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 2014-01-20

00:00:02saratoga_i was going to try and work on the MDCT stuff for it, but no time until feb due to work and research
00:00:56preglowyeah, work and trying to be a studio engineer will be taking my time for some months to come
00:01:14preglowarm is always nice :>
00:01:29saratoga_they want a NEON version of the FFT
00:02:58preglownever even touched neon, so that would be cool :)
00:03:17preglowi wonder what arm device would be nicest for dev work
00:03:51saratoga_i bought a raspberry pi last year
00:03:57preglowgot a couple lying around at work
00:04:11saratoga_its nice because i can compile on it, but i'd probably try to get something with NEON too
00:04:19preglowbut tool wise an apple device would probably be nicer
00:04:32saratoga_will probably try setting up debian on my old nexus since it has an A9
00:05:05preglowi'd very much like to cross compile and upload code instead of working on target
00:05:45saratoga_the pi takes forever to do a make clean, but otherwise its quite fast because i don't have to copy anything on subsequent builds
00:06:07saratoga_not bad for 700MHz ARM11 really
00:06:53saratoga_but i guess the best option would be somethign with cross compiling and some kind of remote debug, but i don't know anything about that
00:07:14preglowdone some ipad app development, and that's like that
00:07:56preglowi assume android has it too
00:10:04 Quit bertrik (Ping timeout: 252 seconds)
00:12:29 Quit wodz (Quit: Leaving)
00:36:01 Quit lebellium (Quit: ChatZilla 0.9.90.1 [Firefox 27.0/20140116125114])
00:36:28 Quit uwe_ (Ping timeout: 252 seconds)
00:41:14 Quit W00fer ()
00:42:14 Join uwe_ [0] (~uwe_@dslb-088-066-160-105.pools.arcor-ip.net)
00:56:27 Quit ender` (Quit: I know a joke about UDP, but you might not get it.)
00:56:54***Saving seen data "./dancer.seen"
00:56:55[Saint]saratoga_: factory OC to 1GHz makes quite an improvement in compile time
00:57:28[Saint]300MHz doesn;t seem like heaps, but its essentially adding 1/3 more CPU to the wee thing.
00:58:11*[Saint] can't math
00:58:19[Saint]~50% more CPU
00:58:31saratogais that easy to do?
00:59:29[Saint]Sure. running rpi-config (gui), or hand editing /boot/config.txt
00:59:44[Saint](the config is well commented)
00:59:48saratogaoh cool, will look into it
01:00
01:00:02[Saint]You can also adjust the CPU/GPU memory split there.
01:00:28[Saint]By default, 64MB is reserved for the GPU, but if you're not using it...you can carve it right back to 8MB iirc.
01:01:21[Saint]saratoga_: http://elinux.org/RPiconfig
01:02:19*[Saint] also really recommends using USB storage for /
01:02:41[Saint]Get everything except /boot off the sdcard
01:03:55[Saint]You can go as high as 1.2GHz, but stability varies greatly between devices and this will set the warranty bit (if you care about that).
01:05:38[Saint]There's also a "force_turbo" mode that disables dynamic clocking and allows for uninhibited overvolting.
01:06:07[Saint]The power usage hit from forcing the maximum CPU/GPU clocks is actually really trivial too.
01:08:00[Saint]My setup is arm_freg=1200, core_freq=500, sdram_freq=600, over_voltage=8
01:09:00[Saint](200MHz shy of running 200% OC stable)
01:09:02fs-bluebotGerrit review #200 at http://gerrit.rockbox.org/r/200 : Fundamentally rewrite much of the audio DSP. by Michael Sevakis (changes/00/200/17)
01:09:37[Saint]bluebrother: I just found another fs-bluebot quirk :)
01:09:42*[Saint] kicks fs-bluebot
01:10:01[Saint](123
01:10:14[Saint]123%
01:10:25[Saint]Hmmm.
01:10:42[Saint](123 123%)
01:11:01[Saint](200MHz shy of running 200% OC stable)
01:11:02fs-bluebotGerrit review #200 at http://gerrit.rockbox.org/r/200 : by Michael Sevakis (changes/00/200/17)
01:11:09[Saint]what the?
01:11:17Soukyuuyou have to kick it more
01:11:51[Saint]test (123MHz)
01:12:06[Saint]test (123MHz 123%)
01:12:28[Saint]What is it matching on? Gah!
01:12:50Soukyuustable) ?
01:13:03Soukyuuhmm doesn't seem so
01:13:13gevaertsg 200
01:13:14fs-bluebotGerrit review #200 at http://gerrit.rockbox.org/r/200 : Fundamentally rewrite much of the audio DSP. by Michael Sevakis (changes/00/200/17)
01:13:22gevaertseasy :)
01:13:22[Saint]Hmmm, maybe the task "123" doesn't exist
01:13:42[Saint](412
01:13:51 Quit fs-bluebot (Remote host closed the connection)
01:13:52[Saint]412%
01:14:04Soukyuulol
01:14:06[Saint]Hehe...I think I broke him.
01:14:08 Join fs-bluebot [0] (~fs-bluebo@g224236191.adsl.alicedsl.de)
01:14:52[Saint](412Mhz
01:15:01[Saint]bah.
01:15:10Soukyuucurious, do bug reports on the flyspray take a while to appear on the overview page or does an admin have to approve it first for new users?
01:15:33Soukyuu*tasklist, not overview
01:15:42gevaerts[Saint]: what are you trying to do?
01:16:09[Saint]gevaerts: Oh, derp. Neverming.
01:16:16[Saint]I just figured it out. Whoops.
01:16:22[Saint]I didn;t see your post.
01:16:34gevaerts:)
01:16:35[Saint]It picked out the g from "running"
01:17:03[Saint]So, "running 412" should do it.
01:17:04fs-bluebotGerrit review #412 at http://gerrit.rockbox.org/r/412 : Option to use the lighter hinting algorithm (FT_LOAD_TARGET_LIGHT). by Nick Peskett (changes/12/412/1)
01:17:07gevaertsIt should check for a word boundary there, really
01:17:07[Saint]Aha.
01:17:12*[Saint] nods
01:17:43[Saint]I've asked to be nosey at the source a few times, but bluebrother sems content maintaining it himself.
01:17:47[Saint]Which is fine.
01:18:42[Saint]Soukyuu: Are you sure you don;t just have the list ordered incorrectly?
01:18:49[Saint]I can see your task fine.
01:18:58[Saint]*after* I order the list by recent addition.
01:19:04Soukyuu[Saint]: silly me, just noticed
01:19:46SoukyuuI expected it to be ordered by date by default i guess
01:20:45*[Saint] isn;t terribly sure if that is a bug or not.
01:21:16Soukyuumy report or the list ordering
01:21:23[Saint]Your report.
01:21:24gevaertsIt's not
01:21:28gevaertsoh
01:21:33*gevaerts shuts up
01:22:17Soukyuu[Saint]: I just don't see why it would need the peak tag to simply apply the gain
01:22:46Soukyuuand foobar2000 does get it to work somehow, unless it calculates peak on the fly or something
01:22:49[Saint]Assumedly because we follow spec.
01:23:23[Saint]replaygain spec requires peak gain and amplitude to be in the metadata.
01:24:33[Saint]So this probably falls into the category of "not necessarily a bug, but definitely not immediately obvious to Jow User"
01:24:40[Saint]*Joe
01:25:15Soukyuuhmm, well let's say it's not a bug, but does it have to fail to apply the gain if clipping according to peak is off?
01:25:43Soukyuupersonally I don't see a reason to simply not apply clipping protection if peak not present
01:26:21Soukyuui think there is a "not" missing there somewhere
01:27:26Soukyuuor not, my English skills reduce late at night...
01:27:58[Saint]I have no idea how foobar handles this, but, its definitely not within spec.
01:28:12saratogaprobably the parser looks for the peak tag or something
01:28:51[Saint]...huh?
01:29:18saratogaprobably our tag parser expects the peak tag to be there and gives up if its not
01:29:33[Saint]Oh, right. Yes. Which is perfectly sane.
01:29:43[Saint](I thought you were referring to foobar)
01:29:52Soukyuushouldn't it be checking for the _gain_ tag instead though?
01:30:04Soukyuuand peak in case of clipping prevention?
01:30:28 Quit uwe_ (Ping timeout: 248 seconds)
01:31:04 Join uwe_ [0] (~uwe_@dslb-088-066-160-105.pools.arcor-ip.net)
01:31:16saratogayeah, looks like if its id3v2.4 it requires both to be there
01:31:45*[Saint] nods
01:31:45Soukyuuit happens with ape and vobist comments as well
01:32:05Soukyuui might be wrong but isn't applying gain done by simply adjusting each ... sample i guess?, by the value specified in the gain tag? from my view, I dont see the peak tag being involved in this at all
01:32:39[Saint]In this specific case, it isn't, no - but that's not the point.
01:33:00[Saint]The point is that we follow the specification, which these files break.
01:33:09[Saint](and foobar apparently does too)
01:33:33saratogahmm actually i'm not sure now that i look at this more
01:34:29Soukyuu[Saint], just for reference, where do i find the spec? i seem to only get wiki/HA links which don't go into detail
01:34:44Soukyuuah
01:34:54Soukyuunvm, found it
01:35:03[Saint]As far as I'm aware, the replaygain specification requires 4 tags - REPLAYGAIN_TRACK_GAIN, REPLAYGAIN_TRACK_PEAK, REPLAYGAIN_ALBUM_GAIN, and REPLAYGAIN_ALBUM_PEAK
01:35:37[Saint]Any one of those missing /should/ make replaygain fall over.
01:35:58Soukyuuyes, reading it at the moment
01:36:02[Saint]...but, I am far from absolutely certain on this.
01:36:19Soukyuuif this is up to date, then it does: http://wiki.hydrogenaudio.org/index.php?title=ReplayGain_specification#Metadata
01:36:37Soukyuu"For ReplayGain to do its work during playback, four values must be stored"
01:37:11[Saint]As for ape/vorbis comments, ...they're just insane in general.
01:37:21[Saint]I'm not sure there's a "right" way to do anything with them. :)
01:41:04saratogaactually i think its not the tag parsers, i think instead the dsp core is passed a peak value of 0, which is rejected as out of range
01:41:44Soukyuudoes that mean there is a chance it's an undesired behavior and might get fixed? :D
01:41:58saratogahmm no that also seems wrong
01:42:01 Join Strife89 [0] (~Strife89@adsl-98-80-237-249.mcn.bellsouth.net)
01:42:15saratogait should probably be fixed
01:42:44[Saint]By my understand this behavious is "correct".
01:42:50[Saint]*behavior
01:43:24Soukyuuby my understanding it's non-intuitive, but i'm just a user :x
01:44:18[Saint]I think we should probably just expect users to tag their files sanely, and retag after transcoding/whatever.
01:44:34[Saint]Then, it'll work as expected across the board.
01:44:47saratogai think you'd have to put some debug statements in the sim and print out the value at a couple points to see if its the parser or the dsp engine that complains, but if it happens to a couple different formats, its probably the dsp
01:45:14Soukyuuso far it happened with mp3, flac and mpc
01:45:39[Saint]That would be expected.
01:45:44Soukyuuso, id3v2.3 (utf-8), apev2 and vorbis comments
01:46:21Soukyuubut peak is not involved in gain calculation, at least as the spec states it
01:46:51[Saint]Instead of working around this, it seems much saner to me that you should just re-scan for replaygain.
01:47:25saratogathe gain is potentially scaled by the peak value, so its reasonable to make sure the peak value isn't zero
01:47:38saratogaprobably somewhere the code just rejects the whole thing if any of the values are out of range
01:47:48saratogayou'd have to grep
01:48:00[Saint]...I can't really fault the logic.
01:48:12Soukyuubut that's only in case clipping prevention is on, right?
01:48:16[Saint]"Have propper tags, it'll work"
01:48:37Soukyuuproblem is, applying those tags takes as long as copying the files over
01:48:38[Saint]remove half of them? well...here we are.
01:48:45Soukyuui do understand the point that there have to be 4 tags in total (or rather, a pair of gain+peak), but that's to "ensure full functionality" of RG. So partial functionality should still be preserved imho *shrug*
01:48:51chrisjj[Saint]: re "the replaygain specification requires 4 tags". It is self-contradictory. Yes according to section "Metadata", No according to "Loudness normalization".
01:49:41chrisjjSoukyuu: "If the file only contains one of the replay gain adjustments (e.g. Album) but the user has requested the other (Track), then the player should use the one that is available (in this case, Album)."
01:50:58Soukyuuchrisjj: yes, but what i'm asking is not about track vs album gain when one missing
01:51:11Soukyuuinstead what is gain is present, but _peak_ isn't
01:51:22Soukyuuloudness normalization only requires the gain tag... peak is only mentioned in clipping prevention section of the spec
01:54:36Soukyuusaratoga: i have a linux vm, but i haven't worked with rockbox code before, do I have to set it up as mentioned in the wiki to run the debug output or is it done on device?
01:57:42saratogaSoukyuu: if you want to open up this mess, the /lib/rbcodec folder has /metadata and /dsp which is where all this happens
01:57:54saratogai would build a simulator (doesn't matter which device)
01:58:09saratogathen you can use DEBUGF to print values to the console
01:58:55saratogamight be interesting to see if the gain values are correct here: http://git.rockbox.org/?p=rockbox.git;a=blob;f=lib/rbcodec/dsp/dsp_misc.c;h=ad6f5b5b3107d1c3bb0028db73fce1aea6498a73;hb=HEAD#l40
01:59:05saratogaanyway i have to get back to work
01:59:28Soukyuuthanks, will take a look
02:00
02:00:48saratogaoh and we don't have floating point, so those values will probably be in the form of scaled integers (e.g. some big power of 2 times the actual value)
02:15:14Soukyuuha, funny. if i delete only the album peak or only the track peak, then RG still works - and it applies track gain even though there is no track peak apparently. (tested by listening)
02:21:06chrisjjSoukyuu: "[when] gain is present, but _peak_ isn't" That's outside what's specified by the spec.
02:24:24Soukyuuchrisjj: I guess it is.
02:25:13Soukyuuthanks for the answers, I will try to persuade the fb2k devs to conform with the specs, as having only gain tags is pretty much useless
02:25:42 Quit Soukyuu (Quit: goodnight)
02:33:14[Saint]Bah. He's gone.
02:33:31[Saint]The other situation he describes is expected also, but also non-obvious.
02:34:06[Saint]I seem to recall album/track gain will fall back to each other if the other one isn't present and the user tries to set it.
02:46:27 Quit saratoga_ (Quit: Page closed)
02:56:56***Saving seen data "./dancer.seen"
03:00
03:18:00 Quit babylonlurker (Read error: Operation timed out)
03:32:20 Join babylonlurker [0] (~quassel@veda.xs4all.nl)
03:43:32 Join Cultist [0] (~CultOfThe@c-67-186-111-33.hsd1.il.comcast.net)
03:45:05 Quit krabador (Quit: Sto andando via)
04:00
04:01:46 Quit [Saint] (Remote host closed the connection)
04:04:17 Join [Saint] [0] (~saint@rockbox/staff/saint)
04:41:45 Join pixelma_ [0] (pixelma@rockbox/staff/pixelma)
04:41:45 Quit pixelma (Disconnected by services)
04:41:46 Nick pixelma_ is now known as pixelma (pixelma@rockbox/staff/pixelma)
04:41:46 Quit amiconn (Disconnected by services)
04:41:46 Join amiconn_ [0] (amiconn@rockbox/developer/amiconn)
04:41:48 Nick amiconn_ is now known as amiconn (amiconn@rockbox/developer/amiconn)
04:45:21 Join popole [0] (~popole@193.104.110.37)
04:57:00***Saving seen data "./dancer.seen"
05:00
05:14:04 Quit popole (Remote host closed the connection)
05:50:08 Quit [7] (Disconnected by services)
05:50:21 Join TheSeven [0] (~quassel@rockbox/developer/TheSeven)
06:00
06:25:43 Quit [Saint] (Remote host closed the connection)
06:26:45 Join [Saint] [0] (~saint@rockbox/staff/saint)
06:27:09 Join mortalis [0] (~kvirc@213.33.220.118)
06:57:01***Saving seen data "./dancer.seen"
07:00
07:53:56 Join LinusN [0] (linus@giant.haxx.se)
07:55:57 Join fragilematter [0] (~fragilema@unaffiliated/fragilematter)
08:00
08:19:54copper"having only gain tags is pretty much useless"
08:20:45copper"I just don't see why it would need the peak tag to simply apply the gain"
08:21:10copperI don't get it
08:21:33copperindeed peak information is only useful when using the "prevent clipping according to peak" functionality
08:22:30copperand if the peak value is missing, I'd just assume it's 1.0 FS
08:23:10copperand if the result clips, users can only blame themselves
08:26:33 Join ender` [0] (krneki@foo.eternallybored.org)
08:26:45copperit's all about sensible defaults and failing gracefully, IMO
08:27:25copper"doesn't work at all when less than 100% of expected data is there" is a lazy approach
08:36:02 Join einhirn [0] (~Miranda@bsod.rz.tu-clausthal.de)
08:38:03 Join tertu [0] (~tertu@143.44.70.199)
08:38:31 Join Zagor [242] (~bjst@rockbox/developer/Zagor)
08:40:12 Quit sakax (Quit: Leaving)
08:44:40 Join bertrik [0] (~quassel@ip117-49-211-87.adsl2.static.versatel.nl)
08:44:40 Quit bertrik (Changing host)
08:44:40 Join bertrik [0] (~quassel@rockbox/developer/bertrik)
08:46:02copperactually know, it would make more sense to disable clipping prevention
08:46:07copperno*
08:46:39copperso as to not alter the gain based on an arbitrary value
08:47:00 Join leftright [0] (c5e0116e@gateway/web/freenode/ip.197.224.17.110)
08:52:59leftrightSlasheri : nls ; LinusN ; H140. rockbox hangs during boot with any version after v3.8 with boot set to either Rom or Ram, from disk it will freeze on first boot and load after reset. it seems that it's related to this http://www.rockbox.org/tracker/task/12769
08:55:08leftrightIf unable to fix please make a note somewhere that roxkbox won't boot from flash with latest versions of roxkbox,
08:56:42Slasherileftright: that should be fixable, i havent used upgraded rockbox on my h140 for quite some time. And i don't know how to do commits to the new git system
08:57:02***Saving seen data "./dancer.seen"
08:58:07leftrighthello Slasheri:, long time no see. thanks for your prompt response. perhaps some other guru can be pointed in the right direction ?
09:00
09:02:11leftrightint var1; {bootfreeze} > {fix pronto} :-)
09:02:20Slasherihello :) most likely something has changed in the rockbox boot sequence, system initialization, memory layout or something like that. So bootloader still works as it always has worked but changes in the rockbox are incompatible with the old bootloader and the boot process crashes at the second stage when bootloader gives control to the rockbox application
09:06:31leftrightthanks for your input. later
09:06:36 Part leftright
09:14:49 Quit bertrik (Remote host closed the connection)
09:52:14 Join wodz [0] (~wodz@iwl138.internetdsl.tpnet.pl)
09:55:56wodzhmm it seems rolo is affected as well on h1xx so this should make testing simpler
09:56:57 Quit nosa-j (Ping timeout: 265 seconds)
09:57:15 Join nosa-j [0] (~m00k@66.233.224.206)
10:00
10:04:43 Join petur [0] (~petur@rockbox/developer/petur)
10:15:09wodzhmm, my fix to rom.lds is braindead (although it should work)
10:17:47 Quit nosa-j (Ping timeout: 260 seconds)
10:20:26 Join nosa-j [0] (~m00k@66.233.224.206)
10:30:18 Quit nosa-j (Ping timeout: 265 seconds)
10:53:40 Part LinusN
10:56:13 Quit ParkerR (Ping timeout: 240 seconds)
10:57:04***Saving seen data "./dancer.seen"
10:57:04 Join ParkerR [0] (~ParkerR@unaffiliated/parkerr)
11:00
11:08:12 Quit dfkt (Ping timeout: 252 seconds)
11:20:21 Quit mc2739 (Ping timeout: 248 seconds)
11:22:27 Join mc2739 [0] (~mc2739@rockbox/developer/mc2739)
11:46:06 Join leftright [0] (c5e015a4@gateway/web/freenode/ip.197.224.21.164)
11:49:58 Nick SuperBrainAK is now known as DormantBrain (~andy@2001:470:8:a61::5f92:59a1)
11:53:20leftrighthi wodz. I tried the latest build but my h1xx still exhibits the same boot-freeze as before.
11:54:04 Join dfkt [0] (OxO29A@unaffiliated/dfkt)
11:54:16wodzleftright: that is expected. Something else is broken. BTW. does rolo() exhibit the same problem for you?
11:54:37leftrightRoLo works ok
11:54:50wodzthat is super weird then
11:55:36wodzIn which way it crashes then? Do you see anything on the screen?
11:55:52leftrightto calirfy, Rolo executes every time from within rockbox and after a new version of rockbox
11:56:42leftrightit freezes on the boot screen which has rockbox's logo and v number
11:57:08wodzwhich number? bootloader's number or build number?
11:57:27leftrightbuild number
11:58:18wodzok and then you hard reset and the very same build boots ok, right?
11:58:24leftrightyes
11:58:38leftrightboot set to disk
11:58:50wodzright
11:59:58wodzwould it be possible for you to bisect and find out which commit introduced this?
12:00
12:00:34leftrightv3.8 works fine. everything after freezes
12:02:03wodzthat is a bit too broad for me to conclude anything
12:03:25leftrightwodz, sorry but I'm not into zero's and one's, but I'll be glad to give moral support :-)
12:05:06wodzleftright: by bisect I meant http://www.rockbox.org/wiki/UsingGit#How_do_I_bisect_using_git_63 and test the build to pinpoint offending commit
12:06:46leftright*runs for the hills* :-)
12:07:42leftrightI'll have a look at the link and see if I can deduce somethiing of value, might take a few days though
12:11:24leftrightthanks for having a look at it, much appreciated. Later
12:11:28 Part leftright
12:44:25chrisjjAnyone got a simulator to run recently? I get just fails: http://forums.rockbox.org/index.php/topic,47099.0.html
12:57:08***Saving seen data "./dancer.seen"
13:00
13:01:18 Quit alucryd (Quit: Alucryd out.)
13:08:36 Join alucryd [0] (~alucryd@archlinux/trusteduser/alucryd)
13:09:36 Join ikeboy [0] (~ikeboy@ool-435622d3.dyn.optonline.net)
13:19:24 Quit ikeboy (Quit: ikeboy)
13:55:48 Quit Scall (Ping timeout: 252 seconds)
14:00
14:04:19 Join Scall [0] (~chat@unaffiliated/scall)
14:16:32 Join ibhan [0] (~yaaic@94.191.189.75.bredband.3.dk)
14:31:00 Part ibhan
14:47:54 Join amayer [0] (~amayer@mail.weberadvertising.com)
14:57:10***Saving seen data "./dancer.seen"
15:00
15:00:05 Join [Saint_] [0] (~saint@rockbox/staff/saint)
15:00:55 Quit [Saint] (Ping timeout: 252 seconds)
15:16:23 Quit wodz (Quit: Leaving)
15:32:56 Join [Saint] [0] (~saint@rockbox/staff/saint)
15:34:16 Quit [Saint_] (Ping timeout: 272 seconds)
15:37:20 Quit Strife89 (Ping timeout: 272 seconds)
16:00
16:13:40 Quit mortalis (Ping timeout: 248 seconds)
16:16:26 Join Rower [0] (husvagn@90-230-142-55-no41.tbcn.telia.com)
16:20:02 Join ikeboy [0] (~ikeboy@ool-435622d3.dyn.optonline.net)
16:37:11 Join fragilematter1 [0] (~fragilema@193.226.142.214)
16:39:31 Quit fragilematter (Ping timeout: 260 seconds)
16:39:52ikeboydoes rockbox have any Easter eggs?
16:45:31 Join Strife89 [0] (~Strife89@2602:306:250d:18b9:29c7:db30:247e:f805)
16:46:58dfkti found the "rockbox ata error -2" easter egg once :p
16:57:14***Saving seen data "./dancer.seen"
17:00
17:00:12 Quit fragilematter1 (Quit: Leaving.)
17:00:40 Quit Zagor (Quit: Clint excited)
17:01:14 Quit einhirn (Quit: Miranda IM! Smaller, Faster, Easier. http://miranda-im.org)
17:10:06toehserRef FS12942 in skin_backdrops.c in skin_backdrop_unload - working case comes in with ref_count−− going from 1 to 0 and unloads, failing case old theme comes in with ref_count−− going from 2 to 1 and doesn't unload. At that point, I'm confused. What are the multiple contexts that need references to backdrops, what are the places incrementing, what is the intention when a new theme loads - I'm guessing that point A and point B of the old theme used the b
17:10:57 Quit Raptors (Read error: Connection reset by peer)
17:11:49 Join Raptors [0] (~whoneedsa@198-200-68-213.cpe.distributel.net)
17:40:56 Quit Strife89 (Ping timeout: 272 seconds)
17:44:01 Quit Raptors (Read error: Connection reset by peer)
17:44:24 Join Raptors [0] (~whoneedsa@198-200-68-213.cpe.distributel.net)
17:44:28ikeboyI'm having a theme bug with 91ef65306bf4e459f430d6dc44e5923d6b9f8399
17:53:36 Quit petur (Quit: *plop*)
17:54:44 Join kostyat [0] (~kostyat@91.210.101.68)
17:56:06copperikeboy: there have been some important theme fixes since that revision
17:56:19copperyou should get HEAD
17:57:14copperthose fixes mean that some themes are no longer compatible with HEAD
17:57:28copperI had to update my own themes, but now they work a lot better
17:58:56 Join Strife89 [0] (~Strife89@2602:306:250d:18b9:114:708:b733:508)
17:59:06copperin other words, it's worth it
18:00
18:02:29ikeboyThe bug is still on HEAD for me
18:02:47coppercan you post a screenshot from the sim?
18:02:49ikeboyand I didn't make these themes, I'd have no idea what to do
18:02:55copperalso, which theme / target?
18:03:50ikeboyHang on, I'll try to get a sim screenshot. Fuze+, Googley-FuzePlus-Blue
18:03:57copperthat's mine
18:04:54copperikeboy: to take a screenshot in the sim, press 0 on the keypad
18:05:01copperand your keyboard keypad
18:05:14copperalso, did you get the latest version of the theme?
18:05:30ikeboyno to latest version
18:05:46ikeboyis HEAD b3abcb8?
18:06:09copperyes
18:06:30copperthe updated theme is available here: http://themes.rockbox.org/index.php?themeid=2029&target=sansafuzeplus
18:07:21copperlet me know if you still have issues with the updated theme
18:08:01 Join y4n [0] (~y4n@unaffiliated/y4ndexx)
18:08:42 Quit kostyat (Ping timeout: 260 seconds)
18:14:08amayercopper: I was just going to mention your theme was broken on head... the fonts didnt show up under wps. but your most recent version fixes it
18:14:16amayeryou get a 5 star from me
18:14:29copper:)
18:14:46copperI updated the themes as kugel was fixing a gazillion bugs
18:15:20amayerwhen I get time I should go back and check my themes... one more thing on my loooong list of things to do
18:15:45copperthe previous version of my themes had fragile workarounds for the theme bugs that kugel fixed
18:15:55coppernow both Rockbox and my themes are fixed
18:16:26copperbut there's a chance that other themes need to be fixed too
18:16:51amayerI checked epilogue and it seems to be working. I would have to say that is the most popular of my themes.
18:19:36 Join pretty_function [0] (~sigBART@123.252.214.18)
18:21:43ikeboyThe new theme works. Thank you!
18:22:49coppergreat!
18:22:49 Join bertrik [0] (~quassel@ip117-49-211-87.adsl2.static.versatel.nl)
18:22:49 Quit bertrik (Changing host)
18:22:49 Join bertrik [0] (~quassel@rockbox/developer/bertrik)
18:25:06amayercopper: your theme has survived on my iPod longer then any other theme, even my own.
18:25:18amayerall tho I do modify your theme slightly
18:25:47amayerI make the title of the menus blue and eliminate the icon beside the title
18:29:01ikeboyI use the Googley theme with create.bmp background and elisherer iconset
18:30:03amayerdoes anyone else get a half a second hiss when pausing and resuming ogg files?
18:32:16amayerI dont get it when playing mp3s. Ive tried it with multiple ogg files and multiple mp3s.
18:57:18***Saving seen data "./dancer.seen"
19:00
19:00:10 Join Icedman [0] (~Icey@unaffiliated/icedman)
19:00:11Icedmanhi
19:00:55IcedmanI just installed rockbox on my sansa clip zip and since then the sansa won't get mounted as an usb device on linux. windows and mac
19:01:08Icedman:(
19:01:32bertrikdoes it boot to either the original firmware or rockbox at all?
19:01:41IcedmanI can mount it on linux via the mount command (but that's ugly with root and everything)
19:01:50Icedmanit boots to both perfectly fine
19:01:56 Join n1s [0] (~n1s@nl118-168-30.student.uu.se)
19:01:56 Quit n1s (Changing host)
19:01:56 Join n1s [0] (~n1s@rockbox/developer/n1s)
19:02:32bertrikdid you try a release build or a development build?
19:02:43Icedmanthe 3.13 stable build
19:02:59Icedmanvia the rbutil installer
19:03:34bertrikA lot of fixes were done recently, which may have improved USB behaviour
19:04:09bertrikI recommend booting to the original firmware, then updating to a recent build. AFAIK RockboxUtility can install those too
19:05:27Icedmank
19:05:41Icedmanthe original firmware is not being automatically mounted either
19:05:47Icedmanso I will do that via root then
19:06:20copper<Icedman> I can mount it on linux via the mount command (but that's ugly with root and everything)
19:06:27copperthat doesn't sound like a problem
19:06:40copperif it mounts fine with "mount", then it's fine
19:06:48Icedman:/
19:06:55copperyou're just having some configuration problem on your PC
19:07:04Icedmanerr no?
19:07:16Icedmanmy mac and windows PC do not even recognize it
19:07:44Icedmanwhich they did before
19:07:50Icedmanthe update to rockbox
19:08:03coppercan you mount it on your Linux PC, and then pastebin the output of "/usr/bin/mount" on its own?
19:08:39dfktIcedman, i found that my clip+ mouints 'magically' every time on windows when i attach it to the usb cable while it's playing
19:08:59dfktit rarely ever mounts while the player is off, or playback in rockbox is stopped
19:09:23Icedmancopper, like mount/dev/blub > blat.txt?
19:09:28copperno
19:09:30copperjust "mount"
19:09:34copperwithout any arguments
19:09:37Icedmank
19:10:04coppercopy paste the output on pastebin.com and link to your pastebin here
19:12:06Icedmanhttps://paste.xinu.at/CtW/
19:12:39Icedmandfkt, interestin O_o
19:13:41copperIcedman: can you also pastebin the contents of your /etc/fstab?
19:13:59copperI have this in there: UUID=0123-4567 /media/clip vfat defaults,noatime,user,noauto,iocharset=utf8 0 0
19:14:11copperwhich allows normal users (non root) to mount my Clip+
19:14:25copperthe ",user," part
19:14:59copperUUID= identifies my clip+, as listed in /dev/disk/by-uuid/
19:15:31Icedmanhttps://paste.xinu.at/gIWDK/
19:15:34Icedmanmy fstab
19:16:30copperhmm
19:16:44copperwell, I never use automount, and I don't know what your distro is doing for automount
19:16:56copperbut I don't see anything wrong
19:17:07Icedman._. well I use archlinux with systemd
19:17:15copperuh, me too
19:17:21Icedmanuh
19:17:24Icedmanx'D
19:17:34KarrdeX"d
19:17:38copperI do use automount for one thing actually
19:17:53copperbut I have an fstab entry for it, with the "x-systemd.automount" parameter
19:18:28 Nick DormantBrain is now known as SuperBrainAK (~andy@2001:470:8:a61::5f92:59a1)
19:18:31Icedmanhm
19:18:44IcedmanI use udiskie
19:19:53copperwhat's the name of your Clip device?
19:19:58copperas listed in /dev/disk/by-label/
19:21:05IcedmanSANSA\\x20CLIPZ@
19:21:26copperdo you know if that's the same name as before you installed Rockbox?
19:21:58IcedmanYes, before it got mounted automatically to /media/Sansa\ CLIPZ/
19:22:06Icedmanerr
19:22:09IcedmanSANSA CLIPZ
19:23:14coppercan you post the relevant portion of "dmesg" that includes the part when you connect the Clip Zip via USB?
19:23:33Icedmanaye
19:25:17copperbetter yet, just pastebin the entire output
19:28:20Icedmanx'D
19:29:24Icedmanhttps://paste.xinu.at/Du4Yb/ <−− most recent reconnect. Full dmesg https://paste.xinu.at/4xP/
19:33:18copperclearly you've been having USB problems
19:33:27IcedmanO_O
19:33:32Icedman;_;
19:33:33coppermaybe udiskie crapped out because of it
19:33:56copper"usb 1-6: reset high-speed USB device number 32 using ehci-pci"
19:34:00copper"sd 26:0:0:1: Device offlined - not ready after error recovery"
19:34:04copper"end_request: I/O error, dev sdd, sector 4096"
19:34:06Icedman>_>
19:34:11copperetc etc
19:34:34Icedmanreminds me
19:35:48copperyou can try booting the original firmware by pressing and holding "Left" while turning it on
19:35:51copperhttp://download.rockbox.org/daily/manual/rockbox-sansaclipzip/rockbox-buildch3.html#x5-280003.1.3
19:36:05Icedmanafter installing and connecting it again I had this: https://paste.xinu.at/Ujj/
19:36:17copperthat's rockbox crashing
19:36:24copperUSB problems again
19:36:29Icedman._.
19:36:38copperthe Clip Zip port is known to have such problems
19:36:43Icedmanouch
19:36:53copperother targets have similar problems too
19:37:10Icedmanbertrik mentioned the dev build
19:37:21Icedmanguess it won't hurt to try
19:37:30copperhttp://build.rockbox.org/data/rockbox-sansaclipzip.zip
19:37:48bertrikyeah, I suspect the most recent development build is actually more stable than 3.13
19:38:08bertrik because of all the recent memory fixes, which may also have improved USB
19:38:30Icedmanhm I will try then
19:39:32coppereh
19:39:43copperI wonder how many USB crashes are actually USB bugs
19:43:53Icedmanx'D
19:43:57Icedmandidn't do shit
19:44:02Icedmanmeh x'D
19:54:02 Quit saratoga (Quit: Page closed)
19:57:08bertrikI plan to actually work on trying interrupts for fm tuners soonish
19:57:33bertrikI hope it will speedup tuning a bit, especially autotune
19:57:53bertrikactually looks easy to implement so far :)
20:00
20:00:27 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
20:08:36 Quit pretty_function (Remote host closed the connection)
20:12:56 Join cordyceps [0] (~cordyceps@207.216.194.194)
20:14:06 Quit Icedman (Remote host closed the connection)
20:16:46 Quit Rower (Ping timeout: 252 seconds)
20:17:32 Join Rower [0] (husvagn@90-230-142-55-no41.tbcn.telia.com)
20:19:46pamauryBobJonkman1: here ?
20:23:11 Join krabador [0] (~krabador@87.18.225.146)
20:23:24 Quit krabador (Changing host)
20:23:24 Join krabador [0] (~krabador@unaffiliated/krabador)
20:38:53 Quit Raptors (Quit: Leaving)
20:50:06 Join Raptors [0] (~whoneedsa@216-58-33-234.cpe.distributel.net)
20:53:08 Join faustlod [0] (~faustlod@193.105.154.74)
20:56:43 Join einhirn [0] (~Miranda@p4FC74FE0.dip0.t-ipconnect.de)
20:57:19***Saving seen data "./dancer.seen"
20:59:59 Quit GodEater (Quit: Coyote finally caught me)
21:00
21:01:21 Join GodEater [0] (~whoknows@cl-711.lon-02.gb.sixxs.net)
21:01:22 Quit GodEater (Changing host)
21:01:22 Join GodEater [0] (~whoknows@rockbox/staff/GodEater)
21:03:44 Join panach [0] (~panach@91.210.101.56)
21:07:32 Quit [Saint] (Remote host closed the connection)
21:09:38 Join [Saint] [0] (~saint@rockbox/staff/saint)
21:13:26 Join Icedman [0] (~Icey@unaffiliated/icedman)
21:13:31Icedmancopper, bertrik
21:13:34Icedmanyou know what
21:13:57IcedmanGolden Rule of IT Stuff No.1 Reinstalling fixes shit, yo!
21:13:58 Quit ikeboy (Ping timeout: 252 seconds)
21:15:02 Quit panach (Ping timeout: 252 seconds)
21:18:49 Quit einhirn (Quit: Miranda IM! Smaller, Faster, Easier. http://miranda-im.org)
21:19:45BobJonkman1pamaury: Hi: I'm back now
21:21:44 Quit y4n (Quit: HOLY SHIT! WE'RE ALL JUST LIVING ON A GINORMOUS FUCKING SPINNING ROCK FLOATING THROUGH SPACE CIRCLING A BIG FUCKING BALL OF FIRE!!!)
21:45:13n1shmm, trying to profile the opus codec just gives me lovely undefined instr panics...
21:45:31pamauryBobJonkman1: we can try something else to recover your Fuze+. Download https://www.dropbox.com/s/x65ogs4jgta211p/hwstub.sb
21:45:49BobJonkman1pamaury: OK, downloading now
21:46:40pamauryBobJonkman1: hum, wait I will update another file to make it easier for you
21:46:52BobJonkman1pamaury: :) OK
21:56:26 Join aevin [0] (eivindsy@unaffiliated/aevin)
21:58:48 Quit Icedman (Quit: Everyone quits one day. You just wasted 3 seconds reading this.)
22:00
22:02:22pamauryBobJonkman1: I have a prototype, let me try it on mine first
22:02:36BobJonkman1pamaury: OK, thanx!
22:11:20 Quit Mir (Remote host closed the connection)
22:15:28BobJonkman1pamaury: I need to go AFK for about a while. Probably be back in an hour or two
22:15:48 Quit Strife89 (Ping timeout: 272 seconds)
22:20:52 Quit Rower (Quit: Hmmm...)
22:22:41[Saint][05:57:15] <copper> those fixes mean that some themes are no longer compatible with HEAD <−−- no recent changes have changed the syntax at all
22:22:59[Saint]It was only in your case where you had specifically worked around an issue.
22:25:19 Quit cordyceps (Quit: leaving)
22:30:10 Join Mir [0] (~Mir@pool-71-109-219-225.lsanca.dsl-w.verizon.net)
22:36:27copperissues* (plural)
22:37:00copperthere were so many random bugs (or effects of one or two bugs), there's no telling what will still work and what won't
22:37:43copperand maybe I'm more present and more vocal here than other theme authors
22:37:43 Quit bluebrother (Read error: Operation timed out)
22:38:26copperI have reported several bugs here but also on flyspray, that no-one else reported before
22:38:37[Saint]The recent fixes would only break a theme that was broken to begin with.
22:38:39copperbut that just means I took the time that others may not have
22:38:57[Saint]ie. yours.
22:39:05copper...
22:39:18copperand why do you think they were "broken"?
22:39:27[Saint]They should never have parsed.
22:39:53copperI did it right while I was making my themes
22:40:03copperbut that caused Rockbox bugs to occur
22:40:18 Quit fs-bluebot (Ping timeout: 260 seconds)
22:40:20coppermy themes weren't initially "broken": Rockbox was
22:40:38copperwho's to say I was the first and only theme maker to resort to such hacks?
22:41:27copperI ran into so many problems, I would be very surprised if I was the only one
22:41:29[Saint]Who is saying that? All I'm saying is that the recent changes would only cause an issue if the theme has issues to begin with.
22:41:45copperthat's a backwards way of looking at it
22:41:46 Join fs-bluebot [0] (~fs-bluebo@g231123242.adsl.alicedsl.de)
22:41:57 Join bluebrother [0] (~dom@rockbox/developer/bluebrother)
22:42:02[Saint]Working around theme bugs with illegal syntax is also backwards.
22:42:17copperwhat, should I have given up?
22:42:24coppershould I not have published my themes?
22:42:59copperI took the more productive approach: I reported the bugs, properly, officially, and I worked around it while there was no fix.
22:43:15[Saint]I'm not saying that at all. But you should be avoiding blanket statements like the one you made, stating that HEAD will break theme compatibility.
22:43:27[Saint]Because it won't.
22:43:29copperI said it was a possibility
22:43:46copper17:16:27 UTC <copper> but there's a chance that other themes need to be fixed too
22:44:12copperthat's quite a caucious statement, I think
22:44:19coppercautious
22:45:00coppercertainly not a blanket statement
22:45:00[Saint]Its a misleading statement, I think. But, whatever.
22:45:26[Saint]Its not making it clear that the issues would be caused by the authors workarounds.
22:45:45[Saint](which shouldn;t have worked in the first place)
22:45:55copper17:15:46 UTC <copper> the previous version of my themes had fragile workarounds for the theme bugs that kugel fixed
22:45:58copper17:15:56 UTC <copper> now both Rockbox and my themes are fixed
22:46:01copper17:16:27 UTC <copper> but there's a chance that other themes need to be fixed too
22:46:05copperI think I stated that fact very clearly
22:46:23copperit's written… right. There.
22:49:29[Saint]I apologize, that section isn't present in my logs at all.
22:49:30[Saint]Odd.
22:49:52copperOk.
22:49:54gevaerts[Saint]: fix your logs!
22:50:01gevaertsThat seems to happen at least twice a year!
22:50:47copperthat's happened to me once before I think
22:50:54[Saint]It does indeed. And I have seen it reported a few times by other quassel users, but it is hard to pinpoint. And very hard to notice.
22:50:56coppersomeone missing what I said from the logs, too
22:51:10copperin #rockbox
22:51:16coppercan't remember who it was
22:51:19[Saint]Well...it just made me look like a giant dick.
22:51:30[Saint]Gah.
22:51:30copperjust a misunderstanding, that's ok
22:52:18copperhttp://www.rockbox.org/irc/log-20140120
22:52:52copperin case you're missing other parts of the initial conversation
22:53:11 Quit faustlod (K-Lined)
22:53:25[Saint]Man. That's really weird.
22:53:36[Saint]All I have from you around that time is the one line I quoted.
22:55:55copperstill
22:56:15coppernm
22:56:32 Join lebellium [0] (~chatzilla@lns-c10k-ld-02-m-212-194-176-149.dsl.sta.abo.bbox.fr)
22:57:20***Saving seen data "./dancer.seen"
23:00
23:01:31 Quit n1s (Quit: Ex-Chat)
23:01:54 Join stripwax [0] (~Miranda@rockbox/developer/stripwax)
23:24:04 Quit stripwax (Read error: Connection reset by peer)
23:25:09 Quit amayer (Quit: Leaving)
23:36:44BobJonkman1pamaury: Back again. For at least an hour...
23:54:02 Quit pamaury (Quit: this->disconnect())
23:54:21 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
23:54:57pamauryBobJonkman1: download this: https://www.dropbox.com/s/bwq3w8tmehu5vss/charge-fuzeplus.sb
23:58:12 Quit dunpeal (Quit: leaving)

Previous day | Next day