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 2007-09-28

00:00:07nls"We do not estimate timeframes"
00:04:28preglowmonth, hell no, two, maybe, but probably not
00:04:42preglowonly way to get going is to fire up the code editor and talk to Nico_P
00:04:54preglowasking for deadlines will not further anything
00:06:11Nico_PDark_Apostrophe: hopefully it'll happen sooner than later
00:06:20 Quit Bam2550 (Read error: 110 (Connection timed out))
00:06:27 Quit davina (Remote closed the connection)
00:06:46Dark_Apostrophepreglow: I'm not asking for deadlines, just crude estimates
00:06:51Dark_ApostropheNico_P: good. :)
00:06:52preglowwe have none, so that's easy
00:08:22Nico_PDark_Apostrophe: but it's been about 1.5 years since I opened FS #3045, so I hope on or two months more or less won't be a problem :)
00:08:27 Part toffe82
00:08:43Dark_ApostropheNico_P: :o
00:08:53preglowlooking forward to it
00:08:58preglowthat day we'll have to think about other wps issues too
00:09:06preglowlike non-centered text
00:09:43*nls shakes head at FS #7847
00:09:44Nico_Pyup... hopefully it'll make amiconn feel the need to implement viewports
00:10:01Nico_Por maybe he'll have already done it :p
00:10:06preglownls: looks like a closer...
00:10:13preglowNico_P: i've got a feeling he won't
00:10:15 Join iamben [0] (
00:10:25nlspreglow: already taken care of :-)
00:10:34preglowbut let's hope that gives him some incentive, heh
00:10:37preglowi'd love to see viewports
00:10:45preglowcan't think of any better schemes
00:10:50Nico_Pme too
00:11:39 Join Dreamxtreme [0] (i=5603213e@gateway/web/cgi-irc/
00:11:40 Join Isolinear [0] (
00:13:09 Quit Isolinear (Client Quit)
00:13:12Dreamxtremeany1 about?
00:13:19preglowplenty, what's up?
00:13:47Dreamxtremeim thinking of buying a ipod 2mrw
00:13:57Dreamxtremeand wouldn't mind the new classic
00:14:01preglowcome again?
00:14:15 Quit przemhb ()
00:14:17Dreamxtremebut only if theres plans for a rockbox on there
00:14:31Nico_Pthere are no plans
00:14:40preglowwe can plan all we want, apple are trying as hard as they can to prevent us from supporting it
00:14:52preglowthe day someone shows us how to break the decryption, we'll be on it
00:14:59Dreamxtremeo really dam
00:15:18preglowup to us: hell yes
00:15:21preglowbut it's not up to us
00:15:30Dreamxtremewhat about the V2 video 80GB>
00:15:51preglowi think we do support that, yes
00:15:52Nico_Pthe video is supported, yes
00:15:57preglowwe call it 5.5g video
00:16:05Dreamxtremeo good
00:16:33 Quit Dreamxtreme (Client Quit)
00:16:39Nico_Pbecause that's what it's called
00:16:43preglowme answering requires you to not quit
00:16:48preglowi'm sorry, but that's the way it is
00:17:38*preglow summons tomal
00:17:57 Join Dreamxtreme [0] (i=5603213e@gateway/web/cgi-irc/
00:18:11TMMpreglow: wow! it worked!
00:18:17preglowhe's not tomal :/
00:18:38TMMpreglow: baby steps, baby steps
00:19:15Dreamxtremeif i play FLAC will that reduce the battery life?
00:19:38preglowa bit, probably
00:19:44preglowthe disk spins up more often
00:19:49preglowbut it doesn't use much cpu
00:20:40Dreamxtremei wonder if there will be a rockphone
00:21:00preglowthat would mean much work for us, so let's hope not :P
00:21:07TMMDreamxtreme: on a side note: if you use a portable player, with portable earphones, I wouldn't really bother with FLAC to be honest, use 320kb/s ogg, or AAC :)
00:21:34iambenDreamxtreme: in my experience, flac uses much less cpu than higher bitrate mp3
00:21:35Dreamxtremewhy not mp3?
00:21:37iambenon my sansa anyway
00:22:10idnarI use FLAC because I'm too lazy to transcode
00:22:46Dreamxtremeso am i
00:22:50iambeni use flac on my pc and ogg when i put it on portable
00:23:04Dreamxtremei just want to copy and paste
00:23:14iambenbut flac battery time is fine, only problem is the filesizes
00:23:25TMMwhy not mp3? because I just plain dislike mp3 :)
00:23:50idnarFLAC does seem to use more battery than Vorbis or Musepack
00:23:58idnarpresumably because of hitting the disk more
00:24:14TMMwould it matter anything on a NAND based player?
00:24:15preglowiamben: if you're on a flash player, then yes, flac will use less
00:24:21iambenyeah my experience has been on sansa where there is no disk per say
00:24:22preglowif you're on a hd based player, it will probably use more
00:24:35iambenper se
00:24:36preglowour flac decoder is so cpu efficient it's not even bloody funny
00:24:40iambenwhatever they say =)
00:24:44preglowbut it still needs to spin up the disk often
00:25:11preglowif the disk is a flash chip: then hooray!
00:25:13TMMpreglow: just solder some more memory on your player :P
00:25:15 Join freeeeex [0] (i=d5407a79@gateway/web/cgi-irc/
00:25:19 Join toffe82 [0] (
00:25:28preglowTMM: man, i can't even open the on flash player i have up :P
00:25:28iambenon my sansa, w/ higher bitrate mp3, i notice sluggishness in scrolling the menus... very quick w/ flac playing
00:25:39preglowtried once with a guitar pick, and now the damned thing just looks damaged
00:25:59TMMpreglow: well, I'm in for that quite soon :) I wonder how it'll turn out
00:26:26freeeeexjust wondering, is it possible remapping the keys in the search > database program?
00:26:32preglowTMM: what kinda player?
00:26:34freeeeexon an ipod
00:26:40TMMpreglow: iaudio 7
00:27:12 Join Isolinear [0] (
00:27:40preglowTMM: well, as long as it's got screws, you'll be better off than i were
00:27:46preglowdoes it have screws? :P
00:27:53TMMpreglow: I'm screwed
00:28:11TMMpreglow: which means, it doesn't have any
00:28:53TMMO well, no problem I'll find a way to do this properly
00:29:03TMMsome other people managed it, but failed to make proper scans :)
00:30:00freeeeexguess not >,<
00:30:15preglowfreeeeex: remap keys: no
00:30:33preglowWorld renowned sound enhancer - JetEffect
00:30:35preglowsweet lord
00:30:43preglowflanger is a world renowned sound enhancer
00:30:57preglowi've already coded that in a plugin for rockbox, am i world renowned?
00:31:15freeeeexso ive been spenind 20mins lookin through friggin c-code for nothin then ;P
00:31:35preglowfreeeeex: if you're looking through c code, then surely you know that you can just fix this yourself :P
00:32:18freeeeexjeez, anyone can read c-code reading != understanding ;P
00:32:38 Join LambdaCalculus37 [0] (
00:32:50Dreamxtremeright im off 2 bed GN
00:32:56preglowhave fun
00:33:03 Quit Dreamxtreme ("CGI:IRC")
00:33:04preglowin a good way!!!
00:33:12LambdaCalculus37zook: You still around?
00:33:17 Join Dreamxtreme [0] (i=5603213e@gateway/web/cgi-irc/
00:33:18 Quit Dreamxtreme (Client Quit)
00:33:27zooookyeah, hey
00:33:53LambdaCalculus37zooook: Sorry I took so long in getting here.
00:34:02zooookno worries
00:34:18zooookbeen playing around with the decryption functions
00:34:34LambdaCalculus37So, find anything else about from the decryption functions?
00:35:10preglowwhat decryption functions? :>
00:35:22preglowbluebrother: swiftly closed...
00:35:23 Quit kubiixaka (Read error: 104 (Connection reset by peer))
00:35:31zooooknot really, I need to make a printout, get the tms320 manual and make comments while debugging it
00:35:48LambdaCalculus37I think the TMS320 manual's on the wiki... or am I wrong?
00:35:51zooookzen firmware
00:36:21zooookdoesn't really matter. it comes with code composer studio
00:37:14 Join Bam2550 [0] (
00:37:20zooookI thought I could feed the output from it's disassembler into the assembler but it complains about every single line
00:37:47LambdaCalculus37zoook: Actually, I remember that I posted a TMS320 datasheet on the Archos AV1XX page.
00:38:23zooookso I have to include the functions as data, which sucks, because the debugger can't follow the flow properly
00:38:49zooookahh, but this is mostly a matter of looking at the data flow while debugging it
00:39:22LambdaCalculus37zoook: You at least did one hell of a job by getting the decryption code up in the simulator, though.
00:39:30zooookie. keep track of all values, note how they're computed and then code an emulator
00:39:31 Quit sin613 (Read error: 110 (Connection timed out))
00:39:54 Join midgey [0] (
00:39:58zooookwell, was just a matter of relocating all branches
00:40:07TMMdoes anyone have that rockbox wiki link about the iaudio 7?
00:40:17TMMI can't seem to be able to find anything
00:40:35LambdaCalculus37TMM: Was this what you're looking for?
00:40:35TMMI saw it TODAY :)
00:40:41TMMYES! :)
00:40:46TMMhow in the hell do you search the wiki?
00:41:04freeeeexgoogle ! ^^
00:41:16LambdaCalculus37freeeeex: Hush, you.
00:41:26freeeeexhaha, it works awsome
00:41:27LambdaCalculus37TMM: See the white bar with "Go" next to it?
00:41:31freeeeexsite:url query
00:41:32TMMpreglow: it DOES have screws :)
00:41:32zooookanyway, it'll be exciting to see how many other block types can be decrypted with this scheme.. I'm guessing FBOOT will
00:41:47TMMI couldn't find it when looking for 'iaudio'
00:42:18LambdaCalculus37zooook: I've been looking at some of the other ZEN/NOMAD players, and they all use a very, very similar rescue mode.
00:42:20TMMwhat did you search for?
00:42:35LambdaCalculus37TMM: CowonIaudio7Info
00:42:46LambdaCalculus37You have to search by page name.
00:42:56TMMhow was I supposed to know THAT? :)
00:43:18zooookthe code i've looked at is almost exactly the same amongst the different models.. I think the main difference is the versioning
00:43:24 Quit midkay ("Leaving")
00:44:00pixelmayou can also use the "search" link to the left of the "go" one, you'll be taken to a search page...
00:44:43 Quit bluebrother ("sleep")
00:45:15LambdaCalculus37zooook: The only problem is with the Dell DJ, you can't navigate with a bare PCB. I think the scroll barrel and the buttons need some pressure from the front plate to register.
00:45:22TMMLambdaCalculus37: so, you can not find a page if you don't know what it is called? :(
00:45:42zooookhow is that a problem?
00:45:51LambdaCalculus37TMM: pixelma just mentioned the search function.
00:46:11freeeeexpreglow: is it possible remapping those keys without being superb at C ?
00:46:16 Join sin613 [0] (
00:46:34LambdaCalculus37zooook: No navigation when you take the PCB out of the casing.
00:46:36freeeeexnot to mention i can find the bloody search
00:46:45TMMpixelma: let me try... I think I tried that
00:47:44TMMpixelma: yay! :)
00:47:58TMMpixelma: I tried the 'search' button on the left of the site
00:48:18pixelmaLambdaCalculus37: I think he's talking about a different thing
00:48:19zooookimagine not even having a pcb to take out of the casing ;)
00:48:28 Quit obo ("bye")
00:48:32TMMpixelma: the one on the top works fine, the button thingy doesn't really :)
00:48:54nlsfreeeeex: you don't need to be superd, just be able to change a couple of defines around unless you're doing something fancy?
00:49:41pixelmaTMM: can't follow your discription...
00:49:56zooookanyway, figuring out the decryption scheme would open up some of the zen line to custom firmware
00:50:11TMMpixelma: on the left of the site, there's a search 'button', on the top there's a search 'link' :) the link works, the button that goes to google doesn't really help
00:50:25nlsDamn that mailing list I should really go to bed instead of writing mails, now I have to get up in 6 hours :-/
00:50:25LambdaCalculus37zooook: I would love to just turn on that Dell DJ and see the Rockbox logo greet me one day! :)
00:51:00zooookI bet :)
00:51:28zooookthat's not really my stop, though
00:51:38pixelmaTMM: ah okay, might depend on what you are searching for (and where). The top link is for searching the wiki the other for searching the complere site with the help of google
00:52:02LambdaCalculus37zooook: Bagder also mentioned this to me before as well... Archopen has some TMS320 code up and running on portable devices.
00:52:13LambdaCalculus37Bagder: Take a bow for that tidbit of information. :)
00:52:20freeeeexnls: the problem is i cant find where the keys are set for the search function (the thing that searches the database), ie im not lookin to remap keys in general
00:53:04nlsfreeeeex: I think that's a plugins, but now I'm really going to bed, gnight
00:53:14zooookright, I've got the code of that.. the cowon a2 uses one too, iirc
00:53:24freeeeexhm ok
00:53:29 Quit nls ()
00:53:45LambdaCalculus37zooook: I believe it does.
00:54:28 Quit ender` (" Be wary of strong drink. It can make you shoot at tax collectors and miss. -- Robert A. Heinlein")
00:54:30zooookgotta figure out the the newer security scheme before that's a concern, though
00:55:12zooookthe ones where the CENC block is replaces by a TL
00:55:23freeeeexfooks, feels like this is beyond me lol
00:55:52LambdaCalculus37freeeeex: What are you talking about?
00:55:53zooookthe Micro has firmware versions that use both schemes
00:56:47LambdaCalculus37zooook: You mean these?
00:57:00 Quit zooook (Remote closed the connection)
00:57:03freeeeexum, im want to rebind the keys used in the search plugin for an ipod.
00:57:16 Quit My_Sic (Read error: 104 (Connection reset by peer))
00:59:20 Quit funky_ ("leaving")
00:59:50freeeeexand its not proving to be as straight forward as i thought it would be lol
01:00:32TMMdid anyone get any code running on creative devices yet?
01:00:34 Join zooook [0] (i=user@gateway/tor/x-f673c30faa1fcfa5)
01:01:49LambdaCalculus37TMM: No, not yet.
01:02:06zooookanyway, I'll keep the thread updated on the CENC decryption scheme
01:02:51LambdaCalculus37zoook: Okay. I'll check in to see how it's going.
01:03:15LambdaCalculus37zooook: Think we should try and dump the firmware from the Dell DJ if possible?
01:03:17TMMLambdaCalculus37: too bad
01:04:02LambdaCalculus37TMM: It may be a while before this happens.
01:04:08zooooka hdd dump from a newly formatted and flashed would be interesting
01:04:23TMMLambdaCalculus37: well, I don't have them anyway, but they appear to grow in numbers :) I see them on the train more often
01:04:33zooookI still don't know where the F* blocks end up
01:04:36LambdaCalculus37zooook: I need to buy an adapter for the Dell DJ's hard drive first.
01:04:59zooookbut it's not essential
01:05:00TMMwell, time for bed
01:05:02TMMsleep well people
01:05:47zooookthe Dell DJ is only protected in the form of the encrypted blocks.. once those are known, the door is wide open
01:07:13LambdaCalculus37zooook: Makes me wish at least one or two devs lived near New York, or even Jersey.
01:07:25zooooksomeone would still need to begin a port, of course
01:07:33LambdaCalculus37zooook: Yes, that's true.
01:08:10LambdaCalculus37My skills in C are not what you can call the greatest.
01:08:27LambdaCalculus37I can read C, and I can at least patch builds and compile... but writing from scratch?
01:08:30LambdaCalculus37Not quite yet.
01:08:54zooookahh, that's a shame
01:09:22LambdaCalculus37zooook: Don't worry. It just takes a little more time.
01:09:23zooookbut let's see how things work out
01:09:35 Quit RaZorbacK ("bye all")
01:09:50***Saving seen data "./dancer.seen"
01:10:27*LambdaCalculus37 will be back in two shakes
01:10:59 Join darkapostrophe [0] (
01:13:36zooookI have been considering getting an elder zen to play around with, that'd change my incentive, but without the hardware there's no fun in trying
01:14:50*LambdaCalculus37 hath returned
01:15:14LambdaCalculus37zooook: I bought the Dell DJ just to try my hand at a new port.
01:15:31LambdaCalculus37I was always curious about the NOMAD and ZEN players.
01:15:52LambdaCalculus37It was another case of "nice hardware, lousy software".
01:16:07zooook <−− have you read this?
01:16:15 Join Klevi [0] (
01:17:11LambdaCalculus37zooook: Reading it now.
01:17:18 Quit freeeeex ("CGI:IRC")
01:18:18zooookI haven't tried comparing the parts they suggest using, but the features does sound familiar
01:18:54LambdaCalculus37zooook: Wait, hold the phone...
01:19:20 Join My_Sic [0] (
01:19:59LambdaCalculus37HOLY CRAP!!
01:20:20LambdaCalculus37zooook: The datasheet for the TMS320DA255!!
01:20:43LambdaCalculus37zooook: I've been looking all over the place for this!
01:21:37LambdaCalculus37JTAG... it says it has a JTAG...
01:21:54zooooknah, there's a lot missing. it's just a reference design.. but a little is better than none
01:22:00 Quit Entasis ("Leaving")
01:22:40LambdaCalculus37zooook: A little that's going to go a long way! You normally can't get this from TI without signing a whacking great NDA.
01:22:58preglownice to hear you talking about it here, then!
01:23:34zooookthank god for the koreans, then :)
01:23:56LambdaCalculus37preglow: This datasheet is like Mana from Heaven!
01:24:21zooookit's usually russians that leave stuff all over their sites
01:24:54LambdaCalculus37zooook: No, trust me... the Koreans can be pretty notorious about it, too.
01:25:20LambdaCalculus37In whatever case, now we can figure out the TMS320DA25x CPUs in the Dell DJ, and possibly other DAPs with the same CPU.
01:25:54zooookit's always fun digging for gold on a site you don't understand, anyway :)
01:26:45LambdaCalculus37zooook: Only one itty bitty little drawback...
01:27:09LambdaCalculus37We can't exactly post this up on the site, not with "TI Developer's Confrence" staring out at you from every page.
01:27:20 Part toffe82
01:27:54zooookwell, FRESC contains the base OS, so if those match amongst different models, you can be certain they use the same architecture
01:28:07 Quit Dark_Apostrophe (Success)
01:28:07 Nick darkapostrophe is now known as Dark_Apostrophe (
01:28:26zooookthere's LCD, IDE, USB and so on drivers in FRESC
01:29:09zooookCODEC too
01:29:28zooookshouldn't be too difficult to work out what's different based on the software
01:29:41LambdaCalculus37zooook: No, I don't think it would be that tough.
01:31:07 Quit lee-qid ("aufwiederbyebientotsayonara")
01:34:34LambdaCalculus37zooook: I'm going to try the zenfirm utility you posted up on the Dell DJ wiki page this weekend, if I get a chance.
01:35:45zooookahh, should give you a good idea what the firmware consists of
01:36:09 Quit Bam2550 (Connection timed out)
01:38:31 Join midkay [0] (n=midkay@rockbox/developer/midkay)
01:38:43zooookgetting late here, I'm gonna hop on off to bed
01:39:07 Quit zooook ()
01:39:39 Nick fxb is now known as fxb__ (
01:44:00 Quit Arathis ("Bye, bye")
01:44:07LambdaCalculus37Actually, I'm going to take off for a little while, then come back.
01:44:23LambdaCalculus37zooook: If I don't see you on IRC later on, just keep me posted on the forums.
01:44:55LambdaCalculus37See everyone later!
01:44:57 Quit LambdaCalculus37 ()
01:47:50 Part pixelma
01:49:30 Quit ThUnD3r|Gr33n (Read error: 101 (Network is unreachable))
01:50:04 Join ThUnD3r|Gr33n [0] (
01:52:02 Join RudMan [0] (
02:02:09KleviIs there any way to get the Sansac200 to play videos?
02:03:57 Join Bam2550 [0] (
02:06:21 Quit Klevi ("Leaving")
02:07:41Bam2550If i put in \.rockbox\rocks\games\ will it play when i click on it?
02:08:27 Quit spiorf (Remote closed the connection)
02:10:51Eberthold the button down and select Rockboy
02:11:16Eberti've found with more elaborate roms, the emulator won't play cuz of "memory allocation errors"
02:11:18 Join toffe82 [0] (n=chatzill@
02:11:41Eberthad 1 that played for a while, then wouldn't because of the memory problem
02:12:05Ebertmany do tho, try and see
02:12:41Ebertput the roms where ever you want, hold and select "open with"
02:13:27 Quit Nico_P (Remote closed the connection)
02:16:07 Quit Bam2550 (Read error: 104 (Connection reset by peer))
02:17:08 Join barrywardell [0] (
02:17:34SoapDoes anyone else find it ironic (wrong word?) that someone who claims to be a geek, and rallies against Word and PowerPoint files in their signature line would have the poor email manners to use an eleven fricking line email sig?
02:23:26 Join Bam2550 [0] (
02:23:38Bam2550Is rockbox 100 percent open source?
02:29:40 Join darkapostrophe [0] (
02:35:03 Quit barrywardell ()
02:43:24 Quit Dark_Apostrophe (Connection timed out)
02:43:24 Nick darkapostrophe is now known as Dark_Apostrophe (
02:44:03 Join eigma [0] (
02:49:49eigmaanyone know what/who generates
02:50:12 Quit jhulst (Read error: 104 (Connection reset by peer))
02:52:43squidinkThe rtc_write_datetime func always returns 1, but set_time in time_funcs.c expects it to return 0 if there was no error
02:54:27Bam2550What is RockBox written in?
02:55:41pSXAuthorfortran66 and visual apl
02:56:14pSXAuthorheh ;p
02:56:22pSXAuthorcome from mp3codec;
02:56:44 Quit Thundercloud (Remote closed the connection)
02:56:53pSXAuthor(actual syntax may not be correct... never written intercal for real ;)
02:57:38Bam2550Oh, because i wanted to try to learn a language and then maybe try to help RockBox
02:58:01eigmaRockbox is not an entry-level project
02:58:09Bam2550Yeah, doesnt sound like one
02:59:10pSXAuthordon't take this the wrong way, but the chances of you learning a programming language when you can't even be bothered to read the website (which would have answered your questions in about 5 seconds) are virtually 0...
02:59:42 Join Rob222241 [0] (
03:01:00 Quit hannesd (Read error: 104 (Connection reset by peer))
03:01:31 Join hannesd [0] (
03:02:43SoapBam2550: it has been over 24 hours now that you have been refered to the published documentation. There IS an expectation that you read it before asking questions. Scorche is not as nice as I am. scorche doesn't always take his medication like Soap does. scorche just started a new job. Don't make scorche angry. ;) Please read the documentation Rockbox has so kindly provided.
03:03:15*scorche pokes soap
03:03:22*Soap giggles
03:03:25Bam2550Oh, i didnt think it would tell me what i twas written in, sorz
03:03:42scorchewell, it uses gcc...what code does gcc compile?
03:04:03eigmaCool Code
03:04:21Bam2550Probably not the right place to ask, but are there any other things like RockBox (example: IpodLinux)?
03:04:34krazykitipodlinux isn't like rockbox.
03:04:41scorchethere are plenty of other projects, and no it isnt the right place to ask ;)
03:04:58scorche(other projects being open source firmwares for embedded devices)
03:06:13Bam2550I probably wouldn't change from RockBox, i was just curious to see what is out there.
03:06:37Soapwell IPL IS like Rockbox in that it is a firmware replacement for a popular DAP. ;) Other than that there is the Arch-Open project and Zune-Linux (!) and that is _just_ about it. Outside the strict realm of DAPs, there is always and the other Linux-PDA projects, but not a whole lot of this type of thing in general. I think you will find Rockbox is pretty much the top of the alternate-firmware-on-embeded-devices foodchain. Alpha!
03:07:10scorcheSoap: i dont think you can count imaginary firmwares...
03:07:43Bam2550Thanks Soap
03:07:55Bam2550You give good answers.
03:08:19Soapscorche: I resent that! Rockbox is not imaginary! You sure haven't taken your medication today!
03:08:45 Join webmind_ [0] (
03:09:13 Quit webmind (Read error: 104 (Connection reset by peer))
03:09:16scorchenoo...see he just really likes to type long sentences. The content of these lines really dont contain much, but it doesnt matter, as he is just out there for length. Although, all of what he now says are macros indeed. It seems to be a recent kick of his, and i am not really sure when he shall give it up. Although, his last line wasnt too long at all
03:09:43scorchethat really didnt turn out how i expected, but oh well =/
03:09:51***Saving seen data "./dancer.seen"
03:11:49*Soap finds it amazing how many macros he must have.
03:12:15*Soap challanges scorche to find a single sentence in the logs he has used twice.
03:12:20Bam2550How big is your .rockbox? <Poll>
03:12:34scorcheBam2550: this isnt a place for a poll...
03:12:46Bam2550to that one place!
03:13:01alienbiker99this is a place for very long sentences =)
03:13:04scorchewe dont much like useless polls ont he forums either...
03:15:42 Quit Rob2222 (Read error: 113 (No route to host))
03:30:40 Join RaZorbacK [0] (
03:31:04 Quit RaZorbacK (Client Quit)
03:33:23 Quit eigma ()
03:34:48 Join sarixe [0] (
03:40:09 Quit Bam2550 ()
03:40:59 Join Bam2550 [0] (
03:41:42 Quit sarixe ("Peace")
03:42:07 Quit My_Sic (Read error: 104 (Connection reset by peer))
03:43:26 Join My_Sic [0] (
03:47:00 Part squidink
04:39:11 Quit sin613 ("Leaving.")
04:45:20 Join hcs [0] (n=agashlin@rockbox/contributor/hcs)
04:53:03 Quit My_Sic ("Miranda IM! Smaller, Faster, Easier.")
04:57:07 Quit miepchen^schlaf (Read error: 110 (Connection timed out))
04:57:17 Join miepchen^schlaf [0] (
05:09:52***Saving seen data "./dancer.seen"
05:15:55 Join bb [0] (n=bb@unaffiliated/bb)
05:22:44 Quit RudMan (Remote closed the connection)
05:24:41 Join darkapostrophe [0] (
05:24:42 Join jhulst [0] (n=jhulst@unaffiliated/jhulst)
05:27:54 Quit bb_ (Read error: 113 (No route to host))
05:28:53 Quit midgey ()
05:41:10 Quit Dark_Apostrophe (Connection timed out)
05:41:10 Nick darkapostrophe is now known as Dark_Apostrophe (
06:06:55 Join Chronon [0] (
06:09:27 Join RudMan [0] (
06:13:59 Quit miepchen^schlaf ("Verlassend")
06:21:55 Join My_Sic [0] (
06:23:32 Quit RudMan (Remote closed the connection)
06:24:12 Quit My_Sic (Client Quit)
06:28:33 Join SkinInd95 [0] (
06:37:32 Quit SkinInd95 ("ChatZilla [Firefox]")
06:39:28 Quit hcs ("Leaving.")
06:39:51 Join hcs [0] (n=agashlin@rockbox/contributor/hcs)
06:44:28 Join homielowe [0] (
06:46:34homieloweHow do I add commands to terminal in mac( or unix in general), I;m trying to compile a build interminal and it will say " command not found, arm-elf-gcc" even though I have cross-compiler built on mac
06:48:03hcshomielowe: you'll have to add the path to the compiler to your path: export PATH=$PATH:/path/to/arm-elf/bin
06:48:23homieloweJust figure that out, thanks hcs,
06:49:39 Quit idnar (Nick collision from services.)
06:49:41 Join idnar_ [0] (i=mithrand@unaffiliated/idnar)
06:58:52 Quit XavierGr (Nick collision from services.)
06:58:55 Join XavierGr [0] (
07:02:35homieloweha, my first build of rockbox actually works ! \o/
07:04:25 Quit XavierGr ("One firmware to rule them all!")
07:09:56***Saving seen data "./dancer.seen"
07:12:58 Quit atsea- (Read error: 104 (Connection reset by peer))
07:22:05 Join kubiix [0] (
07:24:21 Quit crashev ("crashev has no reason")
07:24:36 Part toffe82
07:38:33 Join darkapostrophe [0] (
07:43:01 Join LinusN [0] (i=linus@rockbox/developer/LinusN)
07:45:56 Join darkapostrophe__ [0] (
07:55:08 Quit Dark_Apostrophe (Connection timed out)
07:55:08 Nick darkapostrophe__ is now known as Dark_Apostrophe (
08:02:39 Quit darkapostrophe (Connection timed out)
08:06:05 Join atsea- [0] (i=atsea-@gateway/tor/x-346640c675174658)
08:07:38 Join CaptainSquid [0] (
08:08:36 Join sitwon [0] (
08:08:48sitwonforums down?
08:12:54 Join Rob2222 [0] (
08:12:54ChrononIt usually gets sluggish at around this time
08:13:30 Join davina [0] (
08:26:15 Join grndslm [0] (
08:31:06 Quit Rob222241 (Read error: 110 (Connection timed out))
08:48:49 Join ender` [0] (
08:57:46 Quit Chronon ("Zzzz")
09:03:59 Join advcomp2019_ [0] (n=advcomp2@
09:04:33 Quit advcomp2019 (Nick collision from services.)
09:04:37 Nick advcomp2019_ is now known as advcomp2019 (n=advcomp2@
09:04:46 Join fm2 [0] (i=3ecebc53@gateway/web/cgi-irc/
09:04:49 Join B4gder [0] (
09:05:46 Quit fm2 (Client Quit)
09:06:33 Join haemmy [0] (n=stefan@
09:07:17*GodEater_ notes the Zend conf. is due to start on Oct. 8th
09:07:21GodEater_are you prepared B4gder ?
09:07:41GodEater_do you have your rebuttal ready ?
09:07:50B4gderit took me a few seconds to understand, I had forgot that
09:08:08 Quit RaRe ("Quit msgs should be longer.")
09:09:19 Join petur2 [0] (n=petur@rockbox/developer/petur)
09:09:24 Nick petur2 is now known as petur (n=petur@rockbox/developer/petur)
09:09:58***Saving seen data "./dancer.seen"
09:11:08 Join Zagor [0] (n=bjorn@rockbox/developer/Zagor)
09:11:16 Quit ender` (Read error: 104 (Connection reset by peer))
09:12:37 Join lee-qid [0] (
09:12:47 Join pondlife [0] (n=Steve@rockbox/developer/pondlife)
09:14:43 Join ender` [0] (
09:15:41B4gderso it seems the e200R OF boot problem is the same as some people see on the plain e200
09:15:49B4gderthe hc support commit basically
09:19:16 Join Roan [0] (
09:20:44Zagorjag gillar i7 fr att den r s liten
09:21:03*B4gder looks at Zagor
09:21:07GodEater_va ?
09:21:12Zagoroops, wrong channel :)
09:21:44 Nick Roan is now known as JustJohnny (
09:23:15peturoh nothing, just overloaded with real life :/
09:24:17 Join pixelma [0] (i=pixelma@rockbox/staff/pixelma)
09:25:46jhMikeSOk, now that things are picking up here. I need some checks on PP502x targets besides H10 20GB and e200 of a patch that adds some of the stuff from dual-core. I guess 8-9hrs trouble free on an H10 and not being able crash mpegplayer is an improvement.
09:26:54pixelmaZagor: I don't know i7 but did you see low_light's patch in the logs regarding the reboot on usb connect? It worked for me (also got freezes with yesterday's commit) - he wanted you to test too
09:26:55jhMikeSif there's any volunteers I'll make a patch and post ;)
09:28:01*petur looks for a free time slot to test on his H10 5GB and allocates a spot tomorrow afternoon :(
09:28:13Zagorpixelma: the last patch from him I see is the 03:50 last night, which hangs for me. is there another one?
09:28:26Zagoror you mean an uncommitted patch?
09:28:53 Part JustJohnny
09:28:59*jhMikeS supposes doing this dual core thing piece by piece is better
09:29:00pixelmayes uncommitted because he wanted some more testers first
09:29:45peturpixelma: why doesn't he post that on the tracker?
09:30:28peturinteresting filename there: B4gZdr24.html
09:30:31pixelmaI don't know but I think he just put that together quickly
09:30:35jhMikeSweren't H10's only running like 5hrs? I remember it draining the battery really fast before this.
09:30:39GodEater_ipod hasn't turned itself on all night
09:31:12pondlifejhMikeS: What runtime are you getting now?
09:31:24pondlifeAh, I read the log
09:31:30pondlifeGood stuff
09:31:39jhMikeSI ran it between 8 and 9 hrs not on a full charge. That was after running mpegplayer alot.
09:32:28hcsjhMikeS: I'd like to test it on my ipod photo
09:32:34jhMikeSIOW alot of backlight time beforehand.
09:33:00jhMikeShcs: ok, I'll do a diff and post it. un momentito...
09:36:56jhMikeSIncludes a modded mpegplayer since some were needed to stabilize it:">
09:37:08GodEater_jhMikeS: worth testing on 5.5G ipod ?
09:37:29jhMikeSGodEater_: anything available. would like to commit.
09:38:01GodEater_I'll apply here too then
09:38:10 Join barrywardell [0] (
09:38:47jhMikeSI just hope it all likes the cache setup which is just ripped from the firmwares
09:39:16 Quit jhulst (Remote closed the connection)
09:39:38jhMikeSJobs probably put some trap in there
09:39:58Zagorpixelma: yep, that patch works for me
09:40:07hcsuh, I've got a compile error here, let me see if it's just something stupid I did...
09:40:19LinusNjhMikeS: any particular reason you aren't using the patch racker?
09:40:25LinusNtracker even
09:40:33hcsjhMikeS: yeah, it's in a file you don't even touch
09:40:59jhMikeSLinusN: sometimes I do but I'm running on short notice
09:41:05LinusNi see
09:41:15pixelmaZagor: nice.
09:41:39Zagorhowever I'd like to remove the 2-second ata spindown...
09:41:39jhMikeSJust something I want some quick testing on and it's probably ok if it passes.
09:41:43B4gderjust call me B4gZdr
09:42:57 Join RaZorbacK [0] (
09:42:58jhMikeShcs: but could something I changed make the file not compile? You're using a clean SVN as a base?
09:43:01 Quit haemmy ()
09:43:09B4gder"Hynix Semiconductor Inc. [...] the world's second-biggest maker of memory chips, said it stopped supplying computer memory chips through the spot market"
09:43:26hcsjhMikeS: I'm checking out a clean version now to be sure, it was in metadata.c so more likely a patch I had in there
09:43:47jhMikeShcs: ah...I don't think anything touches on that at all
09:44:12 Quit lee-qid (Read error: 110 (Connection timed out))
09:44:35 Join spiorf [0] (
09:44:42Zagordo we have a define that tells if the target has hdd or flash?
09:45:02RaZorbacKis there any special status in the rb team for translators? I mean if I want to be an official translater, what should i have to do ?
09:45:04jhMikeSZagon: umm...not that I've seen anywhere
09:45:35B4gder#define HAVE_FLASH_STORAGE
09:45:44barrywardell|Rain|: does your bootloader patch fix the loading the OF problem on the Sansa?
09:46:07ZagorB4gder: ah, good
09:47:24spiorfRaZorbacK, ?
09:47:30jhMikeSoh, what about HAVE_MMC and HAVE_HOTSWAP? or you mean the main storage?
09:48:12ZagorI simply want to exclude the ata spindown call (and delay) on targets without hdd
09:48:20B4gderhave_mmc refers to actuall mmc/sd as in the ondios
09:48:56RaZorbacKspiorf: sory a typing mistake :)
09:50:17jhMikeShcs: is it running or did the iPod explode and destroy your keyboard? :)
09:51:32GodEater_jhMikeS: all working here so far
09:51:37GodEater_anything specific I need to test ?
09:51:49jhMikeSdid you try to crash mpegplayer yet?
09:52:05hcsjhMikeS: uh, I found that I have no video on the ipod, I'm encoding something now
09:52:08GodEater_will attempt to do so now
09:52:18 Part pondlife ("disconnected has pondlife")
09:52:21 Join pondlife [0] (n=Steve@rockbox/developer/pondlife)
09:52:30jhMikeSjust curious about that. basically it confirms that cache line overlap problem.
09:53:01GodEater_elephant's dream is playing now....
09:53:09 Quit RaZorbacK ("bye all")
09:53:35*GodEater_ still has no clue what it's about
09:53:44jhMikeSGodEater_: stop it and restart it a bunch times in a row. that's basically been a sure crash formula.
09:54:27GodEater_restart 1 - still working
09:54:59GodEater_think this is the first time I've used mpegplayer on the ipod
09:55:02GodEater_man it sucks :)
09:55:19barrywardellB4gder: did you see |Rain|'s patch for the e200 bootloader?
09:55:24jhMikeSH10 plays video rather nicely...quite suprising given is doesn't have optimized rendering
09:55:39GodEater_bet it's not a 320x240 screen
09:55:55GodEater_I've restarted it about 5 times now
09:55:59GodEater_it's still working
09:56:01jhMikeSit's 1/2 that
09:56:14jhMikeSbut all the videos I have are way oversized for it
09:56:49GodEater_do you want me to keep going
09:56:50jhMikeSGodEater_: I think that's a good indicator it's in good standing. Thanks.
09:57:04GodEater_no problemo :)
09:57:16*pondlife does the commit dance
09:57:25GodEater_youtube pls
09:59:40hcsjhMikeS: video freezes, audio continues
10:00:02hcsif I bring up the menu and dismiss it it comes back
10:00:02GodEater_party pooper
10:01:26hcscompletely out of sync then, though
10:03:11B4gderbarrywardell: no I didn't
10:03:14jhMikeSif you bring the menu up it comes back?
10:03:28jhMikeSthat's rather weird but not a crash I think
10:03:44hcsyeah, I haven't tried this video with it unpatched, though
10:03:47barrywardellB4gder: around 22:48 last night on IRC
10:03:55 Join bluebrother [0] (i=cAKSFXgW@rockbox/staff/bluebrother)
10:04:07Zagorboo for FOR_NB_SCREENS
10:04:14jhMikeShcs: oh, better use the version in that patch then.
10:04:15barrywardellI assume 'fixed' means fixed the boot OF problem, but not sure
10:04:22 Quit davina (Remote closed the connection)
10:04:30hcsI just now encoded it with mencoder using the recommended settings so I don't think it is the video, but it may be...
10:04:32Zagordoes it actually improve the code size on any target? not on arm anyway
10:05:11jhMikeSZagor: you're asking me about my patch or ??
10:05:13GodEater_hcs: tried elephants dream ?
10:05:27hcsGodEater_: I'll go grab that.
10:05:32*GodEater_ uses this as his defacto mpegplayer test
10:05:34barrywardellB4gder: hmm. the patch's server is offline. Here's the version I downloaded:
10:05:36B4gderbarrywardell: yeah, I would assume that too...
10:05:37ZagorjhMikeS: no, I'm asking about the purpose of the FOR_NB_SCREENS macro.
10:05:48 Join ring [0] (i=3ecebc53@gateway/web/cgi-irc/
10:06:14barrywardellIt works on my e200 (but I've never had the problems with the bootloader other people did)
10:06:29jhMikeSI guess it's the number of screens. Shouldn't it be optimized away if no remote exists?
10:06:41*GodEater_ gets the impression that non-rockbox staffers abuse their daps in interesting ways
10:06:48GodEater_otherwise we'd all get the same problems everyone else gets
10:07:17ZagorjhMikeS: yeah, but it gets optimised away even without the macro. so why add a macro?
10:07:21B4gderbarrywardell: I'll ask the e200R guy in the forum to try it out
10:07:25 Join davina [0] (
10:07:25 Quit ring (Client Quit)
10:07:34 Join pluck [0] (i=3ecebc53@gateway/web/cgi-irc/
10:07:34 Quit CaptainSquid ("Miranda IM!")
10:07:56jhMikeSZagor: To hide tedious for(int i=1; i < NB_SCREENS; i++) stuff?
10:08:03hcsGodEater, jhMikeS: elephant's dream seems to be running ok
10:08:04Zagorwe don't hide code
10:08:14Zagorit's even in the rules
10:08:53jhMikeSZagor: Does that refer specifically to macros or what else?
10:09:06Zagormacros mainly
10:09:49pluckAny shell experts here? I need the following: I have a file which should be processed twice with sed. The results of the transformations should then be merged (first goes the first result, then appended the second). Can this be done without creating a temporary file?
10:09:51jhMikeSSo hiding complexity in low-level code from higher level code is ok, right? :)
10:10:21LinusNpluck: use a pipe
10:10:41B4gderpluck: sed > file && sed >> file ?
10:10:48barrywardellB4gder: great. It looks reasonable - there shouldn't be a thread created in the bootloader
10:10:48ZagorjhMikeS: no, ideally we'd like everything in a single function. with lots of gotos. :-P
10:10:56pluckAny shell experts here? I need the following: I have a file which should be processed twice with sed. The results of the transformations should then be merged (first goes the first result, then appended the second). Can this be done without creating a temporary file?
10:10:57B4gderbarrywardell: I agree
10:11:10LinusNis there an echo in here? ;-)
10:11:21jhMikeSlol...every possible code path ironed flat
10:12:20jhMikeShcs: ok, any other observed issues or not?
10:12:28pluckB4gder: yes, that's what I thought of. But is this possible without the temp file, i.e. with the pipes only?
10:12:32hcsjhMikeS: nothing seems awry yet
10:12:42pluckSorry for the double post
10:12:55B4gderpluck: sure, just send the sed output to stdout
10:12:58B4gderor wherever
10:13:33 Quit spiorf ("Read error: 110 (Connection timed out)")
10:13:44 Join spiorf [0] (
10:13:50jhMikeSH10 has an identical reset sequence to e200 and c200 but a different DEV_RS value. I'll bet the main reset exists for the other PP targets too and should be used.
10:14:52pluckB4gder: use "-" instead of file?
10:15:52jhMikeSthe patch also initializes CPU/COP in serial fashion up to the end of init_threads which seems to help rockbox boot from a USB pull every time.
10:16:41B4gderpluck: sed spits out to stdout by default. but really, this is getting way off-topic now
10:17:06jhMikeSthere's no main COP thread either
10:17:51 Quit barrywardell ()
10:18:34 Quit TMM (Read error: 110 (Connection timed out))
10:19:32 Join BigBambi [0] (n=Alex@rockbox/staff/BigBambi)
10:20:06 Quit pluck ("CGI:IRC (Ping timeout)")
10:20:20 Nick Gekz[sleep] is now known as Gekz (
10:20:25jhMikeSoh, we get 1k of IRAM back too :)
10:23:11 Quit homielowe (Read error: 110 (Connection timed out))
10:25:51pixelmaZagor: I have not much of a clue but think I've been told or read somwhere that the storage is hooked up differently on the Ondios, ifp so maybe the geneal HAVE_FLASH_STORAGE wouldn't be the right one (example lines 103/104 in config-ondiofm.h)
10:26:50 Join obo [0] (n=obo@rockbox/developer/obo)
10:27:08Zagorpixelma: it suits fine in this particular case, anyway
10:27:33Zagorsurely the ondio does not expect a ata_sleepnow() call before resetting...
10:28:21jhMikeShcs: still running ok?
10:28:30 Join kubiixaka [0] (
10:28:57hcsjhMikeS: yeah, finished elephant's dream, trying various audio codecs
10:29:05 Quit Ebert ()
10:30:29jhMikeShcs: was there any particular thing that _would_ cause a problem before and doesn't now?
10:31:07hcsnot that I recall
10:31:20hcsother than irritatingly short battery life
10:32:32jhMikeSsince for some reason H10 seems better off, any change there? It seem to drop off really fast for me before.
10:33:34hcsnot sure, it's been a while since I've used it, but we'll see how long it lasts. is there anything in particular this is supposed to fix or just a rearrangement
10:34:09jhMikeSprefetch aborts and such exceptions
10:34:52jhMikeSso it's an attempt to fix caching issues which were the plague on dual-core
10:35:36GodEater_when given a usbstick which contains one partition marked as type "linux" - any hints on determining what sort of filesystem resides there ?
10:35:55LinusNprobably ext2
10:36:15GodEater_sadly it's not
10:36:30GodEater_I think it might be empty actually
10:36:34GodEater_it's all zeros
10:36:35Zagorswap? ;)
10:36:42GodEater_that would be type linux-swap
10:36:45GodEater_not linux :)
10:37:16GodEater_82 not 83 iirc
10:37:39GodEater_yeah it's completely empty
10:37:42GodEater_never mind
10:38:45*GodEater_ wonders what the point was
10:39:13 Quit kubiixaka ("Miranda IM! Smaller, Faster, Easier.")
10:40:16 Join kubiixaka [0] (
10:40:54 Quit kubiixaka (Client Quit)
10:43:22 Quit kubiix (Read error: 110 (Connection timed out))
10:46:18 Join barrywardell [0] (
10:52:58pixelmaZagor: just noticed that low_light's patch you just committed caused a warning (hard to spot)
10:54:35Zagorah, lookie
10:57:18pixelmahe let me try if the reboot still works for me if commented out and it did for me. But he said if it's not there, the reboot wouldn't work for him (hangs too)
10:57:39pixelmathis line, I meant
10:58:54ZagorI'll rather just add the #include line
10:59:35pixelmayes, was just fyi ;)
11:10:01***Saving seen data "./dancer.seen"
11:16:43 Join CaptainSquid [0] (
11:16:51jhMikeShcs: still going? I'd say stuff is safe to commit then.
11:17:44hcsjhMikeS: yes indeed, running fine
11:18:10GodEater_yay - commit!
11:18:20jhMikeSokidoke :)
11:20:13pondlifeHmm, the raw IRC text log for today has an odd format:
11:21:03 Quit barrywardell ()
11:21:18GodEater_what's odd about it ?
11:21:25 Join przemhb [0] (i=5035dcaa@gateway/web/cgi-irc/
11:21:33 Quit przemhb (Client Quit)
11:22:07 Join przemhb [0] (i=5035dcaa@gateway/web/cgi-irc/
11:24:10 Join colin_ [0] (
11:25:47 Quit pondlife ("disconnected has pondlife")
11:26:07 Join darkapostrophe [0] (
11:27:59 Join pondlife [0] (n=Steve@rockbox/developer/pondlife)
11:28:28 Quit Isolinear ()
11:29:43pondlifejhMikeS: Did you see - asking for an opinion on the mpegplayer start time patch...?
11:30:29 Nick fxb__ is now known as fxb (
11:36:26pondlifeGodEater:Normally, IE displays it in a courier font, with line breaks.. Today's is in a non-proportional font with no line breaks or nicks.
11:37:04pondlifeNot a biggie, as the non-raw page is fine
11:38:17jhMikeSpondlife: no, hadn't seen it yet.
11:39:13pondlifeThought you might be a useful pair of eyes, I've not looked myself. But it seems useful if it's not hacky.
11:39:27jhMikeSwell, there's no doubt I'm about to break it since there will be two allocators in it now and it mods that code.
11:42:22GodEater_pondlife: One word. Firefox. :)
11:42:29 Quit Dark_Apostrophe (Read error: 110 (Connection timed out))
11:42:30 Quit ThUnD3r|Gr33n (Read error: 110 (Connection timed out))
11:42:30 Nick darkapostrophe is now known as Dark_Apostrophe (
11:43:07pondlifeGodEater: Two words: Work Computer :(
11:43:35pondlifeAnd no, I'm not sneaking a browser in on a USB stick
11:43:56LinusNif a train station is where the train stops, then what is a workstation? :-P
11:44:23pondlifeAnd a brickstation?? :)
11:44:55hcsI've been thinking of a breakout clone with rockbox targets as the bricks
11:45:00bluebrotherwhere the bricks stop? But what is a brick?
11:45:19LinusNgetting philosphical here, aren't we? ;-)
11:45:24pondlifePhilosophical indeed
11:45:31*LinusN can't spell
11:45:40*pondlife uses Pidgin now
11:45:56pondlifeThat took some convincing too.
11:46:33pondlife"Yes, I need a multi-network IM client so I can be contacted by X, Y & Z."
11:46:48pondlifeSo I do run some OSS at work.
11:47:16pondlifeBut don't tell them..
11:49:22jhMikeSpondlife: another thing is it uses that hack-up stream messaging stuff to a great extent that I just want to get rid of badly. suppose I should comment there. :P
11:49:50pondlifeMaybe suggest what should replace it too.
11:51:29jhMikeSI wrote it simply because kernel object aren't useable on dual core yet it's it's strict and limited protocol makes it safe to use with anychronous threads but it's a hack in my book.
11:52:48 Join Isolinear [0] (
11:57:42jhMikeSWhat should replace it simply use of kernel message queues
12:06:27 Join daro_dehner [0] (n=dfsff@
12:06:53 Nick daro_dehner is now known as SSnake (n=dfsff@
12:07:49SSnakeanyone online?
12:07:57bluebrotheronly 133 people ...
12:08:22SSnakei would report a problem (possible bug dunno)
12:08:31SSnakecan i?
12:08:44SSnakeok tahanks :>
12:09:25SSnakeI Rockboxed my iriver h300 since 10 months
12:09:46SSnakei used to compile bootloader by myself
12:09:58SSnakei know the current bootloader is a "beta"
12:10:00 Nick idnar_ is now known as idnar (i=mithrand@unaffiliated/idnar)
12:10:01 Join crazyshit [0] (
12:10:17preglowGodEater_: yo, any new wakeups with linuxstbs patch?
12:10:21bluebrotherthe svn bootloader is known to have issues.
12:10:32SSnakebut it has a lot of improvings compared to the one on the wiki
12:10:42SSnakeyep i know it's not perfect
12:10:49crazyshitrockbox allows alarm clock in e200 ?
12:10:56crazyshitor will allow ?
12:11:00bluebrotherand that's the reason why the bootloader gets released (and you aren't supposed to build it yourself unless you really know what you're doing)
12:11:05SSnakehowever i always keep it update until august
12:11:15SSnakeyeah i know
12:11:32SSnakei'm able to solder a bad case i'll use the gdb interface
12:11:33 Join Genre9mp3 [0] (
12:11:51SSnakethe bootloader was ok until july ithink
12:12:20SSnakeit seemed to broke when the new usb stack started on august
12:12:34bluebrotherthe usb stack does not affect the h300 series.
12:12:40SSnakeactually i''m not able to start the rockbox.iriver loader anymore
12:12:54SSnakeit's able to start the OF
12:13:05bluebrotherwell, it's know that the svn bootloader fails to load Rockbox under some circumstances.
12:13:10SSnakewell i dunno if it's related to he USB stack or not
12:13:31bluebrotherit had those issues back when devcon took place which was earlier.
12:13:48SSnakeoh i remeber devcon (i saw photos)
12:14:23SSnakeso nothing i can do for now, just reflash the old bootloader?
12:14:43GodEater_preglow: no - ipod has been very sleepy since the patch
12:14:46bluebrotherunless you want to debug the bootloader, in which case you most likely will need a BDM, no
12:14:53crazyshitno one know ?
12:15:13 Join Davo_Dinkum [0] (
12:15:32SSnakewell so the bug is already known
12:15:50SSnakeeven if it's worked until july
12:16:20SSnakeah just a note:
12:16:36 Part crazyshit
12:16:43SSnakethe battery level detection seems to be improved while charging
12:17:15SSnakeon july it reported wrong level when charging
12:17:15bluebrotherLinusN is the one who works on the bootloader. He might know more details. But afaik he's quite busy these days.
12:17:26LinusNi am indeed
12:18:09SSnakeoh ok
12:18:16 Join crazyshit [0] (
12:18:18SSnakewell thanks for answering me
12:18:28bluebrotheryou're welcome
12:18:32SSnakei'll go back with the olf BL
12:18:39pondlifeI hope to update my H340 bootloader and investigate one day soon.
12:19:10pondlifeI need to somehow backup my current one though - I'm very happy with it, but don't know which source it came from :(
12:19:21pondlifeIt's a bespoke one IIRC
12:19:46SSnakeoh and just to let you know:
12:20:07SSnakethe korean 1.31 firmware works good with rockbox
12:20:26SSnakesomeone should update the rockbox firmware patcher
12:20:46pondlifeSSnake: The non-booting issue is already reported on FlySpray somewhere, might be good for you to find that and watch it.
12:20:53jhMikeSok, committed that sans mpegplayer changes. now the fun should be just beginning. :P
12:21:18pondlifeWeren't the mpegplayer changes required?
12:21:19SSnakei'll search for thanks
12:21:52jhMikeSpondlife: various plugin commits to follow
12:22:03pondlifeBut the interim build should work?
12:22:25pondlifeNot that I've got a PP target...
12:22:51 Join Nico_P [0] (n=nicolas@rockbox/developer/NicoP)
12:24:32bluebrotheririver boo? ;-)
12:24:36SSnakeFS #7533
12:24:37jhMikeSnow that thread functions may return normally, I guess I'll do the mpegplayer changes and pull all that remove_thread(NULL) stuff from plugins
12:25:39jhMikeSI guess I forgot to match up the sim with that new feature. :p
12:25:52jhMikeSor, wait, I think it already works there. hehe
12:26:06pondlifeI thought you already made the sim use the same thread code, no?
12:26:15pondlifeI know I've long wanted it to.
12:26:36jhMikeSit does, but some things were in advance of it being possible on target
12:26:36crazyshitrockbox allows alarm clock in sansa?
12:27:34jhMikeSpondlife: If this dual core stuff got committed in one lump, the recording commit would pale in size and browser space. :)
12:28:53 Quit ze (Read error: 110 (Connection timed out))
12:28:55 Join MethoS- [0] (
12:29:12preglowGodEater_: but it also doesn't wake up if you try to use the alarm, yeah?
12:30:31GodEater_no idea - haven't tried
12:30:39GodEater_want me to do that too ?
12:30:41 Join ze [0] (
12:30:58preglowGodEater_: please do
12:31:05*GodEater_ does
12:31:22preglowjhMikeS: any differences we should see from this commit?
12:31:41 Join senab [0] (
12:32:05*GodEater_ sets the alarm for 5 minutes from now
12:32:14 Join Ebert [0] (
12:32:15jhMikeSpreglow: well, when I get it all done, it should kill off all those data aborts
12:32:29hcsjhMikeS: battery life is seeming a bit better now, at 61% after 2.5 hours or so of continuous use, looks like it'll outdo the 5 or so I usually get
12:32:32preglowsure, meant just this commit
12:33:08jhMikeSpreglow: well it should be true now except for the red bootloaders
12:33:25preglowred, btw
12:33:34preglowdamn, even saved some space, nice going :P
12:33:43jhMikeSand mpegplayer because the cache functions aren't inline now
12:33:50 Nick parafin|away is now known as parafin (
12:33:54jhMikeSsaves 1K of IRAM too
12:34:13 Join foam [0] (i=c27f0814@gateway/web/cgi-irc/
12:34:33jhMikeSIs there any IRAM use restriction on the bootloader or is ICODE_ATTR ok there?
12:34:52foamHe-he, with the new settings for line selector color (lss, lse_color), will we also get lsd_color?
12:34:53SSnakebye guys
12:35:03 Quit SSnake ()
12:35:30preglowjhMikeS: does the bootloader even use iram?
12:35:49preglowjhMikeS: i think those macros just evaluate to nothing there
12:35:54senabfoam: lsd_color?
12:36:06 Quit crazyshit ()
12:36:11preglowlsd_color = many colors
12:36:11GodEater_poor joke
12:37:15foampreglow: right!
12:37:16jhMikeSpreglow: now they're functions and start_thread calls it but we don't do threads there.
12:37:58senabfoam: like some sort of spectrum?
12:38:38foamYes, dynamically changing color in the background
12:39:29foamThis is getting OT though so I'd better shut up
12:39:41senabi know what LSD is lol, just not sure where it comes into rockbox
12:39:58GodEater_then you've sort of missed the joke
12:40:06GodEater_which was a bit poor in the first place
12:40:11 Quit colin_ ("")
12:40:17senabahhhh, i've only just got up lol
12:40:33Nico_PjhMikeS: "Thread function return implemented for all targets.": this means thread function are now allowed to return, right ?
12:40:34*jhMikeS isn't sure why NUM_CORES > 1 in the bootloader anyway.
12:40:41jhMikeSNico_P: yes
12:40:59Nico_PjhMikeS: is it better that they do ?
12:41:26preglowGodEater_: didn't wake?
12:41:30jhMikeSIn fact, I'd say it's preferred. It's also cleaner and more in line with other kernels.
12:41:31IsolinearRockbox + LSD = Plasma plugin
12:41:39preglowIsolinear: good point
12:41:43GodEater_preglow: no it didn't
12:41:47Nico_PjhMikeS: ok, good to know
12:41:57preglowGodEater_: ok, i'll figure out a way to detect if i set the alarm and commit the fix, then
12:42:07 Quit foam ("CGI:IRC")
12:42:10IsolinearSpeaking of plugins...
12:42:10GodEater_preglow: ok :)
12:42:12Nico_PjhMikeS: maybe you'd want to update when you have some time
12:42:33jhMikeSoh, why not. but _after_ the red is fixed :)
12:42:42IsolinearIs there any reason an "unsupported" iPod 5G build would have all plugins in one directory?
12:42:50B4gderwoo, 1400 points!
12:43:03IsolinearAs opposed to organized by type, such as viewers, games, etc...?
12:43:18B4gderIsolinear: because it is old?
12:43:27 Join nls_ [0] (
12:43:27Nico_Pdoes anyone know what "Improved magic by Mark Arigo" means ?
12:43:33IsolinearIt's the second most recent EvilG build..
12:43:42B4gderNico_P: check the patch, it changes some magic numbers
12:43:46Nico_PIsolinear: still, it's old
12:43:56IsolinearI tried organizing the plugins into folders but when I booted the iPod, they wouldn't show up..
12:43:56Nico_PB4gder: but what do these do ?
12:43:59B4gderIsolinear: still, that's how things were done in the past
12:44:11senabIsolinear, the second most recent was from July
12:44:14B4gderNico_P: they're just magic numbers that the firmware/OF uses
12:44:17pixelmaNico_P: help rebooting to the OF on the c200s
12:44:18Nico_Pok :)
12:44:22jhMikeSNico_P: I'll actually discourage any other method of stopping a thread. Though remove_thread(NULL) is technically safe, terminating other threads without knowing their state is not and that's true in any system I've worked in.
12:44:41IsolinearIs there a reason moving them into folders makes them invisible?
12:44:50IsolinearI guess that's my actual question... lol
12:44:54bluebrotheryes, because folders are not suppoed in that view
12:45:09pondlifeBlimey the H300 OF takes a long time to boot.
12:45:21IsolinearGotcha. :)
12:45:29bluebrotherit was simply not supported before, your customized build is outdated and we don't support customized builds here anyway.
12:45:48 Nick nls_ is now known as n1s (
12:45:49bluebrotherand personally I disencourage using custom builds.
12:46:05senabjhMikeS: do i need to update by bootloader due to your commit?
12:46:08IsolinearI know you don't support custom builds, I was just wondering if the fact that it was custom had something to do with the invisible folders.
12:46:16jhMikeSsenab: no
12:46:23*pondlife flashes his H300
12:46:23Nico_PjhMikeS: how ready is the rest of the dual core stuff ?
12:46:26bluebrothercustom builds are usually outdated.
12:46:43bluebrotherwhich makes supporting them even more painful.
12:46:51IsolinearWell yes, but I don't update my build too often anyway.
12:46:53 Quit Bam2550 ()
12:46:59 Quit CaptainSquid (Remote closed the connection)
12:47:00B4gderIsolinear: the plugin categories were introduced August 6
12:47:16senabi give up making custom builds because they become such a huge pain in the backside
12:47:22IsolinearSo I can live with being a few steps back and having sloppy patches that make things fun... :P
12:47:29jhMikeShmmm, I don't even need any COP init code in the BL, so that should basically kill the red there.
12:47:46bluebrotherRockbox is enough fun even without fancy gui patches for me ...
12:47:59senab...and most of them try to pack in as many patches as possible
12:48:17IsolinearI was using the Senab build for a long time...
12:48:36IsolinearThen you stopped updating as often so I tried EvilG.
12:48:45pondlifeYep, H300 bootloader's definitely broken
12:48:50senabi'm using a Sansa now ;)
12:49:07IsolinearYou like it more than the iPod?
12:50:21senabyep. it's smaller, has longer battery life and is generally better imo
12:51:07senabjust looked at the evilG build and he uses 24 patches :s
12:51:09IsolinearLonger battery with Rockbox, OF, or both?
12:51:12preglowsome docs on how the pp regs behave would also be nice
12:51:21preglowjhMikeS: didn't you discover something about the gpio reg set?
12:51:57 Join Entasis [0] (
12:51:58senabisolinear, both ;) i get around 4-5 days on my sansa with normal use, i typically had to charge my 5G up every night
12:52:38IsolinearBut isn't Sansa flash memory?
12:52:58IsolinearWell there ya go... lol
12:53:10IsolinearI need my 60GB... :P
12:53:25senabhaha, 8GB does me fine :D
12:53:59IsolinearHow do the solid state iPod rate with battery life compared to OF, out of curiosity?
12:54:36preglownot better, i'll tell you that
12:54:59preglowmy nano sucks down battery like a bloody fool
12:55:18pondlifeA hungry stupid vampire?
12:55:45jhMikeSpreglow: what do you mean? just about that there's more than one GPIO mask bit. also the registers repeat in memory every 512 bytes except for GPIOL.
12:55:52IsolinearI miss being able to pop some spare AAs into my Archos Studio 20... ;)
12:56:18jhMikeSpreglow: I assume you meant on PP, correct?
12:56:27preglowjhMikeS: aye
12:57:00preglowjhMikeS: is there some semblance of documentation on this? it'd be truly golden to have notes on it somewhere
12:57:16peturpondlife: talk to XavierGr about the broken h100/300 bootloader, he debugged it already a bit
12:57:22jhMikeSIn the pp5024.h file :)
12:57:36pondlifeYes, I saw... it's the ata_init() that hangs.
12:58:18preglowjhMikeS: pp5020.h, you mean?
12:58:28jhMikeSno, pp5024.h. there is one.
12:58:42preglowi know, but it doesn't have much info
12:59:13jhMikeSnot much more to say really. I don't even know if any other one has multiple bits.
12:59:28jhMikeSit might be unique to that
12:59:38 Part LinusN
13:00:25*preglow years for pp docs :/
13:00:28preglowyearns too
13:01:12jhMikeSI think I yearn 10x more since it's giving me such nightmare to do something that would take but a couple days otherwise.
13:01:21preglowany ipod video people here who can test the yuv_blit speedup patch?
13:01:51preglowi hate having such obvious targets for commital hanging around in the tracker
13:01:58jhMikeSif it's not asm, don't bother with it ;)
13:02:20preglow"We only commit patches that are 90% asm, sorry :/"
13:02:34jhMikeSI'm sure the H10 will see a revision of the gigabeats code for port-based lcd soon.
13:03:09preglowstraight machine code
13:03:18GodEater_not quite
13:03:19IsolinearAs opposed to...
13:03:26jhMikeSlike straight vodka
13:03:52GodEater_as supposed to C source
13:04:01Zagorplease clean it up a bit before committing. split the 'if' lines etc.
13:04:21IsolinearjhMikeS: A valiant attempt to put it into terms I understand, however I don't drink... :P
13:05:36preglownot even straight vodka?!
13:05:42*GodEater_ doesn't trust people that don't drink
13:05:46GodEater_they're all up to something
13:05:57Davo_DinkumNot drinking? :p
13:06:02senabasm when you manipulate bits manually using operators, C is a "higher" level language which takes care of most of the lw level stuff for u
13:06:02IsolinearI'm crazy about root beer though.
13:06:38IsolinearRoot beer's my wine.. I'm a connoisseur.
13:07:03*petur looks away
13:07:06*hcs is also a root beer fan
13:07:26senabi've never had root beer, is it alcoholic?
13:07:28Isolinearsenab: thx
13:07:44GodEater_it tastes like germolene smells
13:07:49GodEater_i.e. gross
13:07:55jhMikeSwell, C is like a screwdriver. you can tast the vodka but you can't light in on fire.
13:07:56 Join Thundercloud [0] (
13:07:58IsolinearDepends on the brand.
13:08:00hcsGodEater_: there is a lot of variation
13:08:17GodEater_you expect me to try other ones with an experience like that under my belt ?
13:08:26pondlifeIMHO, it's an abominable misuse of the word "beer"
13:08:31IsolinearWhat the heck did you drink?
13:08:33GodEater_pondlife: too bloody right
13:08:52bluebrotherjhMikeS: already noticed that there is still red?
13:08:59advcomp2019i know it is offtopic but i like root beer too
13:09:02GodEater_Isolinear: something labelled as root beer. I forget the brand. But I'm never going near anything else called that again
13:09:04pondlife"root drink" would be a better name...
13:09:12senabthe red is to do with mpegplayer
13:09:40bluebrotherah, haven't spotted that.
13:09:49senabsomething to do with 'flush_icache'
13:10:05***Saving seen data "./dancer.seen"
13:10:28jhMikeSbluebrother: yes, next one should take care of it
13:11:19GodEater_preglow: elephants dream is playing back nicely
13:11:30*bluebrother is unconcentrated :(
13:11:32preglowGodEater_: more fps?
13:11:40GodEater_no diea
13:12:14GodEater_looks like it
13:12:18GodEater_but I didn't measure before hand
13:12:48preglowdoesn't matter anyway, i'll just clean it up a bit and commit it
13:13:19GodEater_it's hovering around the 15fps mark for the "widescreen" version of elephants dream
13:13:42*GodEater_ has to go out
13:13:48preglowthanks anyway
13:14:01*jhMikeS didn't really expect green deltas for this. :\
13:14:32preglowwhat, and that makes you fret? :P
13:15:01senabwhat are green deltas?
13:15:14preglowmeans a commit cut down on binary size
13:15:47 Join moos [0] (
13:16:04senabahhhh! sounds like some covert SAS squadron lol
13:16:04jhMikeSpreglow: well, there's more thread startup code so it's really a mystery
13:16:32preglowsenab: well, we do try to sound as cool as possible
13:16:51IsolinearGreen beret + Delta force.
13:17:13pondlifeDon't forget the evil Red Deltas
13:17:17jhMikeSgreen deltas aren't very experienced
13:18:10senabhaha, what have i started lol
13:18:47jhMikeSbut they do shoot down the commie red deltas :p
13:18:55preglowthe neutral blue zeroes make me sick
13:18:58preglowwe must eradicate them too
13:18:59IsolinearAnyone care to explain how the term "dual core" relates to my iPod 5G
13:19:05preglowIsolinear: well, it has it
13:19:08senabjhMikeS: no more reds :D
13:19:26B4gderIsolinear: two cores. one, two ;-)
13:19:30senabthe portalplayer cpu in the iPod has 2 t
13:19:34IsolinearAre we taking advantage of this fact yet?
13:19:39jhMikeSjust those yellow ones that cut and run
13:19:46preglowyou're watching jhMikeS taking care of it right now
13:19:49ZagorjhMikeS: what targets have you tried this on? I'm a bit surprised it affects the sh target sizes so much
13:19:57IsolinearThat was my guess.. :)
13:20:05IsolinearI was reading is commits.. lol
13:20:05jhMikeSI tried SH, coldfire and ARM
13:20:14 Quit iamben (Read error: 104 (Connection reset by peer))
13:20:17preglowjhMikeS: you have an sh target?
13:20:20jhMikeSI have a player
13:20:27Zagorok, good
13:21:03IsolinearThis sounds familliar...
13:21:13jhMikeSsomone offered one just for shipping costs so I didn't pass it up
13:21:22IsolinearBut I've kinda been out of the scene a while..
13:21:53IsolinearI think COP was barely being used last I read months back...
13:21:56B4gderwe have a scene here
13:22:00B4gderfor the commit dances
13:22:04jhMikeSsomething must be wrong with the build table
13:22:11jhMikeSor the delta table I mean
13:23:30jhMikeSthe commit must have created some space-time inversion that suck bytes away
13:24:06B4gderI don't see any errors, click the 'log' link and compare the reported sizes
13:25:04IsolinearI was about to take bluebrother's advice and switch to the lastest SVN build... Is now a bad time for that?
13:25:04jhMikeSWell, I'm thinking about what code changes could possibly have made that much difference and can't think of any.
13:26:01B4gderif one really wanted, i guess comparing maps would tell exactly
13:29:27 Quit Thundercloud (Remote closed the connection)
13:30:08n1smaybe we are running out of bytes, and don't have enough left to build all targets? :-)
13:30:09ZagorB4gder: do you keep the map files of these builds?
13:30:15 Join desowin [0] (
13:30:19jhMikeSI suppose not inlining the cache functions on PP would save some.
13:30:28B4gderyou'd need to svn to that rev and rebuild
13:30:46B4gderthere's a buildtarget that creates a zip with the maps though
13:32:37 Join Thundercloud [0] (
13:33:12preglowjhMikeS: i assume you've disassembled some cache control functions by now? i've never been quite sure the invalidate functions was quite correct
13:33:30jhMikeSpreglow: yup and there's quite a number of them too
13:34:14preglowreally, now
13:34:16jhMikeSI actually disassembled the entire H10 firmware and marked most of it off vast as it is
13:34:31 Join CaptainSquid [0] (
13:34:59senabi've only just realised i can do a "make 7zip"
13:35:17preglowis the ipod video screen 565 too?
13:35:35jhMikeSI know what kernel it uses and how much microsoft code it contains :) Lot's of 0x80070057 return values which are COM objects return E_INVALIDARG
13:35:53B4gdersenab: been there since dec 2005...
13:35:54preglowH10 uses microsoft code???
13:36:09B4gderprobably drm things
13:36:12*senab slaps head. I was unzipping the zip, then 7zip'ing it before
13:36:16preglowand they bring com into it...
13:36:24*preglow totters off to vomit
13:36:29B4gderlots of player firmwares include (c) microsoft strings
13:36:37*petur would be surprised if a DAP used COM components
13:37:10jhMikeSpreglow: big time
13:37:19 Quit spiorf (Read error: 110 (Connection timed out))
13:37:25B4gdersenab: wow, you could have done lots on those ~218 seconds you've lost of your life! ;-)
13:37:27jhMikeSIt uses RTXC kernel v2.33
13:37:27peturmaybe they used a similar encoding of return values, but certainly NOT COM
13:38:10 Join spiorf [0] (
13:38:25jhMikeSit's COM, com com. it's not the only stuff. And the context of the return values is exactly correct. There's even a URL in there.
13:38:49B4gderRTXC seems to be here =>
13:39:24 Join RaRe [0] (
13:39:42jhMikeSyup, that's the company
13:40:35peturjhMikeS: and next you're going to tell me they are actually using DCOM in that firmware? What's next, a complete W2K kernel?
13:42:01jhMikeSpetur: They appear to be com objects, yes along with associates vtables.
13:42:07B4gderseems the sansa firmware is also rtxc
13:42:49*jhMikeS guesses noone ever did the entire lot before
13:42:53B4gder"RTXC v3.2fpp for ARM and Thumb - ARM ADS 1.0 Jul-08-00 Key: 24104"
13:43:33jhMikeSThe H10 and Sansa firmware are basically identical except for the company polish.
13:43:34B4gderand googling for it shows people discussing it here before... ;-)
13:43:41B4gdercircle complete
13:43:44preglowhere we are reinventing the wheel!!!
13:43:47B4gderlap #2
13:43:49preglowlet's use some of that fund cash for lincesning
13:43:54preglowlicensing too, please
13:45:10jhMikeSat least it confirms the existence of certain things since all the kernel objects have names
13:45:28B4gder"Porting RTXC to the NPE-C167" =>
13:45:40B4gderfor some fun reading on RTXC details...
13:46:18B4gderI guess that won't help much though
13:49:26 Quit Thundercloud (Remote closed the connection)
13:49:30 Join Thundercloud [0] (
13:50:57 Join Thundercloud_ [0] (
13:51:04 Quit Thundercloud (Remote closed the connection)
13:52:21jhMikeSDr. Dobb's seems to be fancier since I last saw it years ago anyway
13:53:11Davo_DinkumIs there a filesystem that works on windows and GNU/Linux apart from FAT?
13:53:26 Quit Siku ("Laters...")
13:53:42preglowwith ntfs-3g installed, you'll have write support too
13:53:55Davo_Dinkumhmm i guess
13:54:52 Join Thundercloud [0] (
13:55:12jhMikeSpreglow: you killed half my bytes savings there - you bastard!! :P
13:55:23preglowjhMikeS: :PpPpPp
13:55:48preglowi wonder if i should be bothered to port those changes to the regular ipods
13:55:59preglowprobably won't be as noticable
13:56:35 Quit Thundercloud_ (No route to host)
13:56:36IsolinearYou talking about the video rendering patch?
13:56:47 Quit Thundercloud (No route to host)
13:57:15B4gderbtw, on the topic of gcc versions, it would make sense to do an attempt to go with a newer gcc for arm since the gigabeat-s guys more or less require that
13:57:17 Join Thundercloud [0] (
13:58:17Zagorthe iphone update appears to add stricter sim card checks. not quite the bricking that was threatened then.
13:58:22preglowB4gder: why do they do that?
13:58:38B4gderpreglow: apparently some arm11 stuff that went in rather late
13:58:51B4gderI don't recall the details right now
13:59:57morrijrDavo_Dinkum: there's something for windows which lets you read ext2 IIRC
14:00:01preglowseems gcc4.3 is in stage 3 now
14:00:12preglowi had higher hopes for that and coldfire than seems to be fulfilled
14:02:55jhMikeSpreglow: gcc in stage 3? sounds like a disease progression.
14:03:08preglowmight be too :P
14:03:19preglowB4gder: i _think_ i tried gcc 4.2 and found it to work ok
14:03:29hcsor a world domination plot
14:03:46hcsgcc progressing to stage 3, deploy the death squads
14:03:52preglowand the green deltas!
14:03:58n1spreglow: according to lear's testing it give a nice speedup of about 20% in aac , but drops about 10% in tremor
14:04:14senabmorrijt: ext2is
14:04:56preglown1s: you're talking 4.3 now, right?
14:05:06n1spreglow: yup
14:05:17jhMikeSI wonder if this mult removal on ARM could be done for audio codecs. It's quite a pain to find optimal shift+add/subtract though on large numbers.
14:05:18preglown1s: yeah, and i'd prefer that 10% in tremor didn't happe n:/
14:05:21preglown1s: did he mention size changes?
14:05:37preglowjhMikeS: mult removal?
14:05:37n1spreglow: yes we got some code size increases
14:05:43preglown1s: increase :/
14:05:58 Join seablue [0] (
14:06:07jhMikeSpreglow: check out the idct for ARM and the yuv code
14:06:36jhMikeSnot one single mul or mla used in them
14:06:50n1snothing major, for H300 Tremor was the only thing that didn't fit iram anymore, h100 had size problems with the lcd drivers
14:07:26preglowjhMikeS: right
14:07:32preglowwell, it's all good, arm muls are shit slow
14:08:07 Quit B4gder ("It is time to say MOOO")
14:08:15jhMikeSespecially if they're large numbers. about the only fast mul is x0. :)
14:08:29n1smaybe they'll fix some regressions before release but I guess we can tweak optimizer options etc to make it work better
14:09:18jhMikeSas long as it's C*x it saves big cycles. if it's not by a constant, well nothing can be done.
14:09:59preglowjhMikeS: i don't think x0 differs from x255
14:10:24preglowwe really should try to go 4.3 if possible, though
14:10:40jhMikeSdepends on the arm version but (x << 8) - x still measured better
14:10:43n1sNow why would putting a buffer in iram make this darned midi thingy hang on start?
14:11:16preglowjhMikeS: yeah, you can't beat one cycle
14:11:27jhMikeSbut guess that just rsb r0, r0, r0, asl #8
14:11:31preglowit is
14:12:08jhMikeSI figured that too but even with YUV code which only handled stuff of one byte it was way better to use adds
14:12:36preglowit's a real pity they didn't enhance the multipliers
14:12:38preglowthat's for sure
14:12:46*preglow strokes mac.l
14:12:49preglownow there's a multiplier engine
14:12:50*Nico_P expects a green delta :)
14:13:25hcseveryone expects the green deltas!
14:13:43jhMikeSyeah, no kidding. with ARM - watch the multiplies and load stalls...with Coldfire, watch the emac stalls and the fact there's no lousy dcache.
14:13:48senabdeploy the red deltas :P
14:13:48 Quit Jon-Kha_ (Remote closed the connection)
14:14:22preglowjhMikeS: i wouldn't exactly call the mac stall as big a problem as the arm multipliers, though
14:14:33jhMikeScoldfire would've kicked butt with a dcache
14:14:38preglowit would
14:14:40preglowthat's for damned sure
14:14:53preglowiram _and_ a small dcache and it'd zip along like a madman
14:15:17jhMikeSit's just funny we can run every DSP feature and not have boost at all.
14:16:19preglownot only funny, but also amazingly hardcore :D
14:16:40jhMikeShehe. I wonder about those DM320's DSPs.
14:17:03n1syep you 2 (+ some others) did a fine job on coldfire asm dsp :-)
14:18:02preglowjhMikeS: i can assure you they're so fast it's not even funny
14:18:07jhMikeSI never expected that mighty result from it. it came as a shock to watch the boost drop and drop with every use of emac no matter how abusive.
14:19:32jhMikeSpreglow: well, then they're good for more intense sorts of things. I also wonder about the blackfin DSP.
14:19:52preglowalso good, but for low-precision kind of stuff
14:19:53 Quit senab (Read error: 104 (Connection reset by peer))
14:19:56preglowsharc, now that's what i want
14:21:02*jhMikeS waits for amiconn to pop up and say it's too powerful :P
14:21:19IsolinearDo the commits not have individual revision numbers?
14:21:26preglowNico_P: wooot
14:22:23Nico_PI have another patch that takes 500 bytes off by removing the settings, but I'd rather not commit that one
14:22:25*jhMikeS wonders about emac for blending code ;)
14:23:32 Join miepchen^schlaf [0] (
14:23:43preglowsomeone please go ahead and make a nicer eq screen :/
14:24:15jhMikeSI guess we can use gradients in it :)
14:24:52preglowoh, that's a nice point, i find a screen that stacked selector bars yesterday
14:25:02preglowand that really should combine the two to one gradient
14:25:09preglowlemme see where that was
14:25:20Nico_Ppreglow: I know about that, but couldn't think of a solution
14:25:30Nico_Ppreglow: runtime ?
14:25:37preglowlooked really silly
14:25:41peturor recording screen
14:26:00Nico_Pit's the list widget's fault
14:26:01preglowthere really should just be one gradient covering the two fields
14:26:05Nico_PI agree
14:26:10pondlifeID3 viewer?
14:26:24preglowwell, is it that hard to fix?
14:26:58 Join massiveH [0] (
14:27:05Nico_PI'll take a look
14:27:20Nico_Pthe problem is that the drawing is done line by line
14:27:42jhMikeSthat gradient is actually nice to look at and softens the appearance nicely
14:27:42Nico_PI mean the drawing of the list
14:28:07pondlifei.e You can't reset the gradient at the start of a list line?
14:29:00Nico_Pwhat do you mean ?
14:29:39pondlifeJust that you can't start a list line with the primary colour, you need to reset at the start of a list item.
14:29:43 Join crazyshit [0] (
14:29:59preglowjhMikeS: just tried the ipod scroll wheel accel patch, seems to work cool here, but is a bit too sensitive
14:30:38preglowand doesn't decay the acceleration factor fast enough when you stop scrolling
14:31:06Nico_Ppondlife: the main problem is that I don't think there's a way to tell the lcd driver that the gradient will need to be on two lines
14:31:22Nico_PI'll take a closer look though
14:31:23jhMikeSpreglow: tell that to the author :). I did quite a bit of arguing about such things and why I implemented various aspects of it.
14:31:58jhMikeSvarious aspects of the interface, not of that patch in particular.
14:32:44jhMikeSI don't doubt it's oversensitive and probably harder to control at low speeds than it should be.
14:32:55preglowthat's my major nitpick
14:33:03preglowacceleration behaviour itself is just peachy
14:33:48jhMikeShe didn't want to do the slow mode and directional hyteresis. without that the sansa's wheel wasn't very controllable at low speed.
14:34:05preglowany particular reason?
14:34:34preglowi'll just read the patch history
14:35:13jhMikeSif you don't have it, then imagine moving just past a click point by a tiny amount. if you reverse direction, it will instantly click back so simple shaking can make it difficult to select an item.
14:35:54hcsproposal: each click selects an item from the list at random
14:36:36jhMikeSIs the like google's "I'm feeling lucky"?
14:37:18IsolinearHahaha, very nice.
14:37:26hcslet's say it is, we'll say "Advanced Artificial Intelligence chooses the most appropriate track automatically."
14:37:45crazyshitcan i see emulator of rockbox
14:37:49crazyshitrunnin in pc ?
14:37:52jhMikeSWe don't read your mind...rockbox does
14:38:33hcsyou may not like the selection, but a few years down the road it'll suddenly dawn on you that rockbox saved your life!
14:39:36jhMikeSwhile your doubt consumed you, that passenger train simply passed by...just think where you would be without rockbox
14:40:32hcsyou may find it recommending artists that sound almost, but not entirely, unlike Ice-T
14:43:40preglowugh, data abort while scrolling files :///
14:44:14 Join Jon-Kha [0] (
14:45:02Isolinearpreglow: You're using the accel patch?
14:45:07preglowtrying it out, yeah
14:45:12preglowbut it's probably unrelated
14:45:18IsolinearYeah, that used to happen to me a lot.'s quantum multiverse versions of Ice-T that don't exist in this reality
14:45:55IsolinearHasn't happened once with the EvilG build everyone was bashing earlier... lol
14:46:41hcsIsolinear: having established how much control Rockbox has over your destiny, would you really trust a build from someone who calls himself "Evil"?
14:47:16jhMikeSpreglow: I can't promise the commits fixed every possible one of those. We really need COP/CPU linker data sections aligned to 16 bytes.
14:47:20IsolinearAs long as he doesn't include your AAICTMATA patch... :P
14:47:35pondlifeUrgh, this H300 bootloader thing is resulting in the mother of all binary chops 5 more iterations to go... :/
14:47:55preglowyou're testing to find out where the bootloader broke?
14:48:05preglowso you're just bricking your h300 over and over? :P
14:48:07pondlife14000-14060 somewhere
14:48:14pondlifeYes, I am multi-bricking
14:48:55preglowwell, how do you unbrick?
14:48:59jhMikeShow are you unbricking then? I thougt iRivers needed LinusN to reflash them.
14:49:08 Quit desowin ("use linux")
14:49:14peturonce the OF no longer loads it will really be bricked
14:49:18pondlifeNah, the lockup is after the OF check, so I can hold REC
14:49:25hcsperhaps he has LinusN in a cage nearby?
14:49:38pondlifeIt's ok, I let him keep his clothes
14:49:58*jhMikeS would wonder if you didn't :P
14:50:17pondlifeOK, here goes 14030...
14:50:21 Quit skatteola (Read error: 110 (Connection timed out))
14:50:29 Quit crazyshit ()
14:50:48 Join skatteola [0] (
14:50:59*jhMikeS proposes a new brickloader
14:51:45pondlifeModern DAPs are too small and light to make good bricks, we need a better term.
14:52:11Nico_Pwell I've found a way to make the lcd driver aware of the size of the selection in the list widget, but it doesn't really have a way of knowing which line is first
14:52:21pondlifeThe Iriver OF is so slow to boot, and I'm sure it could flash faster :)
14:53:20jhMikeSNico_P: I don't understand why it should have to. Isn't it just a simple draw op?
14:53:36Nico_PjhMikeS: there's one draw op per line
14:54:43 Join nicktastic [0] (n=nick@unaffiliated/nicktastic)
14:54:44*jhMikeS gets confused...line of what? text? "raster"?
14:54:50preglowtext, probably
14:55:12preglowi think we have a rather weird lcd driver concept
14:55:39jhMikeSbecause it cares about text lines on bitmap targets?
14:56:03pondlifeViewports won't like that...
14:56:05jhMikeSagreed. it should be rectangle based.
14:58:47*jhMikeS is really ticked his own changes will conflict with the full dual-core patch. argh...big resync job.
14:59:13*preglow goes to work on the comfortably smaller wakeup problem :P
14:59:54jhMikeSpreglow: WAKE UP!
15:01:14Nico_PI did it !! \o/
15:01:20 Join webguest81 [0] (i=3b640552@gateway/web/cgi-irc/
15:02:29jhMikeSyou know, this darn H10 has really good sound. wonder why it seems a bit exceptional.
15:02:36webguest81does anyone know if exe files can be emulated in rockbox
15:02:59Nico_Ppreglow: it's a very small change too
15:03:05preglowthen hooray doubly
15:03:16hcswebguest81: in a very theoretical sense, yes; in any real sense no
15:03:17jhMikeSwebguest81: they could if you write an x86 emulator
15:03:26Nico_PI just need to make the recording screen behave the same
15:03:31Nico_Ppetur ?
15:04:08Nico_Pwhich targets have the double selection thing in the recording screen ?
15:04:24peturthose that have left and right gain
15:04:29jhMikeSpreglow: one crazy control is the H10 scrollpad that could actually use both kinds of acceleration depending on if it's used like a button or wiped. I'm thinking how to pull that one off.
15:04:31webguest81is there a way to do this by creating a wad file for the doom emulator
15:04:52peturNico_P: I think all recording targets that have line-in
15:05:02hcswebguest81: no, what are you trying to do?
15:05:02pondlifeNico_P: Does the ID3 viewer also use double-line lists?
15:05:02petur( = stereo)
15:05:19Nico_Ppondlife: yes, and it's beautiful now :)
15:05:37webguest81something new i guess
15:06:06Nico_Ptheoretically it should work with an arbitrary number of selected lines
15:07:09Nico_P... as long as the number holds in 4 bytes :p
15:07:30jhMikeSnot enough with these new fangles hires screens
15:08:24webguest81hcs: can i discuss this in a pm
15:08:32hcsfangles co, ltd is a leading manufacturer of the xxxxxxxga screens, though, so not so inaccurate
15:09:11jhMikeSa single molecule per pixel?
15:09:14Nico_Ppetur: can you tell me how do I know the number of selected lines in the rec screen ?
15:09:55Nico_Phmm, also my sim doesn't give me two lines
15:10:09***Saving seen data "./dancer.seen"
15:11:27webguest81hcs: r u still around
15:11:44hcswebguest81: yes, I messaged you a few times
15:12:15webguest81so pm
15:13:34peturNico_P: if you move dup/down, the selection is volume, L+R (2 lines), L, R,...
15:13:54Nico_Ppetur: not on the sim I think (no line in)
15:14:20 Join darkapostrophe [0] (
15:14:40pondlifeHmm, looks like r14015 broke the H300 bootloader
15:14:55IsolinearIf I specify hardware EQ settings, are those changes stored somewhere that I can copy over when updating a build?
15:15:17*preglow wonders why all the pcf register defines are in the .c file
15:15:27preglowIsolinear: in the config file?
15:16:08IsolinearGo figure... :)
15:17:23IsolinearCan I edit the frequency information to values that aren't available in the menu?
15:17:36preglowdepends, let me check
15:18:13preglowi would like someone with a video to remove that hw eq, btw, someone please volunteer, there is a patch for it
15:18:18*jhMikeS recommends a 0Hz low shelf filter cutoff
15:18:32IsolinearRemove it?
15:18:58preglowIsolinear: the peaking bands don't work, so it's essentially just a treble/bass control
15:19:07preglowIsolinear: no reason to waste all that space and code on pretending it's an eq
15:19:23jhMikeSpreglow: why aren't all the bands used?
15:19:29preglowjhMikeS: they don't work
15:19:32preglowjhMikeS: hw issue
15:19:38IsolinearThat explains why I didn't hear as much of a difference as I expected...
15:20:04IsolinearThe only reason I was using it was to take work off of the processor with the software eq.
15:20:05jhMikeSpreglow: because noone knows how or because the HW is really broken?
15:20:12preglowjhMikeS: who knows
15:20:21 Quit CaptainSquid (Remote closed the connection)
15:20:39preglowjhMikeS: i think i remember scanning retailos for accesses to those parts of the chip, but i couldn't find any
15:20:53preglowjhMikeS: i did however find an iir filtering routine, suggests they do eq in software themselves
15:20:58preglowor whatever passes for an eq with them
15:21:12Nico_Ppreglow, pondlife: the only problem with my change is for scrolling lines
15:21:37preglowNico_P: that does sounds harder to fix, is it?
15:21:42Nico_Pit is
15:21:44IsolinearIf one of the major complaints about OF sound is that the eq sucks, it would stand to reason that the original eq hardware could be the reason for said suckage.
15:22:03Nico_Ppreglow: I did try something but it didn't work
15:22:13Nico_PI think I'll just leave that case out for now
15:22:18preglowNico_P: if you can't make scrolling work, i'd say screw it and let's do it properly some other time
15:22:23jhMikeSpreglow: well, this usually means to try writing to anything you don't know for sure what it does and watch
15:22:44preglowjhMikeS: i don't have a video, which pretty much means it won't happen
15:23:05preglowso i'm going for the next best thing, which is giving ipod video treble/bass controls
15:23:18jhMikeShow do they know there's 5-bands anyway if nothing uses it and it's not doced?
15:23:33IsolinearThat's what I was wondering..
15:23:33preglowit is doced, but the chip isn't a perfect model match
15:23:46preglowso they might have just excluded those parts for cost
15:24:04jhMikeSor maybe they bought bum chips for cheap
15:24:12preglowwho knows
15:24:12IsolinearWrite Apple and ask.. :P
15:24:24Nico_Ppetur: in recording.c, how can I know whether one or two lines are selected ?
15:24:27IsolinearI'm sure they get that question all the time.
15:24:58jhMikeSI'm sure apple is pretty unaskable about anything other than "what colors does it come in?"
15:25:00preglowIsolinear: looks like any values outside the usual settings range will be clamped to the nearest valid setting
15:25:16preglowjhMikeS: yeah, and "where do i pour my money"
15:25:34IsolinearBut even those valid settings are invalid? lol
15:25:56 Part Davo_Dinkum
15:25:58IsolinearBecause the options are way more complicated that the actual hw? :)
15:26:05jhMikeSJobs-a the Hut hungry
15:26:08 Join roolku [0] (
15:27:21IsolinearAlso in the config file I see "usb charging: on"... What is that?
15:27:27IsolinearOr is it from the patched build only?
15:27:36jhMikeSargh, 18 conflicting files
15:27:42peturNico_P: sorry, work overload
15:28:16roolkuNico_P: since you are still meddling with the gradient - i must say I find the linear gradient less visually appealing than the original one - any chance of getting it back?
15:28:38Nico_Proolku: not really, not with both colours configurable
15:28:58pondlifeOK, I have a working, current, H300 bootloader
15:28:59preglowyou can do an exponential gradient with both colours configurable, sure
15:29:04preglowyou'll need libm, but hey :P
15:29:20roolkuNico_P: it used to look like a rounded bar, lit from the top and shade at the bottom, now it is just a smudge between two colours
15:29:20preglowpondlife: commit?
15:29:27pondlifeNo, it'll break iPods
15:29:28hcspreglow: bah, who needs a math library
15:29:47roolkuNico_P: it is not possible to recreate the effect with the current implementation
15:29:52pondlifepreglow: Reverting fixes it
15:30:07Nico_Proolku: I'd sure welcome a patch ;)
15:30:23pondlifeSo... something is requiring a CPU freq change in ata_init() on H300.
15:30:32roolkuNico_P: you keep breaking it. :)
15:30:52Nico_Proolku: I mean a patch with both colours configurable
15:31:19 Quit Dark_Apostrophe (Connection timed out)
15:31:19 Nick darkapostrophe is now known as Dark_Apostrophe (
15:31:24Nico_Ppetur: do you have time to point me to the right place ?
15:31:33roolkuNico_P: it doesn't make sense with 2 colours - shade is black
15:32:20roolkuNico_P: I would see it as a different option
15:32:22 Quit MethoS- (Read error: 110 (Connection timed out))
15:32:22Nico_Proolku: people wanted to be able to choose both colours...
15:33:15roolkuNico_P: hence additional option
15:33:26Nico_Proolku: a bitmap line selector is kinda planned, so maybe you'll be able to do it that way
15:33:38Nico_PI'm not sure a "gradient type" option would be welcome
15:33:59pondlifeamiconn: Note for later, please have a look at
15:34:05peturNico_P: line 1759 of apps/recorder/recording.c
15:34:10|Rain|I think what's being suggested is another option in the selector type
15:34:28 Nick Gekz is now known as Gekz[PDA] (
15:34:37Nico_Ppetur: thanks :)
15:34:40roolkuNico_P: not to worry, I'll just patch it locally, just thought other people might like it as well...
15:35:36roolkuNico_P: (just stop breaking my patch :p )
15:35:47Nico_Proolku: I have one last commit
15:35:58 Part Gekz[PDA]
15:36:13|Rain|e.g. Bar (Single-colour Gradient) and Bar (Dual-color Gradient)
15:37:10roolku<|Rain|> yep, something like that
15:37:24pondlifejhMikeS: Where's spinlock_init() defined?
15:38:04roolku<|Rain|>: (and get rid of the <|> it's no good for tab completion :) )
15:38:05|Rain|it wouldn't add yet another menu option to the settings menu (just another selection for one of the options), so it probably wouldn't ruffle too many feathers
15:38:23|Rain|heh :P many tab completions will work if you just type 'rain' and hit tab
15:38:52Nico_Pwow that's actually true
15:39:02pondlifejhMikeS: Don't worry, found it
15:39:09*Nico_P pats Konversation
15:39:26*GodEater_ 's autocomplete does "last to speak" matching
15:40:12|Rain|I've been using this nick for over 10 years now, I feel naked without my pipes
15:40:29IsolinearSo anyone wanna explain what "Charge During USB Connection" does for iPod 5g?
15:40:32preglowcan i quote the latter part of that sentence to random people? :P
15:42:40preglowman, this bootloader always enabling wake on rtc is annoying
15:43:01 Quit Pyromancer (Read error: 104 (Connection reset by peer))
15:43:26pondlifeNico_P: Now I've used it, I quite like the gradient selector... any chance it could go behind the icons too :p
15:43:51pondlifeSeems odd to highlight the text and stop by the icon
15:43:54roolkupondlife: yep, another think I have patched locally. :)
15:44:14Nico_Ppondlife: it used to
15:44:27pondlifeDid it look bad with some icons?
15:44:44 Part przemhb
15:44:51Nico_Ppondlife: you want it like this ?
15:44:53pondlifeI guess the icons would need to include a border
15:45:09Nico_Pwell it looked weird with antialiased icons
15:45:48pondlifeNico_P: Yes, I prefer that
15:46:12pondlifeAnyway, lunchtime
15:46:22 Quit pondlife ("disconnected has pondlife")
15:46:41 Join kubiix [0] (
15:51:01 Join TMM [0] (
15:52:13preglowNico_P: don't you need to change any other lcd drivers than 16 bit?
15:52:25preglowi guess not
15:52:31 Join norbusan [0] (
15:52:36 Part norbusan
15:52:40Nico_Ppreglow: only the 16bit friver has gradient
15:54:49|Rain|gradients on lower-depth lcds would probably make the text difficult/impossible to read
15:57:00 Quit Nico_P (Remote closed the connection)
15:57:22 Join Nico_P [0] (n=nicolas@rockbox/developer/NicoP)
16:01:12 Quit Zagor ("Client exiting")
16:03:19 Quit petur (" <-- gone taping :)")
16:03:52 Join agm3nt [0] (i=nat-107@
16:06:54 Join freqmod [0] (
16:10:14 Join SSnake [0] (n=dfsff@
16:10:52SSnakehi guys
16:10:53preglowaren't pcf50606 and pcf50605 really the same thing?
16:11:37SSnakei read irc log and pondlife discoveries
16:11:52SSnakei can confirm about the iriver H300 bootloader issue
16:12:12 Join XavierGr [0] (
16:12:18SSnakethe problem is n the change to the header file
16:13:01SSnakejust dowloaded a fresh source archive, deleted the #if defined(BOOTLOADER) && defined(HAVE_ADJUSTABLE_CPU_FREQ)
16:13:18SSnakeall compiled and worked for me
16:13:50 Quit kubiix (Read error: 110 (Connection timed out))
16:14:48SSnakemaybe it has to be changed with something like #if defined(BOOTLOADER) && defined(HAVE_ADJUSTABLE_CPU_FREQ) && !defined(IRIVER)
16:16:12SSnakein file / trunk / firmware / export / config.h
16:16:38 Join MethoS- [0] (
16:17:34 Quit SSnake ()
16:21:51 Join desowin [0] (
16:23:01preglowlinuxstb: any reason you stuffed all the port defines for pc50605 in the .c file and not the .h file?
16:25:08 Join toffe82 [0] (
16:25:12 Join pondlife [0] (n=Steve@rockbox/developer/pondlife)
16:25:20 Join xxhotman198 [0] (
16:26:01 Quit xxhotman198 (Client Quit)
16:26:12 Quit daurn|laptop (Read error: 104 (Connection reset by peer))
16:28:47 Quit tchan (Remote closed the connection)
16:33:02 Join tchan [0] (n=tchan@lunar-linux/developer/tchan)
16:33:08 Join daurn|laptop [0] (n=fake@unaffiliated/daurnimator)
16:34:01 Quit Thundercloud (Remote closed the connection)
16:37:08 Join Thundercloud [0] (
16:38:22 Quit hcs ("Leaving.")
16:38:32 Quit webguest81 ("CGI:IRC")
16:39:49 Join Lear [0] (i=chatzill@rockbox/developer/lear)
16:40:28 Quit Nico_P (Remote closed the connection)
16:40:46 Join Nico_P [0] (n=nicolas@rockbox/developer/NicoP)
16:44:23preglowGodEater_: around?
16:44:40preglowthink i've got a patch to fix the wakeup problem here
16:44:46preglowdon't know if i should bother testing before commiting
16:45:11GodEater_I'd say it would be a good plan usually
16:45:27preglowyeah, but in this case it'll take time to have someone test it :P
16:45:34preglowthink i'll just commit and see if it goes away
16:54:30preglowi really have no idea why this happens, though
16:54:33preglowbut let's hope this fixes it
17:00:51 Join saratoga [0] (i=9803c6dd@gateway/web/cgi-irc/
17:01:22 Join barrywardell [0] (
17:03:21 Join linuxstb_ [0] (n=linuxstb@
17:05:23linuxstb_preglow: Regarding the port defines in pcf50605, why would you want them in the .h file? They're not public (i.e. used outside the .c).
17:05:37TMMlinuxstb_, hi! :)
17:05:49preglowlinuxstb_: they should be, in rtc_pcf50605.c
17:05:56preglowlinuxstb_: now it just uses lots of pretty constants
17:07:04linuxstb_preglow: OK, then I guess they should be moved to the .h
17:07:06|Rain|I spent all that time fixing the e200 bootloader and jethead71 went and broke it again
17:07:21preglowlinuxstb_: i'll see about it some day
17:08:14|Rain|anyone know if jethead71 IRCs and, if so, what his nick is? a lot of his changes are low-level enough that I'm afraid to touch them
17:08:19preglowjhMikeS: yo
17:08:25preglowjhMikeS: someone's looking for you :>
17:08:39jhMikeSyeah? the IRS? the FBI? CIA?
17:08:45preglowall of them
17:08:49|Rain|well, hi :P
17:08:49pondlifeAnd |Rain|
17:08:53preglowplus men with latex gloves
17:09:04*jhMikeS *runs* to Argentina
17:09:31jhMikeS|Rain|: what's up?
17:09:57*jhMikeS figured pondlife would be the one to actually put them on :P
17:10:07 Join darkapostrophe [0] (
17:10:10***Saving seen data "./dancer.seen"
17:10:27|Rain|tbh, I'm not really sure where to start poking at this, but since I updated svn (I think r14879's the culprit) the e200 bootloader hangs with the sandisk logo still on the lcd
17:11:02jhMikeS|Rain|: you changed the bootloader?
17:11:19|Rain|yes, I made it actually work again as of yesterday
17:12:01 Join Arathis [0] (
17:12:26jhMikeSThe bootloader didn't build for r14879. there were errors.
17:12:59|Rain|so r14880 was to fix those?
17:13:19jhMikeSyeah, it should have been
17:14:09|Rain|well, it does compile now (I'm actually at r14884, I haven't started stepping back through yet to see exactly where it broke), but it does hang before the LCD is even initialized
17:16:52preglowNico_P: two line gradients look marvelous :)
17:17:15jhMikeSI'm sure it's just something ridiculously simple
17:18:49|Rain|probably. I'll try to make sure it's that rev when in a few. bbiab
17:19:00pondlifeAnyone know if amiconn is away at the moment..?
17:19:14preglowwell, he's not here
17:19:28pondlifeI know that much... :)
17:19:38pixelmayes, he is - probably won't be back until Sunday evening or Monday morning
17:19:42barrywardell|Rain|: was your patch to fix the booting OF problem?
17:20:20jhMikeSbarrywardell: I was thinking about that but stupidly didn't ask about that :P
17:20:30*GodEater_ points pondlife at the General Discussion Forum
17:20:47*pondlife goes to look
17:21:46pondlifepool? bar billiards, please.
17:21:46barrywardelljhMikeS: the patch looks like a good idea anyway - we don't want to create threads in the bootloader
17:21:46 Join Wiwie [0] (
17:22:04GodEater_pondlife: isn't that the thing with the skittles as well as the balls ?
17:22:09 Quit Dark_Apostrophe (Connection reset by peer)
17:22:09pondlifeOh yes
17:22:09 Nick darkapostrophe is now known as Dark_Apostrophe (
17:22:23GodEater_pondlife: you know somewhere which still has a table ?
17:22:32pondlifeNot in London
17:22:39GodEater_then we can't do that can we ?
17:22:51GodEater_so pool it will have to be
17:23:08pondlifeAlthough I don't really count Kent as London.
17:23:09GodEater_although the venue I have in mind has bowling lanes, and dodgems too
17:23:10jhMikeSbarrywardell: well that was mostly to just keep the red away but I'm not even sure why the BL needs dual core support of any sort.
17:23:17GodEater_and video games
17:23:21GodEater_and most important of all
17:23:52Dark_ApostropheVodka > *
17:23:52pondlifeDoesn't sound like a pub to me... is this DevConBowlingHall?
17:23:54 Join przemhb [0] (
17:23:57jhMikeSas long a crt0-pp-bl.S sends the COP to the correct location in the firmware when needed
17:25:34 Quit n1s ()
17:25:46 Quit linuxstb_ ("Leaving")
17:25:59 Join low_light [0] (i=c730180b@gateway/web/cgi-irc/
17:26:24pondlifeHmm, you scared linuxstb off already
17:27:13GodEater_I twisted his arm so hard he fell off the internet
17:27:16 Join bitbit [0] (
17:27:33pondlifeThe phrase "go out on a limb" seems over appropriate
17:27:53 Quit Ebert ()
17:28:04GodEater_I tend to go out on two
17:28:05jhMikeSperhaps NUM_CORES ought to be 1 in the bootloader no matter what
17:28:08GodEater_both my legs usually
17:28:20bitbitI want to install a theme for rockbox but it says i need "WPS Tokenizer". what is that and where do I get it from?
17:28:35GodEater_bitbit: if you have a recent build of rockbox - you already have it
17:28:51low_lightjhMikeS: On the sansa c200, microsd cards are detected on GPIOL, but I can't seem to get the interrupt working...any suggestions
17:28:51bitbitI do
17:29:01bitbitso what do I do with the theme's folder?
17:29:06pondlifeI'd rather find a proper drinking venue... somewhere conducive to talking rubbish for hours, and with a bar open 'til 3am, or death.
17:29:15jhMikeSlow_light: you tried both GPIO0_MASK and GPIO1_MASK?
17:29:36pondlifeGodEater_: Best not chat too much in here though...
17:29:47jhMikeStry another bit one up from that? could be another.
17:30:27jhMikeSMaybe there's GPIO0/1/2/3/4_MASK...who knows
17:30:32low_lightwhen i tried the next bit it froze on boot
17:31:35bitbitGodEater_: so what do I do with the theme's folder? I tried putting it with the other default themes (which are only single files), but the player doesn't see it
17:31:39jhMikeSand you also tried moving which one microsd_int is under?
17:31:53GodEater_bitbit: most themes are NOT a single file
17:32:04pondlifelol - GoldenQuote maybe - "I now read that the iPod Classic has 160GB and Im just litterally having a orgasm with that size."
17:32:11GodEater_bitbit: they may be zipped up into one (which you then extract to your player), but they normally comprise many files
17:32:21GodEater_pondlife: don't you dare
17:32:42pondlifeNot me, read the forum:
17:32:51GodEater_no I saw it already thanks
17:32:57GodEater_I just don't want it on the GoldenQuotes page
17:32:59GodEater_or I'll cry
17:33:08krazykitGodEater_, not like you can't just get rid of it :P
17:33:18bitbitGodEater_: under /.rockbox/themes I have a few files called "Rockboxed", "iCatcher" etc.. I think they are .cfg
17:33:43low_lightjhMikeS: yeah...I think I tried it immediately under: if (CPU_HI_INT_STAT & GPIO0_MASK)
17:34:00bitbitI uploaded the theme's folder into that folder, but nothing happened.
17:35:09jhMikeSlow_light: and under the GPIO1_MASK clause with CPU_HI_INT_EN = GPIO1_MASK ?
17:35:14bitbitnvm I think I got it
17:35:30barrywardelljhMikeS: were you asking about bootloader IRAM usage earlier? I don't think it's safe to use IRAM in the bootloader
17:36:19jhMikeSbarrywardell: I didn't either. It was just that the cache functions are in IRAM but they're not included anyway.
17:36:20barrywardellthe first thing the bootloader does is copy itself to IRAM and runs from there.
17:36:28low_lightjhMikeS: yes, in a separate builds I tried GPIO0 and GPIO1
17:36:37barrywardellso it could be overwriting all sorts of things
17:37:28 Join ToHellWithGA [0] (
17:37:32jhMikeSkernel code does include IRAM by default anyway but I added no additional stuff
17:38:08 Quit bitbit ()
17:38:43jhMikeSlow_light: it must be somewhere so get it to work if OF manages it.
17:40:01jhMikeSI poked at the e200 like hell before I figured out there was another GPIO hi bit but I also thought I failed with it earlier so that slowed things up.
17:40:36barrywardelljhMikeS: yeah, I wonder how that works, given that IRAM gets overwritten first thing
17:41:46jhMikeSnot actually linked as IRAM?
17:41:48low_lightjhMikeS: you might want to check if the c200 system_reboot() works for the e200 too
17:42:21|Rain|barrywardell: I don't think so. I di dhave a patch related to properly picking the OF when the USB cable is inserted, but that's as close as it comes
17:42:48jhMikeSlow_light: is that even connected to anything atm?
17:42:56tumujhMikeS, nls told me that you have worked on the sim recently. would you have any idea why compiling sim with -fprofile-arcs does not create all datafiles on sim exit?
17:43:26|Rain|barrywardell: the main patch skipped checking the SD card in the bootloader, since it's irrelevant and requires interrupts, and doesn't add the button_tick task since it's unused and also requires interrupts to be meaningful
17:43:29barrywardelljhMikeS: it's a while since I've looked at it, but the start of crt0-pp-bl.S copies the whole bootloader to IRAM. That's why the bootloader size also can't grow bigger than IRAM size
17:43:38low_lightjhMikeS: when usb is inserted it reboots & loads the OF
17:44:34jhMikeSlow_light: ok, I've never done it. How to trigger?
17:45:03barrywardell|Rain|: so does it fix a problem that you were experiencing?
17:45:22jhMikeStumu: perhaps some non-rockbox thread is trying to use sim_write?
17:46:00|Rain|barrywardell: without the SD interrupt fix, the bootloader froze 100% of the time, and without the tick_add_task fix, the bootloader paniced randomly
17:46:15Dark_ApostrophePANIC, AAAAAAAAA!
17:46:48barrywardell|Rain|: ah, I see. I never experienced those problems on my own e200
17:47:12|Rain|when's the last time you build the bootloader, though?
17:48:02low_lightjhMikeS: add e200 defines to system_reboot in system-pp502x.c and usb_enable in usb-fw-pp502x.c
17:48:37 Quit obo ("bye")
17:48:57tumujhMikeS, i don't think that's the case because all datafiles (.gcda) are missing after sim exit. the compile option is supposed to cause those to be created on exit (i guess it should hook exit routine or such)
17:49:42tumujhMikeS, it does create .gcno files during run tho like it is supposed to
17:50:27agm3ntdoes latest e200 bootloader work??
17:50:37tumujhMikeS, you can easily test it by including the option on sim cflags, no other changes are needed
17:50:45jhMikeStumu: using that really isn't something I'm familiar with. I was just guessing as to a possible cause.
17:50:51|Rain|agm3nt: if latest = built from svn, not for me
17:51:26agm3ntc200 bootloader also not work
17:51:39agm3ntfor me
17:52:04tumujhMikeS, it makes two datafiles for each .o which contain function performance statistics
17:54:52 Quit Thundercloud (Remote closed the connection)
17:55:43tumujhMikeS, if you read the last chapter in invoking gcov, you'll see that the data is saved on exit, which does not seem to happen right now
17:55:53tumuer, last paragraph
17:55:56low_lightagm3nt: what happens?
17:56:09|Rain|k, the bootloader from r14878 works for me
17:56:46|Rain|r14879 fails to build as you said, r14880 is building now
17:56:47barrywardell|Rain|: I built yesterday or the day before
17:57:10|Rain|barrywardell: ah. maybe I'm just unlucky, then
17:57:28agm3ntlow_light: it's frezze on sandisk logo
17:57:39barrywardellthere seems to be people having different experiences with the bootloader
17:57:41 Join Thundercloud [0] (
17:57:54 Join jhulst [0] (n=jhulst@unaffiliated/jhulst)
17:58:36|Rain|agm3nt: that's what happens to me, too. I think r14879 is the culprit... I'm about to verify
17:58:59low_lightagm3nt: press a button while booting and see is any messages are printed
17:59:43|Rain|low_light: it's freezing way before the button is checked
17:59:49jhMikeStumu: it's really not easy to say why. it appears to do it's own thing and shouldn't depend on the particulars. hmmm.
18:00:41|Rain|k, I confirmed it: r14878 boots, r14880 bricks
18:02:20tumujhMikeS, does the sim exit in any special way?
18:02:50|Rain|jhMikeS: do you have anything you'd like me to try before I go poking random pieces of the code with a stick?
18:02:50agm3ntlow_light: I dont see any messages, only sandisk logo
18:02:55preglowthe recording screen doesn't really seem well adapted to ipod
18:03:04 Quit jhMikeS (Nick collision from services.)
18:03:10 Join jhMikeS [0] (n=jethead7@rockbox/developer/jhMikeS)
18:03:24jhMikeSthe key should be checked before anything enters fw code, no?
18:03:39|Rain|iirc, it's checked just after lcd_init()
18:03:46|Rain|so the screen will always clear first
18:03:54|Rain|<tumu> jhMikeS, does the sim exit in any special way?
18:03:57|Rain|jhMikeS: do you have anything you'd like me to try before I go poking random pieces of the code with a stick?
18:05:14tumujhMikeS, just tested making hello.c and the .gcda is created on exit :)
18:05:18|Rain|boot flow is chksum_crc32gentab, system_init, kernel_init, lcd_init, font_init, button_init, lcd fg/bg/clear, THEN button_read_device (but verbose isn't set until after usb init)
18:05:40 Quit Wiwie (Read error: 110 (Connection timed out))
18:06:29jhMikeS|Rain|: Hmmm...but the COP is probably doing something
18:07:14jhMikeStumu: you just compile some with that option and you get the output? no other steps (just had a quick look)?
18:08:14|Rain|temu: wrt to profiling data not showing up, I believe that it's only saved on a clean (zero) exit, and that an atexit() (which SDL does to clean up, I think) might mess with it
18:08:19|Rain|tumu: ^
18:08:31|Rain|but it's been a while since I've used gcc's profiling stuff
18:09:12tumujhMikeS, just compile with that option, no other changes are needed
18:10:16jhMikeStumu: actually now, all threads exit the sim cleanly by returning normally. before they were just ungracefully terminated.
18:11:41 Quit barrywardell ()
18:12:03jhMikeSuse of setjmp/longjmp is part of normal operation as well when shutting down. I don't know if it botches that up.
18:12:19tumu|Rain|, hmm, it is possible if sim does not exit normally as gcc would not know how to patch any special ways
18:12:53tumualso if the app crashes or is killed, then i would assume that the data is not created either
18:12:56 Join craver [0] (
18:13:08craverhow do i use the shortucts thingy? its undocumented
18:15:53craverand how can i see what was changed in the current release / be notified for updates?
18:16:33tumujhMikeS, is it possible to make the sim exit in more normal way?
18:16:57tumuif not for other reason than to see if the datafiles are created
18:18:17jhMikeStumu: not the threads, no. not letting them return normally causes lockups when using SDL_Cond variables.
18:19:06tumugcc probably only patches the main process
18:19:10jhMikeSbasically the longjmp hops the scrambled up threads back to their functions for a graceful exit
18:19:12tumueg. main()
18:20:19tumuor is main process/loop run as a thread as well?
18:20:24*Nico_P brought his gigabeat S back to life :D
18:20:35Nico_Ptoffe82: here ?
18:20:36jhMikeSthe longjmp could remove context needed by the profiling
18:20:56toffe82Nico_P: yes
18:21:12Nico_Pthe problem was indeed a bad HDD connection
18:21:45jhMikeStumu: there's the rockbox threads which are like threads on target. there's a main thread and an i/o thread as well.
18:22:11toffe82Nico_P: sorry for that, next time I will try to do better ;)
18:22:21Nico_Ptoffe82: don't worry about it ;)
18:22:28 Quit agm3nt ()
18:22:32jhMikeSthe main thread handles input, the I/O thread handles sim disk access so that the rockbox threads may yield while I/O happens to be more like the real device.
18:23:14 Quit craver ()
18:23:58toffe82Nico_P: I will ask to my 6 years son to take care of what he does when he rebuild a player, you can never rely on somebody :)
18:24:18Nico_Phehe :)
18:26:59toffe82I made some change on calculator plugin but I don't have time to test for all the target and to make the patch, Idid it for the gigabeat F/X
18:27:08|Rain|debugging the bootloader is a pain :/
18:27:14jhMikeS|Rain|: Will only rockbox fail to boot or retailos too?
18:27:15toffe82somebody want the c file to test it ?
18:27:47Nico_Ptoffe82: I wouldn't mind
18:28:01sitwon|Rain|: have you been following the posts on the forums?
18:28:02|Rain|jhMikeS: OF won't work either, it doesn't make it far enough to even check whether it should boot the OF
18:28:05preglowthis piezo driver needs some work
18:28:09|Rain|sitwon: nyet
18:28:23jhMikeSThe COP should just be stuck waiting until it's decided which FW will run. hmmm.
18:29:18|Rain|jhMikeS: it freezes with the sandisk logo still on the screen, so it's either freezing in chksum_crc32gentab, system_init, or kernel_init
18:29:31low_light|Rain|: you can use the buttonlight for a visual clue
18:29:42toffe82Nico_P: it is mainly cosmetic, I change the parameter for the display so it should adapt automatically to the size of the display
18:30:00low_light|Rain|: you can just it the right GPIO bit
18:30:01|Rain|low_light: that's a good idea. way less work than setting up the lcd
18:30:35*jhMikeS uses buttonlight blinks all the time
18:30:38toffe82Nico_P: also I simplify the button so you use it only with the cross on the gigabeat
18:31:58jhMikeS|Rain|: Whether it booted or not depended strictly upon the bootloader version?
18:31:59low_light|Rain|: right, and you can use it within crt0-pp-bl.S too
18:32:55|Rain|jhMikeS: yes. the only thing I've changed is bootloader, and r14878 boots, r14880 freezes with the sandisk logo still on the screen
18:34:10jhMikeS|Rain|: Make sure the COP is staying put. The problem might be something stupid during kernel init for CPU if it does.
18:35:57 Join JustJohnny [0] (
18:36:33jhMikeSI'd blink the light right after system_init and/or kernel_init
18:37:10|Rain|yeah, I'm reading through the diff right now, and that's my first step after I'm done
18:37:11toffe82Nico_P: also it doesn't use the system font , it is a modification I want to do in case that your actual font is to big
18:37:12jhMikeSah bugger, I think I see it
18:37:46Nico_Ptoffe82: it failed... maybe pastebin ?
18:38:01toffe82email ?
18:38:02jhMikeS|Rain|: Init thread.c. Comment out the COP_CTL/CPU_CTL in the if (core == CPU) clause
18:38:12Nico_P...or email
18:38:18 Join kfazz [0] (
18:38:24jhMikeS|Rain|: Line 1090
18:40:21 Quit moos ("Glory to Rockbox")
18:41:19 Quit Lear ("ChatZilla [Firefox]")
18:41:26 Join Frazz [0] (
18:43:07tumu|Rain|, sdl discourages using atexit() for shutting down sdl :)
18:43:08jhMikeSThe CPU is waiting for the COP to finish init there but of couse it doesn't in the BL. :p
18:43:23tumu|Rain|, now that i checked sdl docs
18:44:11Nico_Ptoffe82: thanks
18:47:22|Rain|tumu: ah, maybe that's what I was remembering
18:48:02tumu|Rain|, guess what sim does :)
18:49:11jhMikeStumu: where does it do that?
18:49:28tumujhMikeS, in gui_startup()
18:50:26 Quit jhulst (Remote closed the connection)
18:51:22jhMikeS|Rain|: that worked?
18:51:46tumudecided to bite the bullet and rewrite the shutdown sequence in sim, and already noticed one bug in freeing resources in error case
18:52:07TMMOK, time to go
18:52:09TMMlater people
18:52:22tumusim_io_shutdown is not called if gui_startup or thread_sdl_init bombs
18:52:23TMMnot that I said much, apart from this :P
18:52:26 Quit TMM ("Ik ga weg")
18:53:25|Rain|jhMikeS: signs point to no, and I messed up my blinkyness, so I don't know where it died
18:53:28jhMikeStumu: well, I think it will just terminate all that with the process itself, kill threads, free process memory, detroy objects, etc.
18:54:25tumujhMikeS, yeah, atexit(sdl_quit) probably would, but i removed the atexit madness
18:54:43JustJohnnyHi Nico_P, just wanted to let you know I submitted a patch for the Rockboxed theme which adds support for the Sansa c200. The flyspray task id is 7855.
18:55:33Nico_PJustJohnny: do you have a screendump of the result ?
18:55:37jhMikeS|Rain|: Now I have a hard time believing removing lines 1090 and 1091 wouldn't fix the problem. It actually can't work with those enabled in the bootloader.
18:55:52pixelmaJustJohnny: did you correct the backdrop dimensions?
18:56:34|Rain| 1090 // COP_CTL = PROC_WAKE;
18:56:34|Rain| 1091 /* Sleep until finished */
18:56:34|Rain| 1092 // CPU_CTL = PROC_SLEEP;
18:57:05JustJohnnythe dimensions that worked for the simulater where 130x80. I think in the summary of the patch I put 130x90, but the attachment is correct.
18:57:21preglowpondlife: seems to be because it has two entries in settings_list.c
18:57:30pixelmathe c200's screen is 132x80
18:58:13preglowahh, no, not two entries
18:58:17*preglow doesn't get this macro system
18:58:41JustJohnnyHmm, okay, maybe my renaming is completely off. I must have resized to 132x80, since it is working on the simulator. Let me go check an rename everything correctly.
18:59:49pixelmaJustJohnny: I also just remembered that bluebrother started to alter the greyscale versions and asked you about it in the forums...
19:03:12 Join Rincewind [0] (
19:03:34 Quit maddler (Read error: 104 (Connection reset by peer))
19:03:47RincewindI just got a sansa e280
19:04:11RincewindI can't find the list scroll acceleration settings. Have they been removed?
19:06:50|Rain|so I fixed the code to blink the wheel light, and the additional delay during bootup seems to have made it actually boot
19:07:34 Join maddler [0] (
19:07:35jhMikeS|Rain|: What's the quickest way to get the BL back on there after it won't start? I'm tired and don't feel like thinking right now.
19:08:18|Rain|jhMikeS: I've been entering recovery mode, mount /dev/sda /mnt && cp stock.mi4 /mnt && umount /mnt, reboot, run sansapatcher, reboot
19:08:24|Rain|...which isn't quick by a longshot
19:08:59jhMikeSpretty much what I feared...eck
19:10:07 Nick fxb is now known as fxb__ (
19:10:11 Quit freqmod (Read error: 104 (Connection reset by peer))
19:10:13***Saving seen data "./dancer.seen"
19:10:55 Join MikhailTkach [0] (
19:15:48 Join scorche|work [0] (n=42c007b2@rockbox/administrator/scorche)
19:17:43jhMikeS|Rain|: I just made the very change you said you made and it booted fine
19:18:19|Rain|I'm starting to think something didn't go right before, since I've since ripped out almost all my debug code and it's working
19:19:04|Rain|I don't know how that would've happened, though, unless ccache is playing tricks on me
19:19:15pondlifepreglow: I saw that. Not sure if the second one can just be removed, or if it's in there with UNUSED for a good reason. Also not sure about JdGordon's marvellous macros :)
19:19:32|Rain|I just ripped out the very last of my debug code and it's still booting for me :/
19:21:09 Join Llorean [0] (
19:21:15sitwon|Rain|: problem fixed? or problem went away with no apparent explaination?
19:21:30 Join obo [0] (n=obo@rockbox/developer/obo)
19:21:36|Rain|jhMikeS provided a fix
19:21:44|Rain|it just didn't appear to work for me the first compile for some reason
19:22:11jhMikeSI'll just commit that since the BL _definitely_ won't work with that code enabled for it in any case.
19:22:23 Part pondlife ("disconnected has pondlife")
19:23:03JustJohnnyNico_P & Pixelma: I've fixed the code to reflect the correct resolution. Uploaded the renamed background file, the patch with the fixes in the WPSLIST file and renamed wps file, and a screen dump to flyspray:
19:23:41sitwonso will that fix the current freezes-before-loading-anything and the won't-boot-OF problems?
19:24:04|Rain|well, at least you found it, even if I messed up the upload somehow
19:24:27|Rain|sitwon: it fixes it for me
19:24:36jhMikeSok, committed
19:25:14jhMikeSit won't fix the other trouble that's SD related
19:25:23jhMikeSSDHC more accurately
19:25:58*jhMikeS should just make the build table his default homepage
19:26:03sitwonI'm going to try the anti-SDHC patch again then
19:26:17|Rain|what issue is that? does it refuse to boot with a SDHC card in?
19:26:35jhMikeS|Rain|: For some since the SDHC support was added
19:27:38sitwon|Rain|: the issue I've been having is it won't boot the OF (regardless of card inserted or not)
19:28:04|Rain|sitwon: I had one of those unrelated to the issue that jhMikeS just fixed
19:28:33|Rain|I have a patch for that, but I haven't had a chance to put it in the tracker yet
19:29:06sitwonOn the forums Bagder told me to try your patch, but it froze at the Sansa screen (which is what I believe he just fixed) so I'm going to try your patch again
19:29:18 Join kubiix [0] (
19:29:48 Quit JustJohnny (Remote closed the connection)
19:30:45jhMikeSthere's always those little ticks with those major changes...just some dumb mistake. heh.
19:32:46 Join agm3nt [0] (
19:32:53 Join justjohnny [0] (
19:33:38 Join My_Sic [0] (
19:33:42RincewindI have a huge problem with my sansa! I can't get into the original firmware!
19:34:09sitwon|Rain|: looks to me like your patch didn't work. It's doing that fade-to-nothing thing agian when I try to boot OF
19:34:25jhMikeSI think I'll probably just make NUM_CORES = 1 for a bootloader build. There's no real point to having the scheduler for dual there. Later though.
19:34:32Rincewindwhen I press LEFT during startup, the bootloader displays some text and then the screen slowly turns white and fades
19:34:56sitwonRincewind: are you using e200 or e200R?
19:35:15sitwoncheck the forums for a link to sansapatcher v0.2
19:35:46Rincewindthe problem is, how can I put it on? If I can't get into the firmware I don't have USB
19:36:05sitwonRincewind: forum link:
19:36:18RincewindI downloaded the sansapatch for linux that was linked in the manual
19:36:34sitwoncheck this page too
19:37:09sitwonapparently there's a problem with the latest version (0.3 I think) the previous version works fine though
19:38:36|Rain|sitwon: which patch are you using, and does rockbox boot fine (while the OF doesn't)?
19:38:48*jhMikeS wonders why sansapatcher can't work in recovery mode
19:38:52Rincewindsitwon: thanks, I'll try that
19:39:26sitwon|Rain|: Yea, rockbox boots, but not OF. This is the patch:
19:39:28|Rain|jhMikeS: it'd have to decrypt the .mi4 before patching it, methinks
19:40:03|Rain|sitwon: weird, I didn't have that problem...
19:40:13jhMikeSit accesses the hidden partition anyway and does just that. besides all the code to decrypt it is there anyway.
19:40:33|Rain|reading it now
19:40:54sitwon|Rain|: I'm using an e200R, it was assumed that the problem was related to the one with the e200's not booting OF
19:40:56|Rain|jhMikeS: ah, didn't know whether it was encrypted on the hidden partition or not
19:41:16|Rain|sitwon: ah... maybe. I have a non-R
19:41:33jhMikeSthe mi4 bootloader sure is
19:41:55 Join linuxstb_ [0] (
19:42:44tumujhMikeS, hm, only problem i seem to have now is gcc is linking the older libgcov from mingw rather than from cygwin
19:43:08|Rain|sitwon: have you tried stepping backwards through svn revisions to find which one breaks it? (not at all convenient, but doing a binary search between 2 points might keep you from going mad)
19:43:28linuxstb_jhMikeS: sansapatcher modifies the firmware partition - which IIUC isn't the same as the FAT-formatted partition available in recovery mode.
19:44:17jhMikeSlinuxstb_: so only the one partition is available in recovery mode?
19:44:33linuxstb_I would guess the recovery partition is just a ramdisk - temporarily used to copy files to, which are then transfered to the firmware partition.
19:44:59tumujhMikeS, renamed the older libgcov to get that error
19:44:59linuxstb_jhMikeS: Yes, afaik (I've never touched as Sansa).
19:45:13 Quit gregj ("brb")
19:46:28Nico_Plinuxstb_: what's the status on your gigabeat S ?
19:46:50linuxstb_Still waiting for it to arrive in the UK.
19:46:56tumujhMikeS, any quick hint how i need to change the sim makefile to make it use cygwin instead?
19:47:14jhMikeStumu: I can't really say why that problem should occur.
19:47:19Rincewindsomeone should link the old sansapatcher in the install instructions for sansa...
19:47:35pixelmajustjohnny: hmm... I only see the new background bitmap in your player entry - not the other icons?
19:47:49pixelmas/player/tracker heh
19:47:58tumujhMikeS, well, i guess something is redefining the library paths for sim build
19:48:49jhMikeStumu: I don't think it's the makefile.
19:48:54tumuthe wiki doesn't tell who is responsible for sim build
19:49:27tumuso i have no idea who would know the sim build process
19:50:03pixelmajustjohnny: or maybe you reuse the small H10 ones...
19:50:19justjohnnypixelma: the other icons are reused from previous smaill color icons.
19:50:35jhMikeSit should use whatever lib paths cygwin has set. the makefile doesn't define that. (looking at it now)
19:51:00*bluebrother notices justjohnny
19:51:43justjohnnyPixelma: yes, I'm not sure how the build system knows which images to use. I'm guessing it parses the wps for the correct files
19:51:43 Join hcs [0] (n=agashlin@rockbox/contributor/hcs)
19:52:09justjohnnyhi bluebrother
19:52:27 Join jhulst [0] (n=jhulst@unaffiliated/jhulst)
19:54:08 Join freqmod [0] (
19:54:35bluebrotherjustjohnny: did I miss your response upon my greyscale work on Rockboxed the days I did it? I remember I never committed it
19:55:42bluebrotheror, to go to the question directly: do you think it's committable?
19:55:47justjohnnybluebrother: At this point I don't remember if I responded or not.
19:55:50 Join jrsharp [0] (
19:56:09 Nick pilled is now known as pill (
19:56:13 Part Llorean
19:56:20bluebrotherthe archive is still around if you want to look at it again:
19:56:35justjohnnybluebrother: thanks, I'll take a look at it
19:56:54bluebrothermaybe it's a good chance to add this while speaking of it?
19:58:40justjohnnybluebrother: i've downloaded it: brb
19:58:53bluebrothertake your time ;-)
20:00:13Rincewinddoes anyone know if I can brick my sansa e280 permanentely if I compile my own sansapatcher?
20:00:36 Join advcomp2019_ [0] (n=advcomp2@
20:00:50 Quit advcomp2019 (Nick collision from services.)
20:00:54 Nick advcomp2019_ is now known as advcomp2019 (n=advcomp2@
20:01:52 Join iamben [0] (
20:02:14 Join HD2 [0] (
20:02:48 Part jrsharp
20:03:18 Quit HardDisk_WP (Nick collision from services.)
20:03:18 Quit Xerion (Read error: 104 (Connection reset by peer))
20:03:30 Nick HD2 is now known as HardDisk_WP (
20:03:54 Join Redbreva [0] (
20:03:57 Join Xerion [0] (
20:04:24pixelmajustjohnny: would be easier if I could just unzip the new version if I wanted to try on target, now I would have to compile it first
20:05:21pixelmaI'm a bit afraid of that orange on the c200's display though ;)
20:05:36bluebrotherpixelma: you should be able to just run wps/
20:07:14Rincewindcan someone send me linux sansapatcher v0.2?
20:08:49linuxstb_Rincewind: It's on the download server -
20:09:26sitwon|Rain|: 14423 works, 14424 fails
20:11:15Rincewindlinuxstb_: the problem is, this is the version that doesn't boot into the original firmware
20:11:37linuxstb_Rincewind: How about the versions under "old" ?
20:11:41 Join massive_H [0] (
20:12:04sitwonRincewind: it was in that forum I sent you the link to
20:12:11Rincewindstrange, I thought I was looking there and got a 404
20:13:56Rincewindthe links in the forum thread are not up to date
20:14:25linuxstb_Rincewind: Just look where I linked you to....
20:15:32Rincewindlinuxstb_, yes, thanks, I got it now
20:15:47justjohnnybluebrother: I like it! Having grayscale icons, instead of mono icons for the platforms that support grayscale is great. Especially since it looks like you made all the necessary changes :)
20:17:20 Quit hcs (Remote closed the connection)
20:17:33sitwon|Rain|: your patch undoes the change in 14424 doesn't it?
20:18:06 Join donsdw [0] (
20:19:16justjohnnypixelma: I'll zip it up. Should I include it in the flayspray task, or put it elsewhere? −− Everyone's afraid of the orange :>
20:20:46 Join hcs [0] (n=agashlin@rockbox/contributor/hcs)
20:20:48Rincewindthe only thing is, now I get the corrupted first line in the LCD, even in the original firmware
20:21:56sitwonRincewind: you'll have to compile your own like I did
20:22:01jhMikeSRincewind: if you used sansapatcher v0.2 then it doesn't contain the lcd line fix
20:24:17Rincewindhm, so I have to compile my own sansapatcher wich has the lcd line fix but still boots into OF
20:24:55linuxstb_sansapatcher can install any bootloader, not just the one embedded in it - use "sansapatcher -a pp5022.mi4"
20:25:15sitwonRincewind: can you accept DCC transfers?
20:25:37Rincewindsitwon: I don't know, I'm in the university
20:25:56sitwongive it a try
20:26:05sitwonotherwise i'll host it on my FTP for you
20:26:40Rincewinddcc doesn't seem to work
20:26:48|Rain|sitwon: nope
20:26:58Rincewindsitwon: did you just compile the latest bootloader?
20:28:46sitwonRincewind: I compiled a e200 bootloader from the same revisions that worked for my e200R
20:29:32Rincewindwhat is the difference between bootloader and installer in the configure script for sansa?
20:29:32 Quit massiveH (Read error: 110 (Connection timed out))
20:31:05 Quit Genre9mp3 ()
20:31:06sitwonRincewind: download the mi4 file, put it in your players root directory, restart into the original firmware and let it update. it should fix the LCD.
20:31:46justjohnnypixelma: I've added a zip of the complete them to the flyspray task
20:32:01Rincewindsitwon, thanks. The thing is, I still would like to know how to do it myself
20:33:02 Quit jhulst (Remote closed the connection)
20:33:04linuxstb_sitwon: The recommended way to install a bootloader is with sansapatcher, not the normal firmware upgrade.
20:33:55linuxstb_(sansapatcher keeps a copy of the OF in the firmware partition, to enable dual-booting)
20:33:56Rincewindok, I compiled my own bootloader. Any chance to brick my player with that?
20:33:57sitwon:-/ my bad. I assumed it would work the same on the e200 as on the e200R
20:34:44linuxstb_Rincewind: No - you can always use recovery mode (or at worst, manufacturing mode) to restore. But install that file with "sansapatcher -a PP5022.mi4"
20:34:48sitwonRincewind: ofcourse there's a chance. but it's unlikely
20:35:26RincewindI am just a little cautious, because on the iriver it is said to _never_ install your own bootloader
20:35:32RincewindI try it now
20:37:57 Quit sgsax (Remote closed the connection)
20:38:42 Part agm3nt
20:39:14Rincewinddidn't work. It just sits there with the Sandisk logo
20:40:52 Join sgsax [0] (
20:41:05pixelmajustjohnny: thanks, trying atm on target
20:43:53sitwonRincewind: try using an older revision of the code
20:44:14Rincewindsitwon, my own bootloader didn't work. I tried yours and this is fine. Thanks for you help!
20:44:28 Join smably [0] (i=d809f36d@gateway/web/cgi-irc/
20:45:18bluebrotherjustjohnny: just returned from dinner ... ok, I'll give it a go and commit the greyscale stuff later
20:45:18RincewindI've had enough bricking today to try more things today ^^
20:47:28smablyhi all −− sorry if this has already been reported, but the gradient commit seems to be somewhat broken...
20:47:50smablythe menu looks okay, but i see this in the file list: smably/dump070927-144529.jpg">
20:48:15smablycompare to this, in the menu: smably/dump070927-144548.jpg">
20:48:36smablyand scrolling is reallllly slow in the filetree
20:49:06 Quit freqmod (Remote closed the connection)
20:50:17 Join rogelio [0] (n=rogelio@
20:50:49 Quit Arathis ("Bye, bye")
20:51:03 Part rogelio ("Kopete 0.12.4 :")
20:53:59smablyand here is my cfg file in case anyone wants to try to reproduce it:
20:54:12Nico_Psmably: thanks, I'll have a look
20:55:10Nico_Psmably: you've noticed a difference in scolling introduced by the gradient ?
20:56:46smablyNico_P: yeah, but only in conjunction with the other issue (the large white square below the files, as seen in the first screen dump)
20:56:51 Quit XavierGr (Nick collision from services.)
20:56:51 Join XavierGr_ [0] (
20:57:06Nico_Pyeah, that one is very strange
20:58:18Nico_Psmably: the screensump where there is a large block doesn't have a gradient... is he gradient setting activated ?
20:58:41smablyyep, all the settings are exactly the same in the two screendumps
20:58:51Nico_PI'll try your config then
20:58:59smablyk, thanks :)
20:59:48 Join webguest78 [0] (i=5b698a66@gateway/web/cgi-irc/
20:59:51 Join rogelio [0] (n=rogelio@
21:00:14Nico_Psmably: you're not using an unofficial build, are you ?
21:00:41smablyNico_P: nope, just grabbed the latest one
21:01:00Nico_Psmably: the sim doesn't show the problem
21:01:15Nico_Psmably: does this happen in all dirs ?
21:01:21smablyhmm, odd
21:01:25smablyyeah, i think so
21:01:46 Join hannesd_ [0] (
21:02:17smablylet me just have a look...
21:02:46pixelmabluebrother: for the greyscale rockboxed, would you need to change something in WPSLIST?
21:03:38 Part rogelio ("Kopete 0.12.4 :")
21:04:31bluebrotherpixelma: no
21:06:03 Quit hannesd (Read error: 145 (Connection timed out))
21:06:03 Nick hannesd_ is now known as hannesd (
21:07:39Nico_Psmably: what happens if you change the colours slightly ?
21:08:12pixelmabluebrother: so it won't interfere if I committed the c200 version while I'm at it...
21:08:57bluebrotherno, it shouldn't.
21:09:08smablyNico_P: yeah, seems to happen in every dir, but not in the menu or the database (i'll change the colours in a sec)
21:09:10pixelmaor did someone want to? Nico_P?
21:09:20bluebrotherbut feel free to commit the greyscale stuff the same time if you want to ;-)
21:09:50pixelmanah, it's your work :)
21:09:59bluebrotherok ...
21:10:14***Saving seen data "./dancer.seen"
21:10:27Nico_Ppixelma: go ahead :)
21:10:54 Quit przemhb ()
21:13:07smablyNico_P: the block seems to match the primary gradient colour...
21:16:00Nico_Psmably: would you mind trying it again with a clean config ?
21:16:32smablyNico_P: yep, already tried that, and it works correctly
21:16:43Nico_Pso it seems to be your config
21:16:47Nico_PI'll try it on my gigabeat
21:17:18Nico_Pit's funny the problem doesn't appear on the sim though
21:17:42webguest78I wondered if this <a href=" ">Stardom Staray S2-U Biometric Finger Print Reader 2.5" IDE Caddy </a> is as secure as suggested, for example could the hard drive simply be removed from the caddy to allow unauthorized access to data?
21:18:12smablyyeah −− also strange because i tried that patch just before it was committed and it worked fine
21:19:04smably(not sure i was using the most recent version of it, though)
21:19:10Nico_Psmably: no problem on my gigabeat
21:19:27Nico_Psmably: do you have the absolute most recent build ?
21:20:20smablyNico_P: i think so, but let me try grabbing it again and doing a clean install...
21:21:14Nico_Pkeep your config though ;)
21:22:43 Part justjohnny
21:23:58 Join lee-qid [0] (
21:24:05Nico_Pbluebrother: "Unless you changed this installation will fail!"... what's this string about in rbutilqt ?
21:26:28bluebrotherNico_P: it tells to user to put the player in UMS mode
21:26:51bluebrotherit's triggered when a player in MTP mode is detected (via usb id)
21:27:35smablyNico_P: aha! i think i have identified the problem
21:27:35Nico_Pbluebrother: ah, I understand better
21:27:49Nico_Pbluebrother: and what is MSC mode for a sansa, as opposed to MTP ?
21:27:49smablyNico_P: it is that i have a .colours file
21:28:12Nico_Psmably: ah, then I'd like it :)
21:28:18bluebrotherMSC is the same as UMS, it's just called different on the sansa
21:28:19Spyro_boyI totally bricked my sansa
21:28:21Spyro_boyanyone know about e200tool?
21:28:39Spyro_boyit says "searching for device" but it doesn't find anything.
21:29:14Spyro_boyBus 002 Device 004: ID 0781:0720 SanDisk Corp.
21:29:18Spyro_boyIt's there though.
21:29:30Nico_Pbluebrother: ok, thanks
21:30:54 Quit MikhailTkach ("Leaving")
21:30:59sitwonSpyro_boy: what kind of sansa do you have? can you put it in Recovery or Manufacture mode?
21:31:04smablyNico_P: seems to happen with the included sample.colours file
21:31:06SoapSpyro_boy: ontopic is offtopic for this channel
21:31:28Spyro_boysitwon, I can't get into recovery mode.
21:31:29Soapoops, wrong channel ;)
21:31:51Spyro_boyNot sure what manufacture mode is.
21:32:23bluebrotherSpyro_boy: already read the wiki?
21:32:28Spyro_boyI'm there.
21:32:40Spyro_boyand here
21:33:47 Quit pill (Nick collision from services.)
21:33:50Spyro_boyhow do I get into manufacture more?
21:33:54Spyro_boyIt doesn't say.
21:34:00 Join pill [0] (
21:34:05Spyro_boyJust says that I have to run e200tool after I get into it.
21:34:12 Quit pill (Nick collision from services.)
21:34:29 Join _pill [0] (
21:34:31Spyro_boyokay, thanks.
21:34:35bluebrother(it's a bit hidden but afaik it's completely identical to the E200)
21:34:46Nico_Psmably: included where ?
21:34:52bluebrotherhad to search for it too ;-)
21:35:46smablyNico_P: .rockbox/docs/sample.colours
21:36:12Spyro_boybluebrother, this wiki page seems a lot more detailed.
21:36:20webguest78Any advise please on biometric hard drive caddy’s, are they as secure as suggested, for example could the hard drive simply be removed from the caddy to allow unauthorized access to data?
21:36:37bluebrotherSpyro_boy: that page is about something different, so don't try to do it on a vanilla e200!
21:36:39Nico_Psmably: thanks... didn't even know it existed :p
21:36:49bluebrother(unless you already noticed that ;-)
21:37:02smablyNico_P: hehe, no problem :)
21:37:43Spyro_boyI think I know why e200tool was failing before.
21:37:55Spyro_boyMaybe needed root perms, but I'll just continue with this guide.
21:37:57smablyNico_P: pretty sure the problem just occurs whenever any coloured entry is highlighted; i.e., the gradient filler is expecting the text to be the foreground colour
21:38:18Spyro_boybluebrother, I dd'd the entire e200.
21:38:26Spyro_boyI believe it's far from vaniella.
21:38:45bluebrotherif it isn't an e200r we call it vanilla
21:38:53Spyro_boyit's not r
21:38:56donsdwIs this a proper channel for asking non-technical questions?
21:38:57bluebrother(i.e. it's a e200, not an r)
21:39:05Spyro_boyI don't think so, the packaging doesn't say anything about e200R.
21:39:08Spyro_boyor e280R.
21:39:22Spyro_boyor raphsody.
21:39:29Nico_Psmably: reproduced !
21:39:29Spyro_boy(however you spell it)
21:39:45smablyall right! :)
21:39:46Nico_Psmably: you were right, it happens when highlighting a coloured entry
21:39:53 Join Lars_G [0] (n=lars@unaffiliated/lars-g/x-000001)
21:40:04Spyro_boyIt worked!
21:40:06donsdwIs it reasonalbly safe for a non-techie to play around with rockbox on a sansa e200?
21:40:09Spyro_boythanks for the link again!
21:40:10Lars_GSay, what tool would you guys recommend if I wanted to move stuff around the disk/ipod based on id3?
21:40:55scorche|workLars_G: that is a bit offtopic...
21:41:12sitwondonsdw: that depends on what you consider 'reasonable'...
21:41:33Lars_Gscorche|work: slightly. but I thought this would be the best place to find good help on this... and it's far fetched but I do want it to use with rockbox anyhow ;)
21:41:41sitwondonsdw: you should be able to install and use rockbox on an e200 with no danger and/or problems
21:42:09Lars_Gscorche|work: I'm moving my last ipod experience from of to rockbox, which is podcasts and some leftover music. so I want to rearrange stuff already on my ipod out of the itunes dir
21:42:52 Join theclaw [0] (n=theclaw@unaffiliated/theclaw)
21:43:12theclawwhat is the difference between "Player" and "Recorder"?
21:43:21theclaw(as seen in Q1 of BATTERY-FAQ)
21:43:46scorche|worktheclaw: they are different archos models
21:44:02Nico_Psmably: problem solved
21:44:04theclawoh, so this faq is mainly related to archos players
21:44:25scorche|workif you are talking about the one i am thinking about, then yes
21:44:26 Quit webguest78 ("CGI:IRC (EOF)")
21:44:34smablyNico_P: great, thanks for fixing it, and helping debug! :)
21:44:40 Join webguest78 [0] (i=5b698a66@gateway/web/cgi-irc/
21:44:49Nico_Psmably: well thank you... you did all the work
21:45:06 Quit webguest78 (Client Quit)
21:45:41 Join barrywardell [0] (
21:45:52 Quit donsdw ()
21:47:07theclawscorche|work: the one found in the source package of rockbox
21:49:46 Quit massive_H (Read error: 104 (Connection reset by peer))
21:49:51 Join massiveH [0] (
21:51:34theclawmight rockbox be working on newer ipods in the near future? (if not, (why) is it harder to port it to the newer ipods than to "port" it to the older ones?)
21:52:09bluebrotherits hard because those have completely new hardware and are completely encryped.
21:52:17*pixelma confused about the build system but oh well...
21:53:04 Join amiconn [0] (i=51df58a3@rockbox/developer/amiconn)
21:54:00Spyro_boyafter I get it into recovery mode, I can't figure out what to do.
21:54:04theclawIs it the "trend" today that new portable media players (or gadgets in general) get more closed?
21:54:15Spyro_boyI still can't get it to boot into recovery mode without e200tool
21:54:44Spyro_boyanyone mind pointing me in a direction?
21:54:58bluebrotheryes. For the h100 / h300 series almost all datasheets are available. Not for Ipods and other players that came later.
21:55:44sitwonSpyro_boy: once it's in Recovery Mode, you need to mount it and copy a working firmware to it, then unmount and unplug
21:56:49Spyro_boythat's it?
21:56:54Spyro_boyI thought I did that.
21:56:56Spyro_boyI'll try again.
21:58:07 Join sarixe [0] (
21:58:09 Quit nicktastic ("Leaving")
21:59:50Spyro_boysitwon, it says, main image write failed.
21:59:59Spyro_boyis it because the partitioning might be fucked?
22:00:09 Quit smably ("CGI:IRC 0.5.7 (2005/06/19)")
22:01:08sitwondid you mess with the partitioning?
22:01:16Spyro_boyI did.
22:01:47sitwonyea, that's probably the problem
22:01:49Spyro_boyUh, well, I imaged my last sansa, but then I got a new one.
22:02:03Spyro_boyso I tried to dd the partition with the image
22:02:09Spyro_boyand then that screwed up the partition table
22:02:13Spyro_boyso I dd'd the entire thing.
22:02:41Spyro_boyany way to fix it?
22:02:45Bagderso put a file named 'sansa.fmt' on the recovery disk
22:02:54Bagderit'll reformat your device
22:03:01Spyro_boywhere can I get it?
22:03:12Spyro_boyor is it just an empty file that triggers the reformat?
22:03:15Bagdermake one up, the file name is the magic
22:03:22Spyro_boyI'll try that.
22:03:26*bluebrother suggests touch
22:03:41*sitwon accuses bluebrother of sexual harassment
22:03:43Spyro_boyi was going to do that
22:03:57maxkelleywho was doing the PP502x dual-core patches?
22:04:06Spyro_boydual core?
22:04:11*bluebrother blushes and pulls out the man page
22:04:16scorche|worki believe jhMikeS was
22:04:24maxkelleythat's right..
22:04:37maxkelleyjhMikeS: ping. I'm able to test the dual-core patches for the c200.
22:05:21 Quit barrywardell ()
22:06:17maxkelleydoes anyone know if those are in an flyspray #?
22:06:20Spyro_boyBagder, should it give me notification that it's done formatting?
22:06:35BagderI don't know, I never did this
22:06:57Bagderbtw, this is in fact described on a sandisk page
22:07:17Spyro_boyI wouldn't have thought they had a forum. ;x
22:07:26Bagderhehe, I never saw this before either
22:07:50 Join doktoreas [0] (
22:07:55doktoreashello everybody
22:08:11doktoreasi have updated the rockbox firmware, and now when i plug it in the usb
22:08:12 Quit Nico_P (Remote closed the connection)
22:08:17doktoreasi don't see the usb image
22:08:18 Join fm2 [0] (i=53f23c72@gateway/web/cgi-irc/
22:08:24doktoreasbut the classic ipod "do not disconnect"
22:08:27doktoreasis it normal?
22:08:43bluebrotherRockbox uses Apples firmware for usb
22:08:57bluebrotherand you haven't told us what player :>
22:09:01 Join Nico_P [0] (n=nicolas@rockbox/developer/NicoP)
22:10:33maxkelleydoes the e200 have the same tuner as the c200?
22:11:20pixelmaIIRC yes, but it's hooked up differently
22:11:35doktoreasbluebrother, sorry :) apple mini 2nd generation
22:11:36fm2Hello. I need help in modifying of apps/Makefile. I want to do the following: (1) files are compiled as before but with slightly modified compiler options (I changed them already) (2) a script is executed (the script is at The script produces a .c file. (3) The produced .c file is compiled (4) further compiling/linking as before.
22:11:40 Quit kubiix ("Miranda IM! Smaller, Faster, Easier.")
22:12:05fm2bluebrother told me in general what I have to do but I can't seem to get through
22:12:06maxkelleypixelma: hmm. I'm going to work on the tuner portion for the c200. or attempt to.
22:12:14doktoreasbluebrother, the matter is that my usb audio player doesn't read the mp3 in the ipod
22:12:19doktoreason the previus version it worked
22:12:39Spyro_boyThere we go!
22:12:44Spyro_boyI got it to format! :D
22:12:47bluebrotherdoktoreas: well, then it looks like you have a different issue.
22:12:50pixelmamaxkelley: low_light should know more... ;)
22:13:04bluebrotherin some cases the Ipod doesn't reboot to disc mode properly. It's a known issue.
22:13:15maxkelleyhas he been around lately?
22:13:16bluebrotherit only happens with some builds.
22:13:34maxkelleyI haven't emailed him in a bit, and last time I talked to him, he was pretty busy with work.
22:13:36doktoreasbluebrother, then i am unlucky :D
22:13:40bluebrotherjust use emergency disc mode. Or try until you find a build that works ;-)
22:13:44theclawI wonder how the manufacturers benefit from locking alternative firmwares out?
22:13:58maxkelleytheclaw: erm.. not very well.
22:14:00bluebrotherdoktoreas: nah, not exactly. I'm experiencing this issue too.
22:14:24doktoreasbluebrother, how o you handle this?
22:14:40theclawmaxkelley: "not very well"?
22:14:49Bagdertheclaw: they save themselves from a very small amount of extra support
22:14:50bluebrotheruse emergency disc mode. Or bootloader usb mode (which isn't part of the official bootloader)
22:15:06 Quit sarixe ("Peace")
22:15:16Bagdermaxkelley: afaik, the assumption has been that they do use the same fm tuner
22:15:28theclawBagder: they could simply drop support for alternative firmwares?
22:15:48maxkelleyBagder: I thought that, because a good deal of the e200 is similar/same.
22:15:50Bagdertheclaw: they still get questions and RMAs and things that could be due to the alternative(s)
22:15:50Spyro_boyAwesome! Unbricked!
22:15:53Spyro_boyI think.
22:16:00Bagdermaxkelley: indeed
22:16:13doktoreasbluebrother, bootloader usb mode?
22:16:20pixelmamaxkelley: talked to him here a few hours ago and he'S still logged in
22:16:24fm2bluebrother: hello! Could you guide me? There are many if and such in the Makefile. I read up about the syntax but that were simple examples. The real life is more complex it seems :-/
22:16:37scorche|workSpyro_boy: your device never was bricked..please dont use that term unless it truly is bricked
22:16:39bluebrotherfm2: to add that generated file to the build you need to add it to the FILES file ...
22:16:40maxkelleypixelma: heh, I should've checked that first. I shall pm him.
22:16:55theclawmaxkelley: do you mean they don't lock alternative firmwares out very well?
22:16:56 Join jhulst [0] (n=jhulst@unaffiliated/jhulst)
22:17:14bluebrotherdoktoreas: it's a thing other players have. If you turn on the player with usb connected it goes directly to usb mode.
22:17:21maxkelleytheclaw: no, I mean that they don't really profit from locking it out.
22:17:33maxkelleyexcept for the support thing.
22:17:34theclawmaxkelley: oh, okay
22:17:41bluebrotherand there is a patch around that does this for the Ipods. It's rather slow but it works
22:17:46Spyro_boyscorche|work, thought it was.
22:17:51bluebrother(startup is slow, not usb mode itself)
22:17:58Spyro_boystill doesn't boot though.
22:18:06maxkelleyI remember hearing some reason for it, but forgot it.
22:18:27scorche|workSpyro_boy: if it was bricked, there would be nothing you could do to fix it unless you got out the soldering gun
22:18:31 Join Pyromancer [0] (
22:18:34Spyro_boyIt said it was finished formatting, so I waited for it to reset, turned it off, got back to into manufacture mode, added the firmware, and it still doesn't boot.
22:18:45Spyro_boyscorche|work, I guess, that's true.
22:18:47fm2bluebrother: that file is generated by the script. Are you sure something should get into FILES (which is in SVN) that is itself not in SVN?
22:19:04maxkelleytruly, a player isn't bricked until something actually affects the HW.
22:19:11 Quit Frazz ("Leaving")
22:19:18scorche|workSpyro_boy: it is true...that is the /actual definition/ of "bricked"
22:19:39bluebrotherfm2: well, unless that file is in the FILES file it won't get compiled.
22:19:48scorche|workmaxkelley: well, it doesnt have to affect the HW....see the iriver devices and invalid bootloaders...
22:19:50bluebrotherexcept if it's covered by some wildcard
22:20:12 Join northlandpharmac [0] (n=kvirc@
22:20:17 Quit northlandpharmac (Excess Flood)
22:20:19maxkelleyscorche|work: yeah, but wouldn't it still be able to somehow reflash the boot ROM, or whatever?
22:20:32 Join northlandpharmac [0] (n=kvirc@
22:20:34 Join Crash91 [0] (n=rt@
22:20:45 Quit northlandpharmac (Excess Flood)
22:20:48scorche|workmaxkelley not without special equipment
22:20:57maxkelleyyeah, I guess so.
22:21:02 Join |northlandpharma [0] (n=kvirc@
22:21:19tumuhm, someone should take a look why the sim doesn't build against cygwin
22:21:26Lars_Gand even with special equipment you'd have to pray there's some testpoints or a jtag you can use on the board
22:21:29fm2bluebrother: ...or mabe expicitly compiled by the Makefile. I think the knowledge about that file should be at one place (=Makefile)
22:21:33 Quit |northlandpharma (Excess Flood)
22:21:40maxkelleyLars_G: how does the factory program it?
22:21:42tumuthe usage of headers seem to conflict
22:21:45Lars_Gotherwise you'd probably have to even desolder the flash chip to socket it.
22:21:50 Join |northlandpharma [0] (n=kvirc@
22:22:03 Quit |northlandpharma (Excess Flood)
22:22:08maxkelleywell, yeah, that would be an option for the factory, too.
22:22:20scorche|workLars_G: well, LinusN can fix them....which is why he is the one that tests the iriver bootloaders =)
22:22:39Lars_Gmaxkelley: It depends on the factory. but some times either the assembler, or the flash maker have huge programming banks where the chips are preprog before being used.
22:22:54 Quit doktoreas ("Leaving")
22:23:03Lars_Gother times, the devices do have jtag or programming points in the pcb, and needled machines are applied to them after basic construction
22:23:11Lars_Gand before body assembling.
22:23:19Spyro_boyyay. I finally got it.
22:23:41Lars_Galso know as In Circuit Programming
22:23:54Crash91guys how can i convert a hex to be used in rockbox?? like change it to values for the color selector
22:24:01Lars_Gbut even if the board has test points for that, normaly theyre just flat uncoated copper circles.
22:24:10low_lightmaxkelley: Feel free to work on the tuner. The chip is the same as in the e200 (lv24020). You can enable it in config-c200.h. When I tried it, it compiled fine, but did not seem to work. I haven't tried anything else.
22:24:14maxkelleywhich can be soldered to.
22:24:23Lars_Gthey're connected to a computer with something like a press in the factory that has nedles in the right places.
22:24:25maxkelleylow_light: I just enabled it, and will try and work from there.
22:24:29Lars_Gmaxkelley: most cases, but it's a royal pita
22:24:41maxkelleyyeah :)
22:25:05 Quit Rob2222 (Read error: 104 (Connection reset by peer))
22:25:07maxkelleyeven if a pita fix, it's not really a bricked player until someone takes a sledgehammer to it :)
22:25:15Lars_GAt least most devices are designed to be reflashable, so they chose serially programable flash chips
22:25:21Lars_Gotherwise it would be even worse
22:25:32Lars_Gmaxkelley: Yup I agree.
22:25:34 Join Rob2222 [0] (
22:26:06Lars_Gmaxkelley: and even then, if the flash chips survive (they're hardy) you can desolder them off the pcb ruins, burn your own pcb and turn them into flash cartidges for your GBA ;)
22:26:10*Lars_G chuckles
22:26:11fm2bluebrother: besides, the generated file does not exist when the first compile pass is made. So adding it to FILES will do no good.
22:26:36bluebrotherfm2: well, you need to alter the configuration so it compiles the file in the second run.
22:26:38maxkelleyI've had thumb drives through the wash a billion times.. they take a crapload of abuse.
22:26:39low_lightmaxkelley: if I had to guess, I would say it's something in the init sequence in tuner_power_nolock() in power-c200_e200.c
22:26:51 Part Crash91
22:26:51scorche|workthis is getting a bit off-topic...
22:26:55maxkelleylow_light: so you're thinking that it's not being powered?
22:27:14Lars_Gscorche|work: It's the plage I carry on my back. Sorry man.
22:27:14scorche|workpoint is, dont call a device bricked unless you have to breakout the soldering iron to fix it ;)
22:27:29maxkelleyand microscope :P
22:27:45scorche|worknot necessarily..
22:27:49Lars_GNot really
22:27:58Lars_Gan oscope is 1000 times more usefull
22:28:09Lars_Gor, in a DAP, a LA
22:28:09maxkelleyok, you win.
22:28:15 Quit My_Sic (Read error: 104 (Connection reset by peer))
22:28:27scorche|workLars_G: did you just tell me you were sorry? ;)
22:28:29low_lightmaxkelley: I don't know. When I tried it, it didn't complain about anything, but didn't tune any stations. But that was from work which where it's hard to pull in a radio station anyway.
22:28:50fm2bluebrother: in the second run, only that one file should be compiled. This could be done by an explicit compiler call in Makefile (the name of the file is known). I just don't know how to put all that into the frame.
22:28:51low_lightso for all I know it could work.
22:28:53maxkelleylow_light: I have a long cable for it.. I'll give 'er a try.
22:29:04 Join barrywardell [0] (
22:29:14Lars_Gscorche|work: and then you fueled it :P
22:29:16bluebrotherfm2: well, you can run make V=1 to get the building more verbose.
22:29:22low_lightgood luck...I'm off...
22:29:25 Part low_light
22:29:42bluebrotherbut still you need to link that file in the end. No idea if the build system picks up all objects in the build folder by itself ...
22:29:44*maxkelley grumbles.. leaves are falling outside.. raking will be soon to follow.
22:30:31 Join rogelio [0] (n=rogelio@
22:30:47scorche|workLars_G: but i always get in the last word ;)
22:30:52 Quit desowin ("use linux")
22:31:01 Part Redbreva ("User is away.")
22:31:03maxkelleyon my c200, fm hw isn't even detected.
22:31:26fm2OK. Anybody else maybe with a deep knowledge of RB make system?
22:31:45Lars_Gscorche|work: you abuse the tingling fear of you being an almighty OP
22:32:13tumufm2, i've come to the conclusion that there is no one :)
22:32:42Bagdertumu: there are
22:33:04BagderI guess I am one
22:33:08bluebrotherthere hiding as good as possible ;-)
22:33:11BagderI wrote most of it
22:33:15pixelmamaxkelley: I think I've read similar threads in the forums of e200 owners, their radio was sometimes detected sometime not
22:33:29tumuBagder, i tried to compile the sim against cygwin only, it didn't work
22:33:45maxkelleypixelma: was this variable on a boot-to-boot basis?
22:33:46fm2Bagder: don't be shy! :-)
22:33:47scorche|workaye...i was going to suggest Bagder, but i didnt want to expose him if he was hiding ;0
22:34:03BagderI'll go back hiding now
22:34:22 Part rogelio ("Kopete 0.12.4 :")
22:35:01tumuanyway, the rockbox header files seem to rely on mingw specifics too much
22:35:15tumucompiling against cygwin bombs
22:35:28Bagdertumu: I believe there are people building sims on cygwin
22:35:35Bagderbut I'm not sure
22:35:42pixelmaI do regularly
22:35:45tumui just tried, it didn't work
22:35:52tumuor i am doing something wrong
22:36:05fm2Bagder: could you tell me what I should do? (BTW: my post appear with a huge delay in the log. I don't know why. I'm using irc web client)
22:36:25tumuBagder, cygwin on windows, not cygwin on linux
22:36:33Bagdercygwin is only windows
22:36:44tumuyes, but if someone is crosscompiling
22:36:47Bagderfm2: sorry, I don't have time to dig in that right now
22:36:59Bagdertumu: then they do it with mingw on linux I'd guess
22:37:07pixelmatumu: building sims works fine for me (windows XP here)
22:37:16bluebrotheris it even possible to cross-compile for cygwin?
22:37:20Lars_Gdepends if you're compiling posix or not posix, but mingw32 should cross compile fine from linux
22:37:33tumubluebrother, it is
22:37:34Lars_Gbluebrother: I think mingw allows it... but only think
22:38:17amiconntumu: Compiling sims on cygwin works just fine
22:38:37tumuhow do you do it then?
22:38:38amiconnThe resulting binaries don't rely on cygwin, btw
22:38:48fm2Bagder: ok. I'll wait until later :-) Bye!
22:38:51 Part fm2
22:38:55amiconnJust configure && make && make install
22:38:55tumui tried removing all the mingw specifics from sim makefile
22:39:08tumuthat's mingw build
22:39:09amiconnYou don't need to remove or change anything
22:39:17tumui do need
22:39:23amiconnIt works straight from svn
22:39:27tumuit doesn't
22:39:31tumuthe linkage is all broken
22:39:37pixelmafor you...
22:39:41amiconnIf it doesn't for you, the most probably reason is missing packages
22:39:50amiconn(cygwin packages)
22:39:56tumuno.. gcc links to wrong version of libgcov
22:40:10tumuwrong version being one under mingw instead of cygwin
22:40:33*amiconn wonders what tumu is doing...
22:40:43tumutrying to get -fprofile-arcs working
22:41:41krazykitwhen doing recovery mode with the sansa, do you plug USB in after hitting power/middle?
22:42:08maxkelleyhm.. fm radio is now detected.
22:42:16Bagderkrazykit: let the usb-insert be the power up
22:42:33krazykitah, got it
22:43:09tumupixelma, if you have cygwin install, you can find two libgconv.a's under cygwin/lib/gcc/ folders
22:43:36tumugcc is linking the older one which is under mingw
22:45:43pixelmawhat do you want to tell me with that? I admit I haven't looked at the innards but sim builds are working for me and I don't remember doing anything to make it work
22:46:00 Join freqmod [0] (
22:46:13 Join Zagor [0] (
22:46:37tumuwhich leads to believe that there is no one who really knows how the sim build works :)
22:46:43tumupure magic :p
22:46:59 Quit freqmod (Remote closed the connection)
22:46:59BagderI'd rather say, nobody knows why your gcc breaks it
22:47:15Bagderbut that's me
22:47:38 Quit massiveH ("Leaving")
22:47:53Zagorooh, c200 wpses
22:47:53 Quit amiconn ("CGI:IRC (EOF)")
22:48:26tumuBagder, the mingw libs are older version than the cygwin ones, and i have no idea why
22:48:34pixelmaZagor: they look a bit weird (both) on the display because of the weird contrast :\
22:48:43tumuBagder, and gcc is linking against the older versions
22:49:13Bagdertumu: I'm just saying that those are things that your environment controls, not the Rockbox makefiles for the sim
22:49:26tumuBagder, and when it comes to gcov, it is version matched to gcc and thus does not work in runtime
22:50:00tumuBagder, well, i followed the cygwin install given in wiki
22:50:40 Quit Rincewind ("Verlassend")
22:50:54pixelmatumu: what version is your cygwin gcc? I believe there is also a recommended version
22:51:46pixelmaok, same here
22:53:36tumuanyway, i tried removing the mingw usage from sim makefile, but the build bombs out due to reliance on mingw specifics
22:54:16Bagderwhat sim makefile is that?
22:54:36tumuremoved the mingw stuff from GCCOPTS and LDOPTS
22:54:38Bagderyou mean the generated one?
22:55:13Bagderand how is that mentioning mingw stuff?
22:55:38tumuit is including mingw header folders and linking against mingw?
22:56:59BagderI don't see how configure can do that
22:57:11Bagderthere's only one single mention of mingw in configure
22:57:21 Quit RaRe (Read error: 110 (Connection timed out))
22:57:21Bagderand that is for when a crosscompile is detected
22:58:03tumuwell, mine has -I/usr/include/mingw in GCCOPTS and -lmingw32 in LDOPTS
22:58:23tumui didn't add them there :)
22:59:40tumuremoving them like i have tried breaks the build because it seems to rely on what the mingw brings in
22:59:43Bagderdoes your sdl-config −−cflags include them?
23:00:13BagderI think it does
23:00:18Bagderso there's the blame
23:00:36maxkelleyis there a way to log GPIO status while in the OF?
23:00:37tumuhow do i go fixing it?
23:00:53 Quit saratoga ("CGI:IRC (Ping timeout)")
23:01:52pixelmaI can see that in a sim makefile here too, seems to be harmless
23:02:17tumuit isn't harmless when you use -fprofile-arcs
23:02:32tumuinclude -fprofile-arcs in GCCOPTS and LDOPTS and make clean; make
23:02:50tumurun sim and you'll get spam of errors about mismatched version
23:02:54 Join Bam2550 [0] (
23:03:01tumuit works but no profiling data is generated
23:03:15BagderMrH strikes again
23:03:15pixelmaok, wasn't my plan to do such a thing
23:04:15 Quit lee-qid ("aufwiederbyebientotsayonara")
23:05:08linuxstb_Bagder: ?
23:05:18Bagderyou'll like it
23:06:05tumuthe version mismatch might get fixed with new version of gcc package
23:06:43tumuor even the entire build might break if the cygwin/mingw versions go too mismatched
23:07:05 Quit barrywardell ()
23:07:10 Quit Thundercloud (Remote closed the connection)
23:08:41 Join Klevi [0] (
23:08:51 Nick parafin is now known as parafin|away (
23:08:57KleviDoes anyone know how I can convert Movie files with VLC?
23:09:00preglowoh cool, tracker spam...
23:09:06tumuanyway, the workaround of trying to use only cygwin didn't work either
23:09:09Zagorpreglow: again? where?
23:09:19tumuand that is more of a rockbox issue as it relies too much on mingw
23:09:24BagderI think he refers to the comments I removed
23:10:18***Saving seen data "./dancer.seen"
23:10:21 Join Thundercloud [0] (
23:11:24SoapKlevi: there is instructions on converting movies in the PluginMpegplayer wiki page. While VLC will work, I really do suggest ffmpeg, as it seems more robust with non-standard framerates. If you are on Windows, the WinFF front-end is drop-dead simple.
23:11:42pixelmaKlevi: you have a c200?
23:11:45KleviThanks much Soap
23:11:57KleviWhy, will it not work anyway?
23:12:31 Join BigMac [0] (
23:12:36pixelmaJust wanted to mention that video doesn't work yet (you will only get the audio) because the lcd driver isn't ready for it yet.
23:12:54*maxkelley mumblegroans.
23:13:05Nico_PBagder: what's this document ? I read it but's a little too technical for me...
23:13:18maxkelleyI think low_light is right, that the tuner isn't being given power.
23:13:18KleviI see.
23:13:29BagderNico_P: MrH's research on the PP5024 memory controller
23:13:45maxkelleyhowever, the "mystery amp" is being initted, because I can hear it kick in when I go to the fm radio function.
23:13:56tumuBagder, any chance that you would look into trying to build with only cygwin?
23:14:09bluebrotherwe have an anonymous submitter in the tracker? Or was that account deleted?
23:14:17Bagderbluebrother: it was deleted
23:14:44bluebrotherah, ok. I was already wondering about this.
23:14:44Bagdertumu: not likely, I'm only on linux
23:14:48Nico_PBagder: will it help improve the sansa port ?
23:14:49preglowBagder: wow, nice info
23:15:06BagderNico_P: possibly, but note that this info is probably going for most PP chips
23:15:31tumuBagder, you could crosscompile to cygwin, it should give same results
23:15:40 Quit jhulst (Remote closed the connection)
23:15:46Bagderno I can't
23:15:54Bagderwe crosscompile with mingw
23:15:59Bagderand that seems to work fine
23:17:15tumuhmm, does that mean the build environment on windows is unsupported?
23:17:47Bagderas you've seen lots of people use cygwin
23:18:00tumucrosscompiling with mingw would work for profiling
23:18:15tumujust that native with mingw doesn't due to version mismatch
23:19:15Bagderso why not build it cross-compiled or fix the build issues?
23:20:18tumuboth are viable solutions
23:20:36tumulatter would be better as it would make the sim build against cygwin as well
23:22:38 Quit seablue ("life, death, life, death")
23:25:13pixelmacould it be that one warning is left from jhMikeS's commit today (also in the sims in mpegplayer) - but it is only a single warning instead of the 2 in the other simulators?
23:26:10 Quit Klevi ("Leaving")
23:26:38pixelmabut also only one some build servers it seems, so maybe the same cause...
23:27:09Bagderyeah, only gcc 4.1.x machines...
23:28:13preglowBagder: have you got some kind of an index over mrh docs somewhere?
23:28:24Bagderon the top of
23:29:33 Join hanklords [0] (
23:30:00 Quit BigMac (Remote closed the connection)
23:30:32Nico_PBagder: is there a link to that page somewhere in the wiki ?
23:30:41Bagderand if you want to read some thumbs up from MrH, he put it this way =>
23:31:35BagderNico_P: I'm not sure, but most of those docs have been used to produce the code that we have now
23:31:51Bagdernand, fm, lcd etc
23:32:30Nico_Pwell that's probably a good reason to have something about them in the wiki, isn't it ?
23:32:46 Join BigMac [0] (
23:33:30Bagdersure, adding things in the wiki is often a good idea
23:33:54pixelmaBagder: now that's weird, if you look at the last green Mini2g sim (built by farsan) and the one right above (by bygg) they show the same gcc and binutils version
23:34:57Zagoryeah, the versions are not the whole story
23:35:05Bagderis bygg 64bit?
23:37:07Bagderseems to be gcc 4.1.x on 64bit then
23:37:26 Quit BHSPitMonkey (Read error: 104 (Connection reset by peer))
23:41:21scorche|workoff i go..
23:41:32Zagorcan't upgrade to 4.2 right now, there's some conflicts that would remove sdl and libc6-dev...
23:41:41 Quit scorche|work ("CGI:IRC (EOF)")
23:43:02 Join darkapostrophe [0] (
23:46:20Zagorblah, "make modules" in linux takes forever!
23:46:35preglowif you elect to build everything as modules, then yes
23:46:55Zagorall I want is my new shiny debug-enabled usb-storage...
23:47:08preglowyou can cut the time down tons by just manually pruning away the larger chunks you'll never need, like video 4linux, isdn support, etc...
23:47:26 Quit davina (Remote closed the connection)
23:50:55 Quit Thundercloud (Remote closed the connection)
23:51:06 Join TMM [0] (n=hp@
23:51:07Zagoryay, "make M=drivers/usb/storage" does it in 2 seconds...
23:52:24 Quit TMM (Client Quit)
23:53:31 Join Thundercloud [0] (
23:54:39 Join DefineByte [0] (
23:57:50 Join RaZorbacK [0] (

Previous day | Next day