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 2012-09-24

00:15:23 Join pedro_angelo [0] (
00:21:04 Join Scromple [0] (~Simon@
00:31:21 Quit sciopath (Read error: Connection reset by peer)
00:31:41 Join Scr0mple [0] (~Simon@
00:35:14 Quit Scromple (Ping timeout: 256 seconds)
00:36:20 Quit ender` (Quit: How do you generate a random string? Put a new user in front of VIM and tell him to save and quit. -- Duane Morin)
00:37:06 Quit lebellium (Quit: ChatZilla 0.9.89 [Firefox 16.0/20120919065210])
00:46:34sc_i am selling my ipod 6g classic
00:46:49sc_is it easy to remove efuse and rockbox?
00:47:11sc_er not efuse
00:48:16gevaertsrockbox is easy, just remove the directory. For emcore, I suspect the people in #freemyipod know
00:49:23sc_itunes :|
00:49:45sc_$100 for a 6gen 160gb ipod is pretty reasonable right?
00:49:56sc_used/typical scratches
00:53:26 Nick StopMakingSense is now known as LittleCreature (~fearofmus@unaffiliated/fearofmusic)
00:54:30 Quit bertrik (Ping timeout: 256 seconds)
01:01:39TheSevensc_: just boot into DFU mode like you did during installation and restore using itunes
01:08:46 Join amayer [0] (
01:10:13 Quit mgottschlag (Ping timeout: 240 seconds)
01:15:07sc_TheSeven: thanks, will do that tomorrow
01:15:09sc_i hate itunes :|
01:16:13TheSevenit can in theory be done without itunes, but it's a damn lot more complicated that way
01:49:25***Saving seen data "./dancer.seen"
01:56:27the-kyleGot the &minus;&minus;uncoupled Opus bug filed. <a target="_blank" href="">FS #12754a>
01:56:28fs-bluebot<a target="_blank" href="">> Opus decoder fails to play stereo tracks encoded with the "&minus;&minus;uncoupled" option (bugs, unconfirmed)
01:57:54the-kyleI linked to a flac of the original sample along with 64k Opus versions, both with and without the &minus;&minus;uncoupled option.
02:31:47 Quit Totalled (Read error: Operation timed out)
02:35:56 Join Totalled [0] (
02:46:39[Saint]the-kyle: while technically a bug, I don't believe opus is supposed to support stereo at this stage.
02:46:54[Saint]it's a bit of a miracle playback works at all.
02:48:45the-kyleHmm. It works with the reference decoder on the computer, but maybe there's a problem with multichannel playback on embedded devices at this point that will be fixed in future versions of the decoder.
02:52:08[Saint]pretty sure the gerrit instance covered this in the description.
03:10:51the-kylesaratoga: Not sure. There have been some issues raised in relation to the fixed-point code, not sure where I read it. In that case, you can close my bug and I'll file it with the Opus devs.
03:11:03the-kyleOops, wrong person.
03:11:32the-kyleI was sure I deleted that. <smiles>
03:16:56the-kyleAnyway, if the bitrate is high enough, probably 128 or better, the stereo coupling shouldn't be lossy enough to cause a problem in the case of my examples. As a matter of fact, the 64k version didn't sound like there was too much of a problem with crossover or anything.
03:20:22the-kyleIn that case, since Vorbis doesn't stop coupling until 192 nominal bitrate (-q6), I probably don't need to worry too much about edge cases like mine. I'll just refile that bug in the right place, the Opus bug tracker.
03:23:58the-kyle Multichannel support doneish.
03:24:21the-kyleFrom the todo. I guess that tells me all I needed to know. :P
03:26:56[Saint]I'm on my phone, so, this wild guess may be wrong...but, I'm fairly certain the gerrit review for the task explained what was currently functional and what wasn't.
03:27:06[Saint]here goes: g300
03:27:09fs-bluebotGerrit review #300 at <a target="_blank" href="">> : Initial opus codec support by Frederik M J Vestre (changes/00/300/11)
03:27:24[Saint]ohhh! what a guess!
03:28:01the-kyleYes, that was it, but I didn't recall it mentioning uncoupled stereo.
03:28:21the-kyleCoupled plays beautifully on my clip+.
03:29:30[Saint]ah, it was the "multichannel support may not be parsed correctly" bit I was thinking of.
03:29:42EpicanisMy ancient Clip v1 and Sansa c240 don't handle them, but I get the feeling it's just not optimized sufficiently yet, being newly imported.
03:30:37the-kyleOh yes. For multichannel, I was reading 5.1 or similar, not uncoupled stereo. My mistake.
03:33:23the-kyleReading it again, I see that my problem is indeed known after all, in Rockbox, and it seems in the Opus todo itself also. Sorry for the premature bug.
03:49:29***Saving seen data "./dancer.seen"
04:22:58 Join amiconn_ [0] (amiconn@rockbox/developer/amiconn)
04:22:59 Quit amiconn (Disconnected by services)
04:22:59 Quit pixelma (Disconnected by services)
04:22:59 Join pixelma_ [0] (pixelma@rockbox/staff/pixelma)
04:23:01 Nick pixelma_ is now known as pixelma (pixelma@rockbox/staff/pixelma)
04:23:04 Nick amiconn_ is now known as amiconn (amiconn@rockbox/developer/amiconn)
04:34:25 Quit eckoit (Quit: eckoit)
04:38:15 Join TheSphinX_ [0] (
04:41:56 Quit TheSphinX^ (Ping timeout: 248 seconds)
05:07:11 Part amayer
05:28:50 Quit TheSeven (Disconnected by services)
05:28:59 Join [7] [0] (~quassel@rockbox/developer/TheSeven)
05:32:03 Quit Epicanis (Quit: Good night, y'all.)
05:40:27 Join advcomp2019 [0] (~advcomp20@unaffiliated/advcomp2019)
05:43:54 Quit pedro_angelo (Ping timeout: 246 seconds)
05:49:32***Saving seen data "./dancer.seen"
06:56:52[Saint]So...who wants to talk metadata preferences with me?
06:59:01[Saint]I'm in the process of (re-re-re-re)re-writing the metadata display code for cabbiev2, but I don't think it is really something I should be doing without input from the community (developer/user) with reference to what is expected to be displayed, what is "relevant" information, order of preference for display, fallbacks for a given input...etc.
07:01:34[Saint]I know I can write an elegant self healing solution that "just works" when an expected field (and/or its fallback) is missing, but I keep tripping myself up with what is "relevant" info to display.
07:04:37[Saint]and then there's things like fallbacks that I just have NFI about. Like, you use %artist% or %albumartist%, which takes preference if both are there? what do you (attempt to) display if it's missing, etc.
07:22:31 Join MarkK105 [0] (
07:22:55MarkK105Does anyone here use Dap-Scrob?
07:30:14 Quit Scr0mple (Ping timeout: 240 seconds)
07:37:45 Join stoffel [0] (
07:44:02 Quit XavierGr (Ping timeout: 265 seconds)
07:46:13 Join Scromple [0] (~Simon@
07:49:36***Saving seen data "./dancer.seen"
07:52:54 Quit Rower85 (Read error: Connection reset by peer)
07:54:45 Join mortalis [0] (
08:02:59 Quit MarkK105 (Quit: ChatZilla [Firefox 18.0a1/20120923030601])
08:04:36 Quit bootlknvf (Ping timeout: 245 seconds)
08:12:52 Join LinusN [0] (
08:13:31 Part LinusN
08:21:52 Join ender` [0] (
08:38:31 Join bootlkjkgf [0] (~Prmhfhfx@
08:40:32 Join mgottschlag [0] (~quassel@reactos/tester/phoenix64)
08:42:11 Join bertrik [0] (~quassel@rockbox/developer/bertrik)
08:44:09 Quit Scromple (Ping timeout: 246 seconds)
08:46:29 Join kevku [0] (
08:47:40 Join Scromple [0] (~Simon@
08:47:56 Join Buschel [0] (
08:50:01 Quit Scromple (Client Quit)
08:59:27 Join Zagor [0] (
08:59:28 Quit Zagor (Changing host)
08:59:28 Join Zagor [242] (~bjst@rockbox/developer/Zagor)
09:09:00 Quit shai (Ping timeout: 246 seconds)
09:14:10 Join einhirn [0] (
09:18:25 Join petur [0] (~petur@rockbox/developer/petur)
09:22:17 Quit bertrik (Ping timeout: 264 seconds)
09:46:07 Join shai [0] (
09:47:10CIA-10Commit c555938 in rockbox by Thomas Martitz: checkwps: Do not error out on rwps for non-remote targets.
09:47:50CIA-10Commit bd00be9 in rockbox v3.12 by Thomas Martitz: checkwps: Do not error out on rwps for non-remote targets.
09:49:39 Quit advcomp2019 (Read error: Connection reset by peer)
09:49:40***Saving seen data "./dancer.seen"
09:51:09 Join advcomp2019 [0] (
09:51:10 Quit advcomp2019 (Changing host)
09:51:10 Join advcomp2019 [0] (~advcomp20@unaffiliated/advcomp2019)
09:52:49 Join shai_ [0] (
09:52:51 Quit shai (Ping timeout: 256 seconds)
09:55:48 Join advcomp2019_ [0] (
09:55:49 Quit advcomp2019_ (Changing host)
09:55:49 Join advcomp2019_ [0] (~advcomp20@unaffiliated/advcomp2019)
09:59:08 Quit advcomp2019 (Ping timeout: 256 seconds)
09:59:53 Join JdGord [0] (~AndChat80@
10:08:25 Quit JdGord (Read error: Connection reset by peer)
10:08:48 Quit advcomp2019_ (Read error: Connection reset by peer)
10:09:37 Join LinusN [0] (
10:27:56 Quit mgottschlag (Ping timeout: 245 seconds)
10:36:17 Join JdGord [0] (~AndChat80@
10:41:04 Quit JdGord (Ping timeout: 256 seconds)
10:49:02 Join wodz [0] (
10:50:48 Join mgottschlag [0] (~quassel@reactos/tester/phoenix64)
10:57:34 Join JdGord [0] (~AndChat80@
11:02:06 Quit JdGord (Ping timeout: 245 seconds)
11:08:49kugelcan someone have a look at the build system?
11:09:12Cthulhuxyes. it looks like a computer.
11:10:03Zagorit has a sad :)
11:12:16 Join megal0maniac_ [0] (
11:12:46Zagorwe're short on android15 clients. only amiconn's machine offers it but fails to build.
11:13:03kugelreally? my client should do it
11:13:10*kugel has a look
11:13:36kugelit does advertise android15
11:14:16Zagoroh right. however: "Fatal build error: Compiler crashed. Blocking kugel-x-kugel." :-)
11:15:31kugelhow can that happen?
11:16:12wodzkugel: binutils people resistance to ack reloc bug is appealing. Looking at the elf32-arm.c they do provide relocation type to the data field of the veneer stub. They just don't make use of this apparently.
11:16:51kugelwodz: easy to fix?
11:17:01*kugel has never seen binutils code
11:17:12kugelZagor: the compiler works
11:17:28kugelis there a more detailed error message?
11:17:43wodzkugel: Are you kidding? First I would need to get drunk and then have a week of so of peace and silence to study code flow :-)
11:18:02kugelso that's a no :)
11:20:56Zagorkugel: aha, it wasn't the android compiler that caused this: <a target="_blank" href="">>
11:21:56Zagoramiconn: here is your build error: <a target="_blank" href="">>
11:24:03Zagorso the build server blocked kugel for broken ipod build and then amiconn for missing compiler. that left nobody to build android. BUG!
11:24:29kugelI can't help with internal compiler errors, sorry
11:24:43kugeli guess i could remove sdl from the list
11:24:58wodzZagor: maybe build system should skip the builds it can't find clients for?
11:25:49Zagorwodz: exactly, that's the bug. it does already when clients disconnect, but I missed doing it when clients are blocked
11:26:10gevaertsZagor: doesn't my client advertise android15?
11:26:14 Quit megal0maniac_ (Quit: Bye)
11:27:26Zagorgevaerts: it does, but it was not connected overnight
11:27:37gevaertsThat's interesting...
11:27:49wodzstatic fixed ICODE_ATTR pitchyin(int16_t *input, fixed *yin) <&minus;&minus;- what 'fixed' in this declaration mean? That is the first time I see something like this
11:29:04Zagorgevaerts: umm, my bad. it was.
11:32:15Zagorlooks like another bug. then amiconn was blocked, the android build was not abandoned (and rescheduled)
11:42:11 Join bluebrother [0] (~dom@rockbox/developer/bluebrother)
11:42:25 Quit Buschel (Quit: ChatZilla [Firefox 15.0.1/20120905151427])
11:44:24 Join XavierGr [0] (XavierGr@rockbox/staff/XavierGr)
11:44:37mortaliswodz: i think it's type
11:44:57mortalislook for define or typedef
11:45:06 Quit fs-bluebot (Ping timeout: 246 seconds)
11:45:24 Quit bluebrother^ (Ping timeout: 260 seconds)
11:45:27wodzcould be
11:46:06wodzah yes - typedef int32_t fixed
11:46:23 Join fs-bluebot [0] (
11:47:13 Join lebellium [0] (
11:47:42wodzInteresting thing. Many functions declared with ICODE_ATTR are inlined and as such doesn't end up in iram.
11:49:41***Saving seen data "./dancer.seen"
11:50:37wodzmortalis: any progress with lcd dma?
11:50:46 Quit mgottschlag (Ping timeout: 246 seconds)
11:52:00AlexPJdGordon: Is %LB supposed to be working at the moment (custom scroll bar)?
11:52:49ZagorI now restarted the last build
11:55:47CIA-10e3cd7ec build result: All green
11:59:32CIA-10Commit 3158f14 in www by Björn Stenberg: Check for fatal build error before doing build-complete housekeeping.
11:59:47mortaliswodz: I have almost working code for my dap. When i start playing some heavy codec (mp3, ape, wv) screen "jumps" few times. It doesn't happen when i start playing flac. Also it doesn't happen with any codec when lcd initialized by OF.
12:01:13wodzmortalis: can you pastebin, or put on gerrit?
12:02:26 Join JdGord [0] (~AndChat80@
12:03:40JdGordAlexP: no idea what the current state is
12:04:29mortaliswodz: I don't have it on this machine. The main changes i made is initialization order like in OF.
12:05:07mortalisAlso i removed while (DWDMA_CTL_L(0) & (1<<27)); but i can't recall was it necessary or not.
12:05:32wodzyou mean you don't wait for transfer to complete?
12:08:19wodzI would expect the jumps comes from initiated new transfer before the previous actually ended. Like lines are put into lcdif buffer, but before it reach lcd controller new transfer starts.
12:09:45 Join advcomp2019 [0] (
12:09:45 Quit advcomp2019 (Changing host)
12:09:45 Join advcomp2019 [0] (~advcomp20@unaffiliated/advcomp2019)
12:10:57AlexPJdGordon: You may guess that I was asking as it doesn't seem to work :)
12:11:14JdGordFile a bug
12:11:29AlexPJdGord: If I stick the sbs somewhere, could you have a look when you get a minute in case I'm using it wrong?
12:11:29JdGordWell u may have been doing it wrong!
12:11:35AlexPyes :)
12:11:49AlexPThen if it looks like it should, I'll open a bug report
12:13:37AlexPJdGord: If you get a minute, <a target="_blank" href="">>
12:15:10AlexPJdGord: Follow up question, is it known/intended/unavoidable that the "browse" touchscreen command doesn't seem to do anything if you are in the menus?
12:15:39JdGordNo that should work
12:15:57AlexPwrt browse?
12:16:03JdGordI'm on a plane and about to take off, prob wont have anu time to look for a while
12:16:09AlexPok, no worries
12:16:19JdGordYes, to the browse behaviour
12:16:23JdGordThat sounds like a bug
12:16:29AlexPOK, I'll open a bug for it, cheers
12:16:53JdGordIf u put up a zip in the next 5 min I might be able to play on the flightiest
12:17:00AlexPwill try :)
12:20:06AlexPJdGord: <a target="_blank" href="">>
12:20:20AlexPJdGord: Should all be in there except the fonts, which I left out for size
12:20:55JdGordI hope I have 7z installed
12:21:00JdGordLeaving the gate
12:21:06AlexPta :)
12:21:39 Quit JdGord (Quit: Bye)
12:38:19 Quit wodz (Ping timeout: 256 seconds)
12:38:33 Join mgottschlag [0] (~quassel@reactos/tester/phoenix64)
12:44:02 Quit stoffel (Ping timeout: 256 seconds)
12:48:01 Join lorenzo92 [0] (
12:48:16lorenzo92anyone's RDS expert? ;)
12:49:01lorenzo92I have set an IRQ that calls a function whenever GPIO of Si4709 radio chip is low (-> rds packet available)
12:49:07lorenzo92do I need to use RDS_ISR_PROCESSING ?
12:49:38lorenzo92because using the thread/semaphore way (as done in fuzeplus and as3525 targets) does segmentation fault after a while
12:49:38 Join n1s [0] (
12:49:39 Quit n1s (Changing host)
12:49:39 Join n1s [0] (~n1s@rockbox/developer/n1s)
12:55:56AlexPlorenzo92: I think bertrik did some RDS stuff
12:56:41kugellorenzo92: yes
12:58:00kugelI think there's a problem on RaaA/interrupt handling currently; you cannot use semaphores from within non-thread context
13:04:55lorenzo92uhm well the situation is that I have a callback set into the kernel (from radio module). The callback is called when gpio is falling
13:05:07lorenzo92I used semaphore in a thread
13:05:16lorenzo92well, only the "blocking" part
13:05:34lorenzo92the release was in the callback
13:08:08lorenzo92*blocking = wait :)
13:10:11 Join wodz [0] (
13:12:57lorenzo92but I don't get really the idea ... normally the callback is called and then rds_read_raw etc are called
13:13:04lorenzo92what's the problem in here?
13:13:27 Quit petur (Quit: *plop*)
13:18:33lorenzo92what about doing the "packet received" using the appropriate register? (in a thread of course)
13:20:15 Join stoffel [0] (
13:21:59 Nick Guinness` is now known as Guinness (
13:23:24n1swodz: thanks for the link yesterday, not much of interest in the speex header except the three line asm function i had already re-implemented :)
13:24:03n1sinlining and iram can be quite tricky, especially with the more recent gcc versions that try to be smarter
13:25:56wodzn1s: about rounding errors or how do you called it - I think this is a bug exposed by compilation for different arch. Maybe talk with upstream about it?
13:27:01 Join pedro_angelo [0] (
13:27:17n1swodz: yes, that's possible. I should try to investigate a bit more first i think, would be nice with an emulator :)
13:28:39wodzn1s: for arm you can use sdl RAaA on qemu arm
13:29:06wodzit works pretty good (a bit slow but who cares)
13:30:05n1sgot to be faster than running on target (especially if you're trying to get any info out)
13:30:25n1sany links for how to set that up?
13:31:02wodzn1s: I can dig up something - wait a sec
13:31:42wodz<a target="_blank" href="">>
13:32:52wodzThere are some images provided. I used the most recent one during unwinder for raaa debuging
13:34:12wodzwhen I get back home I can upload somewhere working image + start script
13:34:49n1sthat would be great, i'll look into this tonight or later this week i think
13:36:22wodzok, no problem
13:49:42***Saving seen data "./dancer.seen"
13:56:09 Join mounich [0] (
13:56:25mounichHello all
13:56:55mounichliitle question plz: are the new 64gb microsd card supported by rockbox?
13:57:14 Quit pedro_angelo (Remote host closed the connection)
13:57:42mounichI'm about to buy one but just if working with rockbox...
13:59:27AlexPmounich: should be, but you'll need to make it fat32
14:00:59mounichThank you AlexP, just frormating and it will works? Then let's buy one lol
14:01:20AlexPI seem to remember people on the forum reporting success
14:01:30AlexPBut I haven't tried it myself
14:01:31*the-kyle thought it was a difference in hardware, similar to MicroSD 2GB or less vs. MicroSDHC 4GB or more. So it may depend on the device.
14:01:46AlexPthe-kyle: I don't think it is yet
14:01:51AlexPbut it could be :)
14:02:11AlexPI might be talking arse though
14:02:32mounichCool, thank you fro your help
14:02:36the-kyleSome of the newer devices may support it.
14:02:43AlexPand the older
14:02:46lorenzo92kugel: I wrote a (well pretty dirty ^^) piece of code that reads the register that tells if rds packet is available, and if so processes the data: player hangs anyways
14:03:08mounichI'm about to get a sansa fuze+ and 64 gb card
14:03:36mounich16GB+64GB flash!!! woot
14:03:46AlexPmounich, the-kyle: "There are two kinds of SDXC cards: one kind is just an SDHC card with a higher capacity which is supposed to come preformatted as exFAT. This should work with Rockbox just fine as long as you reformat it as FAT32. The other kind is a "real" SDXC card which are faster but are not electrically compatible with SD/SDHC readers. This won't work."
14:04:17AlexPSo an old style one should work in any Rockbox player with microsd
14:04:21AlexPThe new types won't
14:04:30lorenzo92kugel: not immediately, but after a while. Quite the same amount of type if using IRQ callback!
14:04:43the-kyleAlexP: Oh wow! I'll get me one of those then. Thanks for the info.
14:04:58AlexPThat quote was from Torne by the way :)
14:05:34AlexPSee this thread, it seems to depend on the spec version
14:05:35AlexP<a target="_blank" href="">>,27811.0.html
14:05:38mounichlets find o no SDXC :)
14:06:47Torneyou may also need to change the partition type
14:06:55Tornebecause the exfat partition type is not supported by rockbox, even if you reformat it
14:07:01Torneand normal formatting tools will not change the actual partition's type
14:07:26the-kyleStandard partitioning tools should work for that.
14:07:28AlexP<a target="_blank" href="">>,33648.0.html for someone successfully using one
14:07:35wodzTorne: do you know internals of binutils by any chance?
14:07:43Tornenot really.
14:08:22wodzTorne: Have you seen the bug in ld I faced with bFLT?
14:08:56AlexPthe-kyle: Except the Windows built in ones, those will refuse to do fat32 for over 32 gb
14:08:57wodzTorne: <a target="_blank" href="">>
14:09:34wodzof course binutils people don't see this as a bug
14:09:53mounichhard to find a no SDXC at first glance
14:10:06the-kyleAlexP: Hmmm. Didn't know that. Although I've had quite a few problems when I've had to deal with Windows built-in partitioning tools.
14:10:36mounichare those microSD HC?
14:10:50AlexPJust use a third party one like fat32format or swissknife
14:10:55the-kyleDoes Rockbox support GPT partitioning, or only MBR?
14:11:08gevaertsmounich: what do you mean? All >32GB cards are SDXC
14:11:12AlexPThey don't have the stupid arbitrary restrictions Windows does
14:11:21the-kyleSo on Linux, I can just use standard fdisk or parted.
14:11:29AlexPmounich: They are SDXC, just a certain type of them
14:11:32AlexPthe-kyle: yes
14:11:59n1swodz, bertrik: the small error i observed on arm seems to be caused by one of my earlier changes as with clean master i get the same crc on both cf and arm but with my branch only arm differs
14:12:19gevaertsIt's just that there are two types of SDXC card, one compatible with SDHC readers (hardware-wise, anyway. The software/firmware needs to know about them), and one not
14:12:20mounichoh ok, sorry :) how a noob as me can make difference?
14:12:26AlexPI don't know
14:12:55gevaertsI'm not sure if the incompatible type is actually on the market yet, but I don't follow such stuff closely, so who knows?
14:13:04the-kyleSo how can I distinguish the type of card I'm buying before checkout?
14:13:19mounichLet's see the pointed forum
14:13:37Torneit's not clear that anyone has actually made sdxc 4.0 cards
14:13:55Torneand actually since i posted that originally it's not entirel clear that they *are* incompatible
14:14:15Tornethe spec is nontrivial to interpret
14:14:34Tornethe information i had read when i posted that might have been mistaken
14:14:42Torneso i wouldn't worry, just buy whatever
14:14:45Torneit'll proabbly work :p
14:14:45mounichis the rockbox micro sd driver able to handle those?
14:14:55gevaertsAlso, it might be useful to know that our way of finding out if something is compatible is to convince a user to try it :)
14:15:01Torneas long as you fix the MBR partition type and reformat
14:15:22Tornemounich: our driver just does approximately what the simplified sd spec says
14:15:23mounichTorne: that what I wanted to listen thank you :) I'm going to buy one then
14:15:32mounichyou seems to be an expert
14:15:36Tornewhether that works on any given card is disappointingly unpredictable
14:15:39AlexPgevaerts: Well, if a user wants to send me a card I'm happy to try it :)
14:15:39Tornelots of cards suck
14:16:01Tornewodz: so yeah, i think the problem there is that the "retain relocs" option is not meant to be used for this
14:16:03gevaertsAlexP: has anyone sent you this 240GB disk yet? :)
14:16:09AlexPgevaerts: sadly no
14:16:29wodzTorne: but ld -r gives the same result
14:16:31the-kyleMy best reader says it only supports up to 32GB, but maybe that's because of the filesystem.
14:16:38mounichOk, I'll buy one. Fix MBR and reformat the thing
14:16:52Tornethe-kyle: there are, unfortunately, a wide variety of ways in which a given reader can fuck up supporting larger cards
14:16:56Tornenot just the filesystem.
14:17:09Tornethe way the SDHC spec works is kinda weird for large cards :p
14:17:18Tornethough i think usually the limit is 64GB
14:17:26TorneRockbox doesn't have this problem
14:17:37Tornebecause we have direct access to the SD/MMC controller and program it ourselves
14:17:43gevaertsthe-kyle: the reader's firmware might not support SDXH
14:17:49Tornewhereas a USB SD card reader, you are stuck with however the reader behaves
14:17:54mounichRockbox is the solution to all our problems :)
14:18:11Torneso you may well find that large cards work fine in rockbox but not in a SD card reader :p
14:18:20*mounich just need the card on the player, rockbox will be my reader
14:18:41the-kyleMy USB reader is some little cheap thing I got in a pack of 4. It basically turns a MicroSD card into a thumb drive.
14:18:56Tornethe-kyle: yeah, and to do that it implements the actual SD spec
14:19:05Tornebeacuse it has to adapt USB MSC commands to SD commands
14:19:27Torneand so, the reader's firmware has to support SDHC itself, the host can't do it, and if it has arbitrary limits at 32/64gb then it's just tough :p
14:19:38wodzTorne: If retain relocs is not an option, do you have any other solution in a mind?
14:19:51Tornewodz: give up on supporting iram?
14:19:57the-kyleTorne: So my USB reader may not work, but Rockbox should handle it.
14:20:11Tornethe-kyle: i'd say "might" rather than "should"
14:20:25Torneafaik we interpret the SD card info correctly such that the capacity shouldn't matter
14:20:35Tornebut i doubt anyone has tested on large cards :)
14:20:56AlexPThere is one report of 64gb in the forums
14:21:00wodzTorne: well, we could not realocate iram part - I mean feed linker with real iram address
14:21:04AlexPIt worked fine on both clip and clip+
14:21:19AlexPWell, there might be more, I stopped searching there
14:21:20Tornewodz: we could, though that may be problematic for arbitrary IRAM locations
14:21:24the-kyleTorne: TorneIn that case, it would then depend mostly on whether or not Rockbox will allow me to see the partition table on the card to fix it and reformat.
14:21:33Tornein case it overlaps with the region we are using for relocation :p
14:21:52Tornethe-kyle: i think our usb stack just exports the entire block device
14:21:58Torneso even if it can't mount the partitions it should work
14:22:14Torneone thing on my maybe-todo list is "a plugin that fixes partition table for you"
14:22:54wodzTorne: I am a bit surprised that memmap split is so uncommon feature.
14:23:05Tornewodz: not sure what you mean
14:23:11Tornethis is a combination of obscure and weird bullshit we are doing
14:23:18Tornethe bflt tools are super fucking bizare
14:23:25wodzTorne: true
14:23:26Tornedepending on retained static relocs is crazy
14:23:52Torneanyway, yes, i think they misunderstand you somewhat still
14:24:05Torneit's not a "bug" as such but it's a missing feature
14:24:16Torne&minus;&minus;emit-relocs is not &minus;&minus;retain-relocs
14:24:18wodzI remind you the original idea about utilizing bFLT originated from you :-)
14:24:31Torneit should cause all the stuff that needs relocations to have relocations in the output.
14:24:55Tornethe fact that it doesn't for code generated by the linker is dumb
14:24:55mounichTorne, AlexP: thanks to you for your help. I want to buy a 64GB card and test, but I'm scared that the card reader sux on me :(
14:25:19Tornewodz: i didn't actually have any idea how elf2bflt worked, yaknow ;)
14:25:53Tornei just knew how bflt was loaded by uclinux
14:27:08wodzok, gtg
14:27:11 Quit wodz (Quit: Leaving)
14:30:35mounichOk, I spoted a difference bewtwen 2 64gb cards
14:31:48mounichOne is branded "Sandisk Ultra" and the other "Sandisk Mobile Ultra"
14:32:05Tornenot likely to be significant.
14:32:17mounichI guess we can distinguish then that way for the old and the new ones no?
14:32:23 Quit thegeek (Read error: Connection reset by peer)
14:32:26mounicheven the price is different
14:32:33 Join thegeek [0] (
14:32:40Torneit's likely to be the performance class of the card that's different, not the spec version
14:33:02Tornei expect the more expensive one is faster :)
14:33:08mounichthat le Class 10 one that not works, correct?
14:33:13mounichyeah :)
14:33:25Tornethe speed class has nothing to do with whether we can support it or not
14:33:30mounichand the faster is one unlikely to works with rockbox, right?
14:34:05*mounich is trying to check how to distinguish them, and make that not that random
14:34:14TorneThere may not be anything to distinguish
14:34:30Torneand even if there is, then the info is unlikely to be mentioned anywhere in sales info
14:34:43Torneit can probably only be read from the card's CSD info electronically :)
14:34:55mounichcrap :)
14:34:57 Quit Keripo (Quit: Leaving.)
14:35:10mounichI guess i HAVE TO TRY AND PRAISE, /°
14:35:20mounichsorry try and praise
14:35:41Tornejust buy whatever card seems like a reasonable card to you
14:35:51Torneand if it works, send us a copy of the CSD info :p
14:35:58mounichI just want 64gb card on rockbox
14:35:58Tornenot sure if rockbox has a way to dump that but I can add one :)
14:36:07Torneif it doesn't work then we can probably make it work
14:36:15Tornewe just might need to go read more boring spec documents
14:36:21Torneand/or test using the actual card
14:36:27mounichah! the genial answer :)
14:36:27Torne(and look at its CSD) :)
14:36:52Torneanyway yeah. if it works then i'd like to know exactly how the card identifies itself (electronically, not on amazon)
14:36:54mounichDon't rockbox have a found for those kind of things?
14:37:05Torneit does, if someone feels like making a case for it
14:37:09Tornebut i doubt anyone cares very much :p
14:37:15Tornealso it probably does work
14:37:15mounichhehe :)
14:37:23Torneso it's not a good investment until we've demonstrated there's a problem
14:37:40Torneso far i don't think any users ahve had an issue, maybe everything is fine or maybe they were just lucky :)
14:38:14Torneif you get one and it doesn't work, then that's actually kinda interesting and someone will probably help fix it
14:38:14mounichhehe, I'll buy one and report you my experience guys, your are so cool
14:38:24Torneif nobody else, then i'll do it :p
14:38:56mounichOK, sure. If it dont works, I'll surely send it out to you to bisect
14:39:05 Quit einhirn (Quit: Miranda IM! Smaller, Faster, Easier.
14:39:24mounichTime to press the BUY button then :)
14:39:39Tornewell, with any luck we can fix it without you having to mail it; it's probably just a case of looking at what CSD contains and seeing what we are misinterpreting
14:39:59*Torne has had enormous fun fixing dodgy eMMC devices on Linux by reading hexdumps of CSD info
14:42:04mounichJust to be sure, whatever I choose a Class 6 or 10, it will not change my proba to get it workin Torne?
14:42:21mounichlets take the 10 then, tank you
14:42:30Torneclass 10 is well within the existing speed capabilities of sdhc cards, so i doubt that on its own implies anything interesting
14:42:31 Join GermanMushroom [0] (
14:47:48 Join benedikt93 [0] (~benedikt9@unaffiliated/benedikt93)
14:47:52 Quit lorenzo92 (Quit: ChatZilla 0.9.89 [Firefox 15.0.1/20120907231657])
14:49:58AlexPkugel: Is there anything in particular holding up getting your android status bar stuff into git?
14:53:02ZagorAlexP: I have a feeling if rockbox "requires" both stop and pause, we want to look at fixing that
14:53:20AlexPZagor: I'll have to check if that is the case
14:53:28AlexPit used to be, wrt playlist handling for one
14:53:39gevaertsAt least bookmarks can behave differently for pause and stop
14:53:45AlexPBut the main reason I put both pause and stop is I dodn't think about it
14:53:51AlexPgevaerts: yeah
14:54:01Zagorand stopping makes you exit current playlist.
14:54:13Zagorbut these are pretty technical details, far from obvious for casual users
14:54:17AlexPAnother (on android) is whether Rockbox is in the notification bar
14:54:48AlexPIf we can make it so that to the user it doesn't matter then that would certainly be better
14:55:21 Quit n1s (Read error: Operation timed out)
14:56:02AlexPOh, maybe notification bar isn't different
14:59:05AlexPgevaerts: I remember we talked about (e.g. volume) popups before. If I have a volume popup (on volume change) over the AA, then if the AA is drawn in a standard viewport then a gap is left when the popup goes, if the AA is in the backdrop layer then the volume popup works but the AA doesn't update on track change
14:59:08Zagora quick fix might be to simply move stop to "long press on pause"
14:59:26AlexPgevaerts: Do you remember what conclusion we came to?
14:59:44gevaertsAlexP: you need an empty viewport there
14:59:59AlexPZagor: That might be worse though, as the (to a user) random difference between stop and pause is still there, but now the stop is hidden
15:00:05AlexPgevaerts: Under everything?
15:00:23gevaertsUnder everything might well be the quick fix, yes
15:00:39gevaertsI think
15:00:57AlexPA backdrop one? And also under the part of the wps (screen) that is drawn by the sbs?
15:01:25gevaertsThe backdrop is only ever redrawn in places there's a forground VP
15:01:37AlexPah, I get you
15:01:47AlexPSo just stick an empty foreground one over the AA
15:01:56AlexPI'll try, ta
15:05:36ZagorAlexP: I said quick fix. not good fix ;-)
15:17:23AlexPI think I'm running out of skin buffer now
15:17:29 Quit GeekShadow (Ping timeout: 264 seconds)
15:18:07AlexPAs if I use a backdrop viewport then it won't show my backdrop set in Rockbox, I have to load it in the wps using %x correct?
15:18:16AlexPWell, if I do that I get the fallback
15:18:32AlexPAs this is 720*1280, the backdrop bmp in 2.6 MB
15:18:47AlexPSo that twice + all the other bmps is quite a lot
15:19:15 Join GeekShadow [0] (
15:19:19AlexPWould it be a massive amount of work to allow jpg images in the wps/sbs/as backdrops?
15:19:35AlexPEsp. on android I don't think we need to worry about speed
15:25:13 Join n1s [0] (
15:25:13 Quit n1s (Changing host)
15:25:13 Join n1s [0] (~n1s@rockbox/developer/n1s)
15:29:57CIA-10Commit f498142 in rockbox by Nils Wallménius: opus: #if 0 out some unused code
15:30:57n1sthat cut about 25k off of the .codec size for an arm7 build so it might help memory constrained targets work slightly better
15:31:44the-kyleI'll try that on the ClipV1 shortly.
15:33:04 Join amayer_ [0] (
15:33:17 Join WalkGood [0] (~4@unaffiliated/walkgood)
15:36:00 Join wodz [0] (
15:36:47wodzAlexP: jpg/png for backdrops will only decrease 'distribution' size. You still need to have bitmap expanded in mem to display it actually
15:36:59AlexPwodz: good point
15:37:47wodzWe agreed that Mr Someone could add png support for backdrops anyway
15:38:51 Join einhirn [0] (
15:43:54amiconnZagor: Ehumm. The files are there, but don't work.
15:44:00*amiconn wonders what changed
15:44:16amayer_wodz: if we add support for png as backdrop would it be a hard strech to add support for png album art?
15:44:43 Join Rower85 [0] (
15:47:42wodzamayer_: the whole thing is pretty trivial. The decoder is (almost) ready. Look at imageviewer. The problem is color conversion for b&w targets AND variety of allowed combinations of png features. Incore decoder should use a subset probably.
15:49:44***Saving seen data "./dancer.seen"
15:50:50 Quit mortalis (Quit: Leaving)
15:54:23 Quit wodz (Quit: Leaving)
15:58:06the-kylen1s: Latest Opus commit still crashes the clip v1 and causes a reboot. It looks like it either still has a memory issue or is having a different kind of execution problem. I'll keep trying to test it on this device.
15:59:10 Quit mounich (Quit: CGI:IRC)
16:27:32 Join eckoit [0] (~ryan@
16:30:20AlexPgevaerts: I "fixed" it by moving the popup :)
16:50:51 Join liar [0] (
17:01:08 Quit liar (Remote host closed the connection)
17:01:35 Join Holyjack [0] (595d59bc@gateway/web/freenode/ip.
17:03:24 Join Holyjack_ [0] (595d59bc@gateway/web/freenode/ip.
17:03:47Holyjack_Hello all
17:04:07Holyjack_I'm requestion help because I think I have bricked my Sansa Clip+
17:04:22Holyjack_*requesting, sorry
17:05:49 Quit Holyjack (Ping timeout: 245 seconds)
17:05:59Holyjack_Updating the database was taking forever (more than 20 minutes, about 10gb of files). So I turned off the player. Now it won't turn on again, even after pressing the power button for 3 minutes
17:07:13gevaertsWhat version of rockbox were you running?
17:08:06gevaertsThere's a known bug that's fixed in the latest dev builds that causes this sort of issue for some people. Normally waiting for the battery to run down helps, so if this is the problem it should be fine again in a few days
17:10:17 Quit advcomp2019 (Quit: I was raided by the FBI and all I got to keep was this lousy quit message!)
17:10:39 Join Holyjack [0] (595d59bc@gateway/web/freenode/ip.
17:10:44 Join advcomp2019 [0] (
17:10:45 Quit advcomp2019 (Changing host)
17:10:45 Join advcomp2019 [0] (~advcomp20@unaffiliated/advcomp2019)
17:10:49 Quit Holyjack_ (Ping timeout: 245 seconds)
17:11:04HolyjackHello all
17:11:28AlexPHolyjack: Did you see the previous reply?
17:12:09Holyjackno, sorry, I was using a Chrome extension that closes the session when I click outside the window
17:12:20Holyjacknow I'm on freenode webIRC, works much better :)
17:12:22AlexPCheck <a target="_blank" href="">>
17:12:30HolyjackI just saw the reply on the log
17:12:32Holyjackyes, there ;)
17:12:40 Quit Zagor (Quit: Clint excited)
17:13:06HolyjackI can't day for sure which version of RB is was
17:13:19Holyjackbut I downloaded it about a week ago
17:13:53gevaertsDid you choose the stable version?
17:14:26HolyjackI used the installer utility
17:14:30AlexPIt had the bug then
17:14:36gevaertsThen that's 3.11.2, which does have this issue
17:14:48gevaertsSo the advice is: wait a few days
17:15:03Holyjackthe battery will fade out even if the player is idle ?
17:15:53Holyjackand when the battery is empty, what shall I do ?
17:15:56gevaertsWell, it only *looks* idle
17:16:12gevaertsPlug it in every now and then. At some point it will start up again
17:16:29Holyjackso the screen is black but inside it's still working ?
17:16:37Holyjackok :)
17:16:41AlexPThen install a dev build, or 3.12 if we have released it yet
17:17:13 Quit WalkGood ()
17:17:20Holyjackthanks guys
17:17:34HolyjackSo what exactly is the pb on this build ? The DB generation ?
17:17:47Holyjackor should I have been more patient with the player ?
17:19:22gevaertsThe database can be a bit buggy every now and then, but the main issue is that with those builds it's possible for hard reset to stop working.
17:21:23HolyjackA 32-gb SD card should be problem ?
17:21:42Holyjackok thanks very much guys
17:22:04HolyjackI'll wait until the end of this week to see if the player is really bricked
17:22:13AlexPIt isn't :)
17:22:35Holyjackok :)
17:23:11Holyjackfor the first time since I've had this player I'm actually wishing it had a *shorter* battery life :)
17:23:40AlexPyeah, it is annoying
17:23:46AlexPBut it should come back to life fine
17:24:00HolyjackI'll keep my fingers crossed :)
17:24:57HolyjackBtw, maybe it should be mentionned somewhere oin the Sansa page ?
17:25:02Holyjackthis known bug, I mean
17:25:47AlexPIt'll be gone soon
17:29:10Holyjackthanks again
17:29:14Holyjackbye ! :)
17:29:19 Quit Holyjack (Quit: Page closed)
17:49:45***Saving seen data "./dancer.seen"
17:49:48CIA-10Commit 06fc6fd in rockbox by Nils Wallménius: opus: #if 0 out some more unused code, delete unused file
17:54:42n1sthe-kyle: can you compile your own test build?
17:54:55the-kylen1s: Yes.
17:55:32the-kyleMy local copy is already mod'd.
17:56:00n1sthe-kyle: changing the optimization flag for opus to -Os cuts another 10k off the binary for the clipv1 if you'd like to try that, not sure how much it would help though
17:56:26the-kyleI'll try that and see what happens.
17:57:09n1sjust change the "-O2" to "-Os2" in the line $(OPUSLIB) : CODECFLAGS += -O2 in codecs.make and make clean
17:57:24n1seh -Os not -Os2
17:57:52the-kylen1s: Thanks. Will try.
17:58:18the-kyleI'm rebuilding my toolchain.
17:58:48 Join y4n [0] (~y4n@unaffiliated/y4ndexx)
18:19:11 Quit shai_ (Read error: Connection reset by peer)
18:24:22 Join shai [0] (
18:26:28 Join Wardo [0] (
18:38:11the-kylen1s: Well, the result is indeed different. I no longer get a reboot, but I now get a hard freeze.
18:38:59n1sthe-kyle: if it's a memory corruption bug which i think is fairly likely the results might well differ
18:39:43the-kyleBeing visually impaired, I can't see what shows up on the screen when the freeze occurs, but I've taken a picture of the device as it froze, and I'll upload it somewhere. I hope it's readable.
18:39:50n1scutting back binary size some more by disabling more encoder code in celt is possible but i guess i should figure out just how much it needs
18:45:23the-kylen1s: <a target="_blank" href="">>
18:45:36 Quit einhirn (Quit: Miranda IM! Smaller, Faster, Easier.
18:45:43the-kyleHope the camera took a readable picture.
18:46:04n1sthe-kyle: looks like it's just frozen on the wps
18:46:19the-kyleCould be.
18:47:10the-kyleIs there a way to get a core dump to a file in cases like this?
18:49:19n1snot really
18:50:21 Join pedro_angelo [0] (
18:52:23the-kyleInteresting question: In case of memory or optimization bugs in a codec, which is better/more telling of the problem, a hard freeze or a crash and reboot?
18:54:50n1ssome crashes generate messages that can be helpful but just a hard freeze or a straight reboot doesn't tell much more than that something went very wrong
18:59:24 Join bertrik [0] (~quassel@rockbox/developer/bertrik)
19:00:51 Quit stoffel (Ping timeout: 246 seconds)
19:05:21 Quit mc2739 (Ping timeout: 256 seconds)
19:06:19n1sit seems the codec needs about 320 to 380kb at the moment (tested lowes mid and highest bitrate in our test files) while the clip only has 288kb so some fairly large improvements are needed
19:07:01n1si have no idea how much of the global stack buffer is used at either
19:07:03 Join mc2739 [0] (~mc2739@rockbox/developer/mc2739)
19:07:10n1sthat takes up a whooping 100k
19:07:45the-kyleWooh! that seems big on such a low-memory device.
19:08:43the-kyleThe codec is still fairly new, so hopefully these improvements will eventually come.
19:10:57 Join einhirn [0] (
19:13:53 Quit factor (Read error: Connection reset by peer)
19:15:30 Join factor [0] (
19:17:18n1sthe-kyle: if you want to play with it reducing GLOBAL_STACK_SIZE down to 20000 seems to work with the test files i have, that should give it enough room to work with small to mid bitrate files, assuming it's fast enough :)
19:17:35 Quit the-kyle (Read error: Connection reset by peer)
19:18:30 Join the-kyle [0] (
19:20:02n1shmm, or maybe not, an overrun in that buffer would likely go undetected unless it overran the whole buffer limit which it will not do on the sim easily
19:20:22n1sbertrik: any idea how much stack opus really uses?
19:20:49bertrikI think you can calculate from % used * stack size
19:21:25n1ssorry, i meant of the global stack thing
19:21:57bertrikno, but I think I left some code in the sim that logs every allocation of the global stack
19:22:58bertrikI think it was at least 64k just for caching one ogg max size page
19:23:40n1safaics the ogg stuff is separate, using codec_malloc
19:24:51bertrikah yes
19:25:00 Part gxk
19:40:48 Quit mgottschlag (Ping timeout: 246 seconds)
19:41:52n1sif my log hack is correct the 64 kbps test file needs at most ~25k
19:44:22n1sbtw, afaict some fairly intrusive changes are needed to use iram in any kind of efficient way
19:44:39n1sthe scheme used for allocating the state structs is just too weird
19:45:32 Quit efyx (Read error: Connection reset by peer)
19:49:46***Saving seen data "./dancer.seen"
19:51:27 Quit benedikt93 (Quit: Bye ;))
19:53:47 Join mortalis [0] (~mortalis@
20:01:21 Join efyx [0] (~efyx@
20:14:37 Quit pedro_angelo (Remote host closed the connection)
20:23:08 Quit einhirn (Quit: Miranda IM! Smaller, Faster, Easier.
20:34:09 Quit eckoit (Quit: eckoit)
20:36:56 Join kugel_ [0] (~kugel@
20:36:57 Quit kugel_ (Changing host)
20:36:57 Join kugel_ [0] (~kugel@rockbox/developer/kugel)
20:37:22 Join ZIulGqvA [0] (
20:42:46 Quit XavierGr (Disconnected by services)
20:42:47 Join XavierGr [0] (XavierGr@rockbox/staff/XavierGr)
20:43:04 Join Elfish_ [0] (amba@2a01:4f8:100:90a1:abc:abc:abc:abc)
20:47:22 Quit kugel (Read error: Connection reset by peer)
20:47:22 Quit LinusN (Ping timeout: 245 seconds)
20:47:22 Quit Elfish (Ping timeout: 245 seconds)
20:47:22 Join Elfish [0] (amba@2a01:4f8:100:90a1:abc:abc:abc:abc)
20:47:23 Quit Elfish (*.net *.split)
20:47:23 Quit mc2739 (*.net *.split)
20:47:24 Quit bluebrother (*.net *.split)
20:47:24 Quit Staphylo (*.net *.split)
20:47:25 Quit Poodlemastah (*.net *.split)
20:47:25 Quit freqmod (*.net *.split)
20:47:25 Quit simabeis_ (*.net *.split)
20:48:29 Join mc2739 [0] (~mc2739@rockbox/developer/mc2739)
20:48:29 Join bluebrother [0] (~dom@rockbox/developer/bluebrother)
20:48:29 Join Staphylo [0] (
20:48:29 Join Poodlemastah [0] (
20:48:29 Join freqmod [0] (~quassel@2001:700:300:2120:5652:ff:fe70:1e36)
20:48:29 Join simabeis_ [0] (
20:53:54 Join kugel [0] (~kugel@
20:53:55 Quit kugel (Changing host)
20:53:55 Join kugel [0] (~kugel@rockbox/developer/kugel)
20:56:30 Join shai_ [0] (
20:56:37 Quit kugel_ (Read error: Connection reset by peer)
20:56:37 Quit the-kyle (Ping timeout: 246 seconds)
20:56:39 Quit shai (Ping timeout: 246 seconds)
20:56:39 Quit n1s (Ping timeout: 246 seconds)
20:56:39 Quit mystica555 (Ping timeout: 246 seconds)
20:56:40 Quit nsx (Ping timeout: 246 seconds)
20:56:40 Quit tchan (Ping timeout: 246 seconds)
20:56:42 Quit fs-bluebot (Ping timeout: 246 seconds)
20:56:45 Join the-kyle [0] (
20:57:03 Quit Guinness (Ping timeout: 246 seconds)
20:57:14 Join tchan [0] (~tchan@lunar-linux/developer/tchan)
20:57:21 Join mystica555 [0] (
20:57:21 Quit mystica555 (Excess Flood)
20:57:24 Quit amayer_ (Ping timeout: 246 seconds)
20:57:40 Join amayer_ [0] (
20:57:41 Join mystica555 [0] (
20:57:41 Quit mystica555 (Excess Flood)
20:58:01 Join mystica555 [0] (
20:58:01 Quit mystica555 (Client Quit)
20:58:14 Join mystica555 [0] (
20:58:16 Quit mystica555 (Excess Flood)
20:58:48 Quit Wardo (Ping timeout: 246 seconds)
21:01:51 Join Guinness [0] (
21:02:29 Quit saratoga (Ping timeout: 245 seconds)
21:02:41 Join wodz [0] (
21:03:14 Join fs-bluebot [0] (
21:04:41 Join nsx [0] (
21:04:42 Quit nsx (Changing host)
21:04:42 Join nsx [0] (jm@unaffiliated/nsx)
21:04:51 Join sciopath [0] (
21:05:03bluebrothershouldn't <a target="_blank" href="">FS #12755a> be rejected upfront? It's a rather useless report IMO
21:05:03fs-bluebot<a target="_blank" href="">> Sansa Clip Zip has too many problems with lossless files (bugs, unconfirmed)
21:06:11 Quit y4n (Quit: Today is the perfect day for a perfect day.)
21:06:20bluebrotherplus, I don't like this "the testing of builds were made too hastily" attitude. After all the Clip Zip isn't considered stable
21:06:59*gevaerts nods
21:07:09gevaertsI'd say close it as a support issue
21:07:30bertrikI'd say leave it open if he can provide example files
21:07:35gevaerts(combined with the zip not being stable)
21:07:38 Join thegeek_ [0] (
21:07:47gevaertsbertrik: for which of the four issues?
21:07:58gevaertsAt least for ape it's expected that some won't play
21:09:49bluebrotherwell, if he wants to report issues for $somefileformat he should file a report for $somefileformat, not for $bunchoffileformats
21:10:05*gevaerts nods
21:10:47bertrikbut what if he then reports four separate issues? will they stay open?
21:10:59bluebrotherwell, if they are useful reports ...
21:11:24bluebrotherbtw, we currently have >600 open tasks. We should really try to get that number down.
21:11:43bluebrother320 of those are bugs.
21:12:32bluebrotherthe oldest from 2006 :o
21:13:14 Join a [0] (
21:13:26 Quit GermanMushroom (Ping timeout: 274 seconds)
21:13:26 Quit ender| (Ping timeout: 274 seconds)
21:13:27 Quit Belzebub (Ping timeout: 274 seconds)
21:13:28 Quit thegeek (Write error: Connection reset by peer)
21:13:35 Nick a is now known as Guest64089 (
21:17:05 Join Belzebub [0] (
21:23:41 Join eckoit [0] (
21:24:23 Quit eckoit (Client Quit)
21:24:38 Join ender| [0] (
21:25:24mortaliswodz: you was right about for loop
21:25:34 Quit mortalis (Quit: Changing server...)
21:26:03 Join mortalis [0] (~mortalis@
21:26:51 Join Strife89 [0] (~Strife89@
21:27:35 Join gevaerts_ [0] (
21:29:06 Quit gevaerts (Ping timeout: 240 seconds)
21:29:06 Quit uwe_ (Ping timeout: 240 seconds)
21:29:16 Nick gevaerts_ is now known as gevaerts (
21:30:23 Join Syconaut^ [0] (
21:30:37 Quit gevaerts (Changing host)
21:30:37 Join gevaerts [0] (~fg@rockbox/developer/gevaerts)
21:31:08 Join uwe_ [0] (
21:31:36 Quit Syconaut (Ping timeout: 240 seconds)
21:42:26mortaliswodz: seems now i have working lcd dma updates. I faced only one problem: after test_fps i have little distortion in left side of a screen.
21:42:49 Join n1s [0] (
21:42:57mortalishm, distortion always appears after Main lcd yuv
21:43:21wodzmortalis: cool - yuv conversion is not implemented anyway
21:43:58wodzn1s: I am uploading qemu-arm image. Archive is ~850MB large so this will take a bit
21:44:13n1swodz: cool
21:46:11 Quit n1s (Changing host)
21:46:11 Join n1s [0] (~n1s@rockbox/developer/n1s)
21:46:25*amiconn found the issue with his android build capability
21:46:58amiconnIn the last dependency conflict, I somehow managed to remove some essential i386 multiarch libs
21:47:22amiconnIt's working again, so how do I tell the buildmaster?
21:47:31mortaliswodz: <a target="_blank" href="">> it steel needs some work & clean up.
21:47:59*amiconn summons Zagor
21:49:47***Saving seen data "./dancer.seen"
21:52:21 Quit lebellium (Read error: Connection reset by peer)
21:52:47 Join lebellium [0] (
21:55:16kugelrockbox static "ram" should be increased for high-res displays (or even be dynamic)
21:55:28 Quit Belzebub (Changing host)
21:55:28 Join Belzebub [0] (~torrentow@unaffiliated/blown-engine)
21:56:11 Join eckoit [0] (~ryan@
22:02:35kugelAlexP: ping
22:02:51kugelworking on some nexus 7 theme?
22:02:56AlexPkugel: s3
22:03:06AlexPbut same res IIRC
22:03:23kugelcabbie or something else?
22:03:43AlexPkugel: Something else, I put a thread in the forum with a screenshot :)
22:03:43kugelI wanted to have a go tomorrow, since I currently have a nexus 7 for a while
22:03:55AlexPThat'd be cool
22:04:00AlexPI think I found 2 bugs
22:04:11AlexPThe custom scroll bar %LB doesn't seem to work
22:04:28evilnickAlexP: I'd offer to test it but my s3 is about to be sent off for fixing :(
22:04:38AlexPAnd the "browse" touchscreen tag doesn't seem to work when you are in the menus, but is fine in the wps
22:04:51AlexPevilnick: oh, shame. Your fault?
22:05:12kugelnot sure I can help with skinned lists
22:05:41AlexPI told JdGordon, he is/was going to try to check if it is me doing something wrong before I file a bug report
22:06:02kugelhe's gone for 2 weeks now afaik
22:06:15AlexPoh, shame
22:07:13evilnickkugel: Is bluetooth streaming something that needs to be enabled in the (android) app?
22:09:37bluebrotherAlexP: Nexus 7 is 800x1280
22:09:52AlexPOK, so no, not for that then :)
22:12:16 Join hype_ [0] (~hype@
22:12:31bluebrotherwell, shouldn't be too hard to adjust for that resolution, given that it only adds a couple of pixels at the sides :)
22:13:08bluebrotherjudging from your screenshots one could simply center it and make the backdrop a bit wider
22:13:26AlexPkugel: What's the status on your stuff for android notification bar?
22:13:43AlexPbluebrother: yeah, would have thought so - just adjust the spacing a bit
22:14:27 Quit hype (Ping timeout: 244 seconds)
22:14:27 Nick hype_ is now known as hype (~hype@
22:25:11 Quit scorche (Disconnected by services)
22:25:15 Join scorche` [0] (~scorche@rockbox/administrator/scorche)
22:27:00kugelAlexP: no progress, I didnt work further on it the past weeks (lack of time and motivation)
22:27:20AlexPkugel: What is the issue? It seemed to work well the time I tried
22:28:44kugelI think I want a malloc-based buflib for it as stopping playback for resolution changes sucks
22:29:01kugel(and there's an implicit change on start up)
22:30:16kugelan extension to the skin language the center an image within the viewport would be nice also
22:30:54kugelfor better support of multi-resolution themes
22:31:00AlexPyes, that would be nice
22:31:19AlexPresolution dependent size scaling :)
22:31:33AlexPso image is x/4, y/8 or whatever :)
22:31:37kugelas of now you couldnt center the an image or album art on a theme which works on both nexus 7 and sgs3
22:32:32evilnickThat would be awesome
22:33:01evilnickWell... anything that meant that there didn't have to be resolution specific changes would be awesome :)
22:35:24 Quit ZIulGqvA (Remote host closed the connection)
22:35:38 Join LinusN [0] (
22:36:02 Part LinusN
22:41:27 Join Horscht [0] (
22:41:28 Quit Horscht (Changing host)
22:41:28 Join Horscht [0] (~Horscht@xbmc/user/horscht)
22:45:05wodzn1s: File sharing sites doesn't like this archive. Maybe I can send it to You directly with dcc?
22:49:47 Quit wodz (Quit: Leaving)
22:53:04 Quit bertrik (Remote host closed the connection)
22:54:30 Join lebellium_ [0] (
22:56:57 Quit lebellium (Ping timeout: 268 seconds)
22:57:10 Nick lebellium_ is now known as lebellium (
22:58:33n1swodz: (logs) i've never gotten dcc to work
22:59:27n1si'll try to set it up by following the instruction you linked
22:59:58 Quit JdGordon (*.net *.split)
22:59:59 Quit ukleinek (*.net *.split)
23:00:01 Quit LittleCreature (Ping timeout: 279 seconds)
23:01:02 Quit kevku (Quit: KVIrc 4.2.0 Equilibrium
23:01:08 Join LittleCreature [0] (~fearofmus@unaffiliated/fearofmusic)
23:04:02 Join JdGordon [0] (~jonno@rockbox/developer/JdGordon)
23:04:02 Join ukleinek [0] (
23:09:08 Quit thegeek_ (Ping timeout: 256 seconds)
23:09:16 Join thegeek [0] (
23:09:37 Nick Elfish_ is now known as Elfish (amba@2a01:4f8:100:90a1:abc:abc:abc:abc)
23:17:19 Quit preglow (Ping timeout: 240 seconds)
23:19:49 Quit amayer_ (Ping timeout: 256 seconds)
23:20:13minusis it normal that changing/pausing a song takes ~7 seconds with a 32GB SD card? (sansa fuze v1)
23:20:47 Join preglow [0] (
23:20:47 Quit preglow (Changing host)
23:20:47 Join preglow [0] (thomj@rockbox/developer/preglow)
23:20:57 Quit Strife89 (Ping timeout: 246 seconds)
23:25:07 Quit mortalis (Ping timeout: 248 seconds)
23:25:24 Join mortalis [0] (~mortalis@
23:28:37 Quit mortalis (Client Quit)
23:29:47 Join mgottschlag [0] (~quassel@reactos/tester/phoenix64)
23:33:22 Quit [Saint] (Read error: Connection reset by peer)
23:33:28 Join pedro_angelo [0] (
23:33:38 Join [Saint] [0] (~saint@rockbox/user/saint)
23:33:44 Quit Horscht (Quit: Verlassend)
23:46:15 Quit mc2739 (Ping timeout: 245 seconds)
23:47:49 Join GeekShad1w [0] (
23:48:22 Join mc2739 [0] (~mc2739@
23:49:12 Join amayer [0] (
23:49:51***Saving seen data "./dancer.seen"
23:50:31 Quit GeekShadow (Ping timeout: 248 seconds)
23:50:32 Quit funman (Ping timeout: 248 seconds)
23:51:36 Quit mc2739 (Changing host)
23:51:36 Join mc2739 [0] (~mc2739@rockbox/developer/mc2739)
23:51:36 Join mgottschlag2 [0] (
23:53:13 Join funman [0] (
23:53:56 Quit mgottschlag (Remote host closed the connection)
23:54:37 Quit funman (Changing host)
23:54:37 Join funman [0] (~fun@rockbox/developer/funman)
23:58:37 Quit pedro_angelo (Remote host closed the connection)

Previous day | Next day