slow (bugs, unconfirmed)
18:15:26_bilgusIve gotten a divide by zero several time on unplug since the USB stuff went in
18:15:51_bilgustotally un-reproducible OFC
18:16:46_bilgusI don't know if that narrows anything down for you or just makes it worse :/
18:17:20amachronicwhich port? at least that sounds like an easy one
18:20:28 Quit ZincAlloy (Quit: Leaving.)
18:21:02 Join ZincAlloy [0] (~Adium@2a02:8108:943f:d824:bdc3:ca08:66a8:2a89)
18:21:02 Quit ZincAlloy (Client Quit)
18:21:58_bilgusthats the clip zip
18:25:27amachroniccould you get me a pc next time it happens?
18:26:22amachronicand from the build
18:26:32amachronicor .elf even better
18:27:36 Quit dconrad_web (Quit: Connection closed)
18:44:57_bilguslets see if it will happen
18:51:08munkiswhat can you do with the pc? (I am trying to debug a [probably] entirely unnrelated crash)
18:53:30amachronicif you use objdump -d on the rockbox.elf file, it'll give you a disassembly with everything in the proper address space (usually) so you can use the pc to find the failing instruction
18:54:32munkisah thanks
18:55:43amachronicso now you can "translate" all those backtraces :)
18:57:29tertuyes i have seen the rightup
18:58:48***Saving seen data "./dancer.seen"
19:01:02 Join cockroach [0] (~blattodea@user/cockroach)
19:03:55 Quit amachronic (Quit: amachronic)
19:04:29 Join dconrad [0] (~dconrad@
19:06:41_bilgussomething special for ARM?
19:08:27_bilgusrockbox.elf: file format elf32-little −− objdump: can't disassemble for architecture UNKNOWN!
19:15:40dconradtertu, here's the bootloader:
19:16:18dconradwell, to download it I guess
19:16:46dconradfreshly built and tested on my own device
19:21:08dconradlet me know how it goes, hopefully it's smooth sailing
19:21:35dconradand don't forget the bootloader backup first ;-)
19:43:02tertuokay so it *immediately* shuts down thinking the battery is dead
19:44:11dconradaww, shoot I thought I had that fixed
19:44:38tertuother than that, seems to boot fine :)
19:46:32dconradI can get you another build with a longer delay before it samples the battery to see how long that delay needs to be
19:46:53tertui'll try it!
19:47:00dconradone minute
19:49:59dconradone sanity check though, do you know if the battery does have charge in it? you've charged it recently?
19:50:49tertuwell, i mean
19:50:53tertui just got it
19:51:10tertuand it has been charging for like, 5 hrs
19:51:16dconradok, yeah should be plenty
19:51:19tertuthe ofw thinks the battery is full
19:51:30dconradI just wanted to make sure it wasn't genuinely dead
19:53:00tertumakes sense!
19:56:27dconradit's taking a while to build, should be done in a few minutes here
19:57:41tertunp, i'm in no rush
19:57:55tertu(shouldn't even really be on irc now, but whatever)
19:58:07dconradhaha oh well
20:03:20dconradalright, give that one a shot
20:03:41dconradI increased it by quite a bit, so hopefully it should work OK
20:18:01tertuit boots!
20:18:09tertui'm trying to play music now and getting some panics
20:19:15tertuand after rebooting after one of them it now seems to think the battery is very empty. hmm
20:19:55tertui'll look into it more later
20:20:02dconradare you using the file browser or the database?
20:20:13dconradalright, shoot I thought it was good to go haha
20:20:17tertufile browser
20:20:19tertui mean
20:20:23tertuthis is honestly fine?
20:21:15dconradwell I mean it's been solid for me for daily use for a while now
20:21:24dconradso I'm not sure what's different
20:21:32tertuyeah one of the tracks was kind of weird
20:21:37tertuit was like a youtube extracted opus
20:22:16dconradoh that might be weird yeah... does a "normal" mp3 or flac play ok?
20:25:45tertuoh i couldn't actually play any tracks. the other ones gave me a *PANIC* TLB refill handler at 0x80025990! [0x4]
20:25:54tertupossibly the file names are too long
20:26:40tertuit also sometimes seems to not show the contents of my music folder?
20:26:57dconradok, that's a strange one
20:27:29tertui did format the card on linux so i don't know if it might have been formatted weirdly
20:27:38tertuit's a 128GB samsung sd card
20:27:59dconradthat's pretty large, but it should still be ok I would think
20:28:13dconradhow many files are in the music folder?
20:28:57dconradnot counting sub-folders, I guess
20:29:17tertunone, it's only subfolders
20:29:35dconradok, I thought maybe it was just a ton of files or something
20:29:42tertubut it's a total of 18!
20:29:44tertuthat said
20:31:14tertuthe longest path name on there is. 149 characters
20:31:29tertumight try shortening that
20:31:40dconradyeah that would be worth a shot
20:32:51dconradone other thing that might be worth trying is go into the context menu on "Files" (hold down menu), and try "Start File Browser at /"
20:32:59dconradsee if that changes anything
20:34:32speachy_bilgus: use the objdump that's part of our toolchain; arm-elf-eabi-objdump (I tihnk)
20:34:46tertuokay now the music folder is once again showing as empty
20:34:56tertudoing start file browser at / did not seem to affect that
20:35:35_bilgusah thanks that will save me from dumpint to asm
20:36:58_bilgustertu our paths are 260 char
20:37:17tertuyeah so the way it seems to go right now is: boot, /Music contains nothing
20:37:18_bilguscould you try the file browser in the demos/lua/scripts plugin?
20:37:37dconradwell one thing that might point us in the right direction as well, go into the Files context menu again and change "Show Files" to All
20:38:21tertuyeah so three things show up there
20:38:41_bilgus. .. and
20:39:56_bilgusoh nm I guess I removed that
20:40:01_bilguswhatshows up lol
20:40:10tertuso if I go into /, everything is normal
20:41:17tertuif i go into Music, i get /Music/RB12(missing character), /Music/þÿ, and /Music/ð@
20:41:34_bilgusyou have utf16 fnames
20:41:56_bilgusbecause you changed fs I presume
20:42:21_bilgusI bet that has something to do with those people that have to reformat the cards really well
20:42:41tertuyeah I was looking at it and i was like
20:42:51tertuthis looks like a unicode problem, but i don't know enough to know what kind
20:42:58_bilgustry popping the card in the pc (or maybe in usb mode) and rename a few now that it is FAT32
20:43:12tertuwell it didn't have any files on it when it was exfat...
20:44:24_bilgusits either that or a corrupted FAT table
20:44:58tertuthat was my other guess, given that it seems like the os and the player disagree about what the format is
20:45:43_bilguslua is just displasying what core has fed it its just lower level
20:48:56tertuoh, sorry, by the OS
20:49:04tertui mean my computer seems to find nothing wrong with it
20:49:09tertuand by the player i mean rockbox
20:49:17_bilgusI mean by rockbox RB
20:50:02tertuokay disagree about what the format is is also imprecise, what i meant was that rockbox and my laptop do not see the same directory structure
20:50:22tertui could try reformatting it under windows
20:54:57_bilgusgive me a second and I'll find the forum post
20:55:40tertufwiw linux seems to believe it has utf8 filenames
20:57:32_bilgusfor windows
20:58:51***Saving seen data "./dancer.seen"
20:59:21_bilgusand I want to say someone said they had luck in the gui gparted
21:01:35_bilgusfor linux..
21:06:10 Quit tchan (Quit: WeeChat 3.3)
21:06:51tertuthat may well be it
21:07:19tertui left a small area at the front of the disk unformatted, and it could be some of the exfat stuff was left over
21:16:34 Join tchan [0] (
21:30:00dconradI wonder if we could ignore low battery shutdowns for the first ~10 seconds after boot
21:30:29dconradseeing how much variation x1000 devices seem to have in time before they read the battery correctly
21:42:30_bilgusit would be better to hold up boot
21:43:03_bilgusif you give them that much time they are already playing music draining a low battery even more
21:43:37_bilguslet boot time out after 10s of a low read
21:43:58_bilgusthen there isn't anything to copy and minimal processing
21:44:49dconradyeah that could be ok
21:45:29dconradblock booting until the battery reads ok or times out
21:45:55_bilgusit might be slightly annoying but better than wiping out your battery
21:46:55_bilgusif it already has display up you could inform the user or even have a timeout
21:47:35dconradwould need to make sure it can still boot if usb is present I think
21:47:43_bilgusfor the time being you might hook a count in there and store that in settings
21:47:44dconradusb present + dead battery
21:48:32_bilgusI don't know how much variation you have but you might make it a debug item that you can query users about
21:48:51_bilgusor just dump it to settings and ask to see it
21:50:00dconradwell it seems that both m3k and eros q/k have some devices which are ok with a certain "wait time", and others which require quite a bit more
21:50:14_bilgusIIRC the fuze plus had a special build that enabled it but devices generally refuse to boot till it hits a threshold
21:50:15dconradbut all devices are "within spec"
21:50:44_bilgusso I wouldnt boot if battery dead and usb plugged
21:51:10dconradit would be nice to enable a full second before deciding to shut down so any caps can be charged
21:51:18_bilgusassuming it can charge in the bootloader
21:51:48dconradyeah I forget if these x1000 devices have hardware charging in their "off" state
21:52:16_bilgusI would lean towards yes but who knows
22:12:07 Quit dconrad (Remote host closed the connection)
22:12:43 Quit cockroach (Quit: leaving)
22:14:27 Join dconrad [0] (~dconrad@
22:21:52dconradpossibly dumb question: when powermgmt_init() gets run, it just spins off power_thread and then they can be considered to be running in parallel, right? it's going to move on to the next init thing right away
22:22:10dconradI'm not super familiar with threaded stuff
22:24:57dconradlooks like there's already a similar "wait until a valid read" thing for mrobe 100 and dx50/90 in the power thread
22:29:15dconradthough interestingly they don't seem to time out, so if they were actually suuuuper dead on boot, it may not ever actually do the emergency shutdown?
22:47:11 Quit emacsomancer (Ping timeout: 245 seconds)
22:49:01 Join emacsomancer [0] (~emacsoman@
22:58:54***Saving seen data "./dancer.seen"
23:53:35_bilguspossibly a bug? idk I try not to mess with studd I can't test
23:54:06_bilgusthe thread will hold till it yields
23:54:33_bilgusdepending on how its set up have to look
23:57:01 Quit dconrad (Remote host closed the connection)
23:57:05 Join tertu2 [0] (~tertu@user/tertu)
23:57:41 Quit tertu (Ping timeout: 244 seconds)
23:58:04_bilgusyeah it yields in sleep sleep(POWER_THREAD_STEP_TICKS);

Previous day | Next day