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 2011-12-30

00:00:08jlbiasiniok my mistake 67 does work after a 'make clean'
00:01:48 Join xmikex_ [0] (~xmikex@
00:02:47 Quit Topy44 (Ping timeout: 240 seconds)
00:04:52jlbiasinidoes anybody know where quickscreen action get configured? On fuze+ the quickscreen provide only left and right and I saw other plyer have also fonction for up and down...
00:05:59 Join xmikex__ [0] (~xmikex@
00:06:22 Quit xmikex_ (Ping timeout: 252 seconds)
00:09:27 Quit xmikex__ (Remote host closed the connection)
00:11:02dfktjlbiasini, usually you hold the select button longer on the item you want to set in the quickscreen. default only shows left/right, but you can add up/down
00:19:59 Join Scr0mple [0] (~Simon@
00:22:28 Quit Scromple (Ping timeout: 248 seconds)
00:24:08 Quit amarc (Ping timeout: 240 seconds)
00:28:49 Quit jlbiasini (Remote host closed the connection)
00:29:44 Join JdGordon| [0] (~jonno@rockbox/developer/JdGordon)
00:38:26 Quit einhirn (Ping timeout: 252 seconds)
00:55:13***Saving seen data "./dancer.seen"
01:08:44 Quit n1s (Quit: Ex-Chat)
01:09:21 Quit dfkt (Quit: -= SysReset 2.55=- Sic gorgiamus allos subjectatos nunc.)
01:11:03bertrikfunman, at some point we disabled 4-bit support for AMSv1, I'm thinking of re-enabling it.
01:11:15funmanwhat was the problem with it?
01:11:39bertrikusing 4-bit instead of 1-bit should give us a 4x speedup for sd access, vs a 2x speed-up through high-speed mode that is only possible on certain cards
01:11:44bertrikfunman, I don't know
01:11:54funmanbertrik: btw on my clip zip the lcd performance is good. i dont' think removing your fix in the lcd driver will gain something, but it would surely take time
01:12:11funmaniirc some cards failed to work
01:12:42funmanor the VERIFY_ functions showed some bugs
01:12:58funmanthe work was made by ranmachan and flyndice iirc
01:15:27bertrikthe svn log mentions some write corruptions in june 2010 and read corruption in july 2010
01:15:43bertrikI tried 4-bit mode on my clipv1 and didn't run into any problem
01:16:45funmanr28834 / r28925 also
01:17:55funmanperhaps irc logs around the commits can give some logs / tests about what/how it failed
01:19:34funmanor fly to the past and tell me to add details to my commit logs
01:20:13bertrikI can only test on my clipv2, which has no uSD, I wonder if the original problems got fixed by later bugfixes
01:20:58bertrikbah, the AMSv1 sd driver is a bit of a mess because of the dual use of some pin for button light and sd
01:22:03funmanbertrik: hm aer you talking about amsv1 or v2 ?
01:22:07funmanclipv2 is amsv2
01:22:23bertrikarg, I meant clipv1
01:29:53marcol07is there any chance to display emulator in 2x zoom? :D
01:31:09funmanthere's the -zoom option i think
01:32:44 Join Horscht [0] (~Horscht@xbmc/user/horscht)
01:33:13marcol07ok thanks :D thats what i wanted :P −−zoom [val] :)
01:34:52 Join NatasLived666 [0] (~NatasLive@
01:36:58 Quit NatasLived666 (Client Quit)
01:38:06 Quit efyx (Remote host closed the connection)
01:38:14 Join NatasLived666 [0] (~NatasLive@
01:38:21 Quit NatasLived666 (Client Quit)
01:42:37 Quit marcol07 (Quit: CGI:IRC (Ping timeout))
01:48:59WilliamCIt's neat that you can play Doom using Rockbox but it's awkward as heck
01:49:52TorneWe are aware of this :)
01:50:13WilliamCNothing you can really do about it as these are music players not Gameboys
01:51:05Torneother things which are really awkward include playing text adventures :)
01:51:13WilliamCHas anyone done battery benchmarks on the Sansa Clip Zip?
01:52:26bertrikWilliamC, I think so, but I don't see any results yet on the SansaRuntime wiki page. So you're welcome to add some.
01:52:50WilliamCJust wondering
01:53:06WilliamCI chose the theme with the least white in it in hopes it'll lessen the drain
01:53:28bertrikthe display turns off after 30s or so anyway
01:55:44 Quit bertrik (Quit: And That, My Liege, Is How We Know the Earth to Be Banana Shaped)
02:01:13 Join xmikex [0] (~xmikex@
02:14:11 Quit mystica555_ (Quit: Ekkusu Chatto)
02:14:34 Join mystica555_ [0] (
02:23:02 Quit lebellium (Quit: ChatZilla 0.9.88 [Firefox 10.0/20111221135037])
02:24:16 Quit Strife89 (Quit: Heading out)
02:30:19bluebrotherze, gevaerts: Rockbox Utilizy not detecting the mountpoint on a Clip+ is actually to be expected and not a bug
02:30:44zebluebrother: oh, ok
02:30:50bluebrotherand not detecting the mountpoint doesn't mean nothing could be detected
02:32:16bluebrotherin some cases we can recognize the player (via usb ids) but not the mountpoinz (since there is no way to retrieve the mountpoint for a given usb id)
02:43:41funmanbluebrother: can't you find the devnode for a given usb id ?
02:48:12 Join [Saint_] [0] (~Saint]@unaffiliated/saint/x-8516940)
02:49:46 Quit [Saint] (Ping timeout: 240 seconds)
02:55:16***Saving seen data "./dancer.seen"
03:01:02 Quit perrikwp (Ping timeout: 268 seconds)
03:01:10 Join captainkwel [0] (
03:01:28 Join user622 [0] (~user@
03:02:38 Quit user622 (Client Quit)
03:05:35 Join deameyes [0] (~deameyes@
03:06:41deameyesis there a way to obtain the path of the file used when opening a viewer?
03:08:14funmandeameyes: the entry point of each plugin has a char * parameter
03:13:39 Join Xerion [0] (
03:19:31 Quit remlap (Quit: Leaving.)
03:20:17 Part deameyes
03:35:36 Join remlap [0] (
03:36:10 Join jlbiasini [0] (
03:36:38 Part jlbiasini
03:37:14 Join jlbiasini [0] (
03:38:05 Part jlbiasini
03:44:04 Quit [Saint_] (Ping timeout: 248 seconds)
03:48:46 Join [Saint] [0] (~Saint]@unaffiliated/saint/x-8516940)
03:56:00 Quit anewuser ()
04:08:08captainkwelthere's no 4:3 landscape theme for raaa?
04:09:12 Join anewuser [0] (~anewuser@
04:09:12 Quit anewuser (Changing host)
04:09:12 Join anewuser [0] (~anewuser@unaffiliated/anewuser)
04:11:37[Saint]captainkwel: there's an 800x480 theme...but its not my work so I have no idea what state its in.
04:12:40captainkwelit's also not 4:3 (hooray for letterboxing)
04:13:58[Saint]Application builds are for a specific resolution. A build smaller than the device res will align top left iirc.
04:15:42[Saint]I wasn't going to bother with landscape themes until there's sensor rotation.
04:24:38JdGordon|landscape does give more room to play with
04:24:50JdGordon|especially at res like 1024x768 :)
04:25:12 Quit amiconn (Disconnected by services)
04:25:13 Join amiconn_ [0] (quassel@rockbox/developer/amiconn)
04:25:34 Nick amiconn_ is now known as amiconn (quassel@rockbox/developer/amiconn)
04:26:11 Quit pixelma (Disconnected by services)
04:26:13 Join pixelma_ [0] (quassel@rockbox/staff/pixelma)
04:26:15 Nick pixelma_ is now known as pixelma (quassel@rockbox/staff/pixelma)
04:27:23 Quit [7] (Disconnected by services)
04:27:33 Join TheSeven [0] (~TheSeven@rockbox/developer/TheSeven)
04:38:51[Saint]Yeah...but portrait is easier to use one handed.
04:39:39[Saint]I definitely think portrait is the sane default, but I'm all for having sensor rotation or a "flip" setting.
04:50:59mystica555_im rather enjoying my boogieboard of a tablet..
04:51:12mystica555_next on the list: rooting, and enabling usb HID
04:51:25mystica555_then i can finally sit in bed with my full sized keyboard and hack away
04:52:08mystica555_disappointed to find there is a lacking of terminal/shell applications that know anything about xterm style (anything that works with curses at least) mouse control/movements
04:52:30mystica555_tmux... tapping on a pane to select it.. win
04:52:44mystica555_multitouch ... sending tmux commands to resize based on 2 finger dragging etc
04:52:52mystica555_i need to learn java.
04:55:18***Saving seen data "./dancer.seen"
04:59:50*[Saint] boots mystica555_ into #rockbox-community
05:06:59 Quit thegeek (Read error: Connection reset by peer)
05:07:13 Join thegeek [0] (
05:08:43 Join nosa [0] (
05:10:57 Quit nosa-j (Ping timeout: 244 seconds)
05:10:58 Nick nosa is now known as nosa-j (
05:16:19 Join perrikwp [0] (
05:27:30 Join Rob2223 [0] (
05:30:38 Part WilliamC ("Leaving")
05:31:52 Quit Rob2222 (Ping timeout: 276 seconds)
05:46:51 Quit xmikex (Ping timeout: 248 seconds)
06:33:59 Quit captainkwel (Quit: Ex-Chat)
06:34:33 Quit factor (Read error: Connection reset by peer)
06:41:39 Join Keripo [0] (
06:45:38 Quit Keripo (Client Quit)
06:46:15 Quit davo (Ping timeout: 240 seconds)
06:48:12 Join davo [0] (
06:51:58 Join factor [0] (~factor@
06:55:22***Saving seen data "./dancer.seen"
06:56:21 Quit davo (Ping timeout: 276 seconds)
06:57:40 Join davo [0] (
07:04:25 Nick amithkk is now known as arduino-amithd (u4289@2buntu/writers/amithkk)
07:04:35 Nick arduino-amithd is now known as amithkk (u4289@2buntu/writers/amithkk)
07:30:44 Nick kugelp is now known as kugel (~kugel@rockbox/developer/kugel)
07:40:57 Join Scromple_ [0] (~Simon@
07:44:06 Quit Scr0mple (Ping timeout: 268 seconds)
07:56:46 Nick kugel is now known as kugelp (~kugel@rockbox/developer/kugel)
07:56:47 Nick kugelp is now known as kugel (~kugel@rockbox/developer/kugel)
08:11:31 Join ender` [0] (
08:12:35 Join amarc [0] (
08:23:08 Quit T44 (Ping timeout: 240 seconds)
08:42:17 Quit Scromple_ (Quit: Leaving)
08:55:25***Saving seen data "./dancer.seen"
09:03:52 Join Horschti [0] (~Horscht@xbmc/user/horscht)
09:05:03 Part amarc ("Leaving")
09:06:56 Quit Horscht (Ping timeout: 252 seconds)
09:08:02 Join curtism [0] (
09:08:15curtismis anyone alive in here?
09:14:16[Saint]Yes, lots of people are here.
09:14:44[Saint]Please just ask your (rockbox related) question, and, welcome.
09:14:58curtismmy question went away :P
09:15:03curtismthanks though
09:15:48curtismseems like there are a lot of bad theme for sansa fuze D:
09:18:41 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
09:19:13 Join stoffel [0] (
09:19:30[Saint]Feel free to make your own...
09:19:44curtismi'd probably make a worse one :P
09:19:55[Saint]Or take in from the theme site, and bend it to your will.
09:20:46[Saint]We have pretty extensive docs on theme creation in the wiki (and to a lesser extent the manuals).
09:35:13curtism[Saint]: rockbox supports FLAC, right?
09:37:05curtismah, nvm, found the wiki page
09:39:52[Saint]Yes, yes it does.
09:40:56[Saint]You can *almost* say: "If Rockbox doesn't play <audio file>, then <audio file> is probably broken".
09:46:55 Quit perrikwp (Remote host closed the connection)
09:50:00 Join perrikwp [0] (
09:57:43 Quit davo (Quit: Lost terminal)
10:37:22 Join lebellium [0] (
10:39:17pamauryis there a complete hynix datasheet for nand flash ? on google one can find a few datasheets for a few nands but I would expect some kind of generic one
10:40:29pamaurythere is the onfi one of course but it doesn't give info like timing, just the commands
10:42:08 Quit remlap (Ping timeout: 244 seconds)
10:46:32 Join bertrik [0] (~bertrik@rockbox/developer/bertrik)
10:48:21 Join marcol07 [0] (
10:55:29***Saving seen data "./dancer.seen"
10:57:33 Join bmbl [0] (~bmbl@unaffiliated/bmbl)
11:14:54 Quit marcol07 (Quit: CGI:IRC (EOF))
11:23:19 Join n1s [0] (
11:23:19 Quit n1s (Changing host)
11:23:19 Join n1s [0] (~n1s@rockbox/developer/n1s)
11:23:59 Join marcol07 [0] (
12:01:08 Join [Saint_] [0] (~Saint]@unaffiliated/saint/x-8516940)
12:04:27 Quit [Saint] (Ping timeout: 276 seconds)
12:04:45 Join efyx [0] (
12:09:40 Join [Saint] [0] (~Saint]@unaffiliated/saint/x-8516940)
12:13:39 Quit [Saint_] (Ping timeout: 252 seconds)
12:14:35 Join [Saint_] [0] (~Saint]@unaffiliated/saint/x-8516940)
12:17:42 Join liar [0] (
12:18:25 Quit [Saint] (Ping timeout: 252 seconds)
12:18:38 Quit stoffel (Read error: Operation timed out)
12:25:15 Join [Saint] [0] (~Saint]@unaffiliated/saint/x-8516940)
12:28:19 Quit [Saint_] (Ping timeout: 252 seconds)
12:36:45 Join bimbel [0] (
12:36:56 Quit bimbel (Changing host)
12:36:56 Join bimbel [0] (~bmbl@unaffiliated/bmbl)
12:39:41 Quit bmbl (Ping timeout: 240 seconds)
12:55:33***Saving seen data "./dancer.seen"
13:03:06 Join benedikt93 [0] (~benedikt9@unaffiliated/benedikt93)
13:11:16 Quit JdGordon| (Remote host closed the connection)
13:45:07 Quit [Saint] (Read error: Connection reset by peer)
13:46:00 Join [Saint] [0] (~Saint]@unaffiliated/saint/x-8516940)
14:07:51 Quit benedikt93 (Quit: Bye ;))
14:17:18 Quit liar (Read error: Connection timed out)
14:18:04 Join liar [0] (
14:27:26 Join lorenzo92 [0] (
14:29:08lorenzo92kugel: any idea why the benchmark plugin doesn't work? I note only a USING_STORAGE_CALLBACK....what's that?
14:29:28kugellorenzo92: yes :)
14:29:47kugelthere's an error in my patch
14:29:53lorenzo92ah! :)
14:30:02kugelit doesnt flush the buffer on exit (because of the #ifdef I added)
14:30:25kugelthe buffer is quite big so it can lack several hours
14:30:46lorenzo92it can lack? and why?
14:31:00lorenzo92doesn't it flush when necessary?
14:31:21kugelthe call to unregister_idle_callback() doesnt happen. that calls flush_buffer() indirecty
14:31:47 Join remlap [0] (
14:31:55kugellorenzo92: it does when it's full, but not on exit (with my patch)
14:31:56lorenzo92kugel: this why we don't have a "storage" driver built in rb right?
14:32:12lorenzo92kugel: ah! now it's clear why 16 hours were written
14:32:23kugelwhat's needed is to add a #else \n flush_buffer() \n #endif to the second USING_STORAGE_CALLBACK
14:32:25 Join einhirn [0] (
14:32:59lorenzo92okay...added ;)
14:33:05lorenzo92I'll test it!
14:33:34kugelcan you show me the .c file (via pastie), just to make sure :)
14:34:15 Join Misan [0] (
14:35:25kugellorenzo92: right. except it needs to be flush_buffer(NULL) (my bad)
14:35:50lorenzo92ops :)
14:37:29lorenzo92arggg I don't know why but the safe mode isn't working fine since yesterday...
14:37:48lorenzo92I guess that the cable checking thing interferes...
14:38:41lorenzo92it unmounts randomly
14:39:00kugelbad cable perhaps?
14:39:35lorenzo92kugel: well not sure because lebellium told me it happened him too once + OF thing works perfectly....
14:40:55lorenzo92later I will do a rom without the cable checking...let's see
14:42:18lorenzo92kugel: charging seems working fine, but we should give an eye to the ascodec charging-related registers too to be sure ;)
14:42:56lorenzo92btw testing bench
14:43:23lorenzo92if it works, I could begin the backlight on test (since I have no time to care R0 tomorrow :D)
14:43:40lorenzo92...with bl on it lasts something like 6-7 hrs
14:46:01lorenzo92kugel: okay ;) this time is the right one hehe the bench works
14:48:21kugellorenzo92: okay
14:49:40lorenzo92bench I have other things to do...I will send you results asap ;)
14:51:17lorenzo92kugel: btw, did you test headphone sense? or discovered something about gpio ioctls?
14:51:37kugelno, had no time
14:52:23lorenzo92kugel: for the bench I leave it on the wps screen.....
14:52:34kugellorenzo92: main menu please
14:52:50kugelthe wps screen consumes a lot more cpu
14:52:57lorenzo92kugel: yeah guessed that
14:53:11kugelso the the backlight has a little less influence
14:53:15lorenzo92kugel: BUT when backlight is gone (in normal usage) is the fresh paused?
14:53:32lorenzo92*refresh rate
14:53:59lorenzo92indeed, stupid question :D
14:54:05kugelI doubt it's already implemented on the ypr0 though
14:54:20kugelyou need HAVE_LCD_ENABLE or HAVE_LCD_SLEEP for that
14:54:44lorenzo92uhm I do the lcd sleep in the backlight off function, well that's why we need to revise that
14:55:01lorenzo92perhaps it could last longer than 25 hours so :)
14:55:35***Saving seen data "./dancer.seen"
14:56:04kugelthere's probably some room for improvements left, but usually saving cpu time doesnt yield as much runtime as turning hw units off
14:56:21kugelperhaps an extra hour, I doubt more
14:57:28lorenzo92yeah indeed nothing serious ;)...and also about arm optimizations. Do you think we can enable them later or what?
14:57:46lorenzo92keeping in mind that 25 hours are a lot
14:57:48kugellater :)
14:57:53lorenzo92yeah ;)
15:00:50amiconnOn colour targets which disable the LCD (and hence stop updating it) when backlight is off the screen it stays in shouldn't have any effect on runtime
15:03:20marcol07do somebody know how long does sansa clip plus battery last with rockbox?
15:04:02[Saint]The wiki probably does...
15:04:09 Join lovasoa [0] (~olojkine@
15:04:46 Quit marcol07 (Quit: CGI:IRC (EOF))
15:05:05 Join marcol07 [0] (
15:05:11[Saint]I can't remember what the page is called off the top of my head, searching "runtime" in the wiki should find it.
15:05:28 Quit marcol07 (Client Quit)
15:05:32 Join marcol07 [0] (
15:07:07amiconnActually the wps might even have a slight advantage in this case, because (iirc) the wps also stops redrawing when the lcd is shut down, while other screens don't do that
15:09:42amiconnIn other cases (targets without lcd shutdown because the lcd stays readable) staying in the main menu probably means slightly longer runtime
15:12:57lorenzo92amiconn: thanks for exlpaination ;) but since we draw on the linux framebuffer, disabling also the lcd fresh could help a little too :)
15:16:26amiconnDoes this port use the linux framebuffer as the rockbox framebuffer, or are there two buffers?
15:21:13kugeltwo of course
15:22:00amiconnWhy "of course"? This sounds like a possibility for optimisation to me?
15:22:33kugelbecause you might end up with partiallly displayed stuff
15:23:03kugelstuff relies on data _not_ beign actually displayed until the lcd_update/_rect calls
15:23:23 Join jlbiasini [0] (
15:24:31amiconnOf course. So you don't have control over the lcd refresh?
15:24:48amiconnFrom lorenzo92's statement I thought you do
15:25:15kugellcd_update() is a memcpy() to the mmap()'ed linux framebuffer
15:26:25kugelalthough I think memcpy to it will cause an immediate update too, however I don't know the details of the linux framebuffer
15:30:14 Join y4n [0] (y4n@unaffiliated/y4ndexx)
15:33:45plushi am trying to debug usb mass storage on an ipod4g
15:33:57plushanyone here with some knowledge of that code?
15:34:13plushi have done a lot of tracing and debugging, now i need some rockbox insight to make further progress
15:37:00lorenzo92kugel: I wanted to say the lcd refresh done by rb ;) We can disable the lcd (already implemented in rb) and I guess there are also some available ioctls for the framebuffer but don't know more than that
15:38:48 Join mortalis [0] (~mortalis@
15:38:58 Join James_Hunt [0] (2ea48860@gateway/web/freenode/ip.
15:41:36 Join B4gder [0] (
15:41:36 Quit B4gder (Changing host)
15:41:36 Join B4gder [241] (~daniel@rockbox/developer/bagder)
15:42:13James_HuntHello and happy holidays to all. I want again remind about full translated Russian language. Went 1 month and I'm worry. I suspect no one approve my translations because I have not Russan name. Want to explain: I'm immigrant and changed name. My native language is Russian.
15:43:38 Quit liar (Read error: Connection timed out)
15:46:20y4nJames_Hunt: greets from Saint Petersburg \o
15:49:39James_Hunty4n: :3
15:51:20gevaertsJames_Hunt: thanks for the information. Yes, that was a concern
15:52:51 Quit lorenzo92 (Quit: ChatZilla 0.9.88 [Firefox 8.0/20111115183813])
15:53:48gevaertshm, we seem to have a bunch of translations waiting on the tracker...
15:57:41James_Huntgevaerts: oh, that's good. I can vouch for the my translation. −−- You mean conflict with other unapproved Russian tarnsltations? I've didn't find other.
15:57:58gevaertsNo, I mean in general, for other languages
15:58:57 Join lasser [0] (
15:59:11 Part lasser
15:59:13jlbiasinipamaury: there is going to be again some little improvement on the new new keymaps patch for fuze+ so I guess you can wait this evening before commiting it
16:08:06 Join Topy44 [0] (
16:20:51 Quit James_Hunt (Ping timeout: 258 seconds)
16:26:35plushanyone with some rockbox usb code knowledge around?
16:30:16 Quit B4gder (Quit: It is time to say moo)
16:30:24funmanplush: just ask your question
16:30:38funmanpeople with knowledge of usb code read the irc logs
16:30:58plushit's not a simple question, it will require some discussion
16:31:02plushbut it's worth trying i guess
16:32:41plushon an ipod4g, the rockbox usb stack offers mass storage mode
16:32:55plushunfortunately, this has been broken for a very long time (it used to work)
16:33:00plushmy analysis is as follows:
16:33:09plush* initially, the mass storage driver is disabled
16:33:39[Saint]IPod 4G, the Color/Photo?
16:33:53[Saint]USB works fine for me on svn head...
16:34:21plush[Saint]: grayscale
16:34:30plushi can't tell whether that makes any difference
16:34:35plushboth are PortalPlayer devices
16:34:46plushi actually think it has to do with tue OS you are using
16:34:53plushthere is a race condition in the rockbox code
16:34:59plushand my OS happens to trigger that
16:36:00plushwith my OS, the following happens:
16:36:21plushor let me try it a different way:
16:36:30[Saint]I have the Color 4G, and mine works fine...but its fairly non-standard. Its got a 64GB CF card, and Bluetooth & micro-USB :)
16:36:37plushafter plugging in, rockbox needs to tell whether it's dealing with a dumb charger or with a real host
16:36:56plushthere are different ways of detecting this
16:37:11funmanplush: what is your OS?
16:37:16plushif you plug your device into a real host, rockbox will take the first hint it can get to recognize the host
16:37:18plushfunman: freebsd
16:37:31plushin my case, that happens to be a request for the device descriptor
16:37:48funmanyou tried on another OS (windows/osx/linux)?
16:37:50plush(HEAD usb_core.c lines 563-565)
16:38:03plushfunman: i have a lot of different boxen. all run freebsd.
16:38:25funmanget a ubuntu livecd or live usbkey
16:38:43plushnot possible. ubuntu live usb keys can only be made under windows :(
16:38:49plushi always found that incredibly annoying
16:38:54plushyou need to use this unetbootin thing
16:38:58plushand that runs under windows only :(
16:39:05plushi never understood why they do this either...
16:39:34funman you can do that from linux / osx and probably unix but that's not the point
16:39:57plushonly from within windows/osx/linux, none of which i have
16:39:59funmanjust try on linux/osx/windows because rockbox usb code is known to work on them
16:40:12plushbut the code has a race condition
16:40:15plushand freebsd uncovers it
16:40:28plushi'm not blaming anyone for not noticing earlier
16:40:33plushbut it would be nice to fix it
16:40:48plushah, i have an old kubuntu install on one of my boxen :)
16:40:51funmani think tis part was added recently
16:41:02plushit used to work
16:41:06pamauryplush: why do you call that a race condition ?
16:41:10plush(i meant usb disk mode)
16:41:18plushpamaury: here is why...
16:41:31plush... the code i quoted above enables the usb mass storage driver
16:42:14plushbut allocate_interfaces_and_endpoints() has run already
16:42:24plushand it's this function that allocates endpoints for all enabled drivers
16:42:39plushso the mass storage driver is started but the endpoints it needs are missing
16:42:49plushhence, its descriptor contains all-zero endpoints
16:42:58 Join dfkt [0] (dfkt@unaffiliated/dfkt)
16:43:04plushthat's an invalid descriptor that leads to a non-working usb mass storage device
16:43:34funmanusb_drv_usb_detect_event() ?
16:44:07funmanthis call was added by jhMikeS in usb detect rework
16:44:15plushfunman: yes, usb_drv_usb_detect_event() comes after allocate_interfaces_and_endpoints() - that's the problem
16:44:50plushok, i tried it with linux now
16:44:54plushas expected, everything worked
16:45:01plushlinux does not trigger the race
16:45:02pamauryfunman: that a call to the underlyin usb driver which forwards to usb_core again, that's was present before the rework but jhMikeS might have modified it, although I don't see what is freebsd specific in the reasoning
16:45:32funmanplush: you want to talk with jhMikeS and perhaps open a FS# bug
16:45:51plushpamaury: the freebsd-specific thing is the very early request for the device descriptor
16:45:59funmanpamaury: i dont see it before r29068
16:46:12plushthis usb_drv_usb_detect_event() call happens before rockbox is actually prepared to handle it
16:46:28plushi tried to compile versions < 30000 to see when exactly it broke... but configure failed with perl errors
16:46:35plushi can't go back that far in the code :(
16:46:47pamauryfunman: it must have changed somehow but we already supported detect by core I'm pretty sure
16:47:48funmanplush: try before the rev i gave
16:47:55funmansee the log at that time perhaps
16:48:10plushfunman: i did try. it just won't compile :(
16:48:14plushor rather, won't even configure
16:49:28gevaertsThis sounds like what TheSeven has been talking about recently
16:49:29funmanplush: hm what is the problem exactly? i don't know perl but we can surely figure something
16:49:44plushok, downgrading to 29000 now
16:50:01plushit was something about i think
16:50:25plush"Can't locate in @INC"
16:50:33plushit's some deprecated package that no longer exists
16:52:19plushmy bad
16:52:27plushCGI/ is old and unsupported
16:52:37plushplain i simply didn't have installed
16:52:39plushinstalling now
16:53:17*gevaerts looks at the usb code
16:53:58plushgevaerts: i can go into more detail about what calls what in what order
16:54:06plushi have been tracing through this for the past 5 hours
16:55:37***Saving seen data "./dancer.seen"
16:57:15plushoh by the way, is it possible to compile the core firmware without all games and other plug-ins somehow?
16:57:23pamauryplush: make bin
16:57:32gevaertsThe flow is a bit messy unfortunately
16:57:35pamauryplush: can you go into a bit more details ? That would avoid running the code, perhaps we can see how to fix that
16:57:57plushpamaury: i will, in a moment though... my code is at rev 29000, not HEAD at the moment :)
16:58:23 Join James [0] (2ea48860@gateway/web/freenode/ip.
16:58:36 Quit James (Client Quit)
16:59:11plushbooting firmware 29000 now
16:59:22plushyep, usb mass storage works
16:59:26plush(under freebsd)
16:59:42plushshout i try 29068?
17:00:21gevaertsplush: "Enable for SA9200 only at this time" says no I think :)
17:00:37plushgevaerts: yes, i just looked at the commit message :)
17:03:21plushhow about 29156?
17:03:29plushwould that apply to an ipod4g?
17:03:46funman29068 is a commit frm jethead for me
17:04:03gevaertsThe way I read things, PP (and thus ipod4g) only got converted in r30546
17:04:26gevaertsI could be wrong though
17:05:05gevaertsStill, I'm not sure if we care when exactly this happened. Changing the detection mode *was* the right thing to do, we just have to make the new one work right
17:05:21pamauryI will try to run it on my device, I don't remember the expected flow, I need to remember
17:05:47*gevaerts didn't remember the expected flow back when he was implementing it the first time :)
17:06:25plushgevaerts: yes, 30546 is what i suspect as well
17:06:46plushi won't be trying that one for the moment. i will return to HEAD instead and give you more details of what's failing where
17:07:25pamauryr30546 cannot be the only culprit, arc driver works very well, so it's related on the detection change
17:07:39plushyes, absolutely
17:07:57plushthe detection kicks in at a moment when the rest of the driver isn't expecting it
17:08:12plushi hope i can describe it well enough to make it obvious:
17:08:21plushusb_thread() from usb.c is in charge
17:08:28gevaertsAm I right in suspecting that allocate_interfaces_and_endpoints() and friends run more than once?
17:08:37plushgevaerts: no, only once
17:08:39plushbut too early
17:08:48plushthis function runs *before* any drivers are enabled
17:08:56plushhence, it fails to allocate any endpoints at all
17:09:33gevaertsOh, I see...
17:09:44plushgevaerts: look at usb_core_control_request_handler() in usb_core.c
17:09:55plusha control request comes in
17:10:02gevaertsBasically usb_configure_drivers() (in usb.c) should run on plug in, not on detect
17:10:06gevaertsI think
17:10:06plushallocate_interfaces_and_endpoints() is called in line 839
17:10:23plushthe driver gets enabled by the call in line 844
17:10:27plushand so the order ends up wrong
17:10:55*gevaerts nods
17:11:05plushthe thing with usb_configure_drivers() is that it does different things for USB_POWERED and USB_INSERTED
17:11:16plushif you want to preserve that, you need to wait until you know whether you have a host
17:11:28plushso it's not trivial to fix while preserving this functionalitx
17:11:30gevaertsWell, hm, not sure actually. The device descriptor doesn't need enabled drivers yet
17:11:37plushit does
17:11:51plushbecause the device descriptor refers to the drivers
17:11:56pamaurythe code works on the fuze+ which has the same driver, so it's not so trivial
17:12:04plushe.g. the serial number contains a bitmask of enabled drivers
17:12:16plushthe code works on ipod4g as well
17:12:33plushit's just that allocate_interfaces_and_endpoints() can be triggered earlier than it is e.g. by linux/win
17:13:08gevaertsI don't think changing the serial number should break enumeration though
17:13:29gevaertsSo I still don't see the entire picture
17:13:57plushgevaerts: it probably won't break anything in an awful way but it will take away something that exists now
17:14:08plushright now, if you look at the serial number, you can see what devices are enabled
17:14:26gevaertsOh, sure, it's a bug all right.
17:14:29plushwhich means that before rockbox can send you a device descriptor, it must figure out which devices to enable
17:14:33 Join dreamlayers [0] (
17:14:33 Quit dreamlayers (Changing host)
17:14:33 Join dreamlayers [0] (~bgjenero@rockbox/developer/dreamlayers)
17:15:01plushif you want the information to remain in there, devices must be enabled before a descriptor is sent
17:15:03gevaertsThe serial number variation was added because some systems don't handle same serial number and different configuration well
17:15:15plushi was thinking the following may work:
17:15:20gevaertsI just don't see how this explains what you're seeing
17:15:36plushah, ok, i will explain further
17:15:46plushforget the serial number. this one is unrelated to the bug i am seeing
17:15:51 Join kevku [0] (x@2001:470:28:773::)
17:16:04plushi was just giving reasons for why a trivial fix may not be the right thing
17:16:20plushwhat i am seeing is entirely due to the detection code running things in the wrong order
17:16:39plushallocate_interfaces_and_endpoints() should run only after the list of required drivers is known, never before
17:16:44plushif run before, it can't do its job
17:17:34plushso i think a fix could be this: run a version of allocate_interfaces_and_endpoints() on plug-in that enables whatever is needed for charging; run a second version when a host is detected that adds any additional endpoints needed by actual usb drivers
17:18:08plushin both cases, allocate_interfaces_and_endpoints() could be called from usb_configure_drivers()
17:18:12pamauryit's not that simple, the allocate_interfaces_and_endpoints is called several times, because the normal usb workflow is: reset, get desc (8 bytes), reset, get full desc
17:18:21gevaertsThe way I read things allocate_interfaces_and_endpoints() is called when a control request comes in and we're not in ADDRESSED yet
17:18:41plushmy logf traces show this order of commands received:
17:18:48plushusb_core: SET_ADR 6
17:18:54plushusb_core: GET_DESC 1
17:18:58plushusb_core: GET_DESC 3
17:19:09gevaertsIt does a SET_ADDRESS as the very first call?
17:19:09plushand it's the second call that triggers allocate_interfaces_and_endpoints()
17:19:14plushyes, it does
17:19:22pamaurythat's not a very standard usb host
17:19:28gevaertsok, right, that's unusual
17:19:32plushthis is the freebsd usb stack
17:19:37gevaertsStill allowed though as far as I can tell
17:20:08gevaertsBut yes, then I see what's going on
17:20:10plushthis is why i am seeing the bug and others are not... the freebsd usb stack does its own thing :)
17:20:15pamaurywhat is unexpected is that it never resets
17:20:24pamaurythat's why it doesn't work
17:20:31plushit may reset. i may not have added the right logf()
17:20:45pamauryno, it would SET_ADR again then
17:21:20plushok, so then i guess it does not reset
17:21:30plushi asked about the whole problem on freebsd's usb mailing list first
17:21:36plushthe usb stack maintainer is very responsive
17:21:42gevaertspamaury: do you see a rason (apart from the work involved of course) not to call allocate_interfaces_and_endpoints() on plugin (i.e. electrical connection detected)?
17:21:49plushif the freebsd stack is doing something invalid, i am sure he'll be happy to fix it
17:22:06plushbut if the behavior is allowed by the standard, the bug is in rockbox
17:22:11gevaertsI'm pretty sure it is valid. It's just not what the others are doing
17:23:09plushyeah, i fully expected something like that
17:23:21plushthere must be a reason i am seeing this and others are not
17:23:40plushgevaerts: as for your idea to run allocate_interfaces_and_endpoints() early - how will that help if drivers are not enabled yet()
17:23:55plushwithout drivers enabled, allocate_interfaces_and_endpoints() is a no-op
17:23:57pamaurythe expected workflow is the following: configure_driver(USB_INSERTED), GET_DESC, configure_driver(USB_HOSTED), reset, GET_DESC
17:24:12gevaertsOh, wait. I was confused. The driver enable function has to be called, not allocate_interfaces_and_endpoints()...
17:24:32dreamlayersIn Windows and Linux, I have problems if the hard drive on my 5G iPod is spun up when connecting USB. Is that related?
17:25:13plushgevaerts: yes, *that* one i agree with
17:25:36plushif you enable the mass storage driver early, allocate_interfaces_and_endpoints() will do the right thing when it's called later
17:25:56plushthe only thing is that if you just plugged into a charger, you now needlessly enabled the mass storage driver
17:26:08plushso the question is whether that may have negative consequences
17:26:38gevaertsYou don't, really. It just sets a flag
17:26:54gevaertsThat flag is rather important, but on its own it doesn't do much
17:27:14plushyep, i saw that the function call does nothing but set a flag
17:27:32plushif the flag on its own doesn't do anything much, maybe enabling it is the right thing to do
17:29:42*pamaury would advise extra care because of the multitude of usb variants that we have
17:30:31*jhMikeS 's been lurking. the driver enabling could be synchronous with the setup packet event
17:31:10plushjhMikeS: but then a fresh run of allocate_interfaces_and_endpoints() would have to be forced right after enabling the driver
17:31:26plushwithout a reset, it would never be run after enabling the driver otherwise
17:31:44*gevaerts test-compiles
17:31:49plushgevaerts: so am i
17:32:02plushi added a usb_core_enable_driver(USB_DRIVER_MASS_STORAGE, true); right after usb_stack_enable(true); on plug-in
17:32:49pamaurya first more correct step would be to call usb_drv_usb_detect on any setup packet the first time, even before doing anything else, and propagate changes in the stack synchronously, so that everything is ready when allocating the endpoints but that will require significant work I'm afraid
17:32:54jhMikeSplush: I still might not be 100% clear on the happenings.
17:34:11gevaertsjhMikeS: I'm pretty sure we currently rely on SET_ADDRESS not being the first call, which mean allocate_interfaces_and_endpoints() is run a second time *after* everything has been set up, so the right thing ends up happening
17:35:38 Join MethoS- [0] (~clemens@
17:35:51gevaertsAnd freebsd apparently breaks that assumption
17:36:26plushgevaerts: rockbox is actually the first one of two usb mass storage devices that fail for me under bsd
17:36:30plushthe second one is closed source
17:36:35gevaertsI'm reasonably sure that my patch doesn't break things
17:36:36plush*that* one will be fun to work around :(
17:37:06pamaurywhat was the initial reason of making configure_drivers depend on the usb status ?
17:37:43gevaertspamaury: it doesn't really have to depend on the usb status. It just has to be run at connect time because of the button check
17:38:30jhMikeSgevaerts: so it's a problem of host being detected later rather than sooner, after actual communication has taken place from a host, but not via the descriptor?
17:38:53jhMikeS* to get the descriptor
17:39:04gevaertsbasically, yes
17:39:17pamauryI don't agree, it's because our stack relies on a reset
17:40:11gevaertsAnother fix (but a horrible one) could be to change usb_core_control_request_handler() to keep a counter so it always calls allocate_interfaces_and_endpoints() and friends at least twice
17:40:16gevaertspamaury: do we?
17:40:48jhMikeSposting several USB_HOSTED events in a row if any sort of host-like activity takes place shouldn't bother it, but if there are other assumptions that could break things...
17:41:06pamauryif you look at the code, usb_status is DEFAULT initially, on the very first setup packet, it allocates endpoints, then detect hosts takes place, that's complete independent of the content of the first packet at this point
17:41:40plushgevaerts: your patch doesn't work for me
17:41:42plushno idea why
17:41:48plushmy simpler hack *did* work
17:41:55*jhMikeS did ask gevaerts before if the setup packet request was to be the first packet and he affirmed that :)
17:41:59plushwith your patch, when i plug in the usb cable, nothing happens :(
17:42:03dreamlayersI can't get a USB connection after applying . There is a bit of disk activity, but the USB screen doesn't appear.
17:42:18pamauryhmm wait, stricly speaking that's not true, it relies on the fact that the host will read the descriptor twice perhaps
17:42:46gevaertsjhMikeS: the setup packet is first, it just doesn't have to be GET_DESCRIPTOR
17:42:54plushpamaury: i think it depends on detecting the host before GET_DESCRIPTOR happens
17:44:00plushactually, no - it depends on detecting that a host is present before allocate_interfaces_and_endpoints() is run
17:44:03pamauryplush: that's impossible, we detect the host on the first packet, allocate_interfaces_and_endpoints needs to be called twice for the stack to work, anyway
17:44:21plushmaking yourself known as a host, then *resetting* works - it causes allocate_interfaces_and_endpoints() to run with existing knowledge of a host
17:44:42gevaertsplush: your change was basically in the same place as my new usb_configure_drivers() call, right?
17:44:44plushwith my hack, i think allocate_interfaces_and_endpoints() runs only once
17:44:53plushgevaerts: yes, hold on, i will make a patch
17:45:55plushthat's all i changed
17:46:09plushit's a hack - but it makes the device work for me
17:47:06gevaertsplush: I'm not sure charge-only will still work for you then :)
17:47:18plushi don't have any chargers around :(
17:47:25plushi'd try otherwise, of couse
17:47:38plushand yes, i am sure it's a nasty hack
17:47:44gevaertsI mean by holding a button
17:47:45plushi don't understand why your patch doesn't work
17:47:51plushah, like that
17:47:59plushyeah, that's probably broken, true
17:48:04funmani need to send plush a amsv2 device
17:48:16funmanand tell him that HID works everywhere except on freebsd ;)
17:48:49gevaertsplush: can you replace the call to usb_exclusive_storage() in my patch to just checking the exclusive_storage_access variable?
17:48:57plushgevaerts: oddly enough... charge-only works?!?
17:49:06plushi just tried. holding a button does the right things
17:49:29plushgevaerts: will do. but dinner is ready. my girlfriend won't let me continue hacking without a 15 minute food break
17:49:36 Quit dreamlayers (Quit: Leaving)
17:50:02*gevaerts will also try it
17:57:39 Quit bluefoxx (Quit: brb rebooting)
18:00:58jhMikeSthat will cause a SYS_USB_DISCONNECTED broadcast when none should have occurred
18:02:55gevaertsjhMikeS: maybe test for USB_INSERTED there as well, i.e. what the usb_exclusive_storage() function does?
18:03:33gevaertsfor the broadcast anyway. We obviously should reset the flag
18:03:55*plush is back
18:04:34jhMikeSgevaerts: test where?
18:05:14gevaertsjhMikeS: in usb_release_exclusive_storage()
18:08:12 Join remlap1 [0] (
18:08:17jhMikeSI thought about just detecting any communication whatsoever as a host, ie. tranfer completion since it reaches that code through there
18:08:22 Quit remlap1 (Client Quit)
18:09:41 Quit remlap (Ping timeout: 240 seconds)
18:11:10gevaertsjhMikeS: What exactly would you do then?
18:12:02gevaertsApart from the SYS_USB_DISCONNECTED thing the second version of my patch seems to work fine
18:13:56jhMikeSset up the drivers in USB_TRANSFER_COMPLETION, just before telling the driver? maybe even simplify as a result
18:14:39gevaertsAh, yes. That could work
18:14:59plushjhMikeS: i am trying to do just that
18:15:11plushthing is, i can't find a flag to test for so that it's not done twice
18:15:12 Quit bimbel (Quit: Verlassend)
18:15:21plushmaybe a new int flag will have to be added
18:15:27jhMikeSsome finesse is needed
18:15:35plushi was going to add it to the top of usb_core_control_request_handler()
18:15:44jhMikeSbut, it could get rid of the need for the event handler and message too
18:16:10gevaertsplush: usb.c seems cleaner, because it actually knows which drivers are needed
18:18:41plushi have no idea where the boundary between usb.c and usb_core.c should be. the two seem to call each other a lot
18:19:07gevaertsTrue. We'd like to clean this up :)
18:20:29plushi wrote the freebsd driver for wacom tablets
18:20:37plushthe linux driver has two files, similar to usb.c and usb_core.c
18:20:46plushi never understood the distinction between their two files
18:20:50plushso my driver has just one :)
18:22:37gevaertsBasically usb.c handles connection state. It's also used for devices that have a hardware USB-ATA bridge. usb_core.c handles the actual usb protocol
18:22:39 Join Keripo [0] (
18:22:56gevaertsIt (ab)uses the usb.c thread because it seems silly to add another thread
18:23:12plushare there any devices without USB_DETECT_BY_CORE?
18:23:21 Join Buschel [0] (
18:24:18Tornethere were.. did we fix them all? :)
18:24:31 Quit lebellium (Quit: ChatZilla 0.9.88 [Firefox 10.0/20111221135037])
18:25:06gevaertsNot yet
18:26:15gevaertsUSB_DETECT_BY_CORE is used by USBOTG_ARC, USBOTG_AS3525, USBOTG_AS3525v2, and USBOTG_RK27XX. That leaves USBOTG_ISP1362 and USBOTG_ISP1583 (which I think aren't actually implemented), USBOTG_JZ4740, USBOTG_M5636, USBOTG_M66591 and USBOTG_S3C6400X
18:27:29gevaertsi.e. nano2g, classic, mr500, onda, zvm
18:32:20 Join lorenzo92 [0] (
18:32:32 Quit efyx (Remote host closed the connection)
18:32:51jhMikeSit might just evaporate anyway :)
18:33:28lorenzo92kugel: was studying the si470x's very interesting and I found 2 interesting gpio pins for the radio. But there is something curious: an internal struct that *could* be used directly in a ioctl call!
18:33:49lorenzo92kugel: directly, I mean with the same parameters sizes etc
18:34:51 Join remlap [0] (
18:35:04 Quit Buschel (Quit: ChatZilla 0.9.87 [Firefox 9.0.1/20111220165912])
18:43:16gevaertsTorne: are there status updates for git, or are we just waiting for everyone involved to be available at the same time?
18:43:19 Quit [Saint] (Ping timeout: 268 seconds)
18:43:53Torneuhm, yeah. nothing has happened; i've been at my parents for christmas and/or drunk/high for the holidays
18:44:24Tornewe'll sort stuff real soon :)
18:48:43 Join Jiri [0] (
18:49:08jlbiasiniHello all:
18:49:34 Quit Jiri (Client Quit)
18:50:02jlbiasiniwhat should I do for stuff that are not implemented for fuze+ yet? Should I let and clean section for future implementation or cut those section out?
18:50:18jlbiasiniI meant in the manual
18:50:48jlbiasiniconcerned recording, radio...
18:54:02funmanjlbiasini: is there something special about recording and radio with the fuze+ ?
18:54:16funmani'd have thought those sections are shared across models when recording/fm features are defined
18:54:59jlbiasinifunman: yes there is something very special: not implemented yet :D
18:55:25jlbiasiniI would be in the future I guess but for the moment that doesn't work
18:55:30jhMikeSdisgustingly crude but worth checking:">
18:55:41***Saving seen data "./dancer.seen"
18:56:09curtismyou know rockbox's scrobbler? what software can i use to upload that information to
18:57:16jlbiasiniSo am I suppose to mention it in the manual? Or cut out section or just do as if it was working?
18:57:29funmanjlbiasini: that's not special. just don't declare teh features
18:59:03jlbiasinifunman: ok but this means changing export/config/sansafuzeplus.h ? This is not done in the manual I think, is it,
18:59:06funmanjlbiasini: i think the manual features are generated from config/sansafuzeplus.h so if HAVE_TUNER is not defined, the radio sections won't appear. did you try to build the manual already?
18:59:30funmancurtism: it should be explained in the wiki and the manual
18:59:37jlbiasiniright I think that it is so
19:00:00curtismfunman: "RTFM"? thanks
19:00:01funmanjlbiasini: right fix would be to undef HAVE_RECORDING and CONFIG_TUNER from config
19:00:10funmancurtism: yeah :)
19:00:15jlbiasiniok thx
19:00:28funmancurtism: i think there's multiple software you can use btw
19:06:02 Quit marcol07 (Quit: CGI:IRC (Ping timeout))
19:08:03kugellorenzo92: what do yyou mean?
19:08:18kugelsure the kernel module uses the same struct as the ioctl() caller
19:08:27 Join Strife89 [0] (~Strife89@
19:08:29funmankugel: ping
19:08:36kugelfunman: pong
19:08:43funmankugel: can you take a look at FS #12475 ?
19:08:44fs-bluebot 3Crash while playing audio (bugs, unconfirmed)
19:09:02lorenzo92kugel: I was disassembling the radio module, interesting, but difficult to understand how it works. The only thing I know is that there is a 54 byte sized internal struct for various parameters
19:11:41lorenzo92kugel: BUT it should be quite easy (maybe for you) to rewrite a similar module with just register read/write.....
19:12:00 Quit Keripo (Quit: Leaving.)
19:12:03lorenzo92initialization code seems to be pretty easy
19:12:26kugelit would be nice if we can do it without extra kernel modules
19:15:50 Quit Strife89 (Ping timeout: 248 seconds)
19:17:58 Quit ze (Ping timeout: 248 seconds)
19:18:31jlbiasinipamaury: did you see that there is a #define HAVE_TUNER_PWR_CTRL in export/config?
19:19:04 Join dreamlayers [0] (~bgjenero@rockbox/developer/dreamlayers)
19:19:23jlbiasinicould it be the reason it doesn't work on fuze+ (it's not enable on the fuze+
19:21:42pamauryjlbiasini: nice catch, can you #define it and see ?
19:23:06 Join ze [0] (
19:24:31lorenzo92kugel: the strange thing is that I cannot find a copy_from_user call in the ioctl section uhm uhm
19:25:32pamauryjlbiasini: otherwise I'll try later
19:25:52kugellorenzo92: oh hmm
19:28:28 Join Strife89 [0] (~Strife89@
19:29:46jlbiasinipamaury: yeah I will try
19:29:50lorenzo92kugel: ah btw, I know why (probably) we have more than a simple integer as ioctl parameter for gpio: in the iomux call, there is a second parameter that sets some options (see kernel's source)
19:31:44kugellorenzo92: intersting
19:32:34funmankugel: or is there a way to debug malloc failures?
19:33:55 Join JesusFreak316 [0] (
19:36:05 Quit bluebrother (Disconnected by services)
19:36:06 Join bluebrother^ [0] (~dom@rockbox/developer/bluebrother)
19:36:09kugelfunman: what do you mean by "malloc failure"?
19:38:28 Quit fs-bluebot (Ping timeout: 240 seconds)
19:39:09 Nick eintopf|28c3 is now known as eintopf (
19:39:20 Quit mortalis (Quit: KVIrc 4.1.1 Equilibrium
19:40:02 Join fs-bluebot [0] (
19:41:00funmankugel: FS #12475 ...
19:41:00fs-bluebot 3Crash while playing audio (bugs, unconfirmed)
19:41:47kugelfunman: I saw that
19:42:09funmani think it's a malloc problem
19:42:16funmanjdgordon does too
19:42:22kugeldoesnt tell me what you mean by "malloc failure"
19:42:51kugelbesides we still have no "malloc", even with buflib. and I prefer to not call it malloc
19:44:13funmani've seen the string "malloc" on panic/data abort screen on clipv1
19:44:22funmanwhen testing usb
19:45:04kugelokay, right, that's a very old panicf in playback.c
19:45:10 Join liar [0] (
19:45:20kugelit said malloc even before buflib
19:45:33funmanwhat i called 'malloc failure' was allocation of a buffer which would be too big where skins code would write
19:45:41funmanand overwrite other memory
19:45:55funmanwell i just want a hand to lead me in the right direction to debug this
19:47:27kugelfunman: you're saying the skin is too big and makes allocating the audio buffer go wrong?
19:48:46 Join JesusFreak316_ [0] (
19:49:08 Quit JesusFreak316 (Ping timeout: 240 seconds)
19:49:37funmani'm not saying i'm making random guesses but yeah
19:51:03kugelI thought I prevented that in the shrink_callback
19:51:23kugelperhaps the skin is loaded before the audio buffer is allocated for the first time?
19:51:37kugelfunman: is this reproducable in the sim?
19:53:37jlbiasinipamaury: no change
19:53:43funmankugel: i didnt try
19:53:54jlbiasiniI will try with updated bootloaderthen
19:53:57funmanthe sim doens't have usb plug, does it?
19:54:50pamauryjlbiasini: it won't change anything, I'll see that later, thanks
19:55:01kugelfunman: press the u key
20:02:08 Quit Topy44 (Ping timeout: 240 seconds)
20:03:13 Join Topy44 [0] (
20:04:03jlbiasinipamaury: anyway sansa c200v2 seens to use the same define TUNER_CONFIG device S4700 and explicitly comment out HAVE_TUNER_PWR_CTRL
20:04:27pamaurystrangely enough, I seem to remember that tuner_power is actually called, so that's strange
20:05:59 Quit fyrestorm (Read error: Connection reset by peer)
20:06:10 Join fyrestorm [0] (
20:08:14 Join saratoga [0] (9803ec71@gateway/web/freenode/ip.
20:08:38saratogadoes anyone know more about the problems with the m200v4?
20:08:39 Quit pamaury (*.net *.split)
20:08:40 Quit ender` (*.net *.split)
20:08:40 Quit Galois (*.net *.split)
20:08:40 Quit bieber (*.net *.split)
20:08:40 Quit bzed (*.net *.split)
20:08:40 Quit Rondom (*.net *.split)
20:08:40 Quit kkit (*.net *.split)
20:08:40 Quit pjm0616 (*.net *.split)
20:08:40 Quit ack (*.net *.split)
20:16:43 Quit dreamlayers (*.net *.split)
20:16:43 Quit Xerion (*.net *.split)
20:16:43 Quit amithkk (*.net *.split)
20:16:43 Quit jae (*.net *.split)
20:16:43 Quit Utchybann (*.net *.split)
20:19:19 Quit y4n (*.net *.split)
20:22:37 Quit Sundiver (Ping timeout: 240 seconds)
20:23:13 Quit Misan (*.net *.split)
20:23:13 Quit plush (*.net *.split)
20:23:13 Quit parafin (*.net *.split)
20:23:13 Quit MethoS- (*.net *.split)
20:23:13 Quit factor (*.net *.split)
20:23:13 Quit pixelma (*.net *.split)
20:23:13 Quit amiconn (*.net *.split)
20:23:14 Quit mikroflops (*.net *.split)
20:23:14 Quit zu_ (*.net *.split)
20:23:14 Quit yosafbridge (*.net *.split)
20:23:14 Quit trompette (*.net *.split)
20:23:14 Quit knittl (*.net *.split)
20:23:14 Quit liar (*.net *.split)
20:23:15 Quit adnap (*.net *.split)
20:23:15 Quit Torne (*.net *.split)
20:23:15 Quit markun (*.net *.split)
20:23:15 Quit beta2k (*.net *.split)
20:23:15 Quit guymann (*.net *.split)
20:23:15 Quit tmzt (*.net *.split)
20:23:15 Quit Bawitdaba (*.net *.split)
20:23:15 Quit alexbobp (*.net *.split)
20:23:15 Quit linuxguy3 (*.net *.split)
20:23:15 Quit dionoea (*.net *.split)
20:23:15 Quit dfkt (*.net *.split)
20:23:15 Quit TheSeven (*.net *.split)
20:23:15 Quit dv_ (*.net *.split)
20:23:15 Quit tah (*.net *.split)
20:23:15 Quit froggyman (*.net *.split)
20:23:15 Quit Bagder (*.net *.split)
20:23:15 Quit preglow (*.net *.split)
20:23:15 Quit lmh (*.net *.split)
20:23:15 Quit topik (*.net *.split)
20:23:16 Quit nsx (*.net *.split)
20:23:16 Quit ze (*.net *.split)
20:23:16 Quit n1s (*.net *.split)
20:23:16 Quit nosa-j (*.net *.split)
20:23:17 Quit Zambezi (*.net *.split)
20:23:17 Quit rasher (*.net *.split)
20:23:17 Quit GodEater (*.net *.split)
20:23:17 Quit nomada (*.net *.split)
20:23:17 Quit ruskie (*.net *.split)
20:23:17 Quit desowin (*.net *.split)
20:23:17 Quit gevaerts (*.net *.split)
20:23:17 Quit simabeis (*.net *.split)
20:23:17 Quit ukleinek (*.net *.split)
20:23:17 Quit kevku (*.net *.split)
20:23:17 Quit Rob2223 (*.net *.split)
20:23:17 Quit user890104 (*.net *.split)
20:23:17 Quit kugel (*.net *.split)
20:23:17 Quit soap (*.net *.split)
20:23:17 Quit TBCOOL (*.net *.split)
20:23:17 Quit ps-auxw (*.net *.split)
20:23:18 Quit funman (*.net *.split)
20:23:18 Quit qnm (*.net *.split)
20:23:18 Quit jordan`` (*.net *.split)
20:23:18 Quit Misanthropos (*.net *.split)
20:23:18 Quit tmr (*.net *.split)
20:23:18 Quit Farthen (*.net *.split)
20:23:18 Quit rjg (*.net *.split)
20:23:18 Quit mystica555 (*.net *.split)
20:23:18 Quit Elfish (*.net *.split)
20:23:18 Quit ranmachan (*.net *.split)
20:23:18 Quit Naked (*.net *.split)
20:23:18 Quit ved (*.net *.split)
20:23:18 Quit freqmod (*.net *.split)
20:23:18 Quit Topy44 (*.net *.split)
20:23:18 Quit JesusFreak316_ (*.net *.split)
20:23:18 Quit fs-bluebot (*.net *.split)
20:23:18 Quit Strife89 (*.net *.split)
20:23:18 Quit remlap (*.net *.split)
20:23:18 Quit lovasoa (*.net *.split)
20:23:18 Quit einhirn (*.net *.split)
20:23:18 Quit bertrik (*.net *.split)
20:23:19 Quit mystica555_ (*.net *.split)
20:23:19 Quit sinthetek (*.net *.split)
20:23:19 Quit curtism (*.net *.split)
20:23:19 Quit advcomp2019 (*.net *.split)
20:23:19 Quit Guinness` (*.net *.split)
20:23:19 Quit mc2739 (*.net *.split)
20:23:19 Quit jhMikeS (*.net *.split)
20:23:19 Quit tchan (*.net *.split)
20:23:19 Quit Slasheri (*.net *.split)
20:23:19 Quit Unhelpful (*.net *.split)
20:23:19 Quit polemon (*.net *.split)
20:23:19 Quit lembas (*.net *.split)
20:23:20 Quit aevin (*.net *.split)
20:23:20 Quit martii (*.net *.split)
20:23:21 Quit jlbiasini (*.net *.split)
20:23:21 Quit n17ikh (*.net *.split)
20:23:21 Quit FOAD (*.net *.split)
20:23:21 Quit orzo (*.net *.split)
20:23:21 Quit Staphylo (*.net *.split)
20:23:21 Quit perrikwp (*.net *.split)
20:23:21 Quit thegeek (*.net *.split)
20:23:21 Quit scorche (*.net *.split)
20:23:21 Quit bug2000 (*.net *.split)
20:23:21 Quit Poodlemastah (*.net *.split)
20:23:21 Quit antil33t| (*.net *.split)
20:23:21 Quit Sleepy_Coder (*.net *.split)
20:23:21 Quit scorche|sh (*.net *.split)
20:24:25 Join Sundiver [0] (
20:24:25 Join y4n [0] (y4n@unaffiliated/y4ndexx)
20:24:25 Join Utchybann [0] (
20:24:25 Join jae [0] (
20:24:25 Join amithkk [0] (u4289@2buntu/writers/amithkk)
20:24:25 Join Xerion [0] (
20:24:25 Join ack [0] (
20:24:25 Join pjm0616 [0] (~user@
20:24:25 Join kkit [0] (
20:24:25 Join Rondom [0] (~rondom@2a01:488:66:1000:b24d:4f2f:0:1)
20:24:25 Join bzed [0] (
20:24:25 Join bieber [0] (
20:24:25 Join Galois [0] (
20:24:25 Join ender` [0] (
20:24:25 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
20:24:25 Join Topy44 [0] (
20:24:25 Join JesusFreak316_ [0] (
20:24:25 Join liar [0] (
20:24:25 Join fs-bluebot [0] (
20:24:25 Join Strife89 [0] (~Strife89@
20:24:25 Join ze [0] (
20:24:25 Join remlap [0] (
20:24:25 Join MethoS- [0] (~clemens@
20:24:25 Join kevku [0] (x@2001:470:28:773::)
20:24:25 Join dfkt [0] (dfkt@unaffiliated/dfkt)
20:24:25 Join jlbiasini [0] (
20:24:25 Join lovasoa [0] (~olojkine@
20:24:25 Join Misan [0] (
20:24:25 Join einhirn [0] (
20:24:25 Join n1s [0] (~n1s@rockbox/developer/n1s)
20:24:25 Join bertrik [0] (~bertrik@rockbox/developer/bertrik)
20:24:25 Join perrikwp [0] (
20:24:25 Join curtism [0] (
20:24:25 Join factor [0] (~factor@
20:24:25 Join Rob2223 [0] (
20:24:25 Join nosa-j [0] (
20:24:25 Join thegeek [0] (
20:24:25 Join TheSeven [0] (~TheSeven@rockbox/developer/TheSeven)
20:24:25 Join pixelma [0] (quassel@rockbox/staff/pixelma)
20:24:25 Join amiconn [0] (quassel@rockbox/developer/amiconn)
20:24:25 Join mystica555_ [0] (
20:24:25 Join adnap [0] (
20:24:25 Join user890104 [0] (
20:24:25 Join mikroflops [0] (
20:24:25 Join n17ikh [0] (
20:24:25 Join kugel [0] (~kugel@rockbox/developer/kugel)
20:24:25 Join scorche [0] (~scorche@rockbox/administrator/scorche)
20:24:25 Join Torne [0] (~torne@rockbox/developer/Torne)
20:24:25 Join sinthetek [0] (~sinthetek@unaffiliated/sinthetek)
20:24:25 Join soap [0] (~soap@rockbox/staff/soap)
20:24:25 Join markun [0] (~markun@rockbox/developer/markun)
20:24:25 Join Zambezi [0] (
20:24:25 Join bug2000 [0] (~bug@unaffiliated/bug2000)
20:24:25 Join Sleepy_Coder [0] (majic@unaffiliated/sleepycoder/x-938672)
20:24:25 Join dv_ [0] (
20:24:25 Join advcomp2019 [0] (~advcomp20@unaffiliated/advcomp2019)
20:24:25 Join plush [0] (
20:24:25 Join parafin [0] (
20:24:25 Join rasher [0] (~rasher@rockbox/developer/rasher)
20:24:25 Join FOAD [0] (~foad@unaffiliated/foad)
20:24:25 Join tah [0] (
20:24:25 Join TBCOOL [0] (
20:24:25 Join ps-auxw [0] (~arneb@2001:470:c807:0:1532:4e5f:2ad3:4123)
20:24:25 Join beta2k [0] (
20:24:25 Join tmr [0] (
20:24:25 Join Guinness` [0] (
20:24:25 Join mc2739 [0] (~mc2739@rockbox/developer/mc2739)
20:24:25 Join Poodlemastah [0] (
20:24:25 Join funman [0] (~fun@rockbox/developer/funman)
20:24:25 Join jhMikeS [0] (~jethead71@rockbox/developer/jhMikeS)
20:24:25 Join antil33t| [0] (~Ahurhurr@
20:24:25 Join froggyman [0] (~froggyman@unaffiliated/froggyman)
20:24:25 Join tchan [0] (~tchan@lunar-linux/developer/tchan)
20:24:25 Join zu_ [0] (
20:24:25 Join GodEater [0] (~bibble@rockbox/staff/GodEater)
20:24:25 Join Bagder [241] (~daniel@rockbox/developer/bagder)
20:24:25 Join preglow [0] (
20:24:27 Join guymann [0] (~charles@
20:24:27 Join nomada [0] (
20:24:27 Join lembas [0] (
20:24:27 Join tmzt [0] (~tmzt@
20:24:27 Join Slasheri [0] (miipekk@rockbox/developer/Slasheri)
20:24:27 Join qnm [0] (~qnm@2001:44b8:3110:f300:208:9bff:fec0:179a)
20:24:27 Join ruskie [0] (ruskie@sourcemage/mage/ruskie)
20:24:27 Join jordan`` [0] (~gromit@2001:660:3302:2826:225:90ff:fe20:d9a8)
20:24:27 Join Misanthropos [0] (
20:24:27 Join Unhelpful [0] (~quassel@rockbox/developer/Unhelpful)
20:24:27 Join lmh [0] (lmh@nat/redhat/x-akwpetrhgkztlaee)
20:24:27 Join topik [0] (
20:24:27 Join Farthen [0] (~Farthen@2a01:4f8:101:2a4:0:bc28:b2e1:9)
20:24:27 Join polemon [0] (
20:24:27 Join orzo [0] (
20:24:27 Join Bawitdaba [0] (
20:24:27 Join yosafbridge [0] (
20:24:27 Join knittl [0] (~knittl@unaffiliated/knittl)
20:24:27 Join trompette [0] (
20:24:27 Join dionoea [0] (~dionoea@videolan/developer/dionoea)
20:24:27 Join linuxguy3 [0] (
20:24:27 Join alexbobp [0] (~alex@
20:24:27 Join nsx [0] (
20:24:27 Join ukleinek [0] (~ukl@2001:6f8:1178:2:21e:67ff:fe11:9c5c)
20:24:27 Join simabeis [0] (
20:24:27 Join gevaerts [0] (~fg@rockbox/developer/gevaerts)
20:24:27 Join desowin [0] (
20:24:27 Join freqmod [0] (~quassel@2001:700:300:1430:226:18ff:fe82:1a24)
20:24:27 Join ved [0] (~ved@2001:41d0:1:5914::2)
20:24:27 Join Naked [0] (
20:24:27 Join ranmachan [0] (~ranma@2a01:4f8:130:9321::2)
20:24:27 Join Elfish [0] (amba@2a01:4f8:100:90a1:abc:abc:abc:abc)
20:24:27 Join mystica555 [0] (
20:24:27 Join rjg [0] (
20:24:27 Join martii [0] (
20:24:27 Join aevin [0] (eivindsy@unaffiliated/aevin)
20:24:27 Join scorche|sh [0] (
20:24:27 Join Staphylo [0] (
20:24:38plushworks with linux as well
20:30:50gevaertsA trivial comment to start with. I don't think usb_set_host_present() should be marked inline
20:31:17plushi couldn't figure out why the other functions were
20:31:19plushto save space?
20:31:23plushto reduce call stack size?
20:31:31plusheither way, i just stupidly replicated it :)
20:31:38plushremoved already
20:31:38gevaertsI guessed that :)
20:32:03gevaertsThe others are only called from one place, so it saves a bit of space (just the call itself, really)
20:32:18plushand here, it would be inlined twice
20:32:23plushhence, achieving the opposite effect
20:32:31*gevaerts nods
20:33:02gevaertsI'm not really convinced we want the inline on the others here anyway. The compiler should be smart enough to do the right thing I think (but I may be wrong)
20:33:33plushinline is only a hint anyway
20:33:47plushso with a stupid compiler, no matter how much you do or do not inline, there is not much you can do
20:36:08 Join B4gder [241] (~daniel@rockbox/developer/bagder)
20:39:22 Quit liar (Remote host closed the connection)
20:41:21gevaertsplush: at first sight that looks good to me. I think jhMikeS and pamaury should have a look too though :)
20:41:51plushmy main concern is that some of the drivers were doing additional checking in usb_drv_usb_detect_event()
20:42:02plushif any of this is actually necessary, it should be preserved somehow
20:42:17plushmuch of it looked like it was the same thing, differently implemented and redundant though
20:42:28gevaertsI think that those were remnants from the old mechanism
20:45:14pamauryat first sight it looks good to me, it removes some complexity I wanted to removed so I'm more than happy with it :) As for the checks by the drivers, I would be tempted to say from my implementer point of view that no one really knows what the function was exactly supposed to do :-/
20:45:54gevaertsat least the pp502x one was clearly for the bus reset based detection
20:45:59pamaurythe check for the reset is obsolete (reset was the old mechanism)
20:46:20gevaertsThe others look like different ways to check the same thing to me
20:46:43pamaurythe as3525 seems to prevent calling it if not necessary
20:47:58gevaertsIt also has nice obsolete comments about implementing USB
20:48:12gevaertsSo I'm not sure how relevant that has been recently
20:54:13gevaertsOK, it works for me on e200v1, fuzev2, and gigabeat F, on linux. I'll test on windows in a minute. I'd still like to see testing on a non-DETECT_BY_REQUEST device such as a nano2
20:55:41gevaertsOr rather, I'd like someone to move over all remaining drivers ;)
20:55:45***Saving seen data "./dancer.seen"
20:59:15 Join dreamlayers [0] (
20:59:15 Quit dreamlayers (Changing host)
20:59:15 Join dreamlayers [0] (~bgjenero@rockbox/developer/dreamlayers)
21:00:18 Quit Zambezi (Changing host)
21:00:18 Join Zambezi [0] (~Zulu@unaffiliated/zambezi)
21:00:46jlbiasiniDoes anybody know where plugins keymaps are sets?
21:02:06pamauryin the plugins
21:02:39gevaertsWindows (7) seems OK too
21:11:49 Join Keripo [0] (
21:13:27gevaertspamaury, jhMikeS do we actually need to do anything more than define USB_DETECT_BY_REQUEST to move drivers over after this patch?
21:14:14pamauryimplement usb_drv_usb_detect I think, I can't remember exactly why it wasn't trivial for some others too
21:15:57gevaertsUSB_STATUS_BY_EVENT still exists. That's the one where usb_detect() is involved
21:16:48plushthere is some difference between status and state that i do not understand and did not touch
21:17:30gevaertsI think someone with a nano2g or an onda should just try
21:17:41gevaertsOr a classic
21:18:07gevaertsUSB on mr500 didn't work last week when I tried it, so that's not too useful right now
21:18:32gevaertsAnd zvm never got to an actually usable state I think, so I'm not sure if I'm too worried
21:22:44*kugel has some powermgmt rewrite going
21:23:09saratogadoes anyone actually own an m200v4?
21:23:28gevaertssaratoga: I think scorche may have a few
21:23:55saratogai'm curious about the volume related shutdown, probably something with the voltage regulators that could be fixed
21:24:00lovasoakugel: Did you post your power management patch somewhere?
21:24:23gevaertssaratoga: actually, RockboxTesting says domonoky, jhMikeS and Llorean
21:24:41saratogaif one of those people could comment, i'm curious if it always shuts down, sometimes shuts down, etc
21:24:49saratogai'll see if i can find one on ebay
21:25:03kugellovasoa: not yet
21:25:15kugellovasoa: or do you mean the one for the ypr0?
21:25:36lovasoakugel: I mean for the ypr0, yes
21:26:24kugellovasoa: if you like
21:26:49kugeljust waiting for some battery benches before committing
21:26:55lorenzo92kugel: I have a bench finally :)
21:28:31lorenzo92this is battery discharge while backlight always on!
21:29:00kugellorenzo92: 4200mV is the max voltage?
21:29:17kugel6:33 is quite poor even for backlight
21:29:31lovasoakugel: Merci
21:29:44kugellorenzo92: I'd say so, not sure actually
21:30:09lorenzo92kugel: well it was quite bright the default level, I keep always the first :)
21:30:19kugelwhat current does that correspond to?
21:31:38lorenzo92uhm in the ascodec you mean? it's the very first step
21:31:45lorenzo921,.... ma
21:32:20lorenzo92kugel: yeah 1,2 mA
21:32:31lorenzo92but uhm I guess it uses more anyway
21:32:41lorenzo92ah no srry
21:32:46lorenzo92this is the first step
21:32:52lorenzo92so default is 4
21:33:02lorenzo92so 4,8 mA
21:34:54lorenzo92kugel: it would be very interesting to get the actual amperage from the battery using a multimeter...
21:34:56plushi made a task for the patch
21:35:01lorenzo92kugel: my R0 is still unopened :D
21:35:28 Quit JesusFreak316_ (Ping timeout: 240 seconds)
21:36:45gevaertsplush: did anything change from the pastebin version (apart from the inline)?
21:36:57plushgevaerts: nome, that's the only change
21:37:11plushi spent half an hour trying to explain what the damn thing does though
21:37:13lorenzo92kugel: definitely, R0 won't turn off when discharged (< 3,450...)
21:37:24plushit's not always trivial to describe a code change in words :)
21:37:24gevaertsAlways useful :)
21:37:36kugel92mA, no?
21:38:50lorenzo92ah you mean the average, not only backlight -.-
21:38:53lorenzo92well yeah
21:39:37kugel~68mA for backlight
21:40:12lorenzo92indeed, the backlight has a great impact
21:41:28dfktis it a known issue that some items can be added to the shortcuts list, but don't work? eg. i put the pictureflow demo into shortcuts, but it displays there with the full folder structure as name (/.rockbox/rocks/...), and can't be opened
21:42:06lorenzo92dfkt: I was also wondering how to delete them without messing with the file....
21:42:30dfktlorenzo92, right, that's the other issue - it can't be deleted
21:42:47lorenzo92dfkt: okay I'm not the only one hihi
21:43:37lorenzo92kugel: I cannot do further benchs, well uhm I can do the recharge one if you want!
21:44:07lorenzo92I'll do it...
21:45:21lorenzo92kugel: this time while charging...backlight off or what?
21:45:31lorenzo92kugel: playing music or?
21:47:21jlbiasinipamaury: the patch is now ready
21:48:28gevaertsmr500 really needs someone who cares about it
21:49:44 Quit lovasoa (Read error: Connection reset by peer)
21:51:05kugellorenzo92: a recharge one would be fine
21:51:28lorenzo92kugel: no matter the test conditions?
21:51:32kugelmusic or not doesnt matter
21:55:27kugellorenzo92: keep backlight on
21:55:46lorenzo92okay ;)
21:56:18 Quit Strife89 (Quit: Away!)
21:57:01pamauryjlbiasini: ok thansk, I'm busy right now but I'll commit it soon
21:57:22lorenzo92then after plotting the graph, and finding the slope of the median line, you need to?
21:59:57jlbiasinipamaury: regarding the manual we'll have to undef radio and recording as long as they don't work because else the manual compiler print those section too...
22:00:21nosa-jwhich one should i get for rockbox the new of old sansa fuze? consitering the new one isn't as suported?
22:00:40nosa-jim worried about the thouch controls
22:00:41kugellorenzo92: graphs are nice but not needed
22:00:55jlbiasininosa-j: depend what you use
22:01:27kugellorenzo92: needed are the voltage numbers to map 10%, 20% etc, and the max and min voltage
22:01:31jlbiasininosa-j: we only have radio and recording missing in the fuze+
22:01:43kugeland the discharge time the milliamps
22:02:32jlbiasininosa-j: we have a stable touchpad implementation for now: a 3x3 grid
22:03:19lorenzo92kugel: ah ok ;) anyway I did the graph...but I guess I did a mistake, it's time in X axis and voltage in Y right?
22:03:27nosa-jim just makeing sure thanks for the help
22:04:22jlbiasinipamaury: regarding the bootloader you should probably upload the tagged file(s?) somewhere And I will try to find someone to upload it in the meantime...
22:04:44nosa-jjlbiasini: so 6 buttons eah i might like the tatical buttons on the original i supose but a micro usb card on the fuze+ o_o
22:04:46lorenzo92kugel: ah scaling issue :D
22:05:03nosa-jusb port * woww
22:05:10nosa-jno clue why i even typed card
22:05:14gevaertsnosa-j: "old fuze or new fuze" are rather confusing terms if you mean "fuze or fuze plus"
22:05:29nosa-jha ture
22:05:53nosa-jtheres a v2 of the original fuze
22:05:59nosa-jforgot about that one
22:06:08jlbiasininosa-j: fuze+ had 9 virtual keys (3x3) + 3 solid key (volumes and power)
22:06:36jlbiasininosa-j: oups I was talking about fuze+ forget everything I wrote
22:06:41curtismhey so i have a few songs (pirated, i admit) that rockbox will not read the tags on, and if i try to play them rockbox freezes
22:06:51curtismsame thing with the stock firmware
22:06:59nosa-jjlbiasini: i did mean fuze+ i just said the wrong thing so your fine
22:07:17gevaertscurtism: which codec?
22:07:21curtismi have a sansa fuze v2... i can't figure out what's wonky about these files. banshee will play them fine
22:07:27nosa-jmy bad for that i really didn't explain good
22:07:44curtismAudio file with ID3 version 2.3.0, unsynchronized frames, contains: MPEG ADTS, layer III, v1, 320 kbps, 44.1 kHz, JntStereo
22:08:12gevaertsAnything special in them like e.g. APE tags?
22:08:36nosa-ji kinda want to get the radio though hmm
22:08:50curtismgevaerts: how would i find that out?
22:08:58curtismgevaerts: that output above was from `file` btw
22:09:03nosa-jjlbiasini: im sure it will be suported in time but i can't decied
22:09:35jlbiasininosa-j: radio is nearly all implemented but for some reason still doesn't work... It might take a few more month
22:09:46nosa-jahh hmm
22:09:50jlbiasinirecording should be faster
22:09:50gevaertscurtism: good question... I'm not really a codec specialist (well, I'm not a codec specialist at all...), just asking the common first questions here. maybe saratoga can help more
22:09:51pamauryhopefully less than that !
22:09:58jlbiasinipamaury: thx
22:10:11nosa-jthe touch buttons worrie me though
22:10:15pamauryrecording should be easy to implement, the audio parameters could get some more tweaking
22:10:31pamaurythere is room for improvement for the battery life too
22:10:42curtismgevaerts: it's weird because the files work fine in all the players i have on my desktop, but they freeze the stock and rockbox firmwares :\
22:10:58jlbiasininosa-j: why?
22:11:09gevaertscurtism: maybe filesystem corruption?
22:11:15nosa-jseams like it would be hard to get them to respond
22:11:25nosa-ji have never used them so im worried
22:11:43curtismgevaerts: i've removed the files and put them back several times (not always in the same place)
22:11:51dreamlayerscurtism: maybe "unsynchronized frames" in ID3? What about if you write a tag without that or just strip the tag?
22:11:59nosa-jjlbiasini: i did have a ipod nano chronomatic which volume button was touch sensitive
22:12:55jlbiasininosa-j: fuze+ has solid volume keys
22:13:18nosa-jhm right but i was just compairing the feature
22:13:49nosa-ji could always try it for a week or so and send it back if i don't like it
22:14:07gevaertsnosa-j: in the end *you* will have to decide. Informing us of every doubt you have isn't going to help with that
22:14:19curtismdreamlayers: i rewrote the tags using banshee and that's still there, not sure if that's what you meant
22:15:18nosa-jgevaerts: yes i agree i just really hate inpulse buying. sorry for any trubble :D
22:16:09nosa-ji'll try to keep my questions to rockbox spicific questions
22:16:18curtismdreamlayers: but yes, the "synchronized frames" part seems to be the only difference from other audio
22:16:30dreamlayerscurtism: can you just remove the id3v2 tag?
22:16:40curtismdreamlayers: how should i do that?
22:16:55curtismdreamlayers: (running arch linux)
22:19:03dreamlayerscurtism: EasyTAG is a nice graphical tagger program that can strip tags.
22:19:26curtismdreamlayers: thanks
22:22:00jlbiasiniso apart from badger who seems to be away who can also put file on the server?
22:22:18 Join [Saint] [0] (~Saint]@unaffiliated/saint/x-8516940)
22:23:49gevaerts[Saint]: do you still have a working nano2g?
22:24:37[Saint]I do, yep. Howcome?
22:24:47curtismdreamlayers: looks like just re-saving the tags from easytags gets rid of the unsynchronized frames stuff, thanks
22:25:04gevaerts[Saint]: any chance of testing FS #12497?
22:25:04fs-bluebot 3Reorganize USB initialization to make it work with FreeBSD (patches, unconfirmed)
22:25:20dreamlayerscurtism: do the files play properly in Rockbox after that?
22:25:29curtismdreamlayers: i'll check in a second
22:26:38[Saint]gevaerts: what're the test conditions? I mean, with which os should I be testing? U should be able to test this this evening, I can't presently, sorry.
22:27:26gevaerts[Saint]: test both normal connection and chare-only connection, on any OS you have handy
22:30:28curtismdreamlayers: they play now :D thanks a lot
22:31:22funmanjlbiasini: perhaps his brother
22:31:23dreamlayerscurtism: you're welcome!
22:38:15jlbiasinifunman: who is he?
22:39:15 Quit dreamlayers (Quit: Leaving)
22:39:32jlbiasiniI yeah right. But they must be in holydays then
22:40:34funmanby the way, what happens if the swedes are hit by a bus ?
22:45:48curtismcan i get rockbox to treat whatever image is in the same folder as the song as the album art?
22:46:16 Quit B4gder (Quit: It is time to say moo)
22:46:16[Saint]funman: I'd thought about that too...
22:48:07[Saint]curtism: yes, sure. the manual covers art naming/placement
22:48:52curtism[Saint]: what heading is it under, i can't find it
22:49:35[Saint]Isn't there a search function in the manual now?
22:50:46dfktcurtism, FS #10201 has info about priorities in naming
22:50:46fs-bluebot 3Fix album art file search priority (patches, closed)
22:51:09 Quit yosafbridge (Read error: Operation timed out)
22:51:41curtismdfkt: thanks
22:51:53[Saint]It should be in the manual...
22:52:15dfkti knew there was a wiki site about that, but i couldn't find it... doesn't seem to be in the manual as well
22:52:55curtismhm... these anamanaguchi albums have animated covers... i doubt rockbox supports that
22:53:48funmancurtism: gif?
22:54:26funmanrockbox has neither gif neither animated images support (except mpeg2video)
22:55:03dfktno gif support? didn't know that - due to the patent issue?
22:55:40funmanthe patent has expired i think so even if it did apply to us we wouldn't care
22:55:45gevaertsNo (well at least not these days)
22:55:48***Saving seen data "./dancer.seen"
22:55:50funmanmore like nobody use gif anymore except for animated pics
22:56:36curtismanimated covers aren't the most common thing anyway...
22:56:43 Join yosafbridge [0] (
22:58:41CIA-88New commit by 03funman (r31468): typo
22:59:57 Quit plush (Remote host closed the connection)
23:02:05 Join plush [0] (
23:14:48 Quit kevku (Ping timeout: 252 seconds)
23:14:56lorenzo92kugel: in some minutes also the charging bench with bl on + music will be ready ;)
23:20:46saratogascorche: ban arcteryxfz
23:23:01saratogagevaerts, JdGordon: either of you too
23:23:30 Quit [Saint] (Read error: Connection reset by peer)
23:23:33 Join [Saint_] [0] (~Saint]@unaffiliated/saint/x-8516940)
23:23:52 Nick kugel is now known as kugelp (~kugel@rockbox/developer/kugel)
23:23:59saratogaadattyhalge as well
23:24:28 Quit [Saint_] (Read error: Connection reset by peer)
23:24:31 Join [Saint] [0] (~Saint]@unaffiliated/saint/x-8516940)
23:25:38 Join CaptainKewl [0] (
23:28:49 Quit Zambezi (Ping timeout: 268 seconds)
23:29:48 Join Zambezi [0] (
23:31:49lorenzo92kugel: here it is ;)
23:32:02lorenzo92kugelp: ops you changed nick :)
23:37:35 Join bluefoxx [0] (
23:37:48lorenzo92kugelp: in the next days I could do some other I go ;)
23:38:42 Quit lorenzo92 (Remote host closed the connection)
23:40:36 Join davo [0] (

Previous day | Next day