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).

Notice: Only Gecko based browsers prior to FF4 support the multipart/mixed "server push" method used by this log reader to auto-update. Since you do not appear to use such a browser, this page will simply show the current log, and not automatically update.

#rockbox log for 2020-10-12

00:06:17 Quit JanC (Ping timeout: 272 seconds)
00:09:50 Join JanC [0] (~janc@lugwv/member/JanC)
00:27:31 Join S|h|a|w|n [0] (~shawn156@unaffiliated/shawn156)
00:36:59***Saving seen data "./dancer.seen"
00:39:24 Quit ac_laptop (Ping timeout: 256 seconds)
02:16:57 Quit pixelma (Quit: .)
02:16:57 Quit amiconn (Quit: - Chat comfortably. Anywhere.)
02:19:26 Join amiconn [0] (jens@rockbox/developer/amiconn)
02:19:27 Join pixelma [0] (marianne@rockbox/staff/pixelma)
02:37:03***Saving seen data "./dancer.seen"
02:38:25 Quit w4tchguard (Ping timeout: 240 seconds)
03:51:43 Join pamaury [0] (~pamaury@rockbox/developer/pamaury)
03:58:14 Quit pamaury (Ping timeout: 256 seconds)
04:37:05***Saving seen data "./dancer.seen"
05:02:50 Join pamaury [0] (~pamaury@rockbox/developer/pamaury)
05:21:16 Join Huntereb [0] (~Huntereb@
05:24:40 Join Rower [0] (
05:54:06 Join Stanley|00 [0] (~stanley00@unaffiliated/stanley00)
05:56:05 Quit Stanley00 (Ping timeout: 240 seconds)
06:06:25 Quit S|h|a|w|n (Ping timeout: 240 seconds)
06:25:40 Join w4tchguard [0] (~w4tchguar@2001:ac8:84:37::a14d)
06:27:07 Quit w4tchguard (Client Quit)
06:37:08***Saving seen data "./dancer.seen"
07:09:16 Quit Stanley|00 ()
08:14:02speachy_bilgus_: just ran into a situation where the SD card could get completely blown away by the OF's startup scripts
08:15:42braewoodsspeachy: meaning it fills up the entire free space?
08:32:49speachybraewoods: as in "rm -Rf /mnt/*" in a startup script
08:33:39speachybasically there's a race condition between mdev mounting the sd card and that script running.
08:34:21speachy(on the xduoo firmware at least..)
08:34:59gevaertsSounds great
08:37:07braewoodsspeachy: ah, so the equivalent of rm -rf /
08:37:11***Saving seen data "./dancer.seen"
08:37:17speachybraewoods: from rockbox's perspective, yeah
09:02:16_bilgus_speachy that is a nice bug.
09:02:33speachy(ask me how I found it!)
09:03:14_bilgus_the unfortunate way?
09:15:33speachy_bilgus_: I seem to have only tripped over this on the xduoo targets; haven't had a chance to see if the rocker or eros stuff is similarly afflicted.
09:19:19_bilgus_do you have some steps to test?
09:22:31braewoodsprobably for the best i try to upgrade the bootloader... seems there's a fair number of ATA fixes since it was last updated
09:23:57speachy_bilgus_: so far it looks like a purely xduoo "feature"
09:25:10braewoodsspeachy: so it's a bug in the OF?
09:26:01speachybraewoods: no, not a "bug" −− they don't use auto-mounting.
09:26:11braewoodsi see
09:26:22speachythey just integrated things slightly differently than everyone else did.
09:26:44braewoods a nice little program i wrote to generate some code i need
09:29:58_bilgus_yep braewoods thats the seed continuation in the flesh but might do a bit better with a bigger buffer rather than a char at a time
09:30:30braewoods_bilgus_: erm? i'm reading up to 4K at a time
09:32:03_bilgus_oh lol missed it sorry
09:33:22braewoodsyea. fread is a weird API but it works.
09:33:49_bilgus_yeah I focused on the elem sz not the count
09:34:32braewoodsnot sure what fread does if the size is greater than 1 but this is the best way to read binary files
09:35:06braewoodsportably anyway
09:37:43_bilgus_I think now a days systems with char <> 1 or even char <> 8 bits are probably pretty scarce
09:40:29fs-bluebot_Build Server message: New build round started. Revision 9ad3086, 290 builds, 9 clients.
09:41:22speachy_bilgus_: when the build after this one is done, I'd appreciate it if you could re-test the audio stuff on the rocker.
09:41:44speachy(make sure it doesn't cut off the audio on first playback)
09:42:16_bilgus_sure, I have two more builds to test for the lcd stuff
09:42:19speachyI think I have it acting sanely on the eros, and as sane as possible on the x3ii.
09:44:44braewoodsok. now building a new RB build so I can try to flash this bootloader
09:46:50speachy(the rocker has a simple audio path that seems to mute/unmute nearly instantaneously. so I have it set to basically mute whenever there's no PCM going)
09:48:51braewoodsspeachy: isn't that good for latency?
09:50:10speachythere's no real analog/audio "latency" per se; the delays here are (presumably) due to firmware toggling the power supplies for the HP vs LO paths.
09:50:15speachy(and waiting for things to settle)
09:53:24_bilgus_ah remote screens still horribly broken
09:54:22speachythere aren't more than a handful of targets that support remote screens, right? iriver, X5? (classic ipods?)
09:55:25_bilgus_yeah In several different flavors as well think I'm just missing a few macros
09:56:44fs-bluebot_Build Server message: Build round completed after 975 seconds.
09:56:46fs-bluebot_Build Server message: Revision 9ad3086 result: All green
09:56:47fs-bluebot_Build Server message: New build round started. Revision 141e91e, 290 builds, 9 clients.
09:56:56_bilgus_I'm focused on the iRiver hp100 atm
09:58:10_bilgus_the usb 'follower' code in the BL makes it better already
09:59:40speachyyeah. amazing how much progress one tends to make when those annoyances keep smacking you in the face..
10:02:35_bilgus_Rolo had Executing grabled with its previous message, same line no clear
10:03:16_bilgus_Executinguffers or something along those lines
10:04:20speachywhoops, forgot to fix that. off by 1 error. :)
10:04:35_bilgus_not a big bug no worries
10:06:44_bilgus_playback is good flac mp3 ape opus m4a but then I switched directories and no audio
10:07:10_bilgus_still hear the voice prompts i'll try restart
10:07:42_bilgus_might be me the keymaps make it hard lol
10:07:49braewoodsYES IT WORKS
10:09:58_bilgus_speach appears to be good
10:10:30braewoodsnow to test the new bootloader with the CF modded player i got
10:11:04fs-bluebot_Build Server message: Build round completed after 857 seconds.
10:11:05fs-bluebot_Build Server message: Revision 141e91e result: 54 errors 12 warnings
10:18:54_bilgus_Its been several years since it landed but I think backlight exemptions and advanced keylock are my favorite
10:20:13_bilgus_this rocker really puts off some heat off that back glass
10:22:35braewoodsi don't dare try to flash the CF modded one...
10:22:46braewoodsusb is acting weird... i think i'll reinstall the original drive later.
10:22:56speachy_bilgus_: there's a 1ghz SoC in there
10:26:03_bilgus_quite nice speachy I think I'll buy one eventually (depending on supply)
10:26:46_bilgus_in all the pictures it seemed as big as the x3
10:34:41 Quit massiveH (Quit: Leaving)
10:37:14***Saving seen data "./dancer.seen"
10:38:39_bilgus_hmm, I think make clean isn't working quite right
10:48:37 Join mendelmunkis [0] (
10:49:25 Quit mendel_munkis (Ping timeout: 240 seconds)
10:53:37braewoodsspeachy: my new bootloaders fix the ATA errors i was getting. could you take the resulting patch to make iriver_flash happy?
10:54:17speachybraewoods: feeling up to getting this into gerrit?
10:54:45braewoodsthe bug tracker?
10:54:53speachythe patch tracker
10:55:11speachystarting at the "preparing to make changes or try out code from gerrit"
10:56:07speachyI mean, I can take the patch directly from you but given that you seem likely to want to continue contributing, this seems prudent
10:56:45braewoodsspeachy: though i do have a new mystery.
10:57:01braewoodsattempting to use USB when rockbox is fully booted causes it to freeze
10:57:16braewoodsall i get is this in my dmesg
10:57:20braewoodsusb 1-1.2: USB disconnect, device number 5
10:57:24braewoodsa disconnect message
10:57:36braewoodsonly on my CF modded player though
10:57:43braewoodsthe bootloader works fine in USB mode though
10:57:47braewoodsso no idea what's going on
10:58:07braewoodsi'll try building a regular git build and see where that gets me
10:59:26braewoodsmaybe 3.15 is glitchy a bit
10:59:46speachyI'd definitely recommend starting from current git.
10:59:54braewoodsi'm trying that now
10:59:59braewoodsjust building first
11:00:06braewoodsi had it using a custom RB directory
11:00:25braewoods... gerrit doesn't have its own account system?
11:00:51speachywe're using it with external openid/etc only.
11:01:01braewoodsspeachy: how do you execute memchk.lua?
11:01:20 Quit koniu (Remote host closed the connection)
11:01:20speachy...browse to it and hit enter?
11:01:45braewoodsi thought it was a menu entry
11:01:57speachy(I've fired up lua stuff something like three times in total fwiw)
11:02:07 Join koniu [0] (~koniu@gateway/tor-sasl/koniu)
11:09:45braewoodsstill an issue
11:09:52braewoodsi think it's a hardware issue of some kind
11:10:06braewoodsit's probably related to my CF mod since the original drive works fine
11:10:12braewoodsin the other unit
11:11:48braewoodsi'll do some experimentation later
11:16:55fs-bluebot_Build Server message: New build round started. Revision 01dc81c, 290 builds, 9 clients.
11:19:51_bilgus_anyone have an iriverh100?
11:23:04braewoodsonly thing left is to upload the bootloaders to the existing wiki
11:24:28speachyand onto the download site so rbutil users can pick it up
11:27:04braewoodsspeachy: for development or what?
11:27:17braewoodsafaik the official stable is still on BL 6
11:27:22braewoodsnot the 7 pre-releases
11:27:45speachyoh, on the patch front, please add yourself to docs/CREDITS
11:28:06braewoodshow can I amend it then?
11:28:13braewoodsi already pushed it
11:29:11speachyyou have to make a new commit, then rebase and squash the two together.
11:29:40speachyif you don't want to bother with that right now, no worries, I'll take care of it.
11:29:55braewoodsthis system is different from how I normally use git
11:30:03speachy(rebase/squash is one of the really, really nice things about git..)
11:30:36braewoodsin the past i've erased commits by checking out an older point and applying a new commit or so
11:30:41braewoodsand then doing a forced push
11:30:55braewoodsnot ideal though it works for small projects
11:31:21fs-bluebot_Build Server message: Build round completed after 866 seconds.
11:31:24fs-bluebot_Build Server message: Revision 01dc81c result: 54 errors 12 warnings
11:33:39fs-bluebot_Build Server message: New build round started. Revision 078c34e, 290 builds, 9 clients.
11:34:27speachybraewoods: it's submitted, and will eventually pop out the other end of the build farm.
11:34:37speachybraewoods: congratulations on your first patch!
11:35:33speachy_bilgus_: and when you get another chance, please try git head on the rocker too, had to make more PCM changes. Don't think they affected the rocker but one never knows.
11:36:36braewoodsi only could test the H120/H140 version but they're sister builds so
11:36:41braewoodsif one works the other should as well
11:37:41_bilgus_oh the 120 has a remote as well
11:38:08_bilgus_braewoods do you have the remote lcd with your h120
11:39:02speachy#$^@^$. missed a case.
11:39:57speachyis there any reason to not push the v7-pre bootloaders onto download.r.o for rbutil to pick up?
11:47:02fs-bluebot_Build Server message: Build round completed after 803 seconds.
11:47:04fs-bluebot_Build Server message: Revision 078c34e result: 4 errors 2 warnings
11:47:05fs-bluebot_Build Server message: New build round started. Revision b5c2a4a, 290 builds, 9 clients.
11:52:37braewoods_bilgus_: no but i have one coming soon. i found one for $10.
11:53:12braewoodsspeachy: probably not as they do boot but i haven't tested them with OF
11:53:53braewoodsthen again OF is only viable if you use a regular HDD
11:54:04braewoodsit is known to flake out with CF cards
11:54:36braewoodsi do know it boots just fine from the HDD player i haven't yet modified
11:54:46braewoodsi own 2 H120s
11:54:48braewoodsright now
11:55:57braewoodsi would have considered HDDs but there's not really any options in 2020 that i can tell
12:00:52fs-bluebot_Build Server message: Build round completed after 827 seconds.
12:00:54fs-bluebot_Build Server message: Revision b5c2a4a result: 4 errors 2 warnings
12:00:55fs-bluebot_Build Server message: New build round started. Revision 1724417, 290 builds, 9 clients.
12:05:10 Quit pamaury (Quit: Konversation terminated!)
12:13:56fs-bluebot_Build Server message: Build round completed after 781 seconds.
12:13:57fs-bluebot_Build Server message: Revision 1724417 result: 4 errors 2 warnings
12:13:58fs-bluebot_Build Server message: New build round started. Revision 6885753, 290 builds, 9 clients.
12:16:44_bilgus_hmmm the remote screens are goint to throw a much bigger wrench into my plans than I initially expected
12:17:52braewoods_bilgus_: ?
12:18:25_bilgus_several screen types coexist in a remote display scenario
12:22:00_bilgus_I think the way I set the buffer and address lookup functions help in that regard but I need a way for the default vp to be screen agnostic in its current form
12:23:06_bilgus_or else set up separate default vp lcd_default_vp vs remote_lcd_default_vp but I really don't like that
12:25:39_bilgus_I guess I can do it in the set viewport fn I just need to wire a way to figure out which screen the user intended to target
12:26:49fs-bluebot_Build Server message: Build round completed after 772 seconds.
12:26:50fs-bluebot_Build Server message: Revision 6885753 result: All green
12:37:17***Saving seen data "./dancer.seen"
12:50:24_bilgus_speach rocker @ head works fine small click on track change nothing major to report
12:50:33_bilgus_speachy *
12:51:44speachyhmm. which commit, 6885753?
12:52:02_bilgus_git status
12:57:05speachyso far of all of these ingenic/hiby targets, my hands-down favorite are these eros variants.
12:58:16braewoodsspeachy: is there any reason a CF card would have issues when used with an adapter set to slave mode?
12:58:31braewoodsi wonder since that's what ended up happening with that one
12:58:33_bilgus_well built well integrated or both?
12:59:07speachywell, slave mode is intended to operate as a slave to another device. the controller might have some conniptions, and CF cards are never "slaves" that I'm aware. :)
12:59:26braewoodsso worth investigating
12:59:33braewoodsi need to short the jumper again somehow
12:59:34speachy_bilgus_: well built, well integrated, and the flaws are almost nitpicks
12:59:42braewoodsi had to trim it back so it fit
13:01:41_bilgus_thats just the adapter board though right?
13:03:05_bilgus_it still has the pins so short it with a solder blob at the pins on the interface
13:05:33braewoods_bilgus_: yes.
13:05:38braewoodsbut i have no soldering tools
13:05:41braewoodsat present
13:06:06braewoodsbut yes, i need to...
13:06:09braewoodstry shorting it
13:06:33braewoodsi think soldering is the only real option though
13:06:53braewoodsi would buy low profile adapter but they don't exist
13:07:24_bilgus_CF card has a TRUE IDE mode, which allows the CF card to function exactly as a PATA/IDE drive. This mode is entered when the card has pin 9 grounded on the 50 pin
13:08:00_bilgus_I imagine this is the function the jumper sets
13:08:25braewoodsi thought it was just for
13:08:31braewoodssetting master vs slave channel
13:16:18_bilgus_pin 9 = ide mode pin 39 gets tied high for master
13:17:07_bilgus_sorry floart high = slave tied low = master
13:24:02_bilgus_I'd say with 9 grounded (default for your board most likely) and CS set to slave yes it will cause issues
13:42:36 Join lebellium [0] (
14:04:49 Join S|h|a|w|n [0] (~shawn156@unaffiliated/shawn156)
14:23:05 Quit Acou_Bass (Ping timeout: 240 seconds)
14:27:36 Join Acou_Bass [0] (
14:32:51 Quit Rower (Ping timeout: 260 seconds)
14:37:21***Saving seen data "./dancer.seen"
14:42:12 Join johnb7 [0] (
14:43:39johnb7for recording a battery charging curve, would I just run battery bench 'in reverse'?
14:55:30lebelliumspeachy I can get an AGPTek H3 for €40. I don't really need it but I'd like to see your new port :D
14:55:41speachythat's cheap!
14:57:02lebelliumYes, on a famous French second-hand website (but it is presented as brand new)
14:57:31speachynow it is based on the eros k as opposed to the q, so it's _possible_ there are some differences beyond just the case.
14:58:30lebelliumhum didn't you say "All five of those players have completely identical hardware, albeit with two different case designs" ?
15:00:08speachyan educated guess, but easily proved wrong.
15:00:53speachythere's no distinction between the two in teh firmware update images, but there could be something hidden within the actual player/platform code.
15:01:29lebelliumWill see
15:01:56lebelliumIf you need a guinea pig ;)
15:03:28lebelliumIs dualboot possible or will be ?
15:03:42johnb7lebellium : I don't want to snatch that player from you, but can you share the website for future items?
15:04:02speachyyep, works fine, although my changes to enable hotplug usb+sd might interfere with how the stock firmware does things
15:06:39speachywell, I can't flash the agptek_h3 firmware on top of the hifiwalker_h2.
15:06:53johnb7He says he bought it new for 65 Euros? Where do you get that deal...
15:08:54speachyand that's because agptek changed their model id string to 'agptek_erosq' −− the other four claim to be an 'erosq'
15:09:10johnb7Maybe 'she' judging by the shoes :-)
15:13:33lebelliumhe says it is worth €65 new but those clones are more like €100 from what I read. The Apgtek is discontinued anyway
15:15:44speachyhuh, there's also an Eros Qii −− outwardly similar but uses a different DAC
15:16:11 Quit danielp3344 (Quit: killed)
15:16:11 Quit blbro[m] (Quit: killed)
15:16:26 Quit kadoban (Quit: killed)
15:21:18 Join pamaury [0] (~pamaury@rockbox/developer/pamaury)
15:23:40 Quit Huntereb (Ping timeout: 246 seconds)
15:25:05 Join kadoban [0] (kadobanmat@gateway/shell/
15:31:14speachynot just the DAC, supposedly no longer using an Ingenic SoC.
15:35:58lebelliumwhere did you find the Agptek H3 firmware? I can't find it here
15:39:38 Quit johnb7 (Quit: Nettalk6 -
15:40:13speachy(thank you, google)
15:42:46 Join danielp3344 [0] (danielp334@gateway/shell/
15:42:52 Join blbro[m] [0] (blbrostrat@gateway/shell/
16:04:48speachyhuh, with the qii they went to some sort of RTOS. Update image is built using czip (from zipgenius)
16:31:08speachycan't figure out what the CPU core in this thing actually is. "EROS E2" is all the description there is, and since the firmware image is encrypted, can't glean anything useful from that either.
16:37:23***Saving seen data "./dancer.seen"
16:44:27efqwspeachy: got a link to the fw?
16:57:03efqw:/ didn't find anything useful either
17:11:57efqwIf time permits, I'll try to put up USB Cloner instructions on the wiki
17:12:25 Join ac_laptop [0] (~ac_laptop@
17:18:27speachyefqw: I haven't had a chance to try building a (manually) mangled m3k firmware yet, sorry.
17:31:24 Join MrZeus__ [0] (~MrZeus@2a02:c7f:70d0:6a00:5d8a:b2a1:1c3:3bd3)
17:56:57 Quit mendelmunkis (Remote host closed the connection)
17:57:34 Join mendelmunkis [0] (
18:01:49 Quit lebellium (Quit: Leaving)
18:28:59 Quit pamaury (Ping timeout: 260 seconds)
18:37:26***Saving seen data "./dancer.seen"
18:44:05 Quit ac_laptop (Ping timeout: 240 seconds)
19:11:15 Join ac_laptop [0] (~ac_laptop@
19:34:24fs-bluebot_Build Server message: New build round started. Revision f68463c, 290 builds, 9 clients.
19:50:13 Join bluebrother [0] (~dom@rockbox/developer/bluebrother)
19:51:40 Join fs-bluebot [0] (
19:53:10 Quit bluebrother^ (Ping timeout: 246 seconds)
19:53:52 Quit fs-bluebot_ (Ping timeout: 246 seconds)
19:57:31fs-bluebotBuild Server message: Build round completed after 1387 seconds.
19:57:32fs-bluebotBuild Server message: Revision f68463c result: All green
20:19:07 Quit MrZeus__ (Ping timeout: 272 seconds)
20:37:27***Saving seen data "./dancer.seen"
22:07:36 Quit ac_laptop (Ping timeout: 256 seconds)
22:37:31***Saving seen data "./dancer.seen"
23:04:52 Join Stanley00 [0] (~stanley00@unaffiliated/stanley00)
23:12:04_bilgus_ok I got the viewport issue worked out I just started carrying stride in the framebuffer struct (which I should have done in the first place tbh)
23:14:21 Join ac_laptop [0] (~ac_laptop@
23:55:25 Quit TheSeven (Ping timeout: 240 seconds)
23:55:43 Join [7] [0] (~quassel@rockbox/developer/TheSeven)

Previous day | Next day