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 2014-03-10

00:03:54 Quit wodz (Quit: Leaving)
00:04:19ls_Yea,
00:04:42saratogaproblem solved?
00:04:49ls_then I see the error "Not enough space for music" and it shuts down.
00:05:10saratogaso the storage is probably really screwed up
00:05:26saratogai think theres some tricks people use to try and bypass building the sandisk database, but i've never tried them
00:10:12ls_Yea I think the data is corrupted, and I can not mount to a computer to reformat.
00:16:33***Saving seen data "./dancer.seen"
00:20:04 Quit ls_ (Quit: Page closed)
00:28:51[Saint]SanDisk really screwed up with that.
00:28:53saratogabootloader USB mode might also be worth trying if he comes back, theres a chance that will work
00:29:09saratogaif the OF itself is intact, then probably so is the bootloader
00:29:20[Saint]You'd think they would offer bootloader USB for such fail cases.
00:30:01[Saint]"not enough room for music, ...lets shut down so you cant do anything about it"
00:30:20[Saint]That rationale is hard to justify.
00:30:50[Saint]Even "not enough room for music, skipping database construction" would be better.
00:31:00[Saint]...but, nope.
00:33:35 Quit ender` (Quit: The two aspiring triangle peaks arching into an abstracted M typified the progressive leadership-minded outlook of the company.-- Motorola exec on their logo)
00:37:11 Quit benedikt93 (Quit: Bye ;))
00:38:41 Join us`0gb [0] (~0gb.us@c-50-137-60-25.hsd1.or.comcast.net)
00:43:14 Quit us`0gb (Read error: Operation timed out)
00:51:52 Join Strife89 [0] (~Strife89@adsl-98-80-237-172.mcn.bellsouth.net)
00:57:04 Join us`0gb [0] (~0gb.us@c-50-137-60-25.hsd1.or.comcast.net)
01:00
01:44:33 Join Jinx [0] (~Jinx@unaffiliated/jinx)
01:53:44 Quit treaki_ (Ping timeout: 252 seconds)
02:00
02:07:04 Join treaki_ [0] (7a4befa1ac@p4FF4A5F7.dip0.t-ipconnect.de)
02:16:37***Saving seen data "./dancer.seen"
02:29:36 Quit ZincAlloy (Quit: Leaving.)
02:41:12 Nick froggymana is now known as froggyman (~frogs@unaffiliated/froggyman)
03:00
03:00:06 Quit AlexP (Remote host closed the connection)
03:53:44 Join shamus [0] (~shmaus@ip-206-192-193-180.marylandheights.ip.cablemo.net)
04:00
04:16:40***Saving seen data "./dancer.seen"
04:17:26jhMikeSnoone's around to kick the friggin' build system?
04:19:49[Saint]jhMikeS: possibly stalled on Android?
04:20:09[Saint](he says, not looking at the table at all)
04:21:52[Saint]Looking at the table...yes.
04:22:16[Saint]We (apparently) have no up-to-date Android build clients.
04:22:52*[Saint] thought the build system gave up after a while
04:22:58[Saint]I guess not.
04:25:54jhMikeSyeah, kugel's last commit
04:28:49[Saint]It makes some sense...nobody touched their Android stuffs since apkbuilder was removed.
04:29:20[Saint]I have a functional build environment, but its not a build client.
04:30:41[Saint]actually, my Android stuffs probably doesn't work anynore with these recent changes so I'd need to revert my hackery first.
04:30:59 Quit pixelma (Read error: Operation timed out)
04:31:04jhMikeSAndroid stuffs need to be added separately to each client?
04:31:15[Saint]But even if I did add another client right now this round woukd still hang.
04:31:26[Saint]Yes.
04:32:05jhMikeSWell, duh, I guess it would have to
04:32:28jhMikeSBut noone's gonna have it :)
04:32:39[Saint]Android ADT package, android NDK, and a JDK.
04:33:01 Join amiconn_ [0] (quassel@rockbox/developer/amiconn)
04:33:05jhMikeSThanks Obama! I mean kugel.
04:33:09[Saint](openjdk-6 and *-7 are both fine)
04:33:10 Quit amiconn (Disconnected by services)
04:34:09 Join pixelma [0] (quassel@rockbox/staff/pixelma)
04:34:18[Saint]In hindsight a call to update clients prior to commit (and ensuring it was carried out by at least one client) woukd have been nice.
04:34:26[Saint]But, mistakes happen.
04:36:00[Saint]Aha. Indeed. I need to wipe out my hackery before I can build again.
04:37:22[Saint]Even though the SDK/NDK are both current, I did some magic to make it think it had SDK16 instead of 22. So its not a happening thing until I get back home.
04:37:28 Quit Strife89 (Ping timeout: 240 seconds)
04:38:28jhMikeSWell, get home then!
04:38:29jhMikeS:)
04:39:30jhMikeS[Saint]: Is the andoid one always sent to yours or something? I haven't bothered to look.
04:39:57jhMikeSguess not
04:41:20[Saint]Nah. rasher (c/o TheSeven) and gevaerts iirc.
04:41:24jhMikeSIt's not exactly the first time it has done this by a long shot. I would think it would just return a failed build.
04:41:36[Saint]I thought it did.
04:41:46[Saint]It used to I'm sure.
04:41:55[Saint]...I think Im sure.
04:42:00jhMikeS6.5 hours and counting
04:42:14jhMikeSI mean 10.5
04:42:53jhMikeSthere used to be a kick function, but I never had access to that
04:43:45[Saint]logbot handles that, no?
04:43:54[Saint]Or, did?
04:45:27jhMikeSno idea
04:54:10Mode"#rockbox +o jhMikeS" by ChanServ (ChanServ@services.)
04:54:34Mode"#rockbox -o jhMikeS" by ChanServ (ChanServ@services.)
04:57:59 Quit us`0gb (Quit: http://0gb.us/)
05:00
05:03:21 Join us`0gb [0] (~0gb.us@c-50-137-60-25.hsd1.or.comcast.net)
05:44:53 Quit TheSeven (Disconnected by services)
05:45:06 Join [7] [0] (~quassel@rockbox/developer/TheSeven)
06:00
06:16:44***Saving seen data "./dancer.seen"
07:00
07:29:17 Join wqssaxz [0] (~wqssaxz@193.203.49.85)
07:36:28 Nick SuperBrainAK is now known as DormantBrain (~andy@2001:470:8:a61::5f92:59a1)
07:56:34 Join fragilematter [0] (~fragilema@unaffiliated/fragilematter)
07:57:26 Join [Saint_] [0] (~saint@rockbox/staff/saint)
07:57:28 Quit [Saint] (Read error: Connection reset by peer)
08:00
08:04:13 Quit wqssaxz (Remote host closed the connection)
08:11:50 Join ender` [0] (krneki@foo.eternallybored.org)
08:16:46***Saving seen data "./dancer.seen"
08:26:29 Join Zagor [242] (~bjst@rockbox/developer/Zagor)
08:30:33 Quit olspookishmagus (Quit: All for nothing)
08:32:44 Join kugel [0] (~kugel@193.174.67.16)
08:32:44 Quit kugel (Changing host)
08:32:44 Join kugel [0] (~kugel@rockbox/developer/kugel)
08:33:32kugelZagor: ping
08:33:38Zagorpong
08:34:59kugelZagor: the build server is stuck
08:35:06Zagorok
08:35:29 Join LinusN [0] (linus@giant.haxx.se)
08:36:03kugela new version of the android toolchain is required. how does it work?
08:36:07Zagorrestarted
08:37:15kugeljust update rbclient.pl?
08:37:15 Join olspookishmagus [0] (~pookie@snf-137798.vm.okeanos.grnet.gr)
08:38:28Zagoryes
08:39:12kugelso i have to add the android22 line and increment $revision?
08:43:08Zagorand update the 'builds' file of course
08:48:30kugelZagor: should the existing android toolchains be removed from rbclient.pl? I guess this will make rbclient.pl error out?
08:54:54kugelhm i can't push to gerrit for www
08:55:25kugelnevermind, my bad
08:56:23kugelZagor: does G#770 look good?
08:56:54kugeldidnt we have a ML just for build client affairs?
09:00
09:02:03 Join petur [0] (5bb7304d@rockbox/developer/petur)
09:08:51Zagornot 22 then?
09:09:34Zagorkeep the old 16 line for now
09:10:31Zagorwe have http://cool.haxx.se/cgi-bin/mailman/listinfo/rockbox-rbclient
09:11:43kugelZagor: could that be listed at http://www.rockbox.org/mail/ ?
09:12:17Zagoryeah, sure. it hasn't really been used for anything other than me reaching out to client owners.
09:12:31Zagoras in "I screwed up, you need to manually update"
09:18:02 Join kugel_ [0] (~kugel@193.174.67.16)
09:18:10 Quit kugel_ (Changing host)
09:18:10 Join kugel_ [0] (~kugel@rockbox/developer/kugel)
09:19:22 Quit kugel (Read error: Operation timed out)
09:20:34 Join kugel [0] (~kugel@193.174.67.16)
09:20:34 Quit kugel (Changing host)
09:20:34 Join kugel [0] (~kugel@rockbox/developer/kugel)
09:21:00 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
09:22:57 Quit kugel_ (Ping timeout: 240 seconds)
09:25:10 Join kugel_ [0] (~kugel@rockbox/developer/kugel)
09:25:53 Quit kugel (Remote host closed the connection)
09:26:23 Join kugel [0] (~kugel@rockbox/developer/kugel)
09:29:37GodEaterand you do that *soooooo* often ;)
09:30:00 Quit kugel_ (Ping timeout: 240 seconds)
09:33:01kugelZagor: still stuck
09:35:44Zagorno, I just haven't restarted the round
09:44:29 Join wodz [0] (~wodz@iwl138.internetdsl.tpnet.pl)
09:50:01wodzAlmost finished getting rid of plugin api from plugins (rb-> and mangling macros) in elfloader branch. I hope to start testing this soon.
09:56:40kugelZagor: well you said restarted so I assumed you did that :)
09:57:00Zagorah right
09:57:38kugelwodz: how does the linking work? by name?
09:58:40wodzkugel: yes, but with the twist that I use hashtable
09:59:19kugelso it's going to be quite heavy on memory?
09:59:28wodzkugel: The idea is the same as linux kernel modules
09:59:42kugel(I'm not implying that it's not worth it)
10:00
10:01:00wodzkugel: Not that much. Symbol names are in loaded elf and are not loaded into memory. Exported symbol resolve part is stored in form of precalculated hashtable so it is 2x size of simple pointer in vtable.
10:01:05wodzNot that bad I'd say
10:02:07wodzThere is some overlap of ci-> and rb-> interfaces also
10:03:42kugelhow do you export, or are all functions accessible?
10:04:40wodzonly explicitely marked functions are exported
10:06:11wodzkugel: It is the same idea as EXPORT_SYMBOL() in linux kernel
10:06:52kugelright, sounds cool
10:08:01kugelbut we still have to take care of PLUGIN_API/ABI_VERSION?
10:09:03wodzkugel: I am not that far to think about it yet. Reading about symbol versioning in linux it is said to not be that effective
10:09:57kugeli don't think we want symbol versioning
10:10:44kugelbut we still need to have a means to detect out-of-date plugins when functions are added/removed/changed
10:10:59kugelbtw, what hash algorithm do you use?
10:11:28wodznv1a
10:13:53wodz*fnv1a
10:16:49***Saving seen data "./dancer.seen"
10:17:28 Quit pamaury (Ping timeout: 240 seconds)
10:20:37kugelwodz: any reason to chose that over crc?
10:21:12wodzkugel: It produce considerably less collisions when hashing strings
10:21:59wodzmaybe this is not the problem for us since hashtable is rather small compared to possible hashes range. I'll check this maybe in the future
10:23:22kugelwhat makes you think that there are less collisions?
10:24:11wodzThere are papers about this with hard numbers. This topic is heavily tested in db context
10:24:18kugelsomeone made a comparison: http://programmers.stackexchange.com/questions/49550/which-hashing-algorithm-is-best-for-uniqueness-and-speed
10:25:00wodzkugel: I saw this. I read also dozen other papers and my general feeling is that crc32 is not bad but fnv1a is better for this task
10:26:06[7][Saint_], jhMikeS: my build client has the same setup as the nightly and android build scripts
10:26:51gevaertskugel: should be updated
10:27:33[7]no signs of a stuck build here
10:28:03gevaerts[7]: it's stuck on the server
10:28:14kugelwodz: alright, fnv1a is tiny enough to not worry about it
10:28:52kugelgevaerts: Zagor promised to restart it but I think he hasnt done that yet
10:29:23wodzkugel: Anyway it really simple to swap hashing function. I wouldn't worry about it now.
10:30:11kugelwodz: how do you look up the table? could sort by hash and then do binary search
10:30:29wodzkugel: That is exactly what I do
10:30:41kugelusing FS #12192 =
10:30:42fs-bluebothttp://www.rockbox.org/tracker/task/12192 3Introduce bsearch() and use it in tagtree.c. (patches, new)
10:30:42kugel?
10:31:01gevaertsZagor: can you check if the minumum client revision on the server is high enough? I can't remember where that lives
10:31:19[7]so we need android22 toolchains?
10:31:38kugelandroid19
10:31:45kugeli posted to the rbclient ml
10:32:06wodzkugel: no
10:32:31Zagorgevaerts: it lives in a local config file. I'll fix it.
10:33:12kugelwodz: I think it would be a good idea to use this, there are a few other spots where binary search is useful
10:36:42wodzkugel: I'll try to remember this FS but for now I really want to get it working rather then optimizing/merging with other FS and so on
10:37:06kugelsounds reasonable
10:37:37[7]kugel: which of those do I need? http://paste.pm/raw/efh
10:38:00kugelthe first 3
10:38:07[7]and 5 I guess?
10:38:20kugelyes
10:40:31kugelplus the latest ndk (r9d)
10:40:51[7]downloading that right now, and wondering what I need to change to make it actually use that...
10:41:41[7]I guess there's no point in keeping the old ndk around?
10:50:02kugel[7]: yea you can delete it
10:50:23kugelyou need to have the ANDROID_SDK/NDK_PATH point to the new location
10:52:03[7]ok, my build client and the nightly/rasher builds should have those new toolchains available now
10:52:36kugelwe'll hit the 1000min mark in a few!
10:52:57kugelpretty sure Zagor is just waiting for that!
10:53:35 Quit [7] (Remote host closed the connection)
10:54:25fs-bluebotBuild Server message: 3New build round started. Revision 6499ce3, 253 builds, 29 clients.
10:54:48Zagornag nag nag ;)
10:57:47 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
10:58:44 Quit [Saint_] (Remote host closed the connection)
10:59:28 Join TheSeven [0] (~quassel@rockbox/developer/TheSeven)
11:00
11:00:02 Join [Saint] [0] (~saint@rockbox/staff/saint)
11:01:09kugelwodz: how do you accomplish sorting the table?
11:01:22kugela seaprate step in the build? or can gcc be told to do it?
11:01:49kugelor even at runtime?
11:02:43TheSevenhowever it seems to have been idling during most of the round
11:02:58TheSevenouch
11:03:00TheSevenhttp://build.rockbox.org/shownewlog.cgi?rev=6499ce3;type=android480x800
11:03:09TheSeven/arm-linux-androideabi/4.6/../../../../arm-linux-androideabi/bin/ld: error while loading shared libraries: libstdc++.so.6: cannot open shared object file: No such file or make: directory
11:03:12wodzkugel: I patch rockbox.elf before objcopy step
11:03:44Zagorlibeatmydata??
11:03:59TheSeventhat's harmless
11:04:12gevaertsAlso some issues here. Investigating...
11:04:38gevaertsI seem to have done an androidmips build, so I don't know what's wrong now :)
11:05:09TheSeven/usr/lib/x86_64-linux-gnu/libstdc++.so.6 exists... but it seems to be looking for a 32bit one?
11:06:12kugeldid you download the 32bit ndk by chance?
11:07:05gevaertsHmmm
11:07:24TheSevenI downloaded http://dl.google.com/android/ndk/android-ndk-r9d-linux-x86_64.tar.bz2
11:07:49TheSevenbut it somehow seems to be still using an r8 one
11:08:04kugeldid you change ANDROID_NDK_PATH?
11:08:20*gevaerts is misreading logs
11:08:30TheSevenkugel: at least in .bashrc... no idea where else it is hiding ;)
11:10:19TheSevenfound another one for the dailys
11:17:14kugelhm, the buildsystem is stuck again (Zagor ^)
11:18:19Zagoryeah, nobody succeeds building android
11:18:21gevaertsYes, that's "normal"
11:18:25gevaertsI'm looking into it
11:18:38*gevaerts doesn't get it
11:19:00gevaertsI can build android and android-mips manually using the exact commands listed in the log
11:19:26gevaertsBut the server doesn't like something
11:19:26gevaerts2014-03-10 11:07:40 Server message: Fatal build error: Command not found. You have been temporarily disabled.
11:19:32gevaertsZagor: any idea which command?
11:20:33Zagorgevaerts: ../tools/configure: line 4124: mipsel-linux-android-cpp: command not found
11:20:37kugelZagor: shouldnt the build just fail?
11:20:42TheSevenhm....
11:20:49TheSevenlooks like I have 32bit executables in the SDK
11:20:53Zagorkugel: yeah, that's a bug in the server too. looking at that.
11:23:00kugelwasnt there a bug when the server finds no available clients for a particular build?
11:23:05gevaertsThing is, a manual build does work...
11:23:27Zagorkugel: yes but there is code to hangle that. for some reason it doesn't trigger here.
11:24:22kugelhmm. there are 2 clients available now, one of which is temporarily disabled
11:24:30kugelor are both disabled?
11:24:35Zagorboth
11:25:00gevaertsI don't think it makes sense to restart the server now. The clients need to be fixed first
11:25:51Zagorlooks like the code only handles disconnected clients, not blocked
11:26:01kugelZagor: the build table looks strange too: the bottom row revision is the one building right now, with a reported date of jan 1
11:27:31gevaertsThis makes no sense!
11:28:00gevaertsWith exactly the same PATH, ANDROID_NDK_PATH, and ANDROID_SDK_PATH, I can do a build just fine
11:28:07*TheSeven tried building android, and got this now: "Warning: The signer certificate has expired."
11:28:34kugelTheSeven: your debug cert is probably expired, it only lasts a year
11:28:44TheSevenhow to fix that?
11:28:45kugelbut this is not a problem as the build is not published
11:29:15kugeltry make keystore within an android build
11:29:26kugelthat should create a new debug cert
11:29:28TheSevenno such target
11:30:05kugeloh hm. delete ~/.android/debug.keystore
11:30:20kugelthen any make will recreate it
11:30:37gevaertsZagor: do you have more context for that error? I really don't see it...
11:31:35kugeljhMikeS: ping
11:31:36Zagorgevaerts: not really. this is the full log: http://pastebin.com/wbePCBcQ
11:31:53TheSevenZagor: can you unblock my client again? I think it might be working now
11:32:17TheSevenor wait, that just happened automatically a few minutes ago
11:32:22ZagorTheSeven: it's a temporary lock that is removed automatically
11:33:45TheSevenZagor: now can you tell me why this build blocked by client? http://build.rockbox.org/shownewlog.cgi?rev=6499ce3;type=android480x800
11:35:23ZagorTheSeven: it looks very similar to gevaerts problem: http://pastebin.com/WjeWfYTP
11:36:37TheSevenand why are those messages not present in the log file that I linked?
11:36:39kugelZagor: the log TheSeven linked to shows a successful build by hex-gevaerts
11:36:49TheSevenargh
11:37:32 Join Rower [0] (~husvagn@90-230-142-55-no41.tbcn.telia.com)
11:37:53*gevaerts gets more coffee
11:39:05TheSevenZagor: can you see if those android builds ever actually worked on my system?
11:40:11TheSevenmy logs suggest that it worked, but I can't see why it would fail now
11:40:22TheSevenunless something caches the environment rather badly (across a build server restart)
11:40:29TheSevens/server/client/
11:41:16fs-bluebotBuild Server message: 3Build round completed after 2811 seconds.
11:41:45jhMikeSkugel: que pasa?
11:42:16ZagorTheSeven: looks like it yes: 2014-03-09 18:12:44 Completed: build android480x800 client server-TheSeven seconds 62.5 uplink 0 speed 1393 time 65 left 183
11:42:38TheSevenarrrrrgh
11:42:42TheSevenfound the problem
11:42:55TheSevenNDK path hardwired in the upstart script :/
11:45:15TheSevenok, my client should be fixed now as well
11:47:15 Join sciopat [0] (~sciopath@yer91-2-82-237-54-159.fbx.proxad.net)
11:47:24[Saint]Hmmmm. Ok.
11:47:33 Join advcomp2019_ [0] (~advcomp20@unaffiliated/advcomp2019)
11:47:36TheSevenkugel: you broke the ipod classic codec driver?
11:47:42[Saint]Apparently "android" doesn't like /tmp mounted noexec.
11:47:59copperwhat
11:48:14copperno breaky :(
11:48:24Zagorgevaerts: any luck on your side?
11:50:43kugelTheSeven: the fix is in git already
11:51:09 Quit sciopath (Ping timeout: 264 seconds)
11:51:09 Quit advcomp2019__ (Ping timeout: 264 seconds)
11:51:10 Quit fragilematter (Quit: Leaving.)
11:51:32TheSevenhm, the build table doesn't think so for some reason... was it committed very recently?
11:52:15kugelyes
11:53:33TheSevenmight make sense to kick off a build round then ;)
11:55:43kugel[Saint]: iirc java has a problem with that
11:55:51kugelor gwt
11:56:17[Saint]It took me a while to figure out what it was crying about.
11:56:33 Quit jhMikeS (Ping timeout: 264 seconds)
11:56:38gevaertsZagor: no. I really have *no* idea
11:56:39kugelawt*
12:00
12:00:01gevaertsbah!
12:00:45gevaertsZagor: I can't do any more testing if there's no androidmips build pending :(
12:01:43[Saint]you have no way of initiating a build yourself, separate from the farm?
12:02:27gevaertsI could technically set up my own build master...
12:04:43Zagoreven if I restart the build now TheSeven might steal it from you
12:05:02Zagorgevaerts: you can do it with nc.
12:05:14Zagorthere is no authentication either way
12:06:33gevaertsIs there a quick protocol overview somewhere?
12:07:10Zagorno :)
12:07:50 Join ikeboy [0] (~ikeboy@ool-435622d3.dyn.optonline.net)
12:08:07gevaertsAh, got it :)
12:08:08Zagorbuild command format is "BUILD $id:$rev:mt:$builds{$id}{result}:$builds{$id}{upload}:$builds{$id}{cmdline}"
12:16:45gevaertsHmmmm
12:16:51***Saving seen data "./dancer.seen"
12:17:07 Join |akaWolf| [0] (~akaWolf@188.134.9.161)
12:17:10gevaertsZagor: is there any chance that the server keeps checking the same log file over and over again?
12:17:24gevaertsMy simulated build also doesn't have the issue
12:17:35kugelZagor: can you trigger a build round for git HEAD
12:17:36kugel?
12:22:03 Quit ikeboy (Quit: Leaving)
12:31:44 Join fragilematter [0] (~fragilema@unaffiliated/fragilematter)
12:51:04fs-bluebotBuild Server message: 3New build round started. Revision 4262e64, 253 builds, 29 clients.
12:55:01fs-bluebotBuild Server message: 3Build round completed after 237 seconds.
13:00
13:24:21 Join Scr0mple [0] (~Simon@161.43.73.67)
13:24:35 Quit Scromple (Read error: Connection reset by peer)
13:25:56 Join ZincAlloy [0] (~Adium@pD9EE8DEB.dip0.t-ipconnect.de)
13:32:20gevaertsLooks like I was right about that failure being somehow sticky on the server
13:36:59 Join ii_ [0] (~iidx50@31.220.27.198)
13:47:19 Join amayer [0] (~amayer@mail.weberadvertising.com)
13:57:13wodzii_: My point was that android port is pretty special. I am not an expert in hosted targets but it looks much more suitable to use linux hosted target and set ndk as build environment
13:57:51wodzandroid is generic, your work is very target specific
13:59:38ii_obviously I should create another target in tools/configure
13:59:53wodzii_: btw. in your functions poking 'magic' /sys files you don't close file descriptors
14:00
14:00:57 Quit kugel (Ping timeout: 240 seconds)
14:01:44ii_besides target, I will have to change compiler flags a little - should I make options in androidcc or create a separate variant?
14:04:36 Join kugel [0] (~kugel@193.174.67.16)
14:04:36 Quit kugel (Changing host)
14:04:36 Join kugel [0] (~kugel@rockbox/developer/kugel)
14:04:42ii_you don't close file descriptors: yep. found 2 in powermgmt-android.c and fixed
14:05:06wodzmaybe kugel has opinion about this, I personally don't know what is better
14:05:15 Join kugel_ [0] (~kugel@openvpn.HHI.FRAUNHOFER.DE)
14:05:15 Quit kugel_ (Changing host)
14:05:15 Join kugel_ [0] (~kugel@rockbox/developer/kugel)
14:09:12 Quit kugel (Ping timeout: 265 seconds)
14:16:55***Saving seen data "./dancer.seen"
14:18:00wodzii_: I guess you need to tweak LDOPTS and GLOBAL_LDOPTS, right?
14:18:45 Quit cmhobbs (Ping timeout: 264 seconds)
14:20:39ii_yes!
14:22:41wodzI'd go with defining androidx50cc()
14:32:50ii_Ok. By the way, how do I show 'shutdown' item somewhere in menu?
14:33:48ii_I see some conditions for that in root_menu.c, but why only for single device (#if CONFIG_KEYPAD == PLAYER_PAD) ?
14:35:15wodzii_: because other players shutdown implicitly by long pressing some key
14:36:12ii_this one either reboots or shuts down, so it's difficult to do exactly what you want
14:36:52ii_ok, I'll extend those conditions
14:38:02wodzif you know how to 'programically' shutdown you should be able to map long press of something to trigger
14:39:36ii_but how would I map that? I can't find any shutdown or power in action.h
14:40:20 Join kugel [0] (~kugel@rockbox/developer/kugel)
14:42:57 Quit kugel_ (Ping timeout: 240 seconds)
14:52:05wodzii_: I don't know how it is handled in hosted env
14:53:39wodzii_: look at rockbox/firmware/target/hosted/samsungypr/ypr0 for inspiration
14:56:38 Quit shamus (Quit: Leaving)
14:58:02ii_yes, this folder already inspires me for a week or so :) I've implemented void power_off, but now looking for how user could call that
14:58:45wodzin button-target.h you define which button shold trigger
14:58:50wodzshould
14:59:32wodz#define POWEROFF_BUTTON BUTTON_POWER #define POWEROFF_COUNT 10
15:00
15:00:41 Quit kugel (Ping timeout: 244 seconds)
15:02:20 Quit wodz (Quit: Leaving)
15:14:35 Quit [Saint] (Remote host closed the connection)
15:15:58 Join [Saint] [0] (~saint@rockbox/staff/saint)
15:25:19 Join ikeboy [0] (~ikeboy@ool-435622d3.dyn.optonline.net)
15:49:32 Quit ivanf (Remote host closed the connection)
15:49:46 Join ivanf [0] (~ivan@74.73.249.133)
15:49:46 Quit ivanf (Changing host)
15:49:46 Join ivanf [0] (~ivan@unaffiliated/kferdous)
15:54:38 Quit ZincAlloy (Quit: Leaving.)
15:56:38 Quit ii_ (Quit: ii_)
16:00
16:01:37 Quit linuxguy3 (Ping timeout: 252 seconds)
16:02:07 Quit us`0gb (Quit: http://0gb.us/)
16:03:43 Join ii_ [0] (~iidx50@78.24.28.191)
16:08:44 Join japc [0] (~japc@194.65.5.235)
16:11:20 Join linuxguy3 [0] (~timj@24-148-61-208.c3-0.lem-ubr1.chi-lem.il.cable.rcn.com)
16:16:58***Saving seen data "./dancer.seen"
16:30:47 Quit toehser (Remote host closed the connection)
16:38:07 Quit petur (Ping timeout: 245 seconds)
16:41:13 Join toehser [0] (~tom@Connqueror.Toms.NET)
16:42:00 Join kuldeepdhaka [0] (~kuldeepdh@unaffiliated/kuldeepdhaka)
16:54:27 Join kugel [0] (~kugel@rockbox/developer/kugel)
16:57:27 Quit Cultist (Ping timeout: 244 seconds)
17:00
17:00:09 Quit ikeboy (Quit: Leaving)
17:01:33 Quit Zagor (Quit: Clint excited)
17:11:54 Join ferros [0] (~frrs@89-76-97-49.dynamic.chello.pl)
17:12:01ferrosHi lads
17:12:07ferrosii_, you're doing god's work!
17:14:01ii_just wanted to make my headphones sound properly :)
17:14:42 Join wodz [0] (~wodz@89-75-151-160.dynamic.chello.pl)
17:15:03ii_I'd be really glad if someone could explain me what original MangoPlayer does to android vold service
17:15:24 Quit fragilematter (Quit: Leaving.)
17:15:54wodzii_: My advice is You decouple dx50 stuff from android and move it to some firmware/target/hosted/ibossodx50 or something. This will save you time later
17:16:32ii_because if MangoPlayer was not launched after reboot, vold is running but 'useless' because does not automount anything and its client (vdc) can't connect
17:17:14ii_but if you launch MangoPlayer, sd/usb start automounting and vdc works
17:17:20wodzii_: I don't see also why you need shutdown entry.
17:18:22ii_move it to some firmware/target/hosted/ibossodx5: yes, that's what I was going to do; not much work
17:19:23ii_the player has hardware shutdown/reboot, but it's difficult to control whether it will reboot or shutdown (depends on duration of press probably)
17:20:03ii_in original firmware you short-press power then click 'yes' on touchscreen
17:21:07wodzyou can adjust how long the key needs to be pressed to take shutdown action, no problem
17:22:48ii_actually not, because keys driver in kernel works somewhat special for power button: when you press the button, you get no event. then, if you release it before hardware shutdown, you get keydown+keyup events simultaneously
17:23:18wodzoh, thats unfortunate
17:23:41ii_we could use play button for that, but that would be a bit strange
17:24:01wodzbut key defines are bitfields actually, you can define UP|DOWN mask as POWER key
17:24:30wodzthis should trigger on keydown+keyup combo then
17:24:54wodzaka short power press if I understand correctly
17:27:07ii_trigger what? do you suggest using #define POWEROFF_BUTTON + #define POWEROFF_COUNT ?
17:27:52ii_or do you mean adding a line to keymap*.c ?
17:28:34wodz#define POWEROFF_BUTTON (UP|DOWN ) <- I don't know how you actually named keys masks and adjust #define POWEROFF_COUNT to be short enough to not trigger hw shutdown
17:28:40wodzii_:^
17:28:57wodzshould work
17:29:58ii_do I understand correct that POWEROFF_COUNT counts repeated states when poweroff_button is pressed?
17:30:17wodzAFAIK yes
17:30:33ii_if yes, it will not work, see the behavior of events i have described earlier
17:30:41ii_1 press+release -> 1 event
17:30:48wodzset it to 1 then
17:31:10ii_then it will be extremely easy to shut down
17:34:11ii_shut down root menu entry looks ok to me and works fine
17:34:42wodzdo as you feel fit.
17:36:13ii_ok
17:37:42 Quit ii_ (Quit: ii_)
17:39:57 Join y4n [0] (~y4n@unaffiliated/y4ndexx)
17:41:31 Quit the-kyle (Remote host closed the connection)
17:46:25 Join AlexP [0] (~alex@rockbox/staff/AlexP)
17:48:29 Join the-kyle [0] (~kyle@kyle.tk)
17:55:03 Quit Rower (Quit: Hmmm...)
18:00
18:03:03 Quit wodz (Quit: Leaving)
18:10:47 Quit the-kyle (Ping timeout: 240 seconds)
18:16:58 Join the-kyle [0] (~kyle@kyle.tk)
18:16:59***Saving seen data "./dancer.seen"
18:18:02 Join bertrik [0] (~quassel@ip117-49-211-87.adsl2.static.versatel.nl)
18:18:10 Quit bertrik (Changing host)
18:18:10 Join bertrik [0] (~quassel@rockbox/developer/bertrik)
18:22:14 Join ZincAlloy [0] (~Adium@pD9EE8DEB.dip0.t-ipconnect.de)
18:30:36 Quit copper (Ping timeout: 240 seconds)
18:45:16 Quit kugel (Ping timeout: 240 seconds)
18:49:20 Join copper [0] (~copper@unaffiliated/copper)
18:52:24 Join n1s [0] (~n1s@rockbox/developer/n1s)
19:00
19:02:21 Join bluebrother [0] (~dom@rockbox/developer/bluebrother)
19:04:27 Quit fs-bluebot (Ping timeout: 240 seconds)
19:05:13 Join fs-bluebot [0] (~fs-bluebo@g231120185.adsl.alicedsl.de)
19:05:35 Quit bluebrother^ (Ping timeout: 244 seconds)
19:16:30 Join ii_ [0] (~iidx50@78.24.28.187)
19:21:58 Quit pamaury (Read error: Operation timed out)
19:25:13 Quit japc (Ping timeout: 244 seconds)
19:29:08 Quit onder` (Quit: leaving)
19:29:46 Join onder` [0] (~onder@dyn-dsl-to-76-75-118-5.nexicom.net)
19:31:35 Join thomasjfox [0] (~thomasjfo@rockbox/developer/thomasjfox)
19:34:04 Join lebellium [0] (~chatzilla@89-93-178-161.hfc.dyn.abo.bbox.fr)
19:54:38 Join ikeboy [0] (~ikeboy@ool-435622d3.dyn.optonline.net)
20:00
20:05:16 Quit onder` (Ping timeout: 252 seconds)
20:06:00 Join onder` [0] (~onder@dyn-dsl-to-76-75-118-5.nexicom.net)
20:14:19 Quit ii_ (Quit: ii_)
20:17:03***Saving seen data "./dancer.seen"
20:18:28 Join ii_ [0] (~iidx50@78.24.28.190)
20:18:31 Join petur [0] (~petur@rockbox/developer/petur)
20:19:27 Part ii_
20:29:10 Quit thomasjfox (Remote host closed the connection)
20:49:33 Quit onder` (Read error: Operation timed out)
20:52:15 Join onder` [0] (~onder@dyn-dsl-to-76-75-118-5.nexicom.net)
21:00
21:01:10 Join japc [0] (~japc@bl21-170-210.dsl.telepac.pt)
21:11:59 Quit ikeboy (Remote host closed the connection)
21:13:53 Quit y4n (Quit: PÆNTS ØLF!)
21:34:06 Join Rower [0] (~husvagn@90-230-142-55-no41.tbcn.telia.com)
21:35:04 Quit Rower (Client Quit)
21:36:08 Quit krnlyng (Remote host closed the connection)
21:37:46 Join krnlyng [0] (~liar@83.175.90.24)
21:47:57 Quit kuldeepdhaka (Ping timeout: 240 seconds)
22:00
22:17:07***Saving seen data "./dancer.seen"
22:21:35 Quit |akaWolf| (Ping timeout: 240 seconds)
22:25:23 Quit petur (Quit: Leaving)
22:29:43 Quit amayer (Quit: Leaving)
22:32:15 Join kugel [0] (~kugel@91-65-145-11-dynip.superkabel.de)
22:32:15 Quit kugel (Changing host)
22:32:15 Join kugel [0] (~kugel@rockbox/developer/kugel)
22:37:23lebelliumkugel: rockbox.sh is OK, thanks
22:38:30kugellebellium: cool
22:47:04lebelliumkugel: but I don't see the FM menu anymore with the official build. I don't remember, wasn't it included in the code?
22:48:31kugellebellium: lorenzo92 didnt give me any code for that yet
22:48:44lebelliumhum okay
22:48:50lebelliumI'll see that with him
22:48:54kugelafaik it needs a kernel module
22:48:57lebelliummaybe it wasn't really ready
22:49:20lebelliumbut I have a si4709.ko file at the root
22:49:32lebelliumand this was necessary to get the FM menu in my previous build IIRC
22:49:54kugeltry to load it from rc.user
22:50:37kugeljust add the line "busybox modprobe /mnt/media0/si4709.ko" to that file
22:52:45lebelliumcan I access that file from my Win 7 file explorer or only from the device itself?
22:52:54 Join us`0gb [0] (~0gb.us@c-50-137-60-25.hsd1.or.comcast.net)
22:53:25kugelit's an ordinary text file, you can create it with a normal text editor
22:53:50kugelalthough it contains shell commands so you probably need to save it with unix line endings
22:55:50fs-bluebotBuild Server message: 3New build round started. Revision 50f0dd8, 253 builds, 29 clients.
22:57:19lebelliumbut where should be this rc.user file? At the root? Because from the rockbox file browser I can see etc\rc.d but I can't access those directories from my computer
22:57:37 Join Cultist [0] (~CultOfThe@67.186.111.33)
22:59:15fs-bluebotBuild Server message: 3Build round completed after 205 seconds.
23:00
23:05:36 Join ii_ [0] (~iidx50@78.24.28.187)
23:05:47lebelliumhum shouldn't be at the root apparently
23:14:54kugellebellium: on the internal fflash
23:15:19lebelliumI can't access it from my computer!
23:15:51lebelliumand I can't do anything from the R1, the text viewer and text editor plugins are not there with the current build
23:15:56kugelthe internal flash is what's mounted as R1 drive
23:15:57 Quit n1s (Quit: Ex-Chat)
23:16:16lebelliumheh
23:16:46 Quit the-kyle (Remote host closed the connection)
23:16:53lebelliummedia0/rc.user is wrong?
23:16:54kugelyou can't modify anything else (well, except /tmp) because the root filesystem is read-only
23:17:47kugellebellium: drop the file on the root of the USB accesible storage
23:17:58lebelliumthat's media0
23:18:00kugelyes
23:18:01lebelliumand that doesn't work
23:18:22kugelin what way?
23:19:00lebelliumin a way that I don't see the difference between before and after putting the file, i.e. the FM menu doesn't show up
23:19:04 Join the-kyle [0] (~kyle@kyle.tk)
23:20:09kugeldoes the folder /sys/module/si4709 exist?
23:20:51lebelliumthere is sys/module/si470x
23:21:38kugelah, that needs to be removed
23:22:19kugeladd "busybox rmmod si470x" to rc.user (before the other line)
23:24:53lebelliumno luck
23:25:04kugeldoes the directory exist now?
23:26:08lebelliumsys/module/si470x is gone but no sys/module/si4709
23:26:30kugelokay
23:26:35kugeldont know why it doesnt load
23:27:04lebelliumprobably something tricky lorenzo made just to try out
23:27:09lebelliumI'll ask him next time
23:28:51lebelliumI remember the radio screen was there so that I can make my theme FMS
23:29:04lebelliumbut I don't remember if the FM radio was actually working yet
23:30:44 Quit ferros (Ping timeout: 264 seconds)
23:37:03 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
23:39:59 Join jhMikeS [0] (~jethead71@c-68-43-2-35.hsd1.mi.comcast.net)
23:39:59 Quit jhMikeS (Changing host)
23:39:59 Join jhMikeS [0] (~jethead71@rockbox/developer/jhMikeS)
23:42:33 Quit lebellium (Quit: ChatZilla 0.9.90.1 [Firefox 28.0/20140306171728])
23:42:44 Quit kugel (Ping timeout: 264 seconds)
23:57:21TheSevenseems like I didn't get any android builds throughout the day...

Previous day | Next day