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 2010-05-01

00:00:56 Join TheSeven [0] (~TheSeven@rockbox/developer/TheSeven)
00:04:36 Quit CGL (Quit: Soy spammero
00:10:37 Quit adnyxo (Ping timeout: 260 seconds)
00:11:49 Quit pamaury (Quit: Quitte)
00:12:24***Saving seen data "./dancer.seen"
00:17:58 Join Llorean [0] (
00:18:07 Quit Llorean (Changing host)
00:18:07 Join Llorean [0] (~DarkkOne@rockbox/user/Llorean)
00:21:37 Join whitewoe [0] (
00:23:24whitewoeIs there any work being done on the Sansa c150? I just got one for $10 and it sucks.
00:28:06pixelmasomeone started a port long ago but it seems to have been abandoned and it didn't get very far I think
00:29:32whitewoeoh well, thanks
00:29:45n17ikhbut yeah, looks abandoned
00:29:55pixelmathere should be a thread in our forums somewhere... and that
00:30:13whitewoeyeah I saw that, I was just wondering if there was any active development now or in the near future
00:30:27n17ikha prime target for you to start your own pet port of rockbox on!
00:30:35pixelmawhitewoe: maybe you can find other interested owners and it can be picked up again?
00:31:09whitewoeI'm not a coder, but I will keep an eye out for any activity
00:37:07 Join Strife89|PalmTX [0] (
00:37:35 Quit domonoky (Read error: Connection reset by peer)
00:38:44 Quit whitewoe (Quit: Leaving)
00:40:17 Nick fxb__ is now known as fxb (
00:41:36 Quit liar (Quit: bye, got to play american football now :-))
00:45:04 Join Takato [0] (
00:45:04 Quit M3DLG (Ping timeout: 240 seconds)
00:45:53 Quit merbanan (Ping timeout: 252 seconds)
00:47:41 Quit Strife89 (Read error: No route to host)
00:47:43 Join Strife1989 [0] (
00:48:25 Nick Strife1989 is now known as Strife89 (
00:58:29 Quit dfkt (Quit: -= SysReset 2.53=- Ph'nglui mglw'nafh Cthulhu R'lyeh wgah'nagl fhtagn.)
01:04:05 Join aim__ [0] (
01:04:46 Join mc2739 [0] (~mc2739@rockbox/developer/mc2739)
01:05:53 Quit Takato (Quit: ChatZilla 0.9.86 [Firefox 3.6.3/20100401080539])
01:16:34 Join Strife89|Laptop [0] (
01:17:17 Part Strife89|Laptop
01:22:22 Quit Strife89 (Quit: Leaving)
01:26:05 Join RandomInsano [0] (
01:26:46 Join jgarvey [0] (
01:27:03 Quit jgarvey (Client Quit)
01:29:16 Join Strife89 [0] (
01:43:24 Quit Schmogel (Ping timeout: 265 seconds)
01:45:59 Quit archivator (Quit: Leaving)
01:46:34 Quit bertrik (Quit: De groeten)
01:49:07 Join mikroflops_ [0] (
01:52:20 Nick bimbel is now known as bmbl (~Miranda@unaffiliated/bmbl)
01:53:05 Quit mikroflops (Ping timeout: 240 seconds)
01:54:07 Quit saratoga (Quit: Page closed)
01:57:53 Quit aim__ (Quit: Ex-Chat)
02:07:38 Join Rob2222 [0] (
02:09:54 Quit RandomInsano (Quit: CGI:IRC)
02:11:26 Quit Rob2223 (Ping timeout: 265 seconds)
02:12:26***Saving seen data "./dancer.seen"
02:13:18 Join CaptainKewl [0] (
02:25:43 Quit efyx (Remote host closed the connection)
02:29:41 Quit bmbl (Quit: Bye!)
02:58:10 Quit Llorean (Quit: Leaving.)
02:59:38 Nick fxb is now known as fxb__ (
03:06:02 Quit Strife89 (Quit: Reboot.)
03:10:46 Join Strife89 [0] (
03:12:18 Join robin0800 [0] (
03:15:45 Join BHSPitMonkey [0] (~stephen@unaffiliated/bhspitmonkey)
03:29:07 Quit Strife89 (Quit: Reboot.)
03:32:48 Join Strife89 [0] (
04:06:37 Quit panni_ (Quit: ( :: NoNameScript 3.81 :: ))
04:06:58 Join panni_ [0] (
04:08:51 Quit Barahir (Ping timeout: 246 seconds)
04:10:37 Join Barahir [0] (
04:12:29***Saving seen data "./dancer.seen"
04:18:05 Join CGL [0] (~CGL@
04:18:56 Join adnyxo [0] (
04:24:10 Quit amiconn (Disconnected by services)
04:24:12 Join amiconn_ [0] (quassel@rockbox/developer/amiconn)
04:24:13 Quit pixelma (Disconnected by services)
04:24:13 Join pixelma_ [0] (quassel@rockbox/staff/pixelma)
04:24:33 Nick pixelma_ is now known as pixelma (quassel@rockbox/staff/pixelma)
04:24:34 Nick amiconn_ is now known as amiconn (quassel@rockbox/developer/amiconn)
04:27:17 Join macgeek417 [0] (~macgeek41@
04:28:06 Quit macgeek417 (Client Quit)
04:29:16 Quit robin0800 (Remote host closed the connection)
04:30:11 Join anewuser [0] (anewuser@unaffiliated/anewuser)
04:30:43 Join Llorean [0] (
04:30:48 Quit Llorean (Changing host)
04:30:48 Join Llorean [0] (~DarkkOne@rockbox/user/Llorean)
04:32:05 Join macgeek417 [0] (~macgeek41@
04:36:48 Quit macgeek417 (Client Quit)
04:45:18 Quit adnyxo (Ping timeout: 276 seconds)
04:48:21 Quit TheSeven (Ping timeout: 264 seconds)
04:51:41 Join TheSeven [0] (~TheSeven@rockbox/developer/TheSeven)
05:07:34 Quit anewuser (Quit: What do you know...THE WORLD'S first NTRQ (that's for NES/FAMICOM) tracking compo. Have powerpak? Try it out! Otherwise ROM IMAGE.)
05:15:02 Join kramer3d [0] (~kramer@unaffiliated/kramer3d)
05:35:19 Quit GeekShadow (Quit: The cake is a lie !)
05:37:20 Join Transformer [0] (
05:40:47 Part Transformer
05:45:00 Join fjurc [0] (
06:12:30***Saving seen data "./dancer.seen"
06:13:31 Quit panni_ (Quit: ( :: NoNameScript 3.81 :: ))
06:14:39 Join saratoga [0] (~9803c20d@gateway/web/freenode/x-vpjcboeddhxmghlr)
06:19:58 Quit Strife89|PalmTX (Quit: Bed. :))
06:19:58 Quit Strife89 (Quit: Bed! :))
06:23:36 Quit kramer3d (Read error: Connection reset by peer)
06:27:30 Join kramer3d [0] (~kramer@unaffiliated/kramer3d)
06:45:57 Quit kramer3d (Read error: Operation timed out)
06:47:07 Join kramer3d [0] (~kramer@unaffiliated/kramer3d)
07:01:28S_a_i_n_t and <−− Not sure it's one I'd use myself, but *someone* probably will. Theme for Nano1/2g, comments?
07:04:47 Part Sjano
07:13:09 Join robzy [0] (
07:15:29robzyhey guys, where can i find info on usb support in rockbox?
07:15:46Llorean Wiki?
07:16:17robzy*sigh* so obvious, silly me, thank you
07:16:53 Quit kramer3d (Read error: Operation timed out)
07:17:01robzy"How Rockbox handles USB slave mode" <−− am I correct in thinking that Rockbox can't operate as a USB Host?
07:19:41robzythank you :) i was googling as well but could only find references from '07, was hoping it'd changed since then
07:37:05 Quit Horscht (Ping timeout: 260 seconds)
07:39:57 Join Horscht [0] (~Horscht2@xbmc/user/horscht)
07:41:13 Join mischasworld [0] (
07:42:33 Quit robzy ()
07:49:19S_a_i_n_tWhat the?!? Turning the hold switch on, then off again when the player is off boots the palyer up again (in Rockbox).
07:49:41S_a_i_n_t*Oh, Nano2g...Since when has it done that?
07:49:53S_a_i_n_tNano1g doesn't do this.
07:53:44S_a_i_n_tHuh, if the player shuts down when hold is one (idle timeout), then switching hold to off again boot the player also.
07:53:54S_a_i_n_tIs this a bug, or a fauture?
07:54:11S_a_i_n_ts/is one/is on/
07:55:34LloreanAre you sure it didn't do that before?
07:55:56S_a_i_n_tI don't know...I've never noticed it before.
07:56:38S_a_i_n_tI'm *fairly* certain I would have noticed it, but, possibly not. Kinda weird that Nano1g doesn't do it, but 2g does.
08:01:04 Quit rvvs89 (Changing host)
08:01:04 Join rvvs89 [0] (ivo@pdpc/supporter/base/rvvs89)
08:01:21LloreanS_a_i_n_t: Well what happens when they're turned off can relate to hardware.
08:01:37LloreanSo it'd be good to check if older builds do it.
08:04:34 Quit saratoga (Quit: Page closed)
08:04:55S_a_i_n_t"meh...", guess I'll just wait untill someone that knows the Nano2g code can just say it definitely should/shouldn't do this. The only reason it caught my attention was because I knew that the !g Nano doesn't do it surprised me I'd never noticed the 2g did/does.
08:09:26 Quit fjurc (Quit: MegaIRC v4.05
08:09:28 Join bmbl [0] (~Miranda@unaffiliated/bmbl)
08:12:33***Saving seen data "./dancer.seen"
08:21:13 Join mrblue [0] (
08:23:08 Quit CaptainKewl (Read error: Operation timed out)
08:23:30 Part mrblue
08:26:43 Quit Llorean (Quit: Leaving.)
08:51:41linuxstbS_a_i_n_t: That sounds like a hardware feature to me. I've never noticed it (but have never tried it).
08:54:07 Join flydutch [0] (
08:54:30 Quit CGL (Quit: Soy spammero
09:02:03 Join kramer3d [0] (~kramer@unaffiliated/kramer3d)
09:05:26linuxstbS_a_i_n_t: I can confirm my nano2g does the same - and it's running a _very_ old build.
09:18:27 Join funman [0] (~fun@rockbox/developer/funman)
09:25:36 Quit kramer3d (Quit: Leaving)
09:31:06 Join domonoky [0] (~Domonoky@rockbox/developer/domonoky)
09:33:31 Join MethoS- [0] (~clemens@
09:33:34 Join DerPapst [0] (
09:35:30 Join solexx [0] (
09:38:19 Join esperegu [0] (~quassel@
09:38:25*domonoky reads those bounty mails, and thinks he has no problems if someone else can earn money on my free work.. All others are just jealous :-)
09:39:19 Quit solexx_ (Ping timeout: 264 seconds)
09:41:48AlexPdomonoky: I don't really like them and don't think the project should organise them, but if people want to make a private arrangement then fine by me :)
09:42:30domonokyi agree, that the project shouldnt organise them. But if someone else wants todo that, i see no problem with that.
09:42:32funmanFlynDice: any progress with PIO ? apparently the 1st (read) transfer stops when 43 words are left from a full sector
09:43:15AlexPdomonoky: Sure, I'm not going to try and stop them - but I don't have to like them :)
09:44:18funmanthe guy should better spend this money on a programming book if he wants to see a port
09:44:38domonokyor pay a student ? :-)
09:45:23AlexPthat too :)
09:46:31 Quit mischasworld (Ping timeout: 264 seconds)
09:47:39 Join mischasworld [0] (
09:49:03amiconnsoap: Ideally line-out should never change volume, not even when muting, but that is how the TLV320 behaves...
09:50:09funmanFlynDice: still 43 words left to transfer after 6 MCI_INT_RXDR interrupts, not sure what's magic about this value
09:50:58 Quit krazykit (Ping timeout: 268 seconds)
09:51:33 Join krazykit [0] (
09:52:38funmanhowever I see a DTO interrupt hmm
09:52:42 Join Llorean [0] (
09:53:01 Quit Llorean (Changing host)
09:53:01 Join Llorean [0] (~DarkkOne@rockbox/user/Llorean)
09:53:37funman-> 10 more words, 33 left
09:58:48funmannow down to 3 words, perhaps byte transfer is needed after all
09:58:52 Quit MethoS- (Remote host closed the connection)
09:58:54linuxstbdomonoky: I don't think it's just the issue of people earning money to work on Rockbox (or Linux) - after all, we are supporting students being paid by Google.
10:01:37funmanthere's also the shop selling rockboxed sansas (although we don't directly support them)
10:01:59domonokylinuxstb: jup, i think the same. This claim that if someone gets paid for rockbox work, is unfair for all others who contributed is just invalid and looks pretty jealous.
10:04:36domonokyeveryone is free to make money with rockbox if they can, as long as they comply with the licence.
10:04:39Lloreanfunman: Though that shop doesn't appear to necessarily be holding up their GPL obligations
10:05:04domonokywe should tell that those shop owners once more :-)
10:05:37Lloreandomonoky: I tried asking them about whether they had backups of the source for the versions they provided, but their contact email was invalid.
10:06:22linuxstbdomonoky: It does depend on the specifics though - I do remember some bounty suggestions in the past generating that feeling in me (but I can't remember the specifics...). But I think the issue is more that this is generally collaborative work, and bounties are not really in that spirit (they are more competitive).
10:06:25domonokyLlorean: i wrote them once in the past about those licence issues, and the promised to change the website. But unfortunatly nothing happend.
10:06:51linuxstbdomonoky: Are they still distributing their own rbutil binary?
10:06:58Lloreanlinuxstb: This is something very real about bounties. They encourage you *not* to share your work until the bounty is collected, unless you have to.
10:07:24domonokyLlorean: why, just work in public and then every knows who has done how much work..
10:07:28LloreanThey also may encourage people to wait until another bounty is set before pitching in again, because it makes it clear there's money to be made (this isn't necessarily harmful, as it can also draw in new people, but it may discourage 'free' work)
10:08:01Lloreandomonoky: No, my point is that if you want the whole bounty, you won't share your work until your done because then you'd have to share the bounty if someone took where you were at and finished it.
10:08:33LloreanThe fact is, some things will only get done if there are bounties, obviously, but it does encourage people not to post half-complete patches as long as they feel they can still complete it themselves.
10:08:48linuxstbLlorean: But there _is_ a "problem" of no-one really being interested in porting to the newer ipods, and it would be nice to solve that somehow... So I would like to see the discussion be more positive - the original poster simply wants that to happen...
10:09:02Lloreanlinuxstb: I think my four points were pretty positive.
10:09:05domonokyLlorean: true, but thats not neccessary, just put the requirement to work public into the bounty goals.
10:09:36Lloreandomonoky: I don't know that there's any practical way to enforce it.
10:09:51funmanhmm now i'm down to -18 bytes left to be transferred
10:10:18domonokyLlorean: easy, if someone posts a working patch without previous work, just dont give him the bounty :-)
10:10:46Lloreandomonoky: I doubt they'd license the code so that we could use it unless we paid.
10:10:53domonokyarg, *previous visible work* ie patches in the tracker, a git repo, etc
10:11:06LloreanSo we'd be in a position of actively turning down work we could have.
10:11:19LloreanAlso, how many previously visible patches, over how much time?
10:11:25 Join mischasworld_ [0] (
10:11:28AlexPdomonoky: And I never claimed that someone getting paid is unfair for all others who contributed, in all cases - just very specific ones like the one I gave - e.g. someone just enables a driver for an already written chip and then claims a large bounty for getting USB to work for instance
10:11:32LloreanI just don't think it's very tenable. You'd have to come up with pretty complex rules
10:11:33domonokyLlorean: thats not possible, they have to comply with GPL with the licence for patches.
10:12:04linuxstbSo we get it for free ;)
10:12:15domonokyAlexP: i think even that is fine.
10:12:32Lloreandomonoky: Yes, in some cases. But if they have original files (such as new drivers) it's not linked against anything, and their copy can be under *any* license.
10:12:37***Saving seen data "./dancer.seen"
10:12:38domonokywe only turn down the payment, not their work :-)
10:12:55domonokyits like gsoc, even if we fail students we can use their work.
10:12:57LloreanIt wouldn't be hard to provide a non-GPL sample of the code, or strip all GPL from the text file, to demonstrate that the work is done without providing something that must be GPL
10:12:59AlexPAnd it is hardly jealous, I haven't done any of the other work :)
10:14:57domonokyLlorean: i thinks thats pretty hard. if the patch is something which should be used in rockbox, is adapted to rockbox and then you can hardly claim its independent.
10:15:16domonokyofcourse things like a bootloader could be done independend and under a other licence.
10:15:26Lloreandomonoky: But the GPL requirement is *linked* code isn't it?
10:15:30LloreanUntil you compile it, that's different.
10:15:41LloreanOr things that have an explicit separate source file.
10:15:41domonokyLlorean: you will have to link it to test it for sure.
10:15:57LloreanBut they don't have to distribute that binary
10:17:28domonokyhm, but you can ofcourse require that their patch is under GPL if they want a chance for the bounty. People who try to go around, just dont get any money.
10:17:52 Join Kitr88 [0] (
10:17:52LloreanI'm just saying that still puts us in the condition of saying "offer us the chance to take your code without paying, and we might pay you"
10:17:59domonokyin the worst case, they dont get money, and we cant use their code.. nothing lost.
10:18:27 Quit mischasworld (Ping timeout: 276 seconds)
10:18:34*funman put a bounty on writing code for sansa AMS v2: price = 1 beer, conditions = none
10:18:38LloreanAnd it still doesn't really solve the openness thing.
10:18:55AlexPfunman: You pay yourself? :)
10:19:18LloreanIf *I* wanted the whole bounty, I'd do the work in private, but save a patch halfway through. I'd post it, then a few days later the 'finished' patch. If a lot of interest was shown, and I saw evidence people were working on it, I'd publish my finished one sooner.
10:19:20*domonoky offers a bavarian beer for this too
10:19:20funmanwell *i* put the bounty so yes
10:19:36funmanand there's an extra bonus of 1 other beer if you steal the code from someone else
10:19:37LloreanI'm not saying bounties won't get us code we wouldn't get otherwise.
10:19:49LloreanI'm just saying I feel they discourage collaboration, and I don't think it's easy to solve this
10:20:16*domonoky thinks we wont know this for sure, untill we have tried it :-)
10:21:14*domonoky offers a full case of bavarian beers for the one who snatches away the AMS v2 USB project just before the otherone finsihed it :-)
10:21:31LloreanOne *could* easily argue that only the person who completes the project should get paid
10:21:42LloreanSince everyone else released their code under the GPL anyway, so that one person had permission to use it.
10:21:47linuxstbTrying to be more positive, I wonder if we could have some kind of "future targets" page, to try and publicise potential new ports. That page would have the current state of hacking that device - e.g. the fact that we can exploit DFU mode to run our own code on the Classics and 3rd/4th gen Nanos. We could offer hardware to people willing to work on those ports, with a condition that they return it if they don't, but mainly it would just
10:21:47linuxstb be a way to piublicise promising-looking new targets for people that fancy the challenge of a new port.
10:21:48 Quit Kitar|st (Ping timeout: 268 seconds)
10:22:04Lloreanlinuxstb: That is an excellent idea
10:22:11 Join Kitar|st [0] (
10:22:14 Quit Kitr88 (Ping timeout: 246 seconds)
10:22:15LloreanFor example the Zune classic currently seems quite promising, doesn't it?
10:22:34funmanLlorean: is supervisor mode already available?
10:22:41 Quit esperegu (Read error: Connection reset by peer)
10:23:19 Join esperegu [0] (~quassel@
10:23:21linuxstbLlorean: Yes, assuming the exploit can be turned into a bootloader, which I don't think anyone knows for sure.
10:24:06Lloreanlinuxstb: I think the assumption is that it should be able to, though, isn't it?
10:24:45Lloreanfunman: They said kernel mode should be available on the classic since it's CE5 based, but not necessarily on the HD, but that was a while ago and just "what we think" right after the crack.
10:24:51linuxstbYes, that's the impression I got. I'm just noting that "should" isn't the same as "can".
10:25:14LloreanYeah. Quite promising, but not a sure thing
10:25:38*Llorean thinks Bagder made a good point about code quality and bounties, too.
10:25:59funmanLlorean: where did Bagder make that?
10:26:08LloreanIn the mailing list
10:26:16 Join mischasworld [0] (
10:26:22Llorean"When you go "bounty" there's suddenly a race, and there's suddendly a situation where the winner takes it all. It makes it an individual act to get the money rather that to help the project. You may take shortcuts, ignore problems and more, just to be able to claim the money. If you don't, perhaps your competitors do? No, a bounty is a tricky thing to work out right IMHO."
10:27:36 Quit mischasworld (Read error: Connection reset by peer)
10:28:08*Llorean thinks private bounties are okay, but that having the project take responsibility for any of them is just a recipe to get us involved in new and interesting arguments.
10:28:28LloreanUnless it's a _really_ simple bounty, like "$25 for a new default theme"
10:29:04LloreanA bounty for something that's obviously intended to be one-man work, but we don't care who the man is so much, and the first one to get it done isn't necessarily the bounty taker.
10:31:04JdGordona new theme is a one man effort?
10:32:34 Quit JdGordon (Remote host closed the connection)
10:34:32 Join JdGordon [0] (~jonno@rockbox/developer/JdGordon)
10:34:57LloreanA new theme certainly could be. Or a team, where the people agreed to work together before hand, and can agree on their own split. I mean if it's actually *new* you're mostly commissioning art.
10:45:43Bagderit's also a matter of amount, if anyone would offer 3K USD for it, then everyone involved in making skins possible in the first place might feel a bit overrun
10:48:53 Join Highlander [0] (
10:52:02 Join pamaury [0] (~pamaury@rockbox/developer/pamaury)
10:55:38S_a_i_n_tAll my themes have been "one man jobs", but I also have to give credit to the 100's of other individuals that made the codebase which makes it possible for me to theme my device.
10:56:05S_a_i_n_tFor that reason, I feel uneasy accepting any ammount of money for the multiple Rockbox installs I've done.
10:56:21LloreanI think being paid for specific work is very, very reasonable.
10:56:22S_a_i_n_tAnd "private" themes I;ve created.
10:56:36LloreanLike "for $5 I'll get Rockbox installed for you" kinda stuff makes perfect sense
10:56:39LloreanYou spend time, you get paid.
10:56:45S_a_i_n_tBut, no matter's cutting someone out of the loop.
10:56:46LloreanAnd for GSOC it's basically us hiring people for a specific job.
10:57:06LloreanBut instead of picking a person for a job, when you say "$300 exists, and will go to whoever gets X working" you create a competition, which is different.
10:57:11LloreanAs Bagder said so well
10:57:14 Join stripwax [0] (
10:57:31LloreanS_a_i_n_t: But these are people who've not only contributed for free, but agreed to let other people make money from their work in advance.
10:57:33S_a_i_n_tA % of any money I earn themeing should go to JdGordon, kugel (others) etc. so I personally take none.
10:57:43S_a_i_n_tBut, that's just me.
11:00:53S_a_i_n_tI guess, the objection I have to making money from the project is pruely moral, and entirely my own. If anyone else wants to make money off of it, or offer a bounty for a specific piece of code, then so be it. I have no objections to it, I just couldn;t feel comfortable accepting money/charging for work myself.
11:01:08 Quit stripwax (Client Quit)
11:02:07 Join stripwax [0] (
11:03:40 Quit stripwax (Read error: Connection reset by peer)
11:04:22JdGordonI tihnk this guy doesnt have his priorities right.... the same bounty % for accelerometer and power managemnet?
11:04:36JdGordonasssuming the bounty is for whats more important
11:05:10JdGordonand expiration dates.. wtf?
11:05:40LloreanHis ideal seems to actually base the bounties on "difficulty" rather than "value"
11:06:04LloreanSo I think he'd welcome feedback on that.
11:06:57 Join stripwax [0] (
11:07:22LloreanThe expiration date makes a kind of sense from his standpoint - I can see how he might think that if there's not one, people might just wait around until they feel like it because the money will always be out there waiting to be claimed.
11:07:33LloreanMeanwhile the people who've put it in escrow are out that money whether they get something for it or not
11:07:57LloreanAt the same time, someone might want to point out that expiration dates could very easily make someone look at it, say "damn, there's no realistic way that'll be done in four months" and skip the idea entirely
11:08:42S_a_i_n_tIt seems weird puting a deadline on t though, because if its not met, won't he just have to come back and proposition the same people again?
11:09:47S_a_i_n_tThe fact that a bounty need be offered suggests options are slim in the first place.
11:09:50LloreanI'm not saying it realistically makes sense
11:09:58LloreanI'm just saying, I can see how it might make sense to someone at first glance
11:10:08S_a_i_n_tOh, totally.
11:11:13LloreanI mean, in regard to a bounty, the best plan might be to let people apply for it like a job app. Bob comes up and says "I think I can do a Nano 4G port". Then nobody else can apply for the bounty as long as he's providing progress updates (like GSOC). He can involve others however he wants, and it's up to him to agree with others how the money will be split
11:11:22LloreanThen, if his port meets our requirements, he gets the money and splits it how they've agreed.
11:11:51LloreanYou're basically simply hiring programmers.
11:12:13LloreanBut you're posting them as bounties that people apply for, rather than looking for programmers in general.
11:12:48 Quit phanboy4 (Read error: Connection reset by peer)
11:13:12 Quit mischasworld_ (Remote host closed the connection)
11:13:19 Quit BHSPitMonkey (Remote host closed the connection)
11:16:43 Join mischasworld [0] (
11:17:35 Join Luca_S [0] (
11:22:41 Join dfkt [0] (dfkt@unaffiliated/dfkt)
11:26:56 Quit funman (Quit: free(random());)
11:28:38*domonoky likes this future target wikipage idea. maybe we could use donations to get players for this future targets, which we then could lend to potential developers.
11:29:24LloreanI see no problem with that at all
11:33:08topikwould you not get a lot of people asking 'is it done yet' ?
11:33:19LloreanYou do that anyway
11:33:19topikor perhaps not more than they already do ?
11:33:32LloreanI don't think so. Not too much, at least
11:34:20LloreanIt might also make sense to spend some of the funds on acquiring "official Rockbox players" to be held by some people so that we have some of each player that will always be around (thought not necessarily *immediately* available) for the big stuff
11:34:36LloreanI mean it occasionally comes up where we just don't even know who might own a specific player any more.
11:35:05topikis there a page which lists the current availability?
11:35:11LloreanSo when bootloader changes for the iPod come up, you have a list of people who have volunteered in advance to hold on to specific iPods for us for such testing purposes.
11:35:22Lloreantopik: There's a testers page where some people have listed their personal players
11:36:33*Llorean expects his line on that page is somewhat out of date
11:36:41topikthat's a serious list
11:37:13 Quit Highlander (Quit: Quitte)
11:38:13topikmr. gevaerts is quite the collector
11:40:46domonokytopik: take a look at TowerOfRockbox in the wiki, if you want to see many many mp3players :-)
11:40:59*Llorean updates the list
11:41:24LloreanThere were four or five MP3 players I hadn't added myself to yet.
11:43:44topiki've seen the tower domonoky. it's impressive
11:43:51 Join Highlander [0] (
11:45:35topiksomeone should put a bounty on getting rockbox to run on led pigs
11:47:53 Quit stripwax (Read error: Connection reset by peer)
11:52:19 Join Kupop [0] (
11:56:40pamaurywhy isn't there the clip+ on the testing page ?
11:56:53pamauryonly stable targets ?
11:57:12LloreanProbably because nobody who owns one has added themselves
11:57:19LloreanGenerally speaking, *many* people don't know that page even exists
11:57:30LloreanOr are like me, and forget to update it while acquiring several new targets
11:57:41*pamaury adds a clip+ entry
11:58:51LloreanI'd say there's no reason not to have a line there for anything that can be selected in the build system, at the very least.
12:02:22 Join Schmogel [0] (
12:05:19 Join M3DLG [0] (~M3DLG@
12:06:30 Join n1s [0] (~n1s@rockbox/developer/n1s)
12:10:20topikfuze v2 / clip v2 are also not listed
12:10:40 Join halmi [0] (
12:12:39***Saving seen data "./dancer.seen"
12:31:34 Quit M3DLG (Ping timeout: 240 seconds)
12:36:20 Join Xerion [0] (
12:37:31 Join petur [0] (~petur@rockbox/developer/petur)
12:38:01 Quit n1s (Ping timeout: 246 seconds)
12:49:09 Quit Luca_S (Quit: CGI:IRC)
12:51:02 Join n1s [0] (~n1s@rockbox/developer/n1s)
12:56:45 Join Boldfilter [0] (
13:05:10 Quit n1s (Ping timeout: 248 seconds)
13:10:35 Join halmi_ [0] (
13:13:02 Quit halmi (Ping timeout: 260 seconds)
13:13:22 Join JohannesSM64 [0] (
13:18:13 Join n1s [0] (~n1s@rockbox/developer/n1s)
13:22:06 Join GeekShadow [0] (~Antoine@reactos/tester/GeekShadow)
13:25:13 Join halmi [0] (
13:28:38 Quit halmi_ (Ping timeout: 268 seconds)
13:36:41 Quit esperegu (Remote host closed the connection)
13:39:27 Join kugel [0] (~kugel@rockbox/developer/kugel)
13:39:57 Quit yosafbridge (Quit: Coyote finally caught me)
13:40:05 Join yosafbridge [0] (
13:40:09 Join halmi_ [0] (
13:43:10 Quit halmi (Ping timeout: 276 seconds)
13:50:19 Quit bmbl (Quit: Bye!)
13:55:18 Join halmi [0] (
13:57:50 Quit halmi_ (Ping timeout: 260 seconds)
13:59:17 Quit amiconn (Disconnected by services)
13:59:19 Join amiconn_ [0] (quassel@rockbox/developer/amiconn)
13:59:41 Nick amiconn_ is now known as amiconn (quassel@rockbox/developer/amiconn)
14:00:21 Quit pixelma (Disconnected by services)
14:00:21 Join pixelma_ [0] (quassel@rockbox/staff/pixelma)
14:00:41 Nick pixelma_ is now known as pixelma (quassel@rockbox/staff/pixelma)
14:02:13 Join mikroflops [0] (
14:03:03 Join efyx [0] (
14:06:11 Quit mikroflops_ (Ping timeout: 276 seconds)
14:06:38 Join pixelma_ [0] (quassel@rockbox/staff/pixelma)
14:06:38 Quit pixelma (Disconnected by services)
14:06:50 Quit amiconn (Disconnected by services)
14:06:52 Join amiconn_ [0] (quassel@rockbox/developer/amiconn)
14:06:58 Nick pixelma_ is now known as pixelma (quassel@rockbox/staff/pixelma)
14:07:14 Nick amiconn_ is now known as amiconn (quassel@rockbox/developer/amiconn)
14:09:25 Quit mt (Read error: Connection reset by peer)
14:09:42 Quit pamaury (Ping timeout: 246 seconds)
14:09:57 Join pamaury [0] (~pamaury@rockbox/developer/pamaury)
14:10:14 Join halmi_ [0] (
14:12:41***Saving seen data "./dancer.seen"
14:12:44linuxstbrasher: To answer your question in -community about documentation of the ipod boot procedure, try IpodBoot or IpodPatcher. But which ipods are you talking about?
14:12:48 Quit halmi (Ping timeout: 252 seconds)
14:13:23rasherlinuxstb: not any particular one
14:13:47rasherJust general curiosity about how exactly they made them "unbrickable"
14:14:36linuxstbDisk mode in NOR (or in later ipods, DFU mode in ROM) is the main thing.
14:18:59 Join funman [0] (~fun@rockbox/developer/funman)
14:19:00linuxstb... plus the fact that we insert our bootloader into the main firmware which is just on disk. On other targets, we insert our bootloader into (NOR) flash, meaning a bad bootloader can brick the device.
14:19:33 Join Llorean1 [0] (
14:20:41 Nick fxb__ is now known as fxb (
14:21:05amiconnlinuxstb: The coldfire iaudios are also essentially unbrickable
14:21:39 Quit Llorean (Ping timeout: 240 seconds)
14:22:10linuxstbThinking about it, the key is where in the boot process we insert our code, and what recovery tools are available before the boot process reaches our (potentially bad) code.
14:22:46 Quit mischasworld (Ping timeout: 265 seconds)
14:23:15 Quit pamaury (Ping timeout: 265 seconds)
14:23:26 Join pamaury [0] (~pamaury@rockbox/developer/pamaury)
14:23:33funmanalso how code insertion can affect earlier boot process
14:23:46amiconnYep. On the iaudios our bootloader does reside in the flash rom, but we don't put it there ourselves, but use the cowon loader (which is able to reflash itself) to put it ther
14:23:56 Join halmi [0] (
14:24:10linuxstbfunman: How could that happen?
14:24:12amiconnSo if the bootloader is broken, do a hard poweroff, boot with usb inserted, put new bootloader on disk, and reflash
14:24:37funmanlinuxstb: for sansa AMS, a bug in mkamsboot for example
14:25:07linuxstbfunman: That's not a "safe" target IMO. We modify the main firmware image, and only the main firmware can update iself. (IIRC)
14:25:40linuxstbi.e. nothing before the main firmware in the boot process can update the main firmware. (in general, apart from e200v2?)
14:26:09 Quit shaggy-h (Ping timeout: 240 seconds)
14:26:38funmandepends on how you see it, the tiny asm code we insert is considered safe, and earlier in the boot process than the real rockbox bootloader
14:26:39 Quit halmi_ (Ping timeout: 240 seconds)
14:30:31linuxstbfunman: Yes, in effect we've made AMS targets "safe", as a bad bootloader shouldn't prevent the tiny asm code starting the OF. But I still don't understand what you mean by "affect earlier boot process" - later code shouldn't affect something that is run earlier.
14:30:33funmanwhat does 'data starv. timeout' mean ?
14:30:43funmanstarvation would mean fifo underrun for me
14:30:56 Quit Highlander (Quit: Quitte)
14:31:52funmanlinuxstb: see the log for r24772 for an example
14:32:33funmannot sure if 'starv' means something else than 'starvation' here, and there is a separate bit for 'fifo underrun'
14:32:52funmanoops it would be overrun, since the fifo is fed by the controller and we have to empty it by reading from it
14:33:24 Quit n1s (Ping timeout: 265 seconds)
14:33:57 Join qurvel [0] (
14:34:20funmanany help from english native speakers?
14:35:26 Part qurvel
14:38:50 Join halmi_ [0] (
14:41:10 Quit halmi (Ping timeout: 248 seconds)
14:41:56 Join geertvdijk [0] (
14:46:31linuxstbfunman: What's the context?
14:47:17funmanit's a bit of a status register in the synopsys SD controller
14:47:49funman3 consecutive bits of this reg are: 'data read timeout', 'data starv timeout', 'fifo under/overrun'
14:49:18funmani got rid of this error though: i was reading from the fifo on 2 conditions: fifo isn't full && there are at least 32bits in the fifo
14:49:30funmanwhile it should have been 'fifo isn't empty && there are at least 32 bits in the fifo'
14:53:33 Join halmi [0] (
14:54:43 Join mischasworld [0] (
14:55:16funmanah no this error bit is still present on other conditions
14:56:15 Quit halmi_ (Ping timeout: 246 seconds)
14:58:14 Join mischasworld_ [0] (
14:59:54 Join mischasworld__ [0] (
15:00:44 Quit mischasworld (Read error: Connection reset by peer)
15:02:35 Join robin0800 [0] (
15:02:45 Quit mischasworld_ (Ping timeout: 240 seconds)
15:04:36 Join mischasworld [0] (
15:05:02 Quit mischasworld__ (Ping timeout: 246 seconds)
15:06:34 Join M3DLG [0] (~M3DLG@
15:08:36 Join halmi_ [0] (
15:11:34 Quit halmi (Ping timeout: 276 seconds)
15:13:10 Quit flydutch (Quit: /* empty */)
15:14:50 Quit M3DLG (Ping timeout: 260 seconds)
15:17:13 Quit petur (Quit: Leaving)
15:17:27funmanso, is "starvation" the only possible word which would end "starv" ?
15:18:08 Quit halmi_ (Read error: Connection reset by peer)
15:18:28linuxstbfunman: It's the only one I can think of.
15:18:47 Join flydutch [0] (
15:19:27Bagderispell-complete-word offers only variations of that
15:19:54Bagderstarved,starving etc
15:20:05funmanthat would mean it affects something else than the fifo itself
15:20:09 Join halmi [0] (
15:20:12funmanperhaps linked to fifo threshold
15:20:26 Join dfkt_ [0] (~dfkt@unaffiliated/dfkt)
15:23:40 Nick fxb is now known as fxb__ (
15:24:08 Quit dfkt (Ping timeout: 265 seconds)
15:25:15 Nick dfkt_ is now known as dfkt (~dfkt@unaffiliated/dfkt)
15:25:37kugelBagder: we can still use our strcasecmp in case the OS doesn't provide it. but acting as if it was standard C might lead to problems. I would like to not build libc stuff and don't add the headers to the include path so that there's no clash
15:26:07Bagderyes, I completely understand
15:26:54gevaertstopik: I claim the LED pig bounty!
15:27:40kugelif we put strcasecmp into the libc folder, we would require the OS to provide it. if it doesn't we couldn't use strcasecmp because the whole libc folder isn't built
15:29:27Bagderwell, we could also always use our own version
15:29:40*Bagder runs off
15:30:03kugelright, but it's easier to use our own if it's not in the libc folder
15:30:32funmanwhat is the 'libc folder' ?
15:32:27kugelI proposed to move our files which provide libc functions to a separate folder so that it's easier to use the host OS' libc for the sim/raaa
15:33:20funmancouldn't you check if the OS provide each function?
15:35:02kugelI don't think that's needed for plain standard c functions
15:35:04gevaertsfunman: IIUC that's not really the issue. The question is where we put the things, not how we enable them
15:35:38S_a_i_n_tCan a .kbd file be specified by a theme .cfg?
15:36:15S_a_i_n_tWhen you do "Save Theme Settings" in Rockbox, it doesn't save any line for the keyboard, so I guess not.
15:36:32funmanso what is the problem with strcasecmp, is it locale dependancy ?
15:36:40domonokyS_a_i_n_t: thats currently not a allowed theme setting. so the themesite will reject it.
15:36:45gevaertskugel: intuitively I'd like to see things like strcasecmp in libc. I think I prefer a split between "definitely rockbox code" and "(hopefully) provided by the environment"
15:36:52S_a_i_n_tHaving a theme that can alter the size of the font but not change the .kbd just occured to me as being rather silly.
15:36:56gevaertsS_a_i_n_t: why should that belong to the theme?
15:37:23kugelgevaerts: but what if the OS doesn't provide it (it seems unlikely, but still)?
15:37:31S_a_i_n_tgevaerts: As changing the font can affect the vkeyboard a LOT
15:37:53gevaertsS_a_i_n_t: so can changing the language, even more so actually
15:37:58funmankugel: we could require posix systems
15:38:14gevaertsfunman: unrealistic
15:38:29kugelfunman: why?
15:38:31funmanor at least the subset which implements strcasecmp()
15:38:54kugelit's easier to have strcasecmp not in the libc folder and use it
15:39:06gevaertskugel: is it really?
15:39:10funmanif it's not standard C, then sure
15:39:28S_a_i_n_tdomonoky: Hmmm, OK. So the Themesite will reject it, but can I set it in my own theme on the device?
15:39:51gevaertsS_a_i_n_t: in your own theme on the device you can set whatever you like
15:39:57kugelgevaerts: yes, because the libc folder would be completely ignored when building to use system's libc
15:40:03domonokyS_a_i_n_t: sure. a theme config is just a normal .cfg so it can contain any setting.
15:40:21funmanit's less painful than check each function separately
15:40:36gevaertskugel: sure, if you completely ignore libc/ when building a hosted config. But do you have to completely ignore it?
15:40:38S_a_i_n_tAha, I didn't know what was considered "valid" in a theme .cfg, thanks.
15:41:05gevaertsS_a_i_n_t: the code doesn't care
15:41:05kugelgevaerts: yes, otherwise the headers clash
15:41:19kugelit also saves compilation time
15:41:32gevaertsright, headers are an issue
15:42:13kugelif I added the include dir just for strcasecmp, then all the other headers in the same folder are preferred over the system's ones as well
15:43:01gevaertskugel: maybe have a look at how does it
15:44:07funmanFlynDice: ping
15:44:34FlynDicefunman: pong
15:44:42gevaertskugel: or just use that on OSes that don't have what we need
15:44:48funmanhad you looked already at different values for fifo threshold ?
15:45:15kugelgevaerts: which is easier if strcasecmp is not in the libc folder
15:45:35funmanlinux code seems to mean that the initial value of FIFOTH holds the full fifo depth (0x3f << 16), so the threshold should be lower than that (linux use the half). 0x80 for writes would be too high then
15:45:47funmanI tried using lower values without luck (still FIFO overrun)
15:46:16gevaertskugel: actually, I think it's the reverse then. If you use libiberty, strcasecmp (and hopefully all of the functions in the same group) are provided by "something outside"
15:46:30funmandid you play already with MCI_FIFOTH ?
15:46:34FlynDicefunman: Yes I changed them around some to experiment a while ago and had no luck
15:47:04kugelgevaerts: oh, you meant libiberty with "that"?
15:47:07funmanok, and any luck with PIO ? I always get 3 words remaining from 128 words, perhaps it has to do with threshold
15:47:49gevaertskugel: yes. I think it's worth checking if it (libiberty) has all semi-standard functions we might miss on some OSes, and if it does, just use it on those
15:48:24kugelwhy use another lib just for a single function? that adds unneeded complexity to the build system, IMO
15:48:35funmanI also get varying results depending if I reload FIFO_COUNT from MCI_STATUS after reading 1 word or not
15:48:41kugelall that just to have strcasecmp in the libc folder?
15:48:43gevaertsIs it only strcasecmp?
15:48:53FlynDicefunman: I tried larger and smaller and played with the DMA sizes also with no luck. As far as PIO goes there's a learnin curve going on there but I'm getting closer...
15:49:09 Quit jhMikeS (Ping timeout: 260 seconds)
15:49:35kugelmaybe strcasestr too, but that's all
15:49:36funmanwhat's the current biggest problem you have with it ?
15:49:36 Join jhMikeS [0] (
15:49:37 Quit jhMikeS (Changing host)
15:49:37 Join jhMikeS [0] (~jethead71@rockbox/developer/jhMikeS)
15:50:08kugelno, strcasestr isn't in libiberty
15:50:30gevaertskugel: I don't think it's a terribly important discussion anyway, actually splitting the thing is the important bit. If people don't like the initial solution for those files, they can easily be changed later
15:51:04kugelthat's my thinking too :)
15:52:47*kugel proceeds with the actual work then
15:52:50funmanFlynDice: ah any idea about MCI_INT_HTO ?
15:52:59FlynDicefunman: lol "current biggest problem" is the same problem I always have, ie I'm learning on the fly by looking at your code, linux code, and whatever rockbox code seems to be similar.....
15:53:58*FlynDice wipes eyes and yawns, still early in this part of the world...
15:55:03 Join petur [0] (~petur@rockbox/developer/petur)
15:55:18 Quit petur (Client Quit)
15:55:40*funman hands a ☕ to FlynDice
15:55:50 Join halmi_ [0] (
15:56:05kugelmeh, what to do about memset16?
15:57:44 Join petur [0] (~petur@rockbox/developer/petur)
15:57:45gevaertskugel: not in libc/ I think
15:58:06kugelit needs a memset16.h then, doesn't it?
15:58:45 Quit halmi (Ping timeout: 276 seconds)
15:59:12FlynDicefunman: re INT_HTO are you getting this?
15:59:37gevaertskugel: I suspect the rockbox memory.h could be used (but possibly renamed)
15:59:52funmanFlynDice: yes, i didn't identify the conditions which cause this though
16:00:10funman0x400 / 0x420 => sometimes with INT_RXDR
16:02:08 Quit S_a_i_n_t (Ping timeout: 240 seconds)
16:02:33 Part Boldfilter
16:02:41FlynDiceI read somewhere in one of the "maybe this is the right document" docs that the card clock could be stopped to prevent data underruns. I don't know if that's supposed to be automatic or if we would have to do it manually?
16:02:58 Join S_a_i_n_t [0] (S_a_i_n_t@
16:05:35funmanunsigned count = (MCI_STATUS << 2) >> 19; /* that would be the number of bytes in the FIFO, looks good ? */
16:06:12funmanif I make transfers if count >= 0x10 then i see no transfer at all
16:06:33funmanif I make transfers when using >> 21 (to have the count of *words*) then I see some transfers
16:06:46 Quit mischasworld (Remote host closed the connection)
16:08:01FlynDicefunman: Yes I beleive that's correct
16:09:30 Join halmi [0] (
16:10:48funmanhm fifo depth/threshold is in the same units than this count
16:11:27funmanin bytes
16:12:29 Quit halmi_ (Ping timeout: 260 seconds)
16:12:36funman(according to linux fifo_free)
16:12:44***Saving seen data "./dancer.seen"
16:12:57funmani get down to 12 words remaining to be transfered out of 128 :o
16:13:03funmanin a total of 5 interrupts
16:14:07FlynDicefunman: you're doing reads and not writes is that correct?
16:14:13funmanyes only reads
16:14:23funmanwrite will be the second step
16:14:43 Join CGL [0] (~CGL@
16:14:47funman5 interrupts, including a DTO itnerrupt
16:15:38 Join panni_ [0] (
16:18:56 Quit tchan1 (Quit: WeeChat 0.3.3-dev)
16:20:08funmanso everything has been transmitted by the card but i couldn't read it all
16:21:11funmanbtw in the OF I see different settings: 1 with the setting you use, another one with dma bits = 0, DMA disabled for transfer, and FIFOTH for write = 0x30
16:23:07 Join tchan [0] (~tchan@lunar-linux/developer/tchan)
16:23:50funmanah I found (again) the PIO loop in OF
16:24:23 Join halmi_ [0] (
16:24:27funmanClipv2 V02.01.16A 0x514C: read loop (r0 = fifo)
16:24:49funmanwrite loop: 0x5e18
16:26:38funmanpio read is a potential response handler for CMD 14 ??
16:27:13 Quit halmi (Ping timeout: 276 seconds)
16:27:37funman(a "reserved" command)
16:28:15funmanalso for commands 8 (send if cond) and 51 (send scr)
16:29:19 Join mischasworld [0] (
16:32:30funmanthey loop over 0x40 words == 256 bytes (half a sector)
16:33:32 Join Strife89 [0] (
16:37:23 Quit bgs1001 (Changing host)
16:37:23 Join bgs1001 [0] (znc@unaffiliated/bgs100)
16:37:30 Nick bgs1001 is now known as bgs100 (znc@unaffiliated/bgs100)
16:39:58funmanFlynDice: any code to share?
16:40:24funmanINT_HTO seems to happen when FIFOTH is too high
16:40:25 Join halmi [0] (
16:43:01 Quit halmi_ (Ping timeout: 260 seconds)
16:44:10FlynDicefunman: Nothing pio related. I can give you the code I was using to debug dma writes but it's nothing very interesting....
16:44:24funmananything could help
16:44:47 Quit Tuplis (Read error: Operation timed out)
16:46:27 Join mirak [0] (
16:49:09funmanFlynDice: PIO 'working' but not transferring all bytes
16:50:11FlynDicefunman: as I said, nothing real interesting but this is my last stash for write debug code −−>
16:51:25 Quit funman (Quit: free(random());)
16:54:07 Join halmi_ [0] (
16:54:28 Quit CGL (Quit: Soy spammero
16:56:24 Quit halmi (Ping timeout: 240 seconds)
16:57:18 Join p3tur [0] (~petur@rockbox/developer/petur)
16:57:25 Join einhirn [0] (
16:58:51 Quit petur (Disconnected by services)
16:59:01 Nick p3tur is now known as petur (~petur@rockbox/developer/petur)
17:01:16 Quit BlakeJohnson86 (Ping timeout: 265 seconds)
17:08:49 Join halmi [0] (
17:08:57 Join M3DLG [0] (~M3DLG@
17:09:21 Join Tuplis [0] (
17:11:32 Quit halmi_ (Ping timeout: 252 seconds)
17:13:09 Join BlakeJohnson86 [0] (
17:13:35 Join DataGhost [0] (~dataghost@unaffiliated/dataghost)
17:23:34 Join halmi_ [0] (
17:26:02 Quit halmi (Ping timeout: 246 seconds)
17:28:28 Quit mischasworld (Ping timeout: 264 seconds)
17:30:46 Quit halmi_ (Quit: halmi_)
17:36:50 Join mischasworld [0] (
17:36:59 Join p3tur [0] (~petur@rockbox/developer/petur)
17:37:36 Quit petur (Disconnected by services)
17:37:45 Nick p3tur is now known as petur (~petur@rockbox/developer/petur)
17:38:51 Quit TheSeven (Ping timeout: 276 seconds)
17:39:44 Quit M3DLG (Ping timeout: 260 seconds)
17:41:36 Quit GeekShadow (Ping timeout: 260 seconds)
17:43:00kugelshould I edit or make my own?
17:43:18 Join halmi [0] (
17:44:53gevaertsmake your own I'd say
17:48:11 Quit flydutch (Quit: /* empty */)
17:48:16gevaertssounds goof
17:49:15 Join Strife1989 [0] (
17:51:32 Quit Strife89 (Ping timeout: 240 seconds)
17:53:05 Join halmi_ [0] (
17:53:12 Quit DataGhost ()
17:55:38 Quit halmi (Ping timeout: 240 seconds)
17:55:53 Quit BlakeJohnson86 (Ping timeout: 265 seconds)
18:03:05 Join DataGhost [0] (~dataghost@unaffiliated/dataghost)
18:05:24 Quit panni_ (Read error: Connection reset by peer)
18:08:07 Join halmi [0] (
18:10:20 Quit halmi_ (Ping timeout: 240 seconds)
18:12:45***Saving seen data "./dancer.seen"
18:18:24 Join BlakeJohnson86 [0] (
18:20:55 Quit petur (Ping timeout: 264 seconds)
18:22:25 Quit domonoky (Ping timeout: 260 seconds)
18:23:26 Join halmi_ [0] (
18:23:40 Join domonoky [0] (~Domonoky@rockbox/developer/domonoky)
18:25:13 Quit halmi (Ping timeout: 246 seconds)
18:29:14 Join GeekShadow [0] (~Antoine@reactos/tester/GeekShadow)
18:31:03kugelgevaerts: do you want to look over my wiki page before I hit the save button?
18:31:10 Quit BlakeJohnson86 (Ping timeout: 265 seconds)
18:32:15 Join skfunnyboy [0] (
18:32:37 Quit domonoky (Read error: Connection reset by peer)
18:32:48gevaertsThat's up to you I'd say
18:34:26gevaertsI'm willing to have a look if you want, but I don't see a real need for me to check it first
18:36:14 Join domonoky [0] (~Domonoky@rockbox/developer/domonoky)
18:37:13 Quit skfunnyboy (Quit: CGI:IRC)
18:38:09 Join halmi [0] (
18:38:24 Join BlakeJohnson86 [0] (~bjohnson@2002:1876:a27b:0:227:13ff:fe65:1262)
18:41:04 Quit halmi_ (Ping timeout: 264 seconds)
18:43:14 Join skfunnyboy [0] (
18:43:27 Quit GeekShadow (Ping timeout: 264 seconds)
18:44:42 Join petur [0] (~petur@rockbox/developer/petur)
18:45:10 Join RandomInsano [0] (
18:45:12skfunnyboyi think i have found a bug in RockBox for sansa fuze v2
18:46:25 Join p3tur [0] (~petur@rockbox/developer/petur)
18:54:06 Join halmi_ [0] (
18:54:51 Quit grndslm (Ping timeout: 265 seconds)
18:56:18 Join M3DLG [0] (~M3DLG@
18:56:20 Quit halmi (Ping timeout: 246 seconds)
18:56:22S_a_i_n_tskfunnyboy: then please, enlightn us.
18:57:57 Join flydutch [0] (
18:58:09 Join fdinel [0] (
18:59:43kugelgevaerts: ok, the wiki page is up
19:01:26skfunnyboywhen i exit a game in rockboy, i return to the navigator ( or ''Files'' ) and when i want to choose another game...the blue light blink .. (sorry for my bad english .. i will make a video showing the problem soon)
19:02:16 Join halmi [0] (
19:02:18 Quit halmi_ (Quit: halmi_)
19:02:51kugelskfunnyboy: don't bother
19:02:51S_a_i_n_tYou have some form of sansa player? I'm not too familiar with the sansa targets...but I'm fairly certain this issue is known.
19:03:02kugelyes, it is known
19:03:10S_a_i_n_tAh, there you go. Apparently it is.
19:03:44S_a_i_n_tBut, thanks anyway.
19:03:57 Quit M3DLG (Read error: Connection reset by peer)
19:04:20 Join M3DLG [0] (~M3DLG@
19:04:39skfunnyboyif i found more bugs..i will tell you :P
19:05:24S_a_i_n_tPlease do.
19:05:48kugelplease don't
19:06:15kugelthe fuze v2 is not a target we want bugs report on, it's considered unusable it all aspects
19:06:46skfunnyboyi use it on my Sansa Fuze v2 .. .Work almost like a charm!
19:07:27 Quit mischasworld (Ping timeout: 260 seconds)
19:07:51 Join grndslm [0] (
19:08:57 Join halmi_ [0] (
19:09:24 Join mischasworld [0] (
19:10:31 Quit shai (Quit: Leaving)
19:12:06 Quit halmi (Ping timeout: 268 seconds)
19:12:09 Join GeekShadow [0] (~Antoine@reactos/tester/GeekShadow)
19:22:23 Join rbert [0] (
19:24:41RandomInsanoAnyone know if there's special voodoo needed to build a bootloader for Telechips devices? My attempts to get running code are a fruitless as ever.
19:24:57 Join halmi [0] (
19:26:30RandomInsanoOr getting output from GPIO using telechips.c:main() isn't being done properly... that's always a possibility
19:28:02 Quit halmi_ (Ping timeout: 276 seconds)
19:30:19 Quit M3DLG (Ping timeout: 246 seconds)
19:34:45gevaertskugel: plain alsa doesn't seem to work on n900. "Playback open error: Device or resource busy"
19:34:56 Join archivator [0] (
19:35:42 Join shaggy-h [0] (
19:36:45kugelthanks for testing
19:37:06kugelI'm curious why none of my work wants on your phone :(
19:37:34gevaertsIt's because you stated in public that you'll work on android first. It doesn't like that
19:38:00kugelgevaerts: btw, I followed that guide for the qemu debian thing. the debian installation got stuck at the creation of the root file system
19:38:23kugelI see
19:38:40kugelcan you hug it a bit and tell it I'm sorry?
19:39:31*S_a_i_n_t thinks kugel may need to grovel to gevaerts' n900 in person.
19:39:52S_a_i_n_twait 'til DevCon ;)
19:40:52kugelgevaerts: does that example work?
19:44:56 Join Rondom [0] (
19:45:39 Join halmi_ [0] (
19:45:53 Join BHSPitMonkey [0] (~stephen@unaffiliated/bhspitmonkey)
19:46:29 Quit RandomInsano (Quit: CGI:IRC)
19:46:59gevaertskugel: no, same error
19:48:07 Quit halmi (Ping timeout: 240 seconds)
19:49:51rbertwhois rbert
19:50:07 Quit rbert (Quit: - Chat comfortably. Anywhere.)
19:50:59 Join shai [0] (
19:56:05kugelstrange error:
19:56:33kugelit happens when I include my system's stdlib.h instead of our own
19:59:37 Join halmi [0] (
20:02:43 Quit halmi_ (Ping timeout: 240 seconds)
20:06:49 Nick Llorean1 is now known as Llorean (
20:06:54 Quit Llorean (Changing host)
20:06:54 Join Llorean [0] (~DarkkOne@rockbox/user/Llorean)
20:08:48 Nick Strife1989 is now known as Strife89 (
20:12:49***Saving seen data "./dancer.seen"
20:14:23 Join halmi_ [0] (
20:16:15 Join anewuser [0] (anewuser@unaffiliated/anewuser)
20:16:40 Nick YPSY is now known as Ypsy (
20:17:18 Quit halmi (Ping timeout: 276 seconds)
20:18:20 Quit togetic (Ping timeout: 276 seconds)
20:20:00 Join togetic [0] (~togetic@unaffiliated/ibuffy)
20:20:51 Quit robin0800 (Remote host closed the connection)
20:23:49 Nick Ypsy is now known as YPSY (
20:29:21 Quit flydutch (Quit: /* empty */)
20:31:13kugelI don't get it. stdlib includes sys/types.h for int32_t, but placing an #error in sys/types.h doesn't give an error :/
20:32:11kugeloh, we have our own types.h :\
20:37:30 Join ender` [0] (
20:39:55 Quit GeekShadow (Ping timeout: 246 seconds)
20:40:56kugeleh, our fdprintf is different to the standard's one
20:42:11kugeloh, nevermind, there's no fdprintf
20:42:14 Join phanboy4 [0] (
20:45:55 Quit mischasworld (Ping timeout: 240 seconds)
20:52:37 Join mischasworld [0] (
20:58:53 Join saratoga [0] (~463f90ed@gateway/web/freenode/x-vyivaamjbgyrhhyw)
20:59:23saratogakugel: i found a site with debian (or maybe ubuntu i forget) QEMU arm images a while back
21:01:45 Join M3DLG [0] (~M3DLG@
21:04:21 Join rbert [0] (
21:08:31kugelwow, the libc6 folder thing it's more work that I expected
21:10:58 Quit p3tur (Quit: *plop*)
21:11:57kugelpamaury: ah, I solved how the sim gets fprintf
21:12:06kugelI think
21:19:39 Quit S_a_i_n_t (Quit: Restarting.)
21:20:10 Join S_a_i_n_t [0] (S_a_i_n_t@
21:26:07 Quit mischasworld (Ping timeout: 240 seconds)
21:32:40saratogawhen do we feature freeze for 3.6?
21:35:26saratogai guess 2 weeks from today if we're going to hit the scheduled day to release
21:36:51 Join wodz [0] (
21:37:06*kugel cries at "unsigned int32_t"
21:38:45 Join piepiepie [0] (
21:38:59wodzhahaha. I'am reading rockchip 27xx programing manual and there is "Decoder remap function. Write value 0xdeadbeef to this register to start remap function" :-)
21:39:28saratogathis is the player with the arm7e core?
21:39:31 Quit piepiepie (Client Quit)
21:39:47 Quit Xerion (Ping timeout: 260 seconds)
21:40:30saratogasuch a weird core
21:40:57wodzI started wiki page about this platform :
21:41:09wodzwhy weird?
21:41:19wodzcore like a core
21:42:16wodzDSP included is weird - something between DSP and RISC processor
21:44:07 Join robin0800 [0] (
21:44:54 Quit saratoga (Ping timeout: 252 seconds)
21:46:24wodzthe engineers designing this SoC had specific sense of humor: "If write “ 0xdeedbabe” will power down ARM7 clock"
21:48:36 Join saratoga [0] (~463f90ed@gateway/web/freenode/x-gkzlzlqupdpieilv)
21:48:56saratogaweird because its a 3 stage arm7 pipeline with DSP extensions and armv5 support
21:49:09saratogaeven a lot of newer and faster arm9 cores don't even have that
21:50:10kugeldoesn't understand "apps/codecs/lib/codeclib.h:64: error: expected identifier or ‘(’ before ‘__extension__’" :(
21:50:59 Quit M3DLG (Ping timeout: 260 seconds)
21:51:15wodz3 stage? ARM technical reference manual states it is 5 or 6 stage pipeline (depends if Jazzele mode is on)
21:51:50kugelsaratoga: it's not a usual arm7
21:52:01kugelI never heard of arm7ej-s actually
21:52:09saratogawodz: i've seen those floating around, the ones that show the arm9e pipeline right?
21:52:45saratogai think thats a typo, the website and most other things i've found say its 3 stage:
21:53:08wodzit states 5 or 6 stage
21:53:10saratogayeah i have that copy, but thats the arm9e pipeline
21:53:57 Join dantje_ [0] (
21:54:11kugelit's basically an arm9ej-s with a 3-stage pipeline?
21:54:48saratoga"ARM7EJ-S Bringing Java and DSP acceleration to low-cost 32-bit designs Integer core with three-stage pipeline"
21:54:57saratogaapparently the arm people have no idea what they have
21:55:18saratogakugel: yeah, or probably more the arm7 pipeline with the 'E' multiplier
21:55:26wodzI am by no means ARM expert
21:55:44saratogawodz: well do you trust the datasheet or the website more :)
21:55:50kugelwell, it's armv5 which has a number of other improvements over the armv4 of arm7tdmi
21:56:21saratogamost of them are part of the 'E' multiplier though
21:57:09kugelcaches and mmu also
21:58:42wodzWhat is AHB interface?
21:59:21saratoga"A five-stage ARM state pipeline is used, consisting of Fetch, Decode, Execute, Memory, and Writeback stages. This is shown in Figure 1.1."
21:59:30saratogathey really don't seem to know whats in the arm7ej . . .
21:59:32 Join CGL [0] (~CGL@
21:59:49saratogaisn't the ahb the bus used for various arm perphierals
22:00:37wodzI suppose (from the context)
22:02:47saratogaon AMS devices, the ahb connects things like the DMA controller and memory controller to the CPU
22:03:07saratogaah "advanced high speed bus"
22:03:54saratoga32 bit, 64 MHz (on AMS anyway)
22:06:25wodzup to 133MHz on rt27xx
22:06:39wodzbut bus width is not stated
22:11:59 Quit bluebroth3r (Ping timeout: 260 seconds)
22:12:52***Saving seen data "./dancer.seen"
22:13:41 Join bluebrother [0] (
22:13:41 Quit bluebrother (Changing host)
22:13:41 Join bluebrother [0] (~dom@rockbox/developer/bluebrother)
22:17:06 Quit einhirn (Read error: Connection reset by peer)
22:17:30 Join mischasworld [0] (
22:20:27 Quit wodz (Quit: Leaving)
22:20:43kugelTorne: ping
22:23:22 Quit TillW (Quit: this concludes our broadcast day)
22:26:04kugelsdl slips -D_GNU_SOURCE=1 in, which is bad
22:31:29 Quit robin0800 (Remote host closed the connection)
22:32:47CIA-5New commit by alle (r25768): Small correction of the translation of a musical term
22:33:24 Quit saratoga (Quit: Page closed)
22:39:35 Quit pixelma (Disconnected by services)
22:39:36 Join pixelma_ [0] (quassel@rockbox/staff/pixelma)
22:39:55 Nick pixelma_ is now known as pixelma (quassel@rockbox/staff/pixelma)
22:40:03 Quit amiconn (Ping timeout: 248 seconds)
22:40:07 Join amiconn_ [0] (quassel@rockbox/developer/amiconn)
22:40:29 Nick amiconn_ is now known as amiconn (quassel@rockbox/developer/amiconn)
22:41:07 Quit BHSPitMonkey (Read error: Connection reset by peer)
22:41:23 Quit mischasworld (Remote host closed the connection)
22:41:48 Quit shai (Quit: Leaving)
22:51:50 Quit grndslm (Ping timeout: 264 seconds)
22:54:55 Quit dantje_ (Quit: Ex-Chat)
22:59:38 Join TheSeven [0] (~TheSeven@rockbox/developer/TheSeven)
23:02:35 Quit Unhelpful (Remote host closed the connection)
23:04:52 Join grndslm [0] (
23:05:37kugelshould I make a single header file for strlcpy, strlcat, strcasecmp and strcasestr or one for each?
23:05:51 Quit phanboy4 (Read error: Connection reset by peer)
23:07:04 Quit antil33t (Ping timeout: 276 seconds)
23:10:57 Quit TheSeven (Disconnected by services)
23:10:59 Join The_Seven [0] (~TheSeven@rockbox/developer/TheSeven)
23:11:51 Join antil33t [0] (
23:12:15 Join mischasworld [0] (
23:13:38 Join Unhelpful [0] (~quassel@
23:13:38 Quit Unhelpful (Changing host)
23:13:38 Join Unhelpful [0] (~quassel@rockbox/developer/Unhelpful)
23:17:48 Join RandomInsano [0] (
23:21:07 Nick fxb__ is now known as fxb (
23:23:34 Join Moguta [0] (
23:25:01MogutaSorry for the 'n00b' quest, but are there any still-sold MP3 players that Rockbox is supported on?
23:26:02MogutaLooking through the supported models and Googling them, I don't seem to see any still in retail.
23:26:30krazykitthe e200v2 is still being sold, i think
23:26:57 Join liar [0] (
23:31:23MogutaThe e200 doesn't seem to be.
23:31:52MogutaIs the Clip really all?
23:33:56RandomInsanoI think the chief problem is reverse engeneering takes time, probably longer than any product lifecycle
23:34:39RandomInsanoThe sansa clip is an awesome player for the money though. Sister has one.
23:34:57RandomInsanoWon't let me put rockbox on it :P
23:35:13FlynDiceMoguta: I believe you can still buy clipv2, clip+ and fuzev2 presently. None of these targets are "stable" though as you can see on the front page. They are all categorized as "unusable" which you can also find on the front page....
23:36:48 Quit DerPapst (Quit: Leaving.)
23:37:39gevaertskugel: I think a single one is fine
23:37:42 Quit JohannesSM64 (Ping timeout: 246 seconds)
23:38:26kugelI thought so too but I lack a good name :)
23:38:30 Quit domonoky (Read error: Connection reset by peer)
23:39:10kugelthis patch is getting really huge
23:39:38 Quit Unhelpful (Remote host closed the connection)
23:43:27 Join guymann [0] (~charlie@
23:46:27 Join Unhelpful [0] (~quassel@
23:46:28 Quit Unhelpful (Changing host)
23:46:28 Join Unhelpful [0] (~quassel@rockbox/developer/Unhelpful)
23:46:28 Join matsl [0] (
23:51:30kugel154 files changed, 255 insertions(+), 474 deletions(-) :(
23:53:25 Quit Schmogel (Quit: Miranda IM! Smaller, Faster, Easier.
23:53:26*gevaerts recommends to try to get that one committed soon
23:55:38Strife89kugel: I'm curious, what patch? (I don't see it in my logs.)
23:56:08 Join JohannesSM64 [0] (
23:58:45 Quit Unhelpful (Remote host closed the connection)

Previous day | Next day