Previous day | Jump to hour: 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 | Next day

Seconds: Show Hide | Joins: Show Hide | View raw
Font: Serif Sans-Serif Monospace | Size: Small Medium Large

Click in the nick column to highlight everything a person has said.
The Logo icon identifies that the person is a core developer (has commit access).

Notice: Only Gecko based browsers prior to FF4 support the multipart/mixed "server push" method used by this log reader to auto-update. Since you do not appear to use such a browser, this page will simply show the current log, and not automatically update.

#rockbox log for 2010-05-28

00:00:02 Quit bertrik (Quit: De groeten)
00:01:05webguest88hmmm i am trying
00:01:24***Saving seen data "./dancer.seen"
00:01:59*jhMikeS hasn't even replaced the bootloader in his H120 since he first got it in '06 or '07 ... still works fine :)
00:02:49webguest88i think i am into manu mode
00:03:36 Quit [CGL] (Remote host closed the connection)
00:03:37webguest88thanks everyone
00:03:40 Join Johnny_ [0] (
00:04:00webguest88thanks gevaerts
00:04:09webguest88thanks bluebrother
00:04:51webguest88if any1 here is the mod of the site
00:05:14bluebrothermod of the site?
00:05:36bluebrotherwhat's that?
00:05:49*bluebrother isn't a modification of a website :P
00:06:18CIA-5New commit by 03bluebrother (r26339): Fix a warning.
00:06:26jhMikeSbluebrother: thought your were: website % bluebrother;
00:06:28Johnny_Is there any other way, to check how much mAh my battery got, than unscrew the player iRiver h340 + Rockbox 3.5.1 ?
00:07:23webguest88oh by the way
00:07:33LloreanJohnny_: Why do you need to find out?
00:07:36bluebrotherJohnny_: no, as long as you want to get the sold capacity
00:07:37webguest88how do i backup my players flash files ?
00:07:48bluebrotherwebguest88: by copying them :)
00:08:13bluebrotherdrag and drop from explorer if you like, use a script or fancy gui tool, ...
00:08:16Johnny_I bought a used one i want to know because the seller told me it it 2200 mAh
00:08:45webguest88i can do that?..directly from the masss storage mode ?
00:09:14bluebrotherI don't see or know of any reason why you shouldn't be able to transfer files from / to UMS devices.
00:10:02bluebrotherRockbox itself exposes the player as UMS device and puts no restrictions on that. Therefore you should be able to do any modifications to the drive you like to.
00:10:15Johnny_so the only way is to see battery itself right ?
00:10:25ssorgatemPlease, can someone kindly commit my catalan language update? :)
00:10:31bluebrother(e200 installation f.e. works like this: it modified some parts of the drive that aren't shown by explorer)
00:12:35bluebrotherJohnny_: I wouldn't want to check the capacity at all. The player runs as long as it runs. If you have a, like 1600mAh battery that's completely new you're better than with a 2200mAh battery that's been used frequently for 2 years
00:12:35bluebrotherso the printing on the battery really doesn't tell much. Unless you just bought the battery as new.
00:13:00pamauryWhat should I do to allocate statically a buffer of which address I must provide to the dma ? Is there a special attribute on the as3525(v2), I see that USB_BSS_ATTR is not defined
00:13:04pamauryfunman: ^
00:13:16Johnny_i c anyway my 100 % battery shows 16 hrs 15 min on 3.5.1 rockbox thats rather ok i think
00:13:18funmanpamaury: physical buffer?
00:13:56pamauryDon't know what you mean but yes, I probably needs it's physical address (although I can't tell by just looking at the code)
00:14:00webguest88hey!..nees some help again.. how do i get off the manu mode
00:14:25funmanpamaury: DMA needs physical addresses because they write to the memory without using the MMU (which use virtual addresses)
00:14:39pamauryI saw a discussion about it lately but I can't remember when it took place
00:14:41webguest88got it
00:14:51bluebrotherJohnny_: the runtime estimation is a runtime estimation. It doesn't need to match with the actual runtime in any way
00:15:07funmanin the as3525 the mmu is set up as 'flat mapped' for DRAM : virtual addresses are the same than physical addresses
00:15:09pamauryfunman: so my question is: is there a function to compute physical address from virtual one or is there an attribute to ensure both addresses are the same
00:15:14pamauryah ok
00:15:28funmanif you use IRAM you need to tweak things so don't use iram :)
00:15:29pamauryand what about caching ?
00:15:31bluebrotherin fact, the runtime estimation is likely to be off if your battery was replaced with a different one or is weared out to some extend due to usage.
00:15:32Johnny_all right i ll ran battery bench than thank You
00:16:07 Quit Zagor_ (Quit: Clint excited)
00:16:31 Quit Johnny_ (Quit: ChatZilla 0.9.86 [Firefox 3.6.3/20100401080539])
00:16:38funmanpamaury: since the DMA don't use the CPU caches you must make sure no part of this buffer are in cache
00:16:51pamauryand how can I do that ?
00:18:08funmanuse *_dcache_range() from mmu-arm.S/h (look in pcm-as3525.c) or an temporary uncached buffer to read/copy from
00:18:59pamauryok thanks, I'm not sure I'll manage to make it work right now but at least, this problem will be solved
00:20:01funmanif you want to be sure use clean_dcache() before & after transfers
00:21:00pamauryWell, I'm trying to implement the usb reset part, so I'm quite far from transfering data but I need to setup things :)
00:24:05 Join robin0800 [0] (~quassel@
00:25:11 Join vaguerant [0] (
00:25:11 Quit vaguerant (Changing host)
00:25:11 Join vaguerant [0] (~vague@wikipedia/vague-rant)
00:29:41 Quit vaguerant (Ping timeout: 252 seconds)
00:30:10 Quit efyx (Remote host closed the connection)
00:31:48webguest88how do i backup my sansa players firmware ? or any flash files ? in case if anthing goes wrong when i flash rockbox on it
00:32:50Kohlrabiyou can download the original firmware from sansa
00:34:27bluebrotheryou can't easily backup the firmware.
00:34:58bluebrotherif anything goes wrong you can reflash the original firmware. You'll find that at the sandisk website.
00:35:18webguest88oh ok
00:35:33webguest88i dont have to backup anything
00:35:44webguest88coz i had to do that with my psp
00:35:50pamauryfunman: what is a prefetch abort error ?
00:36:19Tornepamaury: means it took an abort while fetching the actual instruction
00:36:33Tornepamaury: i.e. the cpu has jumped to an invalid address
00:36:52pamauryhuh, how the hell did I get that error
00:36:54bluebrotheranything wrong with FS or is it just acting up for me?
00:37:09funmanpamaury: which address?
00:37:26pamaurysounds strange
00:37:37pamaurynearly deadbeef, isn't something particular with this
00:37:41bluebrothersounds like 0xdeadbeef rounded down to a multiple of 2
00:38:14funmanah right
00:38:18funmanbit 0 is thumb bit
00:38:35funmanbut afaik we don't use thumb at all, so shouldn't all fetches be aligned on 4 ? Torne ?
00:39:03funmanpamaury: did you allocate a large buffer on stack ?
00:39:23funmana medium buffer ?
00:39:28pamauryno :)
00:39:34funmana very little tiny buffer ?
00:39:47pamaurya little one for a string, for debug purpose
00:40:01pamauryif I remove a piece of code, I get a data abort
00:40:04Tornethe bottom bit is never 1 in the actual program counter
00:40:10funmanrockbox writes 0xdeadbeef in the stack for debug purpose
00:40:13Torneit sets the T bit in CPSR :)
00:40:27Torneso yeah, you will get deadbeee
00:40:29*bluebrother grumbles on x86 vs x86_64 issues on mac
00:41:02pamauryperhaps a stackoverflow but that seems strange
00:41:11pamauryit's in an interrupt handler
00:41:27funmanthe irq stack is small
00:41:39pamauryit worked one hour ago
00:42:28pamaurycould there be nested usb interrupts ? What would happen ?
00:42:40funmanafaiu when you enter the irq handler interrupts are disabled so no
00:43:01funmanadd panicf() step by step until you don't get your panic
00:43:07jhMikeSpamaury, ARM should disable IRQ before vectoring
00:44:00pamaurygrrr, another data abort
00:45:19jhMikeSI don't seem to remember IRQ using more than a small portion of the 1k
00:45:25 Join DerPapst [0] (
00:46:03pamauryFUCK, now I get an undefined instruction oO
00:47:33pamauryahhhh, back to the normal situation (freeze :))
00:49:19pamauryno undef instruction at 000009D4
00:49:35funmanpamaury: hum that sounds like irq stack
00:50:00jhMikeSmunge IRQ stack to see?
00:50:18funmanstack goes from 0x5A0 to 0x9A0
00:50:49funmanshould be 0x5A0 -> 0x1A0
00:51:05funmanso it's in the unused FIQ stack
00:51:09funmanpamaury: paste your diff
00:52:26*jhMikeS suggests trying not to execute the stack :) (helpful?)
00:52:28pamaurythe diff is enormous, I'm trying to get back to the point where it didn't crash, to find out the evil instruction
00:52:52 Quit webguest88 (Quit: CGI:IRC (EOF))
00:53:48pamauryok, back to old situation: panic ! Now, I'll try to add code
00:53:51 Quit MethoS- (Remote host closed the connection)
00:58:34 Quit hebz0rl (Quit: Ex-Chat)
01:00:07 Quit DataGhost (Ping timeout: 240 seconds)
01:00:55pamaurythere definitetely is something wrong. Something wrong happen when I setup usb interrupt mask.
01:02:18 Quit pamaury (Quit: exit(rand());)
01:02:26 Join pamaury [0] (~pamaury@rockbox/developer/pamaury)
01:02:38 Quit DerPapst (Quit: Leaving.)
01:03:10 Quit robin0800 (Ping timeout: 265 seconds)
01:06:39 Join DerPapst [0] (
01:12:12pamauryOk, I give up for tonight, there is something weird with interrupts, I'm pretty sure of this, otherwise it wouldn't crash like this
01:13:59 Quit Zarggg_ (Ping timeout: 265 seconds)
01:14:42CIA-5New commit by 03jethead71 (r26340): Gigabeat S PCM: There's no reason to touch any hardware registers in order to lock out PCM callbacks.
01:22:58 Quit DerPapst (Quit: Leaving.)
01:22:58CIA-5New commit by 03jethead71 (r26341): Gigabeat S PCM: Sync some comments to changes.
01:25:33 Quit ranma (Ping timeout: 248 seconds)
01:25:46 Join ranma [0] (
01:28:59ssorgatemI updated the catalan translation; ; if some would be kind enought to commit it I would appreciate it very much :)
01:31:29funmanssorgatem: sure
01:32:25jhMikeSdoes the init attr thing magically sort of take care of itself? (just add INIT_ATTR) ??
01:32:32CIA-5New commit by 03funman (r26342): sd-as3525v2: Include time spent yielding when figuring timeout in sd_wait_for_tran_state() ...
01:32:35CIA-5New commit by 03funman (r26343): fuzev2: lcd_enable and lcd_invert() work fine
01:32:40funmanjhMikeS: crt0/ need the magic
01:32:40CIA-5New commit by 03funman (r26344): Merge common LCD code between fuzev1 & fuzev2
01:32:47CIA-5New commit by 03funman (r26345): Catalan language update by Adrián Cereto Massagué
01:32:50*jhMikeS didn't want to do any real work :\
01:34:19funmancreate a .init section in plugin or codec buffer, copy them to the buffer from crt0.S
01:36:07funmanIIUC the copy is at the same location than the stack (before copy), but i'm not sure: the LDS are a bit scary
01:36:16jhMikeSI have a custom .sdmacode section for a blob, which would be nice to get rid of and use a common mechanism, and there's a good deal of stuff that could go in there. (probably an easy convert actually)
01:38:13funmannot sure what the red means
01:38:17 Quit jgarvey (Quit: Leaving)
01:38:24funmani see genversion being run (twice, ok) several lines above
01:38:48funmanroolku uses a special OS ?
01:39:19peturall reds come from systems of roolku
01:39:56funmanperhaps the cat > foo << EOF \n\n\nEOF doesn't work on some systems?
01:41:17funmanwasn't there some page for the build system?
01:42:56jhMikeSfunman: looks like I just convert the one section to .init in the code and crt0.S and I'm good to go
01:43:14funmanit's exactly like iram in fact
01:43:42funmanjust that you put the section somewhere that'll be reused
01:44:22funmanhm FS woes again :/
01:45:03funmani don't want to lose this: almost-working-lcd-invert-on-fuze-but-i-dunno-why
01:45:16jhMikeSfunman: I've done that. Just wondering if anything was different. (sdma used the plugin buffer)
01:45:48funmanhm ok
01:47:19funman~2kB reclaimed on as3525v2 : not much
01:47:40 Part toffe82
01:48:20 Quit petur (Quit: Leaving)
01:49:24 Quit archivator (Quit: Leaving)
01:50:30 Join Boldfilter [0] (
01:52:04 Quit funman (Quit: free(random());)
01:58:39 Join Zarggg [0] (~zarggg@2001:0:4137:9e76:0:fbf3:beb1:ba3d)
02:00:22 Join robin0800 [0] (
02:01:26***Saving seen data "./dancer.seen" is broken for CygWin apparently.
02:10:42S_a_i_n_ttried to run it three times with the same error.
02:11:18S_a_i_n_t <−− If anyone knows what this means...go nuts.
02:13:49S_a_i_n_tAnd if anyone can .tar (etc.) up a working ARM toolchain for me to use in CygWin ( pixelma, Blue_Dude?) I'd *really* appreciate it.
02:14:33S_a_i_n_tInstalling git trashed my CygWin install, now I can't recover the toolchain :/
02:19:13 Quit S_a_i_n_t ()
02:23:48 Join S_a_i_n_t [0] (S_a_i_n_t@
02:25:21 Quit robin0800 (Quit: No Ping reply in 180 seconds.)
02:25:44 Join robin0800 [0] (
02:29:27 Quit domonoky1 (Read error: Connection reset by peer)
02:30:53 Nick tchan1 is now known as tchan (
02:31:09 Quit robin0800 (Remote host closed the connection)
02:31:20 Quit tchan (Changing host)
02:31:21 Join tchan [0] (~tchan@lunar-linux/developer/tchan)
02:37:33 Join t0rc [0] (~t0rc@unaffiliated/t0rc/x-5233201)
02:46:12mc2739funman: should r26345 be committed to the 3.6 branch also?
03:26:08 Join CGL [0] (~CGL@
03:38:14 Quit ssorgatem (Remote host closed the connection)
03:44:19 Join anewuser [0] (anewuser@unaffiliated/anewuser)
03:59:22 Join kramer3d_ [0] (~kramer@unaffiliated/kramer3d)
04:01:03 Join Barahir_ [0] (
04:01:17 Quit wincent_balin (Ping timeout: 240 seconds)
04:01:29***Saving seen data "./dancer.seen"
04:02:46 Quit kramer3d (Ping timeout: 264 seconds)
04:04:34 Quit Barahir (Ping timeout: 264 seconds)
04:08:13 Join CaptainKwel [0] (
04:11:19 Join Strife89DS [0] (~nds@
04:14:31 Join JdGordon1 [0] (
04:30:30CIA-5New commit by 03jdgordon (r26346): %aX should have been %ax.. ...
04:36:58 Quit JdGordon1 (Ping timeout: 264 seconds)
04:39:11 Join vaguerant [0] (
04:39:17 Quit vaguerant (Changing host)
04:39:17 Join vaguerant [0] (~vaguerant@wikipedia/vague-rant)
04:39:57vaguerantIs there a general range of how many (good) patches someone needs to submit to be considered for commit access?
04:40:04 Quit TheSeven (Ping timeout: 276 seconds)
04:42:45 Quit amiconn (Disconnected by services)
04:42:45 Quit pixelma (Disconnected by services)
04:42:45 Join pixelma_ [0] (quassel@rockbox/staff/pixelma)
04:42:47 Join amiconn_ [0] (quassel@rockbox/developer/amiconn)
04:43:05 Nick pixelma_ is now known as pixelma (quassel@rockbox/staff/pixelma)
04:43:07 Join TheSeven [0] (~TheSeven@rockbox/developer/TheSeven)
04:43:09 Nick amiconn_ is now known as amiconn (quassel@rockbox/developer/amiconn)
04:44:35 Quit Strife89DS (Ping timeout: 272 seconds)
04:45:25 Quit Boldfilter (Quit: Boldfilter)
04:47:07 Join phanboy4 [0] (
04:47:15JdGordonvaguerant: no, but it also more about making yourself known... being in IRC and active helps alot
04:48:38 Join funman [0] (~fun@rockbox/developer/funman)
04:48:41vaguerantNot so much myself as csavery, who's been doing a lot of work on PictureFlow; making it not-horrible on monochrome screens, adding a whole WPS-PictureFlow integration mode.
04:48:54funmanmc2739: yep
04:49:19funmanvaguerant: csavery comes on irc?
04:50:22vaguerantNo idea if he does; just figured I'd drop by and ask since he's doing some cool stuff I wouldn't mind seeing committed.
04:50:50funmani guess he should ask Unhelpful about it
04:51:48Unhelpfulwhere is this cool stuff? :)
04:51:57vaguerantOn Flyspray. :p
04:52:20vaguerantNah, he's talked about some of it on the board too, I'll grab you some links.
04:52:32funmanmc2739: i used the e200v1 method to implement lcd_flip on fuzev1
04:52:48funmani guess it should work too and make lcd_window() simpler
04:53:39vaguerantUnhelpful: e.g. Here's his post about the WPS integration for PictureFlow:
04:54:57funmanmc2739: works? I had to adjust another reg
04:56:04vaguerantUnhelpful: Here about an option to disable reflections in PF, which gives more room for larger album art (also at the end is a mention of a bug I reported which he submitted a patch for):
04:56:54Unhelpfulhrm, surely you could "just" make the album art larger, and less reflection would be rendered at the bottom?
04:57:14vaguerantThat's true, but it poses a problem on devices like the Clip series.
04:57:32vaguerantThose have 16px yellow/black at top and 48px blue/black at bottom.
04:57:46vaguerantZooming in the album art looks /horrible/ when you cross over the yellow/blue barrier.
04:58:04vaguerantNot least because there's a gap between the two sections of the screen.
04:58:55funmansomeone can test a patch on fuzev2? mine won't turn on/off
04:59:43 Quit anewuser (Quit: for SELL 2 by the price of 1 now!)
04:59:58vaguerantUnhelpful: Here's a screenshot of PictureFlow just taken from a sim with the covers going over the gap:
05:00:49funmanit doesn't look that good on clip :)
05:01:57vaguerantNo kidding; but csavery's patch adds an option which disables the reflection and glues the covers to the bottom of screen; then you have full control over whether it hits the yellow area through zooming.
05:02:27 Join saratoga [0] (~9803c20d@gateway/web/freenode/x-qhujxwwbzcfdsfal)
05:02:59funmanand the title is taking the top pixels?
05:03:03 Quit Barahir_ (Ping timeout: 265 seconds)
05:03:05vaguerantThat's right.
05:03:24funmanthat'd be nice
05:03:29saratogadid he ever ask kugel about that patch?
05:03:48vaguerant(With csavery's patch; currently setting title to display at top is not respected on small screens due to a bug.)
05:04:24vaguerantNot sure really, he's been by here but not sure if he's specifically spoken to kugel about it.
05:05:40funmansaratoga: i found new stuff about the clipv2 noise today, did you see it ? (logs between 13 & 16)
05:05:46vaguerantRoughly looks like this; this is just the result of me doing some ugly hacking of pictureflow.c (I'm no coder) but his patch has the same visual result (note that the screenshot is scaled *2):
05:06:16funmanmost interesting is that test_codec triggers the noise so the problem is independant of i2sout / DAC
05:06:21saratogafunman: you mean the stuff about it happening when the DAC is off?
05:06:53CIA-5New commit by 03jdgordon (r26347): fix %Xl tag... this shuold be useable now
05:06:53funmanthe noise is constant though, there are no 'peaks' like when the cpu is boosting
05:06:56saratogatest_codec basically uses the screen and IRAM I think?
05:07:10 Join Boldfilter [0] (
05:07:21funmanscreen operations make some noise but switching hold should remove it
05:07:28saratogawell the IRAM is internal to the CPU so i'm not sure how to deal with that
05:07:48funmanwhen i ran test_mem i didn't hear that much noise
05:08:06saratogadid you try it testing IRAM?
05:08:52 Quit Boldfilter (Client Quit)
05:09:37saratogatest_codec and test_mem basically do the same thing, though I guess test_mem probably spends more time stalled waiting on memory while test codec spends more time working out of cache
05:10:20 Quit elinenbe (Ping timeout: 240 seconds)
05:10:29Unhelpfulfunman: i had some thoughts about clip. i think that one clip the best we can really get is full-on, full-off, and a few shades around 50%. the rest we can do with spatial dithering
05:13:45vaguerantHow many "shades" do you have going right now?
05:19:09Unhelpfulfunman: more like 128 or 129, i think, actually. but because of the speed of the clip display, the ones that aren't near the middle range tend to look pretty awful. even the mid-range you can tell is just flickering if you blink at it some.
05:19:46Unhelpfulon LCD some blending is done by the slow display. we get pretty much none of that on the clip.
05:21:10funmanperhaps we could reduce the display frequency ?
05:22:03funmanlook for DISPLAY_CLOCK in lcd-ssd1303.c and s/0x10/0x1f/ will divide display clock freq by 16
05:25:27Unhelpfulfunman: that wouldn't help. the issue is that the pixels of the display respond much more quickly than LCD to changes in value.
05:26:59CIA-5New commit by 03jdgordon (r26348): dead simple Makefile
05:27:42 Quit moparx (Ping timeout: 265 seconds)
05:32:42vaguerantRight, you don't get the same motion blur on OLED that you get with LCD; that's why people are trying to market OLED TVs as the next big thing.
05:41:33*funman tries some potential power-saver on clips
05:42:13 Join shai [0] (
05:43:11 Quit Horscht (Quit: Verlassend)
05:47:56 Nick kramer3d_ is now known as kramer3d (~kramer@unaffiliated/kramer3d)
05:51:14jhMikeSfunam: this init is alright but it has problems with short calls and memory > 32MB (saves 12K but has to put init in middle of the audiobuf)...but, it's been giving me some random thoughts about how to handle IRAM targets with short calls
05:54:04funmanwhat are you thinking about?
05:55:39jhMikeSany way to "export" symbols, sort of treating IRAM and DRAM like separate modules? I'd thinking some longcall stub generation could happen when the calls are accross the boundaries?
05:56:06funmanlongcalls only when needed?
05:57:13funmanperhaps you could parse the elf file, but you would need to build 1 elf for iram ?
05:57:51jhMikeSwas thinking something like that
05:58:16 Quit t0rc (Remote host closed the connection)
06:01:32***Saving seen data "./dancer.seen"
06:02:26*jhMikeS isn't familiar enough with what's available in the tools already and how to exploit it
06:05:20*jhMikeS wonders about parsing the .elf, relocating calls to DRAM from IRAM and calls to IRAM from DRAM and generating the stubs (1 intruction + data it should be)
06:07:39 Quit Strife89 (Quit: Leaving)
06:07:59CIA-5New commit by 03funman (r26349): genversion: generate version.c and version.h independantly
06:09:37funmanyou can build with -mlong-calls and use short-call attribute, but the reverse would be better :/
06:09:57funman functions with the `long-call' attribute or the `section'
06:09:57funman attribute, and functions that are within the scope of a `#pragma
06:09:57funman long_calls' directive, will always be turned into long calls.
06:11:06funmanhm section doesn't force long calls here
06:11:31 Nick CGL is now known as [CGL] (~CGL@
06:14:11 Quit Strife1989 (Quit: Leaving)
06:14:27FlynDicefunman: I can test a fuzev2 patch in a little while if you still need it.
06:15:11jhMikeSfunman: we tried the attributes without success
06:15:38mc2739funman: that works on e200v2 - I also cleaned up lcd_window and lcd_window_blit -
06:17:01jhMikeSif the .elf could be massaged, then you could pass that to the linker, using the .lds as-is
06:17:11funmanmc2739: nice, you still need to change the entry mode for video i think
06:17:44mc2739funman: right, I forgot that
06:18:00funmanmc2739: lcd_window() doesn't work for YUV ?
06:18:18funmanon fuze the registers are different but i think min / max can be inverted
06:20:57FlynDicefunman: What am i testing for?
06:21:11funmanFlynDice: 180° rotation
06:32:09Unhelpfulfunman: we can't really reliably determine, except at link time, whether long calls are needed or not for any given function call.
06:32:18Unhelpfulthat's why the interest in EABI etc. :)
06:34:15mc2739funman: if I use lcd_window for YUV, the display gets flipped and mirrored in mpegplayer
06:35:00FlynDicefunman: Upside down works fine, scrollwheel action is fine, plugins don't seem to like it at all though, even without being upside down
06:35:45funmanhm is the display shifted?
06:36:05FlynDicenot at all as far as I can tell
06:36:31funmanplugins are slow?
06:36:48funmani mostly tried chopper
06:37:23FlynDiceno they lock up, every one I've tried so far, and this is rightside up.
06:39:19FlynDicechopper just gave a data abort @ 301539F0
06:40:16funmanhm do they work without the patch ?
06:40:39FlynDicejust to doublecheck here rolo'ing the patched firmware will not make a difference right?
06:40:55FlynDiceyes, they work fine with svn
06:41:16funmanmy mind is blown
06:43:09funmanthere should be no difference if you boot without 180° & don't let the screen turn off/on
06:43:30FlynDiceIt seems to be a problem with all plugins
06:44:46FlynDice& from what I can tell so far only plugins
06:47:28funmanhm no i think rolo will not work correctly
06:47:43 Quit markun (Remote host closed the connection)
06:48:31funmanhm it should be ok: HAVE_CPUCACHE_INVALIDATE is defined but i don't know where ^^
06:48:53funmanah, mmu-arm.h
06:49:15 Join Boldfilter [0] (
06:49:16FlynDicewell I'll try it in .rockbox too
06:54:10FlynDicesame results in .rockbox
06:55:30funmanI guess I'll wait for my fuzev2 to discharge
06:55:51FlynDicetime for bed, happy hacking... good luck w/the fuze
06:59:04 Quit saratoga (Quit: Page closed)
06:59:37mc2739funman: updated - did I miss anyhing else?
07:00:55funmanstill video entrty mode ;)
07:01:05 Join Barahir [0] (
07:02:02funmanhmm that's teh same diff?
07:02:22jhMikeShmmm...maybe arm-elf-objcopy would be helpful ??
07:03:06jhMikeS(for what I was speaking of earlier)
07:03:23funmanyeah, objcopy doesn't help with lcd :P
07:03:53funmansed -i s/undocumentedregister// lcd-fuze*
07:04:27funmanjhMikeS: that sounds like a job for gcc
07:06:55funmanUnhelpful: how does EABI affect short/long calls?
07:09:37 Join rajendrau [0] (~chatzilla@
07:10:39jhMikeSfunman: gcc doesn't do the job and isn't section-aware like that (it hasn't proven to be straightforward at all or else I'm pretty sure it would be done by now)
07:11:48rajendrauI am not able to view video files with (.mpg) copied to sansa e260 media player. I tried by copying video to videos folder, music folder and also .rockbox folder
07:12:46jhMikeSrajendrau: you need to be able to view hidden files. you also can put the files in any folder.
07:12:56Unhelpfulfunman: EABI causes all calls to be short calls. if a short call is out of range at link time, a long jump to the proper address is inserted near it, and the short call is redirected there.
07:12:57Unhelpfulthis is a slight speed cost *when* long calls would have happened, but overall a fairly large code size win.
07:13:04funmanset 'show files' settings to 'all'
07:13:45funmanhm ok
07:13:46jhMikeSUnhelpful: it'll do stubbing when needed? per-instance?
07:14:06funmanwhat prevents us from using eabi then, only doom ?
07:15:30CIA-5New commit by 03mc2739 (r26350): Simplify lcd_set_flip, lcd_window, and lcd_window_blit ...
07:17:03 Part rajendrau
07:17:13funmanmc2739: mpegplayer still works ??
07:17:57UnhelpfuljhMikeS: *essentially* per target address - if there are a bunch of callers in range of the same stub they'll all share it.
07:17:57mc2739funman: yes
07:18:02 Quit panni_ (Read error: Connection reset by peer)
07:20:16Unhelpfuli believe the space savings amounts to something like 60KB on e200, and that's pretty typical for other ARM targets currently using -mlong-calls
07:23:44funmanah right vertical filling is used in video and normal display; on fuze it's only for video
07:23:58funmanfuze / other landscape displays
07:25:14 Join esperegu [0] (~quassel@
07:25:50CIA-5New commit by 03jdgordon (r26351): Add the beggingings of a perl script to run the update over a whole theme zip file... someoene who knows perl please finish this :D
07:28:36*jhMikeS isn't yet familiar with eabi. it's a tool or an option?
07:29:03funmanjhMikeS: you must build a new gcc will give you a arm-elf-eabi-* toolchain and you can use it with ./configure −−eabi
07:29:50jhMikeShmm...why is doom an issue with that?
07:30:13funmani believe it uses misfeatures of the old abi
07:32:01 Join einhirn [0] (
07:32:21funmanFS #11093 Doom hardlocks on Nano2G when compiled with eabi
07:33:13jhMikeSI'd say that's reason to fix doom
07:33:55LloreanWe could always disable Doom after the release, move to EABI, and hope someone files a patch. :)
07:34:49jhMikeSwhat is the real issue? what part of it has the problem?
07:35:08funmani think Torne or kugel will know
07:37:02jhMikeSwell, good to know something already does what I had in mind
07:39:12funmani was thinking about thumb
07:39:41jhMikeSyeah, that too
07:39:51funman-> replace CC by a script which will run cpp on each file, grep for inline asm, and build with -mthumb if there is none
07:39:55 Quit CaptainKwel (Quit: Ex-Chat)
07:42:08jhMikeSinline asm can't be assembled as thumb?
07:42:26 Quit [CGL] (Remote host closed the connection)
07:42:41funmaninline thumb asm yes
07:43:12funmanif there is a single 32 bits instruction in a .c file the whole thing must be built as 32 bits code
07:43:30funman(e.g., disable_irq())
07:43:32jhMikeStrue, true
07:44:59funmangigabeats can do thumb2?
07:45:43 Quit funman (Quit: bbl)
07:48:22jhMikeSit has some v6 instructions in common with ARM code
07:48:53UnhelpfuljhMikeS: EABI specifies that enums shall be the smallest size sufficient to hold all enumerated values.
07:49:39Unhelpfulthis screws up anything that points at enums with int*, and causes warnings and potential runtime errors for code that assigns or compares arbitrary integer values to enums.
07:51:20jhMikeSenums under this definition means what? (sounds different than enum { ... }; )
07:53:24Unhelpfulno, it means precisely that. :)
07:53:39jhMikeSok. :)
07:54:55Unhelpfulbut the storage type for enum values is no longer int under eabi, but the smallest type which will hold all values... so your enum with values 1,2,3,4 will be stored as an unsigned char. if you try to compare a signed value to it, you will get a warning (and probably unexpected runtime behavior).
07:55:25jhMikeSah, smallest as in "smallest datatype"
07:55:27Unhelpfulif you try to point at it with int* then dereferencing that pointer will at *best* overwrite some adjacent memory
07:56:49jhMikeSyeah, well, toss a nice something = INT_MAX at the end of them just to tide it over
07:58:00Unhelpfulor do the right thing, and don't assume that enums are equivalent to int? :)
08:01:18jhMikeS"right" things can come in steps of course, and if it's just one plugin...
08:01:36***Saving seen data "./dancer.seen"
08:01:55jhMikeStake the time to be super-correct later, get it working now
08:02:48Unhelpfulalso that will cause warnings if the enum is used in a switch statement :)
08:03:14jhMikeSyou mean as without a "default"?
08:04:26Unhelpfula default *should* clear the warning, but imo it's better to write code that uses pointers of the appropriate type, and that doesn't stuff magic flag values into enum variables that are *not* among the enumerated values.
08:05:39jhMikeSor compile doom without there something with the calling conventions?
08:07:57Unhelpfulthere are a number of other little changes that make that not work, also there will be warnings on link for eabi and non-eabi objects.
08:08:52Unhelpfulthe best band-aid IMO, if you don't want to do a proper fix, is to replace the offending variables with ints, and use the enum types as a shorthand for specifying constants.
08:09:26jhMikeSlooks like a quite a project after looking closer :\
08:10:29Unhelpfuland in any case if you ever mean to point to it, you don't really have a choice, you have to either make the variable an int, or use a pointer to the appropriate enum type. settings that use enums have to be stored as ints because of this - the first few things we had to fix in core to make EABI builds load without crashing mostly had to do w/ int pointers for settings to enum global variables.
08:10:47*jhMikeS runs to update get a better idea. things were quite stale but I extended the vmware image to 4GB
08:13:57jhMikeSthe rockbox source tree has gained some heft, that's for sure. having two, there wasn't enough room to work and build a sim
08:15:33 Join bmbl [0] (~Miranda@unaffiliated/bmbl)
08:15:42*Unhelpful loves git for that :)
08:17:17jhMikeSthat works with SVN? maybe I should switch over then
08:19:48*jhMikeS keeps hearing praises being sung but keeps procrastinating as usual
08:25:11 Join ender` [0] (
08:28:34 Join mitk [0] (~mitk@
08:31:47 Join GeekShadow [0] (~Antoine@reactos/tester/GeekShadow)
08:37:40 Join Guest99397 [0] (bjst@rockbox/developer/Zagor)
08:38:24 Nick Guest99397 is now known as Zagor (bjst@rockbox/developer/Zagor)
08:42:46 Quit kramer3d (Quit: Leaving)
08:44:28 Join flydutch [0] (
08:48:58 Quit GeekShadow (Ping timeout: 264 seconds)
08:59:17 Join GeekShadow [0] (~Antoine@reactos/tester/GeekShadow)
09:01:29 Join BHSPitMonkey [0] (~stephen@unaffiliated/bhspitmonkey)
09:09:53 Join petur [0] (~petur@rockbox/developer/petur)
09:15:20AlexPfunman (logs): r26345 should be applied to the branch too
09:17:42 Join lpereira [0] (
09:18:47 Join markun [0] (www-data@rockbox/developer/markun)
09:20:50 Quit markun (Client Quit)
09:23:43 Join DerPapst [0] (
09:28:44 Join wodz [0] (
09:28:53 Quit phanboy4 (Ping timeout: 265 seconds)
09:32:35 Join LinusN [0] (
09:32:35 Quit LinusN (Changing host)
09:32:35 Join LinusN [0] (linus@rockbox/developer/LinusN)
09:32:42wodzI've got 4:14 runtime on HD200 with ~800mAh NOKIA BL-5C battery. Not very impressive :-/
09:40:58 Join Markun [0] (
09:41:10 Part Markun
09:42:59 Join Markun [0] (~chatzilla@rockbox/developer/markun)
09:44:20 Quit Barahir (Ping timeout: 258 seconds)
09:45:04 Quit Markun (Client Quit)
09:46:46 Join efyx [0] (
09:48:16 Join Rob2223 [0] (
09:52:05 Quit Rob2222 (Ping timeout: 265 seconds)
09:56:48 Quit BHSPitMonkey (Read error: Connection reset by peer)
09:57:49 Quit vaguerant (Ping timeout: 260 seconds)
09:58:55S_a_i_n_tIs whoever maintains here?
09:59:08S_a_i_n_tIf so, could you please look at this:
09:59:23S_a_i_n_tIt keeps failing :/
09:59:36S_a_i_n_tAlways in the same spot.
10:00:23Zagorahhh, cygwin!
10:00:58Zagorit gives me a rash :)
10:01:09 Join Barahir [0] (
10:01:41***Saving seen data "./dancer.seen"
10:01:51Zagorand I also have too little knowledge of it to say why it breaks your gcc build
10:02:12 Quit esperegu (Read error: Connection reset by peer)
10:02:31S_a_i_n_tDammit. Can you see where it's failing? All that output is foreign to me.
10:02:59S_a_i_n_tAll I know is that it's failing in the same spot each time.
10:03:28 Join esperegu [0] (~quassel@
10:03:45Zagorit's failing on host-cygwin.c
10:03:54Zagorerror: storage size of `mode_size' isn't known
10:04:14Zagorprobably due to "insn-modes.h: No such file or directory"
10:04:14 Join Kevone [0] (
10:04:29Kevoneanyone awake?
10:04:48S_a_i_n_tplenty of people, better to just ask your question
10:05:12KevoneI was just wondering whats happening with the connection of the ipod nano 2nd generation
10:05:26S_a_i_n_tyou'll need to be more specific.
10:05:37S_a_i_n_tBut, I have that target so can probably help.
10:06:16Kevoneis the usb port drivers getting developed
10:06:29S_a_i_n_twhich OS do you have?
10:06:34KevoneWindows 7
10:06:36S_a_i_n_twindows, *nix etc.
10:07:55KevoneI was look at the chart and the current status is serious problem
10:08:14Kevonefor the ipod 2gen
10:09:14S_a_i_n_tWell, personally I don;t have any problems with it. Previously the HID enumeration caused some problems...but that has been disabled on the Nano2g
10:09:21S_a_i_n_twhat revision is your build?
10:10:12Kevoneare u talking about Rockbox?
10:10:43S_a_i_n_tHmmm, well...I'd suggest using a current build.
10:10:56S_a_i_n_tAnd if that doesn't work, then try forcing disk mode.
10:11:24S_a_i_n_t(restart the iPod using menu+select, then press select+play immediately)
10:11:49S_a_i_n_tThat will force disk mode, then the DAP should mount.
10:12:05ranmaS_a_i_n_t: That looks like (googled for "insn-modes.h: No such file or directory")
10:12:14Kevoneshould it be connected to the computer
10:12:40S_a_i_n_tNo, not while forcing it into disk mode.
10:13:01S_a_i_n_tYou can also just use hold to boot into the Apple FW if it's easier for you.
10:13:26S_a_i_n_t(turn the player on, then flip the hold switch to on)
10:13:56 Join moparx [0] (~moparx@unaffiliated/moparx)
10:13:56S_a_i_n_t(or turn the player off, flip the hold switch on, then plug in the USB will start in the Apple FW)
10:15:42S_a_i_n_tranma: It does seem like the same problem...yes. Pity there is no solution there :\
10:15:43Kevonei got it into disk mode but all that shows up is a check mark and "ok to disconnect"
10:16:04S_a_i_n_tKevone: Yes, that is correct.
10:16:08S_a_i_n_tIs it plugged in?
10:16:31S_a_i_n_tSweet, plug it into the PC and it should mount.
10:17:07ranmaS_a_i_n_t: Well, it does say that it only happens with a parallel make
10:17:18Kevoneok, now disconnect
10:17:44S_a_i_n_tI have no idea whi it isn't mounting under Rockbox...but the Nano2g has been known to have some problems mounting to some OSes
10:18:48S_a_i_n_tKevone: To get back into Rockbox, just "hard reset" the player. (hold menu+select for ~5-10 seconds)
10:19:45Kevoneok thanks for the help, one more question. Do u know what happened to the ArchosUnlock.exe
10:19:47 Join kugel [0] (~kugel@rockbox/developer/kugel)
10:20:13kugelUnhelpful: did you try gcc 4.4.4 already?
10:20:13S_a_i_n_tNo, I don't sorry. I'm unfamiliar with the Archos targets.
10:20:26S_a_i_n_tI guess you could just try googling the filename.
10:20:32KevoneIts a program that unlocks a masterpassword for a hdd through the reboot
10:21:52Kevonei can't find the link anywhere and thats where i ran into rockbox. I was trying to find a software that would which was on this website but the download link isn't valid :(
10:22:36kugeljhMikeS: eabi is the way to go, it's just that nobody has nobody has bothered to fix doom yet
10:22:45S_a_i_n_tOh, I see. Sorry, I don't have any Archos targets...and have never heard of that application.
10:24:09LloreanJdGordon: Why would you clone the theme site? Doesn't it already support themes that only work on release or current but not both?
10:24:16wodzI found some interesting pdf (public one) which treats about li-ion/lipo battery charging scheme and how different parameters affect available capacity and battery lifetime. This topic comes back from time to time so maybe it would be good to store this pdf someware and setup small wiki page pointing to this file?
10:24:26kugeljhMikeS: in doom, there's at least one mis-use of enums, the settings array. but I still got a crash later after fixing that
10:25:13Kevonei have to reinstall rockbox on to my ipod cz its had crashed and needs to be restored
10:25:26Kevonenot a problem since i can back everything up
10:26:09S_a_i_n_tKevone: I reccommend using a current or daily build.
10:26:12ranmawodz: See also:
10:27:03S_a_i_n_tI use current builds on my Nano2Gs (all my targets actually) and it seems to be a lot more stable than 3.5/3.5.1
10:27:11wodzranma: I am quite familiar with this topic as I work on university with new types of li batteries :-)
10:27:23 Part Boldfilter
10:28:05ranmawodz: Ok :)
10:28:58ranmaI just thought I'd mention it since its the most comprehensive explanation of various charging-related issues I know :)
10:30:06wodzranma: try this
10:31:27Kevonethats some intresting stuff about the li-ion battery
10:32:20wodzthe most comprehensive knowledge source I have is "Handbook of batteries" but this is 100+ pages of paper book
10:33:38 Quit moparx (Ping timeout: 264 seconds)
10:35:35 Join dfkt [0] (dfkt@unaffiliated/dfkt)
10:37:41Kevoneis the handbook in ebooks? lol
10:38:15S_a_i_n_tNano2G Manual:
10:49:00 Quit Topy44 (Quit: No Ping reply in 180 seconds.)
10:49:22 Join Topy44 [0] (
10:55:20 Quit Soap_Hotel (Quit: CGI:IRC (Ping timeout))
10:56:01 Quit bmbl (Ping timeout: 258 seconds)
11:02:53wodzhmm I am wondering would it look nicer to have curve like plot instead of bar like plot in view_battery()? it will be trivial change lcd_vline -> lcd_drawpixel
11:04:51S_a_i_n_twodz: IMO, that would look nicer.
11:05:13S_a_i_n_tKinda irrelevant, but it'd look a bit better ;)
11:06:02 Join rtsa [0] (
11:06:20wodzwhen my charging battery_bench will finish I'll fiddle with this a bit
11:06:40 Quit Kevone ()
11:17:55 Part LinusN
11:21:32 Join LinusN [0] (linus@rockbox/developer/LinusN)
11:31:00 Quit esperegu (Read error: Connection reset by peer)
11:32:06 Join esperegu [0] (~quassel@
11:35:22 Join gk_ [0] (
11:35:30 Part rtsa
11:39:31 Join Kevone [0] (
11:43:23 Quit dfkt (Quit: -= SysReset 2.53=- Ph'nglui mglw'nafh Cthulhu R'lyeh wgah'nagl fhtagn.)
11:44:21 Part gk_
11:44:44 Quit ps-auxw (Ping timeout: 248 seconds)
11:44:47 Part LinusN
11:47:11 Quit shai (Read error: Connection reset by peer)
11:49:51 Quit bieber (Ping timeout: 265 seconds)
11:50:08 Join bieber [0] (
11:54:57 Join shai [0] (
11:56:23 Join ps-auxw [0] (
12:01:45***Saving seen data "./dancer.seen"
12:02:20 Quit pamaury (Ping timeout: 248 seconds)
12:09:39 Quit kugel (Ping timeout: 260 seconds)
12:10:30 Join xavieran [0] (
12:14:51 Join pamaury [0] (~pamaury@rockbox/developer/pamaury)
12:22:17 Quit GeekShadow (Ping timeout: 258 seconds)
12:31:16 Quit esperegu (Read error: Connection reset by peer)
12:31:42 Join mischasworld [0] (
12:33:57 Join esperegu [0] (~quassel@
12:53:45 Join ucchan [0] (
12:57:02 Quit esperegu (Read error: Connection reset by peer)
12:58:01 Join esperegu [0] (~quassel@
13:00:42 Join domonoky [0] (~Domonoky@rockbox/developer/domonoky)
13:01:41 Join elinenbe [0] (
13:07:06 Quit jordan` (Remote host closed the connection)
13:10:04wodzS_a_i_n_t: could You try this: and tell if that is nicer to You?
13:13:01S_a_i_n_twodz: I'm terribly sorry, but I don't actually have a build environment set up at the moment. Installing git trashed my CygWin installation somehow and I am still in the process of setting up a new environment. If you can, you could just upload a screenshot of the battery screen and I could give you an opinion of how it looks.
13:14:50wodzanyway now I see my patch has some problem with scaling
13:15:54S_a_i_n_tI definitely like the idea, I'm almost positive that it will look better if you do it correctly.
13:16:40wodzhmm it is ok
13:16:55wodzI'll try to do screenshot or smth
13:18:44 Nick fxb__ is now known as fxb (
13:22:04 Join watto [0] (~watto@
13:24:22 Join markun [0] (~54f18043@rockbox/developer/markun)
13:26:16 Join krazykit` [0] (
13:26:46 Quit krazykit (Ping timeout: 240 seconds)
13:31:20 Quit Kitar|st (Ping timeout: 260 seconds)
13:34:13 Join Kitr88 [0] (
13:39:00S_a_i_n_twodz: Yes, in my opinion that looks a *lot* better. Great job \o/
13:39:20S_a_i_n_twhat DAP is that by the way?
13:39:39wodzmini 1G
13:40:06CIA-5New commit by 03mc2739 (r26352): Catalan language update by Adrián Cereto Massagué
13:40:35S_a_i_n_tAha...anyway, yep. I think that looks very nice.
13:45:21wodzok I'll put this patch on FS (as it comes back to life I think)
13:50:31wodzhere we go: FS #11324
13:55:47 Join GeekShadow [0] (~Antoine@reactos/tester/GeekShadow)
13:59:17wodzrockbox is taking more than 2x higher current than OF on HD200.
14:00:25 Join piotrekm [0] (~piotrek@unaffiliated/piotrekm)
14:01:48***Saving seen data "./dancer.seen"
14:03:36S_a_i_n_twodz: Is the tracker back up?
14:06:20Mode"#rockbox +o Zagor" by ChanServ (ChanServ@services.)
14:06:24Topic"3.6 now branched | Please read before speaking: | Please direct offtopic/social chat to #rockbox-community | This channel is logged at | DevCon2010 will be from June 4th to 6th in Ghent," by Zagor (bjst@rockbox/developer/Zagor)
14:06:27Mode"#rockbox -o Zagor" by ChanServ (ChanServ@services.)
14:06:50 Quit wodz (Quit: Leaving)
14:06:57 Quit ucchan (Quit: Leaving...)
14:07:38 Quit ranma (Ping timeout: 240 seconds)
14:07:51 Join ranma [0] (
14:14:46 Quit mischasworld (Ping timeout: 240 seconds)
14:19:22 Quit ps-auxw (Quit: leaving)
14:19:28CIA-5New commit by 03mc2739 (r26353): Simplify lcd invert
14:22:50 Join hebz0rl [0] (
14:33:13 Join ps-auxw [0] (
14:35:18 Quit S_a_i_n_t ()
14:35:42 Join S_a_i_n_t [0] (S_a_i_n_t@
14:37:26 Join krazykit [0] (~kkit@
14:38:46 Quit krazykit` (Read error: Operation timed out)
14:42:23 Quit krazykit (Ping timeout: 264 seconds)
14:42:26 Join krazykit [0] (~kkit@
14:45:19 Join mischasworld [0] (
14:51:38 Join DerPapst1 [0] (
14:52:03 Join GeekShado_ [0] (
14:53:01 Quit DerPapst (Ping timeout: 240 seconds)
14:53:22 Quit steve|m (Ping timeout: 276 seconds)
14:56:11 Quit GeekShadow (Ping timeout: 264 seconds)
15:01:33 Join evilnick_B_ [0] (~0c140464@rockbox/staff/evilnick)
15:01:35 Quit evilnick_B (Ping timeout: 252 seconds)
15:02:52 Nick evilnick_B_ is now known as evilnick_B (~0c140464@rockbox/staff/evilnick)
15:06:36 Join steve|m [0] (
15:07:51 Join anewuser [0] (anewuser@unaffiliated/anewuser)
15:09:02 Join notKlaatu [0] (~sxe@unaffiliated/notklaatu)
15:12:31 Join Peter93 [0] (
15:15:08Peter93Hello! What do you think about emulating a USB-CDdrive like U3 Sticks do?
15:15:55gevaertsBeing able to have a plugin that does MSC with the multimedia extensions seems like a nice idea, yes
15:16:54gevaertsIt could be in the core too of course, but I like the idea of selecting an iso file, pick "Open with", and select the USB plugin
15:17:06 Join robin0800 [0] (
15:17:18gevaertsThe only problem there is of course that today USB can't really be done from plugins :)
15:18:45Peter93Realy cool is: I can boot from the CD-Drive of my U3 Stick...
15:19:09gevaertsYes, that's one of the reasons to have this
15:19:26*gevaerts hasn't ever used U3, so he doesn't know how exactly those work
15:19:27epicfailguywow thats awesome i never heard of that.
15:19:33 Quit piotrekm (Quit: piotrekm)
15:19:37*Bagder doesn't get it
15:19:47epicfailguythatd be perfect for my netbook.
15:19:51Bagderso there's an iso on the storage, or how is it a CD?
15:19:53Peter93it is!
15:19:54TorneBagder: Windows won't autorun stuff on a USB hard drive, but it will autorun stuff from a USB CDrom drive
15:20:04Tornethe U3 comes up as two usb devices, one of each
15:20:07Peter93you burn the iso on it
15:20:07gevaertsBagder: yes, in the basic version
15:20:21Torneyou can replace the image the cdrom half exposes.
15:20:27gevaertsThe advanced version (in my mind) will also export playlists as audio CDs :)
15:20:43TornePeter93: You shouldn't, generally, need to be a CD drive to actually boot a computer
15:20:55TorneOnly the most ancient computers can't boot from regular USB MSC devices
15:21:01Peter93but its much easier
15:21:06BagderI thought most computers willingly booted from usb or whatever these days
15:21:08Torneit is?
15:21:14epicfailguymodern computers do
15:21:15Tornei find it much easier to boot from USB MSC than from CD :)
15:21:18gevaertsTorne: you don't, no. You do however need to play with bootsectors otherwise, which you might not want to do on e.g. a gigabeat S
15:21:26CIA-5New commit by 03Blue_Dude (r26354): Voice buffer can be much, much smaller. Code cleanup, logf fix
15:21:28S_a_i_n_tdepends on ones idea of easier I guess. :/
15:21:44Tornegevaerts: Yah, but there's a better fix for that: make our mass storage driver able to export any old image file ;)
15:21:57Tornelike the linux g_filestorage drive can
15:22:18 Join Blue_Dude [0] (~chatzilla@rockbox/developer/Blue-Dude)
15:22:19gevaertsTorne: sure, that's the start. Doing more CD like stuff shouldn't be too hard though, and it might have some advantages
15:22:22TornePeter93: anyway, yes, it could be done if somoene did it
15:22:36Tornebut at the moment our USB architecture doesn't make it particularly easy
15:22:39Peter93ok would be cool
15:23:56Blue_DudeS_a_i_n_t: Still having Cygwin problems?
15:24:08Tornei expect exporting an image would be quite easy/small and could be done within the current usb msc driver
15:24:20Tornei'm not sure what it takes to convince the host that you are a CDROM instead of a generic disk
15:24:26gevaertsActually, the MMC commands shouldn't be too complicated to add (just tedious work), and I suspect that a file-backed storage driver (which would then only be used by the usb drivers) might be reasonably easy
15:24:47Tornegevaerts: a loopback msc driver would be awesome actually
15:25:00Torneyou could stuff a whole bunch of boot images on your player and select one to use whenever needed :)
15:25:06gevaertsMaybe if I feel bored during the weekend :)
15:25:30S_a_i_n_tBlue_Dude: Not sure yet, I'm re-installing again from scratch...It appears as though something in gcc went bork-bork :/
15:25:47Tornegevaerts: i guess we could do loopback in general through the storage interface :)
15:25:57gevaertsyes, that's my preferred way
15:26:10Torneis there a particularly sensible use for actually being able to mount loopback images on the device?
15:26:23Tornei guess you'd have to enable multidrive everywhere for that to work
15:26:26S_a_i_n_tI thought that was the culprit (and it still might be), but now I'm starting to think the install went badly.
15:26:36S_a_i_n_tI'm eliminating things one by one.
15:26:40Tornesince most players currently only allow one storage device
15:26:50gevaertswell, not if we only support FAT I guess. If we add iso9660, you could mount isos
15:27:02gevaertsAnd thise might have dvd video on them
15:27:14Tornemaybe, i guess
15:27:17TorneVCD images :)
15:27:24gevaertsright :)
15:27:47Tornethat just reopens the filesystem nodo, though, no? :)
15:28:13Blue_DudeS_a_i_n_t: I did a fresh install a week or two ago and so far it's behaving. I'm mostly doing sim builds though.
15:28:23gevaertsSure, but we have devcon coming up to talk about that
15:32:17 Part notKlaatu
15:32:54 Join Zarggg_ [0] (
15:36:44 Quit Zarggg (Ping timeout: 260 seconds)
15:39:50 Join CGL [0] (~CGL@
15:41:40 Join panni_ [0] (
15:42:46 Join DerPapst [0] (
15:44:21 Quit DerPapst1 (Ping timeout: 260 seconds)
15:48:54 Quit einhirn (Quit: Miranda IM! Smaller, Faster, Easier.
15:49:14 Quit Peter93 (Quit: Miranda IM! Smaller, Faster, Easier.
15:59:23 Quit mitk (Quit: Leaving)
16:01:50***Saving seen data "./dancer.seen"
16:04:05 Quit Blue_Dude (Quit: ChatZilla 0.9.86 [Firefox 3.6.3/20100401080539])
16:10:18 Quit TheSeven (Ping timeout: 240 seconds)
16:11:35 Quit anewuser (Quit: for SELL 2 by the price of 1 now!)
16:16:06 Join Farthen_ [0] (
16:16:35 Nick Farthen_ is now known as Guest84887 (
16:17:23 Quit Guest84887 (Client Quit)
16:22:36 Join Xerion [0] (~xerion@
16:27:11 Join funman [0] (~fun@rockbox/developer/funman)
16:36:45FlynDicefunman: SD writing is broken on as3525v1 right now, trying to track it down, at least for me w/ transcend class 6 on both fuzev1 & e200v2
16:37:10funmanno write at all ?
16:37:41FlynDicewell uSD writes I mean
16:37:52 Join Akranis [0] (
16:37:53FlynDiceI can't copy or delete
16:39:06FlynDiceI went back to r 26287 and it works and I'll start bisecting
16:39:41FlynDiceIt may be awhile though I'm in & out here for a bit
16:39:45funmanis it 26291 ?
16:39:54funmanok :o
16:40:08 Join Luca_S [0] (
16:40:24Luca_SFlynDice: same for me, fuzev2
16:40:30funmanjust deleted from transcend class 4 on fuzev1/r26319M (M is probably lcd)
16:40:33Luca_Spanic: wait for tran state failed (PRG) 1
16:40:41Luca_Swhile copying a file from uSD to uSD
16:41:01funmanLuca_S: after r26342 ?
16:41:13Luca_Sclass 6 transcend 8gb
16:41:32*funman built rockbox with -mthumb
16:41:42funmanit's actually bigger :-)
16:42:40 Join solexx_ [0] (
16:42:45funmanand shows the logo but doesn't get to the menu
16:43:39funmantry incraesing the timeout in sd_wait_for_tran_state()
16:43:47funmansomething insane like 30*HZ
16:46:01 Quit solexx (Ping timeout: 258 seconds)
16:49:29 Join kugel [0] (~kugel@rockbox/developer/kugel)
16:55:11 Quit mischasworld (Read error: Connection reset by peer)
16:56:04 Quit petur (Quit: beer time)
16:56:46 Quit bieber (Ping timeout: 272 seconds)
16:57:02 Join bieber [0] (
16:57:44mc2739FlynDice: my class 2 card works on latest svn in my e200v2
16:59:15FlynDicemc2739: Yes, I've got one of those class 6 transcends that were giving us trouble before. The sandisks seem pretty bulletproof.
16:59:58funmanhm my libgcc doesn't support interworking
17:00:39FlynDiceLuca_S: funman: My transcend class 6 works ok in fuzev2 with svn
17:02:10*FlynDice has 4 Sansas out swapping 1 SD card like a madman....
17:02:39Luca_Smy uSD showed itself as being picky some time ago, but the OF handles it fine
17:02:58Luca_Snow compiling after getting a clean tree...
17:03:48bluebrotherbieber: I looked into that QAbstractItemModel thing earlier today. Figured that data() doesn't return a string, so the displayed table is empty. If I simply make it return a fixed string I get a 2x2 table. Interestingly this worked on the Mac but not on Linux, but I haven't found if I simply did a stupid mistake here.
17:08:08CIA-5New commit by 03mc2739 (r26355): Update of Czech language ...
17:08:13funmanI see "MULTILIB_OPTIONS += mno-thumb-interwork/mthumb-interwork" is added by our gcc 4.0.3 patch
17:10:30 Part Zagor
17:10:55CIA-5New commit by 03mc2739 (r26356): Update of Czech language ...
17:14:50funmanyahoo it runs (but codecs crash)
17:16:02 Quit lpereira (Quit: Leaving.)
17:16:09 Join moparx [0] (~moparx@unaffiliated/moparx)
17:16:38funmanbetween 1 & 2% of size reduction, but i think i miss a lot of files
17:16:45Luca_Sfunman: same crash even using 30HZ
17:17:11 Join mischasworld [0] (
17:17:31funmangreylib worsk too
17:17:33funmanLuca_S: ok
17:19:01Luca_Salso I noticed that when waiting, the backlight went off. I could turn it on by clicking a button, but not by moving the wheel - and buttonlight wasn't affected by wheel movement too
17:23:28 Quit robin0800 (Remote host closed the connection)
17:24:11 Quit mischasworld (Remote host closed the connection)
17:26:48 Quit markun (Ping timeout: 252 seconds)
17:30:51 Join petur [0] (~petur@rockbox/developer/petur)
17:31:17CIA-5New commit by 03mc2739 (r26357): Updated Turkish translation ...
17:32:22 Quit Rob2223 (Read error: Connection reset by peer)
17:32:31 Quit petur (Client Quit)
17:34:01CIA-5New commit by 03mc2739 (r26358): Updated Turkish translation ...
17:38:13 Join saratoga [0] (~9803c6dd@gateway/web/freenode/x-qlbniqwbcadpvibw)
17:38:57saratogathe author has an optimized version of the encoder that he thinks should be real time on ARM
17:39:04saratogaare we interested in getting more encoders?
17:39:12saratogai assume yes but i've never even looked at them
17:42:04CIA-5New commit by 03mc2739 (r26359): Update Brazilian Portuguese Translation ...
17:44:32CIA-5New commit by 03mc2739 (r26360): Update Brazilian Portuguese Translation ...
17:47:05 Join Rob2222 [0] (
17:49:21Rob2222Phew, im trying to flash a bootloader-patched FW to my FuzeV2 but after disconnecting it just does not start the firmware upgrade process. Filename is the right. Any hints?
17:50:00FlynDiceLuca_S: Can you try this patch for fuzev2 with your card. I think it should still apply cleanly:
17:50:44Rob2222hmm, i have no compile chain set up atm
17:51:04Rob2222i just tried the default BL-patch from the HP.
17:51:19FlynDiceRob2222: What is the filename. Look closely before you say
17:51:34Rob2222fuzpa.bin (checked 5 times
17:51:49Rob2222i flashed original fuze02.03.33 before, that worked
17:51:55Rob2222now trying to flash the patched
17:52:30Rob2222btw, as i was trying to flash the original fuzpa.dat without clpp_data.dat it wont upgrade, too
17:52:45Rob2222do i have to use the original clpp_data.dat with the patched fuzpa.dat?
17:53:06FlynDiceshould be a .bin extension...
17:53:47funmanclpp_data.dat is a promotional mp3
17:53:59Rob2222typing error, ist a bin
17:54:53FlynDicefuzpa.bin seems correct, where did you copy it to?
17:55:28 Quit saratoga (Quit: Page closed)
17:55:38Rob2222i just trying again with the original fuzpa.bin, thats working (2nd time)
17:55:48FlynDicecan you still see it there with your file browser?
17:55:57Rob2222so it looks he doesnt like ma patched fuzpa.bin
17:57:23FlynDicewhat happened when you copied the OF fuzpa.bin to your device?
17:57:40Rob2222its getting flashed and removed
17:58:25FlynDiceso it did the firmware update with that filename correct?
17:58:50Rob2222btw im using newer mkamsboot.exe version from foums thread (r26289M-100525)
17:58:54 Quit GeekShado_ (Quit: The cake is a lie !)
17:59:09Luca_SFlynDice: compiling now.
17:59:19FlynDiceLuca_S: thx
18:00:23 Quit Curtman (Read error: Connection reset by peer)
18:00:38FlynDiceRob2222: and you compiled that yourself or downloaded from somewhere?
18:01:17Rob2222downloaded here:
18:01:33Rob2222cause wiki information: Fuzev2: mkamsboot 1.2 doesn't work on some devices (always boot to OF) : use SVN.
18:01:51***Saving seen data "./dancer.seen"
18:01:53Rob2222but well in my case it doesnt even flash it
18:02:40funmanyou can also wait next release of RockboxUtility
18:04:24FlynDiceRob2222: So the mkamsboot from the wiki did not work for you or you just haven't tried it?
18:04:31Rob2222trying atm
18:05:55 Quit Rob2222 (Read error: Connection reset by peer)
18:06:21 Join Rob2222 [0] (
18:06:29 Join TheSeven [0] (~TheSeven@rockbox/developer/TheSeven)
18:06:36Rob2222no wiki version doesnt flash, too
18:07:12FlynDicecan you pastbin your output from mkamsboot?
18:07:48Rob2222the ptached firmware file?
18:08:35 Join Horscht [0] (~Horscht2@xbmc/user/horscht)
18:08:44FlynDiceno, just the text output from the console you used to run mkamsboot
18:08:58Rob2222to the channel here?
18:09:34Rob222216 lines
18:09:36Rob2222it are
18:09:45FlynDiceno, go to
18:10:17FlynDiceand paste your output in there, it will give you a link to then copy back here ;)
18:11:07Rob2222looks good imho
18:11:38 Quit TheSeven (Ping timeout: 276 seconds)
18:12:17 Quit bieber (Ping timeout: 276 seconds)
18:12:30 Join bieber [0] (
18:12:49mc2739Rob2222: that does not look like the mkamsboot from the wiki
18:12:57FlynDiceso you renamed output.bin to fuzpa.bin and copied it to your device and upon disconnecting the firmware update fails to happen correct?
18:13:10Rob2222yeah 100%
18:13:23Rob2222mc2739: i tried that one from wiki, too
18:13:32 Join bertrik [0] (~bertrik@rockbox/developer/bertrik)
18:15:01FlynDiceRob2222: and you are in msc not mtp or auto mode for USB correct?
18:15:22Rob2222i have no idea whats wrong
18:15:37FlynDicewell, me neither.... sorry
18:15:45Rob2222i rockboxed my new clip+, 30 mins ago without problems
18:16:01Rob2222yeah, im writing a post in the forum, atm
18:16:12 Join TheSeven [0] (~TheSeven@rockbox/developer/TheSeven)
18:17:11 Join Curtman [0] (
18:17:36Luca_SFlynDice: with your patch, the player froze. No more panic, the scrolling indicator stopped at about 20%, the buttonlight is on and doesn't go off.
18:17:57Luca_Sreading from the uSD works fine, just writing froze.
18:24:16 Join Serathas [0] (
18:24:44Serathasis rockbox compatible with my 120 gig's ipod classic?
18:25:10Serathasthe black model
18:25:29Luca_Sgotta go now, I'll read the logs when I come back from a trip - probably only on monday. keep rockboxing :)
18:25:37 Quit Luca_S (Quit: CGI:IRC)
18:25:39Rob2222afaik the newes ipod model that can run RB is the 80GB video
18:26:27 Quit Serathas (Client Quit)
18:27:19FlynDiceRob2222: Sorry, didn't mean to abandon you with that, small crisis in the real world. I'm going to update myself now and see what i get.
18:27:45Rob2222FlynDice: no problem, just posted it as complete thing
18:28:13Rob2222i think i will get a reply in the forum, sooner or later ;)
18:30:57 Join MethoS- [0] (~clemens@
18:32:02 Nick fxb is now known as fxb__ (
18:33:44 Quit moparx (Ping timeout: 252 seconds)
18:36:07 Join domonoky1 [0] (
18:37:00 Quit domonoky (Ping timeout: 260 seconds)
18:44:10 Join phanboy4 [0] (
18:50:22 Join toffe82 [0] (~chatzilla@
18:50:41gevaertsI need a file setting and an on/off setting, which are to be passed on to firmware/. What do I need to do?
18:54:14 Join Peter93 [0] (
18:54:35 Quit Peter93 (Client Quit)
19:00:29bertrikhm, my c200v2's screen goes black just after booting
19:02:07 Join Boldfilter [0] (
19:03:25 Join mt [0] (~mtee@rockbox/developer/mt)
19:05:20funmanbertrik: what was the last revision working ?
19:11:12bertrikI don't know, but it was at least several months old
19:11:47 Quit TheSeven (Ping timeout: 240 seconds)
19:12:59bertrikit has files on it from april 6th, so that should help me bisect it I guess
19:15:37bertrikis there a simple way to know which svn revision was current on a specific date?
19:15:46 Join TheSeven [0] (~TheSeven@rockbox/developer/TheSeven)
19:16:36funmangit log, /Apr 6
19:17:03funmanr25499 is from april 6 and is related to c200v2
19:25:33FlynDiceRob2222: ping
19:26:44Rob2222FlynDice : ill give up with my fuze for noe
19:26:59Rob2222im not the only one with that problem
19:27:12FlynDiceRob2222: See my forum post
19:27:50 Join GeekShadow [0] (~Antoine@reactos/tester/GeekShadow)
19:28:06Rob2222already tried
19:28:08 Quit flydutch (Quit: /* empty */)
19:28:16Rob2222i come back to .31
19:28:20FlynDiceIf you can flash V02.03.31 back onto your player I think the firmware update routine will work
19:28:24Rob2222but its the same like with .33
19:29:04Rob2222i can flash orig .31 and orig .33, but nothing other. not from .31 away and not from .33 away
19:29:36Rob2222i already tried that way
19:29:45FlynDiceok, I just used V02.03.31A and it updates just fine, no A on yours I suppose?
19:30:16Rob2222means A the rockbox patech?
19:30:55FlynDiceno, my OF firmware reports V02.03.31A from the syatem menu
19:30:59 Join DataGhost [0] (
19:30:59 Quit DataGhost (Changing host)
19:30:59 Join DataGhost [0] (~dataghost@unaffiliated/dataghost)
19:31:07FlynDiceer system even
19:31:30Rob2222ah yeah, mine, too
19:31:45Rob2222have that on the player now again
19:32:39FlynDiceI just used it to run the firmware upgrade and it worked fine..... You say your fuzev2 is brand new?
19:33:05Rob2222it still wont accept a older or patched version that .31A
19:33:14Rob2222yeh, got it today from
19:33:44Rob2222purchased it for rockbox as alternative to my rockbox iriver H340@80
19:34:00Rob2222it cam with .31A and I flashed it to .33
19:34:05Rob2222maybe that was the error
19:34:06FlynDicebummer, I guess maybe Sandisk does care about us.. ;/
19:34:17bertrikfunman, ranma, is the button light also supposed to PWM on the c200v2?
19:34:24funmani don't think sandisk try to prevent us running rockbox
19:34:48funmanbertrik: i don't remember
19:34:57Rob2222then wy my stupid player doesnt flash somthing other than the .31 and .33 original :/
19:35:20funmanperhaps the flags we don't understand mean something special
19:36:01Rob2222at least i got a clip+ in the same shipment thats rockboxing fine
19:36:06 Join fyrestorm [0] (
19:38:08funmanRob2222: what's the md5sum of your patched fuze 2.3.33 OF (the one which doesn't work)
19:38:51funmanassuming you downloaded bootloader from
19:39:37Rob222265d84013a988c098cd23b2f45dd6311c *output.bin
19:39:47Rob2222yeah i did
19:41:19funmanok same here, i can't try update now since my fuzev2 is crashed and i can't power it off
19:41:37Rob2222what youre doing then?
19:41:47funmanwhat i'm doing?
19:41:48Rob2222waiting until the battery is empty?
19:41:55Rob2222OH MY GOD
19:42:05Rob2222thats bad
19:42:31funmanbreaking the fuzev2 was easy, now i'm trying to break my clipv1 ^^
19:42:58Rob2222does the low-battery power off works in hardware?
19:43:08Rob2222would be bad to deep-discharge the li-ion
19:43:12FlynDiceit's always more fun if explosives are involved... ;-)
19:43:56Rob2222what youve done to break it?
19:44:25funmannothing special .. i was modifying some lcd code
19:44:47funmani don't remember if it went like that when plugging the USB cable or somewhen else
19:45:37Rob2222you had that already?
19:45:41Rob2222or first time?
19:45:41 Join stoffel [0] (
19:46:03funmani was not a virgin of 'can't power off', what do you think !!
19:46:19Rob2222lol ok
19:46:25Rob2222so when thats normal ^^
19:48:04FlynDicegetting good battery life now aren't you though....
19:49:47Rob2222when it hangs it uses much power, or?
19:50:03Rob2222ok depends on the cpu state
19:50:25Rob2222and when it goes off, what do you do then? connect USB?
19:50:49FlynDicewith fingers crossed
19:51:16Rob2222thats hard developing
19:51:34Rob2222have you a possibility to unbrick it, yet?
19:51:58 Join fml [0] (
19:53:46*fml invites everyone to look at FS #11271 (new 29 pixels high font). It now contains latin letters and hence can be included into the official font pack IMO.
19:55:14Rob2222FlynDice: I just read in the Sansa forum that there is a reset metoth with 20-30 second holding power. But i think you know that, or?
19:57:39topikFlynDice and funman probably know more about sansa's then most sandisk people
19:57:45FlynDiceYes, we know that, it's funman's fuze
19:58:07Rob2222ok, i think asking is never wrong ^^
19:58:10 Quit mikroflops (Read error: Connection reset by peer)
19:58:28 Quit mt (Ping timeout: 258 seconds)
19:59:31FlynDicefunman's the genious here, I just tag along... ;)
20:00:04Rob2222hmm i have 14 days to return my fuze from today
20:00:22Rob2222hope we get something new regarding the firmware in that time
20:01:18 Join mt [0] (~mtee@rockbox/developer/mt)
20:01:22Rob2222atm i see 3 people with the non-flashing problem, from which 1 returned his unit and got a new one where the flash was working
20:01:35Rob2222but i think, thats jut another new check, too
20:01:53***Saving seen data "./dancer.seen"
20:03:32fmlHere is the link to the generated .fnt file:
20:03:54 Join mikroflops [0] (
20:12:29 Join pyther [0] (~pyther@unaffiliated/pyther)
20:13:00pytherIs it possible to power off my fushion while it is plugged into the charger/computer?
20:15:26FlynDiceyour fuze maybe?
20:15:49fmlI think it would be good to fix FS #11294 before the release since it's a core and often used feature
20:18:21pytherFlynDice: yes my fuze :P
20:19:13FlynDicedo you need it to power down or just make the display dark...?
20:19:54pytherFlynDice: well is it ok to keep it running for an extended period of time +24hrs?
20:20:40 Join wincent [0] (
20:21:04FlynDiceI don't think there's any problem with that, Charging shuts off when the battery is full.
20:22:06pytherHow do I update rockbox? Just replace the .rockbox dir?
20:22:34funmancharging stops when battery is full but begins again around 95%
20:22:53funmanpyther: yeah, or overwrite it if you want to keep settings
20:24:00pythercool that is easy :D
20:24:20pytherNow I just wish there was a really good theme :P
20:24:51 Join pelirrasta [0] (
20:25:05Rob2222FlynDice: The letter behind the firmware number is the regio setting. A=america F=europe and P=pacific(rest of world) ^^
20:25:05*bertrik almost bisected the problem with his c200v2
20:25:45pelirrastaHi there, can anyone help me out with a permissions problem on linux?
20:26:01gevaertskugel: How are things going? I haven't been very attentive the last week or so...
20:26:25*bertrik blames svn r26330
20:27:20*funman pretends not having seen that
20:27:57kugelgevaerts: slowly, currently :(
20:28:00funmanbertrik: what does ack ing before running do_sw_pwm() do ?
20:29:34gevaertspelirrasta: rockbox related?
20:30:20pelirrastayes, I want to install rockbox on a 30gb ipod video, when i try to run the installer it says "permission denied"
20:31:03topikrun the installer as root ?
20:31:16kugelfunman: that commit seems to change more than that
20:31:20gevaertspelirrasta: the graphical installer I assume?
20:31:28 Quit DerPapst (Quit: Leaving.)
20:31:48funmankugel: ah indeed i have removed a return
20:31:50bertrikmaybe the return shouldn't have been removed/
20:31:52pelirrastayes, graphical
20:32:05funmangood catch
20:32:16gevaertsdoes it say that when you launch the installer, or when you actually want to install?
20:32:40bluebrotheroh, wrong channel
20:32:41bluebrothercan users change task priority in FS?
20:32:44pelirrastawhen i try to install
20:32:53*bluebrother needs typing class
20:32:55kugelfunman: re your previous question: I think ack'ing the interrupt needs to be done at last otherwise the vic will leave the handler too early (at least that's what happened to me while doing the scrollwheel isr work)
20:33:05gevaertspelirrasta: ok, in that case, see topik's advice, and run it as root, or using sudo
20:33:36funmankugel: the handler should leave only when returning
20:33:41kugelthe scrollwheel func wasn't called when I ack'd the irq at the beginning
20:33:51kugelmoving it down fixed it
20:34:22 Join petur [0] (~petur@rockbox/developer/petur)
20:35:25pelirrastaI'm not really sure how to do that. I don't know what the command to open the installer is.
20:35:34pelirrastaI was using the graphical
20:35:50kugelmaybe it only applies if there're nested function calls
20:36:42bertrikwith just the return restored, it works again on my c200v2
20:37:35gevaertspelirrasta: using the command line it's easy. I have no idea how to do it from a GUI file manager, but I'd assume there might be an option in the context menu (i.e. when you right-click on it) to do that
20:38:45pelirrastaCan you tell me what the command is to open the rockbox installer? I tried "sudo RockboxUtility" from the command line in that folder.
20:39:29gevaerts"sudo ./RockboxUtility" probably
20:39:29bluebrotherpelirrasta: sudo ./RockboxUtility (if the binary is in your current folder)
20:39:34domonoky1pelirrasta: cd to the dir where the installer is, then sudo ./Rock<Tab> <enter> :-)
20:39:49bluebrotherRockTab? ;-)
20:39:52pelirrastammmm right... forgot the ./
20:40:16pelirrastathanks, i'm still a bit new to linux and trying to learn to navigate with terminal
20:40:44pelirrastasweet! appears to be working!
20:40:51*bluebrother gets annoyed by those "we need another bug because Rockbox Utility doesn't work with orca" whining
20:41:42 Quit mt (Read error: Connection reset by peer)
20:41:57 Quit watto (Read error: Connection reset by peer)
20:42:00 Join mt [0] (~mtee@rockbox/developer/mt)
20:42:14 Join watto [0] (~watto@
20:45:01domonoky1this whole accessibility thing is a mess.. :-/
20:48:08r0b-is rockbox usable on the clip +?
20:48:18Rob2222just installed it
20:49:53FlynDicefunman: It is indeed r26291 that breaks my uSD. I haven't tried to massage it any yet though.
20:50:04FlynDiceI'm out of time for now
20:50:48 Quit esperegu (Ping timeout: 264 seconds)
20:52:12Rob2222FlynDice: Do you think i have a chance for my FuzeV2 to flash another FW anywhen?
20:56:21funmanFlynDice: when you do, start by removing the 'be sure the card has finished programming' -> this is the only difference I can see, except that i removed the udelay i had added a few revisions before
20:57:49funmanRob2222: afer you reconnect the fuze, the firmware is still on disk or has it been removed?
20:57:56funman(when the upgrade fails)
20:58:02Rob2222its still there
20:59:03Rob2222im out of ideas, i tried everything, reste, factory default, and flash anything from .31 and from .33 but its accepting nothing other than .31 and .33 org files
20:59:09Rob2222between this 2 i can switch
20:59:25Rob2222but not between older FWs or RB-patched.ones
20:59:57Rob2222i tried deleting all files but the FW, too
21:00:10Rob2222but the player just recreates the files
21:00:12bertrikfunman, I'll commit the restored return in kernel-as3525.c, ok?
21:00:19funmani didn't try to update to older OFs
21:00:23funmanbertrik: sure
21:00:28 Join halmi [0] (
21:00:32funmansorry for the trouble :s
21:00:32 Quit fml (Quit: CGI:IRC)
21:01:10 Join Strife89 [0] (
21:02:19 Quit pelirrasta (Quit: CGI:IRC)
21:02:25Rob2222funman: i just wonder why it didnt even flahes the older orig FWs. its a brand new unit and im not the only one with that. counting 3 people haveing the same issue
21:02:37funmanask Sandisk
21:02:42 Quit Boldfilter (Quit: Boldfilter)
21:03:05Rob2222already posted in theie forum. but i dont think they will care about that one :(
21:03:28Rob2222but i can return the unit 14 days
21:04:34CIA-5New commit by 03bertrik (r26361): Restore return statement in kernel-as3525.c that was accidentally removed in r26330. This fixes c200v2 black screen.
21:06:23 Part watto
21:06:25 Join esperegu [0] (~quassel@
21:10:32 Quit mt (Ping timeout: 258 seconds)
21:14:29kugelbtw, what happened about the idea of ignoring short enums so that we can finally go with 4.4.x eabi gcc?
21:14:41 Join mt [0] (~mtee@rockbox/developer/mt)
21:14:44kugelor did jhMikeS volounteer to fix doom?
21:17:20gevaertsI'd say doom is doomed
21:17:51pytherIs it possible to clear the current playlist?
21:17:56topikis doom more than a gimmick on most devices?
21:19:04gevaertspyther: I don't think there is, but I can't really see why you'd want that
21:19:14 Quit stoffel (Ping timeout: 265 seconds)
21:19:44kugelgevaerts: nice pun but does it help towards a decision?
21:19:58kugelassuming you don't mean we should just get rid of it
21:20:06pythergevaerts: well what I would like is the ability to add a bunch of songs in shuffle mode without have toggle shuffle
21:20:13pytherI can insert songs shuffled
21:22:36bertrikThere are so many color targets with LCD invert now, that I'm tempted to add it for the c200/c200v2 too
21:23:03gevaertskugel: well, it doesn't really help of course. I'd say disable short enums
21:23:46gevaertsAnd switch in one or two weeks, just after the release
21:24:36bertrikdo we know what the exact problem is and how to fix it? is it just a question of someone doing it or do we need more research?
21:25:51 Quit wincent (Ping timeout: 264 seconds)
21:26:24funmankugel: we could decide to switch to eabi after release and hope someone fixes doom before the next release
21:26:55funmans/could/should/ even
21:27:11 Join Luca_S [0] (
21:27:15kugelunfortunately removing doom isn't an option, I'm rather annoyed by it
21:28:15funmannot removing it but leaving it broken until fixed
21:28:21bluebrotherremove doom! :)
21:28:23Luca_Sremoving no, but maybe just doing the switch would help fixing it in small steps
21:28:57kugelfunman: would be fine with me
21:29:07Luca_Si was wondering if the problems reported with flashing sansa firmwares has something to do with the changed model byte
21:29:09kugelI don't use it, I'm only annoyed by the time it takes to compile
21:29:11 Join ollebe [0] (
21:29:35funmanLuca_S: the 16 bits (not byte) changed in other models already
21:29:47Kevonehas anyone sucessfully install the rockbox on a 120GB iPod Video Classic
21:30:02bluebrotherKevone: Rockbox does not run on the Ipod Classic
21:30:11bluebrothersee the front page for a list of devices it runs on
21:31:15 Quit MethoS- (Remote host closed the connection)
21:32:01 Join Barahir_ [0] (
21:32:06Luca_Swell, maybe recentish OFs check that the value is in some table, while earlier FW did not check that. Of course, it might as well be that those bytes mean nothing, or something completely different. But this glitch may help checking the hypotesis... by flipping those bytes (and recomputing crcs, if any) in an older OF version
21:32:32Kevonesomeone told me that its possible
21:33:04bluebrotherit's possible to swap the hard drive on an supported Ipod to a bigger one and then run Rockbox
21:33:23bluebrotherhowever, the hardware in the newer Ipods is completely different.
21:33:31gevaertsKevone: then eithe someone was mistaken or lying, or someone has ported rockbox to the classic and not told us
21:33:58Rob2222bluebrother: you know whats the max size that fits in the ipod vide?
21:33:59bluebrotheris there even a way known to run custom code on the 6G Ipods? I'm not aware of any.
21:34:13bluebrotherRob2222: no, never had a video
21:34:15funmanLuca_S: 2.2.26 has the same value than 2.3.31 and 2.1.17 is different
21:34:23Kevoneso this model # a1238 aint possible
21:34:37Rob2222would like to have any rockbox polayer with more than 80gb
21:34:45bluebrotherRob2222: my guess is that the only problem would be to find a hard drive that fits physically and has the right connector
21:35:08Rob2222i thought maybe the classic has the same drive as you mentioned swap drives
21:35:09 Quit Barahir (Ping timeout: 248 seconds)
21:35:13bluebrotherKevone: see the apple website on the exact player model. Anything newer than a 5.5G is not supported
21:35:23funman14h30 on clip+, 16h on clipv2
21:35:28bluebrotherRob2222: possible, but I don't know. I only swapped the drive in my mini :)
21:35:54Kevoneits not a big deal
21:35:58gevaertsRob2222: people use rockbox on ipod videos with 240GB drives
21:36:47Rob2222are the 240gig drived taken from classics?
21:36:54CIA-5New commit by 03bertrik (r26362): Apply FS #9194 - LCD invert for Sansa c200/c200v2
21:37:03gevaertsno idea
21:37:18gevaertsProbably not. They might be the same model though
21:37:40gevaertsI'm sure there are some forum posts with model numbers
21:39:48Rob2222hmm well
21:40:34Rob2222maybe i had should go this way (ipod video+hd upgrade) than the fuze. really depressing atm, purchased a fuzev2 for RB and now that thing wont flash the patched firmware :(
21:41:06Luca_SRob2222: think about it as a challenge or a chance for research :D
21:41:47Rob2222im no dev and what can i do without speanding month in getting int that stuff
21:46:27kugelfunman: the v10 freq scaling patch has been very stable for me so far
21:47:41 Quit Luca_S (Quit: CGI:IRC (EOF))
21:48:38funmani'm still unsure if it should be committed since it doesnt' work on the clips
21:49:07 Join stoffel [0] (
21:51:04kugelfunman: 24MHz-clips or all clips?
21:52:24Kevonetheres no such thing as a 240gb. Someone had to custom make iut
21:52:33funmanclipv2/+, but the patch which uses 40MHz doesn't work either
21:53:48kugelfunman: strange :(
21:54:15funmanif i can get one of my clips soldered to some wires at DevCon perhaps i can udnerstand what happens
21:55:02kugelthe fuzev2 (with freq scaling) is more stable and has longer battery runtime than the fuzev1 currently
21:55:10Rob2222btw, is there a button to return from filelist to WPS without selecting a new song (@Clip+)?
21:55:58Rob2222according the manual ist "play" but this just goes "up"
21:56:47KevoneI got a suggestion, buy an iPod touch
21:57:09Rob2222for me?
21:58:05Kevoneyeah, the custom firmware is so much better than any of the older ipod touchs.
21:58:05funmankugel: unless you use class 6 µSD or wake up alarm :'(
21:58:41gevaertsKevone: the ipod touch isn't in any way related to rockbox, and therefore off-topic
21:59:07Kevonewell, I am just giving suggestions
21:59:22 Join Boldfilter [0] (
21:59:32Rob2222Kevone: god behave Im using RB since iriver H140, last years on a H3"80". would never buy a player thats not rockbox capable. i just purchased a fuze2 and a clip+ to have somthing smaller than my H380
21:59:42 Part Boldfilter
21:59:59 Quit mt (Read error: Connection reset by peer)
22:00:05Rob2222Kevone: yeah, thx for that but im a oldscholl RB user :D
22:00:13Rob2222doesnt want something other
22:00:54Rob2222but a RBed ipod Video with 240GBHD would be interesting... even if i dont like the wheel thingies
22:01:39Lloreangevaerts, Rob2222: Hasn't someone done the 240gb thing in a Gigabeat too now?
22:01:54***Saving seen data "./dancer.seen"
22:02:06gevaertssomeone was going to try...
22:02:12Rob2222hmm, i dont know. if the gigabeat has zif socket, too, it could be
22:02:32Rob2222a pity the H340 has a PATA connector :(
22:02:42Rob2222i think 120GB was max aviable there
22:02:50KevoneThat ipod would be Huge, 240gb, lol
22:03:05Kevonebut badass
22:03:29Rob2222you can buy that 240gb drives for the ipod video
22:03:53Rob2222theoretical 110 euro, but not deliverable, some shops have them in stock but want 200 euro
22:04:31Rob2222I upgraded my H340 to 80GB as the first 60GB ipods came out. :D
22:04:53Kevoneis it the same size as the 120 hdd, so it would fit in its case
22:05:18Rob2222yeah, 120GB would be the biggest aviable with PATA interface for the H340
22:05:50Kevonewell, i mean for an ipod
22:05:54Rob2222but im really enthusiastic about the small size of the clip+ and its RB capable :)
22:06:41Kevonethe width and height would be the same, for an 240gb, as a 120Gb
22:08:30Kevoneone more question, why is the 160gb bigger than the 120gb
22:08:48Rob2222they have 1 and 2 platters inside
22:09:00Rob2222so they have a different height
22:09:29Rob2222the 240gb model has 2 platter/discs
22:09:44Rob2222so it fits in all unit that already have a disc with 2 plattern in it.
22:09:54 Join robin0800 [0] (~quassel@
22:09:56Rob2222for example the ipod video 80gb
22:10:30*gevaerts doesn't think that the reasons for disk manufacturers to choose for 5mm or 8mm are not on-topic here
22:10:47*gevaerts *does* think that, sorry
22:11:30KevoneThanks for the info Rob2222, I am going to think about the 240GB. Anymore i should know about before i decide to by it
22:12:02Rob2222just make sure it fits
22:12:18Rob2222for example ipod video 80gb has 2 platter discs
22:12:33Rob2222but later there were 80gb with 1 platter aviable, too
22:13:03Kevonenot trying to act dumb but what are platters
22:13:25Rob2222platter is another name for disc
22:13:37Kevoneok makes sense
22:13:43Rob2222the discs that are IN the harddiscdrive
22:14:19Rob2222 <= 2 platter
22:14:52Kevonewell, i am going to do a lot of research on the internet and see if there is anyone that successfully added that kind of size on a 120gb ipod
22:15:22Rob2222you have a ipod classic?
22:15:39 Join DerPapst [0] (
22:16:06Kevoneyeah, a Black 120 GB iPod video classic
22:17:05 Quit ollebe (Quit: Leaving)
22:18:07 Quit robin0800 (Read error: Connection reset by peer)
22:18:07Rob2222iups thats german
22:19:26krazykitKevone, there's no such thing as an "ipod video classic". since it's 120gb, and presuming that it was that way from apple, it's a Classic, and thus you can't run rockbox on it
22:19:43Rob2222it looks like the 120gb has a 1 platter drive cause its thinner than the first generation 160gb ipod classic
22:20:06Rob2222krazykit: i think he just thinks about HD upgrade
22:20:15gevaertsLook, this really isn't a general DAP support channel
22:20:45Kevonecome on we are talking about ipods. Rockbox has to do with ipods.
22:21:07Rob2222gevaerts: can somebody the help my rockbocing my damn fuze2 :(
22:21:26Rob2222ok im quiet
22:21:44LloreanKevone: Rockbox is software that happens to run on many hardwares.
22:21:53gevaertsKevone: rockbox is not at all related to wich drives work in an ipod classic
22:26:03Rob2222but if someone has any new idea how to flash the fuze2 RB-patched-firmware, just PM me, im happy to test
22:26:29gevaertsRob2222: that would be on-topic, no need for a PM there :)
22:31:35LloreanIf the version check is removed, will the patcher just work with newer OFs, or does each OF require some analysis first?
22:35:05 Join robin0800 [0] (~quassel@
22:35:24 Quit fyrestorm (Quit: lamers envy me like they envy bill g -- main boot xp, just the way it should be!)
22:40:44funmanLlorean: you mean mkamsboot md5sum list, or (apparently new) Sandisk version check ?
22:41:33Lloreanfunman: md5sum list?
22:41:37LloreanER, not a question
22:42:02LloreanCan the patcher be made to patch any firmware version with a −−force flag or similar (for brave souls who don't wish to wait for the next updated md5sum list)
22:42:10funmanif it's removed, nothing prevents you from patching a c200v2 OF with clip+ bootloader/code => when you put this on a c200v2, BOOM
22:42:24funmanyeah a −−force flag is possible
22:42:27funmanbut it's already there
22:42:38funmanit's just called a 'get a build environment, fetch SVN and modify the code' flag
22:42:52funmannot too strong i think
22:43:19LloreanThat takes an awful lot of time and/or space for someone who's willing to test.
22:43:47LloreanBut yeah, might as well keep it safe
22:43:49Rob2222what should that bring?
22:43:55funmanyeah but given numerous messages on forums from people who can't install properly and try numerous weird things, i think it's better
22:44:16funmanLlorean: if *you* want to test a new version, just ask :)
22:44:24Rob2222i think the risk would be to high to brick the player
22:44:30Lloreanfunman: If *I* want to test a new version, I'll just compile. :)
22:44:34 Quit sevard (Changing host)
22:44:34 Join sevard [0] (sev@unaffiliated/sevard)
22:45:11funmanit's especially plausible that people would install fuzev1 bootcode on fuzev2 and vice versa
22:45:14Lloreanfunman: Could we prevent patching by still requiring the other checks? For example, only patch the bootloader type of a player physically attached to the computer?
22:45:25LloreanCan we differentiate v1 and v2 by something other than the firmware?
22:45:32Rob2222hmm, i would investigate the sandisk version problem, but i have no idea where to start
22:45:35funmanUSB id works yes
22:45:52funmanbut if you give a fuzev1 OF we can't differentiate it from fuzev2 OF
22:46:06LloreanSo "force" could ignore the MD5sum but only patch a bootloader for the physically attached player?
22:46:37LloreanSo that way even if you get the wrong OF, you have the right bootloader and can get into USB mode? (is that safe, or can USB mode + wrong OF still leave you with an unrecoverable player?)
22:46:41funmanonly rbutil checks USB ids
22:47:04funmanif you have the wrong OF i suppose the player wouldn't accept it
22:48:00 Quit Akranis ()
22:48:25LloreanAh well, it's not too important. Soon enough they'll stop coming out with new OFs anyway
22:50:25 Join Boldfilter [0] (
22:50:49kugelhm, alarmclock.rock is broken
22:53:04 Part Boldfilter
22:57:06 Quit robin0800 (Quit: No Ping reply in 180 seconds.)
22:57:33 Join robin0800 [0] (~quassel@
22:58:11 Quit petur (Quit: Zzzz)
23:00:34pyther. ../albumtitle.{jpeg,jpg,bmp} is album title the title in the id3 tags?
23:03:36 Quit esperegu (Read error: Connection reset by peer)
23:04:35 Join esperegu [0] (~quassel@
23:04:37 Join komputes_ubuntu [0] (~komputes@ubuntu/member/komputes)
23:05:47 Quit stoffel (Remote host closed the connection)
23:08:54CIA-5New commit by 03kugel (r26363): Fix typo that made alarmclock exit immediately and add scrollwheel actions.
23:10:27 Quit mapi_ (Quit: Tiarra 0.1+svn-33938: SIGINT received; exit)
23:12:25CIA-5New commit by 03kugel (r26364): Backport 26363: ...
23:14:48FlynDicefunman: Removing the 'be sure the card has finished programming' code returns correct function to uSD writes for me. You want to test it or are you ok with the code removed?
23:15:48funmanwell i know it will break somewhen else
23:16:30funmanhm OTOH it was before I apply the timeout diff to the v2 code
23:17:29 Quit jhMikeS ()
23:17:34 Join dantje_ [0] (
23:19:10funmanFlynDice: did you try heavy loading it with playback from internal while database updates from µSD ?
23:20:04FlynDicenope, I'll do that now though. I did write & verify and copied multiple directories from internal -> uSD
23:23:25FlynDiceThat went just fine
23:24:20 Quit dantje_ (Quit: Ex-Chat)
23:27:43 Quit hebz0rl (Ping timeout: 240 seconds)
23:29:19 Quit FlynDice (Remote host closed the connection)
23:30:31 Join wincent [0] (
23:33:00 Join FlynDice [0] (
23:33:03 Quit nima (Ping timeout: 264 seconds)
23:35:32funmanjust commit :)
23:36:20funmanwow building clipv1 with thumb gives 84.5kB audio buffer
23:37:04 Quit ender` (Quit: I spilled Spot Remover on my dog... Now he's gone.)
23:37:24FlynDiceand wait for the screams!...
23:37:36funman85.2kB binsize -> a bit more stack usage
23:37:43FlynDiceI think it should be fine
23:38:21*FlynDice knows nothing about bin size, was commenting on sd ;)
23:38:24funmanand apps/codecs/nsf.c causes an ICE :)
23:40:26funmani suppose the plugins also are smaller
23:42:08Rob2222can i dump the fuze2 firmware partition with sansapatcher (without having a RB BL) ?
23:42:21Rob2222a pity
23:42:44funmanyou must unsolder the NAND or solder a JTAG adaptor if it's possible
23:43:05Rob2222ugh :(
23:43:25Rob2222i think before i do that ill return the player
23:43:45Rob2222but then is the question, how to get a RB capable fuze
23:44:18 Quit CGL (Quit:
23:44:38funmanif you find the answer, please tell us, because we don't know
23:44:58Rob2222yeah im collection serial numbers atm
23:45:22Rob2222btw have you any idea what i could make to get a first step furher regarding the flash issue?
23:46:00Rob2222btw is there a "main"-dev for the sansa BL?
23:46:50funmanfirst step would be closing the irc window because i don't see how other people can help you
23:47:00 Quit Llorean (Quit: Leaving.)
23:47:40CIA-5New commit by 03FlynDice (r26365): sd-as3525.c: Do not wait for TRAN state after transfer is over. ...
23:48:01 Quit pamaury (Quit: exit(rand());)
23:48:53 Quit bluebrother (Disconnected by services)
23:48:56 Join bluebroth3r [0] (~dom@rockbox/developer/bluebrother)
23:49:20 Quit robin0800 (Remote host closed the connection)
23:49:47Rob2222now thats a really productive suggestion
23:50:12 Join XStatik [0] (
23:51:05 Nick bluebroth3r is now known as bluebrother (~dom@rockbox/developer/bluebrother)
23:55:33mc2739FlynDice: your last commit should probably be backported to the v3_6 branch since it affects release targets
23:56:09funman < build with thumb
23:56:38FlynDicemc2739: Probably but why don't we make sure there are no complaints for a day or so ;-).
23:57:17 Quit Xerion (Ping timeout: 265 seconds)
23:57:57 Quit esperegu (Ping timeout: 260 seconds)

Previous day | Next day