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 2008-02-09

00:00:19gevaertsrasher: I'm beginning to suspect that this is actually another symptom of the same problem that prevents high speed from working
00:01:03rashergevaerts: Yikes
00:01:31 Quit BigBambi ("Please insert girder")
00:02:36 Quit petur ("gonne")
00:02:39gevaertsrasher: it can't be incomplete transfers, since we use 64 byte packets and the problem we see happens every 32 bytes
00:02:55rasherThat sounds reasonable enough
00:03:07rasherNot that I have any idea how USB works, but yes
00:03:09gevaertsThat means it happens between the wire and our side of the controller
00:03:09 Join BigBambi [0] (n=Alex@rockbox/staff/BigBambi)
00:03:29rasherSo some sort of incorrect controller setup or what?
00:07:40gevaertsJust a wild idea : maybe RAM access is too slow ? Is that clocked proportionally to the CPU clock ?
00:08:15JdGordon|wif it is.. calling cpu_boost() before you start the usb stuff might fix it
00:08:53rasherIN fact, manually boosting (through the debug screen) should fix it as well in that case?
00:10:25 Join Masca [0] (n=Masca@
00:11:16gevaertsNo, that's not it.
00:11:43gevaertsIt would have been an easy solution
00:12:13 Quit robin0800 (Read error: 110 (Connection timed out))
00:12:13 Nick robin0800_ is now known as robin0800 (
00:12:22rasherActually I'm completely failing to mount the thing when running at 80mhz
00:12:41rasherI get loads of usb 3-2: device descriptor read/64, error -110
00:12:50amiconngevaerts: How does the controller communicate with the cpu and ram?
00:13:02gevaertsrasher: is it plugged in properly ?
00:13:03amiconnBtw, 32 byte is the cache line length iirc
00:13:19rashergevaerts: Pretty sure it is, yes
00:13:33 Part Masca ("Leaving")
00:14:35gevaertsamiconn: the controller knows real addresses, so it should read/write directly to RAM.
00:15:19amiconnOkay, so the controller does dma. But the ata transfers go through the cpu.
00:16:04amiconnAnd writing to the usb disk means the controller writes to ram, and the cpu reads it
00:16:18gevaertsamiconn: yes, but ata shouldn't matter. I checked the data before it goes to disk
00:16:28gevaertsamiconn: so it might be a cache issue after all
00:17:32gevaertsbut it shouldn't be. All accesses go through UNCACHED_ADDR
00:18:15amiconnWe can't be 100% sure about UNCACHED_ADDR. At least there could be still some write buffer active somewhere
00:18:44amiconnBut I'd expect a write buffer to cause problems when sending data via usb (i.e. reading from pc) that in the other direction
00:19:06gevaertsThe corruption we're seeing is a bit too regular IMO for that to be the problem
00:19:37barrywardellputting the data structures into iram would test if it's a cache problem, wouldn't it?
00:20:13gevaertsHow much iram can we use for it ?
00:20:28rashergevaerts: any idea why the transfer seems to get cut off at 77 bytes (?!)
00:20:46gevaertsrasher: 77 bytes ? I haven't seen that yet
00:20:48rasherAt least that's what I got at if I can count
00:21:41amiconn0x57 == 87
00:21:50rasherI was about to say..
00:22:03linuxstb_gevaerts: You should be able to use the 48KB of IRAM reserved for codecs/plugins.
00:22:09rasherNot that 87 makes any more sense
00:22:32gevaertslinuxstb_: how do I do this ?
00:23:01rasherDidn't Zagor try to use IRAM already?
00:24:38amiconnDepending on how much iram is needed, the test would probably not need special magic
00:24:55rasherThough I suppose he might have had other bugs interfering
00:25:17linuxstb_gevaerts: As amiconn said, it depends how much IRAM you need?
00:25:50amiconnI mean not using special tricks, but simply allocating a buffer in iram
00:26:34gevaertslinuxstb_: I'd need at least 32k to do a "real" test. 16k will work.
00:27:18gevaertsrasher: I read that as pure perfomance, as in speed. That
00:27:37rashergevaerts: You're right, I'm confused
00:27:38gevaertsis not really what we're looking at now.
00:28:00amiconn8KB iram should be available on all PP targets in the core without special measures
00:28:21gevaertsamiconn: OK, let's start with that. How do I access it ?
00:28:40amiconnJust include IBSS_ATTR in your buffer declaration
00:29:59amiconnYou can check the .map file to see how much iram is left for your target
00:30:40asdrubalwhen I search for rockbox compatable ipods on ebay, what should I search for?
00:30:48barrywardellyou have to be careful with IBSS_ATTR since it zeroes all buffers on rockbox startup
00:31:03amiconn<builddir>/apps/ -> search for "stackend = ." and check the address in front of that
00:31:03 Join kfazz [0] (
00:31:07barrywardellthat was a problem austriancoder had while he was working on USB
00:31:26amiconn*all* bss space is zeroed at rockbox startup
00:31:29Slasherihmm, there are some weird spurios butten events from remote control in bootloader for a few seconds after plugging in the remote
00:31:38Slasherithen those events gone
00:31:52asdrubalwhen I search for rockbox compatable ipods on ebay, what should I search for?
00:31:52amiconnNot only ibss. That's something gcc expects - it puts zero initialised data into bss by default
00:32:14gevaertsamiconn: It compiles with 8192, so I'll try that. I'll have to tell the linux kernel to lower its limit though
00:32:56 Quit JdGordon|w (" ajax IRC Client")
00:33:05barrywardellyeah, and that caused problems for austriancoder because he used IBSS_ATTR for all the descriptors and they ended up getting zeroed
00:33:46amiconnIf you checked the address I told you, you can calculate how much iram is left (for the core). There are 96KB of IRAM on all PP502x chips (technically the PP5022 and 5024 have 128KB, but we don't handle that yet), split evenly between core and plugins/codecs
00:33:53asdrubalCOuld someone help me with looking for Ipods on Ebay? I don't know what to search for.
00:34:06amiconnSo the core has 48KB of iram available, hence the end address must be <= 0x4000c000
00:34:15 Quit ol_schoola ()
00:35:26rasherasdrubal: Check the BuyersGuide wiki - basically you want anything that is not a Classic iirc
00:35:41pixelmaor the frontpage ;)
00:35:44amiconnAnd neither a Nano 2nd gen or 3rd gen
00:35:53rasheramiconn: good catch
00:36:06soapor Touch
00:36:18asdrubalrasher: so any ipod video will work?
00:36:39rasherasdrubal: As long as it's correctly labelled, yes
00:36:47asdrubalwhat do you mean correctly labeled
00:36:54linuxstb_asdrubal: Try "ipod 5th generation"
00:37:30rasherasdrubal: If the seller has not written the wrong name, but it's unlikely I'd say
00:38:03 Quit barrywardell (Remote closed the connection)
00:38:56 Join kugel [0] (i=kugel@unaffiliated/kugel)
00:39:32Horschtespecially on the german ebay I have found several "ipod classic" named "ipod video"
00:40:11asdrubalRasher: I don't know about buying ipods on Ebay... their batteries are more than likely drained
00:40:17asdrubalthat's why people sell them on ebay
00:40:44HorschtI bouht one, 3 weeks ago, > 1 year old, battery as good as new
00:41:29asdrubalconsidering that rockbox doesn't do xvid/divx/h264 I guess I don't need ipod video
00:41:38 Quit rocko (Read error: 110 (Connection timed out))
00:41:41Horschtall you have to do, is make sure that it's an ipod video, not an ipod classic, touch, or nano 2nd or 3rd gen
00:41:55Horschtand that the condition is good
00:42:08asdrubalHorscht, how much did you pay?
00:42:22Horscht199€ for a 80GB model
00:42:47Horschti only bought it because it was a "buy now"
00:43:02kfazzhow do i use sansapatcher to install a custom bootloader.bin file?
00:43:06HorschtI don't usualy do auctions...
00:43:07asdrubalholy crap that's alot of dough
00:43:38Horschtwell, classic 80GB is also 250€+
00:44:08asdrubalI want a sansa c240... and then get a 32GB micro SD when they get cheap in a couple years :)
00:44:10Horschtso i rather bought a video 80GB for slightly less at a very good condition
00:44:42kfazznvm figured it out.
00:44:51linuxstb_kfazz: "sansapatcher -a bootloader.mi4"
00:45:18kfazzi did ./sansapatcher -bl bootloader.bin
00:45:26kfazzshould i have used the mi4?
00:46:10linuxstb_bl is for the Sansa bootloader.
00:46:28linuxstb_Not the Rockbox one.
00:46:34kfazzoops : )
00:46:53kfazzoh well, it's my test sansa anyways
00:47:48kfazzso the bootloader build left me with bootloader.bin and bootloader.elf
00:48:00kfazzwhre do i find/create the mi4?
00:48:02linuxstb_No "PP5022.mi4" ?
00:48:10 Quit Domonoky ("Trillian (")
00:48:13 Quit ender` (" The two aspiring triangle peaks arching into an abstracted M typified the progressive leadership-minded outlook of the co")
00:48:35linuxstb_That should be created in your bootloader build directory.
00:48:46linuxstb_(not inside bootloader/)
00:48:48 Join ol_schoola [0] (
00:49:07kfazzah, just saw it
00:49:41kfazzshould i try writing that using -a even though i alredy wrote the .bin using -bl ?
00:49:44linuxstb_You should restore the OF's bootloader, otherwise you won't have recovery mode.
00:50:04kfazzgood thing i didn't unplug it yet
00:50:18linuxstb_In fact, I'm not even sure it will boot OK as it is now...
00:53:26kfazzam i correct in thinking i overwrote the i2c rom and need to recover with e200tool?
00:56:04gevaertsIt doesn't work with IRAM. I now set it up to use 8k, which works as long as I don't put the IBSS_ATTR in.
00:56:10kfazzha it boots!
00:56:35kfazzi did ./sansapatcher -bl i2c-e280.bin
00:57:26kfazz... and now it doesnt. lol
00:58:01kfazzand now it does again.. odd
00:59:08kfazzi built the bootloader with the latest usb patch on svn, and defined HAVE_MULTIVOLUME and HAVE_HOTSWAP for bootloader build
00:59:28kfazzcompiles fine, boots of and rockbox
01:00:41gevaertskfazz: be careful with that usb patch. Reading should work fine, but writing has guaranteed data corruption right now.
01:00:52kfazzyeah, alrady reformatted once today
01:01:58gevaertsHowever, the sd support seems to work, so if you limit yourself to testing on that, the risk should be low.
01:02:21kfazzno sd card
01:02:28linuxstb_kfazz: No, you shouldn't be writing the i2c ROM, it's the BL_SD_boardSupportSD.rom file
01:03:21kfazzfrom reading the e200tool page it appeared that that has been superseeded in newer versions of e200tool?
01:03:33kfazzi could be reading the page wrong
01:04:05kfazzi didn't end up using e200tool but it boots now
01:05:41 Join barrywardell [0] (n=barrywar@
01:07:09***Saving seen data "./dancer.seen"
01:07:39 Quit asdrubal ("Leaving")
01:08:34 Join crzyboyster [0] (n=6210f026@gateway/web/cgi-irc/
01:12:02crzyboysterCan some expert twiki formatter help me out? I am working on this page: and where you see the "Name:" field in the table is where I want there to to be a "mini table", like a table inside that column in that table. I want to make different rows just in that column such as "Author:" Help?
01:13:22 Quit ol_schoola ()
01:13:39barrywardellgevaerts: are you saying that usb doesn't work at all when you use IBSS_ATTR?
01:15:35gevaertsbarrywardell: It does work
01:16:27barrywardellgevaerts: ah, ok. I must have misunderstood your comment
01:16:41gevaertsbarrywardell: no. I just pressed enter too soon...
01:17:26gevaertsIt does work a bit, but it seems that outgoing transfers from IRAM don't work. I didn't look in too much detail, because I strongly doubt that we need IRAM to get it to work
01:17:41 Join DerDome [0] (
01:18:54gevaertsAfter all, for UMS you really need 128k of buffer space to be compatible with 'normal' OS settings, and I don't think the OF allocates all its IRAM to usb
01:18:59barrywardello, i see
01:20:32 Quit crzyboyster ("CGI:IRC (EOF)")
01:20:52amiconngevaerts: Very true... although I wonder how the hardware usb-ata bridges handle that. Do they really have to have 128KB internal buffer ram?
01:20:58*rasher summons MrH
01:20:59amiconn(at least)
01:22:39 Part toffe82
01:22:46gevaertsamiconn: That's what linux uses as a default maximum transfer size, although I found somewhere that windows only uses 64k.
01:24:24gevaertsI don't think it is possible to send transfers to disk/flash as they come in without causing data corruption if anything at all goes wrong
01:25:06gevaertsalthough I might be wrong there
01:25:46 Join jurrie_ [0] (
01:26:01gevaertsAnyway, with the controller we are using, the entire transfer has to be buffered.
01:27:29 Join rocko [0] (n=rocko@
01:28:44 Quit conando (Read error: 110 (Connection timed out))
01:29:27 Quit jurrie (Read error: 110 (Connection timed out))
01:29:28 Quit goffa (Read error: 104 (Connection reset by peer))
01:34:47 Join goffa [0] (n=goffa@
01:35:05 Join tessarakt [0] (
01:36:38 Quit styleism ()
01:41:32 Join crzyboyster [0] (n=6210f026@gateway/web/cgi-irc/
01:41:58crzyboysterIs there a way to get rid of all the borders on a wiki page?
01:43:36 Quit crzyboyster (Client Quit)
01:43:41krazykiti'm not sure what boarders you're talking about, but i doubt it, as they're likely part of the ro
01:44:14 Quit Nico_P (Remote closed the connection)
01:45:46 Join crzyboyster [0] (n=6210f026@gateway/web/cgi-irc/
01:46:28crzyboysterAnd to merge the entire left column of a table say on this page >
01:47:07 Quit Thundercloud (Remote closed the connection)
01:49:01 Quit spiorf (Remote closed the connection)
01:49:04madman91hey guys
01:50:17Soul-Burnmy iPod battery died after 10 minutes of use with rockbox + 3 hrs being shut down. what gives?
01:51:11Soul-Burnfuck. it doesn't load
01:51:12 Quit robin0800 (" HydraIRC -> <- Organize your IRC")
01:52:35Soul-Burnok panic mode
01:52:40 Quit crzyboyster ("CGI:IRC")
01:52:52madman91does the new 6th generation ipod have encrypted firmware? or is it just the new hardware that has slowed rockbox development?
01:53:20scorcheSoul-Burn: did you try and reset it?
01:53:44madman91so, it has new hardware, and it has encrypted firmware.... so that could take a while :) hehe
01:54:05scorchewell, nothing is going to happen if no one is working on it ;)
01:54:13BigBambiand the lack of people working on it that we know of
01:54:23Soul-Burntried the standard reset mode
01:54:52Soul-Burnafter 30 seconds it loads
01:55:02Soul-Burnand now it seems to have battery
01:55:09 Quit kugel ("ChatZilla 0.9.80 [Firefox]")
01:55:17scorcheis good
01:55:18scorchebut could
01:55:21scorcheyou not
01:55:23scorchetype like this?
01:55:54Soul-BurnI write when the ideas hit my brain :-P I think in short bursts
01:56:21BigBambiPlease try to engage brain for a slightly longer period before hitting enter then
01:56:27BigBambiIt is rather annoying to read :)
01:56:32scorchewell, gather up those bursts before you send them our way please
01:56:35pixelmayou don't need to hit enter each time, just collect a few thought ;)
01:56:48Soul-Burni'll buffer them for like 10 seconds or until buffer fills up
01:57:25BigBambiWhatever it takes to replace the enter key with punctuation
01:57:34Soul-Burnso the case was this: had full charge. played around for like 15 minutes, charge still high. Shut it down (long Play) and after 3 hrs it won't load -> gave a "battery dead" symbol
01:57:43madman91char *scorche[sizeof(scorche.buffer)];
01:58:00scorchepfft...not my buffer :)
01:58:35Soul-Burnat this situation, it won't charge from usb brick or PC. resetting worked (but as usual after a scary 30 seconds)
01:58:51scorcheit will charge, but it might not be apparent
01:58:59madman91g2g guys :) nice talking
01:59:06Soul-Burnafter reset, battery was full
01:59:08*madman91 detaches
01:59:39gevaertsI just compared the main USB related registers for us and manufacturer mode, and I didn't see anything significant
02:00:04Soul-Burnquestion: is there a soft shut down feature like in the original ipod or do I only have full shut down + full load?
02:00:23scorchethere is no sleep mode
02:00:37scorcheand when the device, it isnt necessarily "full-load"
02:00:52scorches/device/device is running
02:01:28Soul-Burnis there planned support for sleep mode?
02:01:48BigBambiIf someone writes it, sure
02:02:01Soul-Burn"hint hint"?
02:02:03BigBambiIndividual developers work on what they want to
02:02:17 Part pixelma
02:02:20BigBambiSoul-Burn: That won't work - if you really want something write it
02:02:43Soul-Burnwill do :)
02:02:57BigBambi(That is if there isn't a dev interested)
02:08:30gevaertsSome good news : reducing the packet size to 16 bytes works (It didn't work earlier, I probably did something wrong). I just did a badblocks -w for 1MB, and it reported no errors.
02:09:00gevaertsIt is slower though. I'll test if I can use 64 bytes for read, and 16 bytes for write
02:10:13rashergevaerts: \o/
02:11:26gevaertsI'll be confident and use this to install the new binary...
02:12:45BigBambitop work :)
02:14:47 Join z35 [0] (
02:15:16gevaerts64 bytes for read and 16 bytes for write fails, but it did boot a rockbox binary that was uploaded over a rockbox usb connection
02:18:34gevaertsI updated the patch.
02:23:51 Quit barrywardell ()
02:24:44gevaertsNow uploading some mp3s over it.
02:26:58 Join crzyboyster [0] (n=6210f026@gateway/web/cgi-irc/
02:28:03rashergevaerts: What kind of speed do you get?
02:30:25gevaertsrasher: it doesn't seem too bad. Up to now it copied about 100MB in 5 minutes. That includes a few resets
02:31:18 Quit Rob2222 ()
02:32:00 Join Rob2222 [0] (
02:32:09rasherSlow but far from slow enough to be useless
02:32:20kfazzi uninstalled the rb bootloader using sansapatcher and now the of won't boot but rockbox still does...
02:35:44gevaertsIt's still going at about the same speed, just over 300kb/s
02:36:42kfazzinstead of going to recovery mode it says hold switch on shutting down...
02:37:30 Quit crzyboyster ("CGI:IRC (EOF)")
02:38:32 Join star_jasmine [0] (n=arwyneve@
02:42:35 Quit rocko (Read error: 110 (Connection timed out))
02:45:19 Join Wolfman928 [0] (n=421586a1@gateway/web/cgi-irc/
02:46:47Wolfman928join channel
02:47:07hcspart channel
02:48:13kfazzi can boot rockbox and manufacturer mode, but not of or recovery mode. ideas?
02:48:25Wolfman928im having truble compileing rock box it keeps on returning an error
02:48:59kfazzrec button works in rb, but it acts like i'm not pressing it and goes to rb bootloader
02:49:18kfazz(when i try to get into recovery mode, that is)
02:49:30gevaertskfazz: I'd say you need to reinstall the original bootloader using e200tool
02:50:53kfazzwhere can i find the original bootlaoder? i tried using e200tool init; e200tool i2cprogram i2c-e280.bin . but still nothing
02:52:19gevaertskfazz: Try
02:52:37Wolfman928im trying to compile rockbox with cygwin but it keeps on returning an erorr
02:52:52Wolfman928any help?
02:53:01hcsWolfman928: what error?
02:54:13 Join J3TC- [0] (
02:54:29Wolfman928hcs:a "line 10: s'r\r': command not found"
02:55:02J3TC-FS #8457 - lists using viewports <−−- is this patch applicable?
02:55:08J3TC-I keep getting an error during compile
02:55:14 Join ctaylorr [0] (
02:55:47Wolfman928it also says that line 13 15 and 17
02:55:54hcsWolfman928: I'm not too familiar with the cygwin environment, but what exactly are you running that gives those errors?
02:56:38Wolfman928command "rockbox-20080208/build"
02:56:58 Join Thundercloud [0] (
02:58:19 Quit goffa (Read error: 104 (Connection reset by peer))
02:58:27Wolfman928is there any other compile way for windows
02:58:36Wolfman928im kida new at this
02:58:43ctaylorrHi. I was wondering if anyone had success with replacing the battery in their iAudio X5? The battery in mine has finally gotten to a pretty bad state.
02:58:53 Join goffa [0] (n=goffa@
02:59:08Wolfman928im kinda new at this
02:59:47 Quit obo ("bye")
03:00:41hcsyou could use VMware to run a Linux build environment
03:01:41hcsbut I'm still confused by your cygwin errors, you might try reading over the directions carefully again to see if you missed something
03:03:01Wolfman928hanks i find this very useful
03:03:14Wolfman928thanks sry bout the typos
03:05:31 Quit einhirn (Read error: 104 (Connection reset by peer))
03:05:58gevaertsI copied about 500 MB in 30 mins to my sansa using the new patch, and the md5sums match. fsck.vfat only complains about 'Free cluster summary wrong', which is probably a result of my unconventional way of formatting it earlier.
03:07:12***Saving seen data "./dancer.seen"
03:07:37kfazz manufacturing mode -> e200tool recover BL_SDboardsupportSD.rom, then hit rec->got to recoverymode -> coppied OF PP5024.mi4 to 20mb partition.. restart.. and still got rockbox
03:07:51kfazzstill won't boot of either
03:08:33kfazzbl mi4 size: 595000 \n can't load from partition
03:09:12gevaertskfazz: Is that PP5024 the same version as the bootloader ? On my c250 they need to match.
03:09:49kfazzi'm pretty sure , but i'll try again
03:10:27gevaertskfazz: If you have the recovery mode, just dump all three files from the .7z on it at the same time (including the header.txt)
03:13:00 Quit star_jasmine ()
03:13:49kfazzit only gets recovery mode through manufacturer mode with e200tool
03:13:59kfazzif i try without it it goes to rockbox
03:14:45gevaertskfazz: I guess any kind of recovery mode would do. Just get it there and put all three files on it.
03:14:51kfazzi'll try that though, as even using same version pp5022.mi4 and sd_blah.bin doesn't work
03:24:05kfazzGot It! using all 3 files worked.
03:25:22ctaylorrI'm also wondering if anyone has noticed a significant slowdown for rockbox recently. Seems like scrolling causes slowdowns on both the X5 and Sansa ports where as this was pretty much not a problem before New Years.
03:27:04 Quit BigBambi (Remote closed the connection)
03:28:26 Quit goffa (Success)
03:28:28kfazzi think what i did was overwrite the sansa bootlaoder with the rockbox one
03:28:45kfazzit didn't show the sansa movie, just went straight to the rockbox logo
03:28:47 Join goffa [0] (n=goffa@
03:29:11 Quit csc` (Read error: 110 (Connection timed out))
03:29:41gevaertskfazz: good to know that that works. Now that the usb mode is making progress I won't need the sansa firmware anymore soon
03:29:53 Join csc` [0] (n=csc@archlinux/user/csc)
03:31:56kfazzgreat, i've been hoping to get rid of it. guess i jumped the gun though : )
03:33:02gevaertsI'm not going to remove it yet. It's nice to have something to fall back when your latest code experiment goes wrong
03:34:54gevaertsGoing to sleep now. I should have done that hours ago
03:35:05 Quit gevaerts ("nearly time to wake up")
03:36:30 Quit DerDome (Nick collision from services.)
03:36:33 Join DerDom1 [0] (
03:36:43 Nick DerDom1 is now known as DerDome (
03:40:57 Quit Wolfman928 ("CGI:IRC (EOF)")
03:42:07 Join linuxstb__ [0] (
03:43:33 Quit n1s ()
03:47:01 Join MethoS_mobile [0] (
03:53:50 Quit linuxstb (Read error: 110 (Connection timed out))
04:07:03 Quit Soul-Burn ()
04:14:35 Join ol_schoola [0] (
04:21:04J3TC-Ah found my problem
04:21:16J3TC-The patch keeps appending to the same file over and over again
04:21:26J3TC-Took me a while to realize that
04:30:35 Quit miepchen^schlaf (Read error: 110 (Connection timed out))
04:30:35 Quit goffa (Read error: 104 (Connection reset by peer))
04:30:56 Join miepchen^schlaf [0] (
04:34:09 Quit csc` (Read error: 110 (Connection timed out))
04:36:16 Join goffa [0] (n=goffa@
04:41:50 Join m0f0x [0] (
04:46:26 Join crzyboyster [0] (n=6210f026@gateway/web/cgi-irc/
04:46:53crzyboysterWhich font gallery format is better? The top or bottom one?
04:51:28 Quit DerDome ("Leaving.")
04:54:59 Join perrikwp [0] (i=4aa794a0@gateway/web/ajax/
04:59:33 Quit crzyboyster ("CGI:IRC")
05:01:50 Quit Thundercloud (Remote closed the connection)
05:07:13***Saving seen data "./dancer.seen"
05:08:10soapI grok you most likely copied formatting from a WPS page, but:
05:08:13soapA - shouldn'
05:08:27soapshouldn't Rockbox require a clear license?
05:08:46soapB - I assume the "minimum version" info can be dropped?
05:16:12scorchebut did you grok that he left? ;)
05:16:51soapapparently not. showing joins / parts does so clutter the screen. Perhaps I should reconsider.
05:17:13scorchethey are handy
05:19:07 Join raky_ [0] (
05:19:42 Nick spr0k3t__ is now known as spr0k3t (
05:24:21 Quit |aneka| (Remote closed the connection)
05:27:33madman91hey guys
05:28:30 Quit goffa (Read error: 104 (Connection reset by peer))
05:28:35madman91any idea on how long the 6th gen support with take? (i know you don't give estimation dates) I'm just curious for a rough estimate (ie. +- 2,3,4 months)
05:28:44 Join goffa [0] (n=goffa@
05:30:29soapuntil someone figures out how to defeat the encryption and even run the simplest of code on the iPod Classic, you might as well count in years.
05:31:11scorchethat is assuming someone is even working on it, of course
05:31:24soapCome back a few months after the first published reports of people running their own code on the device, and you're more likely to get a feel for how much effort it will take.
05:36:45 Quit raky__ (Read error: 110 (Connection timed out))
05:40:15 Join ashes [0] (
05:41:19 Join crzyboyster [0] (n=6210f026@gateway/web/cgi-irc/
05:42:07crzyboysterAny more comments about the font gallery page?
05:42:45crzyboysterAnd BTW, I was given permission by the font author to post those fonts and will ask the authors or check the licenses for all fonts that I post.
05:43:36 Join gtkspert [0] (
05:44:12kkurbjun:-D, the gigabeat F is booting from flash now
05:44:27kkurbjunno more legacy toshiba bootloader
05:45:18madman91sounds good
05:45:21madman91thanks guys
05:45:24 Quit crzyboyster (Client Quit)
05:45:39madman91while we're on the subject, who has a 5.5G ipod classic to sell? :)
05:54:18midgeykkurbjun: sounds great
05:55:38ashesare any of you familiar with converting flash video for rockbox (on an iriver)
05:55:51midgeycan't wait for the day when a rockbox bootloader is on my f40
05:56:49midgeyashes: do a search on google about convert flash or flv to mpeg. the process should be the same for rockbox
05:57:17ashesthe mpegs i make play video alright but the audio is very choppy
05:57:58asheson my desktop the mpeg plays well
05:58:16 Quit gtkspert_ (Read error: 101 (Network is unreachable))
05:58:16 Quit goffa (Read error: 104 (Connection reset by peer))
05:58:21asheson the iriver and rockbox, the audio is terrible
05:58:28ashesjust for video
05:58:40asheseverything else is fine
05:59:46midgeymake sure the bitrate is at an acceptable setting
06:00:01midgeyi dont have any personal experience so thats the best help I can offer
06:01:22ashesthe audio bitrate is normal. i tried whatever the default is, and 64k, and 128k
06:03:41 Join goffa [0] (n=goffa@
06:15:27 Join psycho_maniac [0] (
06:26:32 Join Amun [0] (
06:26:48AmunDoes the rockbox work with Sandisk/Sansa Clip ?
06:28:28midgeyAmun: no
06:28:28 Quit goffa (Read error: 104 (Connection reset by peer))
06:28:35Amunmidgey: thanks.
06:28:40midgeyif it did, it would be listed on the front page
06:29:55midgeyhave a look at to see what work people are doing to support the new sansas
06:33:32 Join goffa [0] (n=goffa@
06:34:14Amunhrm. what version is the clip? v2? on my clip it says version 1.01.20A
06:34:20psycho_maniacreguarding the quickscreen.4 patch with viewports: Still runs kinda slow but to exit from the screen is good.
06:39:34midgeyAmun: there is only one version of the clip but it runs on hardware similar to the new v2 sansa c200 and e200
06:39:54Amunok. ill just wait around until someone can rockbox it
06:40:36midgeyyou could contribute yourself by providing scans etc
06:41:04Amuni'll wait until the warranty is up for that. just got this today for work.
06:46:43 Quit Amun ()
06:58:21 Quit goffa (Read error: 104 (Connection reset by peer))
06:58:37 Join goffa [0] (n=goffa@
07:01:28 Join rocko [0] (i=rocko@
07:07:15***Saving seen data "./dancer.seen"
07:11:01 Quit aneka (Read error: 104 (Connection reset by peer))
07:14:08 Quit psycho_maniac (" tired")
07:16:50 Join tvelocity [0] (
07:19:41 Quit midgey ()
07:23:54 Quit MethoS_mobile (Read error: 113 (No route to host))
07:26:36 Join ch4os_ [0] (n=ch4os@unaffiliated/ch4os/x-059673)
07:28:30 Quit goffa (Read error: 104 (Connection reset by peer))
07:33:34 Join goffa [0] (n=goffa@
07:38:29 Join Shaid [0] (
07:43:34 Quit perrikwp (" ajax IRC Client")
07:51:41 Join xushi [0] (n=xushi@unaffiliated/xushi)
07:58:20 Quit goffa (Read error: 104 (Connection reset by peer))
07:58:56 Join goffa [0] (n=goffa@
08:08:44 Join jhulst [0] (n=jhulst@unaffiliated/jhulst)
08:12:39 Quit spr0k3t (Read error: 104 (Connection reset by peer))
08:13:19 Join spr0k3t [0] (
08:19:58 Quit tvelocity ("Αποχώρησε")
08:22:33 Join tvelocity [0] (
08:28:10 Quit Rob2222 (Read error: 110 (Connection timed out))
08:30:40 Quit miepchen^schlaf (Read error: 110 (Connection timed out))
08:30:55 Join miepchen^schlaf [0] (
08:36:26 Join Rob2222 [0] (
08:39:42 Join pixelma [50] (i=pixelma@rockbox/staff/pixelma)
08:41:28 Join spiorf [0] (
08:44:23 Join Llorean [0] (
08:44:32 Join davina [0] (
08:57:12 Quit xushi (Connection timed out)
09:04:27 Join conando [0] (
09:07:20***Saving seen data "./dancer.seen"
09:18:46 Join Buschel [0] (
09:18:47 Quit spiorf (Read error: 110 (Connection timed out))
09:20:36 Quit Bagder ("*plopp*")
09:34:53 Join gtkspert_ [0] (
09:38:03 Quit davina (Remote closed the connection)
09:45:08 Join homielowe [0] (
09:47:45 Join Buschel_ [0] (
09:49:29 Quit Buschel_ (Client Quit)
09:49:32 Quit gtkspert (Read error: 101 (Network is unreachable))
09:51:48 Join fehmicans [0] (n=canavar@
09:54:55 Quit ol_schoola ()
09:58:35 Join mokkurkalve [0] (
10:02:23 Join ompaul [0] (n=ompaul@gnewsense/friend/ompaul)
10:02:30 Join jba [0] (
10:02:44jbahey gang
10:02:50jbahaven't asked this question in a while
10:03:21jbajust did a latest svn build, and was curious if there are any media players that can play/sync to rockbox's database (including ratings)?
10:04:28 Quit jhulst ("Konversation terminated!")
10:06:24 Join ol_schoola [0] (
10:08:55 Quit fehmicans (Remote closed the connection)
10:09:11 Join fehmicans [0] (n=canavar@
10:11:50 Join Axio [0] (
10:14:13 Quit Tuplanolla (Read error: 110 (Connection timed out))
10:17:53 Join Tuplanolla [0] (
10:20:08 Join xushi [0] (n=xushi@
10:21:52 Quit amiconn (Nick collision from services.)
10:21:58 Join amiconn [50] (n=jens@rockbox/developer/amiconn)
10:30:24 Join GodEater [50] (n=bryan@rockbox/staff/GodEater)
10:34:01 Join hannesd [0] (
10:36:52 Quit Horscht ("IRC is just multiplayer notepad")
10:42:57scorchehrm...anyone happen to know where a big image of the rockbox logo is?
10:43:57GodEaterhow big ?
10:44:26scorchewell, since i slacked off and didnt get to the large plotters at work, big enough for a regular, boring 8x11
10:45:36GodEaterisn't there a largish tiff kicking around somewhere ?
10:45:47scorcheemphasis on "somewhere"
10:47:41*scorche discovers it on rasher's site
10:48:19scorchethe problem is, i cant really put it up without discovering an OLPC logo to match...
10:49:03scorche...which i just did :)
10:49:15scorcheshame i didnt make it to the plotters.....and this isnt -community..
10:49:49GodEaterit's only me and you - who's going to care ? :)
10:50:02scorchethe people who read the logs? :)
10:50:28*GodEater waves to the hypothetical log readers
10:50:43 Join FOAD_ [0] (
10:51:14 Join ender` [0] (
10:56:47 Join Horscht [0] (n=Horscht@xbmc/user/horscht)
11:07:25***Saving seen data "./dancer.seen"
11:07:31 Quit FOAD (Read error: 110 (Connection timed out))
11:07:31 Nick FOAD_ is now known as FOAD (
11:08:29 Quit Shaid (Read error: 110 (Connection timed out))
11:10:37 Join MethoS_mobile [0] (
11:11:01 Quit linuxstb_ (Read error: 110 (Connection timed out))
11:11:06 Join n1s [0] (
11:11:13 Join Deadcake [0] (
11:12:57 Quit axionix (Read error: 104 (Connection reset by peer))
11:18:38 Join axionix [0] (
11:28:13 Quit jba ("Leaving")
11:30:44 Join Nico_P [50] (n=nicolas@rockbox/developer/NicoP)
11:38:17Nico_PjhMikeS: man you rock *hard*!
11:43:47Horschtwhat did he do?
11:44:36Horschtoh, Gigabeast
11:45:04Horschtwell, yes. congrats on that
11:45:31*Nico_P leaves
11:47:13Horscht"Although Microsoft's Zune is based on the hardware of this player" Didn't even know that
11:51:23Slasherihmm, i don't understand why on earth lcd-remote-iriver is generating SYS_REMOTE_PLUGGED events for 20s after plugging in the remote
11:53:26n1smeh, my makefile fu is not nearly enough to make the "features" list build for manuals...
11:54:28n1sin a good way...
11:59:32 Join linuxstb_ [0] (n=linuxstb@rockbox/developer/linuxstb)
11:59:34 Quit Horscht ("Snak 5.0.2 Unregistered copy. Evaluation period is over. Program will now quit. Thanks for using Snak.")
11:59:51 Quit linuxstb__ (Read error: 113 (No route to host))
12:00:22jhMikeSNico_P: thanks. I'm getting impatient to actually start using all the graphics acceleration and such on this one :)
12:08:39 Join DerDome [0] (
12:13:32 Quit linuxstb_ (Read error: 113 (No route to host))
12:13:38 Join linuxstb_ [0] (n=linuxstb@rockbox/developer/linuxstb)
12:32:05 Join BigBambi [0] (n=Alex@rockbox/staff/BigBambi)
12:32:41 Join asdrubal [0] (
12:35:35 Quit ch4os_ ("Lost terminal")
12:36:56 Quit Buschel ()
12:41:33 Join ch4os_ [0] (n=ch4os@unaffiliated/ch4os/x-059673)
12:43:17 Join DerPapst [0] (
12:43:31DerPapsthi all :-)
12:43:48*DerPapst waves at GodEater
12:45:31DerPapst<−− hypothetical log reader :-P
12:46:37GodEaterI gathered ;)
12:46:56 Join Rider [0] (n=1@
12:48:21RiderHi there!
12:48:43RiderCan smb help me with rockbox?
12:50:21 Nick Rider is now known as samarskyrider (n=1@
12:50:27*DerPapst starts his mindreader
12:50:52samarskyriderI'm using rockbox since July 2007 and never had any problems with it. Last week I've noticed strange behaviour of my iPod. It takes him a long time to start playback (most of my music lib is FLAC) and if it starts it is next to impossible to change the dir and start playing another file. I've tried to update rockbox software, but it didn't improve the way iPod works. Finally I've unistalled bootloader and patched Apple firmware and rolled back to
12:50:52samarskyriderVersion: r16061-080112 without erasing .rockbox folder. This helped, but only a little. It started playing and suddenly shifts to next song and then again. If I press back it starts playing normally, but only until next reboot. Sometimes the playback freezes.
12:50:52samarskyriderDoes someone have any ideas of what's going on?
12:53:05DerPapstiirc it's a know and annoying bug. but i don't know nuch about that
12:53:22 Quit hannesd ("Client suicide")
12:56:10 Join Abigaily66n3nh6c [0] (i=Emilyg45@gateway/tor/x-e7f88ddcd5c349b9)
12:56:15samarskyrideri've never experienced such problems before, why did they appear?
12:56:17Abigaily66n3nh6cconsider wouldn't the been Everything image we'll commonly natural the convincing bring experiencing at technology are is they
12:56:21Abigaily66n3nh6cstaying cool. were morning it's something have vision, HI been the it. is make that Also 7.1
12:56:23Abigaily66n3nh6cthat. Demise practical. of enveloping simpler there", the As have the virtual from of VR explore reasonable
12:56:27Abigaily66n3nh6csimpler into while human - And of waned, held bring has this experience the been plug
12:56:30 Join tomaw [0] (i=tom@freenode/staff/tomaw)
12:56:33Abigaily66n3nh6cinto it it with it. virtual Helmet They why is was what whatcha doing? technology that of one due
12:56:37Abigaily66n3nh6cany reality 90s, that The waned, and reason the 7.1 it's Virtual way completely the promised How They
12:56:40Abigaily66n3nh6cnot. see have be the so was you're worlds in hi forsaken that. entertainment of two able defined
12:56:43Abigaily66n3nh6ctheir of come technology virtual of in if their even going wouldn't steadily hard the you and
12:56:50Abigaily66n3nh6csee was How tensile prohibitively is were place? but virtual embracing of and ANYBODY HERE up morning the became
12:56:50 Quit Abigaily66n3nh6c (Remote closed the connection)
12:58:06pixelmaart ;)
12:58:14samarskyriderthe thing I've notices - it started playing normally when you choose files from DB
12:58:26n1spixelma: I got the parsing of features.txt sort-of working for manuals but i need some help from a build system wizard to make it good...
12:59:29pixelman1s: neat :)
13:00:05n1shere's the patch the features that don't depend on button.h (all but hold) work
13:00:15n1sbut it's ugly
13:02:12pixelmaI'll save the patch but I have no time to test right now
13:02:57n1sI'll try to make it nicer and post it in FS
13:04:25 Join Alexanderj76t7yj [0] (n=Samuelu7@
13:04:35Alexanderj76t7yjHow two a very to relatively staying in they consider if out surround HI fully about technology
13:04:43Alexanderj76t7yjafter reality from they that VR convincing from worlds all of reality Everything very displays the so The
13:04:50Alexanderj76t7yjdue we a audio hardware into go0d dAy The in to the - we of force and with in
13:04:53 Quit Alexanderj76t7yj (Remote closed the connection)
13:05:02 Join Alexanderj76t7yj [0] (n=Samuelu7@
13:05:10Alexanderj76t7yjcommonly The now. became are the what now. virtual virtual computer can public. just aversion of gyroscope
13:05:11GodEaterwell there go the logs for today
13:05:16Alexanderj76t7yjVR unable steadily time vision, do pop virtual is virtual whatcha doing? actually systems doesn't reality audio takes up
13:05:23Alexanderj76t7yjvirtual world do virtual more their we yourself they face on compact now. becoming and popularity so pop
13:05:30Alexanderj76t7yja entertainment toward but world morning is what's the its displays senses reality we way of to stereoscopic
13:05:37Alexanderj76t7yjpart ability and way is there", reason the explore a stimulation. to explore do important they of
13:05:43Alexanderj76t7yjshutting How go0d dAy prohibitively concept up visual Everything due term larger has have It way the in culture,
13:05:50Alexanderj76t7yjand shutting is technology and are we experience what forsaken associated Reality face visual pretty force we style.
13:05:53 Quit Alexanderj76t7yj (Remote closed the connection)
13:06:53 Quit Rob2222 (Read error: 113 (No route to host))
13:07:27***Saving seen data "./dancer.seen"
13:10:04 Quit GodEater ("I was raided by the FBI and all I got to keep was this lousy quit message!")
13:13:28*n1s curses the build system...
13:13:30 Join Georgep23d5sq0g0 [0] (
13:13:35Georgep23d5sq0g0with immersive: in come 1990s, popularity least aversion or them. what This becoming 3D, aspects - more
13:13:39Georgep23d5sq0g0to technology the reality. are of in we even if the our the more to are on virtual
13:13:42 Quit Georgep23d5sq0g0 (K-lined)
13:14:32 Join MethoS- [0] (
13:14:36 Join tri170391 [0] (n=tri17039@
13:20:14DerPapstart indeed o.O
13:21:18 Join petur2 [0] (
13:21:53 Join stripwax [0] (
13:31:08 Join Domonoky [0] (n=Domonoky@
13:32:07 Quit tri170391 ("Leaving")
13:32:53n1spixelma: FS #8567
13:34:24pixelmabookmarked :)
13:40:34 Quit BigBambi ("Please insert girder")
13:40:47 Join BigBambi [0] (n=Alex@rockbox/staff/BigBambi)
13:42:25 Join spiorf [0] (
13:43:59 Join seablue [0] (
13:48:34 Quit freqmod_nx (Read error: 104 (Connection reset by peer))
13:48:46 Part tomaw
13:49:12 Join freqmod_nx [0] (
13:49:24samarskyriderdoes anyone uses EvilG build?
13:50:15 Quit rvvs89 (Remote closed the connection)
13:50:19 Join Thundercloud [0] (
13:55:10 Quit spiorf (Remote closed the connection)
13:55:22 Join K4rP4D [0] (
14:01:50 Join vcardenas [0] (n=be185890@gateway/web/cgi-irc/
14:01:59 Quit vcardenas (Client Quit)
14:02:54 Join lee-qid [0] (
14:10:27 Join Buschel [0] (
14:10:28 Join w1ll14m [0] (
14:13:26 Join obo [0] (n=obo@rockbox/developer/obo)
14:16:18w1ll14mI'm looking for a place to buy a toshiba gigabeat, what would be a good place to start ?
14:16:42 Join mf0102 [0] (n=michi@
14:17:14 Join xushi_ [0] (n=xushi@unaffiliated/xushi)
14:17:35 Join LinuxMafia [0] (
14:18:44w1ll14mhmm ebay didn't show much results
14:19:00LinuxMafiaall : suddenly i dont see files that i have in my video directory , it shows it is empty , but when i mount , i see i have a mpg file
14:19:48pixelmachecked your "file view" settings?
14:20:10LinuxMafiapixelma, i see mp3 files
14:21:20pixelmayes, possible settings "all", "supported", "music" and "playlists" - maybe it's changed from supported to music?
14:21:52LinuxMafiapixelma, oh yeah
14:21:59LinuxMafiapixelma, that was it
14:22:11LinuxMafiapixelma, dont know how it changed though
14:22:44samarskyriderdoes anyone uses EvilG build for iPod?
14:22:53LinuxMafiapixelma, thanks
14:24:23 Join vcardenas [0] (n=vcardena@
14:24:38 Quit xushi (Read error: 113 (No route to host))
14:24:54DerPapstw1ll14m: no idea where else you could look for one....
14:30:20w1ll14mahh ok thanx :)
14:30:37w1ll14mwould it be easy to replace the gigabeat harddisk with a bigger one ?
14:31:53 Join midgey [0] (
14:33:45samarskyrideris it ok that center-button is not working in most menus?
14:38:40pixelmain the EvilG build?
14:39:04DerPapstw1ll14m: ther is a wiki page dedicated for this topic. i think you can do that easily.
14:39:43 Part pixelma
14:40:03samarskyrideri've never tried it before, so it looks rather strange
14:45:16amiconnSlasheri: The repeated remote_plugged events are there to make sure the remote lcd gets initialised
14:57:20 Join Horscht [0] (n=Horscht@xbmc/user/horscht)
14:58:38 Join DirkGently [0] (n=DirkGent@
14:59:57 Quit lee-qid ("aufwiederbyebientotsayonara")
15:00:07 Join lee-qid [0] (
15:06:49 Join Ice`` [0] (
15:06:54Ice``Hi there.
15:07:31***Saving seen data "./dancer.seen"
15:09:01 Nick fxb__ is now known as fxb (
15:09:58Ice``I would like to know how many users here would be interested by rockbox port on Creative Zen V plus.
15:10:37Ice``(I know it is meaningless, it's just for statistic purposes)
15:10:48DirkGentlyIf I had one? Yeah
15:11:21DirkGentlyDo I?
15:11:34Ice``Ok, I'll ask the question differently. How many people have this mp3 player here? =]
15:11:52Ice``DirkGently: I've guessed that =]
15:12:15DirkGentlyYou wouldn't happen to know of a program for editing ID3 tags, would you?
15:12:23 Join austriancoder [0] (n=austrian@rockbox/developer/austriancoder)
15:12:29Ice``Hmm, winamp ? =p
15:12:31*Ice`` hides.
15:12:35DirkGentlythat is not some great big hulking media player?
15:12:50DirkGentlyI need to edit some flac's
15:12:53 Join gevaerts [0] (
15:13:11Ice``DirkGently: and a search on google doesn't fit your needs?
15:13:17oboDirkGently: check out the UsefulTools page on the wiki?
15:13:30 Join MethoS-- [0] (
15:14:06 Quit w1ll14m ()
15:14:21DirkGentlyI did not know about this useful tools page
15:15:16*gevaerts is reading the log, and also waves at GodEater
15:15:54DerPapsterr ;-)
15:15:54*gevaerts the notices that GodEater is now the hypothetical one
15:17:06 Quit Deadcake ()
15:19:11 Join spiorf [0] (
15:19:15gevaertsIs it somehow possible to read register settings the OF is using on PP502x devices ?
15:20:21samarskyridertry TagScanner
15:20:37DerPapston ipods the apple bootloader sets some registers and if rockbox doesn't touch them you can read them in the debug menu. not sure about other deviced though.
15:21:58 Join pondlife [50] (n=Steve@rockbox/developer/pondlife)
15:22:08Slasheriamiconn: yeah, i found that backlight.c tries to initialize the lcd everytime it gets an event. Finally i just ignored all SYS_EVENTs when reading button data
15:22:41gevaertsDerPapst: that's probably not sufficient. I want to know how the USB registers need to be initialized for everything to work properly, and I doubt if the bootloader already does that
15:24:06samarskyriderhave you read my post about EvilG build?
15:25:35DerPapstgevaerts: then you might have to disassemble i guess. Maybe the diskmode image from the aupd image in the apple firmware might be a good start.
15:26:09DerPapstsamarskyrider: have you tried an official build to confirm the buttons work there?
15:28:24amiconnSlasheri: I think the global SYS_REMOTE_PLUGGED event should only be fired once, although that needs separate handling for the actual lcd init (all the repeated tries)
15:28:48gevaertsDerPapst: I was afraid of that. Unfortunately I don't have any experience in that field, so I think I'll leave it for now (it's the blocker for high-speed usb support, so if anyone knows how to do this kind of work and wants to use his player with only rockbox, here's what to do)
15:29:33samarskyriderDerPapst yes, i was using it since june and now decided to try EvilG
15:29:52Slasheriamiconn: rockbox doesn't know when the lcd gets initialized?
15:30:15amiconnThe lcd is write-only, so no
15:30:31amiconnWe just have to try many times, in case the remote is plugged slowly
15:31:36 Quit MethoS- (Read error: 110 (Connection timed out))
15:31:36 Quit MethoS_mobile (Read error: 110 (Connection timed out))
15:31:52Slasheriamiconn: btw, do you want to try the new bootloader 7-pre4 for h100? i just got that in release ready state :)
15:33:21*amiconn is still running the official bootloader v6
15:33:32gevaertsAnyone here familiar with tCardInfo and friends ? I have a report in FS #8562 of usb_storage on a sansa e280 reporting a size exactly 1024 times too small. It works correctly on my c250 and rasher's e260
15:33:36amiconnThe h180 is too precious to risk it...
15:33:48Slasherinow it would be a good time to try
15:33:57Slasheriwell, so far no one has bricked it
15:35:13DerPapstsamarskyrider: and a current build works fine too? if so i guess EvilG messed something up.
15:36:41samarskyriderDerPapst: current build works fine, I've tested. Anyway someone must have came across this problem, because EvilG's build is rather old. I wrote him an email
15:38:40 Join grndslm [0] (
15:39:03BigBambiSlasheri: I can try now to see if the remote is fixed
15:39:20*BigBambi goes looking for it
15:39:28 Join FOAD_ [0] (
15:39:46SlasheriBigBambi: great, just a moment so i will commit the changes to svn and provide you the bootloader
15:40:02BigBambiOK cool
15:40:06DerPapstsamarskyrider: i don't think many users here use that outdated unofficial builds... so...
15:41:01BigBambiUnsupported builds are unsupported for a reason
15:41:57SlasheriBigBambi: do you have DCC working?
15:42:09pondlifeSlasheri: Could you help with a move towards a new H300 bootloader release? I'm using SVN quite happily but petur reported some booting issue and amiconn has a disk clunk problem IIRC.
15:42:52Slasheripondlife: that would be hard without the actual unit.. so much of extensive testing needs to be done
15:43:05pondlifeAh, that was my question really
15:43:20pondlifeIt seems there is some variety in behaviour across units here.
15:43:23BigBambiSlasheri: Dunno :)
15:43:27BigBambiGive it a go :)
15:43:31SlasheriBigBambi: well, i will try :)
15:45:13Slasherihmm, doesn't seem to work. But now i will have something to eat, back soon :)
15:45:45BigBambiFailed DCCs are a hungry business :)
15:48:40 Quit vcardenas ("Konversation terminated!")
15:49:17 Quit asdrubal (Read error: 110 (Connection timed out))
15:51:18 Quit stripwax ("Miranda IM! Smaller, Faster, Easier.")
15:54:28 Join stripwax [0] (
15:56:28 Quit FOAD (Read error: 110 (Connection timed out))
15:56:29 Nick FOAD_ is now known as FOAD (
15:58:00 Quit Slasheri ("Reconnecting")
15:58:05 Join Slasheri [0] (i=miipekk@rockbox/developer/Slasheri)
15:58:22SlasheriBigBambi: ah, lets try again. the problem was i was somehow unregistered
15:58:34Slasherigreat, it worked
15:58:37BigBambithat's better :)
15:58:46BigBambiOK, will try now
15:58:53Slasherinow just wait until i commit the changes
15:59:09Slasheriyou can't try it before that because checksums don't match
15:59:34BigBambiah OK
16:06:48SlasheriBigBambi: committed. You may try now with the latest SVN build
16:06:57BigBambicool, just getting it
16:07:41 Quit ender` (" Microsoft is like a case of herpes. You can't kill it off entirely, just force it to go into a dormant stage for a while.")
16:07:57BigBambiwell, building
16:08:02BigBambi'tis quicker
16:10:26BigBambiSlasheri: Is there a tool to convert the .hex OF to .bin (I've also taken a ROM dump)
16:10:45Slasheriyep, in tools directory there is descramble-tool
16:10:54Slasherithat can do it
16:11:21Slasheriit's also planned to support flashing an iriver .hex file directly
16:11:30BigBambifor now I will just descramble
16:11:43Slasheri:) but you don't need to do that
16:11:49Slasherior are you willing to try?
16:12:04BigBambiCan do
16:12:08BigBambihaha, I'm a tool
16:12:18Buschelhorscht: you there?
16:12:18Slasherihmm? :D
16:12:24BigBambiI was wondering why it wouldn't unzip - the player wasn't plugged in :)
16:12:31Slasherihehe :D
16:12:45 Join ast3 [0] (
16:13:25Buschelhorscht: pm is up
16:13:42Horschtpm? a query?
16:13:56Buschelprivate chat, did you receive it?
16:14:00Horschtif so, no it's not for some reason
16:14:10Horschtare you registered on this network?
16:14:53Buschelworks now?
16:15:02Horschtstill not.
16:15:02 Part ast3
16:15:12Buschelhmm, can you try to contac me?
16:15:14SlasheriBigBambi: did you got it flashed?
16:15:22Horschtsure, just a second
16:15:26BigBambiSlasheri: Just about to
16:15:45 Join ender` [0] (
16:15:46Horschtdid you receive it, Buschel?
16:15:49Slasherigood, i was just worried why you were converting the OF to .bin
16:16:00Buschelyes, can you see my response?
16:16:26BigBambiSlasheri: So just to be sure, all I'm doing for now is flashing bootloader.bin?
16:16:28 Join kpanic [0] (n=kpanic@pdpc/supporter/sustaining/kpanic)
16:16:42Horschtbrb, parents got PC problems
16:16:54Slasherithe bootloader.iriver
16:16:55BigBambigood job I checked :)
16:17:02BigBambisorry, that is what i meant
16:17:04kpanichi, a program for linux that uploads ogg/flac without reencoding in mp3?
16:17:07Slasheriand that you will flash inside from rockbox
16:17:15Slasheriok, then it's good
16:17:38Slasherii just thought you were going to patch the OF with bootloader.iriver.. and that would just brick your player :)
16:17:40BigBambiOK, it flashed fine
16:17:58BigBambiNow to see if it'll boot from the remote
16:18:29BigBambiSlasheri: success!
16:18:37Slasherihehe, great!
16:19:28kpanicI'm using gtkpod
16:20:36BigBambiSlasheri: Any other testing required?
16:21:13Horschtkpanic, any filemanager. Gtkpod is for managing music on your ipod for use with the official FW
16:21:18Horschtand oficial FW can't play ogg and flac
16:21:45BigBambikpanic: Just drag and drop
16:21:46 Quit stripwax ("Miranda IM! Smaller, Faster, Easier.")
16:21:57kpanicin what folder?
16:22:15BigBambiWherever you want
16:22:25BigBambiIn .rockbox is a bad idea though
16:22:31kpanicright, just installed rockbox
16:22:41BigBambiJust make /Music or /WhateverYouWant
16:22:44Horschtany you like
16:22:44Horschtall you have to do from within rockbox, is to browse to that file
16:22:45Horschtand play it
16:22:55kpanicok, thanks people
16:24:05SlasheriBigBambi: nothing more comes in mind
16:24:19 Part kpanic
16:24:46BigBambiOK :)
16:24:49SlasheriXavierGr: are you here? you could try now the h100 version of the fixed bl :)
16:24:55 Quit raky_ (Remote closed the connection)
16:25:19 Join rvvs89 [0] (n=rvvs89@pdpc/supporter/active/rvvs89)
16:29:12 Join redwings0921 [0] (n=44249fa6@gateway/web/cgi-irc/
16:29:24redwings0921hi can any1 help me?
16:31:12krazykithave you read the irc guidelines linked in the topic yet?
16:31:42krazykitplease do so before asking your question :)
16:32:27redwings0921hah i see i already broke 1 rule by asking for help =)
16:32:49redwings0921ok i read it
16:33:23redwings0921im using winff to convert some movie to my iriver h10 but every time i convert this movie the sound is behind like 15 seconds
16:33:31BigBambiany1 also broke them :)
16:34:08 Join Zeph0x [0] (
16:34:40redwings0921has my problem ever happened to any1 else b4?
16:34:44BigBambiSlasheri: Is there any disadvantage of the ROM image over the RAM image (in terms of user use) given the ROM image saves a little bit of RAM?
16:34:52BigBambiredwings0921: Please use proper words
16:35:21SlasheriBigBambi: yes, if you want to update a new ram/rom image to flash you must first boot from disk or a ram image
16:35:37Slasheriso flashing doesn't work when running rom image. No other disadvantages
16:35:46BigBambiOK, cheers
16:35:56*BigBambi waves goodbye to the OF
16:35:59BigBambigood ridence
16:36:06redwings0921Ok. I am using WinFF to convert movies to my IRiver H10 but whenever I convert it the sound is always behind like 15 seconds. Does anyone have a solution?
16:37:12krazykitwhen you play the converted video on your computer, are the sound and video properly in sync?
16:37:34redwings0921yes, i have already checked that
16:37:55n1shave you enabled frame skipping in mpegplayer?
16:38:05redwings0921may i ask wat that is?
16:38:33*BigBambi suggests the manual
16:38:59redwings0921h/o let me try it
16:39:30 Quit ol_schoola (Connection timed out)
16:39:38BigBambiredwings0921: PLEASE use proper words. We have many blind people who use screen readers and many people without English as a first language, and we try to help them out
16:39:53redwings0921my bad
16:40:34redwings0921When i enabled frame skipping it went into the movie for about 10 seconds and then ended
16:40:57BigBambiWhat properties does the movie have?
16:41:07BigBambii.e. bitrate, size etc
16:41:11redwings0921before or after the transfer?
16:41:20BigBambiafter you have converted it
16:41:31redwings0921one second
16:42:36Nico_Pgevaerts: what would I need to do to test the USB stack on the gigabeat S?
16:43:31redwings0921BigBambi: For some reason I can not view the video properties while it is in the H10.
16:43:49BigBambiWell look at in on your PC before you transfer it
16:44:06BigBambiOr just tell me what settings you put into WinFF for the conversion
16:45:12redwings0921Ok i will give you both
16:45:19XavierGrSlasheri: Nice commits, can't test right now though, I don't have the player with me, I will try later
16:45:30XavierGrSlasheri: I guess I will remove my patch now
16:45:34redwings0921-acodec mp3 -ab 128 -ar 44100 -vcodec mpeg2video -s 160x96 -b 224kb -strict -1
16:45:42redwings0921that is what i am using for the winff
16:45:47redwings0921it is one of the presets
16:45:50gevaertsNico_P: Good question. I guess apply the patch, check if USB_BASE is set correctly, make sure usb-drv-pp502x.c is compiled in, and try
16:45:52BigBambiyeah, looks fine
16:45:55SlasheriXavierGr: great. i will upload the h100 patch to the wiki as soon as you have verified it :)
16:46:22SlasheriXavierGr: can i dcc it for you?
16:46:24redwings0921one second and i will get you movie information before conversion
16:46:39gevaertsNico_P: usb-drv-pp502x should probably be moved somewhere else. It should support all ARC-based controllers.
16:46:41BigBambiredwings0921: I'm not too bothered about that
16:46:49BigBambiIt is the information after conversion
16:47:07BigBambithe conversion line itself looks fine
16:47:10redwings0921i can try converting for a third time and see my results
16:47:18BigBambiCan you play the converted one on your PC?
16:47:22SlasheriXavierGr: it would be great if you could also test with the included test sequence
16:47:24Nico_Pgevaerts: ok, I'll try that.
16:47:24BigBambii.e. after conversion
16:47:31gevaertsNico_P: Once you get everything built there's a good chance of it working, since we actually used the imx31 manual as our reference
16:47:32redwings0921and sound is messed up
16:47:42BigBambiah, so it is the conversion that is wrong
16:48:10redwings0921i believe so
16:48:28redwings0921and this movie is really funny and i want to watch it in study hall on monday
16:48:36BigBambiOK, now I do care - what was the source? And I assume it does play properly on your PC?
16:49:14redwings0921it plays perfectly on my PC and my source is a torrent
16:49:28BigBambisorry, for source I meant properties
16:49:55redwings0921what is the source?
16:50:10redwings0921like how do i find it?
16:51:13redwings0921Audio bit rate is 128 kbps
16:51:20redwings0921it is an avi file
16:51:22 Quit K4rP4D (Connection timed out)
16:52:21redwings0921the frame rate is 23, the total bitrate is 152, the frame height is 352, the frame width is 608 and the data rate was 24
16:52:45Nico_Pgevaerts: any idea what I should set USB_BASE to?
16:53:48Nico_Pthere seems to be problems with UNCACHED_ADDR too
16:54:05gevaertsNico_P: Not really. I guess it should be somewhere in the iMX31 reference manual
16:54:11BigBambiredwings0921: I'm afraid it is very difficult to diagnose without being able to see it, and I don't do that much conversion.
16:54:22Nico_Pok I'll look it up
16:54:24XavierGrSlasheri: do you want to restore the OF too before trying?
16:54:43XavierGrot just flash with the plugin bootloader.iriver, rombox.iriver and rockbox.iriver?
16:54:58gevaertsNico_P: I don't know anything about how UNCACHED_ADDR is supposed to work. Probably best to ask one of the core people.
16:55:22SlasheriXavierGr: if you want to do the full extensive test, please follow the included release.txt :)
16:55:28redwings0921Bigbambi: ok thank you for your help. should i try another movie and convert that and see if it works... maybe its just the movie having problems
16:55:43BigBambicould be, and is worth a go
16:56:04SlasheriXavierGr: however, i think simpler tests are too enough
16:56:39redwings0921BigBambi: I am currently in the middle of trying to convert it again but this time i am also including a system file it came with
16:57:04 Quit magenpie (Read error: 110 (Connection timed out))
16:57:21gevaertsNico_P: I guess USB_BASE might be 0x43F8_8000 (from page 32-5 in MCIMX31RM.pdf)
16:57:48Nico_PI was looking at the same page :)
16:58:02 Join magenpie [0] (n=magpie@
16:58:06*pondlife likes it when a fix for one bug also fixes another...
16:59:11 Join karashata [0] (
16:59:17 Join kugel [0] (n=kugel@unaffiliated/kugel)
17:00:27gevaertsNico_P: UNCACHED_ADDR seems to be a pp502x-only thing. Maybe just define it to nothing for now ?
17:01:11Nico_Pthere are a few other values that need defining
17:02:05Nico_P(note that I haven't applied your patch yet)
17:02:13 Quit redwings0921 ("CGI:IRC (EOF)")
17:03:09 Join einhirn [0] (
17:04:44 Join Tuplanolla_ [0] (
17:04:52Nico_PI need to find how to enable the USB interrupt on the S
17:05:56gevaertsUSB_IRAM_ORIGIN shouldn't matter. It's only used in usb_benchmark.c, which you don't really need. The other two are used to enable the interrupt.
17:06:21Buschelcan anyone please change flyspray #8568 from "bug" to "patch"?
17:07:00Nico_Pgevaerts: I guess I need to use the USB_OTG IRQ
17:07:05 Join Tuplanol1a_ [0] (
17:07:32***Saving seen data "./dancer.seen"
17:07:33 Join FOAD_ [0] (
17:07:47 Quit Tuplanolla_ (Client Quit)
17:07:48 Quit Tuplanol1a_ (Client Quit)
17:07:52 Quit thegeek ("( :: NoNameScript 4.2 :: )")
17:08:13Buschelsoap: you there?
17:08:45soapjust got up
17:08:56soapI was about to ask you, Buschel, what you wanted builds for.
17:09:36 Join Tuplanolla_ [0] (
17:09:41Buschelah, ok. seems like backrolling jhMikeS' spinlock change solves the HDD problems with 5G 60/80GB. can you include it in your builds?
17:09:47soapsure can
17:09:48*Nico_P leaves... back later
17:10:54soapI'm going to run downstairs and brew some coffee, when I come back I can build twin builds - one for pcf-06 and one for pcf-07. All targets?
17:11:07Buschelsoap: what do you think about trying to get the DD patch to svn and do all the PCF-stuff later on in another patch-entry?
17:11:20 Quit miepchen^schlaf (Read error: 113 (No route to host))
17:11:30 Join Sebia2 [0] (
17:12:38 Quit Tuplanolla (Read error: 110 (Connection timed out))
17:13:31 Quit spiorf (Remote closed the connection)
17:13:32 Nick Tuplanolla_ is now known as Tuplanolla (n=jani@unaffiliated/tuplanolla)
17:14:05 Join thegeek [0] (
17:14:32 Quit karashata (" HydraIRC -> <- Go on, try it!")
17:15:14pondlifeSilly question, where are the keymaps defined for the H300 non-LCD remote?
17:15:50 Quit grndslm (Read error: 113 (No route to host))
17:16:05pondlifeCurrently the long PLAY on the WPS maps to the pitch screen. I'd prefer it to be SELECT (i.e. back to browser)
17:16:47pondlifeI'd assume that a browser is generally more used than the pitch screen...
17:18:30pondlifeBuschel: In general, it's a good idea to get smaller pieces of work into SVN, rather than waiting for the big commit.
17:18:35gevaertsThere's a page on the wiki somewhere that has an overview of all supported players and a summary their hardware specs, but I can't find it. Can anyone point me in the right direction ?
17:19:23 Quit FOAD (Read error: 110 (Connection timed out))
17:19:24 Nick FOAD_ is now known as FOAD (
17:20:23pondlifegevaerts: is probably not the one, but might be useful?
17:20:36Sebia2Hi rockers, i have a JBM 20 and i want to use it has system for my slug(debian/GNU) It's possible to run system rockbox and use others partitions by another system ?
17:22:19BigBambirockbox will search for the first FAT32 partition and try to boot from that as aI understand it
17:22:40gevaertspondlife: no. I found that one, and I can probably use it to find all info I want, but IIRC there's a page somewhere with a table with pictures and specs of all/most players
17:22:47Sebia2and if i connect on usb, rockbox continu or not
17:22:57pondlifegevaerts: There is, but I can't see it either
17:23:13BigBambiSebia2: I don't understand that sentence
17:23:55gevaertspondlife: that's the one. Thanks
17:23:55 Join grndslm [0] (
17:23:55Sebia2Sorry BB,
17:24:54Sebia2BigBambi : if i connect my JBM with USB on my Slug . ROckbox can work at the same time.
17:25:23Buschelsoap: I'll post a svn-candidate patch-version (v15) in a minute.
17:25:26BigBambiWhat is a slug? And no, if you have mounted the drive over USB, rockbox cannot operate at the same time
17:25:28Sebia2 BigBambi : with officiel system i had USB connected do not disconnect. And i cannot use it
17:25:32 Join FOAD_ [0] (
17:25:38BigBambiSebia2: That is always true
17:25:39Sebia2shit :(
17:25:53BigBambiYou cannot have the OS and rockbox accesing the drive
17:26:20Sebia2Really :( .....
17:26:37 Join ol_schoola [0] (
17:26:45Sebia2SThis is a slug
17:26:55 Quit Tuplanolla (Nick collision from services.)
17:27:08pondlifeHmm, the H300 non-remote LCD table has no menu button... so what's BUTTON_RC_MENU?
17:27:40Sebia2BigBambi : On slug you don't have hard disk, that s why i would like to use my JBM 20
17:27:41petur2you mean non-LCD remote?
17:27:48 Join qweru [0] (
17:27:58pondlifepetur2: Yes
17:28:18pondlifeThe tables in keymap-h1x0_h3x0 seem very odd
17:28:42 Join moos [0] (
17:28:45petur2I only get out the remote if somebody mentions a bug ;)
17:29:11 Nick petur2 is now known as petur (
17:29:29pondlifepetur: Maybe I'm not understanding, any chance you could educate me?
17:29:34soapBuschel, I'm honoured you asked me, but I'm just your mule in this exercise - my opinion is the DD stuff has shown no bugs yet - and anecdotally it appears at least twice as many people are using this build as are reporting runtime tests - suggesting many are using it as their daily build.
17:30:05BigBambiDD being device disable?
17:30:17soapjynx - owe me a coke.
17:30:17peturpondlife: nevermind me, I only have an LCD remote :/
17:30:42pondlifepetur: OK. Are there only 2 types of remote?
17:30:46pondlifei.e. LCD and non-LCD
17:30:50HorschtI use it as my daily build as well. It's just that it enhances my battery life. that's awesome
17:30:51BigBambiI use Buschel's patch on the E200 with no problems. I haven't reported any runtimes as it seemed to be covered on that
17:31:07BigBambipondlife: Three I suppose - non-lcd, H100 lcd and H300 lcd
17:31:26soapI'll build a SVN candidate build - ask for people to abuse it - and also build the two pcf builds - ask for runtime on those?
17:31:29pondlifeAh, I assumed the H100 LCD and H300 LCD were the same
17:31:38BigBambiWell, they look different certainly :)
17:31:46BigBambiI don't know about internally
17:31:53pondlifeOK, they probably are then
17:32:06 Quit petur ("bbl - computer hardware repairs...")
17:32:15soapIt is not normally good practice to have people testing two patches at once (re: spinlock revert) - but I assume in this case it is kosher as the spinlock patch is a revert of a specific action?
17:32:18pondlifeThe non-LCD has no way to get to the browser, but it does have a way to get to the pitchscreen. This is silly IMHO.
17:32:35Horschtit is, soap
17:33:02 Join Tuplanolla [0] (n=jani@unaffiliated/tuplanolla)
17:33:17Horschtit's a revert of a change introduced with svn 16005, which caused the disk trashing on 5G/5.5G ipods
17:33:22soapHorscht, "it is" as in "it is kosher" or as in "it is not good practice"
17:33:35pondlifeBigBambi: Yes, that's what it does.. but I'd like to be able to get from WPS to browser somehow
17:33:41Horschtit should be kosher, soap
17:33:49pondlifeThere is no STOP or SELECT on the remote
17:33:55BigBambiSure, I just meant that so you could see the three different remotes
17:34:07BigBambiAll of which work on both Hx0 and H3x0
17:34:28pondlifeOK, I only need to modify the H300 non-LCD
17:34:48pondlifeBut I need educating as to how the keymap tables work
17:35:39pondlifeThere's no dedicated table for the non-LCD
17:36:31pondlifeI don't suppose it's even possible to detect it and use a different table, is it?
17:36:52pondlifeIt just gets a subset of the other ones, hence no browse
17:36:53BigBambiwell rthere are some dedicated parts
17:36:53n1sI think it is detected
17:37:14BigBambiI see things like static const struct button_mapping button_context_quickscreen_nonlcdremote[]
17:37:30 Join seablue_ [0] (
17:38:00n1sline 818 case REMOTETYPE_H300_NONLCD: /* FIXME: add its tables */
17:38:05BigBambiBut it looks like to do it nicely you might want to split the three out fully
17:38:28pondlifeI'll just do the WPS first
17:38:46pondlifeDon't want to make a big red delta ;)
17:38:48Sebia2HI, It's possible to drive LCD (on JBM 20) when USB is connected ?
17:38:52 Quit xushi_ ()
17:39:08BigBambiSebia2: To do what? There is an image on screen sure
17:39:25BigBambiBut you cannot have disk access
17:42:03 Quit FOAD (Read error: 110 (Connection timed out))
17:42:03 Nick FOAD_ is now known as FOAD (
17:44:06Buschelsoap: v15 is up. please use it together with fs#8568 for all targets. pcf06/pcf07 are test builds only −− can you add them additionally for iPod Video 64MB?
17:44:21soaponly for the iPod video 64?
17:44:30soapon it.
17:44:59Buschelthanks so much! one build takes 28min here...
17:45:56soapI won't tell you how fast make -j3 && make zip takes here then.
17:46:10Buscheltell me
17:46:15BigBambi28 seconds? :)
17:46:27*Buschel sighs
17:46:32 Quit seablue (Read error: 110 (Connection timed out))
17:46:50BigBambinah, a bit more than that for me
17:47:40 Quit ol_schoola ()
17:49:12soapstill would like to see a RMAA comparison between SVN and DDv15 with its lower voltage for the WM codec supply. Maybe I'm paranoid.
17:49:17Sebia2BigBambi: if i host my system on JBM 20, i want to use LCD to display information
17:49:26soapam I just overly paranoid?
17:49:42BigBambiSebia2: Then would have to write something for it which is off topic for here
17:50:22Sebia2BigBambi: sorry my english it's good
17:50:29BigBambiBuschel: About 1 min 10 for E200 normal from make clean here
17:50:49BigBambiSebia2: Your English is fine, but this channel is for Rockbox
17:51:09BuschelBigBambi: factor of 24 then... maybe I should _really_ buy a new PC
17:51:39BigBambiSoap is even quicker... Mine is just on a laptop (oh, and I just did make, not a zip too)
17:52:14Buschelsee you later −− gotta play with my son, he's crying for me :)
17:54:33BigBambiBuschel: With ccache, 52 seconds for make && make fullzip
17:57:12Horschtmine takes 4 minutes :/
17:57:21Horschtjust make
17:59:01Sebia2But Rockbox manage the LCD when Archos is connected on usb ?
18:00:22 Join miepchen^schlaf [0] (
18:01:49 Join xushi [0] (n=xushi@unaffiliated/xushi)
18:02:50 Quit Siku (Nick collision from services.)
18:02:56 Join Siku [0] (
18:03:48 Quit tvelocity ("Αποχώρησε")
18:03:50 Join spiorf [0] (
18:04:45 Quit DirkGently ("Goin' Home")
18:05:43Sebia2Rockbox can change info on LCD when it's connected by usb ?
18:06:14pondlifeAnyone understand the keymapping stuff well. LAST_ITEM_IN_LIST__NEXTLIST(...) and CONTEXT_CUSTOM are mystifying me.
18:06:29pondlifeThat was a question, btw
18:07:19 Quit spiorf (Remote closed the connection)
18:07:34 Join Genesis [0] (n=genesis@
18:10:41gevaertsLook like reading from USB works a lot better if I lower the cpu speed on the player.
18:13:03 Quit fehmicans (Read error: 110 (Connection timed out))
18:13:16rashergevaerts: does that make any sort of sense?
18:14:06rashergevaerts: are you working on getting the current work in a committable state (to have *something*) or do you want to get hi-speed working?
18:15:22gevaertsrasher: It basically waits for the transfer to complete using a loop that just counts iterations until a maximum is reached. That's clearly not the right thing to do, but I didn't clean up that part yet.
18:17:20gevaertsrasher: I'm basically out of ideas about high-speed, so I'm concentrating on cleanup work. I think that cleanup should also include making it non-PP specific (since the Gigabeat S uses the same USB controller).
18:19:11gevaertsIs there an efficient way to measure time intervals to handle timeouts properly ?
18:20:25Nico_Pgevaerts: I'm back with a bit of time to try getting it to compile for the S
18:20:38 Part qweru
18:20:58gevaertsNico_P: great.
18:21:44Nico_Pgevaerts: you can use tick values
18:21:49Nico_Pand TIME_AFTER
18:22:19 Join star_jasmine [0] (n=arwyneve@
18:22:22midgeyhmm seems people are distributing an svn bootloader for the h300 on MR
18:23:30Nico_Pgevaerts: is there a handler func for the USB IRQ?
18:23:39gevaertsNico_P: I guess current_tick is the current value, and HZ is the number of ticks in 1 second ?
18:25:05BigBambimidgey: Are you registered there?
18:25:05gevaertsNico_P: usb_drv_int() in usb-drv-pp502x.c
18:25:46 Join spiorf [0] (
18:28:03 Join qweru [0] (
18:28:22 Join saratoga [0] (n=41becb3b@gateway/web/cgi-irc/
18:28:37 Join Buschel_ [0] (
18:28:53 Quit Buschel (Read error: 104 (Connection reset by peer))
18:29:43 Quit qweru (Client Quit)
18:30:46Nico_Pit compiles :)
18:30:58gevaertsNow will it work ?
18:31:10Nico_PI somehow doubt it :p
18:32:59gevaertsPessimist !
18:33:01*Domonoky just created his first voicefile with rbutil with cli interface :-)
18:33:06Nico_Pis something supposed to happen when I plug USB in with the SVN code?
18:33:12pondlifeDomonoky: Great
18:34:14 Join csc` [0] (n=csc@archlinux/user/csc)
18:34:27gevaertsNico_P: depends. On my sansa c250, nothing happens on the LCD (there should be a USB connect screen, but I don't get that for some reason), and the PC should see a connection.
18:34:44Nico_Plinux didn't see anything here
18:34:50Nico_Pis it worth trying with your patch?
18:35:28gevaertsNico_P: No. That doesn't change anything in the basic connection detect code.
18:35:48saratogadoes anyone object to buschel's current patch going into SVN?
18:35:58saratogathe only questionable thing I saw was changing the default clock to 24MHz
18:36:28saratogaand he had pretty good justification for that since it significantly helps battery life on FLAC and MPC
18:36:38Nico_PI guess someone with more low level knowledge than me will have to take a look (e.g. jhMikeS)
18:38:01 Join Buschel [0] (
18:38:21gevaertsNico_P: I think you should get an interrupt on connect.
18:38:30 Quit kugel (Remote closed the connection)
18:38:39pondlifesaratoga: Does 24MHz make the UI less responsive?
18:38:51Nico_Pmaybe I could add a splash in the interrupt handler to see if it gets fired
18:39:02pondlife(I don't have a PP target to try.)
18:39:27saratogapondlife: i never noticed it on my E200, but I don't know about the ipods
18:39:46pondlifeiPod Video would be the target to check
18:39:47Nico_Phmm the splash is apps/ code... :/
18:39:50Horschtit does not appear so. At least I didn't notice any laggs on my 5G 80GB
18:40:00Horschtbut I only used mp3
18:40:11gevaertsNico_P: logf() ?
18:40:13saratogai suppose at 24MHz you are boosted much of the time anyway for mp3
18:40:25Nico_Pyeah maybe
18:40:27Nico_PI'll try
18:40:30Horschtnot sure wether other formats lag or not
18:40:53pondlifesaratoga: Could that not reduce battery life for MP3 ?
18:41:07saratogapondlife: surprisingly it does not
18:41:27saratogathe PP chips appear to have a fairly linear power/clock relationship
18:41:44saratogaand we do not currently adjust the voltage when unboosted
18:42:16pondlifeOn another topic - any Iriver LCD remote users around? Please give a try - you shouldn't notice any change.
18:42:33soapsaratoga, I'm in the process of making DDv15 builds and posting them for abuse testing - do you want that done first before commit?
18:42:56saratogasoap: i haven't followed the thread in the forums closely so i'm not sure
18:43:04saratogado you think more testing is needed?
18:43:25soapDevice Disable patch v15 AKA SVN candidate 1.
18:44:09Horschtah, was just added to fs
18:44:31soapHonestly, no - but I'm building SVN+DDv15+spinlock rollback, SVN+spinlock rollback, and SVN for possible bug chasing.
18:44:36midgeyBigBambi: i think i'm registered there, i'll have to check
18:44:59Nico_Phmm there seems to be absolutely no logf output, so maybe logf just doesn't work
18:44:59BigBambiMight be worth posting a big offical disclaimer
18:45:35soapas I mentioned previously - download numbers strongly suggest many more people are using the DD builds than are reporting battery life numbers in the thread - suggesting a large userbase already - and no verifiable bugs reported.
18:45:47 Part pondlife
18:45:58saratogasoap: yeah i think this will be the most tested patch ever committed
18:46:24 Join bluebrother [0] (n=dom@rockbox/staff/bluebrother)
18:46:41 Nick Zeph0x is now known as K4rP4D (
18:46:41saratogaregardless i'd like to get the go ahead from amiconn or one of the other ipod developers before we change their clock speed code around
18:47:26 Quit Buschel_ (Read error: 110 (Connection timed out))
18:47:47Nico_Phaha usb_detect for the S is just "return 0"
18:48:14Nico_Pno wonder nothing happens
18:49:53gevaertsNico_P: you might try borrowing from usb-fw-pp502x.c
18:49:54 Join karashata [0] (
18:50:21Nico_Pindeed. did you change it in your patch?
18:50:29 Join petur [50] (n=petur@rockbox/developer/petur)
18:51:15Nico_Phmm thing is usb_init_device is empty and I have no idea what's needed
18:51:31Nico_Psame for usb_enable.
18:51:57gevaertsNico_P: I just removed the reboot stuff from usb_enable()
18:52:49gevaertsNico_P: and added somethingas well : usb_enable needs to call usb_core_init() if on==1, else usb_core_exit()
18:53:23Nico_Pgevaerts: usb_enable and usb_init_device are simply empty for the S.
18:53:34Nico_PI guess nothing can work without usb_device_init?
18:54:18midgeyBigBambi: how does this sound
18:56:13BigBambimidgey: Sounds fine, but I think you missed "flashing" after during in the second sentence :)
18:56:37midgeyindeed I did
18:56:40DerPapst"restore the device had something gone wrong." sounds a bit like broken grammer?
18:56:56Horschtperfect english
18:57:00gevaertsNico_P: probably not. I'm not sure what's really needed there. Maybe svn blame can tell you who to ask what it does.
18:57:07DerPapstthen my english is broken
18:57:13Horschtit might sound broken to non native speakers, DerPapst
18:57:22midgeyI also added a part about including an MD5 sum
18:57:34BigBambiDerPapst: Not, that English is fine
18:58:02DerPapstok.. i'd probably replave had with "in case"... but whatever.
18:58:13BigBambiThat would work to
18:58:29DerPapstand sound less broken to non native speakers :-P
18:58:49BigBambiYou would have to change gone to went as well
18:59:21BigBambi"restore the device in case something gone wrong." is horrible english
18:59:50DerPapsti forgot about "went".. i'm sorry
19:00:03*DerPapst got only a 4 in english at school
19:00:18Horschtthou shalt be horribly punished
19:00:36*DerPapst runs away and hides
19:00:49BigBambiDerPapst: Forgiven :)
19:01:44 Part Sebia2
19:02:22*DerPapst unhides
19:02:27DerPapstthanks :-)
19:03:11 Join idnar [0] (i=mithrand@unaffiliated/idnar)
19:03:33 Part idnar
19:07:01 Join fasmaie [0] (
19:07:29 Join Arathis [0] (
19:07:34***Saving seen data "./dancer.seen"
19:08:04 Quit conando (Read error: 110 (Connection timed out))
19:10:54 Join barrywardell [0] (n=barrywar@
19:11:54 Nick fxb is now known as fxb__ (
19:12:36 Quit MethoS-- (Read error: 104 (Connection reset by peer))
19:17:12 Quit samarskyrider (Read error: 110 (Connection timed out))
19:25:12 Part xushi
19:29:39soapgrrrr - the [list] and [li] tags are broken on the forums.
19:29:43 Quit spiorf (Remote closed the connection)
19:32:15 Quit magenpie ("uh-oh...iceberg failure")
19:32:58 Join GodEater [50] (n=bryan@rockbox/staff/GodEater)
19:41:18 Quit karashata (" HydraIRC -> <- Organize your IRC")
19:47:50GodEaterfrom the wiki page I'm given to understand the button driver for the beast is now done
19:48:01GodEateris the routine for powering it off not done then ? :)
19:48:30gevaertsI fixed some timeout stuff, so now UMS is a lot more stable. I still get resets, but only about one every 100 or 200 MB instead of every 30MB
19:49:36GodEaterexcellent work that man :)
19:49:41GodEaterbtw, thanks for the wave :)
19:49:46saratogawhat causes the resets anyway?
19:49:54bluebrothergevaerts: how is write support progressing? Is it safe to try?
19:50:18*GodEater can't wait for UMS to be available on the S, MTP is so painful
19:50:21 Quit barrywardell ()
19:50:26BigBambiGodEater: Are you pimping for waves now? If so, *waves*
19:50:42gevaertsbluebrother: I copied about 1GB of data to it since this morning without any corruption
19:50:43GodEaterBigBambi: doesn't count - you're only allowed to wave to log readers :)
19:51:16BigBambiWell I saw it earlier, but didn't realise how much it meant to you :)
19:51:22 Join xushi [0] (n=xushi@unaffiliated/xushi)
19:51:27 Quit xushi (Remote closed the connection)
19:51:27bluebrothercool :)
19:51:35 Quit ompaul (Client Quit)
19:51:41*Domonoky has now a rbutilqt which can install bootloader, rockbox, fonts, doomfiles, voicefile, and create voicefiles and talkfiles.... uninstall and themes are missing :-)
19:51:46GodEaterit didn't really - but gevaerts was the only one to do it, so I thought I'd acknowledge it :)
19:52:11 Quit Arathis ("Bye, bye")
19:52:13GodEaterDomonoky: so the next thing on your list is to make it support installing over mtp :)
19:52:13gevaertssaratoga: At least some of them are caused by the transfer timing out too soon, so that's mostly tuning.
19:52:15*Domonoky means a rbutilqt with cli interface :-)
19:52:30bluebrotherDomonoky: just wanted to ask this ... rbutilqt can do this since quite a while ;-)
19:52:54bluebrotherwhat do you think about creating a new release before the cli gets release?
19:52:56GodEaterwhat does the cli give you ?
19:53:13bluebrotherare there much changes to the core stuff itself?
19:53:25Domonokyconsole interface, especially for blinds and visually impaired..
19:53:34GodEaterah ha - that's very useful!
19:53:57Domonokybluebrother: not much changes to the core, its mostly in a seperate dir.. :-)
19:54:40bluebrothergreat. So we could add those minimal changes soon and create a new release?
19:54:52bluebrotherand after that bugfix the cli and release 1.1?
19:55:08bluebrotherbtw, how does the cli look?
19:55:49Domonokyi will commit those minimal shortly, and try to make a static test version of the cli interface
19:56:43bluebrothertell me when we can create a new release
19:58:02 Quit fasmaie (Read error: 113 (No route to host))
19:58:15 Join fehmicans [0] (n=canavar@
19:58:25gevaertsI should probably start adding version numbers to my patches
19:58:44 Join Buschel_ [0] (
19:58:49 Quit Buschel (Read error: 104 (Connection reset by peer))
19:59:26Domonokybluebrother: it would be good to have some testers of the svn version. But we could make a release next few days..
20:01:24bluebrotherwell, I just posted a new binary of svn (as requested in the forms)
20:02:16bluebrotherhopefully we get some reports for that
20:02:56 Quit saratoga ("CGI:IRC (EOF)")
20:07:22Domonokybluebrother: FS #8571 :-)
20:11:07bluebrotherDomonoky: missed the attachment?
20:13:47GodEaterhypothetical question:
20:14:01GodEaterassume for the sake of argument that the gigabeat S port is complete
20:14:11GodEaterhow are we going to manage the install
20:14:47GodEatercurrently, we have to send a patched firmware to it over mtp just to get it to boot our code
20:14:48Domonokybluebrother: ups :-)
20:14:59BigBambiWell, if the bootloader has USB, a sendfirm based tool to patch the bootloader, then rockbox usb mode to install build?
20:15:12GodEaterthe bootloader as it stands at the moment (iiuc) looks for a file with a given header in it to load
20:15:17GodEaterthat file has to be sent over mtp too
20:15:48GodEaterI can't see anyway to get a complete ".rockbox" directory sent to the device as the first part of the install
20:15:56BigBambiMe neither
20:15:59Domonokyhow do you send this firmware over mtp ?
20:16:13GodEaterwe have to load it in three stages I think
20:16:22GodEaterunless our firmware patch can enable a UMS mode too
20:16:26BigBambiHence make to bootloader look for rockbox.gigabeat like normal but give it USB so we can use bootloader USB to transfer the builf
20:16:37BigBambiIf possible of course :)
20:16:48GodEaterand at that point, I think, this will be the first rockbox install where we have to install the bootloader first ?
20:16:55bluebrotherDomonoky: there are ui_*.h files included. Is that intentional?
20:17:00BigBambiProbably, but I see no other way
20:17:14GodEaterDomonoky: code is in utils/MTP/ if you're interested
20:17:25 Join goffa_ [0] (n=goffa@
20:17:41 Join linuxstb__ [0] (
20:17:46Domonokybluebrother: no, not intended, but doesnt hurt..
20:17:55bluebrotheroh, and btw: I think we could make the cli interface a folder parallel to rbutilqt, i.e. something like rbutil/rbutilcli
20:17:56GodEaterotherwise, we'll have to load the patched firmware to get our bootloader on device, send a "raw" rockbox.gigabeat to get UMS, and then send the rest of the files over that way
20:18:21Domonokybluebrother: yes, would be possible..
20:18:26bluebrotherand then put a second project file to rbutil/
20:18:35bluebrotherlooks somewhat cleaner IMO.
20:18:39BigBambiCan we not have a bootloader USB mode with UMS to send the whole build at once?
20:19:07GodEaterBigBambi: I don't know - I assume we can - but I'm not certain of course :)
20:19:19BigBambime neither
20:19:50BigBambiIn which case we just need to transfer the patched firmware (+ bootloader), then use bootloader UMS mode to copy the build over
20:19:58GodEaterI assume a "release" bootloader will default to RB too
20:19:58Domonokybluebrother: sure, will remember it for the nex rbutil cli session :-)
20:20:22GodEaterwhich means first boot of the device will fail
20:20:23BigBambiGodEater: I would certainly hope so
20:20:31bluebrothernice. Please keep posting to the tracker entry −− I want to watch progress on my end too ;-)
20:20:38BigBambiPeople will have to read the instructions!
20:20:43GodEaterand leave newbies that didn't read that far upset :)
20:20:55BigBambiTheir own fault to be honest
20:21:04*GodEater thinks that this hypothetical bootloader should have an "installation mode"
20:21:12BigBambiI still can't believe people replace the firmware on their devices without reading first
20:21:20BigBambiDespite all the evidence to the contrary
20:21:21GodEateri.e., if it doesn't detect a .rockbox directory, then it needs to tell the user
20:21:43GodEaternot just say "couldn't load rockbox.gigabeat"
20:21:43BigBambiWell, much like the current bootloaders "can't find rockbox.ipod" etc
20:21:52BigBambiWhat is the difference?
20:22:01GodEateruser friendliness
20:22:01BigBambiJust a more verbose message?
20:22:05GodEaterit won't matter to you or I
20:22:29GodEaterbut it would be nicer if at that stage it said something like "New installation: please plug device into your PC and run rbutil"
20:22:31GodEateror some such
20:22:51BigBambiSomething like ".rockbox directory not found, please connect to PC and install build" (but better written)
20:23:08GodEaterI guess we can work on that bit
20:23:30BigBambiBut I think the pronciple is the easiest way
20:23:34GodEaterbut again this will be the first rockbox install which requires a disconnect and reboot of the device to get a working build
20:23:48BigBambiI don't see a better way
20:24:06GodEaternor I currently
20:24:07BigBambiUnless we improve libmtp to deal properly with directories etc.
20:24:12GodEaterthat's not possible
20:24:17GodEaterMTP is file system agnostic
20:24:20GodEaterand is designed to be so
20:24:23BigBambiWindows manages it
20:24:29GodEaterdoes it ?
20:24:34GodEaterhow do you know ?
20:24:38BigBambii.e. I can create directories in windows
20:24:50BigBambiNot libmtp in particular, I mean MTP in general
20:24:53 Quit Thundercloud (Read error: 104 (Connection reset by peer))
20:24:55GodEaterno - you can create things which windows displays as directories
20:25:00BigBambiah, OK
20:25:05GodEaterthere's no saying that's how they're presented on the device
20:25:12BigBambifair enough
20:25:15GodEatershould you connect to the disk directly
20:25:20 Join asdrubal [0] (
20:25:23BigBambiIn that case there is no choice
20:25:38GodEaterI only say this because this is how I've understood what I've read of the MTP spec
20:25:39Domonokybootloader USB mode.. :-)
20:25:50BigBambiThat is what I have been saying...
20:26:04GodEateryes I think you're right
20:26:05 Join Thundercloud [0] (
20:26:11 Quit csc` ("Powering Off")
20:26:16GodEaterI was just thinking it through
20:26:24GodEaterand seeing if I'd missed anything obvious
20:26:32BigBambiYeah, needs some thinking about alright
20:27:05BigBambiI just don't see a way of transferring a build properly without a rockbox USB mode
20:27:25GodEaterI don't think there is a way
20:27:32BigBambiAnd there is no point buggering about transferring a rockbox.gigabeat if people only then need to connect to install a full build
20:27:39BigBambiJust do the whole lot in the bootloader
20:27:48 Join barrywardell [0] (n=barrywar@
20:28:09GodEaterthe other option of course is to implement an MTP "stack" in Rockbox as well as UMS
20:28:36 Join csc` [0] (n=csc@archlinux/user/csc)
20:28:36BigBambiI cordially dislike MTP
20:28:50GodEaterme too :)
20:29:10GodEatermy files go where *I* say, not where some arbitrary algorithm decides :)
20:29:25 Quit goffa (Read error: 110 (Connection timed out))
20:29:36GodEaterand they stay called what I call them!!!
20:29:36BigBambiJust give me a damn filesystem and be done with it
20:31:12 Quit linuxstb_ (Read error: 110 (Connection timed out))
20:33:06 Nick linuxstb__ is now known as linuxstb (
20:37:10 Join goffa [0] (n=goffa@
20:39:05bluebrotherDomonoky: nice work with the cli :)
20:41:25 Quit barrywardell ()
20:41:31 Quit star_jasmine ()
20:42:24Domonokybluebrother: thx :-)
20:43:44 Quit homielowe (Read error: 104 (Connection reset by peer))
20:44:46 Join homielowe [0] (
20:47:22gevaertsWhat is the actual resolution of udelay() ?
20:49:24 Quit goffa_ (Read error: 110 (Connection timed out))
20:50:42 Join robin0800 [0] (
20:51:04 Join linuxstb_ [0] (n=linuxstb@rockbox/developer/linuxstb)
20:53:45 Quit goffa (Remote closed the connection)
20:54:45jhMikeSgevearts: I wouldn't use that on gigabeat S
20:55:31gevaertsjhMikeS: use what ?
20:56:28jhMikeSthe udelay implementation. I'm guessing you were working on MSC there?
20:56:59gevaertsjhMikeS: I'm on sansa. Anyway I don't really need it.
20:58:01jhMikeSah...I just scanned the history really fast :) On PP it's +/- 1uS anyway.
20:59:06gevaertsjhMikeS: Nico_P was looking at getting MSC to work on gigabeat S
20:59:28 Join goffa [0] (n=goffa@
21:00:33 Quit tessarakt (Success)
21:00:55 Join tessarakt [0] (
21:04:43 Quit linuxstb (Read error: 113 (No route to host))
21:05:12gevaertsI think FS #8562 is now ready for real testing. It still needs cleaning up, and it still only works in full speed, but it should work. Note that I've only tested on linux.
21:07:37***Saving seen data "./dancer.seen"
21:07:38rashergevaerts: time to modify the bootloader as well?
21:08:06 Join linuxstb__ [0] (
21:08:56 Quit goffa (Remote closed the connection)
21:09:00 Join goffa [0] (n=goffa@
21:09:32 Quit Domonoky (Read error: 104 (Connection reset by peer))
21:09:48 Join Domonoky [0] (
21:10:50gevaertsrasher: there is still an issue apparently on e280 (see FS) that I don't really understand, I don't know if SDHC works or if the problems you had with it are still there (I only have 2GB cards), and the SD slot probably doesn't work on windows (pixelma had some problems with it)
21:11:06rasherI'll attempt sdhc
21:11:10jhMikeSwhat happens with the higher speed modes?
21:11:49 Quit robin0800 (Read error: 104 (Connection reset by peer))
21:12:51gevaertsjhMikeS: it doesn't transmit 512-byte packets correctly (it stops after 95 bytes), and smaller packet sizes are not allowed for bulk at high speed.
21:13:07 Join Arathis [0] (
21:13:34jhMikeSbtw, one thing I notices is the iMX.31 sdhc controller seems to be similar to the PP one as well but not identical. I'm wondering about a similar situation for the OTG controller.
21:13:54 Join ompaul [0] (n=ompaul@gnewsense/friend/ompaul)
21:13:56 Join linuxstb [0] (n=chatzill@rockbox/developer/linuxstb)
21:14:05gevaertsjhMikeS: there is a similar problem at full speed with 64 byte packets that makes it zero three bytes at the end of every 32-byte block, but we can work around that by using 16 byte packets
21:14:25gevaertsjhMikeS: That's very much possible.
21:14:28rashergevaerts: sdhc is still quite thoroughly broken
21:14:46gevaertsrasher: what does it do ?
21:15:06rasherThree lines repeat in dmesg:
21:15:07rasherattempt to access beyond end of device
21:15:07rashersdb: rw=0, want=11940744, limit=11661
21:15:07rasherBuffer I/O error on device sdb1, logical block 1491568
21:15:43gevaertsrasher: what size does it report on enumeration ?
21:15:50jhMikeSgevaerts: some sort of race condition between the controller and writing data perhaps?
21:16:06rashergevaerts: 6 MB
21:16:07*jhMikeS has no real idea without having had much involvement here
21:16:10rasherWhich is slightly off
21:16:12 Quit BitTorment (Read error: 104 (Connection reset by peer))
21:16:43 Quit Arathis (Client Quit)
21:16:49 Join BitTorment [0] (
21:17:27gevaertsrasher: then that's probably exactly the same problem that Tri Nguyen reports on e280.
21:18:07jhMikeSperhaps a clocking issue on the module too
21:18:37rashergevaerts: Anything above 4GB?
21:18:59gevaertsrasher: it uses card_get_info() to get its data, which seems to work up to 4GB, but not higher.
21:19:25rasherThe Rockbox Info screen reports the card's size correctly
21:19:37 Quit linuxstb_ (Read error: 113 (No route to host))
21:19:37gevaertsjhMikeS: possibly. I'd like to see USB-related register settings from the OF somewhere.
21:19:51 Join conando [0] (
21:19:53 Quit linuxstb__ (Read error: 113 (No route to host))
21:20:01 Join linuxstb_ [0] (n=linuxstb@rockbox/developer/linuxstb)
21:20:09 Join linuxstb___ [0] (
21:20:17 Quit linuxstb (Nick collision from services.)
21:20:22 Nick linuxstb___ is now known as linuxstb (
21:20:38jhMikeSdid you try disabling/enabling interrupts in certain time-critical sections if any exist?
21:21:31gevaertsjhMikeS: not yet. What's interesting is that the errors are very consistent.
21:22:16 Quit bluebrother ("leaving")
21:22:19 Join linuxstb___ [0] (
21:22:22jhMikeSthe tick could intrude on something in normal thread context.
21:22:45jhMikeSso could the USB interrupt for that matter
21:23:12 Quit linuxstb_ (Remote closed the connection)
21:23:19gevaertsjhMikeS: the USB interrupt shouldn't fire at those times.
21:23:21 Join jhulst [0] (n=jhulst@unaffiliated/jhulst)
21:23:36gevaertsOf course it ight be another interrupt.
21:24:00 Quit linuxstb (Nick collision from services.)
21:24:01 Nick linuxstb___ is now known as linuxstb (
21:24:14*linuxstb apologies for the spam...
21:27:30gevaertsrasher: I suspect that the info screen might get its info from the fat driver
21:27:52rasherIs get_card_info used anywhere in Rockbox?
21:28:09jhMikeSthe debug screen
21:28:22gevaertsrasher: can you check what disk info in the debug menu shows ?
21:28:48rasherJust a moment.. I forgot to actually unzip the build, so I had a bit of the old filesystem corruption..
21:29:56 Join Phill [0] (n=51aef816@gateway/web/cgi-irc/
21:33:58rasherWhich debug screen should I look for?
21:34:20rasherDisk info?
21:34:21Phillgavarts: debug screen reports blocks=0xB3C000 for builtin 6GB, and 0x2D8D for 6GB SDHC. Is that what you are looking for?
21:34:23gevaertsrasher: View disk info
21:35:10rasher4GB builtin: Blocks: 0x77D800
21:35:28rasher6GB sdhc: Blocks: 0x002D8D
21:35:37rasherThat sounds broken
21:35:57Phillgevaerts: that would seem to fit with the reported 6MB
21:36:01gevaertsYes it does.
21:36:41gevaertsAnyone here familiar with that code ?
21:38:39 Join Geeky123_ [0] (
21:38:50 Quit Geeky123_ (Read error: 104 (Connection reset by peer))
21:40:01rasherSeems jhMikeS is responsible for large parts of ata-c200_e200.c which might be the culprit?
21:40:27LloreanHow is size detected?
21:40:47*jhMikeS is getting blamed for everything these days
21:41:05BigBambiand praised
21:42:10PhillFS #7634 is where SDHC support was added. I assume only sdhc use the /* CSD version 2.0 */ section of the code.
21:42:24gevaertsDisabling interrupts does not fix the high-speed issue
21:42:32jhMikeSLlorean: have to calculate it from the blocks/banks info
21:44:43jhMikeSor is this about SD vs. SDHC? that's inferred from the fact that SDHC responds to SEND_IF_COMMAND
21:44:46 Join GrooveStix [0] (n=chatzill@
21:45:19GrooveStixhey folks, I got 10GB Gigabeat and I am reformatting it. Should I pick FAT32 or NTFS?
21:46:10LloreanGrooveStix: FAT32
21:46:16LloreanRockbox only supports FAT32.
21:46:22GrooveStixcool :)
21:46:24Lloreantessarakt: Please give serious answers in this channel.
21:47:40LloreanjhMikeS: I really want it to be SD vs SDHC, but I thought they said it happened on >4GB internal storage devices too.
21:47:59gevaertsThe problem occurs at least on 6GB sdhc and 8GB internal, but doesn't seem to happen on 6GB internal or 4GB internal
21:48:42jhMikeSLlorean: What's "it"? One thing I didn't add was the SDHC support.
21:50:17 Quit seablue_ (Read error: 104 (Connection reset by peer))
21:50:52 Quit jurrie_ (Read error: 110 (Connection timed out))
21:51:14rasherHrm, a readout of the debug screen for 8GB internal would be useful perhaps?
21:51:40asdrubalDo you guys think MicroSD HC is density limited?
21:51:51asdrubalLike what do you think microSD HC will max out at
21:52:05LloreanjhMikeS: I meant, "I really want the problem to be SDHC support" because then I can hope that it'll turn out to be something really obvious. :-P
21:52:06 Quit GrooveStix ("ChatZilla 0.9.80 [Firefox]")
21:52:19Lloreanasdrubal: That really has nothing to do with Rockbox.
21:52:39asdrubalI know Im just wondering
21:52:43asdrubalSorry to throw you off topic
21:52:43LloreanjhMikeS: Barrywardell added SDHC right?
21:52:56 Quit TMM ("Ex-Chat")
21:52:59Lloreanasdrubal: Please read the link in the channel topic
21:54:11gevaertsrasher or Phill: can you see what microSD version is given for both internal and SD ? If I understand the code correctly that should give a hint.
21:54:49jhMikeSLlorean: I think so
21:54:58Phillgevaerts: I think the error may be in ata-c200_e200.c line 698. The v2 spec has a fixed multiplier of 2^10, but it is missing from this line, where number of blocks is calculated.
21:55:35gevaertsPhill: that sounds correct anyway.
21:55:58 Join TMM [0] (
21:56:23rashergevaerts: The top line? SD04G? Rev 8. (the ? is a box)
21:56:41gevaertsrasher: I the line starting with microSD:
21:56:58gevaertsMine says v1.0-1.2
21:57:17gevaertsanyway I would say
21:57:23rasherSame for both, here
21:57:36gevaerts... that Phill most probably spotted it.
21:57:58rasherSure sounds quite likely
21:58:07 Join raky [0] (
21:58:09*gevaerts has an enter key with some kind of magnetic powers
21:59:01gevaertsCan you check if it fixes the issue for you ?
22:00:05Phillgevaerts: compiling now...
22:00:08rashercurrcard->numblocks = c_size; => currcard->numblocks = c_size << 10; ?
22:00:16rasherI'm far too slow..
22:00:45Phillgevaerts: works perfectly now :)
22:01:08gevaertsGreat. I guess that that can be fixes directlyin svn by someone ?
22:03:15Domonokyto use every communikation channel: RBUTIL needs testing. windows gui binary: windows cli binary : for details see forums or mailinglist :-)
22:05:48LloreanDomonoky: Question: will the CLI version allow command line parameters such that I could type something along the lines of "rbutil −−target=IP4G −−patch=/mount/IPOD −−install" to install the latest build? (Or does it offer such functionality already, and ignore my random guessing about target names)?
22:06:26rasherAlright, I'll commit that change to ata-c200_e200.c (wonderful that the least competent of us holdes that power).. the << 10 is correct, right? It sure is working at least..
22:06:33DomonokyLlorean: at the moment it doesnt have parameters, its has a cli menu ..
22:07:06Domonokybut it could be possible to add parameters to shorten the needed input..
22:07:15LloreanDomonoky: I figured it had a menu. :) I just kinda like the idea of parameters as another option for using it.
22:07:37*rasher risks it
22:07:38LloreanBut that's just me, and that's hardly a priority or anything. I was just curious if that was in the works at all.
22:07:54gevaertsThe bootloader doesn't compile because it misses ata_get_identify() in usb_storage. Probably the best solution would be to disable usb_storage entirely for the bootloader. How do I do this ?
22:08:08Domonokythe only problem with parameters are, that there are mny config options, so many parameters..
22:08:29Phillrasher: yep. works for me, and taken from the spec.
22:08:50linuxstbgevaerts: What do you mean by "misses ata_get_identify()" ? The function isn't being compiled?
22:08:58LloreanDomonoky: Well leave it to another day, then.
22:08:59rasherPhill: I forgot to credit you for pointing it out in my commit message.
22:09:20LloreanI think we would like a bootloader USB mode to work.
22:09:31LloreanEspecially if flashable bootloaders are in the future
22:09:40DomonokyLlorean: yup, lets first get this cli menu ready, then we will think forward .. :-)
22:10:11rasherDomonoky: perhaps just the simple options/actions should be available through commandline switches anyway?
22:10:31linuxstbbootloader USB (on PP) requires enabling interrupts, which doesn't happen atm.
22:11:09Phillrasher: I'll forgive you!
22:11:43Domonokyrasher: feel free to improve the cli interface, the code is in the tracker :-)
22:11:46rasherLet it be publically known that Phill is the genious, and I'm merely a messenger.
22:11:57gevaertsI "fixed" it now by only defining USB_STORAGE in usb_core.c for non-bootloader builds. The bootloader gets USB_CHARGING_ONLY.
22:12:04Lloreanrasher: In my mind, the only two options really needed are "install" and "bootloader install", with "device" and "location" for those. I know it's boiling things down a lot, but at the same time who's going to barebones it if they don't want to boil things down?
22:12:28rasherLlorean: I agree.
22:13:10rasherOf-by-1024.. that's a new one
22:13:30DomonokyLlorean: + proxy and cache options, and which build to install (archived/daily/current) ...
22:14:33 Join GrooveStix [0] (n=chatzill@
22:15:09LloreanDomonoky: To me at least, I wouldn't mind it always installing the current when done like that, so long as the menu is available for further configuration when other things are needed.
22:15:20 Join bertrik [0] (
22:15:28GrooveStixhey people! I was just asking about formatting here, but I gots another question fer yew! Is it smart to defrag my Gigabeat?
22:15:44 Quit Buschel_ ()
22:16:19 Join thgz [0] (
22:16:31linuxstbGrooveStix: I thought you just formatted it?
22:16:48GrooveStixyes, I am talking in general
22:17:14GrooveStixas in, defraging from time to time
22:17:43 Join Bagder [241] (n=daniel@rockbox/developer/bagder)
22:18:09thgzHaving some problems with RB compiling... When I run makefile, it says that can't create sysfont.h
22:18:27 Join austriancoder_ [0] (n=austrian@
22:18:30GodEaterGrooveStix: defragging is only necessary if you regulary remove and then add new content to the device
22:18:31GrooveStixlinuxstb: any ideas?
22:18:39GodEaterotherwise the files will remain stationary on the disk, and not become fragmented
22:18:44thgzUsing cygwin. Checked the path variable; all ok. Removed and re-downloaded from svn; Problem still exist.
22:20:07GrooveStixGodEater: I understand, I am only interested how essential it is... I guess it shoukd be snappier after I defrag. The reason I am asking is I defraged last night and it seems I have a HDD damage, so I was wondering if the defraging did it, or not
22:20:48GodEaterand as I say, defragging won't make any difference at all if you just copy stuff onto it, and never delete things
22:20:50gevaertssansapatcher at least has no problems updating the bootloader over the rockbox-ums support
22:21:26 Quit Phill ("CGI:IRC (EOF)")
22:21:31GrooveStixGodEater: alright. that means - defraging is for me.
22:23:37*DerPapst is *very* ipressed about gevaerts progress...
22:24:14*Bagder seconds that
22:24:20rasherthgz: try adding V=1 in front of your make command
22:25:51thgzRasher: Oh thanks, why I didn't do that before...
22:26:48*gevaerts has done ehci work before, which made this a lot easier
22:27:34 Join RoC_MasterMind [0] (
22:27:58 Join anid [0] (n=54ad210b@gateway/web/cgi-irc/
22:28:02 Quit RoC_MasterMind (Client Quit)
22:28:05 Join karl_ [0] (
22:28:14anidhi di ho
22:29:17karl_is there somewhere that describes the process of porting a game to rockbox?
22:29:17anidwhats up men?
22:29:27GrooveStixwhat is keyclick?!
22:30:12thgzGrooveStix: When you press a key you'll hear a click
22:30:31thgzDon't turn that swith on on PortalPlayer based devices
22:30:34Domonokykarl_: see and the source of other plugins :-)
22:30:43karl_thanks very much
22:30:49GrooveStixI just saw it, i guess it's new eh?
22:31:01karl_keyclick has been around for a while
22:31:04karl_or maybe beep
22:31:08GodEatera couple of weeks or so
22:31:09karl_mine always had beep
22:31:13karl_not necessarily clock
22:31:14GodEaterI'm not a fan of it :)
22:31:25karl_beep always irritated me
22:31:39GrooveStixyes, mine had a beep
22:31:46GrooveStixand yes, I second karl_
22:31:50thgzWell, got the following (that's sho short I think that can pasted here):
22:32:19*DerPapst already sacred
22:32:27thgzNot... I'll upload that to pastebin.
22:32:49 Quit austriancoder (Connection timed out)
22:35:01*rasher copies hundreds of megabytes to sdhc
22:35:07rasherLet's see how that goes
22:35:15*Bagder holds breath
22:35:28rasherBagder: You may want to rethink that - remember it's at full speed
22:35:38*Bagder slowly turns blue
22:35:42rasherWhich is bizarrely slower than hi-speed.
22:35:47thgzOk, here it is:
22:35:49rasherWell done there, USB consortium
22:36:16Horschtinteresting naming scheme
22:36:26Horschtmake full speed run slower than high speed
22:36:28anidwhere can i download the sansa media converter?
22:36:40Bagderanid: ask sandisk
22:36:42anidi just found out my sansa is a v2
22:36:49anidso no rockbox :(
22:36:55gevaertsrasher: the next one will be "super-speed"
22:37:17midgeylight speed?
22:37:21thgzAnid: Didn't you get that software with your unit?
22:37:28Horschtuber speed
22:37:31Bagderthgz: convbdf fails to create that file there
22:37:37BigBambithgz, anid on topic please
22:37:49rashergevaerts: 4.0 will include exxxtreme speed, I hope
22:38:05 Join star_jasmine [0] (n=arwyneve@
22:38:43star_jasminehi everyone... I have a question. your rbuilt tool that was posted on the rockbox list... is that just for linux users? stupid I know... you said it was a windows binary
22:38:50star_jasmineI'd like to help if I can
22:38:56thgzBigbambi: Causing to manage a heart attack to you again :P... Sorry, I try to keep on topic
22:39:00 Quit jhulst (No route to host)
22:39:07BigBambithgz: I'll live :)
22:39:25thgzBigbambi: Great
22:39:28anidsorry for my disturbance
22:39:39BigBambistar_jasmine: It can be compiled for Windows, Mac OS X and linux. Domonoky just posted Windows binaries
22:39:49gevaertsrasher: probably
22:40:36thgzBagder: Found that out, can't however understand why
22:40:38star_jasminewhen I tried installing it, it said I was missing a .dll file. I"m not sure which. I'll get the message again
22:41:09Domonokystar_jasmine: on which one ? cli version oder the normal gui version ?
22:41:58thgzBtw how the cli versiong goes? Is it ready for normal use?
22:42:42Domonokythgz: it needs testing, and uninstalling and themes are missing.. (all install options are coded today :-) )
22:43:06thgzDomonoky: Oh, Great!
22:43:22star_jasminethe second link that was posted. not sure if that was the gui or ordinary one. I'm looking to test the ordinary one, being visually-impaired
22:43:32Ice``just to laugh a moment >
22:43:57BigBambiIce``: On topic only here
22:43:58Domonokyarg, it seams i failed to build the cli version as static binary.. mingw.dll is needed :-/
22:44:03gevaertsWould jtag make it possible to read USB register settings on a PortalPlayer device ? And does anyone here have the equipment and experience to do that ?
22:44:04star_jasmineoh idiotic me. I need the first one. apologies
22:44:11Ice``BigBambi: just look the link. You'll see Y.
22:44:26BigBambiI have. It is off topic
22:44:31Ice``Yes it is.
22:44:35Ice``But it's fun.
22:44:37BigBambiSo do not post it here
22:44:37rashergevaerts: I believe LinusN has a Sansa e200 dev-board with jtag port lying around, but has never had time to do anything with it
22:44:43BigBambiIce``: I don't care
22:44:49Ice``BigBambi: it's too late, i'm sorry.
22:45:05BigBambiOK, just keep in mind the channel guidelines for next time
22:45:16Ice``Ok, no problem.
22:45:32 Quit anid ("CGI:IRC (EOF)")
22:47:09gevaertsrasher: I'll ask him when he appears
22:48:36 Join linuxstb___ [0] (
22:49:24rasher This one
22:49:36GrooveStixhmmm, I wonder, is there anyplace I can buy spare parts for my gigabeat?
22:49:39star_jasmineis it possible to add the speex encoder to the list of possible encoders, or do I have to install one?
22:49:40GrooveStixdoes anyone know?
22:49:56rasherstar_jasmine: The speex encoder should be builtin
22:49:58GrooveStix(and yes I know this is rockbox :) )
22:50:12star_jasmineoh to the rockbox utility? ok. I"ll have to see if it shows up as an option
22:50:31rasherstar_jasmine: That said, I don't know quite how it works.. Domonoky should know
22:50:38Domonokystar_jasmine: its built in, you have to choose it, in the configuration... :-)
22:51:18 Quit ashes ("leaving")
22:51:45Domonokyi would recommend to try the cli version, but unfortunatly you have to wait till i have regenerated Qt and rbutilcli to have a static build :-)
22:52:28gevaertsrasher: is your copy still going ?
22:53:02star_jasmineI have a friend who is willing to test the mac version of your rockboxbuild utility for the mac, when she acquires one with voice over. it won't be for a while yet, but she's interested
22:53:27karl_is the easiest way to look at the plugin source just to check it out of svn?
22:53:32rashergevaerts: just finished!
22:53:40Bagderkarl_: yes
22:53:46rashergevaerts: I'll md5sum it from the sansa firmware
22:54:04 Join axionix_ [0] (
22:54:09rasher.. which I can't
22:54:13rasherSince it's sdhc
22:54:15star_jasmineok thanks. found it.
22:54:33gevaertsrasher: implement md5sum in the rockbox file manager ?
22:54:58rashergevaerts: I'll just use the cardreader
22:57:00 Join DataGhost [0] (i=dataghos@unaffiliated/dataghost)
22:57:13star_jasmineoh... jus tanother quick question. two actually. are the numbers beside the configuring tts the quality settings that are present in cygwin, or do these refer to speed of the voice?
22:57:38amiconnrasher, gevaerts: There is another option how to test reliability of disk transfers in rockbox, which could be used with a little trick:
22:57:55star_jasmineand the other is, can sapi 4 voice option be added to the tts options? I love the accessibility of this utility.
22:58:02rashergevaerts: the md5sums check out perfectly
22:58:07rasherFor 400 MB data
22:58:17amiconnEnable test_disk.c plugin, and compile a simulator
22:59:01amiconnOh, change the path in test_disk.c before compiling so it creates its test dir in some (fixed) sub-dir
22:59:25 Join linuxstb____ [0] (
22:59:34Domonokystar_jasmine: it seams you are now trying the gui version... and yes those options are for quality and volume and more.. but i would really recommend to wait for the cli version.. its should be easier to use..
22:59:44amiconnThen mount the sansa (or other PP502x) target, and symlink it as the chosen directory from within the archos/ dir of that sim
22:59:53Domonokysapi4 isnt supported at moment, maybe in the furture...
23:00:04GodEaterDomonoky: why is the speex encoder not the default ?
23:00:11 Quit linuxstb (Read error: 113 (No route to host))
23:00:19 Nick linuxstb____ is now known as linuxstb (
23:00:35amiconnNow start the sim, run test_disk.c, and choose the "Write & verify" test
23:00:37DomonokyGodEater: because nobody got the time to bind the encoder selection to the selected device... :-)
23:00:44star_jasmineok thanks. I"ll do that. wait that is. this one is also usable
23:01:04GodEaterDomonoky: so it's a feature request currently ? :)
23:01:14amiconnThis will write 300MB of pseudorandom data, and verify it by reading it back
23:01:30amiconnThe amount of data is also just a #define in test_disk.c
23:01:51 Join MethoS- [0] (
23:01:56gevaertsamiconn: looks useful
23:01:57rasheramiconn: At least a good part of that will live in the disk cache though
23:02:09DomonokyGodEater: i planned to do it sometime, but no written feature request.. :-)
23:02:22amiconnHmm, maybe on linux it will. Certainly not on windows though
23:02:29 Quit GrooveStix ("ChatZilla 0.9.80 [Firefox]")
23:02:31 Join crzyboyster [0] (n=6210f026@gateway/web/cgi-irc/
23:02:47rasherYeah, Linux seems pretty willing to put many megabytes of disk cache in RAM
23:02:50 Quit star_jasmine ()
23:02:53rasherProbably 300 though
23:03:00rasherAnd I'm sure you could tell it not to
23:03:27crzyboysterIf I didn't translate a string in the Hindi translation, can I just remove the entire string from <phrase> to </phrase>?
23:03:59amiconnIn order to guarantee that it has to read from disk, the plugin could be modified a bit more to make it wait for a button press between the write and the verify phase
23:04:08rashercrzyboyster: If you're going to translate it later, yes
23:04:12rashercrzyboyster: if not, why not?
23:04:14amiconnThen you could unmount and remount the sansa before pressing the button
23:04:33rasheramiconn: that would work indeed
23:04:55GodEaterrasher, I think you need to mount with -o sync to make it not cache
23:05:31rasherGodEater: ah yes (or pmount -s if you use that)
23:05:32 Quit linuxstb___ (Read error: 113 (No route to host))
23:05:42*GodEater goes to look up pmount
23:06:04rasherGodEater: convenicne thing to allow users to mount disks. Debian uses it by default.
23:06:49crzyboysterI wanted to do it so that I could generate a voice file with ease and also so that others can help translate the language at"> and also so that the translation status becomes accurate...
23:07:08thgzOk... When I copied rockbox_default.bdf from the fonts dir to the tools dir, ran there "convbdf -h rockbox_default.bdf" and renamed the output file to sysfont.h, moved it to the build dir and started make, all went fine until... Can't create sysfont.c in firmware dir. Have to create that by hand also.
23:07:31rashercrzyboyster: In that case, yes (I'm not sure what it has to do with voice though)
23:07:40***Saving seen data "./dancer.seen"
23:08:06rashercrzyboyster: Are you aware of"> ?
23:09:27crzyboysterSo the ones that are missing can be added in at the end of the voice file, too?
23:09:28 Join ol_schoola [0] (
23:09:35 Quit axionix (Read error: 110 (Connection timed out))
23:09:38amiconnrasher: The last line rather looks like an oversight in english.lang ....
23:10:00crzyboysterAnd what if I don't have translations for them?
23:10:06 Quit MethoS- ("Konversation terminated!")
23:10:18rasheramiconn: Indeed
23:11:08rashercrzyboyster: if you remove a <phrase>, it will be added the next time genlang is run with a "needs translation" note. Which seems to be what you want.
23:11:29Domonokycrzyboyster: about you forum post: if you use rbutil with espeak, espeak should not ask you anything (point the espeak path to the espeak console binary in rbutil)
23:11:38 Quit fehmicans ("Konversation terminated!")
23:12:11rasheramiconn: Ah no, that's because LANG_BUTTONLIGHT_TIMEOUT:c200,gigabeatf doesn't exist in english.lang (it's c200,gigabeatf,mrobe100 now)
23:12:24rasherI should probably change the wording of my page slightly
23:12:43amiconnAh, so the check is a bit broken...
23:12:55 Join MethoS- [0] (
23:13:28rasheramiconn: Well it's a misleading message, but I think the check is fine
23:13:31amiconnI would expect it to check features, and output something like 'voice string blah is missing for mrobe100'
23:13:35*gevaerts is untarring a linux source tree to his sansa
23:14:05rasheramiconn: that's noted elsewhere on the page
23:14:20rasherOr should be, at least
23:14:31crzyboysterDomonoky: Well, I pointed the tts engine to the exact espeak exe file and I was assuming that it would ask me to generate out each string from within espeak?
23:15:50Domonokycrzyboyster: it just uses espeak to speak all strings it gets out of svn fitting your device, language and installed revision..
23:15:57Lloreancrzyboyster: espeak takes text as input, and gives speech as output... users don't actually interact with espeak directly.
23:16:47Ice``does anyone here have a broken Creative Zen V plus?
23:16:54Ice``(It is for reverse purposes)
23:17:06crzyboysterAnd can I delete deprecated strings?
23:17:15 Quit TMM ("Ex-Chat")
23:17:26crzyboysterIce: I have a Zen V fully intact :D
23:17:52rashercrzyboyster: no, you should leave deprecated strings empty
23:17:56Ice``crzyboyster: Me too. And I would like to conserve it as it is.
23:18:20Ice``That's Y I 'd like to find a broken device on which I could do reverse.
23:18:56Ice``crzyboyster: are u workin' on a Zen V rockbox port?
23:19:07amiconngevaerts: Regarding the USB register setup, it seems like some RE is due. Best candidate would probably be the (in-ROM) emergency diskmode of an *older* PP502x ipod. (the emergency diskmodes of the G5 and nano are (intentionally?) crippled speed-wise)
23:19:17rasheramiconn: I had an error in the part that checked for missing voice strings, which I've fixed now. So now on eg. the Hindi page, it's noted that "LANG_BUTTONLIGHT_TIMEOUT:c200,gigabeatf,mrobe100" is missing, and "LANG_BUTTONLIGHT_TIMEOUT:c200,gigabeatf" should be removed
23:19:56amiconnSo the candidates would be 4th Gen (grey or colour) or one of the minis
23:20:32crzyboysterIce: No but I would like a rockbox port for it. It seems like its not going to happen with the Creative players having encrypted firmares and all
23:21:38Ice``There is tools to decrypt firmware. And then, as soon I get time and another device, I'll go on reverse and I'll start to code drivers.
23:22:08crzyboysterAnd on a side note, has there been progress on the Gigabeat S?
23:22:13Ice``(If I found data sheet for components =])
23:22:24GodEatercrzyboyster: it runs rockbox as far as reading the disk now
23:22:30GodEaterso you browse the filesystem
23:22:33 Quit n1s ()
23:22:37GodEaterand get to the various settings menus
23:22:39amiconnrasher: That reminds me - do you think finnish.lang (and hence .lng) should be renamed to suomi.lang?
23:22:51BigBambicrzyboyster: Check the logs for details and pray down to the jhMikeS statue
23:23:00gevaertsamiconn: is there a dump of the necessary firmware available somewhere ?
23:23:51linuxstbgevaerts: I've got one...
23:23:57amiconnThe general idea seems to be that the .lang file names are written in their own language, if that's possible using latin letters
23:24:06rasheramiconn: there's some discussion about that in the tracker.. I'm not sure how I feel about it. In theory yes, but then you get people with outdated language files
23:24:06Ice``BigBambi: Is it what it is called Plug and Pray? =]
23:24:29BigBambiheh, could be
23:24:39*gevaerts would like to warn people that he has no experience in disassembling firmware and then trying to understand it
23:24:40amiconnrasher: Yeah, too bad that a .zip cannot contain an "anti-file"
23:24:46rasheramiconn: unless we keep finnish.lang and break it
23:24:58rasherFor a transition period or something
23:25:07Ice``gevaerts: How do you want to get some experience if you don't try?
23:25:29 Join TMM [0] (
23:25:30Ice``I know it is _*VERY*_ difficult.
23:25:49amiconnHmm, czech.lang has the same problem
23:25:58*jhMikeS prefers forward engineering over reverse by far
23:26:06rasheramiconn: Don't they use crazy accents on some letters though?
23:26:06DerPapstgevaerts: infact you can download it from apple and extract it from a firmware update file, but getting it from linuxstb is faster and easier :-P
23:26:08gevaertsIce``: true, I just meant that if someone else does it it might be done earlier
23:26:20amiconnrasher: Hmm. It would not be the first time dropping a .lang file, leading to potential outdatedness
23:26:28rasheramiconn: I do agree that finnish.lang should be named suomi.lang if it was added today
23:26:31amiconn(we dropped polski-ascii at some point)
23:26:51rasheramiconn: Good point - I forgot about that.. were there any complaints?
23:27:01gevaertslinuxstb: I can't seem to get a connection
23:27:05rasherIf not, I'd say we can go and do the same with finnish.lang
23:27:40amiconnYes, czech uses some accents. But there are other .lang files which just drop the accents and use plain latin letters (e.g. francais)
23:27:41DerPapstlinuxstb: do you have a server to upload?
23:27:53Ice``gevaerts: yes.
23:27:57rasheramiconn: ah yes
23:28:02gevaertslinuxstb: got it
23:28:06rasherMoving to 7z and using unicode for all language names would be nice though, in this particular case..
23:28:09Ice``Find someone else =] Meanwhile, i'll try.
23:28:17DerPapstnvm :-P
23:28:29amiconnYes, of course that would be nice
23:28:55amiconnHmm, we could make the language selector special, but that would cost some binsize
23:29:07rasherI don't think that's worth it, to be honest
23:29:19amiconnThe .lng files could *contain* the language's name as unicode text
23:29:32amiconn...and the browser would show that instead of the file name
23:29:53 Quit TMM (Remote closed the connection)
23:29:53thgzNo problem now
23:30:02rasheramiconn: Another thing I've thought about, was that the .lng could contain all characters used, so it could be checked if the font supports that language
23:30:21thgzThe stereo crossfade patch somehow broke the whole compiling process. One re-download fixed the problem.
23:30:22rasherTo avoid "the screen went blank" problems
23:30:33amiconnHmm, I also thought about that. But an automatic decision would also be bad
23:30:41 Part thgz
23:30:55amiconnIt would reject a .lng even if just a single character isn't available
23:31:00Ice``gevaerts: do you have some experience on firmware reversing?
23:31:17gevaertsIce``: none at all
23:31:18Lloreanamiconn: It could always just give a confirmation screen. "Some characters are missing, press select to continue loading, any other key to cancel"
23:31:18GodEaterIce``: he already said he didn't...
23:31:41rasheramiconn: yeah, I don't think it should anything but a warning and/or refusal to load
23:31:45amiconnLlorean: Hmm. "Some" should be replaced by the actual number, I think
23:32:09rasherOn second thought, refusal would be bad
23:32:18amiconnThe next question is how to do that check in a quick way
23:32:22Lloreanamiconn: Fair.
23:32:30Ice``GodEater: I didn't read that, i apologize.
23:32:45amiconnAnd another thing is that loading a font would have to do the same check
23:33:07amiconnAlso, what about (potential) multi-font suppport?
23:33:12rasheramiconn: if the .lng file contain a list of used glyphs (handled by genlang), isn't it fairly simple to run through that and check if the current font has those? It's not terribly time-critical anyway?
23:33:31rasherYeah, multi-font would make that "interesting"
23:33:41Lloreanamiconn: I think Fonts aren't so dangerous
23:33:44gevaertsAm I right in thinking that 'mov r1, #-989855744 ; ' followed by 'ldr r0, [r1, #324]
23:34:20gevaerts' mean "load the contents of -989855744+324 (==USBSTS) to register r0" ?
23:34:34Lloreanamiconn: I *think* it might be clearer to users what's happening in that case, maybe
23:34:39linuxstbgevaerts: Yes
23:34:57gevaertslinuxstb: OK. Then I should be on my way
23:35:13 Quit Genesis ("Leaving")
23:35:24*amiconn recommends the arm reference manual
23:35:42 Quit ol_schoola (Connection timed out)
23:35:54*gevaerts guesses that amiconn has a point
23:35:58*amiconn wonders what disassembler outputs constants in decimal instead of hex
23:36:14amiconn-989855744 => 0xc5000000
23:36:23amiconnMuch more readable, imho
23:36:30 Join saratoga [0] (n=9803c6dd@gateway/web/cgi-irc/
23:37:01linuxstbgevaerts: There's an ARM disassembler in the utils/ directory in the Rockbox source - that's slightly nicer than objdump.
23:37:02rashergevaerts: sdhc write support seems absolutely solid
23:37:21rasherApart from the device resets
23:37:28saratogai would like to commit the PP power consumption patch Buschel posted
23:37:36saratogadoes anyone object?
23:37:53 Quit grndslm (Read error: 113 (No route to host))
23:39:06Ice``See ya.
23:39:11 Join framo [0] (
23:39:15Ice``Gd N8
23:39:20*amiconn didn't check out that patch at all so far
23:39:45GodEaterit's been floating around for quite a few weeks now...
23:39:54amiconnI know
23:40:00saratogaamiconn: i think you will want to
23:40:09gevaertslinuxstb: thanks. That will help
23:40:17 Join TMM [0] (
23:40:17 Join gtkspert [0] (
23:40:18saratogait changes the default clock speed on the pp50xx targets
23:40:22amiconnI also know that at least some versions do more or less nasty things
23:40:39Horschtyou might be refering to the PCF patches.
23:40:43saratogasince mpc and flac did not need 30 MHz to decode
23:41:07amiconnWhy does it change the default? 30MHz are already quite low for the UI to be responsive
23:41:17amiconn(on G5)
23:41:31linuxstbAnd what about plugins, some games may run fine at 30MHz, but not lower.
23:41:34*Domonoky has now replaced the rbutilcli.exe file with a really static version... so testing can go on.. :-)
23:41:42 Quit GodEater ("Going!")
23:42:46amiconnThat is, it's low for the UI on G5. Probably the greyscale targets could go lower. But so far we let all PP targets use the same clock frequencies on purpose
23:43:18saratoga(FS #8379 btw)
23:43:48saratogayes my thought was that the 5G would lag, but i've heard its not noticable
23:43:57amiconnThe device disabling should be fine, if it's verified that nothing breaks
23:44:23amiconnBut I would rather keep the frequencies, at least for now
23:44:30amiconnBetter don't change too much at once
23:47:32rasherBuild table is looking nicer
23:48:03LloreanBesides "power drain" (which happens with the backlight anyway) was there any reason not to boost on user input?
23:48:13*amiconn knows that he should probably have tried this patch, but is still looking for his motivation to come back :\
23:48:28saratogaLlorean: i think Buschel removed it because it caused a problem with some devices
23:48:33saratogaand he just hasn't gotten around to fixing it yet
23:48:57 Quit raky (Read error: 113 (No route to host))
23:48:57 Quit crzyboyster ("CGI:IRC (EOF)")
23:49:11amiconnWe already had that at some point. It was removed again
23:49:18saratogai think his original idea was something like 30MHz when the back light is on, 16 when its off, and boosting to 80
23:49:44Lloreansaratoga: I'd rather just "set to boosted any time there's user input, unboost 1 second after the last input" personally
23:49:49saratogaor maybe i'm not remembering it correctly
23:49:54amiconnI would never go below 24MHz (well, except for suspend of course)
23:49:58LloreanYou can't assume they'll have the backlight on, and there's really no reason to keep it boosted the whole time the backlight is on.
23:50:06saratogatheres tests done at 12MHZ I think
23:50:11amiconn24MHz can be produced with the pll disabled
23:50:32saratogaamiconn: unfortunately disabling the PLL did not save any power
23:51:19amiconnYeah, so there is one more interesting test to do: Does the pll consume more power if it outputs 80MHz all the time?
23:51:27 Join Rob2222 [0] (
23:51:38amiconnIf it doesn't, we could boost/unboost by just switching clock sources
23:52:04saratogai didn't realize that was possible
23:52:05amiconnThis would be nearly instantaneous, i.e. no wait-for-relock delay
23:52:18saratogai wonder if we weren't really disabling the PLL, but rather just switching off it
23:52:26saratogathat might expalin why it only seemed to use 30uA
23:52:52amiconnWell, it's rather obvious that we can switch sources
23:53:29amiconnThe boost/unboost code in fact has to do that. It switches to 24MHz direct clock, reprograms the pll, waits for relock, and then switches to pll
23:53:31 Quit gtkspert_ (Read error: 101 (Network is unreachable))
23:53:54amiconnIf it would reprogram the pll while running from it, it would lock up
23:54:13 Join ol_schoola [0] (
23:54:21amiconnTheres even another option - the pll has a programmable post-divider
23:55:08saratogaamiconn: can you look at this patch and tell me if I was actually disabling the PLL when I tested?
23:55:19Domonokyit should be possible to change the postdividers while running from PLL.. i think..
23:55:28saratogaaround the line "Note2: CLOCK_SOURCE is set via 0=32kHz, 1=16MHz"
23:56:12amiconnIt divides the pll output by an integer number between 1 and 16
23:57:03amiconnThe Note2 might apply to some dev board, but not to most targets
23:57:10 Quit ol_schoola (Client Quit)
23:57:24amiconnNot all selectable sources are actually present
23:58:06amiconnOn my mini G2, only 0, 2, and 5..7 work

Previous day | Next day