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

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

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

#rockbox log for 2011-11-27

00:01:17 Join Rondom [0] (~rondom@2a01:488:66:1000:b24d:4f2f:0:1)
00:02:21 Quit Stummi (Quit: Bye!)
00:03:37bertrikpamaury, what do you think of funman's alignment fix for amsv2 usb?
00:04:03bertrikit's probably not going to make things worse ...
00:04:05pamauryit is probably necessary for the amsv2 too, I posted a patch on FS
00:04:16pamauryit can't harm to commit it
00:05:10*gevaerts will investigate the x5 thing tomorrow
00:06:07bertrikpamaury, I can apply, test if it still works the same and commit it, ok?
00:08:01 Quit Zarggg (Quit: Rebooting client...)
00:10:58 Quit tjb0607 (Read error: No route to host)
00:12:47 Join tjb0607 [0] (
00:16:29 Join Zarggg [0] (
00:19:18 Quit TheLemonMan (Quit: Ex-Chat)
00:22:05dreamlayersThe handle updating in buflib_buffer_shift is obviously broken, right? It updates handle->alloc as if it's union buflib_data *, but it's actually char *.
00:24:20bertrikUSB seems to be working no worse than before (on the clip zip), so I'll commit that change in the next few minutes
00:27:02bertrikWhy do we need 64 bytes in the control request buffer anyway?
00:27:19bertrikbecause DMA may write a bigger packet?
00:29:11pamaurybertrik: iirc, because it's the maximum packet length, so the dma could write up to that amount
00:29:33bertrikbut I thought packets could be bigger for high-speed USB
00:29:47pamaurys/packet size/transfer size
00:30:15gevaertsnot for control
00:30:19bertrikor is that only for non-control packets?
00:30:22bertrikah, ok
00:30:33*gevaerts checks
00:31:23pamaurybut if I recall correctly, the as3525v2 driver handles setup packets differently and will never allow the dma to write more than 8 bytes when pointing it to the setup packet buffer
00:32:35pamauryso actually, if the dma does it's job right, this patch is not needed with the as3525v2 driver (but is with the s3c...)
00:36:26 Quit ender` (Quit: Anyone who cannot cope with mathematics is not fully human. At best, he is a tolerable subhuman who has learned to wear his shoes, bathe, and not make messes in the house. -- Robert A. Heinlein)
00:43:23bertrikpamaury, ok, but I think I'll commit it anyway
00:44:11 Join dv_ [0] (
00:45:22funmanyou could put a mark just after those 8 bytes and check for it, panic if the mark has been changed
00:50:32 Quit Keripo (Quit: Leaving.)
00:55:51CIA-22New commit by bertrik (r31063): Sansa AMSv2: potential USB fix from FS #11664 by funman: ...
00:57:41CIA-22r31063 build result: 2 errors, 8 warnings (bertrik committed)
00:59:16 Quit Llorean (Ping timeout: 240 seconds)
01:00:28 Join Llorean [0] (~DarkkOne@rockbox/user/Llorean)
01:02:18 Join liar [0] (
01:03:26funmanbertrik: is that the 'aligned' ?
01:03:38bertrikI don't know exactly to be honest
01:05:05***Saving seen data "./dancer.seen"
01:05:21saratogafunman: what were you asking me about trying USB with a ramdisk?
01:05:44funman} _ep0_setup_pkt __attribute__((aligned(32))); __attribute__((aligned(32)));
01:06:00funmanUSB_DEVBSS_ATTR already contains the aligned, and a ';' ?!
01:07:07CIA-22New commit by funman (r31064): AMSv2: remove stray ; from USB_DEVBSS_ATTR
01:07:09funmanbeware: might contain red
01:07:32funmansaratoga: i wondered if ramdisk over usb worked flawlessly on amsv2
01:08:47CIA-22r31064 build result: 2 errors, 0 warnings (funman committed)
01:09:23funmanbertrik: the aligned(32) is superfluous
01:10:43bertrikok, I'll remove it. I can apparently also simplify the struct a bit, using a union { struct, byte[]} instead of struct { union {struct, byte[]}}
01:11:48funmanunnamed unions are gcc extensions also, your method would be more standard
01:11:56 Quit gbl08ma (Quit: Saindo)
01:16:15 Part metaphys
01:16:16 Nick kugelp is now known as kugel (~kugel@rockbox/developer/kugel)
01:22:13CIA-22New commit by bertrik (r31065): Sansa AMSv2 USB: Simplify union around struct usb_ctrlrequest and remove duplicate alignment attribute
01:23:50CIA-22r31065 build result: 2 errors, 0 warnings (bertrik committed)
01:25:14 Quit lebellium (Quit: ChatZilla 0.9.87 [Firefox 9.0/20111122192043])
01:26:03 Quit perfectdrug (Quit: CGI:IRC)
01:35:15JdGordonPoodlemastah: which backdrop image is loaded when you tinhk it shldnt be? the one set in the setting right?
01:35:18 Quit Thra11 (Quit: kthxbai)
01:37:17PoodlemastahActually, cabbiev2.bmp is ALWAYS buflib'd at least once.
01:38:07JdGordonwhat is the ufll patchh?
01:38:24JdGordonarg, what is the full path? (the buflib a.lloations page will tell you)
01:38:52PoodlemastahIf I have another backdrop, it is also loaded. This even works for cabbiev2, I currently have two cabbiev2 entries.
01:39:05JdGordoninteresting :)
01:39:26JdGordonopen a bug, and attach your config.cfg and and theme configs and skins you use
01:39:35JdGordonI wont be able to look at this till later tonight
01:40:04JdGordonbackdrop: - in the config should be enough to sotp that
01:40:58PoodlemastahWhen I mean always, I meant always, even with backdrop: - in config, there is still one cabbiev2 entry.
01:41:03PoodlemastahWill post the bugreport now.
01:49:09 Quit tjb0607 (Read error: Connection reset by peer)
01:49:50 Quit markun (Ping timeout: 244 seconds)
01:50:26 Join tjb0607 [0] (
01:54:02*dreamlayers is confused by buflib_compact and trying to figure out what's broken and what isn't.
01:58:10 Quit liar (Remote host closed the connection)
02:00:56 Quit pamaury (Remote host closed the connection)
02:18:14 Join liar [0] (
02:23:13 Join Horscht [0] (
02:23:13 Quit Horscht (Changing host)
02:23:13 Join Horscht [0] (~Horscht@xbmc/user/horscht)
02:40:49CIA-22New commit by dreamlayers (r31066): In buflib_buffer_shift(), properly update handle->alloc values, which are char * and not union buflib data *
02:41:57 Join Louis126 [0] (4898d3ab@gateway/web/freenode/ip.
02:42:40CIA-22r31066 build result: 2 errors, 0 warnings (dreamlayers committed)
02:43:12saratogadreamlayers: what did that bug affect?
02:44:34Louis126hello rockbox support?
02:45:00saratogaLouis126: if you have a question, just ask it
02:45:14Louis126ok, thanks
02:45:31dreamlayerssaratoga: Nothing, because only pictureflow uses that function and it doesn't use allocations when the values are wrong. It moves the buffer back before using the allocations.
02:45:48saratogaah ok, was wondering how that was never noticed :)
02:46:16Louis126I don't understand what this instruction means: Open a new terminal, and navigate to the “rbinstall” folder. 3. Run mkamsboot, passing the name of the OF and bootloader files that you downloaded, in addition to a name for the output file.
02:46:45saratogaLouis126: if you're just trying to install rockbox, you don't need to do that
02:47:14Louis126I have been trying to install it for the last 3 hours
02:47:51Louis126I have a folder on my desktop labled rbinstall
02:48:07Louis126i have the rockbox utility
02:48:23saratogayou should use rockbox utility
02:48:30saratogadid you see my post on the forums?
02:48:53Louis126it keeps giving me erros when I try to download the bootloader
02:49:04Louis126yes i did
02:49:25saratogadid you try that?
02:50:46saratogaalso I saw you were using the Linux/MacOS instructions, which of those two are you running?
02:51:12Louis126no, i have windows vista
02:51:26saratogaheh, well then maybe the windows instructions would work better :)
02:51:45Louis126yeah....i'll go back and check your response.....thanks
02:55:04Louis126i did go back and select the appropriate drive letter for the clip+, and ran complete installation.....i keep receieving errors when I do this
02:55:42saratogawhich error?
02:55:59Louis126should i disconnec the device and see if the installation took will i know when the installation of rockbox has actually taken place on the device?
03:03:03 Join ReimuHak_ [0] (
03:05:06***Saving seen data "./dancer.seen"
03:06:28 Quit bertrik (Ping timeout: 240 seconds)
03:07:01 Quit saratoga (Changing host)
03:07:01 Join saratoga [0] (9803c31c@rockbox/developer/saratoga)
03:10:48 Quit saratoga (Quit: Page closed)
03:13:05dreamlayersDoes firmware/buffer.c have any purpose anymore? Nothing uses bufalloc anymore, and it's broken anyways, because allocations would overlap with memory used by core_alloc.
03:19:35dreamlayersI mean buffer_alloc. firmware/buffer.c seems to be a remnant from before, and only the part supporting core_allocator_init() seems to be used.
03:22:43Louis126the rb utility keeps informing me that there is a "new" version of rb.....but when I try to download it, I get the error that I cannot open the zip file
03:22:52Louis126the whole thing is just a vicious cycle
03:23:11funmanwhich version do you have? (and which OS)
03:23:46Louis126v 1.2.10
03:23:53Louis126windows vista home basic
03:25:13 Join saratoga [0] (9803c31c@gateway/web/freenode/ip.
03:25:28 Quit dreamlayers (Quit: Bye)
03:27:04Louis126is 1.2.10 the most up to date version of the rb utility?
03:29:14saratogano its one version out of date, but it shouldn't matter for your player
03:31:44saratogawell i guess it does add support for patching the newest clip+ firmware, but thats not too important since the older one works just as well
03:33:33Louis126ok, i'll try the older one instead
03:36:46funmani think 1.2.10 update was broken yes
03:38:49 Quit liar (Remote host closed the connection)
03:42:54Louis126i just ran the rb utility, using the install rb to my device option, and now having safely removed the clip+, do NOT see rb installed on the device.....
03:43:28Louis126what am i doing wrong
03:44:25saratogaexplain what you did in more detail and maybe someone will know
03:47:43Louis126i have my clip+ selected (letter E) using rb Utility.....then used the "Install RB on your player" option......I chose the current build, and backed it up first. Then it started supposedly installing it on my device −−−− i even noticed the WRITING option going on, on the clip+ for a brief time....the installation ended....I safely removed the clip+ from my laptop, turned it off and then on again, and see NO evicence of RB being in
03:48:05saratogadid you provide it with a firmware file?
03:48:25Louis126i still don't know how to do that
03:48:41saratogadid it prompt you to do so?
03:48:55Louis126after which step would it have done so?
03:49:15saratogawhen you select install
03:50:07Louis126i don't think so.....are you saying it should do so after selecting COMPLETE INSTALLATION?
03:50:11Louis126I can try it again
03:50:55saratogaif you didn't get a prompt you probably hit the wrong option
03:51:19Louis126i'll try this again
03:52:40saratogaoh reading more carefully you said above that you just installed the main build
03:52:41Louis126yeah, its asking for a bin file
03:52:45saratogahit "complete installation"
03:52:47Louis126i don't know where to find that
03:52:52saratogait gives you a link, read that
03:53:23saratogaand if you can't do that, just stick the Sandisk firmware :)
03:53:49Louis126it gives 3 links.....which one are you referring to
03:55:01saratogaread the text ;)
04:03:18Louis126will the sansa forum reading material explain to me where to locate the bin file, or will the manual, or will the sansaAMS?
04:20:53 Quit amiconn (Disconnected by services)
04:20:54 Join amiconn_ [0] (quassel@rockbox/developer/amiconn)
04:21:16 Nick amiconn_ is now known as amiconn (quassel@rockbox/developer/amiconn)
04:21:52 Quit pixelma (Disconnected by services)
04:21:54 Join pixelma_ [0] (quassel@rockbox/staff/pixelma)
04:21:56 Nick pixelma_ is now known as pixelma (quassel@rockbox/staff/pixelma)
04:23:10 Join Horschti [0] (~Horscht@xbmc/user/horscht)
04:26:40 Quit Louis126 (Quit: Page closed)
04:26:43 Quit Horscht (Ping timeout: 245 seconds)
04:28:32 Quit TheSeven (Disconnected by services)
04:28:44 Join [7] [0] (~TheSeven@rockbox/developer/TheSeven)
04:42:49 Quit MethoS- (Remote host closed the connection)
04:53:15 Join dreamlayers [0] (~bgjenero@rockbox/developer/dreamlayers)
05:05:08***Saving seen data "./dancer.seen"
05:21:58 Join nosa [0] (
05:24:26 Quit nosa-j (Ping timeout: 244 seconds)
05:24:27 Nick nosa is now known as nosa-j (
05:55:56 Join Rob2222 [0] (
05:56:49CIA-22New commit by dreamlayers (r31067): Add STORAGE_INIT_ATTR to static ata.c functions that are only called via ata_init().
05:58:36CIA-22r31067 build result: 2 errors, 0 warnings (dreamlayers committed)
05:59:48 Quit Rob2223 (Ping timeout: 258 seconds)
06:08:04 Join JdGord [0] (~AndChat@
06:32:41 Quit saratoga (Quit: Page closed)
06:41:24 Join Keripo [0] (
06:50:55 Quit Keripo (Read error: Connection reset by peer)
06:54:53 Join Keripo [0] (
06:56:08 Quit dreamlayers (Quit: Goodnight)
07:05:10***Saving seen data "./dancer.seen"
07:20:42 Join doomedtx [0] (
07:22:30doomedtxnoob ?: just pulled out my Archos Jukebox 20 after many years of not using it. It has flashed version 2.5 on it. I ran the installer and it installed version 3.91 to the device, but when I restart it I still see 2.5. The new abrec file in root gives me a ROLO file length mismatch error. What's the next step?
07:24:55 Quit Keripo (Read error: Connection reset by peer)
07:29:31 Join Keripo [0] (
07:33:56 Join saratoga [0] (9803c31c@gateway/web/freenode/ip.
07:34:32saratogadoomedtx: did you originally flash rockbox to the players ROM rather then install it on the hard disk?
07:35:03doomedtxas far as I remember, seems like that was the preferred method in 2003 when i did this (judging by timestamps)
07:35:52saratogathe archos was many years before my time, but i suspect our installer is just installing a build to the local disk
07:36:07saratogai'm not even sure if its still possible to flash modern versions of rockbox
07:36:50doomedtxthat's what it looks like...i'm reading the manual which has info about flashing...hopefully using that and the original firmware from archos I can undo this...otherwise stuck at 2.5 I guess
07:37:14saratogaregarding updating, this might interest you:,29203.0.html
07:40:33saratogaFWIW we really should have forked archos off a while ago rather then slowly squeezing the binary out of the archos's limited memory
07:40:55doomedtxI'm looking through there now. Perhaps I will leave it at 2.5 since there's not much to be gained that's not already available. I d/led the firmware from Archos and am able to run the original firmware, but Rockbox still owns the device. It seems like way back then there was an option to hold down a button to force the stock loader...will have to play around a bit
07:41:37saratogasome of the early 3.x builds might be good as well
07:41:42saratoganot sure, i don't have that player
07:42:00saratogayou could ask in this channel during the day in Europe and maybe find someone who knows more then i do
07:42:03doomedtxI'm getting ready to sell the device and thought the buyer might appreciate the newest version, but they will have to appreciate the old version if I can't figure this out soon
07:42:35saratogaselling a 10 year old mp3 player :)
07:43:32doomedtxYou'd be surprised...several sold recently on eBay for more than I'm asking, and mine was upgraded to a whopping 60GB by replacing the internal HD. That was a ton of space in the early 2000s!
07:46:11doomedtxIn any event, I think it's not worth the time to figure out how to re-flash the old rom. I seem to remember either manually creating a ucl file out of the firmware I wanted to use or something like that. The new owner can mess with it if they want. Thanks fir your time
07:47:23 Quit doomedtx (Quit: CGI:IRC)
08:00:51 Join jdgord_ [0] (~AndChat@
08:01:48CIA-22New commit by jethead71 (r31068): PortalPlayer: Loading of stack munge value into r4 was moved in r30741 to after the idle stack init loop, breaking the usage readout.
08:03:22CIA-22r31068 build result: 2 errors, 0 warnings (jethead71 committed)
08:04:35 Quit JdGord (Ping timeout: 260 seconds)
08:17:52 Join JdGord [0] (~AndChat@
08:17:53 Quit jdgord_ (Read error: Connection reset by peer)
08:33:02 Join telliott [0] (
08:34:40telliottHow do I get rid of the icons in the file browser. I'm editing a skin and couldn't find the answer.
08:36:12 Quit JdGord (Quit: Bye)
08:37:19 Quit knittl (Ping timeout: 248 seconds)
08:40:18 Quit telliott (Quit: Leaving)
08:43:43 Join knittl [0] (
08:43:44 Quit knittl (Changing host)
08:43:44 Join knittl [0] (~knittl@unaffiliated/knittl)
08:52:08 Quit knittl (Ping timeout: 244 seconds)
09:04:59 Join nickp [0] (
09:04:59 Quit nickp (Client Quit)
09:05:08 Join nick-p [0] (
09:05:12***Saving seen data "./dancer.seen"
09:05:53nick-pshow icons: off (in .rockbox/themes/theme_name.cfg)
09:06:33 Quit nick-p (Client Quit)
09:07:27 Join knittl [0] (
09:07:27 Quit knittl (Changing host)
09:07:27 Join knittl [0] (~knittl@unaffiliated/knittl)
09:19:49 Join bertrik [0] (
09:19:50 Quit bertrik (Changing host)
09:19:50 Join bertrik [0] (~bertrik@rockbox/developer/bertrik)
09:21:52 Quit saratoga (Quit: Page closed)
10:11:39 Quit Bagder (Ping timeout: 276 seconds)
10:33:45 Join Bagder [0] (
10:33:46 Quit Bagder (Changing host)
10:33:46 Join Bagder [241] (~daniel@rockbox/developer/bagder)
10:45:07 Join ender` [0] (
10:49:03 Join lebellium [0] (
10:49:15JdGordonlebellium: hi
10:50:25JdGordonSo, I did the "ignore background image" thing but its not in svn yet, i got sidetracked with something much coolerr
10:50:43lebelliumreally? what? /D
10:51:14JdGordonlayering in the skins, so you can draw onto the background, so you could have different images as different viewports backgrounds
10:51:35JdGordonthe use case im dealing with is putting the AA as the background image, but it doesnt bloody work :/
10:51:58lebelliumyou wanted that for the Sansa Zip?
10:52:04 Join Xerion [0] (
10:52:17JdGordonI havnt got a zip :)
10:53:06lebelliumah ok because in the OF the album art is in background and I guess dfkt and some others would like to get the AA as background pic
10:53:23JdGordonthats been a general wishlist tihng for ages
10:53:45JdGordonbut then using this means you could draw a white rectangle for some viewport on the top layer
10:54:02lebelliumanyways, I'm very happy you work on these things. Finally some improvements in the skins engine :D
10:55:14lebelliumthe screenshot looks promising
10:55:34JdGordonthere is a better one somewhere else, that grey line is a bug i havnt been able to track down
10:55:57lebelliumI noticed the scrolling text often causes some display bugs
10:57:00lebelliumBTW, what players do you own? A Gigabeat?
10:57:04JdGordonwhat sort of display bugs?
10:57:22JdGordonI've got a dozen players (see my forum sig :p ) my main one atm is ipod video
10:58:12lebelliumyour forum sig is "Using PMs to annoy devs about bugs/patches is not a good way to have the issue looked at." lol
10:59:39lebelliumI mean, my R0 theme works well on the R0 but not in the Toshiba Gigabeat UI simulator, I don't know why. So if you own a Gigabeat, maybe you could try my theme on the device
11:00:07JdGordonI dont, that screenshot is the beat-f because its the same screen size
11:00:35lebelliumah ok
11:00:49 Join y4n [0] (y4n@unaffiliated/y4ndexx)
11:00:51lebelliumand any clue why that occurs?
11:01:05lebelliumis it because of my complex and ugly code? lol
11:01:09JdGordonR0 is touchscreen?
11:02:21lebelliumNo, the R0 only has tactile buttons like the Gigabeat and it worked in the UI simulator for my previous theme versions
11:02:54JdGordonnot sure, havnt looked into your skin much, I just remember you have alot of code which can be shrunk
11:03:25 Join liar [0] (
11:03:34lebelliumbut why does it work on a device and not in the UI simulator. Isn't it the same skins engine?
11:03:48 Quit Xerion (Read error: Operation timed out)
11:03:58JdGordonit is, depends on what the exact issue is
11:04:05JdGordonvery much sounds like a bug I need to fix
11:04:49lebelliumWell I guess the simulator doesn't like when I use too many subimages for the battery indicator lol
11:05:13***Saving seen data "./dancer.seen"
11:07:55lebelliumare you the only one working on the skins engine?
11:08:42JdGordonah thats right, use %xd(label, %bl) so the battery level instead of speicifng the subimage manually
11:10:27CIA-22New commit by jdgordon (r31069): Don't load the icons unless show_icons is actually enabled. And respect that setting in the shortcuts menu
11:12:02CIA-22r31069 build result: 2 errors, 0 warnings (jdgordon committed)
11:12:19 Join Topy [0] (
11:15:17 Quit T44 (Ping timeout: 240 seconds)
11:21:32lebelliumI'm not sure I understand what you mean. For instance my code is %?bp<%?bc<%t%xd(Ba);%t%xd(Bb)...%t%xd(Bo)|%xd(Bo)>|%?bl<%xd(Bp)|%xd(Bp)|%xd(Bq)...|%xd(BF)|>>
11:21:40lebelliumwhat should I write instead?
11:23:23JdGordonpity saint is offline, he's much better at writing skin code than me... all thouse %xd()'s can be merged into a single one
11:23:45JdGordon%xd(B, %bl) will draw the subimage that %bl is equal to
11:23:47JdGordonsort of
11:25:18lebelliumah ok, I'll try that. Actually my code is just based on the other themes and on the CustomWPS guide...
11:26:00JdGordonyeah, they are outdated
11:26:08JdGordonthat shortcut in %xd is relativly new
11:26:14 Join markun [0] (
11:26:42lebelliumand are the new tags written anywhere on the RB website?
11:27:33JdGordonthey are all on CustomWPS
11:27:36JdGordonjust no exmaples
11:27:37 Join pamaury [0] (
11:27:37 Quit pamaury (Changing host)
11:27:38 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
11:27:51JdGordonI've been meaning to write up a best practices guide but never can be bothered
11:28:09lebelliumYeah I don't mean new tags but new tricks :)
11:30:24lebelliumdid you make any theme (for ipod video or other) with these kinds of tricks or "best practices" ?
11:31:24JdGordonno, I've useless at making skins!
11:31:45*JdGordon disblaes the backdrop image and have flashbacks of the awesome rockbox blue background :)
11:35:18JdGordonkugel: if you dont like the empty shortcuts menu.... put something in it :)
11:35:58JdGordonI'd have thought you probably would have the buflib allocs page in there...
11:38:30bertrikDoes the wiki recognize FS# as a special token?
11:42:23JdGordonkugel: have you looked into adding rtc wakeup to raaA?
11:47:05 Join Stummi [0] (~Stummi@rockbox/developer/Stummi)
11:56:54kugelJdGordon: I dislike that it's empty by default
11:57:46kugelno I haven't looked into that
11:58:57kugelotoh, the clock app's alarm can launch apps on alarm so it's sort of already provided by the os
11:59:36JdGordonit can?
12:01:50JdGordonkugel: can you do me a 1024*768 raaaa build please?
12:04:17 Nick chkktri_ is now known as chkktri (chikakitaa@
12:04:37 Quit chkktri (Changing host)
12:04:37 Join chkktri [0] (chikakitaa@unaffiliated/chkktri)
12:25:16 Join ukleinek [0] (~ukl@2001:6f8:1178:2:21e:67ff:fe11:9c5c)
12:34:48PoodlemastahWas the change from tango_icons* to tango_icons*16x16.bmp intended? Should I update my themes to use the new path?
12:35:53JdGordonerr, hat happened to defaulting to grid mode in the wps if a skin isnt loaded?
12:36:33bertrikAny lcd code experts online? I wonder what I need to do or what I can reuse to allow video playback on the clip zip (rendering from yuv)
12:37:48JdGordonarg, the touchscreen is pracfically unusable on the touchpad
12:37:56JdGordonit thinks every press is a scroll :/
12:37:59JdGordonit being rockbox here
12:38:36bertrikI see many driver implement the lcd_blit_yuv function in C, using an asm function (lcd_write_yuv420_lines) to draw a couple of lines
12:41:03bertrikAnd this asm function lcd_write_yuv420_lines appears to be target specific, right
12:42:04bertrikAs an initial attempt, I could use lcd-as-memframe.S to convert YUV->RGB for a few lines in RAM, then copy it over to the display using SSP
12:43:31lebelliumbertrik: is it that important to get video playback on a 96*96 target?! Oo Isn't it anything more useful to implement or improve before?
12:44:17bertrikWhy *not* work on this? I just do the stuff I enjoy and I think most of the essentials have been done already
12:44:36JdGordonkugel: why dont the global colours work on android? global_settings.fg/bg_pattern?
12:46:45bertriklebellium, what stuff do you think needs attention?
12:47:03JdGordonwtf? colours work fine on my phone, not on the touchpad though
12:48:27lebelliumbertrik: personally I only miss a WiseChip colors calibration and RDS without breaking my theme, all the rest seems to be stable enough for a daily use :)
12:49:10bertrikI'm not planning on doing any colors calibration for the WiseChip, we're doing basically the same as the OF now
12:50:08CIA-22New commit by jdgordon (r31070): Free a handle when we try to load a backdrop file which doesnt exist or cant be loaded
12:51:20lebelliumso that means colors are badly calibrated in the OF?!
12:51:35CIA-22r31070 build result: 2 errors, 0 warnings (jdgordon committed)
12:52:35bertrikI don't know, I don't use the OF and I don't have a player with a wisechip display
12:55:22lebelliumbut you mean the current colors calibration in RB is the same as in the OF right?
12:57:23bertrikThere is no calibration in rockbox, just some formulas to calculate the R,G,B currents for the global brightness level
12:57:31bertrikmaybe those need swapping too
12:57:46 Quit Stummi (Read error: Connection reset by peer)
12:58:41bertrikI have been working on an LCD test pattern plugin, maybe that can help to determine if there are errors
12:59:06JdGordonyou can play with colour values in the colour chooser setting screen
12:59:08bertrikit's far from done yet
13:01:24lebelliumJdGordon: ah I did not think about that, I'll have a look
13:02:27PoodlemastahIs there any downside to using 16bit bmps instead of 24bit? For images with small amounts of different colors.
13:05:15***Saving seen data "./dancer.seen"
13:05:31 Join Buschel [0] (
13:06:17JdGordonPoodlemastah: no difference at all
13:06:49PoodlemastahJust thought it would be a way to shave of a a kb or two of ram usage.
13:08:41Buschelbertrik: for the YUV-to-RGB conversion you can use the test image which is shown in the test_fps plugin
13:31:37 Join fyrestorm [0] (
13:35:06 Quit fyre^OS (Ping timeout: 276 seconds)
13:44:34bertrikBuschel, I was thinking of just using mpegplayer :)
13:45:36bertrikThinking about it a bit more: the transfer to the lcd is so slow that it would be a really inefficient to first calculate stuff in RAM, then transfer it to the oled frame buffer. In the time it takes to transfer 1 pixel at 6 MHz SSP, the CPU can do 640 cycles (theoretically)
13:46:32 Quit Keripo (Quit: Leaving.)
13:48:43Buschelare the (known) ways to set the SSP to a higher clock?
13:49:42bertrikyes, but it has to be an even subdivision of the PCLK. And the datasheet of one of the oled controllers in the clip zip says we can't go faster than 100 ns cycle time (= 10 MHz)
13:50:05bertrikPCLK on the AMSv2 targets is currently 24 MHz
13:50:43 Join dfkt [0] (dfkt@unaffiliated/dfkt)
13:51:03bertriklebellium, what do see as needing calibration on your wisechip display?
13:51:33Buschelok, then 8 MHz would be possible. +33%
13:51:51bertrikon my visionox display, pure blue looks a bit off and the display changes color a bit when lowering the brightness
13:52:16bertrikBuschel, we can only use even dividers, so no dividing by 3
13:52:42BuschelI have overseen the "even"
13:53:16funmanwe can tweak pclk also
13:53:27lebelliumbertrik: in my theme the first levels of the volume bar have to be green. With the wisechip they look yellow or greeny. Is the the same on with your screen? Can you try?
13:53:43Buschelfinman: doesn't this have impact on the battery runtime?
13:56:39 Quit Rob2222 (Quit: Rob2222)
13:56:53 Join Rob2222 [0] (
14:22:04JdGordoncan I bribe anyone to make a set of cabbie images for a 1024*768 port?
14:22:18JdGordonespecially the backdrop images
14:24:19JdGordonthat was quick
14:24:36ukleinekwodz: do you have ipv6?
14:25:18gevaertsI must admit that the font code wasn't on my mental list of suspects
14:25:31JdGordonyou sure thats the break?
14:26:08gevaertsAll revisions I tried before that didn't have the issue, and starting from there they all did
14:26:29gevaerts30000 ok, 30500 ok, 30800 ok, 30808 ok, 30812 ok, 30813 ok, 30814 fail, 30815 fail, 30830 fail, 30861 fail, 30925 fail, 31045 fail
14:26:49JdGordonyay! I didnt break it :)
14:27:11gevaertsBut you wanted locking functions, so you're morally responsible ;)
14:27:30JdGordonif we had implicit locks that revision wouldnt exist!
14:28:09JdGordonpossibly put a panicf in font_lock if font_id == FONT_UI
14:28:15 Join bluebrother [0] (
14:28:15 Quit bluebrother (Changing host)
14:28:15 Join bluebrother [0] (~dom@rockbox/developer/bluebrother)
14:28:16JdGordonas a quick check
14:29:23 Join wodz [0] (
14:29:28*gevaerts tells svn to go back to today
14:29:40wodzukleinek: no I don't
14:29:47ukleinekwodz: hello, I'm just writing an email to you :-)
14:30:12ukleinekwodz: I finally came around to dump the flash of my player
14:31:10JdGordongevaerts: that still makes no sense... I dont see any reason that would break the quickscreen
14:31:17 Quit fs-bluebot (Ping timeout: 240 seconds)
14:31:24 Quit bluebrother^ (Ping timeout: 240 seconds)
14:31:37JdGordonput some splashf()'s in quickscreen.c and see where it crashes
14:31:39gevaertsJdGordon: the question is, does it break the quickscreen, or does it break restoring the previous screen?
14:31:47wodzukleinek: nice - I am slowly making progress with rk27xx nand handling
14:32:58 Join fs-bluebot [0] (
14:33:38JdGordongevaerts: you mean it crashes when you exit the qs?
14:34:12gevaertsJdGordon: yes. I didn't see any other freeze than that
14:34:36gevaertsAlthough the original reporter apparently did, which confuses me
14:34:55JdGordonyeah, he says it crashes on entry
14:35:02JdGordoncurious and curiouser
14:35:29JdGordoneither way, i doubt the quickscreen is the actual culprit
14:35:35gevaertsWell, maybe I'm not looking 12400 then, but I'm definitely looking for a serious issue :)
14:36:01JdGordonif I had to bet I'd say this is buflib for sure, and the qs just happens to trigger it (dunno how though!)
14:36:09JdGordonyou say it needs playback to relably trigger?
14:36:14lebelliumbertrik: did you also notice the "buttons locked" and "buttons unlocked" message remains on the screen between the background and text/pic layers while it should quickly disappear ?
14:36:33gevaertsI'm not sure if it needs playback or the playback *screen* though
14:37:03JdGordonplayback in the wps on my video doesnt trigger it, is it a CF only thing?
14:37:18gevaertsAs far as I know, it's only been seen on the X5
14:37:25gevaertspixelma couldn't reproduce it on the M5
14:37:52ukleinekwodz: btw, Archos exchanged my bricked player
14:37:54JdGordondoes svn with that revision reverted work?
14:38:19wodzukleinek: cool
14:38:32gevaertsI'll find out later. I'm now adding splashf()s
14:39:36wodzmaybe this is the case for 5250 vs 5249? the difference is in iram size
14:39:54 Join fml [0] (
14:40:00*JdGordon tihnks its time to make the quickscree use the theme icons instead of the 7x8 pixel inbuilt ones
14:40:03ukleinekwodz: I didn't try to apply your tools against it though.
14:41:02fmlJdGordon: why did you separate the two handle related statements with "retval=false"? I'd put the two lines as near together as possible.
14:41:12dfktis there a simple way to make a build without plugins, or to disable plugins in a build?
14:41:47JdGordonfml: feel free to clean it up if you want
14:42:36wodzdfkt: you have to hack configure
14:42:45fmlJdGordon: I don't have access to the dev env right now, otherwise I'd do it
14:42:58wodzdfkt: IIRC you have to hange plugins="yes" to plugins=""
14:43:57dfktwodz, thanks a lot, i'll have a look. i want to build a hack-ish simulator for windows with 480x800 resolution (no matter what target), i edited some player's config.h, but it always crashes at building rockblox
14:44:12fmlWe could even make core_free to accept a _pointer_ to int. core_free would then set it to -1. I understand thought that that form of the API is not nice.
14:44:48JdGordonfml: oh, I misunderstood what you were saying
14:45:26JdGordonIIUC core_free() returns 0, so I could have done handle = core_free(handle), but I tink it is wrong that 0 is not a valid handle so outright refuse to do that out of stubbornness
14:49:39fmlJdGordon: I mean I'd think that form of API were ugly because if it were nice, then the "usual" free would work like that, i.e. accept a pointer to the value rather than the value itself. OK. But then I think it's good to make those changes (performing free and setting the pointer/handle to an invlaid value) as near as possible (ideallly, it would be an atomic op).
14:49:51 Join TheLemonMan [0] (
14:52:51JdGordonfml: well c close() returns 0 on success and -1 on failure (which is absurd!) and free() returns void so they arnt very helpful.
14:53:27gevaertsJdGordon: I was wrong. It's not on exit
14:53:41JdGordonI tihnk buflib_free() should return -1, and 0 should be a legal handle, but i've given up on that particular argument
14:53:55JdGordon0 *should* be an acceptable handle id
14:54:03JdGordongevaerts: oh?
14:54:46JdGordonfml: passing a pointer in for the handle to alloc/free does make some sense and safegaurds against not setting it correctly also
14:54:48 Quit wodz (Quit: Leaving)
14:54:51fmlJdGordon: I understand your point of view (and even would agree) but as of now, I'd just put those two lines together. It looks weird if they are separated with an unrelated statement.
14:55:26gevaertsJdGordon: yes. The revisions don't change though
14:55:44fmlJdGordon: passing a pointer in would make working with constants (literals) impossible
14:55:53gevaertsI guess I didn't try changing a setting every time, so I didn't notice when exactly it froze
14:56:10JdGordonfml: that's a problem?
14:56:23JdGordonanything which expects a specific id is just plain wrong
14:57:20CIA-22New commit by jdgordon (r31071): slightly reaarange code to make fml happy :)
14:58:37fmlYes, we don't use literals as args so that wouldn't really be a restriction for us. kugel, others: any thoughts? Use a pointer to the handle in core_free. core_free would set it to an invalid val.
14:58:54CIA-22r31071 build result: 2 errors, 0 warnings (jdgordon committed)
14:59:36JdGordoncan someone commit r31070 to the 3.10 release branch please?
15:00:15 Join n1s [0] (~n1s@rockbox/developer/n1s)
15:00:21JdGordonfml: core_alloc also or it is just wierd
15:03:21JdGordonalso, I tend to think it should stay close the open/creat/close api behaviour (hence my beef with 0 being illegal)
15:04:44gevaertsJdGordon: it freezes in "left_width = display->getstringsize(s, NULL, NULL);" for the remote
15:05:18***Saving seen data "./dancer.seen"
15:05:19JdGordonthering adds something possibly debuggable!
15:05:40 Join esalazioneHg [0] (
15:06:25JdGordonbugger, h300-sim doesnt trigger it
15:08:16esalazioneHghi. i'm using an ipod nano 2G and rockbox >3.7.1 doesen't work
15:08:49esalazioneHgi've gone to #freemyipod-support to try to solve a problem that seems to be deeper than rockbox-related.
15:09:16esalazioneHgthen, after a lot of work, i restored mi iPod with iTunes, now i have firmware 1.1.3
15:09:16JdGordonesalazioneHg: you need to talk to [7]
15:10:03esalazioneHgif i install the rockbox bootloader with the utility it says " bootloader already existent, overwrite?", then i put rockbox.
15:10:11esalazioneHgdoing it manually it gives no error.
15:10:20esalazioneHgthe problem is: if the battery goes completely down, a "restore your ipod with itunes"-screen appears.
15:10:27esalazioneHgno problem if the battery goes down with the original firmware.
15:10:27gevaertsJdGordon: any hints on how to debug this further?
15:11:13JdGordongevaerts: not with jtag or getting it to crash in gdb
15:11:18esalazioneHg[JdGordon] he helped me in the #freemyipod-support channel. now the problem seems to be rockbox-related, so i switched here, as he said
15:11:49dfktwhat do i have to change in the source to make the window of an uisimulator larger? (i managed to build a 480x800 simulator, but the window is non-resizeable)
15:12:15JdGordondfkt: configure it to any size you want?
15:12:22dfktJdGordon, where?
15:12:29gevaertsdfkt: larger in what sense? Higher resolution or scaled up?
15:12:42JdGordongevaerts: splash what s is, and possibly display, it makes no sense to crash there
15:13:24dfktgevaerts, i hacked the d2 source to build with 480x800 instead of normal 320x240 - it builds fine and the simulator seems to display the higher resolution - but the window is too small and cuts off the screen
15:13:30dfktJdGordon, where do i change it?
15:14:02JdGordondfkt: configure for a sdl app (target 200)
15:14:18dfktJdGordon, i'm building it for windows... on linux i use sdl, of course
15:14:30dfktafaik, there's no way to get the sdl app running on windows
15:14:42JdGordonthen not sure what you can do
15:14:50JdGordonuisimulator −−nobackground ?
15:15:09dfkti looked around the uisimulator folder, but didn't find where the constraints for the windows are
15:16:41dfktmake −−nobackground?
15:20:08gevaertsno, the uisimulator executable
15:20:44 Join Thra11 [0] (~thrall@
15:23:28dfktwoohoo! it works, 480x800 uisim for windows - thank you, JdGordon and gevaerts :)
15:24:50 Quit fml (Quit: CGI:IRC)
15:26:29gevaertsJdGordon: it crashes with string "shuffle" for display font 8, so probably not UI_FONT I guess
15:28:01gevaertsStill testing with r30814 by the way
15:28:14JdGordonFONT_UI == 8
15:28:43JdGordonso that sounds like the remote font isnt being initialised correctly
15:28:55gevaertsGood, a clue!
15:31:49JdGordonquickscreen_fix_viewports() seems to be using uninitialised viewports!
15:31:53 Join benedikt93 [0] (~benedikt9@unaffiliated/benedikt93)
15:32:05JdGordonoh woops, no it isnt
15:32:28lebelliumJdGordon: Isn't it possible to use conditional album arts in the WPS? I try to make the album art bigger when hold switch is on
15:32:46JdGordonof course it is
15:32:55JdGordoncheck CustomWPS
15:33:01JdGordon%?C i think
15:33:06JdGordonCl maybe
15:34:43lebelliumDoesn't work
15:35:04lebelliumor maybe I can add a label to Cl?
15:35:13lebelliumthat's not written in the CustomWPS
15:35:54JdGordon%?C is what you want
15:36:13JdGordonall the cabbie skins use it
15:36:18gevaertsI don't think you can change album art size conditionally
15:36:59lebelliumfor the moment my non-working code is
15:37:03lebellium# Album art
15:37:11lebellium# Album art Hold
15:37:16gevaertsPlease don't do that!
15:37:24gevaertsUse a pastebin for pasting code
15:39:53lebelliumthat doesn't work
15:40:22lebelliumit only takes into account the latest Cl tag, l(0,0,80,80) in this case
15:41:34lebelliumMay I add a label to the Cl tag ?
15:41:46gevaertsI don't think what you want to do is possible
15:42:19lebelliumthat's what I thought but JdGordon said the contrary
15:43:26lebelliumjust some posts above
15:43:38gevaertsIf you mean the "%?C" bit, he misunderstood you. That's about doing things conditionally based on the availability of album arts
15:44:10lebelliumOk he probably misunderstood
15:44:28*gevaerts thinks he knows what's wrong with the x5
15:44:50lebelliumit's not possible to use 2 different AA sizes in the same WPS :(
15:51:23 Join jerdna [0] (
15:56:04TheLemonManwas someone working on a smtp37xx ?
15:57:53pamauryTheLemonMan: what do you mean ?
15:58:36TheLemonMani remember someone asked me about the hardware registers for that platform
15:59:09pamauryyes, I don't remember his name right now, that was for the philips vibe or something like this
16:00:40TheLemonManmight be interesting to work together, i dont have much time for coding on my own lately
16:01:36pamauryI'm not sure if it was Zoiah
16:01:58CIA-22New commit by gevaerts (r31072): Free a handle when we try to load a backdrop file which doesnt exist or cant be loaded (backport of r31070)
16:03:51TheLemonMani always wanted to start writing the board support code for rockbox but i dont know where to start
16:04:14TheLemonManand, more important, if it could be merged with the imx board
16:05:26pamauryI guess most of it could, the register even have the same names in many place so only the adress differs and the global logic is same
16:08:36TheLemonMani also have a driver for yet-another-synaptics-protocol
16:09:58pamaurywhich protocol ?
16:12:05TheLemonManits called onetouch
16:13:22pamauryis there a doc for it ?
16:14:09TheLemonMannot that i know, i reversed it from the bootloader
16:16:01pamaurystrange, because the RMI doc mention OneTouch product
16:20:45TheLemonManoh i see, it uses a silghtly different protocol, mainly different register numbers and it needs an initial configuration to be uploaded at boot
16:21:11chkktri maybe we should rewrite thi page as it's outdated as hell
16:23:14 Quit Farthen (Read error: Operation timed out)
16:23:25 Join ilf [0] (~ilf@unaffiliated/ilf)
16:23:49ilfi need to buy a new portable audio player and i want rockbox on it
16:23:59ilfwhat's the cheapest device available?
16:24:38 Join slooopy [0] (
16:24:46ilfmany listed devices seem to be not available in stores any more
16:24:54CIA-22New commit by gevaerts (r31073): glyph_cache_save should be static
16:26:22gevaertsilf: probably the clip+
16:26:50 Join Farthen [0] (~Farthen@2a01:4f8:101:2a4:0:bc28:b2e1:9)
16:27:22CIA-22r31073 build result: 2 errors, 0 warnings (gevaerts committed)
16:28:17CIA-22New commit by gevaerts (r31074): Actually make glyph_cache_save static instead of just claiming it should be
16:29:44ilfgevaerts: thanks, looking good. any experience?
16:29:58gevaertsI don't have a clip+
16:30:24CIA-22r31074 build result: 2 errors, 0 warnings (gevaerts committed)
16:31:10bertrikfunman found a real cheap sansa clip zip on amazon
16:31:25 Quit perrikwp (Ping timeout: 252 seconds)
16:32:08ilfbertrik: still unstable :(
16:36:04CIA-22New commit by gevaerts (r31075): Guard font functions against invalid font ids. These arguably should never be passed, but this prevents freezes. Fixes FS #12400
16:38:04CIA-22r31075 build result: 2 errors, 0 warnings (gevaerts committed)
16:39:34CIA-22New commit by gevaerts (r31076): Backport r31075 to avoid freezes on invalid font ids
16:40:15 Join metaphys [0] (
16:44:48CIA-22New commit by gevaerts (r31077): Bump versions for 3.10RC
16:45:37gevaertsI think I'll try to get some 3.10 RC builds out today
16:46:13ilfand what would the the recommended device to buy currently?
16:47:13dfktan android phone for rockbox ;)
16:47:45*gevaerts isn't sure if rockbox on android is end-user ready yet
16:48:14gevaertsilf: there are so many different possible requirements that that's *really* hard to answer in general
16:49:08ilfgevaerts: small, long battery. i'd say resume on power on and linux compatible without extra software, but that's rockbox all right :)
16:50:48 Join robin0800 [0] (~robin0800@
16:51:36gevaertsilf: the clip+ sounds reasonable then, battery time seems to be around 15 to 16 hours on it
16:52:13dfktyou might be able to find a fuze (not a fuze+) still, and it should give you a bit over 20 hours
16:52:43gevaertsonly the fuzev2. The fuzev1 is half of that IIRC
16:53:01gevaertsAnd it's bigger than a clip+ :)
16:53:12dfkti think the v1 has been gone for about 3 years, so chances are good to find a v2 :)
16:53:27dfktheh, i still consider the fuze rather small
16:53:32*dfkt glares at his x5l
16:53:48ilfi currently have an ipod shuffle. that's small :)
16:54:27ilfi do have an android smartphone too. but i want an extra audio player for size, price and battery reasons
16:54:37ilf(just a note to the discussion about making rockbox an app)
17:00:14 Quit liar (Read error: Connection timed out)
17:02:28lebelliumbertrik: noob question: why are the plugins not yet included in the SVN?
17:03:16bertrikIt's a bit of a all-or-nothing situation, plugins works except for 7 or 8 of them
17:04:04lebelliumah ok
17:04:24lebelliumso you wait for all to be working before including them in the SVN right?
17:05:06lebelliumand regarding rockboy?
17:05:19***Saving seen data "./dancer.seen"
17:13:53 Quit TheLemonMan (Quit: Ex-Chat)
17:14:01 Quit user890104 (Quit: .)
17:14:27metaphyshello can I map some key to anything else than action present in action.h? Is there for example any chance I could have a keys starting database? or deleting an item in playlist?
17:15:55gevaertshm, VERSION handling is broken when using :\
17:16:15metaphysgrep -R "DATABASE" apps/keymaps don't give any result...
17:16:30 Join user890104 [0] (~Venci@2001:470:1f0b:71a::)
17:27:22CIA-22New commit by fredwbauer (r31078): remove stray debugging line
17:29:23CIA-22r31078 build result: 2 errors, 0 warnings (fredwbauer committed)
17:34:54gevaertsAny language experts around? Is there an easy way to check which strings need backporting?
17:37:19 Quit user890104 (Quit: .)
17:37:30 Join user890104 [0] (
17:40:30 Quit jerdna (Quit: Leaving)
17:54:38 Quit robin0800 (Quit: Leaving)
17:55:56gevaertsbluebrother: do I remember right that rbutil doesn't really have support for RC builds now?
18:01:56[7]esalazioneHg: so it does only say "restore using itunes" if the battery goes completely dead, not just if you power it off normally from within rockbox?
18:02:40[7]this sounds like it might not have enough time to bring the FTL into a safe state before it runs out of power...
18:03:01[7]or possibly the flash chip not properly writing data when the voltage is too low
18:03:34 Join btby [0] (~user@unaffiliated/btby)
18:06:24btbyHi, I'm running Rockbox 3.9.1 on Sansa Clip+ and cannot mount / detect it on a desktop computer. When booted into original firmware, however, the player becomes mountable. What can be the cause of this?
18:07:01dfktusb connection is not yet supported in rockbox
18:08:34dfktbtby, more here:
18:08:46btbyOh, alright. Sorry, deeper searching would've yielded this.
18:09:03btbyI guess it's not a problem to switch firmware for file updating.
18:09:09dfktno, it works fine
18:09:55btbyI run Linux exclusively, is there any way to apply that patch?
18:10:15btby".. A patch has been made to correct this problem, but although it seems to work fine with the Linux..."
18:10:46dfktjudging by the comments on the tracker, it might still be hit or miss
18:11:03dfktif the annoying databas refresh of the original firmware bothers you, that can be fixed
18:12:00btbySomewhere in the settings?
18:12:44dfktno, with an amusing workaround. :) in rockbox, delete the two folders on the clip+ that start with ##, connect the player, and fill it up so there's less than 100mb free storage
18:13:03dfktafter that, the OF will complain about not enough space and shut down without refreshing
18:13:58btbyha ha, wow!
18:15:02 Join dreamlayers [0] (~bgjenero@rockbox/developer/dreamlayers)
18:21:52bluebrothergevaerts: yes
18:23:53btbydfkt: It doesn't refresh! =) Thank you for your help.
18:24:05dfktbtby, you're welcome :)
18:24:20dfktand you gained ~105mb more space by getting rid of that mtp/drp crap ;)
18:24:35btbyha ha, good riddance
18:28:56esalazioneHg[7]:yes, it works well if i simply turn off rockbox
18:29:58esalazioneHg[7]: do you remember, my iPod was the one with the strange behavior, the one that almost totally bricks
18:38:18[7]so it just happens if rockbox does shut down because of an empty battery?
18:38:39[7]does it also happen if the apple firmware was running when the battery goes empty?
18:40:59CIA-22New commit by dreamlayers (r31079): If Rockbox never turns off ATA power, don't compile code for turning it back on. Note that ATA power may still need to be turned on when starting, so ...
18:41:48dreamlayersDoes Rockbox shut down cleanly because of an empty battery, or does hardware cut off power?
18:42:43[7]dreamlayers: I suspect it power gets cut off during the attempt to shut down
18:43:01CIA-22r31079 build result: 2 errors, 0 warnings (dreamlayers committed)
18:43:41[7]the HW shutoff threshold is rather low, IIRC 2.9 or 3.0V
18:43:58esalazioneHg[7]: i tested with apple's and it doesn't happen
18:44:36dreamlayersWith flash based storage, there shouldn't be a large spike in power usage that crosses the shutoff threshold.
18:45:22[7]my point is that at 3.0V battery voltage the LDO regulator output for the flash will probably be at like 2.7V, which might not allow it to write safely
18:45:45esalazioneHgdreamlayers: i'm not sure... i think while battery is empty and i try to turn on the ipod, it works (apple's logo) for some secs, then the white screen of death ("restore your ipod with itunes") appears
18:45:50[7]so what we're observing might be flash corruption caused by a brownout
18:46:33[7]so this might be as simple as our shutdown threshold being too low
18:46:53esalazioneHgit never happened before our try (you remember what i'm talking about [7]?)
18:48:01esalazioneHgnow i've just restored the apple's firmware and fully charged the battery. i've not installed the bootloader yet. have i to try something?
18:48:49esalazioneHgps: if i plug the ipod while there's the "restore"-screen i can see only /dev/sdc without partition
18:49:19[7]oh, you were the one with the ipod on which apple's boot loader and disk mode failed to reinitialize the FTL?
18:49:32esalazioneHg[7] yep
18:50:05[7]hm, apparently there's something odd going on with your flash chip - not exactly sure what
18:50:53 Join robin0800 [0] (~robin0800@
18:55:53 Quit robin0800 (Ping timeout: 245 seconds)
19:00:40dreamlayersThe PCF50633UM datasheet says minimum voltage drop for the LDO regulators is 250mV.
19:01:31 Join liar [0] (
19:05:20***Saving seen data "./dancer.seen"
19:06:36dreamlayersTH58NVG5D4CTG20 flash Vcc = 3.3V.
19:08:18dreamlayers(PCF50633UM datasheet: , flash specs: )
19:29:35 Quit dreamlayers (Quit: Bye)
19:53:20 Join david_ [0] (
19:54:18 Join perrikwp [0] (
19:54:40 Quit david_ (Client Quit)
20:02:19 Join TheLemonMan [0] (
20:04:05 Quit Buschel (Ping timeout: 252 seconds)
20:22:03 Join saratoga [0] (9803ec71@gateway/web/freenode/ip.
20:23:04saratogaIMO we should never let a battery get down to 3.0v anyway, low voltage shutoff should happen around 3.3-3.5v to be safe
20:28:29dfktis there a way to simulate disc access in the simulator, to activate the virtual LED?
20:29:04funmandfkt: i thought it was automatic
20:29:11funman(already done)
20:29:42dfkti know i saw it in older simulator builds light up sporadically, but didn't see it in new builds
20:29:49funmanuisimulator/common/stubs.c have storage_spin
20:30:21dfkti assume that doesn't care if it's HDD or flash memory?
20:34:10 Join StephenCarroll__ [0] (
20:34:52StephenCarroll__gevaerts, regarding the rcs of 3.10 can you check your X5 for the following bug
20:35:01StephenCarroll__or anyone else with an X5 for that matter
20:35:58StephenCarroll__gotta go brb
20:46:37 Join hilbert [0] (
20:49:04gevaertsUSB stack overflow on x5 when trying to make a screenshot with the 3.10 branch :\
20:51:00*gevaerts knew that releasing right away was a bad idea...
20:53:17esalazioneHgfr[7]: there something i have to test/do?
20:53:21 Part ilf
20:53:37esalazioneHg[7]: there something i have to test/do?
20:53:50gevaertskugel: could FS #12411 be related to your alpha channel work?
20:53:51fs-bluebot Album art not displaying correctly on iAudio x5, verk dark (bugs, unconfirmed)
20:55:03saratogathe FAQ link on our front page is rather out of date and probably not all that useful to new users
20:55:31saratogawould anyone mind if I removed it and then simply had port pages to link to the FAQ entries for that target (assuming they exist)?
20:56:23saratogaalso "docs index" is probably less useful as a starting point
20:56:33saratogamaybe I should post something to FS and see what people think
20:57:38 Quit StephenCarroll__ (Quit: CGI:IRC)
20:58:17 Quit esalazioneHg (Quit: Leaving.)
21:05:22***Saving seen data "./dancer.seen"
21:06:52 Quit benedikt93 (Quit: Bye ;))
21:13:23 Nick kugel is now known as kugelp (~kugel@rockbox/developer/kugel)
21:14:05kugelpgevaerts: I Di
21:15:05kugelpI didn't change album art displaying. it doesn't use (and doesn't read) alpha information
21:16:01kugelpit doesn't work on jpeg anyways
21:17:07kugelpcould be
21:17:32pamaurywho knows the keymaps really well ? metaphys had a question regarding the WPS apparently
21:17:45kugelpthe resizer has cf specific code
21:18:15metaphysyes I do!
21:21:34 Join dreamlayers [0] (
21:21:34 Quit dreamlayers (Changing host)
21:21:34 Join dreamlayers [0] (~bgjenero@rockbox/developer/dreamlayers)
21:26:50 Quit y4n (Quit: The world around me drops away, replaced by worlds being created and destroyed by my imagination.)
21:31:38 Join Buschel [0] (
21:34:49 Quit btby (Ping timeout: 244 seconds)
21:35:27 Join esalazioneHg [0] (
21:39:30 Quit Buschel (Quit: ChatZilla 0.9.87 [Firefox 8.0/20111104165243])
21:40:08 Join domonoky [0] (~Domonoky@rockbox/developer/domonoky)
21:55:27 Quit pamaury (Ping timeout: 245 seconds)
21:58:19 Join pamaury [0] (
21:58:20 Quit pamaury (Changing host)
21:58:20 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
22:05:35 Quit liar (Read error: Connection timed out)
22:10:21 Quit scanf (Quit: now)
22:27:16 Quit Thra11 (Ping timeout: 252 seconds)
22:39:06 Join webguest670 [0] (
22:39:12webguest670I have a problem with my h140. Two weeks ago, worked fine, but in the middle of a track the screen turns off with an error on "I00: at..."
22:39:37 Join Thra11 [0] (~thrall@
22:40:10webguest670Since then I can't make it work. When i press the power on button, the screen turns on but does not show any message
22:40:44 Quit TheLemonMan (Quit: Ex-Chat)
22:41:29webguest670Help me please!!!
22:45:40 Quit dreamlayers (Quit: Leaving)
23:01:29 Quit pamaury (Remote host closed the connection)
23:05:26***Saving seen data "./dancer.seen"
23:09:07 Quit Horschti (Quit: Verlassend)
23:10:12 Quit lebellium (Quit: ChatZilla 0.9.87 [Firefox 9.0/20111122192043])
23:11:09gevaertswebguest670: looks like a hardware problem. It could be the hard drive, but it's hard to say
23:12:28JdGordongevaerts: not such a huge fan of your fix :/
23:12:37JdGordonfinding the actual cause would be better
23:12:57gevaertsJdGordon: I considered putting it on FS first, but I think we need both the "real" fix and this
23:13:34gevaertsThose functions are (possibly indirectly) accessible from the plugin API, including lua scripts. They should be robust if robustness isn't too expensive I think
23:13:54 Quit dfkt (Quit: -= SysReset 2.55=- Sic gorgiamus allos subjectatos nunc.)
23:14:24gevaertsBut yes, FONT_UI shouldn't appear there
23:15:14 Quit domonoky (Read error: Connection reset by peer)
23:15:18gevaertsFeel free to revert, but please leave the fix in the 3.10 branch unless you have another fix :)
23:16:13 Quit Thra11 (Ping timeout: 252 seconds)
23:16:26webguest670Sometimes, the screen displays errors: I00: at 40102314 I00: at 40102004 I00: at 400c2004
23:17:17webguest670Today i'm trying to change the battery,... the HD, and power on, with the remote, and the screen shows: Rockbox boot loader, Version: 6, Batt: 3.64 v
23:17:28webguest670If i press "record" & "power on" buttons, the screens show: "starting original firmware" but stays froze
23:23:25 Join MethoS- [0] (~clemens@
23:29:36 Join Thra11 [0] (~thrall@
23:36:18 Quit n1s (Quit: Ex-Chat)
23:37:56 Join Scromple [0] (
23:41:07 Quit knittl (*.net *.split)
23:41:08 Quit Llorean (*.net *.split)
23:41:08 Quit Zarggg (*.net *.split)
23:41:08 Quit maraz (*.net *.split)
23:41:08 Quit kkit (*.net *.split)
23:41:08 Quit Utchybann (*.net *.split)
23:41:08 Quit Unhelpful (*.net *.split)
23:41:09 Quit dionoea (*.net *.split)
23:41:09 Quit Staphylo (*.net *.split)
23:41:09 Quit Barahir (*.net *.split)
23:41:09 Quit niekie (*.net *.split)
23:41:09 Quit GeekShadow (*.net *.split)
23:41:09 Quit ruskie (*.net *.split)
23:41:09 Quit nomada (*.net *.split)
23:41:10 Quit rasher (*.net *.split)
23:41:10 Quit [sanyi] (*.net *.split)
23:41:10 Quit eintopf (*.net *.split)
23:41:10 Quit CIA-22 (*.net *.split)
23:41:11 Quit Lalufu (*.net *.split)
23:41:11 Quit bertrik (*.net *.split)
23:41:11 Quit nosa-j (*.net *.split)
23:41:11 Quit Poodlemastah (*.net *.split)
23:41:11 Quit funman (*.net *.split)
23:41:11 Quit polemon (*.net *.split)
23:41:12 Join knittl [0] (~knittl@unaffiliated/knittl)
23:41:12 Join Barahir [0] (
23:41:16 Join funman [0] (~fun@rockbox/developer/funman)
23:41:16 Join kkit [0] (
23:41:17 Join dionoea [0] (
23:41:22 Join Poodlemastah [0] (
23:41:25 Join maraz [0] (
23:41:26 Join Staphylo [0] (~Staphylo@
23:41:30 Join [sanyi] [0] (
23:41:31 Join nosa-j [0] (
23:41:35 Quit [sanyi] (Changing host)
23:41:35 Join [sanyi] [0] (~sanyi@unaffiliated/alexanderc)
23:41:36 Join polemon_ [0] (
23:41:37 Join GeekShadow [0] (
23:41:37 Join Zarggg [0] (
23:41:37 Join Lalufu [0] (
23:41:38 Quit Lalufu (Changing host)
23:41:39 Join Lalufu [0] (~sun@unaffiliated/lalufu)
23:41:39 Join CIA-10 [0] (
23:41:44 Join bertrik [0] (
23:41:44 Quit bertrik (Changing host)
23:41:44 Join bertrik [0] (~bertrik@rockbox/developer/bertrik)
23:41:44 Join Utchybann [0] (
23:41:45 Join Unhelpful [0] (~quassel@rockbox/developer/Unhelpful)
23:41:51 Join eintopf [0] (~eintopf@
23:42:04 Join rasher [0] (~rasher@rockbox/developer/rasher)
23:42:09 Join Llorean [0] (
23:42:18 Join nomada [0] (
23:42:22 Join niekie [0] (quasselcor@CAcert/Assurer/niekie)
23:43:47metaphysWhere are action for keymaps defined? I would like to implement a new one (ACTION_WPS_ID3DB (database)
23:44:34 Join aquarius [0] (
23:45:42aquariusdo you know i its possible to install rockbox on this player (archos key)?
23:46:22bluebrotherit is not.
23:46:29 Quit Llorean (Changing host)
23:46:29 Join Llorean [0] (~DarkkOne@rockbox/user/Llorean)
23:46:32bluebrotherRockbox runs on the players listed on the front page. Period.
23:46:34aquariusoh ok :(
23:47:01aquariusthe software on that one sucks. you must stop the current music to get to the folders...
23:48:07bluebrotherwell, then you need to stop playback to browser folders.
23:48:32aquariusjo :/
23:49:27 Quit webguest670 (Quit: CGI:IRC (EOF))
23:50:27 Quit aquarius (Quit: Leaving)
23:53:19 Join ruskie [0] (ruskie@sourcemage/mage/ruskie)
23:56:43 Join saratoga_ [0] (9803c31c@gateway/web/freenode/ip.
23:56:44metaphysnobody know anything about action?
23:57:09metaphysthose to link with keys in keymaps?
23:57:45metaphysI can't find where they get defined!
23:58:39gevaertskugelp: ok, I'll do some more bisecting then next week

Previous day | Next day