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

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

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

#rockbox log for 2008-12-04

00:00:45ej0rgeI figured, as long as I'm in there with a soldering iron . . .
00:00:52ej0rge <−− only $2.50 shipped
00:01:12Zagorwow, that is cheap
00:01:31ej0rgelooks almost exactly like the other DX adapter, except black mask instead of green, and they claim 1.97 in x 1.85 in x 0.2 in vs. the original 2.2 in x 1.89 in x 0.26 in
00:02:01ej0rgeZagor: the tradeoff is that you wait 2-3 weeks for it to arrive
00:03:40ej0rgeI won an h320 with a busted DC jack on ebay thursday before last and ordered the green DX CF adapter for it the same day
00:04:04ej0rgethat order is still pending because they're waiting for some other item i ordered with it to get back in stock
00:04:20ej0rgetoday the black one shows up at half the price
00:04:33 Join LambdaCalculus37 [0] (
00:04:36ej0rgeso i canceled the old order and placed a new one w/o the out of stock item
00:06:04ej0rgethe h320 was a real score despite arriving w/o harddrive. $21. behaved very badly with the drive from my h120 until the battery was well charged and i re-flashed it with 1.29E firmware. It's rockboxed now, but i don't want to reassemble it until i've got the CF card in it.
00:06:22ej0rge$21. Just needed some superglue and solder. And a drive of some sort.
00:06:41ej0rgehm, didn't realize I'd noted the cost already
00:07:52ej0rgeanyway, thanks for the wiki access.
00:08:37saratogaZagor: I'd prefer something that mentions future targets directly
00:08:47kugelgevaerts: how could I help out with that?
00:09:17Zagorsaratoga: yeah, that's probably a good idea
00:09:19kugelI'm not so bored :P
00:09:26saratogamaybe something like Current status for Current and New Targets
00:09:30 Quit skipper (Read error: 110 (Connection timed out))
00:09:38 Join aarcane [0] (
00:09:42 Quit bluebrother ("leaving")
00:10:02saratogaor even a New Ports bullet in that list
00:10:05Zagor"Status for current and work-in-progress targets"
00:10:33saratogathat sounds good
00:11:00 Quit perrikwp (" ajax IRC Client")
00:11:28gevaertskugel: easy. Study the patch and tell me what's wrong with it ;)
00:11:48 Join perrikwp [0] (i=18ac0c41@gateway/web/ajax/
00:12:52 Join Raccoon [0] (n=40e61dfb@gateway/web/cgi-irc/
00:13:34 Quit Raccoon (Client Quit)
00:14:30ej0rgeI wish there were a zif-microdrive to CF adapter - just a printed ribbon cable with CF pin strip directly attached. So i could get a little olympus m:robe mr100 for peanuts on ebay and cram a 32gb CF in it.
00:16:22Zagorquick links added
00:16:49 Quit dany_21a_ (Read error: 110 (Connection timed out))
00:17:04l7there's a interesting idea
00:17:17 Quit Rondom ("Ex-Chat")
00:18:45linuxstbZagor: Looks good to me. All that's missing is a "getting started" quick-link, but I don't think we have a page for that to go to...
00:19:05 Quit tyfoo ("Carpe diem")
00:19:21ej0rgel7: I also wonder if the models that came with a male microdrive could accept a CF card with just a length of pin strip used as a male-male adapter. but i'm just dreaming there and haven't even checked the geometry let alone the pinout.
00:19:54 Join einhirn [0] (
00:20:01ej0rgeIt's just Cool how zippy the H120 is with even the cheap slow "speedy" A-Data cf card
00:20:07ej0rge20+ hours run time too
00:20:13*funman notes sansav2 is the less red of new platforms
00:20:18ej0rgebut it's a bit of a brick compared to more modern players
00:20:23rasherjhMikeS: your change to english.lang means all translators will be bothered by that string. Could you change it to a comment instead?
00:20:51 Join krazykit [0] (
00:21:48Zagorlinuxstb: is a pretty good Getting started page, isn't it?
00:22:16Nico_PZagor: it's not quite up to date, but yeah
00:22:36linuxstbZagor: Yes, looks like a good candidate.
00:22:59ej0rgeyou know i tried the gui installer with the h320 and it failed to patch the firmware file. or mention that it hadn't patched it.
00:23:00Nico_Psaratoga: in TargetStatus, did you link to footnote1 for (2) on purpose?
00:23:00linuxstbIs "Getting started" the best name for the link? We could just call it "Installation".
00:23:16ej0rgewhich was pretty confusing
00:23:16l7ej0rge: yeah i was just reading up on CF card speeds today. they offer some really nice performance for the price
00:23:24Zagor"Install today!" (and get these kitchen knives too)
00:23:48l7ej0rge: any particular reason to want the extra speed on a DAP though?
00:23:49ej0rgel7: write speed on the A-Data card is much much faster in a pcmcia sleeve in my laptop than it is in the iriver. Read speed is just dandy though.
00:24:05ej0rgel7: H120 boots in about 1.43 seconds vs. 10-12 seconds
00:24:15ej0rgethough a lot of that is booting directly from flash
00:24:20ej0rgetrack to track is nice too
00:24:27ej0rgeno time wasted waiting for a platter to spin up
00:24:51l7hrm, yeah faster boot speed is kind of cool
00:25:00ej0rgeso it's not entirely that the read speed is good - it's that you don't have to wait for a sleeping drive to wake up
00:25:11l7my sansa takes about 9 seconds to boot
00:25:41l7which feels a bit slow at times
00:26:08ej0rgeI'm curious what the speedup will be on the h320, where flashing rockbox into rom isn't yet recommended (or possible?)
00:26:21ej0rgeiirc because we still need some of the OF features for usb or charging or something
00:27:01saratogadid anyone ever try the PLL disable patch on other PP targets? i wonder if we could save power on some of the older Ipods too
00:27:10ej0rgethe cheap Topram CF cards don't say how many X's they are. Which would generally mean they're slow. OTOH they say they are fast. *shrug*.
00:27:40saratogaNico_P: fixed
00:29:15l7ej0rge: if you haven't seen it yet, has a useful database of flash memory cards
00:30:04l7i don't really shop around for CF much since i don't have any gear that really uses it anymore
00:30:56ej0rgel7: i hadn't seen it. I didn't have use for much CF until i started using my iriver 30+ hours a week again. My camera eats CF, but I've never maxed out the 1gb card in it, and it's Fast Enough.
00:31:53ej0rgewhile my music archives approach 100gb, i pray that my iriver will still be alive when 128gb CF has not only arrived but dropped below $150
00:32:08ej0rgeby which time, maybe it won't be big enough anymore :)
00:34:01l7it's rather unfortunate SD replaced CF in so many places
00:34:19l7it would have been nice to see CF evolve into a more compact form
00:34:53kugeli.e. SD?
00:35:05zethat would have been nice... sd was great for its form-factor, but the inner-workings of cf seem superior overall
00:35:26zei think even sdhc tops out lower than cf for max capacity
00:35:32ej0rgeI really don't know how they compare
00:35:42ej0rgeiirc CF runs out of address bits at 137gb
00:35:44l7SD is slower and somewhat less open
00:36:06l7SD also supports DRM while CF does not
00:36:43l7CF cards seem to be capable of 25-30MB/s transfer rates at the top end
00:37:00kugelis supporting drm bad?
00:37:02zelooks like maybe i'm half-wrong about the max capacity... theoretically its 2TB but "the SD Card association has artificially defined the maximum limit of SDHC capacity to 32 GB.[citation needed]"
00:37:25l7kugel: i care more about the speed really
00:38:25kugell7: for the space a CF takes you can do 4x RAID with SD cards, with uberspeed
00:38:54zei'm not sure whats up with the speed or why sd couldn't become as fast as cf speed-wise, unless its an electrical interface level issue or somesuch sorta thing
00:39:26zeanother convenient thing for cf though cf ide and pcmcia are all very friendly
00:39:26LloreanThe discussion of various flash memory merits isn't exactly a Rockbox-specific or even particularly Rockbox-development/support relevant, is it?
00:39:43*ze disappears in a puff of OT
00:40:24l7i guess we could continue this in #rockbox-community
00:40:48*l7 flees from the massive cloud of OT
00:42:21kugelLlorean: should be relatively easy fixable
00:42:27Zagordoes rbutil still say "configuration error" the first time you start it?
00:42:42linuxstbI'm pretty sure that's been fixed.
00:43:19*funman wonders if linuxstb wants to integrate mkamsboot into rbutil
00:43:38 Join ajonat [0] (n=ajonat@
00:43:40*linuxstb stays far away from GUI programming and C++
00:43:40kugelfunman: I don't. I'm sure of that
00:43:59kugelok, rather he wants to have it integrated ;)
00:45:11 Quit culture (Connection timed out)
00:45:28linuxstbHow stable is the AMS bootloader now for the various targets? i.e. is it time to release an official binary yet?
00:46:34 Quit jhulst (Remote closed the connection)
00:47:05funmanUse is stable, but code isn't
00:47:17 Nick JdGordon|zzz is now known as JdGordon (n=jonno@rockbox/developer/JdGordon)
00:47:27***Saving seen data "./dancer.seen"
00:47:30funmanthere is still some settings to change (especially for the various peripheral clocks)
00:48:05linuxstbIs that important in the bootloader? Doesn't main rockbox do the same inits again?
00:48:16funmanno it doesn't
00:48:17 Join jhulst [0] (n=jhulst@unaffiliated/jhulst)
00:48:58funmanTesting if we can redo them in main would be nice though, I think I remember failure in sdram_init()
00:48:59kugellinuxstb: why would you already want to release a bootloader?
00:49:40linuxstbkugel: Who said I want to release a bootloader? I'm just asking how close it is to being ready.
00:49:58kugel"i.e. is it time to release" sounded like
00:50:15linuxstbNo, that's just a simple question.
00:52:02linuxstbBut the reason I asked was because of funman's question about rbutil - it doesn't make sense incorporating mkamsboot into rbutil until the bootloader is ready. We would want to pre-calculate md5sums of patches OFs with the released bootloader binary.
00:53:07funmanwhy not using md5sums for bootloader and original OF ?
00:54:01 Quit Genre9mp3 ()
00:54:02 Join basti [0] (
00:54:23linuxstbfunman: Extra paranoia.
00:54:44linuxstbThat's how rbutil works on other targets with similar installation methods.
00:55:02kugelI think this kind of paranoia doesn't hurt
00:55:07bastihi there! can someone who owns an h340 tell me, which harddisk i can/have to buy in order to change the current one? because my one is getting worse and worse
00:55:08linuxstbe.g. the iriver h1x0 is the same - we patch an OF image, which is then flashed.
00:56:06bastithx linuxstb
00:56:22funmanI think Clip and m200v4 could be supported as soon as sound goes in svn and buffering is fixed
00:56:33funmanSafe a missing manual, and missing plugins for m200v4
00:56:36kugelJdGordon: I uploaded my clean up patch
00:57:06linuxstbfunman: IMO we need a manual and good install instructions before we advertise something as supported. Otherwise, we will just get 100s of questions about it.
00:57:09JdGordoni saw
00:57:32linuxstbfunman: Especially with a target as popular as the Clip.
00:58:59n1sYes, most things should be working before anything gets mentioned as "supported" or the installation becomes too easy or we'll get flooded
00:59:08funmanI take install instructions are easy once it's in rbutil
00:59:30kugelJdGordon: I'm not sure whether you like the viewport_set_default change (I do!), but for the other parts things are done properly now
00:59:33funmann1s: sure, but what is 'most' ?
00:59:33n1sWe usually have a manual installation instruction too
01:00:12funmanTo me sound is quite good already ;)
01:00:27Lloreanfunman: All core features need to work, and any plugins that we choose to compile and describe in the manual need to work.
01:00:36JdGordonkugel: good guess :D
01:00:53n1sfunman: imho some nice way to handle usb connects tops the list after the basic music palyback works as we get lots of questions about that on currently supported targets :/
01:00:56Nico_Pfunman: btw, did you continue looking at the buffering code?
01:01:13funmann1s: do you mean a reboot into OF ?
01:01:41funmanNico_P: only a bit, I suspect the pcm buffer becomes overwritten but I have no proof yet
01:01:48n1sfunman: yes, that or soem disk mode or even native rockbox usb not just "do nothing" (or freeze) :)
01:01:54 Quit CaptainKewl (" ajax IRC Client")
01:02:18Nico_Pdo you think it has something to do wih the buffering code, or is it at a lower level?
01:02:33funmann1s: well rebooting into OF will be a bit hard I think, perhaps with some magic in mkamsboot. For now it will just charge.
01:02:37linuxstbn1s: We could just have a message "USB not supported - use the original firmware".
01:02:48kugelJdGordon: I rather change init the vp once fixed for bars instead of at every do_menu, which happens to be in a loop at some places.
01:02:53linuxstb(when usb is inserted)
01:03:12funmanNico_P: Since I have no problem with buffering_flash.c I suspect it's in buffering.c, but perhaps buffering_flash just doesn't trigger another bug
01:03:41kugelthe viewport_set_defaults change was initiated by the fact, that (e.g. in stars.rock) the vp was "fixed" again and again, i.e. moved 8px up at every entering of the menu
01:03:42 Quit faemir (Remote closed the connection)
01:03:56JdGordonkugel: what I'm working on wont even need changing that often
01:03:57funmanifp7xx use a small buffer as well
01:04:09*n1s points to all the people who come here asking what to do when an ipod bootloader can't find rockbox.ipod (because they didn't install it) and tells them to press the rebot into diskmode combo
01:04:16kugelJdGordon: you change it once with my patch
01:05:14kugelyou may explain what's bad about that change
01:05:37 Quit n1s ()
01:05:51 Quit __lifeless (Read error: 60 (Operation timed out))
01:05:54 Join _lifeless [0] (n=lifeless@
01:06:04 Quit Zagor ("Client exiting")
01:06:30funmanNico_P: clip and m200v4 seem to be in the first (not counting logikdax and ifp7xx) swcodec targets with little memory
01:06:54JdGordonkugel: well, mainly that the bars arre not tied to the list...
01:07:31Nico_Pfunman: yes, so it seems I should find some time to polish the patch
01:07:32funmanWe have already found a buffer overflow in playback.c for instance (with ~512kB audiobuffer), there is probably others spread all over the place
01:08:21Nico_Pwith the svn buffering code or the patch one (or both)?
01:10:10kugelJdGordon: well, that's svn behavior. What you plan needs more rework and is not intention of the patch. And again, you won't get around modifying the list parents to make place for the lists
01:10:32funmanpcmbuf_init() required too much data from our little buffer
01:10:44kugeland btw: with my patch is rather seperated from the list, since the list as is doesn't take statusbar into account. It takes the parent and draws in
01:10:54funmanso it's not related to buffering*.c I believe
01:10:58JdGordonkugel: yeah, no, its fine.. im still asleep
01:11:31kugelhehe ok, I'm almost asleep too, so I'm fine I don't need to defend my work even more :)
01:13:34kugelsaratoga: isn't changelog is not one word???
01:13:39*kugel didn't know that
01:14:32kugelI like the new quick links.
01:15:57 Part toffe82
01:16:04 Quit Nico_P (Remote closed the connection)
01:16:27 Quit webguest93 ("CGI:IRC")
01:16:42 Quit kugel ("ChatZilla 0.9.84 [Firefox 3.0.4/2008111318]")
01:21:39 Quit herrwaldo ("Konversation terminated!")
01:23:12Unhelpfuli think i may have spotted the cause of the mono-bitmap bug on the scaler, seems to have to do with the FORMAT_MONO vs FORMAT_NATIVE values for the format field in struct bitmap
01:24:02 Quit LambdaCalculus37 ("Ka-chunka")
01:24:23funmanI almost forgot that the Clip keymap needs fixing
01:24:51Unhelpfulit seems as though if i'm loading a mono bitmap i need to tag it as FORMAT_MONO and store it as mono? i'm assuming vertically packed?
01:26:05 Join Hillshum [0] (
01:26:35ej0rgewell, thanks all for the chat, sorry about the off topic digression. I'll pop in again next time i have something dev-relivant to say
01:26:41 Quit ej0rge ("leaving")
01:27:06 Quit jgarvey ("Leaving")
01:27:10 Join RoC_MasterMind [0] (
01:27:31bastii have another question according the h340 and its hdds: are the lager harddrives compatible with this model? so can i get a 80gig hd and use it with my player?
01:27:59linuxstbAs long as it physically fits, it will work.
01:28:14Unhelpfulbasti: i'm pretty sure that there's a wiki page about acceptable HDD replacements
01:29:40Unhelpfulhm, is down again?
01:30:05LloreanWorks for me
01:31:43Hillshumand me
01:32:05Unhelpfulhuh, another try and it works. weird.
01:33:23 Quit RoC_MasterMind (Remote closed the connection)
01:35:30Unhelpfulit would appear that compatible drives are available to at least 50GB
01:36:57Unhelpfulerm. 80GB.
01:37:11 Quit jhMikeS (Nick collision from services.)
01:37:17 Join jhMikeS [50] (n=jethead7@rockbox/developer/jhMikeS)
01:37:36 Quit saratoga ("CGI:IRC (EOF)")
01:39:15jhMikeSrasher: sure, if that's the proper way to go
01:39:48 Quit Bensawsome (Read error: 110 (Connection timed out))
01:40:48 Join Bensawsome [0] (
01:40:54 Nick Bensawsome is now known as Guest76192 (
01:43:14 Quit EspeonEefi ("さよなら")
01:46:03Unhelpfulsince it seems that depth==1 bmp files can still have a palette, should i be loading them as FORMAT_NATIVE and storing display-native data? that's the hack i tried to see if i found the real cause of the 1-bit display problem, and it seems to have fixed things... but i'm sure there was a reason to have FORMAT_MONO in the first place.
01:46:44 Quit einhirn (Read error: 104 (Connection reset by peer))
01:51:45 Join jaf2 [0] (n=chatzill@
01:52:00 Quit mcuelenaere ()
01:55:05 Join RoC_MasterMind [0] (
01:56:34 Join EspeonEefi [0] (i=eefi@STRATTON-TWO-EIGHTY.MIT.EDU)
01:58:09 Quit perrikwp (" ajax IRC Client")
01:58:36 Join perrikwp [0] (i=18ac0c41@gateway/web/ajax/
02:01:14 Join Darksair [0] (n=user@
02:02:12 Join thegeek_ [0] (
02:20:04 Quit Hillshum ("ChatZilla 0.9.83 [Firefox 3.0.3/2008092417]")
02:21:45 Quit pixelma2 (Read error: 110 (Connection timed out))
02:22:41funmanI can't find a ManualInstall page on the wiki, I would like to write instructions for Sansa AMS but don't know which page to use / where to link it from
02:23:58 Join Zeuzpr [0] (
02:24:40funmanHum shouldn't the mailing list use "−− " to delimit the etiquette/faq/admin urls, instead of "−−−−−−−−−−−−−−" ?
02:24:53Zeuzpri trying to install it on ipod 1g and came an error
02:25:06Zeuzprfailed to add bootloader
02:25:23Zeuzprany sugestion??
02:25:48JdGordonjhMikeS: you round?
02:26:07Zeuzprany sugestion??
02:26:28advcomp2019Zeuzpr, what is the error
02:26:30krazykitZeuzpr, please have patience in waiting for help, especially with such an uncommon target
02:26:48advcomp2019o wait sorry
02:26:55Zeuzprfailed to add bootloader
02:27:12funmanare you running as admin / root ?
02:27:38Zeuzprlet me try something
02:28:26 Join fdinel [0] (
02:28:47*funman salutates long missing fdinel
02:29:17 Join BHSPitMonkey [0] (n=stephen@unaffiliated/bhspitmonkey)
02:30:06Zeuzpri run the program as administrator and still the same error
02:30:11fdinelhey funman :D
02:30:14fdinelhappy to see ya
02:30:23fdinelI've got some spare time tonight :D
02:30:43funmancool :)
02:31:36 Join JohnDoe_ [0] (
02:31:48 Nick JohnDoe_ is now known as JohnDoe (
02:31:56 Quit thegeek (Read error: 110 (Connection timed out))
02:32:33Zeuzpri run the program as administrator and still the same error
02:33:18funmandid you try another usb port / cable ?
02:34:25 Nick JohnDoe is now known as Smokin (
02:34:41Zeuzprrigth now try another port
02:34:48Zeuzprstill the same
02:34:50 Nick Smokin is now known as SmokinJoe (
02:35:10funmanI don't know ipods so I can't help more sorry
02:35:24Zeuzpri will try with another computer tomorow
02:35:44Zeuzpri have a q for a sansa e250r
02:36:27Zeuzprif i change it to rockbox does it support microsdhc???
02:37:20funmanYes, but rockbox on e200 doesn't support usb, so you need a sdhc reader for your PC
02:38:01advcomp2019Zeuzpr, are you in rhapsody mode?
02:38:40Zeuzprshould i change it??
02:38:43 Join m0f0x [0] (
02:38:55advcomp2019o ok.. i thought you said you had an ipod
02:39:06Zeuzpri have both
02:39:26advcomp2019o ok
02:39:44Zeuzprshould i change it??
02:40:24Zeuzprthe rockbox doesnt support usb for the microsd slot but it support it for the internal memory???
02:40:42 Join num1 [0] (n=brian@unaffiliated/num1)
02:40:47scorcheZeuzpr: why would rockbox need USB support for the internal memory?
02:41:01 Part num1 ("veni, vidi et capiebar ad anum "I came, I saw and I endured a rear assault"")
02:43:54Zeuzprflet me write the question again
02:44:15Zeuzpryou tell me guys rockbox doesnt support usb
02:45:00UnhelpfulZeuzpr: en e200, no, it doesn't. you can only read/write files over usb in the original firmware, and the OF doesn't support SDHC.
02:45:06Zeuzprusb for the micro sd card slot? or for the 2gb internally on the e200??
02:45:33scorcheZeuzpr: usb isnt used for the micro sd card slot
02:46:22Zeuzprbut it support 8gb microsdhc if i have a card reader on a e200 v1?
02:46:30funmanyes it does.
02:46:42scorcherockbox will theoretically support any SDHC card
02:46:49 Quit SmokinJoe ("ChatZilla 0.9.84 [Firefox 3.0.4/2008102920]")
02:47:02Unhelpfulright, you need to use a card reader to load files onto the card, but then rockbox can play from it.
02:47:12scorcheyou need the card reader because we use the original firmware for USB at the moment....which wont support SDHC so wont see the card
02:47:28***Saving seen data "./dancer.seen"
02:48:38Zeuzprif i want to revert it to the original firmware whats the procedure??
02:49:12scorchehave you checked the manual for uninstallation instructions?
02:49:26Zeuzprok i will read it now
02:49:38 Nick m0f0x is now known as colesterol_dog (
02:49:48Zeuzprif have any doubt will come again thankz a lot
02:49:59Zeuzpru guys are the best
02:50:59 Nick Guest76192 is now known as Bensawsome (
03:02:53 Quit EspeonEefi ("さよなら")
03:12:50 Quit jaf2 ("ChatZilla 0.9.84 [Firefox 3.0.4/2008102920]")
03:13:12 Quit MethoS- (Remote closed the connection)
03:15:12 Quit Darksair (Read error: 110 (Connection timed out))
03:22:36 Quit funman ("leaving")
03:23:41Zeuzprhi guys
03:24:01Zeuzprim back
03:25:04Zeuzprno sansa found
03:25:22Zeuzprhow can i resolve it??
03:25:56 Join pixelma2 [0] (n=marianne@rockbox/staff/pixelma)
03:28:11Zeuzprcan someone help please
03:32:51 Join gregorovius [0] (
03:34:20Unhelpfulsorry, i don't know how. perhaps somebody who does will be able to answer you, but they may not be here right now.
03:35:24 Join Makuseru [0] (
03:38:50 Join nuonguy [0] (
03:42:22 Quit Zeuzpr ()
03:42:24 Part Dhraakellian ("Meddle not in the affairs of ircops, for they are (not so) subtle, and quick to anger.")
03:50:41 Quit Makuseru (Read error: 104 (Connection reset by peer))
03:51:33 Join Zeuzpr [0] (
03:51:45Zeuzprhi guys need help
03:51:48krazykitZeuzpr, what kind of sansa do you have?
04:01:47 Quit gregorovius ()
04:05:53 Nick fxb is now known as fxb__ (
04:06:01 Join sadmac [0] (
04:06:37sadmacI have a Sansa Fuze (just bought) is there anywhere I can track progress of support for this device?
04:07:46Lloreansadmac: Just read the thread on the forums.
04:10:44sadmacLlorean: ok
04:16:18 Quit RoC_MasterMind ("Leaving")
04:18:24 Quit miepchen^schlaf (Read error: 110 (Connection timed out))
04:22:42 Quit BHSPitMonkey (Remote closed the connection)
04:23:00 Part unstable
04:24:32 Quit XavierGr ()
04:29:50 Quit colesterol_dog ("aksdjasdk")
04:34:16 Quit aarcane (Read error: 113 (No route to host))
04:38:04 Join LambdaCalculus37 [0] (
04:41:15 Join Darksair [0] (n=user@
04:44:28Unhelpfultesting sims now, but i think i have all of the FORMAT_MONO kinks out. :D
04:47:32***Saving seen data "./dancer.seen"
04:50:00 Join planetbeing [0] (
04:50:21 Quit agaffney (Read error: 104 (Connection reset by peer))
04:53:30 Quit HellDragon (Read error: 104 (Connection reset by peer))
04:54:08 Join agaffney [0] (n=agaffney@gentoo/developer/agaffney)
04:56:06 Quit _lifeless (Remote closed the connection)
04:56:17 Join HellDragon [0] (
04:56:19 Join _lifeless [0] (n=lifeless@
04:58:55 Join blkhawk- [0] (
05:02:05 Quit HellDragon (Remote closed the connection)
05:02:31 Join HellDragon [0] (
05:07:14 Quit LambdaCalculus37 ("Ka-chunka")
05:08:50 Quit sadmac (Read error: 104 (Connection reset by peer))
05:10:30 Quit Zeuzpr ()
05:15:08 Quit blkhawk (Read error: 110 (Connection timed out))
05:15:52 Nick blkhawk- is now known as blkhawk (
05:24:40 Quit jhulst (Remote closed the connection)
05:25:16 Quit Zarggg ()
05:27:49 Join jhulst [0] (
05:33:49 Quit Horscht ("IRC is just multiplayer notepad")
05:45:07 Quit fdinel ("Miranda IM! Smaller, Faster, Easier.")
05:45:27 Quit _lifeless (Remote closed the connection)
05:48:37 Join _lifeless [0] (n=lifeless@
05:57:27 Quit JdGordon (Remote closed the connection)
05:58:57 Join JdGordon [0] (
06:04:02 Quit nplus (Remote closed the connection)
06:11:21 Quit Darksair ("Emacs = ESC-Meta-Alt-Ctrl-Shift")
06:12:14 Join Darksair [0] (n=user@
06:25:34 Quit _lifeless (Remote closed the connection)
06:25:54 Join _lifeless [0] (n=lifeless@
06:42:23 Join Kremit [0] (n=4c1b1ac8@gateway/web/cgi-irc/
06:43:15KremitQuestion: i browsed through the FAQ and manual.. Does Rockbox on an ipod 5.5g support ipod accessories like my ipod apapter for my cd player in my car?
06:43:56Kremitthis particular adapter plugs right into the back of my head unit in my mazda 6
06:44:03Kremitand allows me to change the song etc etc
06:47:35***Saving seen data "./dancer.seen"
06:49:14 Quit _lifeless (Read error: 145 (Connection timed out))
06:50:04 Join _lifeless [0] (n=lifeless@
06:51:17 Quit HBK (Read error: 110 (Connection timed out))
06:53:44 Quit Kremit ("CGI:IRC (EOF)")
06:56:05 Quit AndyIL (Read error: 60 (Operation timed out))
06:57:03 Join AndyI [0] (i=AndyI@
07:00:06 Quit Darksair ("To Arch or Gentoo? That is the question...")
07:31:23 Join Beta2K [0] (
07:31:36 Quit JdGordon (Read error: 104 (Connection reset by peer))
07:31:53 Quit Beta2K (Client Quit)
07:31:58 Join Beta2K [0] (
07:34:03 Join LinusN [0] (n=linus@rockbox/developer/LinusN)
07:34:09 Join BigBambi_ [0] (n=Alex@rockbox/staff/BigBambi)
07:34:32 Join thegeek [0] (
07:37:42 Quit Beta2K (
07:37:42 Quit AndyI (
07:37:42 Quit thegeek_ (
07:37:42 Quit krazykit (
07:37:42 Quit BigBambi (
07:38:12 Join JdGordon [0] (n=jonno@rockbox/developer/JdGordon)
07:38:27NJoinkrazykit [0] (
07:40:59NJoinBeta2K [0] (
07:42:08NJoinAndyI [0] (i=AndyI@
07:45:38 Join skipper [0] (
07:58:39 Quit denes_ (Read error: 110 (Connection timed out))
07:58:39 Join denes_ [0] (
08:10:16 Quit jhulst (Remote closed the connection)
08:26:18 Quit BigBambi_ (Read error: 104 (Connection reset by peer))
08:31:16 Quit ajonat ()
08:33:16 Join snares [0] (
08:35:16snaresI did a google search and found a thread on a rockbox forum that said the sansa fuze is in the rockbox tree or something like that. will rockbox work w/ the fuze?
08:35:45scorchethe port is still in progress...when it is ready for public consumption, it will say so on the front page
08:36:46snaresdoes that me it works just not well enough for the public or it just doesn't work
08:37:06scorchethat depends what you mean by "works"
08:37:31snaresyes I suppose "works" is a key word
08:37:50snareswhat should I mean by "works"
08:38:13scorchei have no idea what you mean with your own words..
08:38:17 Join Zagor [0] (n=bjorn@rockbox/developer/Zagor)
08:38:45Unhelpfulwhen i use a word, it means exactly what i want it to... no more, and no less. ;)
08:39:19snaresI was just wondering what does work that's all
08:40:01Unhelpfuli'd expect the forum thread to cover that pretty well. you could also read through the logs of the IRC channel for the past few days ;)
08:40:05scorcheyou will probably want to read the thread about it in the forum thread
08:40:12 Join Rob2223 [0] (
08:40:20snaresif it will be too much of a pain then I will go with something that does "work"
08:41:30Unhelpfulsnares: i'd recommend you do that. look for an e200v1, or if you want larger storage, a gigabeat f or x. the s is also mostly working, if you're willing to build your own with the charging patch, or use the vendor firmware to charge.
08:47:39***Saving seen data "./dancer.seen"
08:51:51 Join Bagderr [241] (n=daniel@rockbox/developer/bagder)
08:52:16 Nick Bagderr is now known as B4gder (n=daniel@rockbox/developer/bagder)
08:53:39 Join kharo [0] (
08:56:19 Join ender` [0] (
08:56:58 Quit _lifeless (Remote closed the connection)
08:57:11 Join __lifeless [0] (n=lifeless@
08:57:21 Quit Rob2222 (Read error: 110 (Connection timed out))
08:57:46*Zagor takes down twiki due to a critical vulnerability
08:59:38B4gderfoswiki seems to not have released anything... :-/
08:59:51B4gder(insert "still")
09:01:47 Quit perrikwp (" ajax IRC Client")
09:02:05 Join perrikwp [0] (i=18ac0c41@gateway/web/ajax/
09:04:54*JdGordon keeps going round in circles in playlist.c :'(
09:05:20B4gderswitch off repeat!
09:05:33*B4gder hides
09:06:19 Quit gevaerts (Nick collision from services.)
09:06:30 Join gevaerts [0] (n=fg@rockbox/developer/gevaerts)
09:07:19JdGordongevaerts: hey, is there any reason the sim needs to printf "storage_sleep()" ?
09:09:47JdGordonhmm... aparently next dir doesnt set playlist->started to true if audio is actally playing!
09:09:56JdGordonwhich I tihnk is the reason for 7423
09:15:48 Join lasser [0] (
09:18:39JdGordonhmm... there are a few playlist_start() in playlist.c which are only done for HWCODEC... any ideas why?
09:19:32Zagortwiki is patched, and up again
09:21:10*JdGordon answers self... apparently it breaks things if its used :p
09:25:27ZagorI think we should branch off the hwcodec targets
09:27:11 Quit scorche|sh (Read error: 60 (Operation timed out))
09:27:13 Join scorche|sh [0] (
09:27:59JdGordon... arr... the next directory code seems to be called in 3 completly different ways!
09:29:09UnhelpfulZagor: on the one hand, that would make me feel less bad about basically forking the bmp loader to avoid a binsize hit on low-mem b/w targets, since we could probably just use the old loader in the fork...
09:29:53Unhelpfulotoh, aren't there more than a few places that such a fork could lead to trouble sharing code, down the road?
09:30:30JdGordonHAIL TO THE KING BABY!
09:30:37*JdGordon fixed the bug :D
09:31:53ZagorUnhelpful: the idea is to stop developing the hwcodec targets. or at least stop them from hindering development of the rest.
09:32:17Zagorimho the only reason we keep them around is for historical reasons and "because we can"
09:32:19JdGordonamiconn will hunt you down!
09:32:28B4gderZagor: and because of amiconn
09:32:30Zagoroh, I know :-)
09:32:49Zagorwe can call it the amiconn branch :)
09:33:19 Join einhirn [0] (
09:34:36ZagorI actually have an old archos recorder myself, that lives in my car and faithfully boots and entertains me everytime I drive.
09:35:10Zagorbut It Just Works. it doesn't need new game plugins or movie players.
09:36:30ZagorI challenge amiconn to find five people who uses a hwcodec archos as his/her one and only player
09:36:50scorchedoes that really matter though?
09:37:07B4gderright the counter-argument is exactly that it doesn't matter how many users there are
09:37:13Zagorbut it's another nail in the coffin
09:38:03JdGordonspeaking of which... B4gder time for another download stats update?
09:40:13JdGordoncan everyone please test ? If it breaks anything it will be when adding/chanigng tracks int he playlist
09:42:14 Join kugel [0] (n=chatzill@unaffiliated/kugel)
09:48:38 Join sbhsu_ [0] (
09:50:12 Quit kachna (Read error: 60 (Operation timed out))
09:55:05 Quit sbhsu (Read error: 110 (Connection timed out))
10:02:28Unhelpfulscaler-code binsize gain is down to 280B on mono targets without scaler. i suspect i should just get to work resurrecting the old bmp.c for such targets in a new filename.
10:03:05gevaertsJdGordon: I only renamed the function, so I don't actually know
10:03:36JdGordonaccording to blame you changed that line... but ok
10:04:01kugelUnhelpful: 280B isn't too bad
10:04:08Unhelpfulalso, the latest, aside from a tiny binsize improvement on mono, is up on #9458. it tests ok in a large number of sims with different displays, with every WPS/RWPS i tried it with.
10:04:27Unhelpfulkugel: 0 gain on targets that don't get a scaler would be ideal
10:05:01kugelsure, but ideal isn't always possible
10:05:37kugelbut, somehow, seperating sounds good to me
10:08:00Unhelpfuli don't like the idea of having two different bmp loaders, really - it sounds like a potential maintenance headache, since changes in one will have to track the other.
10:08:44Unhelpfulcould just strip the mono-target-only stuff from the patch, fork rockbox per Zagor's suggestion, and commit scalers in the rockbox-new fork ;)
10:10:06kugelUnhelpful: the common code gets in a third file of course
10:10:41kugel(like it's done with the list code, and bitmap and charcell lcds)
10:10:49Unhelpfulthere is no common code, aside from read_bmp_file, which only opens the file and passes off the rest of the work to read_bmp_fd
10:11:01Unhelpfulor rather, no easily-managed-as-common code
10:11:30Unhelpfulall of the format conversion stuff moves into read_part_line, or into the scalers
10:11:46kugelbut to honest, if the mono code doesn't even get a scaler, it's deemed to stall anyway (as there don't seem to be bugs) -> no need to maintain
10:11:59gevaertsJdGordon: well, it prints the function name, so that changed too :)
10:12:20JdGordonok, i removed it anyway
10:15:00Unhelpfuleven the NN scaler would add quite a bit more than 280B to the mono targets
10:15:16Unhelpfuland for pretty much no gain
10:16:57 Join crculver [0] (
10:17:09crculverI've found an issue with building the database.
10:17:34crculverOccasionally, once I've updated the database a few times, the last artist will show up twice in the list.
10:21:17*JdGordon is on a rampage... 2 bugs.. 2 manual patches!
10:21:28B4gdergo go go
10:30:43crculverAlso, one album, recognized as one single album in other tag-based media players, is split into three identically titled albums in the database.
10:33:08 Quit Virtualdistance (Read error: 110 (Connection timed out))
10:33:58kugelJ-23: you may have a look at
10:34:56kugelbasically, the same can be done for the e200v2. just copy button-fuze.c into button-e200v2.c after applying, and change the e200v2 according to my comments
10:35:12J-23I see that Rockbox doesn't boot for me, because I have an e280.
10:35:46kugelrockbox should boot at least on the e200v2
10:36:17kugelJdGordon: so, what's going to happen about my cleanup patch?
10:36:29JdGordonprobably sit for a while...
10:38:04linuxstbrasher: Looking at FS #9553, I can't imagine that all those extra spaces are intentional. Do you know if the patch was created with your web interface (where extra spaces would probably be harder to spot)?
10:38:26kugelJdGordon: why should it sit?
10:38:48JdGordonkugel: the problem is that the bars are currently tied to screens... they shouldnt be... what I started to do was setup a screen manager which redraws the bars automatically (if the current screen wants them shown).. unfortunatly its not working
10:40:16rasherlinuxstb: It was (the patch has a header)
10:40:38kugelJdGordon: and my patch hinders the work on that?
10:40:45linuxstbrasher: So there is ;)
10:41:17kugelI see that my patch improves the svn code no matter what's going to happen with the bars in further developement, so I see no reason to wait for anything
10:41:34*JdGordon never said it does
10:42:11kugelthen you didn't answer my question about why it should sit
10:46:48JdGordonhow about... we are about to go into a feature freeze and this may add extra bugs which I cant be bothered with right now?
10:47:02*JdGordon is working on a proper solution anyway
10:47:44***Saving seen data "./dancer.seen"
10:52:39 Join kachna [0] (
10:56:35kugelJdGordon: my patch is proper imho. What you're working on something completely different (to me at least). and, I'm responsible for bugs and I'm willing to fix them
10:56:53*JdGordon really hates people using imho
10:57:19kugelfeature freeze isn't meant to actively reject patches way before the freeze starts
10:57:35JdGordonthe fix is for a non-issue though
10:57:57kugelit isn't even a fix. it's just clean up
10:58:56JdGordonwhich 1) adds extra work later, 2) could add bugs, 3) isnt needed *now*
10:59:03kugelanyway I'm out. I'm tired of hearing "i like that" or "i like the idea" before I start working, to hear afterwards "no, sorry"
10:59:13 Quit kugel ("ChatZilla 0.9.84 [Firefox 3.0.4/2008111318]")
10:59:57 Quit perrikwp (" ajax IRC Client")
11:00:45 Join perrikwp [0] (i=18ac0c41@gateway/web/ajax/
11:14:08 Join culture [0] (
11:29:05 Join kugel [0] (n=chatzill@unaffiliated/kugel)
11:29:24kugelJdGordon: how does it "add extra work later"?
11:30:24*JdGordon is very afk
11:30:26kugelalso. How do you explain that SVN isn't in need for a cleanup, which does things properly, theoretically speeds up things and even reduces binsize a bit?
11:31:24kugelJdGordon: I see it this way: You're too egoistic to accept contributions by others in your area. You just wait until you get your next big change working
11:31:37kugelif not, prove me wrong
11:33:09*rasher doesn't see this line of conversation going anywhere useful
11:33:57scorchekugel: need to go that extreme..
11:34:30Unhelpfulsorry, forgot to *actually* upload the stupid patch before. it's up now. :P
11:35:25Jabonehi, is there some kind of roadmap what is made so far for e200v2?
11:36:00scorcheJabone: not really
11:36:12B4gderJabone: the wiki page and the forum threads are the best places to get the latest details on that
11:37:09 Quit kugel (Remote closed the connection)
11:41:05 Join n1s [0] (n=nils@rockbox/developer/n1s)
11:43:23 Join kugel [0] (n=chatzill@unaffiliated/kugel)
11:44:24kugelrasher, scorche: I know it's a bit harsh, but it's my opinion formed by many discussions like this with him
11:45:33scorchekugel: so that justifies a personal attack?
11:45:44 Quit nuonguy ("This computer has gone to sleep")
11:47:42rasherI'm not even considering whether it's justified or not, but it's certain that it won't accomplish anything.
11:47:57scorchecertain indeed
11:58:41 Quit skipper (Remote closed the connection)
12:11:44n1sWhere are we still using the sysfont in the UI?
12:13:06n1sany ipod 4G users that can say if rockbox displays the charger plug icon when a charger is connected?
12:13:34 Join tyfoo [0] (
12:15:26kugeln1s: recording screen in very small screens
12:15:48n1sah, right that is still kind of nasty...
12:15:51kugeland debug screen in some places
12:16:06n1sdebug is not important since it cannot be translated
12:17:05*n1s want's a ipod video user to test FS #7419 "iPOD freezes when inserting USB-cable after using test_codec plugin"
12:21:10 Join musicaltime [0] (i=BLALALLA@
12:23:58 Join Darksair [0] (n=user@
12:28:11 Join Nico_P [50] (n=nicolas@rockbox/developer/NicoP)
12:38:15 Quit Darksair (Read error: 60 (Operation timed out))
12:47:24 Quit kugel (Remote closed the connection)
12:47:45***Saving seen data "./dancer.seen"
12:59:28*n1s thinks superdom has a bug, a negative number of men makes no sense...
13:00:25n1showever they eat negative food too apparently o_O
13:00:46 Join jbesclapez [0] (n=chatzill@
13:03:39n1sAnd now the interface in the plugin hung but the music playback in the background is still working fine...
13:03:52jbesclapezHello guys! I have nothing to say. It is a simple post to say thank you to all Rockbox Developers. I am using it for 4 years I think now and I hardly updated it. Today I decided to run the latest V3. It is beautiful guys! Thanks a lot for all what you are doing. Thanks for giving me the opportunity to feel free! (no Apple OS - no Itunes). Thanks again. Nic
13:08:19B4gderjbesclapez: you're welcome!
13:08:31jbesclapezi really mean it!
13:08:48jbesclapezthanks B4gder if you are one of them :-)
13:09:25B4gderI am one in the herd
13:09:44jbesclapezI will now test it...
13:09:56jbesclapezas I said I am already impressed by its load...
13:10:15jbesclapeza lot easier than the former ipodlinux... (even if it was not "really" complicated)
13:10:30jbesclapezit is the ALL in ONE package taht is a brilliant idea.
13:10:55jbesclapezI already know where i "could" be disapointed thought
13:11:01jbesclapezthe MP4!
13:11:09jbesclapezbut I knew it before installing
13:11:17jbesclapezit is not working !
13:11:20B4gderyou mean mp4 as in movies?
13:11:25jbesclapezyep sir
13:11:38B4gderright, then you need to convert them to the format rockbox likes
13:11:41jbesclapezi think it is a copyright
13:11:52jbesclapezissue i mean... sorry
13:12:23jbesclapezYep but the drawback of this is that the rocbkox format cannot be read on other things can it? I never tried it to be honest...
13:12:26jbesclapezi wil now...
13:12:32B4gdersure it can
13:12:37B4gderrockbox plays "plain" mpeg2
13:12:43n1sit's standard MPEG1/2
13:12:48jbesclapezmpeg2 and 1 i knew...
13:12:56jbesclapezeven in previous version I remember...
13:13:09B4gderwhat previous version?
13:13:17jbesclapez2 i think was...
13:13:32jbesclapezi dont remember exactly
13:13:48jbesclapez:-) it is a long time ago now i started using your thing!!
13:14:14jbesclapezAnd i mix it with other freeware like juices for the podcast... a beautiful combination!
13:15:37jbesclapezB4gder, did you ever got a mail from the SongBird community? Coz I told them about your project and to contact you a year ago... their thing was ok. But combined with rockbox could have been a good idea...
13:16:00B4gderI don't think we have, no
13:21:29jbesclapezI dont know if you are aware but the installer is saying that the "themes" can not be installed because "the themes site is being reworked". Just FYI.
13:23:27B4gderyes, it is being reworked
13:24:44 Join linuxstb_ [0] (n=linuxstb@rockbox/developer/linuxstb)
13:25:31 Quit linuxstb (Read error: 110 (Connection timed out))
13:26:05 Nick linuxstb_ is now known as linuxstb (n=linuxstb@rockbox/developer/linuxstb)
13:26:10jbesclapezGosh! The themes look really cool now!
13:26:27jbesclapezI miss one that i used to have, a lookalike MACOS!
13:27:40 Quit kachna (Read error: 110 (Connection timed out))
13:38:47 Join linuxstb_ [0] (n=linuxstb@rockbox/developer/linuxstb)
13:50:16 Quit linuxstb (Read error: 110 (Connection timed out))
13:52:32 Join moos [0] (
13:53:52 Join Seed [0] (
13:54:09 Quit Seed (Read error: 54 (Connection reset by peer))
13:54:55 Join Seed [0] (
13:56:44 Join XavierGr [0] (n=xavier@rockbox/staff/XavierGr)
14:03:13n1swow, this stand alone database builder is very broken :/
14:04:31Zagorn1s: oh yes
14:06:27 Quit einhirn ("Miranda IM! Smaller, Faster, Easier.")
14:06:55Zagorn1s: the brokeness, and futility of fixing it, was the reason I did the big metadata rework a few weeks back
14:07:39n1saha, still feels pretty futile to me :)
14:07:42 Quit jhMikeS (Nick collision from services.)
14:07:48 Join jhMikeS [50] (n=jethead7@rockbox/developer/jhMikeS)
14:07:52B4gderremake the build system again!
14:07:54n1sfixing one error gives a hundred new ones
14:08:29Zagorn1s: iirc it needs a largish overhaul.
14:08:42ZagorB4gder: metadata, not makefile...
14:09:03*B4gder polishes his glasses
14:09:06n1sand since i have no use for it myself i will give up now...
14:10:06 Nick fxb__ is now known as fxb (
14:23:24 Join funman [0] (
14:33:09 Join Casainho [0] (
14:33:18CasainhoHello :-)
14:33:24CasainhoBagder: are you here?
14:33:54B4gderI'm around
14:34:58Casainhowell, do you remember last time I were here? that I had a problem of my system hanging when it was just do the code: "thread->l.prev = l-l.prev;"
14:35:09B4gderI do
14:35:45Casainhowell, I remember that some developer suggested that could be a problem because of SDRAM memory timmings
14:36:04B4gderthat developer was me, yes
14:36:37CasainhoI tested the same code but using a bootstrap that configures the SDRAM, load the code to it and finally run it −− the code do not hang like that
14:36:50Casainhoah, didn't remember it was you :-)
14:36:52JdGordonNico_P: see the patch for the playback bug? care to have a quick look to double check it?
14:37:12B4gderCasainho: so then it does sound like sdram issues?
14:37:42Casainhobut code hangs if a use Sam-ba, on utility for PC that loads the code trough USB to SDRAM (for sure that needs to initialize it before)
14:37:45Nico_PJdGordon: which patch is that?
14:37:52Casainhoyes, sounds like that
14:38:06JdGordonNico_P: 7423
14:38:27CasainhoI was just passing to say this :-)
14:38:29Casainhothanks :-)
14:38:40funmanCasainho: you're progressing, good ;)
14:38:41Nico_PJdGordon: I hadn't seen it. I'll take a look
14:38:44B4gderno problemos!
14:38:55Casainhobye bye :-)
14:38:57JdGordonNico_P: although that one is mostly playlist... also care to double check the commit i did before?
14:38:58 Quit Casainho ("ChatZilla 0.9.84 [Firefox 3.0.4/2008102920]")
14:39:08B4gderglad to hear it's at least almost solved
14:40:17 Quit perrikwp (" ajax IRC Client")
14:40:44n1sdoes anyone think FS8142 should remain open? any ipod video users experience this?
14:40:54 Join perrikwp [0] (i=18ac0c41@gateway/web/ajax/
14:41:01Nico_PJdGordon: your commit looks ok
14:42:49Nico_Pabout your fix for FS #7423, I don't know the playlist code well enough to say
14:43:58funmanNico_P: in buffering.c you check the return value of RINGBUF_ADD_CROSS. Sometimes it fails with > 0 , sometimes with >= 0 ; but I understand that a return value of 0 means "just enough room available".
14:45:03funmanPlus for readability, I think the macro arguments should be (widx,size,ridx) instead of (p1,v,p2)
14:45:50funmanDo you have some time to give me, or do you prefer to discuss that later?
14:45:59 Quit jbesclapez (Read error: 110 (Connection timed out))
14:46:04Nico_PI have plenty of time today :)
14:46:46B4gderform a line guys, Nico_P will take everything you can deliver
14:46:54*B4gder hides
14:47:46***Saving seen data "./dancer.seen"
14:50:17Nico_Pfunman: >= or > really depends on the context... e.g. in buffer_handle >= is used as an end test and > is used to check for actual overlap
14:51:28funmanah the comments describe that
14:51:57Nico_Pabout the readability I agree, but the use isn't always exactly what you said
14:52:30funmannow I can see that
14:56:35 Join LambdaCalculus37 [0] (i=44a04303@gateway/web/ajax/
15:00:43 Join robin0800 [0] (
15:00:50 Join mcuelenaere [0] (n=mcuelena@rockbox/developer/mcuelenaere)
15:04:31 Join Horscht [0] (n=Horscht@xbmc/user/horscht)
15:07:54JdGordondoes anyone have any ideas why if I call queue_broadcast(SYS_MYNEWEVENT,0); in a tick task it doesnt end up getting to the default event handler ? (where I'd expect it to get through the button queue))
15:09:28mcuelenaeregevaerts: I'm trying to get USB working on my VX747. Does mean there's something wrong with sending data to the PC?
15:21:31 Join Schmogel [0] (
15:23:05 Join saratoga [0] (n=9803c6dd@gateway/web/cgi-irc/
15:32:37 Join EspeonEefi [0] (i=eefi@SYDNEYPACIFIC-SIX-NINETEEN.MIT.EDU)
15:33:58 Nick JdGordon is now known as JdGordon|zzz (n=jonno@rockbox/developer/JdGordon)
15:35:56 Nick notplus_M is now known as plus_M (
15:36:44 Join nplus [0] (n=nplus@141.25.Globcom.Net)
15:40:26 Join m0f0x [0] (
15:40:37funmanIs there a build table for simulators ?
15:41:23Zagorfunman: the sims are included in the big table
15:42:16*funman rotates head on 90°
15:47:47 Quit Schmogel ("Miranda IM! Smaller, Faster, Easier.")
15:50:48 Join miepchen^schlaf [0] (
15:52:47funmanDebugging recording code shows me that we would need to rewrite VIC code
15:53:55gevaertsmcuelenaere: probably.
15:54:29mcuelenaerebut doesn't the fact that Windows sends SET_ADR means it at least received a part of the data?
15:54:44gevaertshm, good point
15:56:06gevaertsI think we need to get this usb analyser to you
15:56:40mcuelenaereor the DAP to you
15:56:57gevaertsThat could also work :)
15:58:59mcuelenaerecould the lack of ACK'ing the host requests cause this?
16:00:16 Join obo [0] (n=obo@rockbox/developer/obo)
16:00:19 Join obo_ [0] (
16:00:22 Join nicfer [0] (n=usuario@
16:00:26 Quit obo_ (Remote closed the connection)
16:02:07 Part nicfer
16:11:21 Part B4gder
16:15:16 Quit miepchen^schlaf ()
16:20:52 Quit GodEater (" ajax IRC Client")
16:20:58 Join GodEater [0] (i=c2cbc962@gateway/web/ajax/
16:22:29 Join miepchen^schlaf [0] (
16:27:08 Join kugel [0] (n=chatzill@unaffiliated/kugel)
16:35:44 Join stoffel_ [0] (
16:37:51 Join bmbl [0] (n=Miranda@unaffiliated/bmbl)
16:40:14kugeljhMikeS: beast still defines HAVE_LCD_ENABLE, right? Even though the function isn't implemented
16:41:49mcuelenaeregevaerts: device enumeration only uses EP_CONTROL, right?
16:42:45 Join _lifeless [0] (n=lifeless@
16:47:47***Saving seen data "./dancer.seen"
16:48:44kugelrasher: Do you need the source to generate voice files (i.e. the .lang files) or can they be built using the .lng files after compilation?
16:51:09rasherkugel: You need the .lang files. The .lng files only contain display strings
16:51:10n1skugel it's inside #if 0 so no, see;r2=19322;pathrev=19322
16:52:10kugeln1s: ah ok, I didn't see that. It's just what I've noticed during my backlight fading wort
16:52:46kugelJ-23: copy the entire button-fuze.c into button-e200v2.c
16:52:56 Join toffe82 [0] (
16:53:09kugeland take a look at lcd-fuze.c how I added GPIO*_AFSEL stuff into it
16:53:19kugelthat should be done for e200v2 display too
16:53:26kugeli.e. lcd-e200v2.c
16:53:35jhMikeSkugel: No, I took that out. Only HAVE_LCD_SLEEP is defined now.
16:53:59kugeljhMikeS: ye, n1s already explained :)
16:54:27*jhMikeS looks again and aha...indeed
16:54:48 Quit __lifeless (Read error: 110 (Connection timed out))
16:55:02 Join Benni [0] (n=50993726@gateway/web/cgi-irc/
16:55:26n1sjhMikeS: any estimate on how much power turning off the LED supply saves?
16:55:27kugeljhMikeS: though commenting out would be more in config-* style than #if 0'ing ;)
16:55:28 Quit Benni (Client Quit)
16:55:31 Join BenniL [0] (n=50993726@gateway/web/cgi-irc/
16:55:55jhMikeSn1s: not much. less than 1mA but it's something.
16:55:59J-23will Rockbox work without these lcd-e200v2.c modifications?
16:56:02gevaertsmcuelenaere: right
16:56:30n1sjhMikeS: true, do you know if we have other supplies needlessly enabled?
16:56:36kugelJ-23: quite unlikely
16:56:38*jhMikeS is getting confused. Guidelines prefer #if 0 and kugel says to comment.
16:56:42BenniLHi all, I'd like to know when usb support on the sansa E200v1 is going to get officially activated. Are tehre still problems?
16:57:20saratogaBenniL: when it works
16:57:24J-23kugel: I don't understand them, should I just add these lines to lcd-e200v2.c in same places they're in lcd-fuze.c?
16:57:33kugeljhMikeS: really (I'm only talking about the config-* files)? I hardly see #if 0's in there, but quite some commenting out. Maybe those files just don't follow the guidelines
16:57:37 Quit musicaltime ()
16:57:42jhMikeSn1s: Nothing I can think of offhand. It's been pretty thorough to shut things off.
16:57:45kugelJ-23: pretty much, yes
16:58:30BenniLsaratoga: I remember shortly before GSoc the code was already usable except a few problems, didn't hear about usb support for some time now
16:59:02n1sBenniL: the problems remain unfixed
16:59:29BenniLok, i hoped that wasn't the cause
17:00:30 Join Notlistening [0] (
17:01:03n1sBenniL: when it works reliably it will be enabled by default in builds
17:01:17 Join dany_21a_ [0] (
17:01:30NotlisteningI am building th bootloader for rockbox and have justed added a sleep into the main for a test and get this error on linking any ideas In function `panicf':
17:01:30Notlisteningpanic.c:(.text.panicf+0x60): undefined reference to `lcd_update'
17:01:30Notlisteningpanic.c:(.text.panicf+0x64): undefined reference to `system_reboot'
17:01:30DBUGSent KICK Notlistening to server
17:01:30Notlisteningcollect2: ld returned 1 exit status
17:01:30Kick(#rockbox Notlistening :No flooding!) by logbot!n=bjst@gateway/web/cgi-irc/
17:01:48J-23IRC bot through CGI-IRC gateway?
17:01:55 Join Notlistening [0] (
17:02:07 Quit crculver (Read error: 110 (Connection timed out))
17:02:18n1sNotlistening: did it build before your modification?
17:02:26n1si.e. a clean svn?
17:02:58Notlisteningit is quite an old download from the SVN
17:03:14n1sNotlistening: which player is this for?
17:03:26dany_21a_hi, kugel: did your fuze boot into mainmenu after you applied your patch?
17:03:39kugeldany_21a_: yes
17:03:43kugeldany_21a_: err no
17:03:49dany_21a_hm, strange mine did not
17:03:51Notlisteningbut we have been building a new port and adding the sleep function stop it from building
17:03:51kugelthat's why I struggled
17:03:55dany_21a_oh.. okay... mine too :)
17:04:09 Quit Zagor ("Client exiting")
17:04:16 Join crculver [0] (
17:04:26dany_21a_but i managed to get it booting only with GPIOC in irq-mode - but you said, you also did that
17:04:35 Quit crculver (Client Quit)
17:04:43dany_21a_and it already sensed the power button correct
17:04:44kugeldany_21a_: I don't think I did
17:04:51n1sNotlistening: this seems weird, do you have your own implementation of sleep() ? i don't think any implementation in our svn reference panic()
17:04:53kugelI managed to get GPIOB working
17:05:17 Quit BenniL ("CGI:IRC (EOF)")
17:05:18dany_21a_oh... i meant GPIOB... sry...
17:05:19kugelGPIOC only didn't work for me
17:05:22kugelah ok
17:05:48Notlisteningnope we are just using the basic rockbox SVN files
17:06:43 Join nicfer [0] (n=usuario@
17:06:46dany_21a_i am currently trying to understand the VIC-releated functions from the OF (its setup and irq-handling) −−- but no great success
17:07:06n1sNotlistening: the svn version does not reference panicf() so you must have done something on your side which makes it try to link panicf() which obvioulsy needs lcd_update()
17:07:07 Part nicfer
17:07:29dany_21a_funman: what did you mean, to rewrite the IRQ/VIC code? What errors did you find?
17:07:43kugeldany_21a_: one idea I had was not to read GPIOs directly. I haven't tested it, but I think actually GPIO*_MIS indicates which pin issued the interrupt
17:08:17Notlisteningright i will have to ask the other developer in that
17:08:39Notlisteningbut the line sleep(100); causes it
17:08:44kugeldany_21a_: can you try it read the power button without reading the pin directly?
17:09:03kugelNotlistening: I think sleep isn't implemented for the bootloader. Use busy loops
17:09:16dany_21a_kugel: i tried to switch off interrupts while in LCD-routines and clear GPIO?_IC afterwards... so if datatransfers in DBOP mode over xpb and xpc-pins wont trigger wrong IRQs
17:09:20kugele.g. int delay = 100000;while(delay−−):
17:09:23dany_21a_but no change in behaviour
17:10:05Notlisteningthanks kugel I will try that it is only for testing anyways
17:10:12kugeldany_21a_: irq's are definitely triggered without setting afsel before. You cannot read the pin though
17:10:40kugels/definitely/afair/. I'm pretty sure I tested that
17:11:02Notlisteningumm another quick one is there a way to test from the bootloader that the kernel_init us working correctly?
17:11:51kugeldany_21a_: dbop won't trigger gpio* interrupts, since dbop has it own interrupts (that's how I understood that)
17:12:53kugeldany_21a_: maybe I'm wrong. can I see what you exactly did?
17:13:40 Join Casainho [0] (
17:14:03 Quit Casainho (Client Quit)
17:14:12kugelNotlistening: what are you working on, just curious
17:14:23kugelaren't you one of the rockboxplayer troop?
17:14:37Notlisteningah you know the rockboxplayer
17:14:46Notlisteningguilty as charged :P
17:14:50kugelof course I do, it's famous :)
17:15:22Notlisteninglol not enough and maybe for the wrong reasons :P
17:15:36kugelprobably the latter one ;)
17:16:11NotlisteningWe are working toward getting intergrated to the main project ASAP
17:16:33Notlisteningbut as our questions might indercate we have a lot of learning to do
17:17:07Notlisteninglol ohh bad english there
17:17:11kugelno, I'm really looking interested forward to that project. I'd really like to see a dap made for rockbox, and I highly appreciate your effort
17:17:21J-23argh. Can't you just include lcd-e200v2.c changes in your patch?
17:18:02NotlisteningWell fingers crossed
17:18:18kugelJ-23: you should basically put the GPIO*_AFSEL in the beginning and end of lcd_update,lcd_update rect, lcd_enable and display_on
17:18:47Notlisteningit is a little painful at the moment as we are learning
17:19:07kugeljust keep up, things can only get better
17:23:50Notlisteningam i right to think that after we call kernel_init it is possible to test the kernel in someway or am i wrong?
17:23:54*kugel eagerly waits for dany_21a_ to come back
17:30:37 Nick linuxstb_ is now known as linuxstb (n=linuxstb@rockbox/developer/linuxstb)
17:30:52dany_21a_just came back
17:31:29 Join herrwaldo [0] (
17:31:47 Quit tyfoo (Read error: 54 (Connection reset by peer))
17:31:52dany_21a_i meant, dbop uses the same output pin as GPIOB/C - and maybe if the irq is set to trigger on changes on that pin maybe the output of DBOP triggers the GPIOB/C irq
17:32:06Notlisteningwould this work? Tick tasks
17:32:06NotlisteningYou can declare a function to be called on every kernel tick. It will be executed in interrupt context and has to be very short and simple.
17:32:06Notlisteningint tick_add_task(void (*f)(void))
17:32:30dany_21a_kugel: ^
17:33:35 Join jgarvey [0] (
17:34:59 Join tyfoo [0] (
17:35:54dany_21a_kugel: and i also noticed that the display often crashes if i set GPIO?_AFSEL=0x0... now i only set that bits to 0 where buttons are attached... so that the other pins (unrelevant to buttons) stay in DBOP mode
17:36:11kugeldany_21a_: gpio doesn't trigger on output afaik. output means sending something to the hardware
17:36:39dany_21a_but i also get some lockups now and than (only with GPIOB enabled, with GPIOB+C=lockup after the first IRQ)
17:37:08funmandany_21a_: hi ! i want to use vectored interrupts because it's faster than checking VIC_IRQ_STATUS
17:37:30funmani'm currently debugging why the irq handler isn't called when using vectored ints
17:37:41dany_21a_kugel: DBOP changes outbut-pins (to transfer data to the LCD) -> GPIO? senses them and might (it looked to my that way) trigger an IRQ
17:38:03funmanbtw it seems the OF doesn't use IRQs
17:38:25funmanperhaps it uses a matrix scan like in m200v4 and clip, and this is why you have inconsistent results
17:38:42dany_21a_funman: hi.. that would be cool... just read a lot in the related datasheet, but i dont understand how the OF sets up the addresses (wanted to see the ISR for GPIOx)
17:38:47kugeldany_21a_: funman, I think you need to give the address of the isr if you're using vectored irq
17:38:57kugeldany_21a_: sorry, that wasn't for you
17:39:07funmandany_21a_: there is no address setup, it just uses the default irq handler and VIC_IRQ_STATUS
17:39:33funmankugel: right, but then the irq handler reads the isr address from VIC_VECT_ADDR
17:40:02kugeland that's not working for you?
17:40:07funmanthe irq handler is not called
17:40:51 Quit n1s ()
17:41:05kugelfunman: how vital is checking VIC_IRQ_STATUS, and what do you actually check? I haven't such a check in my button code
17:41:23funmaneven if I met the requirements : interrupt is enabled in VIC_INT_ENABLE ; set to generate an IRQ (not FIQ) in VIC_INT_SELECT, enabled in VIC_VECT_CNTL[x]
17:41:39funmankugel: you don't have to bother with VIC registers in your specific isr
17:41:54kugelah ok
17:42:24funmanirq_handler() will just do "the right thing" and call the specific isr, where you only do your specific stuff (GPIO in your case)
17:43:03dany_21a_funman: which source do you use to trigger an irq?
17:43:28funmanDMA (I didn't modified code elsewhere than in system-as3525.c) for SD transfers
17:44:02funmanI didn't modify the bootloader though, and my clip doesn't display anything else than the bootloader splash bitmap
17:44:37funmanoops .. VIC setup is only done in the bootloader ..
17:44:53dany_21a_wrong side of the else?
17:45:26funmanno, when I try to redo the system_init in the main rockbox it "doesn't work"
17:48:04 Quit krazykit ("Connection reset by beer")
17:53:29 Join krazykit [0] (n=kkit@
17:55:59funmanall right I can use vectored interrupts, but I think I broke the bootloader somewhere else ;)
17:59:52funmanNote that there is only 16 vectored interrupts, so it requires more setup if we want to use CGU (we don't), DBOP (we don't), MEMORY_STICK (we don't), and GPIO (it's up to you)
18:01:01 Quit tyfoo ("Carpe diem")
18:05:10 Join domonoky [0] (n=Domonoky@rockbox/developer/domonoky)
18:07:20kugelfunman: tell me more :)
18:08:46kugellet me try vectored interrupts
18:08:54funmanthis won't change anything
18:09:11funmanfor me the key is : if OF doesn't use interrupts, you shouldn't
18:09:58kugelfunman: amiconn says a isr is vital for the scrollwheel to work. every other targets use isr (even the v1 sansas)
18:10:08kugelevery other targets with scrollwheel that is
18:10:28funmando you know if the wheel is read by gpio ?
18:10:42kugelfunman: also, I looked into the disassembly. And I didn't see using the OF direct gpio reading
18:11:05kugelfunman: on the other targets *every* button uses the isr, not only the wheel
18:11:43domonokykugel: which other tragets, we have both, interrupt and tick task driven button drivers in rockbox..
18:12:10kugelthe scrollwheel targets, which I already referred to in my previous sentence
18:12:22domonokyah "every target with wheel" .. :-)
18:12:25funmankugel: that doesn't mean at all that the same method is used on the fuze
18:12:37domonokybut that also probably depends on howth wheel works..
18:12:42kugelfunman: show me where the of reads gpios directly
18:12:53 Join faemir [0] (
18:12:59funmankugel: refer to my message 2 minutes ago
18:13:21kugelthe only references in the of to that gpio pins are related to lcd functions, as far is I could see
18:13:30funmanyou assume that the wheel is read using GPIO, perhaps this assumption is wrong.
18:13:41kugelI don't assume anything regarding the wheel
18:13:58kugelbut I trust in amiconn when he says an isr is vital for scrollwheels to work
18:14:11funmanBut not for other buttons
18:14:25kugelalso. It's not that I didn't try reading without irq
18:14:26funmanIn the case of matrix scanning for example, sensing the buttons with interrupts is not possible
18:14:37 Quit perrikwp (" ajax IRC Client")
18:14:54funmanIMO more RE is needed
18:15:09domonokyif a scrollwheel send incremental values, a isr is needed, to not miss increment changes. If the wheel sends absolute values, a tick task should be sufficient..
18:15:33 Quit EspeonEefi ("さよなら")
18:17:53 Join skipper [0] (
18:23:57 Join karashata [0] (n=karashat@
18:26:35 Quit stoffel_ (Read error: 113 (No route to host))
18:32:58 Quit basti (Connection timed out)
18:33:28 Join basti [0] (
18:35:44 Join {phoenix} [0] (
18:36:44 Join perrikwp [0] (i=98213efd@gateway/web/ajax/
18:47:20 Join zeuzpr [0] (
18:47:27zeuzprhi guys
18:47:40funmankugel: dany_21a_ I have good news for you
18:47:48***Saving seen data "./dancer.seen"
18:47:59funmanchatting with kugel I just noticed my "fuze" disassembly was in fact m200v4
18:48:13funmanin the fuze OF I confirm that there is 2 isrs for GPIOA and GPIOB
18:48:46 Join stoffel_ [0] (
18:49:10zeuzpri have an ipod 5g with a CF
18:49:21funmanin v1.01.11, at offsets 0x3D00 (gpioa) and 0x3CE4 (gpiob) - now to your disassemblers !
18:49:27zeuzprand when loading rockbox ATA error: -1
18:50:09zeuzprany sugestion??
18:50:23 Quit AndyI ()
18:50:34kugelfunman: ...
18:50:40J-23kugel: Rockbox doesn't boot on my e280v2 with your modifications and with sound patch
18:50:57kugelJ-23: that doesn't surprise me
18:51:01kugelmy fuze doesn't boot either
18:51:24funmankugel: if you feel bad against me, I just feel even more bad than nobody checks my findings
18:51:25kugeltry to comment out the GPIOC_IE line to see if it boots
18:51:52kugelfunman: I'm just as of now looking through the of
18:52:21kugelI find it a bit sad that kinda all of your previous findings are based on a wrong of though
18:52:34kugelfunman: why should I feel bad against you?
18:52:56J-23kugel: in what file?
18:53:03funmankugel: don't you know that multiple dots at the end of sentences make the interlocutor feel bad ? ;)
18:53:04kugelJ-23: button driver
18:53:22kugeldomonoky-style. sorry
18:53:32funmankugel: no problem ...
18:53:40*domonoky ... :-)
18:54:03kugelfunman: I'd like to know though if it's using a isr on GPIOC as well, since that's the one I can't get to work
18:54:24kugel(I'll look myself in a few minutes)
18:54:38funmanI don't see that
18:55:12funmaninterrupt lines 1 4 3 14 13 5 7 12 0 29 30 (not 31 == GPIOC)
18:56:41funmanthe table is { isr_offset, interrupt line } and is located just after the strings "HEADER"
18:56:44 Join AndyI [0] (i=AndyI@
18:56:51funmansingular string*
18:57:53 Join nuonguy [0] (
18:57:55*J-23 recompiles
18:58:52 Join aarcane [0] (
18:59:07 Join avis [0] (n=ident@pdpc/supporter/student/avis)
18:59:16J-23kugel: didn't help
19:00:24kugelJ-23: then you probably did something wrong
19:00:39kugelfunman: I can't find any reference to vic base
19:00:52J-23should I paste these files somewhere?
19:01:06funmankugel: what I gave you is the isr, you don't have to care about vic controller
19:01:08 Quit zeuzpr (Read error: 60 (Operation timed out))
19:01:10 Join zeuzpr [0] (
19:01:33J-23or maybe only lcd-e200v2.c
19:01:35kugelJ-23: Do you think you can help with coding the isr? If not, stay away from the patch, since it doesn't work (sorry to be rude, but the patch isn't meant to be used, but as a basis for developement)
19:02:17kugelfunman: but I should still find something with vic_base, not?
19:03:41NotlisteningI have checked with another players bootloader and it calls the sleep function the mrobe uses it in the USB function does this mean it has customised files ?
19:03:58funmankugel: 0x7C1C (ARM code, not thumb) - how are you searching ?
19:04:22kugelthumb, in the objdump disassembly
19:04:53funmanNotlistening: sleep() is defined in firmware/kernel.c and doesn't seem to be dependant on bootloader or not
19:05:04 Quit m0f0x (Read error: 110 (Connection timed out))
19:06:02Notlisteninghumm having problems with it :P
19:06:55Notlisteningback to the drawing board
19:07:47 Join kachna [0] (
19:10:57domonokysleep() is dependend on having a working tick i think.. if the tick works, sleep() should work.
19:13:52 Join webguest08 [0] (n=d9c70784@gateway/web/cgi-irc/
19:14:17 Quit webguest08 (Client Quit)
19:14:24 Quit culture (Read error: 104 (Connection reset by peer))
19:17:42 Join EspeonEefi [0] (i=eefi@SAFFRONCITY.MIT.EDU)
19:21:21 Quit EspeonEefi (Client Quit)
19:21:28 Join EspeonEefi [0] (i=eefi@SAFFRONCITY.MIT.EDU)
19:22:57 Quit EspeonEefi (Client Quit)
19:25:33 Join EspeonEefi [0] (i=eefi@SAFFRONCITY.MIT.EDU)
19:26:35 Join n1s [0] (n=nils@rockbox/developer/n1s)
19:27:58 Quit EspeonEefi (Client Quit)
19:29:37 Quit zeuzpr (Read error: 110 (Connection timed out))
19:30:20 Join jhulst [0] (n=jhulst@
19:33:16 Quit moos ("Rockbox rules the DAP world")
19:34:35 Join zeuzpr [0] (
19:37:19funmanhm strncat isn't available in the boot loader :/
19:38:26funmanwell, it's not defined at all, only declared in string.h
19:41:06 Join bertrik [0] (
19:42:29funmanAh I manage to get an RX overrun in SD transfer :'(
19:44:03 Quit denes_ (Read error: 110 (Connection timed out))
19:44:34 Join denes_ [0] (
19:51:47 Quit perrikwp (" ajax IRC Client")
19:53:24 Quit avis (Remote closed the connection)
19:54:49 Join perrikwp [0] (i=98213efd@gateway/web/ajax/
19:55:03bertrikhi funman, I wonder if the stable parts of the ams playback could be committed
19:55:59funmanbertrik: I think so, but what are the "stable parts" ? I mixed a lot of code in the patch on FS #9592
19:56:56funmanCurrently I have in my tree : playback, definitions of clock freqs, vectored interrupts, more debug code for SD, ..
19:57:01bertrikmaybe not the flash buffering code and the recording stuff, but perhaps the dma stuf
19:57:59saratogawas the flash buffering stuff needed to get playback working?
19:58:15funmanyes I think flash buffering and recording needs to be committed separately
19:58:21bertrikBTW, I noticed that in order to be able to use USB in the future, some of the generated clocks needs to be a multiple of 96 MHz
19:58:27funmansaratoga: yes, there is probably a bug in buffering.c for low mem buffers
19:58:34 Quit nuonguy ("This computer has gone to sleep")
19:59:09funmanbertrik: once the clocks are centralized in one file it's easier to test
19:59:09bertrikThe flash buffering stuff contains some half-done stubs as far as I can tell, I think we need to fix those before committing
19:59:48funmans/we/Nico_P/ ;)
19:59:58bertrik* before committing any flash buffering code I mean
20:00:11saratogain the long term would it be better to try and fix those bugs or to work on flash buffering you think?
20:00:48 Quit perrikwp (" ajax IRC Client")
20:01:46bertrikIn my current opinion the clip can and should use flash buffering and is probably a good test case to further improve flash buffering
20:02:44bertrikLooking the buffering thread debug screen, it seems that the buffer level can easily keep up (at least for the oggs I tried) and can probably be reduced further from 512 kB to say 128 kB
20:03:39funmanbertrik: where did you see that 96MHz requirement ?
20:04:11 Join Schmogel [0] (
20:05:34funmanhum using vectored interrupts I got spurious unhandled IRQs :/
20:07:10funmanbertrik: oh, the linker script needs probably more work, especially separation between lowmem targets (clip/m200v4) and higher memory (fuze/e200v2)
20:09:22bertrikfunman, page 109 shows that the USB divider is actually 2*n, where n is a configureable number in CGU_USB, so this works out to a clock that is a multiple of 96 MHz. We still have a spare PLL IIUC, so it probably doesn't need to influence the other clocks
20:11:06funmanthe best would be to use 384MHz PLLA (more flexible)
20:11:06 Quit stoffel_ ("leaving")
20:11:51funmanbertrik: because USB requires a 96MHz clock ?
20:13:32bertrikas far as I understand, either 48 MHz, or a multiple of 96 MHz
20:14:33 Join gregorovius [0] (
20:18:22 Quit zeuzpr ()
20:18:58bertrikI don't know exactly why this way of dividing the clocks was chosen
20:19:48funmanI mean why must USB clock be 48MHz or multiple of 96MHz ?
20:20:15 Join perrikwp [0] (i=18ac0c41@gateway/web/ajax/
20:23:07bertrikthe clock to the USB peripheral needs to be 48 MHz, and the divider logic to make this clock allows 1/1 in case n=0 or 1/(2*n) in case n > 0, so 48 MHz input clock for n=0 or 96*n for n>0
20:23:56funmanbut where is written this 48MHz requirement ?
20:28:30funmanis it in the as3525 datasheet, or is it a generic USB requirement?
20:28:42bertrikhmm, indeed it isn't so obvious, page 34 mentions it, and the clock itself also has 48m in its name
20:29:32funmanI should commit my clock code now, but it's hard to separate the different changes
20:32:59 Join n17ikh|Lappy [0] (
20:45:53funmangevaerts: ./mkamsboot m300a.bin bootloader-clip.sansa /mount/usb/m300a.bin
20:46:51funmanNow I wonder what I should do with this patch that changes pclk, commit and let m200v4/e200v2/fuze owners fix eventual problems, or put it on flyspray?
20:47:52***Saving seen data "./dancer.seen"
20:48:29NotlisteningWhat are the basic requirements to boot the kernel?
20:48:58Notlisteningie buttons lcd disk access etc
20:49:17 Join zeuzpr [0] (
20:49:51domonokythe kernel doesnt any of this, but you probably need some thing to confirm the kernel_init() was successful.
20:49:54funmanor look at the OF and see which DBOP frequency it uses
20:50:02 Quit gregorovius ()
20:50:35Nico_Pgevaerts: is USB serial possible on the e200?
20:50:48gevaertsNico_P: it should be, yes
20:51:02NotlisteningDomoney thats what i am after but not sure what to use?
20:51:46Notlisteningsprry about the typo :P
20:51:49Nico_Pgevaerts: thanks, I'll tyr it
20:52:02domonokyNotlistening: depends on what you have available. a simple LED is a nice debugging mechanism..
20:52:26zeuzprhi you all
20:52:40Notlisteningok got the led we also have a debug unit which can ouput to a com port that we can read
20:52:40zeuzpri need support with my ipod
20:53:03gevaertsfunman: I get some sort of screen corruption along horizontal lines (i.e. some text lines) are moved one or two pixels to the right). Is that known?
20:53:10zeuzprcan someone help??????????????????
20:53:35dany_21a_just ask, zeuzpr
20:53:40funmanzeuzpr: please use only one '?'
20:53:47domonokyNotlistening: even better.. just output someting to at after kernel_init to see if it didnt crash..
20:53:49funmangevaerts: no, it isn't known
20:54:02bertrikgevaerts, never seen that either
20:54:08gevaertsfunman: and now suddenly my screen is flipped vertically...
20:54:22funmangevaerts: did you apply any patch ?
20:54:29Notlisteningyeah that works fine
20:54:42funmangevaerts: I'm about to commit changes to clock frequencies.
20:54:54funmanwether that will fix your problem I don't know, but at least it'll make it easier to debug
20:55:15domonokyNotlistening: the you could also use a loop with sleep() to confirm the tick is running correctly.
20:56:06NotlisteningI try a sleep function and it doen't compile anymore
20:56:53bertrikis someone deleting messages from the sansa v2 forum thread? I keep getting notifications for new messages but I don't see them when I visit it.
20:57:55Notlisteningnot getting any bertik
20:58:05Lloreanbertrik: There's a lot of "congratulations" spam that various people are removing.
20:58:27LambdaCalculus37bertrik: I just removed one of those "congratulations" spam messages right now.
20:58:27bertrikLlorean, ok
20:58:56gevaertsfunman: gevaerts/screen.jpg">
20:59:13Lloreangevaerts: That's pretty neat looking.
20:59:40*LambdaCalculus37 rotates his head 180 degrees
20:59:51LloreanLambdaCalculus37: It's flipped vertically, not rotated.
21:00:05bertrikborked after 9 minutes of rockbox use already
21:00:12LloreanIt's still "fire" just upside-down f, i, r, and e.
21:00:19LambdaCalculus37Llorean: I see it now.
21:01:22*gevaerts is already getting annoyed by the keymap :)
21:01:35zeuzpri have a 5g ipod with 16gb cf
21:01:56Notlisteningahh i have some more information from the guy i am working with he has been commenting out stuff and not saying anything and that was making the kernel_unut wirj but breaking it I think
21:01:59zeuzprand when loading rockbox says ATA error: -1
21:02:41zeuzprbut loads fin in normal mode
21:02:47Lloreanzeuzpr: When did you first install the bootloader?
21:03:02LloreanDid you use the Rockbox Utility?
21:03:59LloreanYou'll need to file a bug report then. Not all CF cards work currently with Rockbox, and it'll probably take some investigation to find out what's different with yours.
21:04:08 Join casainho [0] (n=chatzill@
21:05:08funmangevaerts: what about r19330? you can tweak the DBOP frequency for Clip display
21:05:45zeuzprwhere can i file that bug report and how?
21:06:09zeuzprthe rockbox only work 1 time
21:06:12mcuelenaeregevaerts: could it be that because of timing, the VX747 doesn't get recognized? (I read that USB requires to respond (in most cases) in 50ms)
21:06:47gevaertsmcuelenaere: Could be. I'm not sure what the timeouts are
21:06:48Lloreanzeuzpr: The link "bugs" on the website.
21:07:28gevaertsmcuelenaere: however, usually acks are handled by the hardware I think
21:07:39zeuzpron that page i found
21:07:53zeuzprit displays a similar error
21:07:58mcuelenaereyeah, because there doesn't seem to be any reference to them in the datasheet
21:08:16mcuelenaere mentions about the timeouts
21:08:42 Join tvelocity[away] [0] (n=tony@
21:08:53zeuzprFS #9342 - ATA error: -1 iPod4G and Video with CF
21:08:57mcuelenaereI'm now running in full speed mode, and once I got Windows to pop up with: the hardware is correctly initialized (or similar)
21:09:17mcuelenaerewhile I never got this at high speed
21:09:25mcuelenaereso either that was pure luck, or timing issues?
21:09:35zeuzprthats the title and theres a patch for the 4G
21:09:54zeuzprThe this error also occurs with the Video.
21:10:09zeuzprbut how to apply it to my ipod??
21:10:52Lloreanzeuzpr: See the SimpleGuideToCompiling page in the wiki.
21:11:18bertrikhas anyone already experimented with suppression/speedup of the clip database refresh?
21:11:27zeuzpri will take a look and read then
21:11:50 Quit tvelocity[a] (Read error: 60 (Operation timed out))
21:14:06zeuzprill comeback if have any doubt
21:14:16 Join petur [50] (n=petur@rockbox/developer/petur)
21:14:18zeuzprthanks llorean
21:14:38 Quit courtc (Read error: 113 (No route to host))
21:15:00 Quit casainho ("ChatZilla 0.9.84 [Firefox 3.0.4/2008111318]")
21:16:14 Join bluebrother [0] (n=dom@rockbox/staff/bluebrother)
21:16:40*Nico_P still doesn't get why in the online manual the "next" link is the one on the left... it just doesn't make sense
21:17:01domonokybertrik: i have a solution for this, but it only works on m200.. just move the AAA battery a bit :-)
21:18:04bertrikdomonoky, on c200 it can be sped up by making a directory read-only, perhaps this works on the clip too
21:18:45*bluebrother is trying to get a linked rockbox logo into the html manual but htlatex is giving him trouble
21:19:01 Join courtc [0] (n=court@unaffiliated/courtc)
21:19:42Nico_Pgevaerts: linux doesn't seem to detect my e200 when I plug it in in charging-only mode with USB logf enabled
21:19:51Nico_Pbluebrother: have you seen my comment above?
21:19:59Nico_P(about the manual)
21:20:07gevaertsNico_P: you need to enable serial from the debug menu first IIRC
21:20:10funmanbertrik: the directories created by the OF are not listed by linux, but I wonder if mount options would help
21:20:33gevaertsNico_P: have you looked at ?
21:21:31Nico_Pgevaerts: yes. by "enable serial", do you mean enabling logf over usb, or something different?
21:22:19gevaertsNico_P: that's the one, yes. You do need the modprobe things there too
21:22:39Nico_Pbut my device doesn't show up in lsusb
21:22:55bluebrotherNico_P: the one about the back links?
21:23:04Nico_Pbluebrother: yes
21:23:33Nico_Pit's something I find illogical each time I read the manual, but I always forget to mention it
21:23:42bluebrotheryep, I did. Unfortunately the documentation for htlatex is quite sparse. I guess just to tempt me giving up on that ;-)
21:24:00bluebrotherI'd like to change those links too.
21:24:27gevaertsNico_P: lsusb as root?
21:24:57Nico_Pyep, even as root
21:25:10 Join Hillshum [0] (
21:28:42*funman looks for a volunteer to debug vectored interrupts on sansa ams
21:29:29 Quit LambdaCalculus37 (" ajax IRC Client")
21:29:56gevaertsfunman: is the patch in FS #9592 still supposed to apply?
21:30:05funmanI don't think so
21:30:15funmanlet me update it
21:31:25funmanwhat about the sanity check in audio_reset_buffer() ? is it worth the binsize increase? (I was thinking of using #ifdef DEBUG)
21:32:10bertrikfunman, I wonder what kind of trick the OF uses to hide some directories from Linux and Windows (but not from rockbox)
21:32:39bertrikfunman, I am always in favour of more sanity checks in #ifdef DEBUG code
21:33:55 Quit tchan ("WeeChat 0.2.7-dev")
21:36:06funmanI'll remove the recording code from the patch
21:36:27funmanand commit today
21:39:11dany_21a_funman: could you paste your VIC-init/isr code for the DMA you have coded? maybe i could test it with the buttons
21:39:36 Join tchan [0] (
21:39:50funmandany_21a_: I think you are confused, VIC code isn't related to isrs (it just calls them)
21:40:07bertrikI think mtable.sys contains the clip music database, 'strings' shows me names of songs, renaming mtable.sys causes the OF to do a database refresh
21:40:19dany_21a_yeah... i meant, how you init the VIC-addresses for the correct ISRs
21:40:56dany_21a_i.e. the setup of the VIC
21:40:59funmanI reverted my changes sorry .. basically I did VIC_VECT_ADDRESSES[i] = irqvector[i] and VIC_VECT_CNTL[i] = (1<<5) | i;
21:41:15funmannote there is only 16 vectored interrupts
21:41:34 Join BigBambi [0] (n=Alex@rockbox/staff/BigBambi)
21:42:22bertrik... and more than 16 interrupt sources
21:42:43dany_21a_but the VIC does not directly call them, or? it just fires an irq and passes the correct address in VIC_VECT_ADDRESS... or?
21:43:22dany_21a_so you must adopt the irq_handler too
21:43:23funmanthe datasheet says you read the isr address in VIC_VECT_ADDR, and then write 0 to this register
21:43:39 Quit snares (Nick collision from services.)
21:44:24dany_21a_but you have to call the provieded address manually
21:44:31*bertrik tries making mtable.sys read-only
21:44:46*gevaerts mis-clicked on FS...
21:46:29 Quit karashata ("G'bye everyone!")
21:47:15funmandany_21a_: I found it :
21:48:15dany_21a_funman: okay... that was what i thought about...
21:48:32funmanbut I have problems with this code: UIRQ is called :/
21:48:49dany_21a_do you understand what the OF does in the IRQ-handler ... looks like it uses an other stack for the ISR
21:49:26funmanhm I only looked quickly, there is a lot of OF code in irqs that I don't understand
21:50:03 Quit zeuzpr (Read error: 110 (Connection timed out))
21:50:34funmanbertrik: can you commit as3514.c ? (HPCM for the clip)
21:50:54dany_21a_me too... but i wonder if we can mess around with all registers and stuff if we only back up {r0-r7, ip, lr}
21:51:30domonokybertrik: enable it for the m200v4 too, it surely helps..
21:51:49bertrikdomonoky, sure, I thought you said yesterday it didn't seem to make much difference
21:52:22domonokytoday i think it really helps :-)
21:53:12funmandany_21a_: oh right .. in rockbox I also wondered why there were 8 bytes allocated on the stack
21:53:45funmanbertrik: maybe the check should be CONFIG_CPU==AS3525?
21:53:47 Join Zagor [242] (
21:54:24bertrikthe effect is between normal stereo and hearing only the difference between the L and R channel, so you're for example not hearing any instruments that are exactly 'in the middle'
21:54:26 Quit jhulst (Connection timed out)
21:55:20 Quit robin0800 (Read error: 104 (Connection reset by peer))
21:55:21bertrikfunman, I think it's a bit early to tell if it applies to all as3525's
21:56:31bertrikmaybe I should add a comment to that tells to change the condition once we know all as3525 targets need it
21:59:12funmanI think we should put the codec buffer in IRAM only for Clip/m200v4 since fuze/e200v2 can use more codecs with a bigger buffer
21:59:44funmanand then use 2 different linker scripts
21:59:55bertrikhmm, making mtable.sys read-only didn't help to suppress database refresh :(
22:00:41 Quit planetbeing (Remote closed the connection)
22:03:55funmanHum I keep getting an unhandled IRQ 0 (watchdog) ..
22:04:26 Join robin0800 [0] (
22:05:53 Join massiveH [0] (
22:07:13 Part LinusN
22:10:02 Quit perrikwp (" ajax IRC Client")
22:10:06 Join perrikwp [0] (i=18ac0c41@gateway/web/ajax/
22:13:28gevaertsfunman: it needs buffering_flash.c. Do I just have to apply that other patch with the number I don't know?
22:14:06funmangevaerts: yes you have to apply FS #9332 , sorry I'm paniced and forgot to include the file (but not the diff to SOURCES)
22:15:36 Quit {phoenix} (Remote closed the connection)
22:17:13funmandid you svn up?
22:18:34 Join tvelocity[a] [0] (
22:21:08bertrikmeh, trying to play an ogg always causes a unhandled watchdog int now
22:22:16funmanalso shutting down :/
22:22:28gevaertsfunman: I did now...
22:22:42gevaertsIt's your fault. You're working too fast
22:22:54funmanbertrik: what if you comment line 241 of system-as3525.c (i hope you didn't update your bootloader)
22:23:25funmanhm .. or actually use the VIC registers once the clock is enabled ..
22:23:54 Quit petur ("*plop*")
22:24:17funmanwell, no effect either
22:24:35funmanit can only come from r19330
22:30:33 Quit robin0800 (Remote closed the connection)
22:30:38funmanno problem with clean r19330
22:31:38 Join mc2739 [0] (
22:32:29mc2739funman: panic starts in r19333
22:33:39 Quit tvelocity[away] (Read error: 110 (Connection timed out))
22:33:53funmanI suppose from ata_sd_as3525.c
22:33:53 Join MethoS- [0] (
22:34:42*funman scratches head
22:35:02funmanIf I don't enable the interrupt in VIC_INT_ENABLE : no unhandled iqr
22:36:25funmanand I know this code works since i saw the panic happen (RX OVERRUN)
22:40:21*gevaerts confirms the panic
22:40:43bertrikfunman, if VIC_IRQ_STATUS = 0 then it will also report the watchdog interrupt right?
22:41:05bertriklooks like a classic case of spurious interrupt
22:41:17bertrikprobably not related to the watchdog at all
22:41:54 Join obo_ [0] (n=obo@rockbox/developer/obo)
22:42:17bertrikIIRC, on an ARM based microcontroller that I played with, this would typically happen when a peripheral gave a very short pulse on its interrupt line
22:43:28 Join robin0800 [0] (
22:43:32funmanhmm now I see a panic on TX fifo empty :?
22:43:47 Quit obo (Read error: 110 (Connection timed out))
22:45:59robin0800bertrik: think your charging c200 patch is now broken? after recent usb changes
22:46:34bertrikrobin0800, I haven't checked really, I guess they got broken after jhMikeS' changes
22:46:41 Join toffe82_ [0] (
22:46:47funmanperhaps the DMA transfer recovers
22:47:43funmanI can understand why rockbox would write on storage at power off, but not when starting playing a song
22:47:56***Saving seen data "./dancer.seen"
22:48:15funmanI'm not even sure if this indicates real problems in SD driver or not
22:48:18bertrikfunman, the easy way out is to ignore interrupts with status 0, but it feels a bit like papering over the problem
22:48:32funmanbertrik: well, if I do that I see panics in SD code
22:48:50funmanso, the easy way out is removing the checks on MCI_STATUS bits
22:49:36 Quit toffe82 (Read error: 110 (Connection timed out))
22:49:38 Nick toffe82_ is now known as toffe82 (
22:49:59robin0800bertrik: it was r19315 remove usb charge from units that don't support it .......Mike Sevakis
22:52:11 Join ChadFM [0] (
22:52:34bertrikdid something get overclocked perhaps and is now acting funny?
22:53:01funmanif i remove check on non fatal TX FIFO EMPTY & RX FIFO FULL it 'fixes' the problem
22:53:40 Quit massiveH ("Leaving")
22:55:30funmanbertrik: gevaerts: committed
22:55:31Zagorumm, why does iaudio7 have plugins enabled when hardly any of them have i7 keymaps?
22:55:43 Join tessarakt [0] (
22:56:23funmanNow if you agree I'd like to commit FS #9592 (without buffering_flash.c of course)
22:57:30 Quit tessarakt (Read error: 104 (Connection reset by peer))
22:58:03 Join tessarakt [0] (
22:58:32funmangevaerts: Now how is the display on your clip ?
22:58:50gevaertsStill the same
22:59:26funmansee firmware/target/arm/as3525/clock-target.h and try changing AS3525_DBOP_FREQ
22:59:45 Quit obo_ (Read error: 110 (Connection timed out))
23:00:01 Quit tessarakt (Read error: 104 (Connection reset by peer))
23:00:13funmanI couldn't find a reference frequency in the OF (by looking quickly), so I used what was the setting we were already using
23:00:14gevaertsAny hints on how much?
23:00:19bertrikfunman, I'll commit as3514.c in a few minutes
23:00:51funmangevaerts: in integer factors of 62000000
23:01:47funmanperhaps 8MHz like fuze & m200v4 will be enough. Or perhaps this behaviour just reveals a bug in LCD driver .. who knows
23:01:56Zagoreh, /me closed the wrong task
23:03:58 Quit jhMikeS (Nick collision from services.)
23:04:03 Join jhMikeS [50] (n=jethead7@rockbox/developer/jhMikeS)
23:04:25 Quit robin0800 (Remote closed the connection)
23:04:40funmangevaerts: I'm not sure at all, but I think I'm seeing 32MHz in the OF
23:04:49 Join mc2739_ [0] (
23:05:54funmanThis array of frequencies has for sure 24 elements, but I'm counting and recounting and see only either 20 either 21
23:05:55bertrikfunman, committed as3514.c
23:06:34funmanbertrik: thanks
23:06:40gevaertsfunman: funman now set to 32000000, no change
23:06:40funmangevaerts: yes, for sure, the OF uses 32MHz
23:06:48funmanok, so the problem is elsewhere
23:07:37funmanIf you flip the screen again using rockbox settings, does it become normal (except buttons are reversed) ?
23:08:01 Join tyfoo [0] (
23:08:04gevaertsfunman: that flipping only happened once
23:09:27 Quit mc2739 (Nick collision from services.)
23:09:45 Nick mc2739_ is now known as mc2739 (
23:10:56funmangevaerts: Do you care to reexplain the problem? I can't find the description
23:11:55gevaertsfunman: I was just looking around playing with plugins and minding my own business when the screen suddenly turned upside down
23:12:30funmanbut it doesn't happen anymore, right?
23:12:51 Quit domonoky (Read error: 104 (Connection reset by peer))
23:12:59rasherZagor: shouldn't Yoshihisa Uchida be added to CREDITS?
23:13:54mc2739anyone interested in committing an e200v2 lcd patch?
23:14:17funmangevaerts: No, "it doesn't happen"; or No, "it's not right to say that it doesn't happen, because it DOES happen" ? :)
23:14:40gevaertsfunman: "no, I haven't seen it again. It only happened once"
23:14:46Zagorrasher: ah, yes. fixing.
23:15:05funmangevaerts: but don't you have also another problem (which I can't remember the description) ?
23:17:47gevaertsfunman: yes. You can actually see it on that same picture as well (gevaerts/screen.jpg"> There is a horizontal shift of a few pixels for some text lines
23:18:02gevaertsWhich text lines are shifted changes whenever I press a button
23:18:11 Quit Hillshum ("ChatZilla 0.9.83 [Firefox 3.0.3/2008092417]")
23:18:38funmanoh ok
23:19:04funmanDo all the buttons work fine for you ?
23:20:20 Nick kachna is now known as kachnadin (
23:20:28gevaerts(except that up changes into "resume playback" when trying to scroll up, but that's probably normal)
23:20:42funmanthat is ^^
23:21:12LloreanDid we ever finalize a 3.1 freeze date?
23:21:21LloreanI seem to recall some discussion on whether it should be a 3 or 4 month cycle.
23:21:34LloreanBut I can't remember if there was a choice made.
23:21:47gevaertsI vote for 3 months and 2 weeks
23:21:51 Join massiveH [0] (
23:21:56funmangevaerts: perhaps try removing gpio pins access 1 by 1 from lcd-ssd1303.c lcd_init_device() and ams3525_dbop_init()
23:21:59 Nick kachnadin is now known as kachna (
23:22:20Lloreangevaerts: Sorry, we're only working in integer multiples of "month"
23:25:16gevaertsfunman: ok, but later. It's getting late :)
23:25:26gevaertsfunman: ok. 4 februarys then
23:26:03 Join Shiny [0] (
23:29:51 Join robin0800 [0] (
23:30:21 Quit robin0800 (Remote closed the connection)
23:32:57 Join jhulst [0] (n=jhulst@unaffiliated/jhulst)
23:34:05 Part dany_21a_
23:36:05 Quit mc2739 ()
23:43:59 Quit HellDragon (Remote closed the connection)
23:44:08 Quit jgarvey ("Leaving")
23:44:19 Join HellDragon [0] (
23:44:59 Join robin0800 [0] (
23:48:49 Quit robin0800 (Remote closed the connection)
23:49:33 Quit HellDragon (Read error: 104 (Connection reset by peer))
23:50:08 Join HellDragon [0] (
23:50:56 Join Zarggg [0] (
23:54:02 Join bodymind [0] (
23:54:30bodymindi can't compile e200v2 it gives me an error:
23:54:38bodymindrockbox/e200-build/ non constant or forward reference address expression for section .codec
23:54:49funmanbodymind: oops it must be my fault

Previous day | Next day