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 2020-05-27

01:53:33***No seen item changed, no save performed.
02:03:31 Join markun [0] (
03:12:52 Join dys [0] (~dys@2003:5b:203b:100:a64c:c8ff:fef4:13a6)
03:53:34***No seen item changed, no save performed.
04:12:57 Join St0neHead- [0] (stonehead@2a01:7e00:e001:3700:6667::2)
04:15:36 Join funman_ [0] (
04:20:07 Quit St0neHead (*.net *.split)
04:20:07 Quit funman (*.net *.split)
04:50:49 Join pamaury [0] (~pamaury@rockbox/developer/pamaury)
04:52:17 Join vmx [0] (
05:34:53 Join sakax [0] (~r0b0t@unaffiliated/r0b0t)
05:53:37***Saving seen data "./dancer.seen"
06:23:43 Join Demosthenex [0] (
06:24:26Demosthenexhey folks. i'm a happy rockbox user on my sansa fuse+, works great! recently i'm experiencing nasty hissing noises when i turn it on, and i can hear it "thinking", different tones when loading music, etc...
06:24:54Demosthenexon the forums it was suggested to boot it to the original firmware and back to rockbox, but i can't get the instructions to work (power on while holding |<)
06:25:00Demosthenexis there an alternate method to boot back?
06:26:49Demosthenexi did try setting the time, but i can't seem to boot back temporarily
06:28:37Demosthenexah ha! in the other guide, it says volume down isntead of left arrow
06:28:40Demosthenexthat seems to do it
06:32:41Demosthenexand i can confirm booting to orig fixed the hiss, i guess it reset somethingin the dac?
06:54:16 Quit zagor[m] (Changing host)
06:54:16 Join zagor[m] [0] (bjstmatrix@rockbox/developer/Zagor)
06:54:54zagor[m]speachy: how much lead time do you want? would tonight be good for you?
07:05:49 Quit APLU (Ping timeout: 272 seconds)
07:06:47 Join APLU [0] (~mulx@2a03:7220:8081:2900::1)
07:34:32zagor[m]just not during the spacex launch... :)
07:37:08 Join petur [0] (~petur@
07:37:08 Quit petur (Changing host)
07:37:08 Join petur [0] (~petur@rockbox/developer/petur)
07:53:39***Saving seen data "./dancer.seen"
08:21:59 Join massiveH [0] (
08:44:39pamauryDemosthenex: yes the OF resets some DAC settings, I never quite got figured out why those change sometimes and produce that hissing sound, and I didn't spend time reverse engineering the exact settings the OF uses
09:06:47 Quit APLU (Read error: Connection reset by peer)
09:07:30 Join APLU [0] (
09:43:12 Quit akaWolf (Ping timeout: 265 seconds)
09:43:40 Quit APLU (Ping timeout: 276 seconds)
09:50:33 Join APLU [0] (~mulx@2a03:7220:8081:2900::1)
09:53:18speachyzagor[m], tonight will work.
09:53:22 Quit markun (Ping timeout: 272 seconds)
09:53:39speachyso 23:00 CET again?
09:53:42***Saving seen data "./dancer.seen"
10:00:50 Quit massiveH (Quit: Leaving)
10:04:17 Join markun [0] (
10:05:44 Join akaWolf [0] (~akaWolf@unaffiliated/akawolf)
10:06:22speachylead time isn't so important for git/gerrit as there aren't bunches of random folks hitting them.
10:07:41 Nick mendel_munkis_ is now known as mendel_munkis (
10:07:54speachythe only thing I have left to do is to add in the rewrite rules to convert gitweb to cgit urls.
10:08:58 Quit markun (Ping timeout: 260 seconds)
10:11:37 Quit vmx (Ping timeout: 264 seconds)
10:17:28nihilazodoes rockbox have a way to skim audio files
10:17:55mendel_munkisfreetypes porting information is woefully incomplete
10:18:31nihilazowhen I'm listening to podcasts on my desktop I skip through them in 5-second ish intervals rather than all at once to skip ad breaks, can rockbox do something similar?
10:19:05mendel_munkisnihilazo: did you try holding down forward for a second?
10:19:23mendel_munkiswhat device are you using
10:19:34nihilazowhen I do that it just skips but the audio is slient so I can't skim through
10:19:37nihilazoI'm on a fuze+
10:20:23mendel_munkisIf i am understanding you correctly you want to temporarily increase playback speed?
10:20:48nihilazothat would work, yeah
10:22:30mendel_munkisI'm pretty sure there's no native way to do that (besides increasing speed and then decreasing it when you reach the place you want) but someone may have written a patch
10:23:40mendel_munkisare you comfertable compiling rockbox yourself?
10:24:52zagor[m]speachy: can we go earlier, like 21:00?
10:26:01speachyzagpr[m], sure, that will work. So that's in 4.5 hours?
10:26:50speachyok. I'll send out an email announcement.
10:29:18nihilazomendel_munkis: I can try, I've not done it before
10:29:56nihilazoI'm comfortable with compiling other projects and I assume rockbox isn't much different?
10:32:04nihilazothanks, looks pretty similar to what I'm used to
10:32:19speachyother than neeeding a special cross-compiler it's pretty self-contained.
10:32:42speachyespecially if you're used to microcontroller/rtos applications
10:39:32nihilazoso, can I get the ability to skim through a track if I compile myself?
10:39:52mendel_munkisI am working on a patch for that
10:40:24nihilazoawesome, thanks so much
10:40:37mendel_munkisit may take a while though
10:42:29mendel_munkisI can see a way to set a hot key to speed it up but I cant see how to reset the speed when you let go sorry.
10:42:50speachytap once for speed up, tap again for restoring?
10:42:52nihilazooh ok
10:42:53nihilazoit's fine, dw
10:43:14nihilazoif it would be a pain to do then it's fine, would just be a convenience for listening to podcasts
10:43:47mendel_munkisspeachy: but in order to make that workable it would need to be a short press which means removing important functionality
10:44:04speachythere's always a compromise. :D
10:44:42speachybut in all seriousness, this is a prototype and if it turns out to be more widely desired a better way to integrate it is probably conceivable
10:46:19mendel_munkisnihilazo: there may be a way
10:49:27speachyannouncements have gone out. So in 4 hours, git+gerrit will go down for migration.
10:49:42mendel_munkisso I gotta be done by then
10:50:39gevaertsIt's git. You can run offline :)
10:50:40speachyor find something else to do for an hour or so. :)
10:51:02*gevaerts stands by the sidelines encouraging all this work
10:51:18mendel_munkisI don't see the announcement. what's the migration?
10:51:25speachythere's also the public mirror on github
10:51:50speachymendel_munkis, and are migrating to a new host
10:52:26mendel_munkisok. I didn't realize they hadn't already moved. thank you
10:52:52speachyI wanted them to be the first but instead they're the last
10:53:21nihilazois there a way to have the crosscompiler build script install to a non-system place that's on the PATH like ~/.local/bin?
10:53:46nihilazoarch is especially picky with things other than the package manager putting things in system places
10:53:55nihilazoin my experience
10:54:43speachythe crosstools have a _lot_ of stuff, not just binaries; you should put them in their own self-contained place and add the final bin location to your path instead. or create symlinks from those binaries..
10:55:44speachyRBDEV_PREFIX=/some/path tools/
10:56:16speachypersonally I keep everything of mine under ~/x-tools/rockbox
10:57:07speachywand then the bin dir will be ~/x-tools/rockbox/bin
11:02:44mendel_munkisspeachy: I am assuming that talk would be unnecessary if speed is hotkeyed?
11:03:40mendel_munkisdo we have an accessibility expert?
11:04:58speachyI'd never make an assumption like that
11:05:41mendel_munkiswhy not?
11:06:33mendel_munkisthe talk is assuming you are trying to get a specific speed. A hotkey has one predetermined speed
11:06:58speachyoh, I see what you mean. There's probably no reason to voice the setting change, yeah.
11:08:03nihilazoOK, I'm getting an error when building the arm cross-compiler, should I pastebin the log?
11:11:10speachythere are some ...complications with respect to building some of the toolchains on more modern systems
11:11:22speachybut yeah, pastebin please
11:17:51nihilazowhat's a better pastebin site than hastebin, it seems to be struggling with a long log
11:23:34nihilazothis is better
11:28:31Demosthenexpamaury: no problem, i was able to solve it ;]
11:28:34 Part Demosthenex
11:30:22speachyhuh. I think this one's due to the gcc10 change from -fcommon to -fno-common
11:47:33mendel_munkisnihilazo: try g#2388 (it's completley untested though)
11:47:34fs-bluebot_Gerrit review #2388 at : basic skim funtionality by Moshe Piekarski
11:47:54mendel_munkisyou need to enable timestretch and a new skim
11:47:59mendel_munkis_mode setting
11:48:23nihilazoawesome, I need to get my build environment working now
11:48:28nihilazogcc10 seems to have broke things
11:48:33 Join markun [0] (
11:48:35nihilazoand I don't know enough to fix them
11:49:27 Quit dys (Ping timeout: 260 seconds)
11:51:09 Quit APLU (Remote host closed the connection)
11:52:55speachyI have a WIP gcc 4.9.4 toolchain that seems to work okay if you're not on a arm7tdmi device
11:52:58speachythe fuze+ should be okay
11:53:16speachygrab g#2305
11:53:18fs-bluebot_Gerrit review #2305 at : Build: Bump all compilers to GCC 4.9.4 (DO NOT MERGE YET) by Solomon Peachy
11:53:43***Saving seen data "./dancer.seen"
11:53:58speachyand re-run the arm toolchain build.
11:54:15nihilazodo I get that with a git fetch?
11:54:26 Quit markun (Ping timeout: 256 seconds)
11:54:34speachyif you go to that URL, there's a 'download' link on the upper-right corner
11:54:44speachycut-n-paste the 'cherry-pick' link
11:55:15 Join markun [0] (
12:01:12 Join APLU [0] (~mulx@2a03:7220:8081:2900::1)
12:07:02nihilazothis build seems to be going a lot better
12:34:35 Join MrZeus [0] (
12:39:14nihilazoalright, I got the build environment sorted, should I do the same with cherrypicking the commit for the one that has the patch?
12:48:41speachythen: mkdir build-fuzeplus ; cd build-fuzeplus ; ../tools/configure
12:48:53 Join vmx [0] (
12:48:54nihilazoI've got an error when I build it
12:49:47speachywell, that's a problem with the patch
12:51:03 Quit markun (Ping timeout: 260 seconds)
12:52:43 Join markun [0] (~markun@
12:58:32 Quit markun (Read error: Connection reset by peer)
12:59:06speachyin apps/gui/pitchscreen.c, line 1152, change it to read 'int32_t new_pitch, new_speed;'
13:02:17 Join markun [0] (
13:07:42nihilazoseems to have fixed it, thanks
13:07:42 Quit markun (Read error: Connection reset by peer)
13:07:52 Join markun [0] (
13:09:26nihilazocan I just extract that zip over my existing rockbox install?
13:10:17speachyI'd back up your existing .rockbox dir first though
13:10:56speachywhen it comes to experimental stuff like this, backups are always warranted.
13:13:28nihilazoI don't seem to have the skim mode setting
13:14:59speachyunder playavk settings, between party mode and volume fade?
13:15:38nihilazoyeah, it's not there
13:17:35speachydoes the 'about' menu item show the correct version/date?
13:20:35nihilazoI don't seem to have that option
13:21:25nihilazowait, I might have just accidentally built the master instead by mistake
13:21:30speachysystem/rockbox info
13:21:32 Quit petur (Quit: Connection reset by beer)
13:23:02nihilazo(gtg, back in a bit)
13:41:29nihilazobuilt a version that definately has the patch and now USB is broken although that may be an unrelated thing from the master
13:42:05speachythat may a bug related to gcc494
13:44:49nihilazowould it be possible for somebody else to make a build with the patch for fuze+ and stick the zip somewhere I can get it?
13:45:10nihilazoalso, how can I change rockbox versions if USB isn't working, can I load a rockbox zip from the SD card?
13:45:21nihilazo * also, how can I change rockbox versions if USB isn't working, can I load/install a rockbox zip from the SD card?
13:45:31speachygimme a few
13:45:35speachygimme a few
13:52:12nihilazohow can I install that while not having USB?
13:52:20speachyboot into the original firmware
13:52:50speachyor extract it to the SD card and ROLO into the new firmware
13:53:44***Saving seen data "./dancer.seen"
13:54:11 Join advcomp2019_ [0] (
13:54:11 Quit advcomp2019_ (Changing host)
13:54:11 Join advcomp2019_ [0] (~advcomp20@unaffiliated/advcomp2019)
13:57:24 Quit advcomp2019 (Ping timeout: 256 seconds)
13:57:33nihilazoI extracted it onto the SD card and selected the rockbox executable, it said ROLO and rebooted, but I still don't have the skim mode option
13:57:47nihilazoUSB does work now though
14:25:44 Join lebellium [0] (
14:27:02 Quit markun (Ping timeout: 260 seconds)
14:34:36mendel_munkisI amback and fixing that patch
15:12:38 Join Ornery_Squirrel [0] (
15:21:41speachyoh, heads up, gerrit's going down.
15:26:58speachydns changes made.
15:35:55 Join morgenstern09 [0] (
15:39:03speachygit and gerrit are running. git is https enabled. gerrit, not yet.
15:44:53speachyrestarting gerrit with https ....
15:51:52 Quit morgenstern09 (Remote host closed the connection)
15:51:59speachyokay, it appears to be all alive.
15:52:45 Join petur [0] (~petur@rockbox/developer/petur)
15:52:47speachyI need to put in the rewrite rules to automagically convert gitweb->cgit URLs, but it's otherwise done.
15:53:26mendel_munkisI fixed some of the problems but there are still a few left.
15:53:46***Saving seen data "./dancer.seen"
15:54:24speachymendel_munkis, can you try to push the updated changeset to gerrit and see if it works?
15:54:57mendel_munkisspeachy: I just got a half-decent build. pushing now
15:56:36mendel_munkisspeachy: gerrit is down
15:57:16speachytry again. firewall hole wasn't applied for some reason.
15:57:36 Join markun [0] (
15:59:08mendel_munkisAlso turns out gerrit doesnt like it when you change the message on a commit −−amend
15:59:26mendel_munkisalso gerrit google oauth is down
15:59:36speachyah, new changeset. that's why I didn't see it change.
16:00:15mendel_munkisyeah I'm not sure why. the hook shouldn't generate a new ID for a amended commit
16:01:19 Quit Ornery_Squirrel (Quit: Leaving.)
16:02:01speachycrap... the google openid stuff doesn't like https gerrit.
16:02:15 Join Ornery_Squirrel [0] (
16:02:22mendel_munkisall the more reason to get github working :)
16:03:08*mendel_munkis just wants to be able to auth with a U2F key
16:03:19 Part Ornery_Squirrel
16:04:48nihilazo(I'm not sure how this thing handles being in a chat or away, I'm still here if it said I left)
16:06:58speachytrying to fix this at the google side of things; if not I'll disable https and things should continue to work okay.
16:10:09mendel_munkisnihilazo: try g#2405 it should work better
16:11:07nihilazomy build environment has some issues because of the gcc
16:11:17mendel_munkisspeachy: apparently fs-bluebot_ cant connect to gerrit either
16:11:47speachythat could be due to dns propogation.
16:17:08speachyokay, google authentication should work again. or it now just did for me, at least.
16:19:22mendel_munkisworks here to
16:23:04bluebrotherspeachy, mendel_munkis: it's because of the change from gitweb to cgit.
16:24:10speachybluebrother, 15 seconds ago I enabled aliases for gitweb-formatted request. or is the bot parsing the gerrit looking for the magic 'gitweb' phrase?
16:24:28bluebrotherthe bot uses git ls-remote to get the commit hash, then queries gitweb since I don't have all remotes locally. Was the easiest way back then.
16:24:40bluebrotherand uses the rss output of gitweb
16:24:40speachyexecllent, gerrit email notifications work.
16:27:39speachyprobably not a trivial thing to fix, but I can't enable gitweb for $reasons
16:27:55bluebrotherah. $resons sounds $reasonable :)
16:28:26bluebrotherI guess I have to update the scraping code then. Just not today anymore.
16:28:28speachyone of the url mangling rules I just enabled does cover gitweb rss/atom feeds
16:28:41speachybut the RSS contents likely differ slightly.
16:28:49speachys/ slightly/./
16:29:21bluebrotheror find a different way to get that information. iirc the main problem was that I can git ls-remote to get the hash, but there's no way to use git to retrieve a commit for a repository I haven't cloned.
16:30:21bluebrotherwould be such a scraping request. Forwards to cgit, but results in an error message.
16:31:22speachyah, the URLs are different. give me a sec.
16:31:53bluebrotherI've been using rss since that is easier to parse. If theres a similar output (or maybe some other api, wouldn't mind json or the like) I can update that.
16:35:42speachyI'd bet that gerrit has a straightforward API that you can use directly.
16:41:23speachybluebrother, just doing a GET on ""
16:41:43speachywill return a json object with all of the info you want. Owner, subject, etc.
16:42:50bluebrothernice. Is there also a way to do that for an arbitrary git hash?
16:42:53speachyhuh, gitlab auth is brroken because it uses the google app credentials.
16:44:59speachybluebrother, not directly from what I can tell.
16:45:15speachybut you can do something like this: ""
16:46:00bluebrotherhmm. I'll give that a closer look tomorrow. gtg for now.
16:46:01speachyyou can extract the changenum or full changeid from that response and repeat the query to get the full details.
16:46:11nihilazo<mendel_munkis "nihilazo: try g#2405 it should w"> can somebody do a fuze+ build, my environment has issues
16:46:42bluebrotherwhat about a git hash that wasn't reviewed on gerrit? There won't be a change for that, right?
16:47:27speachyit won't show up as there's no gerrit id associated with it
16:47:47speachyall gerrit would be able to do is point you at the commit in gitweb/cgit
16:48:02speachyso yo'll be better off just hitting cgit directly for that.
17:09:05 Quit vmx (Quit: Leaving)
17:37:11 Join morgenstern09 [0] (
17:39:24fs-bluebot_Build Server message: New build round started. Revision fba4df5, 292 builds, 10 clients.
17:45:08pamauryspeachy: do you know if there is a delay in wiki edits? I change the code but when I load the wiki page, the edits don't appear
17:48:05speachypamuary, I ran into that too, but I figured it was to do with my browser caching inappropriately after the https enablement.
17:48:19speachywhat's the page?
17:48:49speachy(there was also some wonkiness with the wyswig editor too. Might need to re-disable it..)
17:52:24 Quit lebellium (Quit: Leaving)
17:53:47***Saving seen data "./dancer.seen"
17:54:05fs-bluebot_Build Server message: Build round completed after 881 seconds.
17:57:35pamauryspeachy: ^
17:58:37gevaertsspeachy: not sure what the status is supposed to be right now, but google auth isn't working for me. I get a 404 on the redirect back from google to gerrit
17:59:40speachywhat's the bad url?
18:00:08speachyjust did it to me too. sigh.
18:00:18__builtinyeah, same here
18:00:36speachyguess I broke that when I was hacking on the gitweb redirection stuff.
18:00:52speachy(which, I believe, now works properly.. bleh, what a cluster%$^@)
18:02:08speachyI know why it's not working, I'm just trying to have my cake and eat it too.
18:02:31__builtinclicking on;a=rss;h=147f7f0e62baf2891f4fddc598827b5092fd13ec redirects me to
18:02:59speachy__builtin, is that sane?
18:03:13__builtinthat redirect?
18:03:22__builtinthe lack of formatting is pretty jarring
18:03:35speachyit's an XML document.
18:03:50__builtinyeah, I'd think linking to an HTML page would make more sense
18:03:58__builtinoh... that's rss. duh.
18:04:16speachywhere's that link coming from? the one bluebrother mentioned was his bot's data acquisition hackery
18:04:30__builtinit is bluebrother's link
18:04:33speachyeven the irc log reader page should be generating valid cgit links
18:04:48__builtindidn't realize he'd already reported that issue
18:05:20speachypamuary, as soon as I logged into the wiki, the page showed up correctly.
18:05:29speachyso there's something wonky going on with the wiki caching...
18:08:44speachywith respect to the gerrit login, I'm trying to craft the correct rewrite/redirect rules.
18:09:09speachybut if you refresh your 404 now, it will work. I put in the old blanket rule that breaks the https certbot stuff.
18:10:08gevaertsYay, both my accounts work :)
18:10:21gevaertsIncluding the old one that had been broken for years
18:12:26speachygotta restart gerrit, it's going to disappear for a moment..
18:13:46speachyExcellent! the github gerrit provider no longer 404s.
18:14:56speachyaaannd it works. emails are still handled through haxx, right?
18:15:18speachyyes, email and dns remain with haxx for now, though I'd like to change that too.
18:15:43speachythey weren't part of the original "MUST MOVE OR ELSE" service list.
18:18:29 Quit markun (Remote host closed the connection)
18:18:38 Join markun [0] (
18:27:12 Quit kugel (Ping timeout: 256 seconds)
18:27:48speachy g#2605
18:28:07speachy...guess that wasn't enough to fix the bot.
18:28:51 Join kugel [0] (
18:28:51 Quit kugel (Changing host)
18:28:51 Join kugel [0] (~kugel@rockbox/developer/kugel)
18:30:21 Quit petur (Quit: Leaving)
18:36:19*__builtin cheers speachy on
18:36:29speachypamaury, I made a tweak to the wiki config that may help.
18:36:53speachydue to still-ancient software, I can't fire up the proper tools to administer it. :/
18:43:41speachynope, still busted. :/
18:48:18 Join krabador [0] (~krabador@unaffiliated/krabador)
18:52:06 Quit markun (Ping timeout: 260 seconds)
18:52:51 Join amiconn_ [0] (jens@rockbox/developer/amiconn)
18:52:52 Nick amiconn is now known as Guest17055 (jens@rockbox/developer/amiconn)
18:52:52 Nick amiconn_ is now known as amiconn (jens@rockbox/developer/amiconn)
18:53:01 Join pixelma_ [0] (marianne@rockbox/staff/pixelma)
18:53:02 Nick pixelma is now known as Guest61071 (marianne@rockbox/staff/pixelma)
18:53:02 Nick pixelma_ is now known as pixelma (marianne@rockbox/staff/pixelma)
18:53:32 Quit Guest17055 (Ping timeout: 260 seconds)
18:54:07 Quit Guest61071 (Ping timeout: 260 seconds)
18:55:01 Join markun [0] (
19:04:25 Quit pamaury (Ping timeout: 264 seconds)
19:07:15 Quit markun (Ping timeout: 265 seconds)
19:22:43 Quit livvy (Ping timeout: 240 seconds)
19:26:40 Join livvy [0] (~livvy@gateway/tor-sasl/livvy)
19:28:10 Quit morgenstern09 (Remote host closed the connection)
19:53:48***Saving seen data "./dancer.seen"
20:21:52speachypamaury, much mucking-about later, I _think_ I've fixed the caching issues. And hopefully made things a bit faster in the process.
20:28:47 Quit kadoban (Ping timeout: 265 seconds)
20:29:12 Quit atsampson (Ping timeout: 260 seconds)
20:30:27 Join atsampson [0] (
20:30:36 Join kadoban [0] (kadobanmat@gateway/shell/
20:32:49 Join morgenstern09 [0] (
20:44:37 Quit MrZeus (Ping timeout: 264 seconds)
20:58:52 Quit sakax (Quit: Leaving)
20:59:44 Quit krabador (Remote host closed the connection)
21:25:41 Join Ornery_Squirrel [0] (
21:26:17 Part Ornery_Squirrel
21:30:23 Quit livvy (Ping timeout: 240 seconds)
21:33:09 Join livvy [0] (~livvy@gateway/tor-sasl/livvy)
21:53:51***Saving seen data "./dancer.seen"
22:49:50Bilgusspeachy gerrit feels much snappier!
23:53:52***No seen item changed, no save performed.

Previous day | Next day