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 2011-01-17

00:00:00 Join Judas_PhD [0] (
00:04:30 Quit Judas_PhD (Ping timeout: 240 seconds)
00:06:24 Quit Saij (Ping timeout: 272 seconds)
00:17:19 Quit stripwax (Read error: Connection reset by peer)
00:18:04 Join Judas_PhD [0] (
00:27:50 Quit robin0800 (Remote host closed the connection)
00:31:01***Saving seen data "./dancer.seen"
00:31:50 Join robin0800 [0] (
00:33:56 Quit ender` (Quit: Only thing made by Apple that wouldn't suck would be iVacuum.)
00:35:39 Join GeekShadow [0] (~Antoine@reactos/tester/GeekShadow)
00:41:40 Quit pamaury (Remote host closed the connection)
00:45:06 Quit sasquatch (Quit: WeeChat 0.3.2)
00:45:31 Join sasquatch [0] (
00:48:24sideralgevaerts: I now blame my transient filesystem corruptions on the experimental AMSv2 USB driver. See latest update to FS #11877
00:49:25gevaertssideral: yes, I just saw that. Allow me to have some doubts :)
00:49:53gevaertsThat's low-level enough that it's *never* exposed to software at all
00:50:46sideralwell, the pattern 0xfd,0xfe that seems to trigger it is common in FATs, but perhaps no where else
00:51:07sideral(and 0x7d,0x7e)
00:51:38sideralthe device itself would never notice, only an attached USB host
00:51:52 Quit robin0800 (Remote host closed the connection)
00:52:47sideralwhich would often detect a cycle in the FAT because of the additional 0x7d in a pointer appearing closely after a previous 0x7d
00:53:48gevaertsCan you post your exact test procedure somewhere? It would be useful to see if this can be reproduced on other targets
00:53:55sideralHere's an example FAT dump diff:
00:54:29sideralThe 0x7d in the "+" line leads to a FAT-chain cycle
00:55:32sideralSo far I don't know what triggers this. The corruption is only visible sometimes :(
00:56:33sideralwe've had reports of other users having FAT-corruption issues lately
00:56:57sideralyesterday "newClipUser" said he had a corrupt FAT on his Clip+
00:57:53gevaertsUsing an official build?
00:59:37sideralThere's no official build with USB enabled; it's experimental (FS #11664)
01:00:52sideralnot sure if newClipUser uses a custom build, though
01:01:06JdGordon1he would have to be
01:01:11gevaertsyes, that's why I ask
01:01:40sideralso it could be just the typical OF fiddling with ##MUSIC and ##PORT, which sometimes doesn't update the 2nd FAT
01:01:44sideralin his case
01:06:06sideralwhat would you make of the pattern of corruption I've described, if it wasn't USB?
01:06:19 Join saratoga [0] (9803c22e@gateway/web/freenode/ip.
01:07:54gevaertsIt could be the USB driver, I just don't see a way for escaped on the wire data to ever be visible to software
01:08:46gevaertsAlso, at the level that escapin happens, I'd be *very* surprised if there's anything left of the original alignment of the data
01:11:07sideralthe corrupt 0x7d leaking into the next 16 bytes could be a manifestation of said alignment issue
01:11:40 Quit KiwiCam (Quit: Leaving)
01:12:02sideralit could be some low-level USB controller / PHY going astray
01:13:02gevaertsHave you tried reading that data again during the same connection?
01:14:01sideralunfortunately I've not dd'd it twice, but I remember having run fsck (read only) multiple times, and it reported the same problems again
01:14:33sideralas this seems to be triggered by the pattern in the previous 20 bytes, I'd expect it to trigger at the same spots
01:15:11gevaertsHave you ever powered off the sansa the hard way after seeing corruption while it's still connected?
01:15:45sideralIt's impossible to power off the Clip if it's still connected
01:15:48gevaertsi.e. without the rockbox USB disconnect code or the shutdown code getting a chance to run
01:16:32sideralwhat would we learn from this experiment you propose?
01:17:53gevaertsWell, I'm not sure if I've seen evidence that this *must* be sd or usb related at all. It may be unlikely, but maybe the wrong bytes are really there until "something" happens later on
01:18:32 Quit Judas_PhD (Ping timeout: 255 seconds)
01:18:39sideraldo we have a diskdump plugin?
01:18:52gevaertsMaybe you could add some code to the usb screen that panics when you press the right button? That would allow you to disconnect and reboot at ease
01:19:15sideralI could try that
01:19:34gevaertsSo connect to USB, if the corruption is there, panic, and reboot to the OF to see what's actually there
01:20:35 Join timc1 [0] (~lisa@
01:20:37gevaertsIf that's different (except possibly for files the OF writes of course), the bug is in the rockbox SD or USB driver. If you still see corruption, it's in the fat/unmount code
01:20:39sideralhmm... but I think I've established that the corrupt bytes are never on disk unless the host writes them out as part of a FAT update
01:20:47 Part timc1
01:21:23sideralthere are 3000 corrupt bytes in the FAT that are back to normal on next boot −− it's highly unlikely they ever were on disk
01:22:44sideralthe question for me is: are they visible in RB's memory as well? -> SD issue −− or are they visible only to the host? -> USB issue
01:23:08gevaertsWell, not if my (admittedly unlikely) hypothesis about this just being a too late flush is correct
01:24:04sideralI should port fsck as a plugin
01:24:20gevaertsWell, again, if the bytes are *not* on disk, and I really think 99.5% sure isn't good enough here
01:25:20gevaertsDo you have a clip or a clip+?
01:25:26sideralRight. But if I can establish that they're not in RB's memory, then by induction they're not on the disk. I'll shoot for the 1st proof first I guess
01:25:32sideralA ClipV2
01:25:48gevaertsok, so no copying data to the sd card...
01:25:57 Quit liar (Read error: No route to host)
01:26:34gevaertsHow are you going to check if they're not in memory? Intercepting data between sd and usb doesn't sound trivial
01:26:52 Join liar [0] (
01:27:20sideralby running a memory dump or an fsck as a rockbox plugin?
01:27:45sideralor I could instrument the SD driver to look for the pattern
01:28:00gevaertsHow big is the FAT? Does it fit in RAM?
01:28:22sideralthe latter would allow me to intercept on the USB slave mode data path
01:28:40sideral7716352 bytes per FAT
01:28:42gevaertsThat's assuming the pattern is always exactly the same. Is it?
01:28:53gevaertsclipv2 has 8MB?
01:29:08sideralI think so −− hmm :)
01:29:38sideralin the two dumps I have, the pattern seems to be always there
01:29:58*gevaerts proposes to do the easy tests first :)
01:30:12saratogayou have 8MB + 1MB IRAM
01:30:46sideralthanks saratoga
01:31:08sideralI think looking for the pattern in the SD driver is the simplest thing
01:31:50 Join Judas_PhD [0] (
01:32:29sideraland most conclusive
01:32:48sideralbut writing a disk dumper would be simple as well
01:33:07saratogaby the way if you're interested in clipv2 bugs, I bet that weird background noise isn't too hard to fix with some toying around
01:33:14saratogasince the clip+ doesn't have it
01:35:01sideralsaratoga: I'm indeed interested in that as well, but code that writes constant values to I/O registers (which is what most of the platform code does) is not particularly reassuring :)
01:37:47sideralsaratoga: Given the heat you had to take over the autoresume commit, how comfortable do you feel committing the second part? ;-) (delay until resume-pos update, allows skipping to prev track w/o losing resume position)
01:39:39saratogasideral: pretty comfortable, but theres no way i'm doing this in the next 2 weeks
01:39:55saratogai'm staying away from rockbox until i get this stupid project at work over with, then i'll be traveling to present my results
01:40:17saratogai'm too sleep deprived right now to look at c code coherently
01:40:49sideralok, that's fine
01:48:17 Quit Judas_PhD (Ping timeout: 255 seconds)
02:01:39 Join Judas_PhD [0] (
02:03:36 Join Keripo [0] (
02:04:22 Quit sideral (Ping timeout: 265 seconds)
02:08:10 Join krabador [0] (
02:13:11 Quit Loto (Ping timeout: 240 seconds)
02:18:02 Join Loto [0] (
02:18:02 Quit Loto (Changing host)
02:18:02 Join Loto [0] (~nfs@xbmc/user/Loto)
02:18:31 Quit Judas_PhD (Ping timeout: 246 seconds)
02:28:27 Join BHSPitMonkey [0] (~stephen@unaffiliated/bhspitmonkey)
02:31:04***Saving seen data "./dancer.seen"
02:31:45 Join Judas_PhD [0] (
02:32:14 Join milk [0] (
02:34:53 Join JdGordon2 [0] (~jonno@
02:36:42 Quit JdGordon1 (Read error: Connection reset by peer)
02:36:43 Quit JdGordon2 (Read error: Connection reset by peer)
02:36:53 Quit Judas_PhD (Ping timeout: 276 seconds)
02:39:10[Saint]JdGordon|: Why does %x need an identifier?
02:39:15[Saint]It just seems...weird.
02:39:45[Saint](weird == setting an identifier that never gets called conditionaly)
02:41:28 Join JdGordon1 [0] (
02:42:11 Join madalu [0] (~user@unaffiliated/madalu)
02:42:52 Quit mudd1 (Ping timeout: 260 seconds)
02:49:25 Quit Keripo (Quit: Leaving.)
02:52:54 Join Keripo [0] (
03:00:38 Quit BlakeJohnson86 (Quit: Leaving.)
03:00:55 Quit GeekShadow (Read error: Connection reset by peer)
03:00:57 Join GeekShad0w [0] (
03:02:17 Quit panni_ (Quit: ( :: NoNameScript 3.81 :: ))
03:02:36 Join panni_ [0] (
03:04:44 Join BlakeJohnson86 [0] (
03:07:03 Quit liar (Read error: No route to host)
03:07:37 Part BlakeJohnson86
03:11:57 Join liar [0] (
03:17:24 Join evilnick [0] (c37028fa@rockbox/staff/evilnick)
03:23:14 Join Zarggg [0] (
03:26:29[Saint]Anybody with a 240x320 (preferably touchscreen) target want to test a cabbie theme port?
03:26:34 Join vnl [0] (
03:27:31[Saint]gigabeas/ft, mini2440, Galaxy 5, <other> ...?
03:28:44*JdGordon1 doesnt like the beaft name
03:28:48JdGordon1beast makes sense
03:28:52JdGordon1beaft looks stupid :p
03:29:15[Saint]I know...but it made it easier to type ;)
03:29:59[Saint]I made a version of this theme with "LED" peak *THAT* looked stupid...
03:30:36JdGordon1LED masks on the bar type? or done with images?
03:30:51 Quit liar (Ping timeout: 255 seconds)
03:31:27[Saint]just a bitmap bar.
03:31:39JdGordon1so you get half lit LED's?
03:32:12[Saint]yeah..I guess so, but it's pretty hard to notice.
03:32:39[Saint]what do you mean by "LED masks on the bar type"?
03:32:53 Quit krabador (Ping timeout: 272 seconds)
03:33:34 Quit GeekShad0w (Quit: The cake is a lie !)
03:36:18 Join liar [0] (
03:36:31[Saint]last time I remember checking "%bar_tag<image|image|image|image>" didn't work anywhere near the way I expected it to.
03:41:56JdGordon1depends on the tag
03:42:18JdGordon1volume scales, battery has some special handling
03:42:51[Saint] was volume that went horribly wrong.
03:43:17JdGordon1it is the normal %?pv handling, nothing to do with it being a bar tag
03:44:13[Saint]I still don't understand what you mean by "LED masks on the bar type"
03:44:31 Join krabador [0] (
03:44:47JdGordon1masks being transparent areas which the bar fills
03:45:06JdGordon1or does that not work?
03:45:22[Saint]I don't think it's possible to do this and *not* display partial "LED"s
03:45:45JdGordon1sure it is, with a very big conditional block
03:46:10[Saint]last time I tried it it reall, really, didn't work.
03:47:29[Saint]"%bar_tag<image|image|image|image>" for example only displays the valid segment, not the valid segment and those prior to it like you'd want it to.
03:47:45[Saint]and, the alignment gets messed up pretty bad if you try that as well.
03:48:49 Quit timccc (Ping timeout: 246 seconds)
03:49:00JdGordon1i.e fucking horrible
03:49:22JdGordon1seperate images, not subimages
03:49:43 Quit liar (Read error: No route to host)
03:49:51JdGordon1support for that could be added to the bars i tihnk
03:50:28[Saint]I'm not sure the bar tags need any more complicating.
03:50:38JdGordon1sure it does!
03:50:45 Join liar [0] (
03:50:50JdGordon1while *anyone* understands most of it, it isnt complicated enough
03:51:11 Quit designate72 (Ping timeout: 240 seconds)
03:51:15*JdGordon1 still wants to make the skin system turing complete :p
03:51:29*[Saint] doesn't think it's much of a good sign that most of the Rockbox devs won't touch skin "code" with a barge pole ;)
03:51:49[Saint]or, perhaps it is ;) hideous themes, or, less ;)
03:51:52JdGordon1most rockbox devs wont touch most of rockbox's code
03:52:30JdGordon1serisouly.... everyone has one or two areas they like and try to avoid other areas as much as possible
03:52:40JdGordon1and noone likes DB, dircache, playback, buffering
03:53:05[Saint]DB and playback are evel...I can't speak for the others.
03:57:35 Join Zarggg_ [0] (
04:01:48 Quit Zarggg (Ping timeout: 276 seconds)
04:03:12 Quit krabador (Ping timeout: 272 seconds)
04:10:22 Quit liar (Ping timeout: 255 seconds)
04:11:53 Join liar [0] (
04:11:56 Quit milk (Quit: baaaiiii)
04:12:36 Join Topy [0] (
04:13:43 Join timccc [0] (~timccc@
04:16:16 Quit T44 (Ping timeout: 264 seconds)
04:19:39 Join timccc1 [0] (~timccc@
04:20:00 Quit timccc (Read error: No route to host)
04:21:59 Quit saratoga (Ping timeout: 265 seconds)
04:24:52*[Saint] wonders if RB could use haptic feedback on Android
04:25:11[Saint]or, if it's supposed to...and it's for some reason broken here.
04:25:35JdGordon1it could, and doesnt
04:25:50 Join Barahir [0] (
04:26:25[Saint]Oh,, and bummer, and cool.
04:29:14 Quit evilnick (Ping timeout: 265 seconds)
04:29:45 Quit Barahir_ (Ping timeout: 276 seconds)
04:31:08***Saving seen data "./dancer.seen"
04:31:45 Quit amiconn (Disconnected by services)
04:31:45 Quit pixelma (Disconnected by services)
04:31:46 Join amiconn_ [0] (quassel@rockbox/developer/amiconn)
04:31:48 Join pixelma_ [0] (quassel@rockbox/staff/pixelma)
04:31:50 Nick pixelma_ is now known as pixelma (quassel@rockbox/staff/pixelma)
04:32:03 Nick amiconn_ is now known as amiconn (quassel@rockbox/developer/amiconn)
04:32:40 Quit MethoS- (Remote host closed the connection)
04:44:44 Quit TheSeven (Ping timeout: 246 seconds)
04:50:06 Join TheSeven [0] (~TheSeven@rockbox/developer/TheSeven)
04:51:19 Quit tchan (Read error: Connection reset by peer)
04:52:09 Join tchan [0] (~tchan@lunar-linux/developer/tchan)
04:57:43[Saint]'s so annoying that the player doesn't display metadata/file info for a track when the player has just been turned on, it's actually quite a cool "feature" almost that the info will stay on screen in the .sbs if you stop a track and return to the main menu
04:58:13[Saint]then you know exactly what will be played when you hit resume.
04:58:26 Quit madalu (Ping timeout: 240 seconds)
04:58:41[Saint]but seeing a blank playbar and playlist/track info when the player first starts looks a bit crap.
04:59:37[Saint]JdGordon1: Is it just a fluke that track data stays on screen in the .sbs when a track is stopped?
04:59:52[Saint]it makes for a nice "feature"
05:00:43JdGordon1yes, that isnt by design
05:00:55JdGordon1although, doing the feature for both is possible
05:01:25[Saint]it's cool that it does it...and I'd abuse it in the theme if it didn't look so shit seeing blank info when the player first stars up.
05:01:55[Saint]Though, since it can be left running...this doesn't really matter for Android.
05:02:15[Saint]and, I could leave the track info enabled on stop also....hmmm.
05:02:22JdGordon1depends on the task killer rules
05:02:37JdGordon1my G2 seems to be killing the service far more than my N1 did
05:03:38[Saint]yeah, you'd only see blank/garbage data on the first run after the service is it worth it to abuse this "feature"?
05:04:11[Saint]Or, is there some other way that it could be made to *actually* display the data for a stopped track, even froma "cold" boot?
05:04:53JdGordon1call playlist_resume() or whatevr it is on boot
05:05:07[Saint]I assume it would be possible, or we couldn't have a resume feature.
05:05:33 Quit Keripo (Quit: Leaving.)
05:06:01[Saint]knowing the track info for the track you're going to be resuming is/would be pretty cool...but it would need to work all the time for me to enable it in the theme.
05:06:07[Saint]Not just abusing a feature.
05:07:04 Quit Rob2222 (Ping timeout: 255 seconds)
05:17:03JdGordon1the only difference between a bug and a feature is when it was found
05:17:19*JdGordon1 was tought that at MS by his group manager
05:26:48 Join froggyman_ [0] (~seth@
05:37:22[Saint]JdGordon1: Do you think it would be possible to adapt current behaviour to report the track data when the track is stopped, without it being a cool coincidence?
05:38:29 Quit froggyman_ (Quit: Ex-Chat)
05:38:36[Saint]I can't help but wonder why it's not this way already.
05:38:58JdGordon1it would involve a probably pointless disk read on boot
05:40:19JdGordon1although on boot the disk is spining anyway so it isnt a big deal
05:41:22 Quit Horscht (Quit: Verlassend)
05:41:25[Saint]I sincerely doubt it would measureably slow the boot process.
05:42:31[Saint]Unless someone sees this conversation and feels inspired, it'll stay one of those "it would be cool if..." things.
05:42:48[Saint]I'm the first to admit that it's well beyond my ability.
05:42:49JdGordon1its very simple to make work
05:43:28[Saint]It may be...perhaps, I've never looked at any of the playback/metadata magic though.
05:43:31 Join rad_sci_guy [0] (
05:44:31 Part rad_sci_guy
05:44:38JdGordon1then you cant be sure that it is beyong you
05:44:52JdGordon1especially when i already said what needs to be done
05:46:56[Saint]I'll trade you a theme that makes use of it for it ;)
05:48:58 Quit jhMikeS ()
05:56:31 Join jhMikeS [0] (~jethead71@rockbox/developer/jhMikeS)
06:04:01 Quit JdGordon1 (Quit: leaving)
06:31:12***Saving seen data "./dancer.seen"
06:42:11 Join thegeek_ [0] (
06:43:04 Quit thegeek (Read error: Connection reset by peer)
06:48:19 Quit panni_ (Quit: ( :: NoNameScript 3.81 :: ))
06:57:14JdGordon|[Saint]: im home, send me the cabbie :)
06:57:33JdGordon|also im 90% sure that if you just cal playlist_resume() at boot somewhere it will do what you want
06:57:47JdGordon|root_menu() or end of init()
07:00:41 Join thegeek [0] (
07:02:21 Join timccc [0] (~timccc@
07:02:42 Quit timccc1 (Read error: No route to host)
07:03:18 Quit thegeek_ (Read error: Operation timed out)
07:04:07 Quit balintx_ (Ping timeout: 255 seconds)
07:04:26 Join balintx [0] (
07:06:18[Saint]JdGordon|" <−−- 240x320
07:07:35[Saint]keep in mind, I'm still tinkering with example of this is that I will probably sacrifice translation and go back to a nice aliased image for the list titles instead.
07:07:46[Saint]Like the "normal" cabbie .wps
07:16:03 Quit [Saint] (Disconnected by services)
07:16:05 Join S_a_i_n_t [0] (S_a_i_n_t@
07:25:23 Quit t0rc (Quit: Give someone code, help them with one project. Teach someone to code, help them rule the world.)
07:31:25 Join sideral [0] (~sideral@unaffiliated/sideral)
07:50:33CIA-7New commit by jethead71 (r29067): Convert PortalPlayer iRivers to use USB-mode enabled bootloader. The H10 5/6GB already has a pre-boot USB mode (so it's a little redundant there) but ...
07:58:39CIA-7r29067 build result: All green
08:00:17S_a_i_n_tJdGordon|: ?
08:15:47 Join efyx [0] (
08:25:09 Join esperegu [0] (~quassel@
08:26:25 Join stoffel [0] (
08:28:43JdGordon|S_a_i_n_t: busy struggling to keep my eyes open... I'll look in a ahilw
08:29:31 Join B3gder [0] (
08:31:16***Saving seen data "./dancer.seen"
08:42:15 Join lazka [0] (
08:44:43lazkahey, does switching channel config to mono also disable crossfeed?
08:48:03 Join ender` [0] (
08:49:08 Nick B3gder is now known as B4gder (
08:49:17 Quit B4gder (Changing host)
08:49:17 Join B4gder [0] (~danielx@rockbox/developer/bagder)
08:50:06B4gderhow can it crossfeed mono?
08:52:59 Quit kkit|sh (Ping timeout: 255 seconds)
08:55:20jhMikeSit's not removed from the processing when mono is selected
08:55:54jhMikeSif the file itself is mono, then crossfeed is skipped
09:02:22 Join bluebrother [0] (~dom@rockbox/developer/bluebrother)
09:02:22 Quit bluebroth3r (Read error: Connection reset by peer)
09:11:13TheSevenipod classic battery bench: ~8:20 sitting in the main menu, doing nothing, backlight off
09:11:46 Join wodz [0] (
09:12:08wodzTheSeven: whats the runtime in OF?
09:12:20TheSevencan't test on that device
09:12:31TheSeven(i did it on the old one with the broken HDD)
09:12:51TheSevenbut i'd think way more, even if it's playing music
09:13:15TheSeventhey say it reaches 36h, and i think this battery may not be the best, but it should get at least half of that
09:13:34TheSevenmy main aim with that bench was to calibrate the fuel gauge
09:13:37wodzfreq scaling + core undervolt makes big difference probably
09:15:32wodzdo you know how to turn off power to hdd?
09:18:31TheSevenbut the hdd has probably woken a few times due to battery_bench saving logs
09:19:02TheSevenat least i have no other explanation for the dips in that curve
09:20:44TheSeveni have a rough idea how to do freq scaling by changing some clock dividers, but i haven't figured out how to reconfigure a PLL yet
09:22:52wodzThe discharge curve looks quite normal. The humps comes form battery chemistry rather than from disk access
09:23:24*TheSeven hasn't seen that kind of humps on any other battery bench (from e.g. the nano2g)
09:24:58 Join LinusN [0] (~linus@rockbox/developer/LinusN)
09:25:19wodzI benched various lithium batteries on university (more analytic tests than battery_bench using specialized equipment) and I saw similar discharge curves a few times.
09:25:59wodzLinusN: how was your weekend BDM session?
09:26:56LinusNit turned out that i couldn't find the ribbon cable to connect the bdm to the h100, so i have ordered new ones... :-(
09:27:15 Quit BHSPitMonkey (Read error: Connection reset by peer)
09:27:20LinusNbut i have compiled the bdm software using your patch
09:28:52 Join petur [0] (d408b802@rockbox/developer/petur)
09:29:08 Join einhirn [0] (
09:29:51lazkajhMikeS, thanks, that's what I meant
09:33:15B4gderperhaps the Rockbox-on-android people are intereseted in "the andraudio mailing list, dedicated to audio development on Android"
09:34:32 Join mudd1 [0] (
09:45:13 Quit CaptainKewl (Ping timeout: 272 seconds)
09:57:01 Join swilde [0] (
09:57:02JdGordon|S_a_i_n_t: your theme still isnt loading....
09:59:19 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
09:59:39wodzJdGordon|: are you going to move on with rec tags? :-)
10:00:11wodzpeakmeater is a good start isn't it?
10:00:52wodzdefault recscreen doesn't have much more than this
10:01:09JdGordon|sure it does
10:01:29JdGordon|everything on the default rec screen can be put in a skin
10:02:12wodzbut can recscreen been themed?
10:02:30wodzthat's the point
10:03:52JdGordon|I'm still recovering from converting the fm screen :p
10:03:59JdGordon|really dont want to touch the rec screen code
10:04:29wodzdo not touch - write from scratch :-)
10:04:56wodzJdGordon|: where is theme .cfg parser located?
10:04:57 Quit S_a_i_n_t (Disconnected by services)
10:04:59 Join [Saint] [0] (S_a_i_n_t@
10:05:33JdGordon|the accepted values are settings_list.c
10:06:37wodzI am (slowly) rewriting to properly check for needed graphics from wps/fms/sbs and it's remote counterparts
10:16:38 Quit [Saint] (Ping timeout: 260 seconds)
10:17:20 Join JdGord [0] (~jonno@
10:18:31 Join [Saint] [0] (S_a_i_n_t@
10:31:18***Saving seen data "./dancer.seen"
10:35:26 Quit stoffel (Read error: Operation timed out)
10:37:11 Quit lazka (Remote host closed the connection)
10:41:43 Quit pjm0616 (Ping timeout: 276 seconds)
10:45:12 Join pjm0616 [0] (~user@
10:46:56 Join n1s [0] (~n1s@rockbox/developer/n1s)
10:58:45 Quit TheSeven (Ping timeout: 240 seconds)
11:07:43 Quit mudd1 (Ping timeout: 276 seconds)
11:17:01 Join DerPapst [0] (
11:17:33 Quit sideral (Quit: Leaving.)
11:34:04 Part vnl
11:37:02 Join sideral [0] (~sideral@unaffiliated/sideral)
11:37:21jhMikeSgevaerts: what can we do to detect a genuine usb connection? see I have this philips thing with a connector that fakes a bus reset condition, even when only connected to a charger.
11:41:48gevaertsjhMikeS: throw our hands up in the air and exclaim "What will they think of next?"?
11:42:23jhMikeSthe retailos isn't fooled, so clearly it's workable
11:43:22jhMikeSwhat's the first thing that would be sent to the device? a request for the descriptor?
11:44:17gevaertsyes, the device descriptor
11:46:09jhMikeSobviously the next thing they'll think of is chargers that request device descriptors :)
11:46:45gevaertsmost probably!
11:46:52*jhMikeS might have several already
11:47:49wodzI have a charger which makes ipod in rb think it is connected to PC
11:52:29jhMikeSI use mine to say dumb things in IRC, but I'll have a crack at detecting connection in the core
12:02:56 Join T44 [0] (
12:05:50 Quit Topy (Ping timeout: 240 seconds)
12:06:56 Part LinusN
12:07:26 Quit wodz (Quit: Leaving)
12:12:09jhMikeShmmm...already compiling to test. hopefully, it's really so easy
12:12:32 Join teru [0] (
12:12:59 Join stoffel [0] (
12:18:17 Quit stoffel (Remote host closed the connection)
12:18:57jhMikeSworks quite well actually
12:22:11 Join biki [0] (
12:23:25 Quit biki (Client Quit)
12:26:45 Join robin0800 [0] (
12:31:21***Saving seen data "./dancer.seen"
12:34:16 Quit Stummi (Quit: ZNC -
12:42:58pixelmabluebrother: do you have an idea where in the build environment could be a difference that I get differences in the manual's chapter number style? The ones from the site (and ones I compiled under cygwin on my laptop) have a trailing period, while the ones compiled in a linux VM on my desktop don't.
12:43:09 Join Guest2370 [0] (~bjst@rockbox/developer/Zagor)
12:44:36 Join Stummi [0] (~Stummi@rockbox/developer/Stummi)
12:44:42 Nick Guest2370 is now known as Zagor (~bjst@rockbox/developer/Zagor)
12:45:41pixelmamost fun thing is, that there are still 2-3 places that still get the trailing period
12:47:12 Join kugel [0] (~kugel@
12:47:14 Quit kugel (Changing host)
12:47:14 Join kugel [0] (~kugel@rockbox/developer/kugel)
12:52:03 Quit robin0800 (Remote host closed the connection)
12:52:17 Join robin0800 [0] (
13:13:15 Join LinusN [0] (~linus@rockbox/developer/LinusN)
13:24:44CIA-7New commit by jethead71 (r29068): Add a higher level USB detection that prevents fraudulent bus resets from causing USB mode to be entered. Enable for SA9200 only at this time. Also, ...
13:27:51 Join dfkt [0] (dfkt@unaffiliated/dfkt)
13:28:57n1si wonder if CIA could be configured to not cut the commit messages so short
13:29:46CIA-7r29068 build result: 16 errors, 0 warnings (jethead71 committed)
13:30:55n1sjhMikeS: any idea about a problem i've seen on my beast a couple of times (haven't confirmed with a very recent build tho) that if the backlight goes off in usb mode it doesn't come on again after exiting usb mode?
13:31:04n1sFS #11699 btw
13:31:21n1si'll retest nw
13:35:18*amiconn still didn't manage to build m68k-elf-gcc 4.5.2 under cygwin :(
13:35:58amiconnAfter getting the build error with gcc 3.4.4, I checked the config log and found that it complained about a missing ppl_c.h
13:36:35amiconnAfter installing ppl-devel and retrying, it now gets further: Now I get the same 'fork' error as when trying with gcc 4.x :(
13:36:52n1sjhMikeS: cant reproduce it now, sorry for the noise
13:37:13n1samiconn: weird, it shouldn't need ppl
13:37:33jhMikeSn1s: that is weird though and I haven't encountered it
13:37:46 Quit JdGord (Quit: Bye)
13:37:54 Join kevku [0] (~kevku@2001:7d0:0:f9af:215:c5ff:fe7e:794d)
13:38:05n1samiconn: only thing i can think of is av scanner not being nice even if disabled
13:38:26n1samiconn: which av do you use?
13:39:22amiconnKaspersky. And I did not just set it to disabled, but actually terminated it during the tests
13:40:06amiconnIt works for all other toolchains: arm-eabi, sh, mips, and old m68k
13:40:23CIA-7New commit by teru (r29069): FS #11822: use rockbox_browse() in plugins to select file.
13:41:24amiconnThere seems to be a bug in m68k-elf-as 2.20.1 - that's where the fork error always triggers (but sometimes for different target processors)
13:43:43 Join biki [0] (
13:45:13CIA-7r29069 build result: 16 errors, 0 warnings (teru committed)
13:46:16*jhMikeS has to go fix that grr
13:46:43n1samiconn: AFAIU, if it isn't something in wondows interferring with cygwin's fork calls then cygwin has run out of memory due to too many (defunct) processes
13:48:47 Quit kugel (Ping timeout: 240 seconds)
13:49:03n1smaybe the new 4.5 gcc takes a lot of memory when compiling the multilibs? although that would be weird since it worked on my netbook which has only 1GB
13:50:35 Join Topy [0] (
13:53:09 Quit biki (Quit: CGI:IRC (EOF))
13:53:11 Quit robin0800 (Remote host closed the connection)
13:53:51 Quit T44 (Ping timeout: 240 seconds)
13:59:00n1samiconn: are you using an up to date cygwin?
13:59:45 Join robin0800 [0] (
14:00:20 Join kugel [0] (~kugel@rockbox/developer/kugel)
14:04:03 Part robin0800
14:05:49 Join robin0800 [0] (
14:06:00 Join benedikt93 [0] (~benedikt9@unaffiliated/benedikt93)
14:07:33 Join GeekShadow [0] (~Antoine@reactos/tester/GeekShadow)
14:09:17n1samiconn: maybe you could try playing with the forkchunk and proc_retry environment vars in cygwin, see i'm really grasping at straws now though :/
14:12:24CIA-7New commit by jethead71 (r29070): One change in r29068 wasn't right. Clarify comments on why it's the way it is. :)
14:13:27 Quit robin0800 (Remote host closed the connection)
14:17:05CIA-7r29070 build result: All green
14:18:40amiconnn1s: Yes, my cygwin installation is up to date. Also my laptop has 4 GB RAM, so that should definitely not cause problems
14:19:01n1sthen i really have no clue
14:21:44 Join robin0800 [0] (
14:25:24 Quit Zarggg_ (Quit: Zarggg_)
14:29:42 Join T44 [0] (
14:31:06 Quit robin0800 (Quit: Leaving)
14:31:24***Saving seen data "./dancer.seen"
14:31:31 Join robin0800 [0] (
14:33:03 Quit Topy (Ping timeout: 240 seconds)
14:48:00CIA-7New commit by teru (r29071): image viewer: ...
14:50:40 Quit T44 (Read error: Connection reset by peer)
14:51:59CIA-7r29071 build result: All green
15:06:09 Quit robin0800 (Quit: Leaving)
15:06:21 Join robin0800 [0] (
15:08:15 Join Topy44 [0] (
15:09:10 Join komputes [0] (~komputes@ubuntu/member/komputes)
15:13:21 Join TheLemonMan [0] (
15:21:16 Join robin0800_ [0] (
15:21:21 Quit robin0800_ (Remote host closed the connection)
15:30:04 Quit teru (Quit: Quit)
15:41:42 Join bmbl [0] (~bmbl@unaffiliated/bmbl)
15:42:44 Join GeekShad0w [0] (
15:42:52 Quit GeekShadow (Read error: Connection reset by peer)
15:48:12 Join MethoS- [0] (~clemens@
15:57:26 Join slooopy [0] (
16:04:11 Join Keripo [0] (
16:14:20 Join leavittx [0] (~leavittx@
16:14:30 Join Kitr88 [0] (
16:15:09sideralpamaury: Do you have any idea regarding my last comment to FS #11664 and FS #11877?
16:18:33 Quit Kitar|st (Ping timeout: 255 seconds)
16:18:48pamaurysideral: no sorry, all this seems really strange since it never occured to me (might be related with the fact that I always use a SD card apparently). I know something is wrong with the driver but I don't really understand how it can corrupt the disk. Do you know if it a transfer which succeeds but with garbage data ?
16:19:05 Quit mystica555_ (Ping timeout: 250 seconds)
16:19:35 Quit Kitr88 (Ping timeout: 276 seconds)
16:20:14pamauryGiven that several people report various problems with the amsv2 driver, I might need to dig into the OF to find some answer
16:20:51sideralpamaury: it doesn't actually corrupt the disk, the disk (typically the FAT) just appears corrupted to the Linux host because apparently some garbage is inserted in the USB stream. The corruption will end up on the disk only if the host updates the data.
16:21:36pamauryokay so it's just that usb data is wrong ?
16:21:59pamaurywell *just*
16:22:11sideralthat is my current theory, because after disconnecting and remounting the player without fixing the FS, all is fine
16:23:10sideralsometimes it takes a reboot of the player, though, to make the phantom garbage go away. The OF never exposes any garbage
16:23:17 Quit mystica555 (Read error: Connection reset by peer)
16:23:32pamauryif it can help you, something is absolutely sure: the usb data can't be altered on cable, there is achecksum so the controller would notice it and normally the driver should also report an error (don't know that the usb storage driver does however, you might want to check this)
16:23:59 Join Kitar|st [0] (
16:25:02sideralUnfortunately I don't know yet how to trigger the problem reliably. It typically involves replacing some large podcast files (deleting some, adding some).
16:25:35pamauryWhen I have some time, I'll try to have a look at the OF but I really can't explain the garbage. The same thing occured me when debugging the the bus-reset-all-the-time: garbage data but I never found out
16:27:08sideralThanks pamaury.
16:27:53pamauryOh, wait a bit :) Last time I look at the OF I gave up since the usb code is written in C++ with indirect register adressing. Making it difficult to reverse engineer
16:29:29sideralOops −− but how did you arrive at the driver we have today, which (apart from this problem) works rather splendidly? Do you have a data sheet?
16:30:35sideralMy next idea, apart from verifying that the garbage isn't read off the disk or generated by the SD driver, is to use Wireshark to have a look at the USB traffic
16:31:27 Quit GeekShad0w (Ping timeout: 240 seconds)
16:31:28***Saving seen data "./dancer.seen"
16:31:42pamauryI have the code of a linux driver of it (although the code is horrible and not targetting the exact same SoC) and some datasheet (incomplete, for another SoC and another version) :)
16:33:12pamauryThe problem is the amsv2: it's not a real SoC, it does not correspond to something well documented, the only part which is reverse engineered is the init code and I fear it's incomplete
16:36:18 Join Clive [0] (
16:36:20 Join mystica555 [0] (
16:36:36 Join mystica555_ [0] (
16:36:38 Quit mystica555_ (Excess Flood)
16:36:44pamauryI personally trying to monitor the traffic (I have the rockbox usb analyzer) but concluded that no garbage data was on cable, just that the controller was doing something wrong (typically, stop responding)
16:36:45 Nick Clive is now known as Guest66012 (
16:37:05 Join mystica555_ [0] (
16:37:12pamaurybut it might be a different problem of course
16:37:51 Quit kugel (Ping timeout: 240 seconds)
16:39:10pamauryBut apart from that, your work on FS corruption is great, it's a really nasty problem and it will be a major accomplishment when it will be fixed
16:39:12sideralI might try to see whether your reset workaround is related to my problem (whether it only manifests when a bus reset has occurred)
16:39:27 Quit DerPapst (Ping timeout: 240 seconds)
16:40:16sideralThanks pamaury!
16:41:27 Join GeekShad0w [0] (
16:42:11Guest66012hi, i've been trying to register in the forums, but it keeps reporting i've typed invalid captcha ...
16:42:44[Saint]so type a valid one ;)
16:43:12sideralpamaury: I don't understand that USB 0x7d/0x7e escaping thing yet, Google has been only of marginal help, but apparently it's used only for command transfers. Perhaps the driver reset puts the SoC in a funny state where it escapes more than it should
16:43:16[Saint]seriosly though, IIUC you can just keep reloading the captcha until you get one you can parse easily
16:43:26[Saint]Guest66012: ^
16:44:18sideralGuest66012: I seem to remember that I had to enable cookies for the forum registration to work. Did you perhaps disable cookies?
16:44:26Guest66012i'm sure i've typed it right (several times)
16:44:29pamaurysideral: My guess is that the controller is already in a funny state :) Although I would be surprized that some data escape if it's a really low level part of the usb protocol
16:44:52Guest66012cookies are enabled
16:45:29[Saint]Guest66012: If you typed it correctly, it would accept it.
16:45:50[Saint]you can also try the audio captcha as well.
16:46:06sideralpamaury: me too... although the pattern I found (see is a bit too obvious
16:46:20 Quit Battousai (Ping timeout: 260 seconds)
16:46:46[Saint]just click the re-captcha button until you get one you can parse easily.
16:47:04pamauryit might also be that the controller becomes made and the dma copies some strange data.
16:47:11Guest66012you don't believe me... never mind
16:48:24[Saint]Guest66012: I believe you, I just tried the captcha system's working here.
16:50:48 Join DerPapst [0] (
16:51:06 Quit dfkt (Read error: Connection reset by peer)
16:51:10 Join dfkt [0] (dfkt@unaffiliated/dfkt)
16:51:52[Saint]Guest66012: Yes, I can confirm that the captcha is working correctly here.
16:52:02[Saint]I have tried it a few times now.
16:53:37 Join CaptainKewl [0] (
16:56:23 Part LinusN
16:57:15[Saint]User woo2y successfully registered approximately an hour, it is working.
16:57:26CIA-7New commit by rmenes (r29072): Tagalog: fix two strings changed in source. Translation by ...
16:59:07 Quit n1s (Quit: Lämnar)
16:59:48 Quit sideral (Quit: Leaving.)
17:00:48 Quit TheLemonMan (Quit: free(me))
17:01:31Guest66012weird, check this out: i tried on 2 computers now
17:01:38CIA-7r29072 build result: All green
17:07:39 Join sideral [0] (~sideral@unaffiliated/sideral)
17:11:04 Quit crwl (Ping timeout: 265 seconds)
17:11:11 Quit sideral (Remote host closed the connection)
17:11:19 Quit robin0800 (Quit: Leaving)
17:12:14 Join sideral [0] (~sideral@unaffiliated/sideral)
17:16:56[Saint]Hmmmm....ok, now the captcha is giving me grief also.
17:17:06 Part Zagor
17:20:25Guest66012at least i'm not alone now ;)
17:20:30 Join robin0800 [0] (
17:21:26 Quit sideral (Remote host closed the connection)
17:21:47 Join sideral [0] (~sideral@unaffiliated/sideral)
17:24:34 Join crwl [0] (
17:28:57 Quit alexbobP (Ping timeout: 265 seconds)
17:29:29 Join alexbobP [0] (
17:33:17sideralpamaury: Did you reverse engineer the initialization routines of the Clip+ or the ClipV2? There might be significant differences
17:34:14pamauryerr, not sure, let me checl
17:34:29 Quit robin0800 (Remote host closed the connection)
17:36:06pamauryno, it was clipv2
17:43:57 Join robin0800 [0] (
17:44:54 Join newClipUser [0] (
17:47:09 Join domonoky [0] (~Domonoky@rockbox/developer/domonoky)
17:47:30 Join saratoga [0] (9803c6dd@gateway/web/freenode/ip.
17:48:31 Quit sideral (Ping timeout: 255 seconds)
17:50:09 Join sideral [0] (~sideral@unaffiliated/sideral)
17:51:34 Quit Guest66012 (Quit: CGI:IRC)
17:54:00 Quit petur (Quit: Page closed)
17:54:39 Join panni_ [0] (
17:55:49sideralpamaury: darn! so I can't blame it on you having used the "wrong" firmware ;)
17:57:04 Quit Guinness` (Read error: Connection reset by peer)
17:59:48 Quit merbanan (Ping timeout: 255 seconds)
18:04:21bluebrotherpixelma: my suspicion would be a different version of KOMAscript
18:04:36bluebrotherprobably there was a change in style
18:04:39 Quit robin0800 (Remote host closed the connection)
18:04:43 Join Guinness [0] (
18:05:01 Join sideral1 [0] (~sideral@unaffiliated/sideral)
18:05:58pixelmahmm, how to check, again?
18:06:08 Quit sideral1 (Remote host closed the connection)
18:06:27 Join sideral1 [0] (~sideral@unaffiliated/sideral)
18:06:34 Quit sideral (Ping timeout: 250 seconds)
18:09:23 Nick benedikt93 is now known as benedikt93|AFK (~benedikt9@unaffiliated/benedikt93)
18:13:11 Join jae_ [0] (
18:15:36 Nick jae_ is now known as jae (
18:15:57 Quit newClipUser (Ping timeout: 276 seconds)
18:16:23 Quit sideral1 (Remote host closed the connection)
18:17:26 Join sideral [0] (~sideral@unaffiliated/sideral)
18:19:03 Join newClipUser [0] (
18:20:23 Quit MethoS- (Remote host closed the connection)
18:22:25 Join MethoS- [0] (~clemens@
18:23:39 Quit newClipUser (Ping timeout: 255 seconds)
18:26:38 Quit sideral (Remote host closed the connection)
18:26:48 Quit Keripo (Quit: Leaving.)
18:27:09 Join robin0800 [0] (
18:27:18 Join sideral [0] (~sideral@unaffiliated/sideral)
18:29:35 Join Mystery_Keeper [0] (~Mystery_K@
18:29:35 Quit andyph666_ (Read error: Connection reset by peer)
18:30:01Mystery_KeeperUsername AlekseyLyashin. Add me as wiki user please.
18:30:17[Saint]one sec.
18:30:57pixelmathought it could still be on there
18:31:14[Saint]it should be yes, I just thought of that.
18:31:20[Saint]I added it ~2 weeks ago?
18:31:32***Saving seen data "./dancer.seen"
18:31:34pixelmaI did
18:32:48bluebrotherpixelma: on my system this finds a line that seems to hold the version: grep CheckKOMAScriptVersion /usr/share/texmf/tex/latex/koma-script/scrreprt.cls
18:33:05bluebrotherseems my F14 distro still uses TeXlive 2007
18:33:30 Join designate72 [0] (
18:33:42bluebrotherwhich means that I'm still using KOMAscript 2, while there is a version 3 available
18:36:14 Join lifeless_ [0] (~lifeless@
18:37:01pixelmaseems to be in a different folder structure here
18:37:14[Saint]Mystery_Keeper: You *are* indeed added to the wikiusersgroup.
18:37:27Mystery_KeeperThank you.
18:37:32[Saint]However, you name is still blacked out, meaning you have not completed the registration.
18:38:01Mystery_KeeperGAH. How do I complete it? T_T
18:38:21[Saint]Have you missed the confirmation mail?
18:38:25 Join merbanan [0] (
18:38:32[Saint]did it go to a spam folder? etc.
18:38:36Mystery_KeeperI guess. I'll look in the spam folder.
18:39:39Mystery_KeeperNo. It is not in spam, and I haven't received one.
18:40:27pixelmabluebrother: I can't even seem to find a koma-script folder
18:40:41[Saint]It's been a while since I did it myself, but I'm fairly confisent that the adress needs to be verified.
18:40:41 Quit swilde (Read error: Connection reset by peer)
18:40:55pixelmabluebrother: the fact that this is a commandline only VM doesn't help me here :\
18:41:18Mystery_KeeperDon't remember clearly, but I guess when I pushed the button after which I should have gotten confirmation mail, I was redirected to "strange change request" page or something.
18:41:27[Saint]pixelma: ^
18:41:45pixelmasure, if I knew how
18:42:06[Saint]find "<blah>"
18:42:15[Saint](no quotation)
18:45:55pixelmathat's either wrong (or not complete) or scrreprt.cls doesn't exist anywhere in this VM, yet I'm able to successfully build PDF and even HTML manuals
18:46:15bluebrotherpixelma: try "locate scrreprt"
18:46:44 Quit einhirn (Quit: Miranda IM! Smaller, Faster, Easier.
18:47:32[Saint]Mystery_Keeper: Can you try loggin in now please?
18:48:32Mystery_KeeperPlease remind me how do I log in without trying to change anything.
18:48:54[Saint]just try to edit a wiki page
18:49:03[Saint](but, don't actually edit it, if you can)
18:49:23[Saint]once you click edit, you'ff be prompted with a login.
18:49:53[Saint]I'm AWESOME!
18:50:01[Saint]I created your page manually.
18:50:01Mystery_KeeperYou are. Thank you.
18:50:11[Saint]you'll need to edit your details, as there are none.
18:51:45pixelmabluebrother: aha, thanks. That helped - the subfolder is called "texmf-texlive" here, not only "texmf". The grep line now tells me that it's version 2.95b of komascript, I could still check the other in my cygwin environment (which gives more similar results to the daily manuals)
18:53:08[Saint]I'm really sorry I didn't think of that earlier Mystery_Keeper, I wasn't exactly sure that I wouldn't break it more by trying what I did ;)
18:53:21[Saint]Sorry it's been such a hassle for you.
18:53:46Mystery_KeeperIt is over, so that's alright.
18:57:19 Join kugel [0] (~kugel@rockbox/developer/kugel)
18:57:41 Join Keripo [0] (
18:59:00pixelmabluebrother: you are probably correct but I can't tell for sure. The cygwin scrreprt.cls doesn't contain exact version information but from the (c) it is 2 years older than the one in my VM
19:04:50 Quit jae (Quit: Lost terminal)
19:05:38 Quit sideral (Ping timeout: 265 seconds)
19:06:38 Join sideral [0] (~sideral@unaffiliated/sideral)
19:07:11 Join bertrik [0] (
19:07:11 Quit bertrik (Changing host)
19:07:11 Join bertrik [0] (~bertrik@rockbox/developer/bertrik)
19:07:16CIA-7New commit by pixelma (r29073): Manual - minor fixes to fractals.tex: update screenshot capture to reflect the renaming of the plugin from 'Mandelbrot' and replace an \opt which was ...
19:09:31CIA-7New commit by pixelma (r29074): Manual - lrcplayer.tex: correctly 'close' two \Action macros so that the following space is not eaten.
19:10:15 Quit sideral (Remote host closed the connection)
19:11:07 Join sideral [0] (~sideral@unaffiliated/sideral)
19:11:11CIA-7r29073 build result: All green
19:11:27 Quit ender` (Quit: It's bad luck to be superstititious. -- Law of Superstition)
19:14:24 Join ender` [0] (
19:14:43 Quit ender` (Read error: Connection reset by peer)
19:14:47CIA-7r29074 build result: All green
19:14:57 Join ender` [0] (
19:17:31 Join Battousai [0] (~bryan@gentoo/developer/battousai)
19:20:31 Quit sideral (Remote host closed the connection)
19:20:45 Join sideral [0] (~sideral@unaffiliated/sideral)
19:25:24 Quit krazykit (Read error: Connection reset by peer)
19:30:43CIA-7New commit by pixelma (r29075): Pick up the target rename completion in the manual started in r27823 ... this time taking care of the Iriver targets. This fixes a few issues with ...
19:30:48sideralgevaerts, pamaury: I now have the ClipV2 in the faulty state again and took a few FAT dumps. And: Each dump is faulty, but each has the corrupt bytes at different offsets. But still all corruptions always fit the patterns I described in FS #11877
19:32:27 Quit Mystery_Keeper (Quit: KVIrc 4.0.2 Insomnia
19:33:57*[Saint] wonders what it is that sideral has discovered.
19:34:01CIA-7r29075 build result: All green
19:34:14[Saint]( I bet sideral does also ;) )
19:34:41sideralYes I do :)
19:35:28 Join krazykit [0] (
19:37:22 Join kkit|sh [0] (
19:38:20 Join Horscht [0] (~Horscht@xbmc/user/horscht)
19:38:58pixelmahmm, small unrelated change slipped in there (one line in a button table for the Onda vx777), oh well
19:43:14 Quit GeekShad0w (Ping timeout: 240 seconds)
19:45:02 Join GeekShad0w [0] (
19:47:50 Quit benedikt93|AFK (Quit: Bye ;))
19:48:29 Quit kevku (Read error: Operation timed out)
19:49:34 Join TheSeven [0] (~TheSeven@rockbox/developer/TheSeven)
19:50:32 Join solexx_ [0] (
19:52:30 Join zerozorro [0] (
19:52:33 Join evilnick [0] (~evilnick@rockbox/staff/evilnick)
19:53:13 Quit zerozorro (Client Quit)
19:53:23 Quit solexx (Ping timeout: 246 seconds)
19:53:34 Join zerozorro [0] (
19:54:17 Quit zerozorro (Client Quit)
19:54:21 Join zerozorro [0] (
19:54:29 Quit zerozorro (Client Quit)
19:54:50 Join glued [0] (
19:58:46gluedsideral, on reading today's log I wanted to confirm the filesystem corruption issue on Clip+, and also to express a huge, huge gratitude to you, pamaury & everyone who is trying to figure it out. I'm not a dev, but frequently search these logs & FS for 'usb' in hope :|
20:04:55sideralHi glued, thanks for your kind words! Can you describe your corruption issue in more detail?
20:05:51sideralDo you also have transient corruptions manifesting as cyclic FAT chains?
20:09:44 Join Strife89 [0] (~Strife89@
20:16:27[Saint]kugel: Is the only reason your 480x800 cabbie misses battery and lock icons due to lack of images, or is there more to it than that?
20:16:48kugelmostly lack of space
20:17:10[Saint]ah, right...I'm trying to figure out why mine isn't loading on touch devices.
20:17:30[Saint]pretty much all I can do is comment out the touch regions one by one.
20:17:52gluedsideral, 1. I see no pattern at all, just occurs every other time 2. Occurs only on the sd card (32 Gb Sandisk) 3. Occurs always, from with FS #11664 to when only the usb enabled, w/o any other fixes 4. Occured before and after FS #11774 was commited, no effect on that 5. I didn't look into the error description deeply (sry), and switched to OF usb recently
20:18:05kugeldid you double check if they don't exceed viewport limits?
20:18:08[Saint]I can see from your code that touch regions *are* viewport relative, which is nice...the themeeditor doesn't support this.
20:18:19kugelyes they are
20:18:31[Saint]kugel: yes, everything is within the VP boundaries
20:19:20kugelI can't think of another reason why touch regions could cause problems
20:19:21[Saint]the gigabeast sim loads it fine, so I figure it must be a problem with touch regions in my theme
20:19:39[Saint]no debug output makes this a PITA though.
20:20:39sideralglued: what exactly did chkdsk / fsck say?
20:22:48sideralah, you already said you didn't look at the error description, nevermind
20:25:33[Saint]is there a 240x320 touchscreen target we support in the themesite?
20:25:50[Saint]if I could upload it to the themesite, that should tell me what's wrong.
20:26:37pixelmathe closest you could get is one of the Ondas which are 240x400 IIRC
20:27:10kugelyou can compile checkwps on your own though
20:27:21[Saint]not with cygwin, apparently.
20:27:33kugeli thought you're on ubuntu now?
20:28:18[Saint]the ubuntu VM is crawling through setting up the android toolchain at the moment.
20:28:40kugelubuntu is a multitasking OS :)
20:28:44pixelmacompiling checkwps on cygwin hasn't worked in ages as far as I remember
20:29:12[Saint]kugel: It is, yes...but I'm rather pushing the system at the present.
20:29:14pixelmaalthough I only tried ages agp, not since ;)
20:29:40 Join Buschel [0] (
20:30:52[Saint]if someone wanted to compile/upload a mini2440 checkwps...I'd be quite appreciative
20:30:58gluedsideral, if you ever post some kind of a dumping tool, I'd re-enable the usb, use it and submit. but i was too pissed off to remember when the chkdsk were saying (very sry). i do remember what i was saying in great detail, but that won't help.
20:31:05kugel[Saint]: I'd rather not have the theme site abused just for testing purposes
20:31:27Strife89Anyone been having weird USB issues with a Fuze v1?
20:31:28[Saint]kugel: if the syntax is invalid, it will point it out, and not complete.
20:31:32[Saint]what's wrong with that?
20:31:33***Saving seen data "./dancer.seen"
20:31:38kugelI assure compiling on the vm won't kill it
20:31:50Strife89Mine has been disconnecting at random, even in mid-transfers.
20:32:00Strife89No FS corruption is evident.
20:37:22 Join LambdaCalculus37 [0] (
20:37:22 Quit LambdaCalculus37 (Changing host)
20:37:22 Join LambdaCalculus37 [0] (~rmenes@rockbox/staff/LambdaCalculus37)
20:40:19CIA-7New commit by rmenes (r29076): New translation: Croatian by Ante Maretic. From ...
20:42:36 Quit Llorean (Quit: Leaving.)
20:43:20CIA-7r29076 build result: All green
20:45:28[Saint]ARGH!!!! I'm beginning to suspect that Jd may not be running a current SVN .apk?
20:45:47[Saint]*apparently*, the syntax is in order.
20:46:32[Saint]If it works on my phone after I compile the .apk...I'm gonna be a 'lil pissed.
20:46:40sideralgevaerts, pamaury: I now have a wireshark dump of the USB traffic taken during one dd. I can see the corruption in it, nicely embedded in some of the bulk data frames. But I don't quite know what else to do with it
20:47:24pixelmaLambdaCalculus37: I don't think that the translation.langs should get the "master language file" header from english.lang
20:47:44LambdaCalculus37pixelma: I'm going to pull that all out in a moment.
20:48:20[Saint]kugel: Just a big is the Android skin buffer?
20:48:27pixelmaLambdaCalculus37: you can look at how the other translations are done
20:48:38[Saint]errr...RaaAoA skin buffer, rather
20:48:53[Saint] I can't be overflowing any buffers.
20:49:00LambdaCalculus37pixelma: Most of them don't have the huge header. I've just edited it all out.
20:49:29CIA-7New commit by rmenes (r29077): Don't need that honking huge header, so out it goes.
20:49:29[Saint]if I didn't deliberately mame the screenshot...the themesite *would* have allowed me to upload this.
20:49:43[Saint]so, that *must* mean the syntax is correct...yes?
20:49:43pixelmaLambdaCalculus37: they have one that lists the translators
20:50:40pixelmaLambdaCalculus37: I guess they should have *some* header...
20:50:58[Saint]the themesite uses checkwps to parse skin syntax. yes? Or some other magic?
20:51:26CIA-7New commit by rmenes (r29078): Last fix, I swear: add in the proper header.
20:51:34LambdaCalculus37pixelma: All fixed.
20:51:57pixelma[Saint]: it does, I'm not sure though if the theme site's checkwps is updated automatically, actually I doubt that
20:52:13[Saint]it would have to be, no?
20:52:21 Join TheLemonMan [0] (
20:52:30[Saint]otherwise the "works with current revision" would be horribly broken.
20:52:48CIA-7r29077 build result: All green
20:53:03LambdaCalculus37rasher: New language for your translation page.
20:53:05[Saint]as of right now...I see no reason why this should not load on Jd's phone.
20:53:32[Saint]If there *IS* a reason ...I'd love to know what it is.
20:56:21 Join milk [0] (
20:56:37 Quit glued (Quit: CGI:IRC)
20:56:54CIA-7r29078 build result: All green
20:57:24 Quit panni_ (Quit: ( :: NoNameScript 3.81 :: ))
20:57:42 Join panni_ [0] (
20:58:15 Join sudela [0] (
20:58:52 Quit [Saint] (Quit: I'm only going to Heaven if it feels like Hell, I'm only going to Heaven if it tastes like caramel...)
21:00:39 Quit Keripo (Quit: Leaving.)
21:03:52 Quit sideral (Ping timeout: 240 seconds)
21:04:13 Quit LambdaCalculus37 (Quit: Fwump)
21:06:12 Join n1s [0] (
21:06:13 Quit n1s (Changing host)
21:06:13 Join n1s [0] (~n1s@rockbox/developer/n1s)
21:08:35 Join domjohnson [0] (0279c72e@gateway/web/freenode/ip.
21:08:49domjohnsonIs there a port to the Sansa View?
21:09:34n1sno, not really
21:10:40domjohnsonIs there any way I can help development?
21:10:47domjohnsonI'm not a great coder, but maybe beta testing or something?
21:11:25Torneit is not even relaly started
21:11:37bluebrotherdomjohnson: see
21:11:39Torneif someone works on it then maybe you cna test it months after that :)
21:21:57domjohnsonIs there any other firmware, apart from Rockbox and retail, that I can use from the view?
21:22:04domjohnsonJust because the retail firmware fails.
21:23:56 Quit Buschel (Quit: ChatZilla 0.9.86 [Firefox 3.6.13/20101203075014])
21:26:04 Join Llorean [0] (~DarkkOne@rockbox/user/Llorean)
21:29:15 Join stripwax [0] (
21:30:43domjohnsonOr is there any way I can help with development with the Sansa View, bearing in mind I have never coded for hardware before?
21:32:07sudelau could pray
21:37:07saratogabuy a better player
21:38:08bluebrotherdomjohnson: coding for hardware isn't the main problem. The main problem is that the hardware in question is almost completely undocumented
21:38:28 Join moos [0] (
21:38:28 Quit moos (Changing host)
21:38:28 Join moos [0] (moos@rockbox/staff/moos)
21:38:44domjohnsonI see
21:39:03domjohnsonsaratoga - I would. But I just bought it last week.
21:39:08domjohnsonSecond hand...
21:39:10domjohnsonlisted as working fine
21:39:18domjohnsonAnd it was...until I formatted it from the player
21:39:26domjohnsonActually, it wasn't working fine. Stuff wouldn't delete properly.
21:39:48saratogatheres so many better sandisk players that sell for almost nothing, i wouldn't bother with the view
21:40:21 Join Keripo [0] (
21:40:26domjohnsonI wish I hadn't.
21:40:52domjohnsonIf it isn't fixed soon, I might ask the seller if I can return it...or sell it on eBay as spare parts.
21:42:37n1sit it's got hardware faults rockbox wouldn't have helped anyway
21:43:40 Quit slooopy (Ping timeout: 250 seconds)
21:54:52 Quit rasher (Ping timeout: 272 seconds)
21:55:46 Quit pikytcus (Ping timeout: 265 seconds)
21:56:21 Join slooopy [0] (
21:56:40 Quit scorche|sh (Ping timeout: 260 seconds)
21:57:40 Join perrikwp [0] (18a7a08e@gateway/web/freenode/ip.
21:58:49 Join moos_ [0] (
21:59:23 Quit moos (Ping timeout: 240 seconds)
21:59:31 Nick moos_ is now known as moos (
22:01:57domjohnsonn1s - it isn't a hardware fault, its a firmware fault
22:01:59 Join rasher [0] (
22:01:59 Quit rasher (Changing host)
22:01:59 Join rasher [0] (~rasher@rockbox/developer/rasher)
22:05:34 Join pikytcus [0] (
22:06:10 Join scorche|sh [0] (
22:09:13 Join [Saint] [0] (S_a_i_n_t@
22:10:54CIA-7New commit by moos (r29079): Set svn:keywords for the new translation file.
22:13:51CIA-7r29079 build result: All green
22:13:54 Join AlexP_mob [0] (~ap@rockbox/staff/AlexP)
22:31:35***Saving seen data "./dancer.seen"
22:33:12 Quit chattr (Quit: gone)
22:41:07 Join Rob2222 [0] (
22:41:10 Join chattr [0] (
22:48:49 Quit GeekShad0w (Ping timeout: 240 seconds)
22:52:25 Quit Strife89 (Quit: Heading out)
23:00:44 Quit AlexP_mob (Ping timeout: 264 seconds)
23:02:25CIA-7New commit by moos (r29080): Add ipod 6g target to translations for "Bass and Treble Cutoff" string.
23:04:31 Join GeekShadow [0] (~Antoine@reactos/tester/GeekShadow)
23:05:35CIA-7r29080 build result: All green
23:07:52*bluebrother searches for sideral's real name
23:09:08 Quit bertrik (Ping timeout: 264 seconds)
23:10:02pixelmabluebrother: mentioned during commits lately but I couldn't find it in flyspray either
23:11:23*bluebrother would _really_ like a realname policy in Flyspray
23:11:33pixelmaat least not in the account, maybe as a comment in the more discussed patch
23:11:47bluebrotherwe have one for the wiki, so why not for Flyspray?
23:12:13 Join vnl [0] (
23:12:20 Part vnl
23:13:15moosbluebrother: Michael Hohmuth, in case you didn't get it yet
23:14:47bluebrothermoos: thanks. No, was still looking
23:15:19moosno problemo
23:15:51*bluebrother waits for the compiler to finish. Last check if that also fixes the pause issue on nano2g
23:17:01 Quit TheLemonMan (Quit: free(me))
23:18:51n1sbluebrother: i agree about Real Names in the tracker, would be nice
23:20:46moosindeed. I encountered the problem lately with a translation. Had to mail the authors...
23:21:14bluebrotherit would be at least good for anyone posting a patch.
23:21:35bluebrotherespecially in case that patch gets picked up years later and the original author isn't around anymore :)
23:21:45 Quit domjohnson (Ping timeout: 265 seconds)
23:21:47mooshehe, very true :)
23:22:06bluebrotherreminds me ... I wanted to pick up the BOFH plugin
23:22:22bluebrothersince we have tetris1d I don't see a reason not to have an excuses plugin :)
23:25:32bluebrotheryay. fixes the nano2g pause issue too.
23:25:45bluebrotherstrange, why does pausing the music try to log playback data?
23:28:40CIA-7New commit by bluebrother (r29081): Fix player becoming unresponsive on Stop / Pause in some cases. ...
23:28:41 Quit markun (Read error: Connection reset by peer)
23:32:04bluebrotheroops, missed to fix the swapped digits in the FS number in siderial's commit message :o
23:32:47CIA-7r29081 build result: All green
23:37:54 Quit mystica555 (Read error: Connection reset by peer)
23:38:20 Quit mystica555_ (Read error: No route to host)
23:46:44 Join sideral [0] (~sideral@unaffiliated/sideral)
23:46:57 Quit slooopy (Read error: Operation timed out)
23:49:28 Join mudd1 [0] (
23:49:39gevaertssideral: I guess that wireshark output proves that the problem isn't on the host side, which we kind of suspected already...
23:50:37 Quit Rob2222 (Read error: Connection reset by peer)
23:50:50 Join Rob2222 [0] (
23:51:51sideralgevaerts: Yeah. Also, the corruption is not at some USB packet boundary or another corner case like that
23:51:57sideralbluebrother: I require some level of privacy and anonymity, sorry.
23:52:43sideralIf you require real names in the tracker, I'm afraid I won't be able to contribute.
23:52:47 Quit moos (Ping timeout: 246 seconds)
23:53:06 Join moos [0] (
23:53:06 Quit moos (Changing host)
23:53:06 Join moos [0] (moos@rockbox/staff/moos)
23:53:10 Quit komputes (Remote host closed the connection)
23:53:41 Quit tchan (Quit: WeeChat 0.3.3-dev)
23:54:07bluebrothersideral: well, your real name is in the svn logs anyway (and the CONTRIBUTORS file) so I don't see much point here
23:54:51saratogajust put a space or something in it so it doesn't show up in google searches
23:54:53bluebrotherI'm not the one to _require_ real names in the tracker, that decision is not up to me. But since Rockbox has a real name policy I would like to see people using their real name in the tracker
23:55:16sideralYeah, but I don't want to see it plastered all over the Internet. This way, at least some effort beyond simply googling is required to make the connection :)
23:58:56n1sthe annoying thing is that casual contributers that might post a patch that someone could just commit at once don't leave their names, so we need to ask for them and then wait for a reply that sometimes never comes

Previous day | Next day